Joe Friel - Jim Vance - Triathlon Science-Human Kinetics Inc

Download as pdf or txt
Download as pdf or txt
You are on page 1of 664

Triathlon

Science
Joe Friel
Jim Vance
Editors

Human Kinetics
Library of Congress Cataloging-in-Publication Data
Triathlon science / Joe Friel, Jim Vance, editors.
p. cm.
Includes bibliographical references and index.
1. Triathlon. 2. Sports sciences. I. Friel, Joe. II. Vance, Jim.
GV1060.73.T77 2013
796.42'57--dc23
2012036423
ISBN-10: 1-4504-2380-9 (print)
ISBN-13: 978-1-4504-2380-9 (print)

Copyright © 2013 by Human Kinetics, Inc.

All rights reserved. Except for use in a review, the reproduction or utilization of this work in any form
or by any electronic, mechanical, or other means, now known or hereafter invented, including xero-
graphy, photocopying, and recording, and in any information storage and retrieval system, is forbidden
without the written permission of the publisher.
This publication is written and published to provide accurate and authoritative information relevant
to the subject matter presented. It is published and sold with the understanding that the author and
publisher are not engaged in rendering legal, medical, or other professional services by reason of their
authorship or publication of this work. If medical or other expert assistance is required, the services
of a competent professional person should be sought.
The web addresses cited in this text were current as of December 2012, unless otherwise noted.
Developmental Editor: Cynthia McEntire; Assistant Editor: Elizabeth Evans; Copyeditor: Bob
Replinger; Indexer: Nan Badgett; Permissions Manager: Martha Gullo; Graphic Designer: Joe Buck;
Graphic Artist: Kim McFarland; Cover Designer: Keith Blomberg; Photograph (cover): Daniel Swee/
Action Plus/Icon SMI; Photographs (interior): Neil Bernstein, except where otherwise noted; Photo
Asset Manager: Laura Fitch; Visual Production Assistant: Joyce Brumfield; Photo Production Mana-
ger: Jason Allen; Art Manager: Kelly Hendren; Associate Art Manager: Alan L. Wilborn; Illustrations:
© Human Kinetics, except where otherwise noted; Printer: Sheridan Books
Human Kinetics books are available at special discounts for bulk purchase. Special editions or book
excerpts can also be created to specification. For details, contact the Special Sales Manager at Human
Kinetics.
Printed in the United States of America 10 9 8 7 6 5 4 3 2 1
The paper in this book is certified under a sustainable forestry program. The paper in this book was
manufactured using responsible forestry methods.
Human Kinetics
Website: www.HumanKinetics.com
United States: Human Kinetics Australia: Human Kinetics
P.O. Box 5076 57A Price Avenue
Champaign, IL 61825-5076 Lower Mitcham, South Australia 5062
800-747-4457 08 8372 0999
e-mail: humank@hkusa.com e-mail: info@hkaustralia.com
Canada: Human Kinetics New Zealand: Human Kinetics
475 Devonshire Road Unit 100 P.O. Box 80
Windsor, ON N8Y 2L5 Torrens Park, South Australia 5062
800-465-7301 (in Canada only) 0800 222 062
e-mail: info@hkcanada.com e-mail: info@hknewzealand.com
Europe: Human Kinetics
107 Bradford Road
Stanningley
Leeds LS28 6AT, United Kingdom
+44 (0) 113 255 5665
e-mail: hk@hkeurope.com E5643
Contents
Introduction: The Tri-Knowledge Advantage . . . . . . . . . . . . . . . . ix
Joe Friel

Part I Physical Attributes of Triathletes 1


Chapter 1 Physiology and the Multisport Athlete . . . . . . . . . . 3
Randall L. Wilber

Chapter 2 Genetics and Inheritance in Triathlon


Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Ildus I. Ahmetov and Malcolm Collins

Chapter 3 Gender and Age Considerations in Triathlon . . . . . 39


Romuald Lepers

Part II Technical Execution and Efficiency


in Each Event 55
Chapter 4 Swimming Biomechanics for Triathlon . . . . . . . . . 57
David Pease

Chapter 5 Cycling Biomechanics for Triathlon . . . . . . . . . . . . 71


Jeff Broker and Sean Langlais

Chapter 6 Running Biomechanics for Triathlon . . . . . . . . . . . 87


George M. Dallam

Part III Environmental Factors and Equipment


Options 99
Chapter 7 In the Water . . . . . . . . . . . . . . . . . . . . . . . . . . . .101
Bruce R. Mason and Gina Sacilotto

iii
iv } Contents

Chapter 8 On the Bike . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113


Jeff Broker and Sean Langlais

Chapter 9 For the Run . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129


George M. Dallam

Chapter 10 Triathlon Training Technologies . . . . . . . . . . . . . . 143


Hunter Allen

Part IV Physiological Function in Triathlon


Training 159
Chapter 11 Aerobic Capacity . . . . . . . . . . . . . . . . . . . . . . . . 161
Ross Tucker

Chapter 12 Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171


Ross Tucker

Chapter 13 Anaerobic Threshold . . . . . . . . . . . . . . . . . . . . . 181


Ross Tucker

Chapter 14 Muscle Types and Triathlon Performance . . . . . . . 191


Ross Tucker

Chapter 15 Fatigue Resistance and Recovery . . . . . . . . . . . . . 199


Ross Tucker

Part V Training Modes and Methods for


Triathletes 209
Chapter 16 Warm-Up and Cool-Down . . . . . . . . . . . . . . . . . 211
David Warden

Chapter 17 Flexibility and Core Strength . . . . . . . . . . . . . . . 223


David Warden

Chapter 18 Strength Training . . . . . . . . . . . . . . . . . . . . . . . . 249


David Warden
Contents |v

Chapter 19 General and Specific Training . . . . . . . . . . . . . . . 269


David Warden

Chapter 20 Interval Training . . . . . . . . . . . . . . . . . . . . . . . . 285


David Warden

Part VI Training Strategies in Triathlon 297


Chapter 21 Duration, Frequency, and Intensity . . . . . . . . . . . 299
Stephen J. McGregor

Chapter 22 Periodization . . . . . . . . . . . . . . . . . . . . . . . . . . . 311


Stephen J. McGregor

Chapter 23 Tapering and Peaking for Races . . . . . . . . . . . . . 327


Stephen J. McGregor

Chapter 24 Physiology of Overtraining . . . . . . . . . . . . . . . . . 339


Matt Fitzgerald

Part VII Training Base Building for Triathlon 351


Chapter 25 Swim Base Building . . . . . . . . . . . . . . . . . . . . . . 353
Jim Vance

Chapter 26 Bike Base Building . . . . . . . . . . . . . . . . . . . . . . . 363


Neal Henderson

Chapter 27 Run Base Building . . . . . . . . . . . . . . . . . . . . . . . 373


George M. Dallam

Part VIII Multisport Event-Specific Training


and Racing Tactics 385
Chapter 28 Sprint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
Gale Bernhardt

Chapter 29 Olympic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399


Gale Bernhardt
vi } Contents

Chapter 30 Half-Ironman . . . . . . . . . . . . . . . . . . . . . . . . . . 409


Gale Bernhardt

Chapter 31 Ironman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419


Gale Bernhardt

Chapter 32 Duathlon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431


Gale Bernhardt

Chapter 33 Combination Workout Training . . . . . . . . . . . . . . 441


Gale Bernhardt

Part IX Sports Medicine for Triathletes 449


Chapter 34 Triathlete Body Maintenance and Medical Care . . 451
John Post

Chapter 35 Triathlon Injuries and Preventive Measures . . . . . 461


Nathan Koch

Chapter 36 Triathlon Injury Recovery Techniques . . . . . . . . . 497


Nathan Koch

Part X Nutrition for Triathletes 513


Chapter 37 Energy Needs, Sources, and Utilization . . . . . . . . 515
Bob Seebohar

Chapter 38 Nutrition Periodization . . . . . . . . . . . . . . . . . . . . 525


Bob Seebohar

Chapter 39 Nutrient Timing for Triathlon Training


and Racing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535
Bob Seebohar

Chapter 40 Supplements for Triathletes . . . . . . . . . . . . . . . . 543


Bob Seebohar
Contents | vii

Part XI Psychology of Multisport 553


Chapter 41 Mental Toughness for Triathlon . . . . . . . . . . . . . 555
JoAnn Dahlkoetter

Chapter 42 Psychology of Triathlon Training . . . . . . . . . . . . . 563


JoAnn Dahlkoetter

Chapter 43 Mental Skills for Peak Triathlon Performance . . . . 573


JoAnn Dahlkoetter

Epilogue: The Application of Science in Triathlon . . . . . . . . 583


Joe Friel

References 585
Index 621
About the Editors 639
About the Contributors 641
This page intentionally left blank.
Introduction
The Tri-Knowledge Advantage
—Joe Friel, MSc

W hether you are a triathlete, coach, sport scientist, or student of the


sport, if you are passionate about the science of triathlon, you are in
for a treat. Human Kinetics, Jim Vance, and I have brought together some of
the most knowledgeable people in the many fields related to sport science
to contribute to this book. Here you will find the scientific underpinnings
of a wide range of triathlon topics from physiology, gender, and genetics to
biomechanics, training methods, and psychology—and more. The book is
a treasure trove of information that will help you understand what it takes
to succeed in this complex sport.
No project like this has ever been done for triathlon. I promise that you
will learn a lot. I know I did, having read each of the chapters before publi-
cation. And I’ve been around the sport for a long time.
I came to triathlon in 1983 when I did my first race in Longmont, Colo-
rado, a race just short of what we now call the Olympic distance. It was a
pool swim. After many years of only running before trying this new sport,
my stroke mechanics were pathetic. In those days we didn’t believe that
technique was important. It was all about building fitness in the pool with
endless intervals. The bike I rode was decent for the early 1980s—a Fuji
road bike that probably weighed 25 pounds (11 kg). In those days there was
no such thing as a triathlon bike. It would be 6 more years until the first
of those appeared in bike shops. I came from a running background, but
my ego was deflated on that race day by how slowly I ran after an hour of
pushing my limits on the bike. Of course, power meters, GPS devices, and
accelerometers were unheard of. I would get my first heart rate monitor later
that summer. Racing back then was only about going as hard as you could
for as long as you could.
In those early days we had no reason to think that triathlon would be
anything other than a fun way for swimmers, cyclists, and runners to com-
pete with each other. We never dreamed that it would one day become the
fastest growing sport in the world or that in less than 20 years it would be
an event in the Olympic Games. Triathlon was just something we all did as
a break from our “real” sports.
In the 1970s I had gone back to school to work on a master’s degree in
exercise science, in part so I could become a better runner. I had high aspira-
tions as an athlete and had always been curious about how to train to run

ix
x } Introduction

faster. I figured the answers to all my training questions could be found


within those ivy-covered walls. After 3 years of study I came away with
few answers, but with much better questions. And along the way I learned
of the role of research in seeking answers to such questions. Ever since then
I have been a student of the science of training for sport. To this day I read
research studies the way most people read novels—just for fun. So when
I was asked to be a part of this book project, I jumped at the opportunity.
I knew I would need help with such a huge project, so I asked Jim Vance
to come onboard to assist. Jim has been a coaching associate with me for
4 years. He also comes from a running background, having competed in
track and field at the University of Nebraska. After graduation, he took up
mountain bike racing, which eventually led him to triathlon. He soon turned
pro and moved up to the Ironman distance. Jim was a public school teacher
before becoming a coach, so he understands how to help athletes become
more fit and faster by learning as well as doing. I was Jim’s coach before he
retired from professional racing to become a full-time coach. So I knew that
we could work well together on this book.
We started on the project in the spring of 2011 by deciding what should be
included, and we then made a list of people we would like to have onboard
as contributors. The list was impressive—and we got them all. They each
fall into one of three categories: academics with advanced degrees, coaches
with high levels of certification and unparalleled experience, and medical
practitioners known for their treatment of endurance athletes. Several of
the contributors fit in two of these categories. All are athletes, and many
have competed or continue to compete at remarkably high levels given
their busy lives. They come from five countries—Australia, France, Russia,
South Africa, and the United States—on four continents. You can read brief
biographies about each of them in the About the Contributors section. Their
accomplishments are remarkable. In the sport of triathlon, never has such a
high-powered group with so much cutting-edge information been brought
together with a single mission: to help you understand better what it takes
to excel as a triathlete.
Here’s a brief overview to what you can expect from the contributors to
Triathlon Science:
In part I you will learn about the physiology of triathlon from Randy
Wilber, PhD; about genetics from Ildus Ahmetov, PhD, and Malcolm
Collins, PhD; and about the training needs of special populations from
Romuald Lepers, PhD.
Part II takes an in-depth look at the movement skills of each of the three
disciplines. Chapters are written by David Pease, PhD, from the Aus-
tralian Institute of Sport; Jeff Broker, PhD; and George Dallam, PhD.
Part III examines the effect of environmental conditions and equipment,
including computer software, on performance. Contributors are Bruce
Introduction | xi

Mason, PhD; Gina Sacilotto; Jeff Broker, PhD; Sean Langlais; George
Dallam, PhD; and coach Hunter Allen.
Ross Tucker, PhD, of the Sports Science Institute of South Africa, con-
tributed all of part IV on the physiological determinants of success in
triathlon.
U.S.-based coach David Warden applies much of what preceded his part V
with a discussion of the application of science to training for triathlon.
David is well known for in-depth discussions of the science of training
in his Tri-Talk podcasts.
In part VI Stephen McGregor, PhD, and coach and author Matt Fitzgerald
provide an overview to training in three disciplines using a periodiza-
tion-based model.
In part VII Jim Vance; Neal Henderson; and George Dallam, PhD, explain
the details of building general, or base, fitness as a component of the
periodization model.
Long-time coach, author, and associate Gale Bernhardt shows how to
build beyond base fitness to prepare to race at the standard distances
in part VIII.
In part IX two of the leaders in sports medicine for endurance athletes,
John Post, MD, and Nate Koch, PT, ATC, describe their cutting-edge
injury prevention and treatment techniques.
Well-known coach, author, and nutritionist Bob Seebohar explains how
triathletes can fuel their bodies to perform better in part X.
Finally, in part XI noted sports psychologist JoAnn Dahlkoetter, PhD,
describes mental preparation for competition. This aspect of the sport
is probably the least understood among triathletes at all levels.
In all, the book contains 43 chapters, each written by a leader in the field
to explain the cutting-edge concepts and methods that will help serious tri-
athletes train and race to their potential. I’m certain that after you’ve read
Triathlon Science you will have a much deeper understanding of what it takes
to train and perform better in all three disciplines.
This page intentionally left blank.
Pa r t

I
Physical Attributes
of Triathletes
This page intentionally left blank.
CHAPTEr
1
Physiology and the
Multisport Athlete
—Randall L. Wilber, PhD, FACSM

T his chapter provides advanced information on exercise physiology nec-


essary for coaching or participating in triathlon. Here we will examine
the athlete’s energy production systems, cardiopulmonary physiology, and
skeletal muscle contraction.
The sport of triathlon presents a unique physiological challenge in endur-
ance sport. Whether you participate recreationally in the local sprint triathlon
or are dedicated to competing in the grueling Ironman triathlon, the three-in-
one triathlon format requires a demanding but intelligent and well-balanced
training program. Knowledge of the science of triathlon has become critical
to success in the sport, particularly in the Ironman race format.
Although each of the three disciplines in triathlon offer unique training
challenges, some basic physiological principles can be applied to swimming,
cycling, and running. This chapter will help triathletes and their coaches
understand the basic physiology of endurance sport. Three primary topics will
be presented: energy production, cardiopulmonary physiology, and skeletal
muscle contraction. Better scientific knowledge means better performance.
The goal of this chapter, therefore, is to build a strong physiological foun-
dation from which triathletes and coaches of triathletes can design effective
and successful training programs. Added to the basic scientific information
presented here is expertise in swimming, cycling, and running provided in
subsequent chapters.

Energy Production
Every triathlete has experienced the bonk, most likely in competition but per-
haps on a long training ride or run. A significant amount of physical energy
is required to train for and compete in triathlon. This section deals with the
physiology of energy production. It will help you understand how your body

3
4 } Wilber

produces energy. Practical knowledge of the three energy systems can help
you design effective workouts and pace yourself more effectively during races.
What exactly is energy? How is energy produced and used in a triathlete’s
body? The answers to these questions rely heavily on the sciences of biology
and biochemistry.

Energy Systems
The basic unit of energy within the human body is adenosine triphosphate
(ATP). A simple way to think of a molecule of ATP is as an energy dollar bill.
Millions of molecules of ATP in the human body are providing energy. Tri-
athletes are constantly using and replenishing ATP, even when not exercising.
ATP utilization and production is similar to the daily scenario of spending
money to pay bills and maintain a lifestyle.
The molecular structure of ATP is shown in figure 1.1. ATP is made up of
three unique subunits: adenine, ribose, and the phosphate groups. Notice the
wavy lines that connect the three phosphate groups, each of which represents
a high-energy bond.

H H
N

N
N
O O O H
–O P O P O P O CH2 N N H
O Adenine
O– O– O– H H
Phosphate groups
H H

HO OH
Ribose
Adenosine
Adenosine monophosphate (AMP)
Adenosine diphosphate (ADP)
Adenosine triphosphate (ATP)

„„Figure 1.1 Structure of adenosine triphosphate (ATP).

E5643/Friel/Fig. 1.1/448502/TimB/R2-kh
The basic biochemical reaction whereby ATP produces energy is shown
in figure 1.2. A single molecule of ATP is represented on the left side of the
reaction. When ATP comes in contact with water and the enzyme ATPase,
one of its high-energy bonds is broken, or cleaved, which releases a burst of
chemical energy. This burst of chemical energy can be used for several impor-
tant physiological functions including nerve transmission, blood circulation,
tissue synthesis, glandular secretion, digestion, and the physiological func-
tion that we will focus on later in this chapter, skeletal muscle contraction.
Physiology and the Multisport Athlete |5

ATPase
ATP ADP + Pi + ENERGY
H2O

SKELETAL MUSCLE CONTRACTION


Nerve transmission
Blood circulation
Tissue synthesis
Glandular secretion
Digestion

„„Figure 1.2 Biochemical conversion of ATP to ADP + Pi + energy.


E5643/Friel/Fig. 1.2/448503/TimB/R1

There are three energy-producing systems (see figure 1.3): the adenosine
triphosphate–creatine phosphate (ATP–CP) system, which is immediate;
glycolysis, which is short term; and oxidative phosphorylation, which is long
term. The three energy systems are similar in that they all produce ATP, but
they differ in how quickly and how much ATP they produce. Two of the three
energy systems—immediate and short term—are anaerobic energy systems.
In other words, these two energy systems do not require oxygen to produce
ATP. In contrast, the long-term energy system is aerobically dependent, requir-
ing oxygen to produce ATP.

ATP

Anaerobic Aerobic

1 2 3
Immediate Short-term Long-term
ATP/CP Glycolysis Oxidative Phosphorylation
<10 sec. 30-120 sec. >5 min.

ATP/CP Glucose Glucose/fat

„„Figure 1.3 The three energy-producing systems.


E5643/Friel/Fig. 1.3/448504/TimB/R3-kh

Immediate Energy System


The first energy system is the immediate energy system, one of the two anaerobic
energy systems. The technical name for this energy system is the adenosine tri-
phosphate (ATP)–creatine phosphate (CP) system. The immediate energy system
is for immediate use. The biochemical reactions involved in this system are
shown in figure 1.4. Notice that the first reaction is the same one described
6 } Wilber

earlier in the chapter that showed the conversion of ATP to chemical energy
(see figure 1.2). Recall that one of the high-energy bonds was cleaved in that
reaction, which resulted in adenosine triphosphate (ATP), which contains
three phosphate groups, being converted to adenosine diphosphate (ADP),
which contains two phosphate groups. Notice that ADP is not simply thrown
away after the initial reaction (figure 1.4). Rather, it goes through a recycling
process with CP, which donates its phosphate group to ADP (two phosphate
groups) to produce a new molecule of ATP (three phosphate groups).
The immediate energy system has the advantage of producing ATP quickly
but the disadvantage of producing only a limited supply of ATP. In terms of
athletic performance, the immediate energy system is the dominant energy
system during high-intensity, short-duration exercise lasting approximately
10 seconds or less. The immediate energy system is used in athletic per-
formances like the 100-meter sprint in track, 10-meter diving events, and
weightlifting events.

ATPase
ATP ADP + Pi + ENERGY
H 2O

CK
CP + ADP ATP + Cr

„„Figure 1.4 Two basic biochemical reactions of the


ATP–CP energy system, the 1.4/448505/TimB/R1
E5643/Friel/Fig. immediate energy system.

Short-Term Energy System


The second energy system is the short-term energy system. Like the immediate
energy system, it is anaerobic because it does not require oxygen to produce
ATP. The technical name for this system is glycolysis because the first of
several biochemical reactions in this energy system involves the conversion
of glycogen to free glucose. A simplified version of glycolysis is shown in
figure 1.5. One molecule of glucose is converted to two molecules of pyruvic
acid, and then, in the absence of oxygen, the two molecules of pyruvic acid
are converted to two molecules of lactic acid. Notice that in glycolysis two
molecules of ATP are produced per one molecule of glucose.
The short-term energy system has the advantage of producing more ATP
than the immediate energy system but the disadvantage of taking longer to do
so. Another disadvantage is that the short-term energy system produces lactic
acid, which is quickly converted to lactate and positively charged hydrogen
ions (H+) (see figure 1.5). High concentrations of H+ create the acidic burning
sensation in exercising skeletal muscle that, along with other biochemical,
neural, and biomechanical factors, contributes to premature fatigue.
Physiology and the Multisport Athlete |7

Glucose
PFK

2 pyruvic acid

NO O2

2 lactic acid + 2 ATP

C 3H 5O 8-
+ H+
Lactate

„„Figure 1.5 Biochemical reactions involved in glycolysis,


the short-term energy system.
E5643/Friel/Fig. PFK = phosphofructokinase.
1.5/448506/TimB/R2-kh

In terms of athletic performance, the short-term energy system is the


dominant energy system during high-intensity, moderate-duration exercise
lasting approximately 30 to 120 seconds. Examples of the short-term energy
system being used in athletic performance include the 400-meter sprint in
track, 100-meter sprint in swimming, and 1,000-meter track cycling event.

Long-Term Energy System


The third energy system is the long-term energy system, which is aerobic in
nature and requires oxygen to produce ATP. The technical name for this
energy system is oxidative phosphorylation. A simplified version of this relatively
complex energy system is shown in figure 1.6. Notice that the long-term energy
system starts out the same way as the short-term energy system does as a
single molecule of glucose converted to two molecules of pyruvic acid. Because
oxygen is available, however, pyruvic acid is not converted to lactic acid as
in the short-term energy system. Rather, pyruvic acid is converted to acetyl
coenzyme A (acetyl CoA), which then enters one of several mitochondria in
Glucose
PFK

2 pyruvic acid
O2

2 acetyl CoA

Krebs
Electron transport system 32 ATP
cycle

„„Figure 1.6 Biochemical reactions involved in oxidative phosphorylation,


the long-term energy system. PFK = phosphofructokinase.
E5643/Friel/Fig. 1.6/448507/TimB/R1
8 } Wilber

the cell and goes through a series of biochemical reactions (Krebs cycle and
electron transport system [ETS]) that ultimately produce 32 molecules of ATP.
The long-term energy system has the advantage of producing very large
amounts of ATP compared with the other energy systems, but it has the
disadvantage of taking more time to produce that relatively large amount
of ATP. The long-term energy system takes longer because it uses oxygen to
produce ATP. The only place in the cell where oxygen can be used to produce
ATP is in the mitochondrion, which is essentially a large ATP factory that
requires several steps for processing, increasing the time needed for the final
production of ATP.
In terms of athletic performance, the long-term energy system is the
dominant energy system in low- to moderate-intensity, long-duration exer-
cise lasting longer than 5 minutes. The long-term energy system is used in
athletic performances like the marathon, 800-meter swim, and road cycling
events. The long-term energy system is the dominant energy system used
during triathlon training and racing, but it is not the only energy system
used in triathlon.

Energy Dynamics During Exercise


The previous section examined each of the three energy systems separately.
But note that more than one energy system can function at once. Consider a
symphony orchestra, in which several different instrument groups play soft
or moderate or loud depending on the musical score. At the beginning of
the symphony, the string group may be loud, the woodwind group may be
moderate, and the percussion group may be soft. These musical emphases
may be reversed by the end of the symphony to reflect soft music by the
string group and loud music by the percussion group. The same is true for
energy production during exercise. Each of the three energy systems is in a
state of dynamic flux. Like the various instrument groups, each of the energy
systems is operating constantly during exercise but at a different level of ATP
production depending on the intensity and duration of the exercise.
An example of the symphony orchestra effect is shown in figure 1.7, which
shows energy dynamics during a cycling road race. During pack riding, the
exercise intensity is moderate and the duration is relatively long. As described
in the previous section, the dominant energy system during moderate-inten-
sity, long-duration exercise is the oxidative phosphorylation (long-term) energy
system. Notice that it is dominant but not the only energy system used during
pack riding. The other two energy systems are active but playing softly.
During a hill climb, the intensity picks up but the duration is shorter
compared with pack riding. This type of high-intensity, moderate-duration
exercise requires the glycolytic (short-term) energy system to play loudest, the
ATP–CP system to play louder, and the oxidative phosphorylation system to
play softer versus pack riding. Finally, notice how the energy dynamics are
reversed during the high-intensity, short-duration final sprint to the finish.
Physiology and the Multisport Athlete |9

100 ATP/CP
90 Glycolysis
Oxidative phosphorylation
80

Energy contribution (%)


70
60
50
40
30
20
10
0
Pack riding Hill climb Final sprint

„„Figure 1.7 Energy dynamics during a cycling road race.


E5643/Friel/Fig. 1.7/448509/TimB/R1
Here the ATP–CP system is clearly the loudest, whereas the glycolytic (short-
term) and oxidative phosphorylation (long-term) energy systems are relatively
quiet.
In triathlon the dominant energy system is the oxidative phosphorylation
system. But remember the role of the ATP–CP and glycolytic energy systems
to triathlon performance. Knowing when to train and how much time to
devote to training each of the three energy systems is an important ingredient
of success in triathlon and is reflective of a well-designed and scientifically
based training plan.

Cardiopulmonary Physiology
Because triathlon relies heavily on the oxidative phosphorylation energy
system, understanding the basic concepts of cardiopulmonary physiology is
important. The term cardiopulmonary refers to the heart and lungs and to the
way in which those vital organs work in synchrony to ensure that the blood
is carrying oxygen and nutrients optimally to the working skeletal muscles
during exercise.

Cardiopulmonary Anatomy
The primary anatomical structures of the cardiopulmonary system are illus-
trated in figure 1.8, which shows the lungs (top), heart (center), and skeletal
muscles (bottom). An anatomical tour of the cardiopulmonary system begins
in the lungs. Blood passes through the capillary beds of the lungs where it
unloads carbon dioxide (CO2) and picks up oxygen (O2). This oxygen-enriched
blood travels from the lungs to the heart through the pulmonary vein.
Oxygen-enriched blood initially enters the heart in the left atrium and
then flows into the left ventricle. When the heart contracts, oxygen-enriched
10 } Wilber

blood is ejected from the left ventricle and exits the heart through the aorta.
The aorta ultimately branches into several smaller arteries that carry oxygen-
enriched blood to the entire body. After the oxygen-enriched blood reaches,
for example, the leg muscles during cycling or running, it unloads O2 and
picks up CO2. Blood exiting the exercising muscles is oxygen reduced and
returns to the heart through the venous system.
Oxygen-reduced blood is ultimately delivered to the heart by two large
veins, the superior and inferior vena cava. The vena cavae deliver oxygen-
reduced blood to the right atrium of the heart, where it then flows into the
right ventricle. When the heart contracts, oxygen-reduced blood is ejected by
the right ventricle and travels through the pulmonary artery to the lungs,
which brings it back to the starting point of the tour.

Lung
From To
pulmonary pulmonary
artery Capillaries vein
Alveoli
PO2 = 40 PO2 = 105 PO2 = 100
PCO2 = 46 PCO2 = 40 PCO2 = 40
O2
CO2

Right atrium
and ventricle

Left atrium
and ventricle

O2
Systemic Systemic
veins CO2 arteries

PO2 = 40 PCO2 = 46 Muscle PO2 = 100 PCO2 = 40


fibers

Capillaries

Muscle

„ Figure 1.8 Cardiopulmonary anatomy and partial pressure gradients that promote
O2 and CO2 movement inE5643/Friel/Fig
the body. PO21.8/448510/alw/pulled-R1
= O2 partial pressure, PCO2 = CO2 partial pressure.

Oxygen Transport
Triathlon is heavily dependent on the oxidative phosphorylation energy
system for ATP. In the previous section, O2 transport was referred to in general
terms: oxygen enriched and oxygen reduced. In this section, O2 transport is
Physiology and the Multisport Athlete | 11

described in more detail, focusing on the gas physics and physiology of O2


transport.
What exactly carries oxygen around in the body? A small percentage of
O2 travels through the body dissolved in the fluid portion of the blood. But
the primary way by which O2 is transported through the body is by the red
blood cells (RBCs), also called erythrocytes.
Human blood contains trillions of red blood cells. The RBC portion of the
blood is referred to as the hematocrit (Hct), which is expressed as a percent-
age of the total number of RBCs relative to total blood volume. Hematocrits
for healthy people residing at low elevation range from 35 to 45 percent for
women and 40 to 50 percent for men. A single RBC contains about 250 million
molecules of hemoglobin (Hb). The hemoglobin molecule transports oxygen
throughout the body. A single molecule of Hb can transport four molecules of
O2. Thus, a single RBC has the capacity to transport 1 billion molecules of O2.
Next, it is important to understand how oxygen-reduced blood becomes
oxygen-enriched blood in the lungs (see figure 1.8). The entire process of O2
transport is regulated by changes in the partial pressure of oxygen (PO2) that
take place from the moment ambient air is inhaled through the nose and
mouth until it reaches the body’s tissues and organs. PO2 decreases as the
inspired air moves from the nose and mouth to the lungs.
Specifically, the PO2 of inspired air at sea level is approximately 149 mil-
limeters of mercury (mm Hg), which drops to 105 mm Hg in the lungs. As
previously described, blood entering the lungs through the pulmonary
arteries contains RBCs that are relatively low in O2, or oxygen reduced. The
PO2 of this oxygen-reduced blood is approximately 40 mm Hg. This pressure
difference (105 mm Hg in the lungs versus 40 mm Hg in the oxygen-reduced
blood), or pressure gradient, favors the diffusion of O2 from the lungs to the
oxygen-reduced blood where it binds to hemoglobin molecules. The diffusion
of O2 from the lungs to the blood takes only about 0.75 seconds and occurs
across a sheer membrane in the pulmonary capillaries that is approximately
1/10,000 the width of a facial tissue. Because of this pressure gradient and
diffusion of O2 in the lungs, oxygen-enriched blood exits the lungs with a
PO2 of 100 mm Hg. The oxygen-enriched blood is transported by the pulmo-
nary vein to the left ventricle of the heart, from which it is then circulated
throughout the body.
When oxygen-enriched blood arrives at the capillary bed of a skeletal
muscle, the pressure gradient favors the release of O2 from hemoglobin (PO2
of approximately 100 mm Hg) to the skeletal muscle (PO2 of about 30 mm
Hg) (figure 1.8). The oxygen unloaded in the skeletal muscle can now be
used by the mitochondria to produce ATP by the oxidative phosphorylation
energy system. Finally, the blood exits the skeletal muscle capillary bed in
an oxygen-reduced state (PO2 of about 40 mm Hg) and returns to the right
ventricle of the heart to repeat the process of oxygenation in the lungs and
O2 transport throughout the body.
12 } Wilber

Training Effects
The ability to transport oxygen efficiently is clearly an important factor con-
tributing to optimal performance in triathlon, which is heavily dependent
on oxidative energy production. Of course, one question that immediately
comes to mind among triathletes and coaches is, Can the cardiopulmonary
system and oxygen transport capabilities be improved through training?
The answer is yes.
One way to improve O2 transport is to undertake altitude training, which
has the effect of increasing the number of RBCs and Hb molecules, resulting
in an increased capacity to get oxygen to the exercising muscles. (Unfortu-
nately, some unethical athletes have chosen to induce the same physiological
effect by using illegal pharmacological ergogenic aids such as recombinant
human erythropoietin [rhEPO].) Many athletes, however, do not have the
time or resources to undergo altitude training for a duration that will bring
about an increase in RBC and Hb.
Note that several positive cardiopulmonary training effects can be acquired
at sea level by a well-designed and scientifically sound training program that
involves regular endurance training. Regular endurance training is defined
here as a minimum of 30 to 45 minutes per training session and a minimum
of three to five training sessions per week for at least 8 weeks. Several ben-
eficial cardiopulmonary adaptations occur:
• Decrease in resting and exercise heart rate
• Increase in total blood volume
• Increase in cardiac output
• Increase in exercise respiratory capacity
.
• Increase in maximal oxygen uptake (VO2max)
• Improvement in lactate threshold (LT)
• Improvement in maximal exercise performance
• Improvement in exercise economy
• Improvement in endurance performance
• Improvement in heat tolerance
• Decrease in total body weight
• Decrease in body fat
• Decrease in blood pressure (if moderate or high blood pressure exists)

Lower Resting and Exercise Heart Rates


Through regular endurance training, the heart becomes stronger because
of progressive overload. Because the heart is stronger, heart rate at rest and
during exercise is lower than it was before the endurance training program
began. During exercise, athletes can expect heart rate to be lower at a specific
Physiology and the Multisport Athlete | 13

workload. For example, if heart rate taken immediately after running 800
meters on the first day of training was 175 beats per minute (bpm), after 8
weeks of endurance training heart rate should be significantly lower after
running 800 meters at the same pace. The magnitude of the reduction in
exercise heart rate will be difficult to determine because it will vary from
person to person.
Recovery heart rate will also improve because of endurance training. Using
the previous example, if it took 3 minutes for the heart rate to drop from 175
bpm to 125 bpm after running 800 meters on the first day of training, after
8 weeks of endurance training heart rate should drop to the same rate in
much less than 3 minutes. Again, the improvement in recovery heart rate
will vary from person to person. Despite this individual variability, after a
minimum of 8 weeks of endurance training, improvements should occur in
heart rate at rest (lower), during exercise at the same workload (lower), and
during recovery following a hard effort (less time to recover).

Increased Blood Volume and Cardiac Output


Endurance training increases the level of specific hormones that regulate the
body’s volume of blood. These hormones act to increase the fluid portion of
the blood, called plasma. The overall effect of this hormonal response is an
increase in total blood volume. An increase in total blood volume, along with
the fact that the heart is stronger and more powerful, means that the heart can
pump more blood through the body in a single minute, an amount referred
to as cardiac output.
An increase in cardiac output is important because more blood is delivered
to the brain, liver, kidneys, and other important organs. During endurance
exercise, increased cardiac output is important because more blood is deliv-
ered to the working skeletal muscles. As a result, more oxygen is delivered
to the exercising muscles for energy, and carbon dioxide and other metabolic
byproducts are removed from the muscles more rapidly.

Higher Maximal Oxygen Uptake


Endurance training improves the capacity of the lungs during exercise. Both
respiratory rate (breaths per minute) and tidal volume (liters of air per breath)
improve. These improvements . in lung capacity contribute to an increase in
maximal oxygen uptake (VO2max).
Maximal oxygen uptake is defined as the highest volume of oxygen that
an athlete’s
. body is capable of taking in and using for aerobic energy produc-
tion. VO2max can be expressed in absolute units (liters of oxygen per minute
[L . min−1]) or relative units (milliliters
. of oxygen per kilogram of body mass
per minute [ml . kg−1 . min−1]). VO2max is usually expressed in milliliters of
oxygen per kilogram body weight . per minute (ml kg min ) to allow. com-
. −1 . −1

parisons between individuals. VO2max can rise to levels of 65 to 75 ml kg−1 .


min−1 and 75 to 85 ml . kg−1 . min−1 in well-trained female and male endurance
14 } Wilber

athletes, respectively. By comparison, typical values for untrained females and


males may range from 35 to 40 ml . kg−1 . min−1 and 45 to 50 ml . kg−1 . min−1,
respectively. .
An improvement in VO2max is important because it means that more
oxygen is available to the.exercising muscles for energy production. Research
has shown that a high VO2max is one of several physiological factors that
contribute to success in triathlon.

Higher Lactate Threshold and Maximal Exercise Performance


A couple of additional physiological factors recently identified by scientific
research to be important contributors to endurance performance are lactate
threshold (LT) and maximal exercise performance. These physiological
parameters are typically measured under laboratory conditions. The lactate
threshold represents the point during an increasingly demanding endurance
training session or race that requires a greater contribution from the glycolytic
energy system and a lesser contribution from the oxidative phosphorylation
energy system. As a result, lactate production exceeds lactate removal, and
blood lactate levels increase exponentially.
In evaluating LT capabilities in triathletes, swimming velocity in meters
per second (m . sec−1), cycling power output (watts [W], watts per kilogram
of body weight [W . kg−1]) and running velocity (m . min−1) are the measure-
ments of interest. The higher the LT is, the better endurance performance
is. A triathlete can expect to see significant improvement in these LT para-
meters by executing a well-designed endurance training program, both over
the course of a single season and from season to season, depending on how
many years he has been in training.
Maximal exercise performance is simply the objective quantification of an
endurance athlete’s athletic capability at the point of volitional exhaustion that
occurs at the conclusion of a laboratory-based maximal exercise test, such as
a treadmill test. In evaluating maximal exercise performance in triathletes,
the same physiological measurements are of interest, although now under
maximal-effort (versus LT-effort) conditions: maximal swimming velocity
(m . sec−1), maximal cycling power output (W . kg−1), and maximal running
velocity (m . min−1).
As with the LT, the higher the maximal exercise performance is, the better
endurance performance is. An athlete can expect to see the same type of
improvement in maximal exercise performance as seen in the LT within a
single season and from season to season by executing a well-designed endur-
ance training program.

Improved Economy
Another physiological factor contributing to endurance performance is
economy. The concept of physiological economy is the same as fuel efficiency
or economy in an automobile. A more economical or efficient car uses less
Physiology and the Multisport Athlete | 15

fuel at a specific speed and achieves greater miles per gallon than a less
economical car.
The same is true for endurance
. athletes. For example, athlete AA and
athlete BB have a similar VO2max value of 65 ml . kg−1 . min−1. Athlete AA,
however, uses 50 ml . kg−1 . min−1 while running at a 5:00 per mile (3:06 per
km) pace in the first half of a 10K race, whereas athlete BB uses 53 ml . kg−1 .
min−1 while running at the same 5:00 per mile pace in the first half of a 10K
race. Thus, athlete AA is more efficient and economical in terms of energy
expenditure than athlete BB because he uses less oxygen at the same pace.
Logically, then, athlete AA will have a competitive advantage over athlete
BB over the second half of the race because of better physiological economy.
Physiological economy can be affected by several factors including use
of a well-designed endurance training program, as well as running bio-
mechanics, uphill running, training in heat and humidity, bungee running,
and plyometric training.

Additional Training Effects


The combined effect of the training-induced improvements in cardiac output,
maximal oxygen uptake, lactate threshold, economy, and maximal exercise
performance clearly have a positive effect on endurance performance.
In addition, the ability to work and exercise in heat and humidity improves
significantly through endurance training. As mentioned earlier in this section,
the body produces more plasma volume and total blood volume because of
endurance training. Total blood volume acts like a radiator coolant found in a
car or truck. By having more plasma and greater blood volume, an athlete is
able to produce more sweat and dissipate heat more effectively from the body,
particularly when exercising in a hot and humid environment. This benefit
is particularly helpful for triathletes who compete in tropical environments.
Endurance training can lower total body weight and reduce body fat. This
result is probably not a major concern to well-trained triathletes, who are
typically lean, but it may become more important as athletes get older and
have less time to train than they did earlier in their careers.
For people with moderate to high blood pressure, regular endurance
training can have a significant lowering effect, thereby decreasing the risk
of cardiovascular disease and premature death. Similar to total body weight
and body fat, high blood pressure is probably not a major concern to most
well-trained triathletes, whose blood pressure is typically normal. But elevated
blood pressure may become an issue as an athlete gets older and potentially
less active.

Skeletal Muscle Contraction


The physiological process of skeletal muscle contraction is operating during
daily activities, even when not exercising. Walking up stairs, lifting a book,
16 } Wilber

even reading this sentence involves skeletal muscle contraction. Of course,


the process of skeletal muscle contraction is extremely active during exer-
cise, and it is an important training consideration for people who compete in
triathlon. This section describes the process of skeletal muscle contraction,
focusing on the unique anatomical structure of skeletal muscle fiber and the
fascinating step-by-step process of muscular contraction.

Skeletal Muscle Anatomy


The anatomical structure of skeletal muscle is similar to a suspension-bridge
cable. A bridge cable has several internal smaller-diameter cables wrapped
in an overlapping configuration that significantly enhances the strength and
stability of the mother cable. The anatomical structure of skeletal muscle is
similar in that deeper into the muscle, the tightly bundled muscle fibers are
progressively smaller in diameter. They are reinforced by various connective
and overlapping anatomical structures that provide additional support.
Figure 1.9 shows this suspension-bridge cable characteristic of a skeletal
muscle. In viewing the figure, notice how the muscle fibers become progres-
sively smaller in diameter. Also, notice the connective and supportive tubelike
structures that surround each sequential layer of skeletal muscle. The main
structure in skeletal muscle fiber is the sarcomere. The sarcomere is impor-
tant because it is the basic unit of all skeletal muscle contraction. Figure 1.10
shows a detailed version of the sarcomere and surrounding structures. Note
the structures that surround the sarcomere, described in the next section:
T-tubule, tubules of the sarcoplasmic reticulum, and terminal cisternae of the
sacroplasmic reticulum. Notice also the mitochondria, where ATP is produced
by oxidative phosphorylation.
Finally, notice the two most important structures in skeletal muscle contrac-
tion, actin and myosin, as illustrated in figure 1.11. The anatomical structure of
myosin is shown in the inset. Notice that myosin is made up of a tail segment
and several large heads. An important feature of myosin is that the heads
have the ability to move, for coupling and contraction, as detailed in the next
section. The anatomical structure of actin is shown in the inset. Actin is the
double strand of egg-shaped structures lined up in an end-to-end configura-
tion. Notice also the thinner protein strand tropomyosin, which overlaps the
outer surface of actin, and the troponin complex, which is attached to and
positioned at regular intervals on tropomyosin.

Contraction of Skeletal Muscle


A sequence of neural, biochemical, and physiological events allows skeletal
muscle contraction to take place. The process of skeletal muscle contraction
occurs in three phases: excitation, coupling, and contraction. Each of these
phases involves several steps.
Bone
Tendon
Muscle
Epimysium
Endomysium
Perimysium
Fasciculus
Muscle fibers
Muscle fiber

Myofibril

Epimysium
Perimysium
Nuclei
Blood vessel Muscle fiber
in middle of
Endomysium a fascicle

„ Figure 1.9 Structure of skeletal muscle.


E5643/Friel/Fig 1.9/448513/alw/pulled-R1

Transverse Sarcoplasmic Opening Nucleus


tubules reticulum into
(T-tubules) T-tubule

Mitochondria

Myofibril

Sarcoplasm
Plasmalemma
Terminal cisternae

„ Figure 1.10 Skeletal muscle sarcomere and surrounding structures including the
T-tubules, tubules of the sarcoplasmic reticulum,
E5643/Friel/Fig and the terminal cisternae of the
1.10/448514/alw/pulled-R3-kh
sarcoplasmic reticulum.

17
18 } Wilber

Myosin
Actin

Sarcomere

Z-disk M-line
H-zone A-band
I-band

Nebulin

Actin
Tropomyosin Thin filament:
Troponin Actin, troponin, tropomyosin
Thick filament:
Myosin Z-disk
M-line

Titin
Myosin Actin
Myosin
head

„ Figure 1.11 Contractile filaments, actin and myosin, shown individually and in
relation to one another.

The excitation phase of skeletal muscle contraction is shown in figure 1.12.


E5643/Friel/Fig 1.11/448515/alw/pulled-R2-kh
Excitation refers to the neural impulse that serves to spark the sequence of
biochemical and physiological steps that result in skeletal muscle contraction.
The excitation phase includes three key steps:
Physiology and the Multisport Athlete | 19

1. Motor nerves imbedded in the muscle fire off electrical impulses called
action potentials. These action potentials move through the muscle fiber
like electricity traveling through a power line (figure 1.12a).
2. The action potential moves along the sarcolemma and down the T-tubules
of the sarcoplasmic reticulum (figure 1.12b).
3. The action potential triggers the release of calcium (Ca2+) from the ter-
minal cisternae of the sarcoplasmic reticulum (figure 1.12b).
The coupling phase of skeletal muscle contraction is shown in figure 1.12c.
Coupling refers to the interconnection of the contractile filaments, actin and
myosin. The key steps in the coupling phase of skeletal muscle contraction
are as follows:
1. Ca2+ binds to the troponin complex.
2. The troponin complex changes its shape and configuration, thereby
allowing tropomyosin to recede into the space between the actin strands.
3. As tropomyosin recedes from the outer surface of actin, it no longer blocks
the outer surface of actin from interfacing with myosin.
4. The binding sites on actin are now fully exposed. The myosin heads
quickly attach, or couple, to actin at the binding sites.

Terminal cisternae Sarcoplasmic reticulum


Motor neuron Plasmalemma

Axon terminal T-tubule


Storage vesicle

Synaptic cleft

ACh
receptor
ACh
Ca2+
Muscle fiber

a
Tropomyosin b
Troponin Ca2+
Actin Active site
Myosin head
bound to active site

„„Figure 1.12 Excitation phase and coupling phase of skeletal muscle contraction.

E5643/Friel/Fig. 1.12/448516/Pulled/R2-kh
20 } Wilber

The contraction phase of skeletal muscle contraction is shown in figure 1.13.


Contraction refers to the sequence of events whereby myosin essentially pulls
on actin, thereby drawing the two contractile filaments closer together, result-
ing in muscular contraction. This phase is typically referred to as the sliding
filament theory. Note that the contraction phase of skeletal muscle contraction
cannot occur unless ATP is present. The key steps in the contraction phase
of skeletal muscle contraction are as follows:
1. Myosin and actin are attached. This process is called myosin–actin cross
bridging.
2. ADP and Pi are released from the myosin head. When ADP and Pi are
released, the myosin head pivots and pulls the actin filament toward the
middle of the sarcomere. This step is called the power stroke.
3. A new ATP molecule attaches to the myosin head. When this occurs, the
myosin head briefly releases from its attachment to actin.
4. The new ATP molecule is quickly converted to ADP and Pi and energy.
When this occurs, the myosin head quickly reattaches to actin.
5. Steps 1 through 4 are repeated continuously, provided that ATP is sup-
plied at the appropriate rate. (If ATP is not present and myosin–actin
cross bridging occurs, then the muscle is in a state of rigor mortis.)
A good analogy for the contraction phase is a tug-of-war contest. In tug-
of-war, the hands of several people are constantly grabbing, pulling, quickly
releasing, and regrabbing a rope. Similarly, in the contraction phase several
myosin heads are constantly grabbing (step 1 above), pulling (step 2), quickly
releasing (step 3), and regrabbing (step 4) on an actin “rope.” The contrac-
tion phase cannot continue indefinitely. Skeletal muscle will return to the
relaxed, noncontractile state when the motor nerve action potential ceases,
which in turn shuts off the release of Ca2+ from the terminal cisternae of the
sarcoplasmic reticulum. Without Ca2+ present, tropomyosin and the troponin
complex resume their noncontractile positions, where they serve to block
myosin from attaching to actin.

Skeletal Muscle Fiber Types


Traditional exercise physiology textbooks typically identify three basic types
of human skeletal muscle fiber. Recent animal and human research, however,
has provided compelling evidence of several additional pure and hybrid
skeletal muscle fiber types.1, 2, 3 Nevertheless, for the purpose of this chapter
we focus on the three main skeletal muscle fiber types:
Type I: slow oxidative (SO); for example, the soleus muscle
Type IIa: fast oxidative glycolytic (FOG); for example, the diaphragm
Type IIx: fast glycolytic (FG); for example, the gastrocnemius muscle
Physiology and the Multisport Athlete | 21

Myosin filament
45o
Myosin ATP
binding
1 Tight binding in the rigor sites
binding
state. The cross-bridge is site 2 ATP binds to its binding site
2 3 4 on the myosin. Myosin then
at a 45° angle relative to 1
the filaments. dissociates from actin.
G-actin molecule

ADP
ATP

1 2 3 4 1 2 3 4
5

6 At the end of the 3 The ATPase


power stroke, the activity of myosin
myosin head releases hydrolyzes the ATP.
ADP and resumes the ADP and Pi remain
tightly bound rigor bound to myosin.
state.

Pi ADP
Pi

1 2 3 4
5 1 2 3 4

Actin filament moves


toward M-line.

90o
5 Release of Pi initiates the Pi 4 The myosin head swings
power stroke. The myosin over and binds weakly to
head rotates on its hinge, 3
a new actin molecule.
1 2 4
pushing the actin filament The cross-bridge is now
past it. at 90° relative to the
filaments.

„„Figure 1.13 Contraction phase of skeletal muscle contraction.


SILVERTHORN, DEE UNGLAUB, HUMAN PHYSIOLOGY: AN INTEGRATED APPROACH, 4th, © 2007. Printed and
electronically reproduced by permission of Pearson Education, Inc., Upper Saddle River, New Jersey.

Note that most muscles have a certain percentage of type I, type IIa, and
E5643/Friel/Fig 1.14/448518/alw/pulled-R1
type IIx fibers. In other words, the soleus muscle is not made up exclusively
of type I fibers. Rather, it is composed predominantly of type I fibers but
Note: Image sizw 2/3-2
includes lower percentages of type IIa and type IIx fibers. Similarly, the dia-
to fit image
phragm muscle is predominantly composed of type IIa fibers, whereas the
gastrocnemius muscle is primarily made up of type IIx fibers.
Type I and type IIx fibers have distinct metabolic features. Type I fibers
are designed to facilitate oxidative phosphorylation energy production and
are found in relative abundance in endurance athletes. In contrast, type IIx
fibers are designed to facilitate glycolytic energy production and are found
22 } Wilber

in relative abundance in sprint and power athletes. Type IIa fibers are essen-
tially a hybrid of type I and type IIx fibers and therefore have the capability
of producing ATP by oxidative phosphorylation or glycolytic metabolism.

Training Effects
In describing the effect of training on skeletal muscle, a useful first step is
to categorize the specific types of training that alter the physiological and
biochemical characteristics of skeletal muscle. Skeletal muscle can be affected
by aerobic endurance training, anaerobic training, and resistance training.

Aerobic Endurance Training


Endurance training stresses and challenges type I (slow oxidative) muscle
fibers more than it does type IIx (fast glycolytic) muscle fibers. As a result,
type I fibers tend to enlarge with endurance training. Although the percentage
of type I and type IIx fibers does not appear to change, endurance training
may cause type IIx fibers to take on more type IIa (fast oxidative glycolytic)
fiber characteristics.
The number of capillaries supplying each muscle fiber increases with
endurance training. The capillary bed is a microscopic, meshlike structure
that is embedded deep in the muscle. The capillaries serve as a transfer
point whereby oxygen and nutrients (glucose) are delivered to the exercising
muscles by arterial blood, whereas carbon dioxide and metabolic by-products
(lactate + H+) are removed by venous blood. This delivery-and-pickup process
is enhanced if the number of capillaries is increased, thereby allowing the
exercising muscles to perform more efficiently.
Endurance training increases both the number and size of the mitochondria
in skeletal muscle, particularly for type I muscle fibers. The mitochondria are
microscopic, capsule-shaped units located in the muscle cell that are essential
for the production of ATP by the oxidative phosphorylation energy system
(figure 1.6). By increasing both the size and number of mitochondria through
endurance training, oxidative energy production is enhanced.
The activity of many oxidative enzymes is increased with endurance train-
ing. Figure 1.6 is a simplified representation of the oxidative energy system.
It shows how two molecules of acetyl CoA enter a mitochondrion, move into
the Krebs cycle, and then move on to the electron transport system (ETS) to
produce 32 molecules of ATP. The Krebs cycle is a series of biochemical reac-
tions that are essential to the production of the 32 molecules of ATP, which
are ultimately synthesized in the ETS. Most of the oxidative enzymes that
are enhanced by endurance training are located in the Krebs cycle phase of
the oxidative phosphorylation energy system. Similar to the increase in mito-
chondria, an increase in the activity of oxidative enzymes serves to enhance
oxidative energy production.
Finally, endurance training increases muscle myoglobin (Mb) content by
75 to 80 percent. Myoglobin is the “smaller brother” of hemoglobin and has
Physiology and the Multisport Athlete | 23

many similar structural characteristics. Like hemoglobin, myoglobin has as its


primary physiological function the transport of oxygen. Whereas hemoglobin
carries oxygen from the lungs to the exercising muscles by the bloodstream,
myoglobin picks up oxygen after it has been dropped off in the capillary bed
by hemoglobin. Next, myoglobin transports oxygen to the mitochondria where
it is used to produce ATP by the oxidative phosphorylation energy system. By
increasing myoglobin content through endurance training, oxygen delivery
within the exercising muscle is enhanced.

Anaerobic Training
Anaerobic training increases ATP–CP and glycolytic enzymes in skeletal
muscle. Some of these enzymes are shown in figure 1.4 (ATP–CP) and figure
1.5 (glycolytic). Similar to the increase in the oxidative enzymes, increases in
the ATP–CP and glycolytic enzymes serve to enhance the production of ATP
by those two energy systems.
Skeletal muscle buffering capacity is enhanced by anaerobic training. As
described earlier, lactate and H+ are metabolites that are produced in the gly-
colytic energy system. High concentrations of H+ can slow the release of Ca2+
in the excitation phase of skeletal muscle contraction, thereby contributing to
premature fatigue of the muscle. Because of anaerobic training, the amount
of bicarbonate (HCO3−) in skeletal muscle is increased.
As shown in figure 1.14, bicarbonate acts as an effective buffer for reduc-
ing acidosis in the exercising muscle. Bicarbonate essentially picks up the
potentially detrimental H+ and subsequently removes it safely from the body
in the form of H2O and CO2.

C 3H 5O 8– + H+ + HCO3– H2CO3 H2O + CO2


Lactate Bicarbonate

„„Figure 1.14 Bicarbonate reduces acidosis in exercising muscle.

E5643/Friel/Fig. 1.15/448519/TimB/R2-kh

Resistance Training
Although triathletes use various types of resistance training programs (heavy
weight and low repetitions versus moderate weight and high repetitions), the
following physiological adaptations occur because of regular resistance train-
ing. Regular resistance training means a minimum of three to five training
sessions per week for at least 8 weeks.
An increase in the size of the skeletal muscle fiber is known as hypertrophy.
Most research studies have shown that regular resistance training in combi-
nation with an adequate diet will produce skeletal muscle hypertrophy. The
degree of skeletal muscle hypertrophy will vary depending on the specific
resistance training program used, as defined by weight, repetitions, and
number of training sessions per week.
24 } Wilber

Regular resistance training also increases the number of active muscle


motor units. A single muscle motor unit is made up of several muscle fibers
along with the nerve that innervates them and stimulates the muscles fibers
to contract in unison. After several weeks of resistance training, the skeletal
muscle produces additional active motor units.
Muscular strength increases because of resistance training. Muscular
strength is defined as the maximum force that is generated by a muscle or
muscle group. Muscular strength is usually measured using a one-repetition
maximum lift (1RM), or the maximum amount of weight that a person can lift
just once. An athlete who can bench press 300 pounds (136 kg) in a 1RM has
twice the muscular strength of an athlete who can bench press 150 pounds
(68 kg).
Resistance training also increases muscular power, which is not the same
as muscular strength. Muscular power is the explosive aspect of strength and
is the product of muscular strength and the speed of a specific movement.
For example, athlete AA and athlete BB can both bench press 150 pounds (68
kg) in a 1RM. Athlete AA, however, completes the lift in 1.0 second, whereas
athlete BB can complete the lift in 0.5 second. Although athlete BB has the
same muscular strength as athlete AA, he has twice the muscular power
because he can lift the same weight in half the time.
An important performance characteristic for triathletes is muscular endur-
ance, which is also enhanced by regular resistance training.4, 5, 6 Muscular
endurance refers to the capacity to sustain repeated muscular actions, such
as when running for an extended period. It also refers to the ability to sus-
tain fixed or static muscular actions for an extended period, such as when
attempting to pin an opponent in wrestling. Muscular endurance is usually
measured by counting the number of repetitions that an athlete can perform
at a fixed percentage of her 1RM. For example, if an athlete bench presses
140 pounds (64 kg) in a 1RM, her muscular endurance can be measured by
counting how many repetitions she can complete at 75 percent of 1RM, or
105 pounds (48 kg).

Conclusion
After reading this chapter, triathletes and their coaches should feel confident
in their understanding of the physiology of triathlon. Triathletes should apply
the scientific principles of the three energy systems (ATP–CP, glycolysis, oxi-
dative phosphorylation) to the design of daily workouts to meet the specific
goals of their training programs. The principles of the three energy systems
should be applied to race-day strategy for all triathlon formats—sprint,
Olympic, and Ironman.
An understanding of cardiopulmonary anatomy and physiology will help
triathletes and coaches apply these principles to create training programs that
are designed to enhance the four most important contributors to endurance
Physiology and the Multisport Athlete | 25
.
performance: VO2max, economy, lactate threshold, and maximal exercise
performance (swimming and running velocity, cycling power output).
Knowledge of skeletal muscle anatomy and physiology will optimize the
triathlete’s training capacity while simultaneously reducing the chance of
injury or overtraining. Applying this knowledge will maximize positive
training effects on skeletal muscle that are derived from aerobic training (e.g.,
enhancement of oxidative enzymes) and anaerobic training (e.g., enhanced
muscle buffering capacity), which ultimately serve to improve race perfor-
mance.
This page intentionally left blank.
CHAPTER
2
Genetics and Inheritance
in Triathlon Performance
—Ildus I. Ahmetov, PhD, and Malcolm Collins, PhD

A thletic performance and the occurrence of sports-related injuries are


both multifactorial conditions, which are determined by the complex and
poorly understood interactions of both environmental and genetic factors.
Although much work has been done to identify the nongenetic components
associated with performance and susceptibility to injuries, an ever-growing
body of research is investigating the genetic contribution to these observable
characteristics or traits, known in the field of genetics as phenotypes.
It has long been recognized that the interindividual variability of physical
performance traits and the ability to become an elite athlete have a strong
genetic basis. The genetic factors that influence these performance and
sports-related injury phenotypes are now being sought. Several family, twin,
case-control, and cross-sectional studies suggest an important role of genetics
along with epigenetic (i.e., stable and heritable changes in gene expression)
and environmental factors in the determination of individual differences in
athletic performance and training responses.
Research has shown that a genetic component of the variance in any phe-
notype (i.e., height, muscle mass, athlete status, and so on) is determined by
small changes in the structure of DNA (composed of about 3 billion genetic
letters or nucleotides and about 25,000 genes), which are called polymorphisms.
In other words, any position in the genome where more than one nucleotide
base is found among a population of individuals is known as a polymor-
phism, and each variant letter found at a polymorphism is known as an allele.
In most cases, two different alleles are found at any polymorphism in the
genome (e.g., the nucleotides G and C are both possible at position –634 of
gene VEGFA; see figure 2.1). Across a population, the frequency of each allele
can be measured. Because most genes come in pairs (inherited from father
and mother), each person has a pair of alleles within his or her genome for
any polymorphism. This combination of alleles at any variant site is known
as a genotype (for example GG, GC, and CC genotypes).

27
28 } Ahmetov and Collins

GCAGCGAAAGCGACAGGGGCA VEGFA C allele

Position –634 of DNA sequence

GCAGCGAAAGGGACAGGGGCA VEGFA G allele

„„Figure 2.1 Scheme of possible genetic variations (alleles) at position


–634 of gene VEGFA. This example
E5643/Friel/Fig. shows that people may possess two
2.1/448520/TimB/R1
different types of alleles (G or C) of a given DNA sequence.

The human genome has no less than 50 million polymorphic variants,


which make all individuals different. The most common type of DNA
sequence variant is the single-nucleotide polymorphism, or SNP. Other types
of DNA polymorphisms exist as well. Insertion/deletion (I/D) polymor-
phisms are the presence or absence of a stretch of specific DNA nucleotides
at a certain position in the genome (e.g., I/D polymorphism of the ACE gene).
Another type of genetic variation is known as a repeat polymorphism. These
polymorphisms are repetitive stretches of DNA sequences (e.g., CAG repeats
in the AR gene).
Genetic variations can affect the amount and structure of mRNA/protein
and therefore may account for the main share of genetic factors in human
phenotypic variability, but most of these polymorphisms remain to be discov-
ered. If any allele is associated with a certain phenotype, it is called a DNA
(or genetic) marker. Based on final effect, DNA variations generally can be
classified as genetic markers associated with endurance athlete status, height,
increased proportion of slow-twitch muscle fibers, risk of injury, and so on.
These and other genetic markers are considered in the sections that follow.

DNA Effect on Height, Body Composition,


and Muscle Fiber Composition
Physical characteristics and body composition are heritable traits and are
known to be fundamental to excellence in athletic performance. Specific
athletic events require different body type, height, weight, and muscle fiber
composition for maximal performance. Mean height, body mass index, and
percentage of slow-twitch muscle fibers of elite triathletes reflect the preferred
anthropometric and morphologic measures for this type of sport.

Height
The mean body height of elite male and female triathletes typically is around
71 inches (180 cm) and 67 inches (170 cm), respectively (e.g., 2012 Olympic
champions Alistair Brownlee and Nicola Spirig are 176 cm and 166 cm tall,
respectively.) The consensus within the scientific literature is that height is
Genetics and Inheritance in Triathlon Performance | 29

highly heritable; 80 percent of its variance is determined by small contribu-


tions of multiple genes.1 Of note, heritability refers to the proportion of the
phenotypic variance attributable to all genetic effects. Initially, 47 common
genetic variants, which explain 5 percent of the variance in height within
Caucasians, were identified during a meta-analysis of genomewide associa-
tion studies (GWAS).1
Recently Yang and colleagues2 estimated that approximately 295,000
common SNPs explained 45 percent of the variance and that the remaining 35
percent of the genetic variance could be explained by genetic linkage between
the causal variants and the genotyped SNPs. In summary, besides the effects
of nutrition, which determines most of the 20 percent of the environmental
variance,1 height is determined by small contributions of sequence variants
within thousands of genes.

Body Composition
Body composition is most often viewed in the context of the two-compartment
model: Body mass is a combination of fat-free mass (FFM; includes muscle
mass, bone mass, and other lean components) and fat mass. Total body fat is
defined as the absolute amount of energy stored in the form of triacylglycerol
in the body. Total body fat content is generally estimated using the body mass
index (BMI = kg/m2). The mean BMI of elite male and female triathletes is
around 23.0 and 20.5 kg/m2, respectively.
Any estimate of genetic effects on BMI is influenced in unknown propor-
tions by the contribution of the genotype to fat mass, muscle mass, skeletal
mass, and other components. Heritability estimates range from 44 to 90 per-
cent for BMI. An excessive amount of fat is associated with greater risk of a
variety of morbid conditions. Low body fat content is associated with better
performance in most endurance activities. Researchers have found that the
athletes
. with lower body fat percentage had higher maximum oxygen uptake
( VO2max). In other words, the athletes with lower body fat percentage seemed
to use oxygen most efficiently, and the excess of body fat was reported to be a
deterrent to physical performance. Furthermore, percent body fat is negatively
correlated with total race time in male Ironman triathletes.3
On the other hand, muscle mass (with heritability from 52 to 90 percent)
is considered a favorable factor for endurance performance. Skeletal muscle
is the largest tissue in the body and the main energy-consuming and work-
producing tissue, providing the propulsive force to perform physical activities.
Support for a role of genes in variation in body fat and muscle content has
been obtained by several approaches including association studies. To date,
more than 50 gene polymorphisms associated with BMI have been reported.
These polymorphisms are located in genes mainly involved in appetite con-
trol, fat absorption, lipid storage, and fatty acid oxidation.4
The genetics of fat-free mass or muscle mass is less studied. The genetic
markers associated with increased FFM or muscle mass are located within
30 } Ahmetov and Collins

genes involved in regulation of vascular tone, muscle contraction, hormone


metabolism, and cell signaling.5

Muscle Fiber Composition


The ability to perform aerobic or anaerobic exercise varies widely among
people, partially depending on muscle fiber composition. In untrained people,
the proportion of slow-twitch (type I) fibers in the vastus lateralis muscle is
typically around 50 percent (range 5 to 90 percent), and it is unusual for them
to undergo conversion to fast-twitch fibers. Endurance-oriented athletes are
reported to have a remarkably high proportion of type I fibers in their trained
muscle groups, whereas sprinters and weightlifters have muscles that consist
predominantly of type IIa and IIx fibers.
Indeed, triathletes have a high percentage of type I fibers in their gastroc-
nemius, vastus lateralis, and posterior deltoid muscles.6 It has been suggested
that the genetic component for the observed variability in the proportion of
type I fibers in human muscles is of the order of 40 to 50 percent, indicating
that muscle fiber type composition is determined by both genotype and envi-
ronment.7 The genetic variance is that portion of the individual differences
associated with differences in DNA sequence at relevant genes and other
DNA regions. It incorporates the effects of single genes, gene–environment
interactions, and a variety of gene–gene interactions.
Environmental variance depends on factors such as nutritional habits, level
of habitual physical activity, nonheritable intrauterine influences, and a variety
of other lifestyle components and factors from the social and physical envi-
ronment. Such environmental factors modulate muscle phenotype through
epigenetic mechanisms, activation of transcription factors, and activation of
transcriptional coactivators.
Environmental and genetic factors associated with muscle fiber composi-
tion are shown in table 2.1. To date, reports have shown that five gene polymor-
phisms are associated with muscle fiber composition. These polymorphisms
are located in genes involved in biochemical processes in mitochondria, glu-
cose and lipid metabolism, cytoskeletal function, hypoxia-related growth of

Table 2.1 Factors Associated With Muscle Fiber Composition


Factors that induce increased Factors that induce increased
Factors proportion of ST fibers proportion of FT fibers
Environmental Tonic activity; reduced thyroid hormone Phasic pattern of activity;
(intrinsic or level (hypothyroidism); high-intensity increased thyroid hormone level
extrinsic) endurance training (hyperthyroidism); resistance training;
spaceflight and unloading; spinal cord
injuries
Genetic Gene variants associated with high Gene variants associated with high
percentage of ST fibers (e.g., ACE I, percentage of FT fibers (e.g., ACE D,
ACTN3 577X, HIF1A Pro582, PPARA ACTN3 R577, HIF1A 582Ser, PPARA
rs4253778 G, VEGFR2472Gln) rs4253778 C, VEGFR2 His472)
Genetics and Inheritance in Triathlon Performance | 31

new blood vessels, and circulatory homeostasis. Interestingly, most of these


gene variants are associated with physical performance, athlete status, and
various metabolic and cardiovascular diseases.8, 9

DNA Effect on Endurance Training


The effects from aerobic exercise differ greatly among people, depending
on lifestyle factors and genetic backgrounds. An understanding of genetic
backgrounds will help to clarify criteria of daily physical activities and
appropriate exercise for individuals including athletes, making it possible
to apply individualized preventive medicine and medical care. Note that at
the molecular and cellular levels, people with the same genotype respond
more similarly to training than do those with different genotypes, indicating
that genes play an important role in determining individual differences in
response to training.5
Research on aerobic endurance clearly shows that some people respond
more to training. than others.10 In the same study the maximal heritability
estimate of the VO2max response to training adjusted for age and sex was
reported to be 47 percent. This result means that genetically gifted athletes
have a much greater response to training.
For example, evidence that many of the world’s best endurance runners
originate from distinct regions of Ethiopia and Kenya, rather than being
evenly distributed throughout their respective countries, appears to sustain
the idea that the success of East African runners is genetically determined.
Studies have shown that African distance runners have reduced lactic acid
accumulation in muscles, increased resistance to fatigue, and increased
oxidative enzyme activity, which equates with high levels of aerobic energy
production.
At least 65 genetic markers have been shown to be associated with endur-
ance phenotypes in response to training.5, 8, 11, 12 These genes are primarily
involved in ATP, glucose and lipid metabolism, hypoxia, mitochondrial
processes, regulation of muscle fiber type composition, oxygen transport,
vascular tone, and the development of new blood vessels.
Most recently, Bouchard and colleagues12 have identified a panel
. of 21 gene
polymorphisms that account for 49 percent of the variance in VO2max train-
ability. The study included 473 sedentary adults who followed a standardized
20-week exercise program. Subjects who carried 9 .or fewer favorable alleles
at these 21 gene polymorphisms improved their VO2max by 221 milliliters
per minute (ml/min), whereas those who carried 19 or more of these alleles
gained, on average, 604 ml/min.

Genes and Triathlete Status


Athlete status is a heritable trait: Around 66 percent of the variance in ath-
lete status is explained by additive genetic factors. The remaining variance
32 } Ahmetov and Collins

is because of nonshared environmental factors. Fifteen genetic markers


have been reported to be associated with triathlete status (table 2.2). For
instance, one elite Russian triathlete possesses the following combination
of seven favorable genotypes: PPARA GG, PPARD CC, PPARGC1AGly/Gly,
PPARGC1BAla/Pro, UCP2Ala/Val, ACTN3 RX, and ACE ID (carrier of 10
favorable alleles).
All three triathlon disciplines are endurance events. Therefore, it is not
surprising that investigators have shown that genetic variants associated with
endurance performance are associated with performance during the Iron-
man triathlon.13, 14, 15, 16 Some common sequence variants are associated with
the performance of Ironman triathletes.13, 14 Many of these sequence variants
are associated with improvements in biochemical pathways within skeletal
muscle that improve endurance performance.17
A certain genotype is associated with reduced sprint and power perfor-
mance in several studies.18 It was not associated with ultraendurance perfor-
mance in the South African Ironman triathlons study,19 but the frequency
of this genotype was significantly lower in Russian Olympic triathletes in
comparison with controls.20 In the same group of Russian triathletes the fre-
quencies of specific variants located in genes involved in lipid and glucose
metabolism, fatty acid oxidation, muscle fiber regulation, mitochondrial
processes, muscle fiber regulation, and thermoregulation were significantly
higher than in nonathletes.8, 21, 22
Different sets of sequence variants have generally been associated with
endurance performance and sprint or power events in a sport-independent
manner. Discipline-specific associations, however, have recently been reported

Table 2.2 Gene Variants Associated With Triathlete Status


Favorable
Gene Polymorphism marker Cohort
ACE Alu I/D I Caucasian Ironman triathletes
ACTN3 R577X R Russian Olympic-distance triathletes
BDKRB2 +9/−9 −9 Caucasian Ironman triathletes
COL5A1 rs12722 C/T rs12722 T Caucasian Ironman triathletes
COL6A1 rs35796750 C/T rs35796750 T Caucasian Ironman triathletes
EPAS1 rs1867785A/G rs1867785G Australian Ironman and Olympic-distance
rs11689011C/T rs11689011T triathletes
NOS3 Glu298Asp Glu298 Caucasian Ironman triathletes
PPARA rs4253778G/C rs4253778 G Russian Olympic-distance triathletes
PPARD rs2016520 T/C rs2016520C Russian Olympic-distance triathletes
PPARGC1A Gly482Ser Gly482 Russian and Polish Olympic-distance triathletes
PPARGC1B Ala203Pro 203Pro Russian Olympic-distance triathletes
TFAM Ser12Thr 12Thr Russian Olympic-distance triathletes
UCP2 Ala55Val 55Val Russian Olympic-distance triathletes
UCP3 rs1800849 C/T rs1800849 T Russian Olympic-distance triathletes
Genetics and Inheritance in Triathlon Performance | 33

with the cycling and running legs of the Ironman triathlon.15, 16 Variants
within the genes that produce the collagens that form a fibrous protein that
connects and supports body tissues are associated with the cycling and run-
ning times in an Ironman triathlon, respectively.15, 16 Types V and VI collagen
are important structural proteins within the connective tissue components
of the musculoskeletal system. These results, however, need to be confirmed
in other similar subjects.
Adding to the complexity, multiple genes that remain to be investigated,
each encoding for proteins involved in different biological systems and having
a small influence on the systems, are likely involved in determining the innate
performance ability of a triathlete.23 Therefore, the limited number of stud-
ies (South Africa Ironman, Russian Olympic distance, Australian Ironman
and Olympic distance) investigating the genetic contribution to triathlon
performance and the extremely small number of sequence variants analyzed
to date should not be overinterpreted by athletes and coaches. More study
is needed before affective assessment of genes for performance can be done.

Genetic Contribution to Injuries


Triathletes are at increased risk for muscle, tendon, ligament, and other inju-
ries as a result of training and competition.24 Thirty-nine percent of the 433
triathletes who completed a general medical history questionnaire during the
2006 and 2007 South African Ironman Triathlons reported suffering from a
painful, swollen, or stiff tendons or ligaments during their triathlon careers
(unpublished data). Acute and chronic or overuse musculoskeletal tissue
injuries can negatively affect an athlete’s ability to train and perform during
competition. In extreme cases these injuries could end an athlete’s competi-
tive career or prevent him or her from reaching full potential. Certain athletes
seem to be predisposed to specific injuries, whereas others, in spite of training
harder and longer, appear to be protected.25, 23
No single factor causes musculoskeletal tissue injuries; instead, typical
injury risk models include multiple intrinsic and extrinsic factors that are
associated with these injuries.23 A typical injury risk model for Achilles tendon
injuries is illustrated in figure 2.2.26 The individually associated factors do
not cause these injuries. Instead, they are believed to increase the risk for
musculoskeletal injuries to a lesser or greater extent.
The current level of scientific evidence used to include risk factors in the
risk models varies significantly. The inclusion of increased age and weight, for
example, as an intrinsic risk factor for Achilles tendon injuries is based on weak
scientific evidence, whereas strong evidence indicates that a previous injury
increases the risk for Achilles tendon injuries.26 The current injury risk models
are therefore unable to identify athletes who are at risk for Achilles tendon
injuries. The biological processes that result in these injuries are poorly under-
stood, and the traditional injury risk models have not been developed on the
underlying biological and biomechanical processes that cause these injuries.
34 } Ahmetov and Collins

Intrinsic risk factors Extrinsic risk factors


G + E Previous injury Physical activity
G + E Flexibility Medications
G + E Muscle strength Occupation
G + E Age Training errors
G + E Weight Shoes and equipment
Tendon temperature Environmental conditions
Blood supply Surface
Biomechanics Nutrition and smoking Injuring event
G Male sex
Predisposed Susceptible
Systemic disease Achilles
Achilles Achilles
G Blood group tendon
tendon tendon
injuries
injuries injuries

„„Figure 2.2 Relationship between intrinsic and extrinsic risk factors as the role of
the injuring event in the etiologyE5643/Friel/Fig.
of Achilles tendon injuries.23, 26 Many of the individual
2.2/448523/TimB/R2-kh
intrinsic risk factors are also in their own right multifactorial phenotypes determined by,
to a lesser or greater extent, both genetic (nature, G) and environmental (nurture, E)
factors. A high level of scientific evidence suggests that a previous Achilles tendon injury
(listed in black boxes) is an intrinsic risk factor for a subsequent injury. A moderate to low
level of evidence suggests the intrinsic (gray boxes) and extrinsic (white boxes) are risk
factors. Intrinsic risk factors merely predispose athletes to the injury. The predisposed
athlete becomes more susceptible to injury when exposed to the appropriate extrinsic
factors. A specific event results in a complete or partial rupture or the volume of tissue
damage becomes symptomatic. Blood supply refers to the blood supply in the tendon.

Most of the intrinsic factors included in the typical risk models are in their
own right multifactorial traits determined to varying degrees by both genetic
(nature) and environmental (nurture) factors (figure 2.2).23 Many are polygenic
traits, and each individual gene has a small effect on the phenotype. Added
together, they have a significant contribution.23 In spite of this, specific genetic
sequence variants within genes that contribute to the structural components
of tendons and ligaments have recently been identified as intrinsic risk factors
for some tendon and ligament injuries. Genetic variants previously associated
with musculoskeletal soft tissue injuries have been extensively reviewed and
are summarized in table 2.3.23, 25
Whether these genes are directly or indirectly associated with these
injuries remains to be determined. For example, one particular genotype
has been reported to protect athletes from developing chronic Achilles
tendinopathy.27 It was recently reported that this genotype also protected
healthy, physically active people against an age-related decline in range of
motion measurements. Because this gene is associated with both range of
motion measurements and chronic Achilles tendinopathy, it is not surpris-
ing that flexibility has been reported to be an intrinsic risk factor for Achilles
tendon injuries.27 The involved gene, which produces type V collagen, plays
a critical role in the regulation of type I collagen fibril assembly and lateral
growth in tendons, ligaments, and other connective tissues (fibrillogenesis).
Genetics and Inheritance in Triathlon Performance | 35

Table 2.3 Summary of the Genes Previously Associated With Musculoskeletal


Soft-Tissue Injuries
The genes encode for structural components of the matrix (collagens and glycopro-
teins), signaling molecules (cytokines and growth factors), and extracellular matrix
(ECM) proteinases (MMPs) have been identified as intrinsic risk factors for some
tendon and ligament injuries.

Matrix Signaling molecules ECM proteinases


Gene Injury Gene Injury Gene Injury
COL1A1 ACL ruptures GDF5 Achilles MMP3 Achilles
tendinopathy tendinopathy
COL5A1 Achilles tendinopathy IL-1βa MMP10b
Female ACL ruptures IL-1RNa Achilles MMP1b ACL ruptures
COL12A1 Female ACL ruptures IL-6a tendinopathy MMP3b
MMP12b
TNC Achilles tendinopathy

The data within this table is more extensively described in Tucker and Collins.23
COL1A1, α1 chain of type I collagen; COL5A1, α1 chain of type V collagen; COL12A1, α1 chain of
type XII collagen; TNC, tenascin C; GDF5, growth and differentiation factor 5; IL-1β, interleukin-1β;
IL-1RN, interleukin-1 receptor antagonist; IL-6, interleukin-6; MMP3, matrix metalloproteinase 3;
MMP10, matrix metalloproteinase 10; MMP1, matrix metalloproteinase 1; MMP12, matrix metallo-
proteinase 12; ACL, anterior cruciate ligament.
a
A pathway-based approach was used to investigate the association of genes within the inflamma-
tory pathway and Achilles tendinopathy.
b
These four MMP genes have been mapped to chromosome 11q22 and a haplotype within this region
associated with ACL rupture.

Collins and Posthumus27 have recently suggested that the type V collagen
connect in the Achilles tendon and other tissues may alter their mechanical
properties and susceptibility to injuries.
Note that none of the genetic risk factors or any of the other intrinsic risk
factors causes musculoskeletal injuries. They merely modulate or contribute
to the risk for these injuries. Predisposed athletes need to be exposed to
appropriate extrinsic factors, and an event that causes the rupture, tear, or
the volume of accumulated damage within the tissue becomes symptomatic.

Physiological Responses
During Participation in Triathlons
Besides being subject to individual variations in performance and suscepti-
bility to injury, not all triathletes respond identically to the stresses of train-
ing and participating in a triathlon. These physiological responses are also
complex phenotypes determined by both genes and environment. This notion
is best illustrated by the large individual variation in body weight changes
during participation in the Ironman triathlon and the practical recommenda-
tions on how much and how often triathletes should drink while competing.
36 } Ahmetov and Collins

Weight losses, mainly because of dehydration, of up to 12 percent have


been reported.28 Inappropriate replacement of this fluid loss during exercise
can lead to disregulation of sodium homeostasis, resulting in either hyper-
natremia, hyponatremia, or in rare cases potentially fatal encephalopathy
(EAHE).28 EAHE is caused by abnormal fluid retention in athletes who drink
to excess during prolonged exercise, usually lasting more than 4 hours, and
is preventable when athletes drink only to thirst during exercise. It has been
proposed that drinking behavior, water losses, and electrolyte homeostasis
during exercise are determined partially by genetics. An association of cer-
tain genes with weight loss during the South African Ironman triathlons
has previously been reported.29, 30 The products of these genes regulate water
and electrolyte homeostasis by controlling thirst and water reabsorption.29, 30

Practical Recommendations
Regarding Genetic Profiling
Large variations occur within the responses to training and participation in
a triathlon, performance, and susceptibility to sports-related injuries. This
variation is caused by complex interactions of environmental and other factors
with an athlete’s genetic makeup. Although these traits have an inheritable
component, athletes and coaches need to understand that no single genetic
test or panel of genetic tests can be used to determine performance, response
to training, or susceptibility to injuries.
Recognize as well that because all these phenotypes are multifactorial,
genetic testing can never be used to predict or diagnose any of these phe-
notypes. Predictive and diagnostic genetic testing is limited to the rare,
classical genetic disorders caused by a single genetic mutation. But genetic
testing may one day be incorporated into multifactorial models, consisting of
both genetic and nongenetic components, to determine risk for injuries and
response to training. These models would enable clinicians and coaches to
optimize training programs to reduce the likelihood of injury and maximize
training responses. The ethical issues associated with the development and
implementation of any model including genetic testing, however, need to be
considered.

Conclusion
Physical performance is a complex phenotype influenced by multiple envi-
ronmental and genetic factors, and variation in human physical performance
and athletic ability has long been recognized as having a strong heritable
component. The question is no longer whether there is a genetic component to
athletic potential and endurance and strength trainability, but exactly which
genes are involved and by which mechanisms and pathways they exert their
Genetics and Inheritance in Triathlon Performance | 37

effect. Our current progress toward answering these questions represents


only the first steps toward understanding the genetic factors that influence
human physical performance.
The current review provides evidence that hundreds of genetic mark-
ers are linked to morphological phenotypes, elite athlete status, training
responses, and risk of injuries. Of those, 15 gene polymorphisms are linked
with triathlete performance. Although more replication studies are needed,
the preliminary data suggest an opportunity to use some of these genetic
markers in an individually tailored prescription of lifestyle and exercise for
health and sport performance.
This page intentionally left blank.
CHAPTER
3
Gender and Age
Considerations
in Triathlon
—Romuald Lepers, PhD

B oth gender and age can affect performance in triathlon. During these last
decades, female triathletes have reduced the gap in triathlon with their
male counterparts, especially in long-distance triathlons such as Ironman
distance. Female triathletes such as Chrissie Wellington have led triathlon
performance to a higher level. In addition to the improvement in female
triathlete performance, an increase in participation of masters triathletes
(more than 40 years old) has been observed recently. Masters triathletes
can still achieve extremely high levels of performance: Dave Scott finished
second overall at the Hawaii Ironman Triathlon at 42 years old in 1996; more
recently, in 2012, Natascha Badmann finished 6th at Hawaii at 46 years old.
Moreover, some unique older triathletes (up to 80 years of age) have begun
to push the limits of the interaction between aging and human endurance.
In this context, we could wonder how training and outcome expectations
for the standard triathlon distances should be modified to the unique needs
required by gender and age.

Gender Differences
in Triathlon Performance
The gender difference in endurance performance has received considerable
attention in these last decades, but most studies have focused on running.
Previous studies that have investigated the participation and performance
trends of female endurance athletes in running reported an increase in
female participation and an improvement in performance during the last
three decades.1

39
40 } Lepers

Although some authors have suggested that the gap in gender difference
in endurance performance could be closed, more recent studies could not
confirm this assumption, showing that the gender difference in endurance
performance is no longer diminishing.2 For elite runners the gender differ-
ence in marathon running performance has remained about the same since
the 1980s at a difference of approximately 10 percent.3 Gender difference
in triathlon regarding participation and performances has been less in-
vestigated.

Physiological Considerations
Physiological and morphological characteristics may be responsible for gender
difference in triathlon performance. Little information was available on the
physiological determinants of endurance performance in women until the
1970s. The current available data, while limited in comparison
. to those avail-
able on men, suggest that maximal oxygen uptake (VO2max), lactate thresh-
old, and running economy interact in women as determinants of endurance
performance in a manner similar to that in men. .
The current explanation of gender differences in VO2max among elite
athletes when expressed relative to body mass is twofold:
• First, elite females have more body fat than males
. (about 13 percent versus
about 5 percent). Much of the difference in VO2max disappears when it
is expressed relative to lean body mass.
• Second, the hemoglobin concentration of elite athletes is 5 to 10 percent
lower in women than in men.
Concerning lactate threshold, there is no reason to believe that values will
be lower in women than in men because mitochondrial adaptations in the
skeletal muscles of highly trained male and female athletes appear similar.
Finally, the average oxygen cost to run a given speed (i.e., running economy)
by groups of elite male and female athletes appears to be similar and prob-
ably plays the same role in determining success in endurance performance.
Therefore, the major physiological reason to explain .the slower record perfor-
mances by women than men is probably the lower VO2max values observed
in
. women. Although a number of elite male and female athletes have similar
VO2max values, these values are at the low end of the elite range for men
versus the upper end of the elite range for women.
Male triathletes possess a larger muscle mass, correlating with greater mus-
cular strength and lower relative body fat compared with females. Low body
fat is an important predictor variable for total time performance in triathlon.
For example, Knechtle, Knechtle, and Rosemann4 showed that low body fat
was associated with faster race times in male Ironman triathletes but not in
females. Males retain on average 7 to 9 percent less percent body fat than
females, which is likely an advantage for males. Therefore, gender differences
in percentage body fat, oxygen carrying capacity, and muscle mass appear
Gender and Age Considerations in Triathlon | 41

to be responsible for gender difference in triathlon performance. Note also


that pregnancy and the menstruation cycle may affect training and racing.

Gender Difference in Participation


and Performance Density
The number of females competing in triathlon has increased progressively
since the 1980s. For example, the number of females finishing the Hawaii
Ironman Triathlon increased from 20 in 1981 (6 percent of the participants),
to more than 470 in 2010 (27 percent of the participants). By comparison,
women represented 32 percent of the participants at the New York Marathon
during the last 10 years. In Europe, a progressive rise in the number of female
Ironman finishers also occurred, but the rate was lower. For example in 2010
females accounted for 8, 11, and 13 percent of the field at the France, Austria,
and Switzerland Ironman triathlons, respectively. The lower rate of female
participation in Europe is presumably explained by the fact that Ironman
triathlon in Europe is younger compared with Hawaii Ironman Triathlon
(first event held in 1978). Female participation in European Ironman triathlons
will probably increase in the future.
For short-distance triathlon, the rate of female participation appears
greater than for Ironman triathlon. For example, in 2010 at the Zurich
(Switzerland) short-distance triathlon, females accounted for 26 percent of
the field, but females made up only 13 percent for the Ironman distance
held in the same city. Interestingly, the increase in female participation this
last decade at the Zurich short-distance triathlon only appears for females
between 40 and 54 years old and not for younger female triathletes. In
contrast, the participation of male triathletes at this event did not change
during the last decade.
Nowadays, female triathletes have the same opportunities to train and
compete as males do. Females may be gaining more competitive opportuni-
ties as they age, or they may be seeking out competition later in life, after
childbirth, or later in their careers. Another aspect linked to the increase in
participation of female triathletes is motivation. If winning prize money can
be an additional motivation for elite triathletes, having fun and staying in
good health may be the main motivations for recreational female triathletes.
The performance density in triathlon, quantified by the time difference
between the winner and 10th place, is greater in males than in females,
whatever the event (see table 3.1). For example, at the Hawaii Ironman
Triathlon between 1981 and 2010, the average time difference between the
winner and 10th place was 5.8 percent for the males and 7.5 percent for the
females. During the last 5 years, the time difference between the winner and
10th place decreased for both males (about 3.1 percent) and females (about
5.7 percent), suggesting that elite female performance density will probably
become similar to what is seen among males in the future.
42 } Lepers

Table 3.1 Time Difference Between the Winner and 10th Place, Expressed as a
Percentage of the Winner’s Performance for Males and Females at Various Events
in 2009
Difference 10th to 1st in 2009 Male Female
Olympic Triathlon World Championship 0.7% 1.1%
Grand final (with drafting)
Gold Coast, Australia
Xterra Triathlon World Championship 4.5% 11.7%
Maui, Hawaii, USA
Half-Ironman World Championship 3.5% 6.8%
Clearwater, Florida, USA
ITU Long-Distance World Championship 3.8% 9.8%
Perth, Australia
Ironman World Championship 2.3% 8.8%
Kona, Hawaii, USA

Long-Distance Triathlon:
The Example of the Hawaii Ironman Triathlon
The analysis of male and female performances during the Hawaii Ironman
Triathlon World Championship, considered as the premier race in the field of
long-distance triathlon, provides accurate insights into gender difference in
long-distance triathlon. A study conducted by Lepers5 in 2008 showed that
overall performance time of elite male and female triathletes at the event
decreased rapidly between 1981 and the late 1980s and then plateaued thereaf-
ter for both males and females (figure 3.1). During the last two decades, while
swimming times for males and females and running times in males tended
to stagnate, running times in females marginally improved. In contrast,
cycling performance over time was more stochastic, presumably caused by
the substantial effect of wind conditions. Between 1988 and 2010, the gender
difference remained stable and practically identical for swimming (+0.8 per-
cent per decade), increased a little for cycling (+1.3 percent per decade) and
decreased somewhat more for running (−1.8 percent per decade).

Gender Difference in Swimming


is Smaller Than in Cycling or Running
The average gender difference in swimming performance at the Hawaii
Ironman over the last 25 years was 10.3 percent (figure 3.2). The gender dif-
ference in Ironman 3.8-kilometer swimming is consistent with values found
for various swimming events from 50- to 400-meter freestyle but is lower than
the gender difference found for 100-meter freestyle. It has been suggested
that swimming gender difference became progressively less with increasing
distance between 50 and 1,500 meters.6
Gender and Age Considerations in Triathlon | 43

110 Males
100 Females

Swim time (min)


90
80
70
60
50
40

480
440
Cycle time (min)

400
460
320
280
240

330
Run time (min)

300
270
240
210
180
150

880
810
Total time (min)

740
670
600
530
460
1981 1985 1990 1995 2000 2005 2010
Year

„„Figure 3.1 Swimming, cycling, running, and total performance times at


the Hawaii Ironman Triathlon for the top 10 males and females from 1981
E5643/Friel/Fig. 3.1/448526/TimB/R1
to 2010. Values are mean ± SD. Two races took place in 1982, in February
and October.

Two factors could explain the reduced gender difference for longer dis-
tances, especially for ocean swimming compared with sprint events in the
pool. First, the denser salt water compared with fresh water would likely raise
more of a female’s body out of the water because females have more body fat
than males do. This positioning would reduce surface area in the water and
total drag compared with swimming in fresh water and give some specific
advantage to females. Second, according to swimming studies, at lower velo-
city a woman’s drag coefficient drops somewhat compared with a man’s in
any similar water condition.
44 } Lepers

15

14 **
**

Difference in time (%)


13

12

11

10

8
12.8 10.3 13.0 13.2
7
Total Swim Cycle Run

„„Figure 3.2 Average gender difference in time performance for total


E5643/Friel/Fig. 3.3/448529/TimB/R1
event, swimming, cycling, and running at the Hawaii Ironman Triathlon for
the top 10 males and females from 1986 to 2010. Values are means ± SE.
**Significantly different from swim, P < 0.01.

The swimming gender difference in time (10.3 percent) appears on average


smaller compared with cycling (13.0 percent) and running (13.2 percent) (figure
3.2). The difference between swimming and the two other disciplines could
be explained in part by the biological gender difference in relative body fat
(body mass in females is 7 to 9 percent higher). Indeed, greater body fat may
represent a limit in weight-bearing activities such as running, but in contrast
it increases buoyancy in water.
Moreover, it has been shown that the underwater torque, a measure of the
tendency of the feet to sink, is lower for females than for males. In addition,
the mechanical efficiency of swimming corrected to body surface area is
greater for females than for males. Upper-body strength differences between
males and females are larger compared with differences in the lower body,
so presumably buoyancy issues outweigh this difference. These factors could
explain the reduced difference between males and females in swimming
compared with running and cycling.

Cycling Gender Difference Comparisons Are Difficult


The average gender difference in cycling at the Hawaii Ironman is 13.0 per-
cent in performance time. Cycling performance comparison between genders
in single cycling events is difficult because cycling does not have an official
time-trial championship with a distance close to 180 kilometers at the same
distance for males and females. For example, at the 2007 world cycling time-
trial championships, the difference between the male champion’s pace and the
female champion’s pace was 11.5 percent, but males rode 44.9 kilometers and
females rode 25.1 kilometers. Schumacher, Mueller, and Keul7 reported that in
track cycling the gender gap difference between males and females appeared
constant (about 11 percent) for distances between 200 and 1,000 meters.
Gender and Age Considerations in Triathlon | 45

The available data suggest that the difference in cycling between males and
females is of similar magnitude for much longer time-trial cycling. Greater
muscle mass and aerobic capacity in males, even expressed relative to lean body
mass, may represent an advantage during long-distance cycling, especially on
a relatively flat course such as Ironman cycling, where cycling approximates
a non-weight-bearing sport. Indeed, it has been shown that absolute power
output, which among elites is greater for males than females, is associated
with successful performance. In addition, a significant correlation has been
reported between 40-kilometer time-trial performance and body mass.8

Running Gender Difference is Slightly Decreasing


Over the last 25 years, the top 10 males have run the Hawaii Ironman mara-
thon on average 13.2 percent faster than the top 10 females. The physiological
differences between males and females in running performance that are well
identified in the literature persist in the marathon of an Ironman. Morpho-
logical (body fatness) and physiological gender differences, such as oxygen
carrying capacity (hemoglobin concentration) may partly account for the
gender difference in distance running performance.
Interestingly, during the last 25 years elite female triathletes improved
their running time by 0.8 minutes per year while running time remained
stable for males. The reasons for such an improvement in female running
performance at Ironman distance is not clear because both males and females
had the opportunity to use new training methods (e.g., altitude training) and
advances in nutrition. If females continue to improve their running perfor-
mance at Ironman in the future, they could reduce the gender difference
in the marathon and therefore in overall performance. The best example is
Chrissie Wellington, who astonishingly reduced the gap in running with her
male counterpart in Roth Ironman (Germany) to the world’s fastest Ironman
distance performance for both males and females (see table 3.2).

Table 3.2 World Best Total Event Performance Times (h:min:s) for Males and
Females With Corresponding Split Times Without Transition Times for Swimming,
Cycling, and Running at the Hawaii Ironman Triathlon and Roth Ironman (Germany)
Gender difference is expressed as a percentage of the male’s time.
Swim Cycle Run
Ironman course records Total 3.8 km 180 km 42 km
Hawaii Ironman Triathlon
Male, Craig Alexander (2011) 8:03:56 51:56 4:24:05 2:44:02
Female, Chrissie Wellington (2009) 8:54:02 54:31 4:52:06 3:03:05
Gender difference 10.3% 5.0% 10.6% 11.6%
Roth Ironman (Germany)
Male, Andreas Raelert (2011) 7:41:33 46:18 4:11:43 2:40:52
Female, Chrissie Wellington (2011) 8:18:05 49:49 4:40:39 2:44:35
Gender difference 7.9% 7.6% 11.5% 2.6%
46 } Lepers

Observations on ultratriathlons such as double- or triple-Ironman-distance


triathlon show greater gender difference in performance compared with
Ironman distance.9, 10 For example, the gender difference in total time perfor-
mance for both double- and triple-Ironman distance triathlons is close to 19
percent, which is greater than that found for an Ironman (about 13 percent).
But nonphysiological factors may have contributed to these observations. For
example, the widening of gender difference with increased distance could
be caused in part by the fewer number of female finishers in ultratriathlons.

Comparison Between Short and Long Distances


Surprisingly, gender differences in short-distance triathlon (consisting of a
1.5-kilometer swim, 40-kilometer bike, and 10-kilometer run) performances
for elite athletes have been less investigated. One reason may be that the top
international Olympic-distance triathlons (i.e., World Championship Series
events) have all been draft legal for several years and therefore finding a
reference of high-level Olympic-distance triathlon without drafting for com-
parison is difficult. The addition of several nondrafting international-distance
triathlons that offer a competitive platform for professional triathletes should
help in analyzing gender difference in short-distance triathlon performance
in the future.
Gender differences in times for swimming, cycling, running, and total
event have been compared for triathlons with distances from Olympic to
Ironman distance for the top 10 elite males and females (figure 3.3). Gender
difference in time for swimming was lower for Olympic distance triathlon (5.4

Olympic distance Off road


Half Ironman Ironman
21
Time gender difference (%)

18

15 **
**
12 **
**

9 **

3
Swim Cycle Run

„„Figure 3.3 Mean percentage difference in time for swimming, cycling, and running
at four triathlon events of different types (2007, 2008, and 2009; data pooled) between
E5643/Friel/Fig. 3.2/448527/TimB/R2-kh
the top 10 females and males: Olympic distance, World Cup Triathlon (Des Moines,
Iowa, USA); off-road, Xterra Triathlon World Championship (Makena, Hawaii, USA);
half-ironman, Half-Ironman Triathlon World Championship (Clearwater, Florida, USA);
Ironman, Ironman Triathlon World Championship (Kona, Hawaii, USA). Values are means
± SE. **P < 0.01, significantly different from off-road triathlon.
Gender and Age Considerations in Triathlon | 47

percent) than for other triathlons (about 10 to 13 percent). For cycling, gender
difference did not differ between the three conventional distances (about 10
to 13 percent). Gender difference in time for running was lower for Ironman
triathlon (9.7 percent) than for other triathlons (about 14 to 18 percent).

Gender and Age Interaction


in Triathlon Performance
Knowing that the physiological (e.g., muscle strength, oxygen carrying capac-
ity) and morphological (e.g., percentage of body fat, muscle mass) functional
characteristics change with advancing age, gender difference in triathlon
performance may also change with advancing age. After age 55, the decline
in athletic performance increases exponentially in both sexes, but this decline
is typically more pronounced in women than in men.11 This change in gender
difference with advancing age has been examined at the Hawaii Ironman
Triathlon12 (see figure 3.4).
Figure 3.4 shows that gender difference in total event performance time
increased significantly with advancing age from 55 years. Male triathletes at
the age of 60 years were on average 27 percent slower than the 30- to 40-year-
old triathletes, and the difference reached 38 percent for women.
Age- and gender-related differences in swimming, cycling, and running
performances likely result from physiological, sociological, and psychologi-
cal changes that occur. The exact reasons for these sex-related differences
26
$**
24

22 **
Gender difference (%)

20

18

16

14

12

10
18-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64
Age groups (years)

„„Figure 3.4 Average gender difference in time for total event at the Hawaii
Ironman Triathlon (2006, 2007, 2008 pool data). Values are means ± 95 percent CI.
E5643/Friel/Fig. 3.4/448530/TimB/R1
**Significantly different from all age groups from 20 to 24 to 50 to 55 years, P < 0.01.
$ Significantly different from age group 55 to 59 years, P < 0.05.
Reprinted, by permission, from R. Lepers and N.A. Maffiuletti, 2011, “Age and gender interactions in ultraendurance
performance: Insight from the triathlon,” Medicine & Science in Sports & Exercise 43(1), 134-139.
48 } Lepers

are not clear. A greater decline of one or more physiological determinants


of endurance performance for women compared with men (e.g., maximal
oxygen uptake, lactate threshold, and exercise economy) has not been evi-
denced.
Age-related changes in body composition (i.e., increase in percentage body
fat and loss of muscular mass), hormonal changes, and fluid balance changes
(e.g., decline in the thirst mechanism) could also differ between males and
females and affect triathlon performance. Finally, the reduction in training
volume and intensity could contribute to the larger declines in endurance
performance of elderly athletes. Differences in terms of years of training,
training volume, and intensity between elderly men and women triathletes
performing Ironman triathlon may exist, but further work is required to
clarify this.
Interpretation of cross-sectional comparisons of triathlon performance
times across ages and sexes must be made carefully. Compared with men,
fewer women are competing in triathlon events, especially in the older age
categories. For example, the percentage of women participating at the Hawaii
Ironman Triathlon during 2007 to 2009 corresponded on average to 27 percent,
but women finishers in the age group 60 to 64 years represented only 3 percent
of the women’s field. This participation difference will no doubt diminish
over the next couple of decades. As a result, triathlon performances of the
oldest women will probably improve more rapidly than those of the oldest
men as the new generation of well-trained young female athletes moves into
older age group competition.

Age-Related Decline
in Triathlon Performance
Even if it is possible for an 80-year-old male athlete to finish an Ironman
triathlon in less than 17 hours (e.g., Lew Hollander finished the Hawaii
Ironman Triathlon in 15:48 in 2010; see table 3.3), the gradient in declining
performances increased notably after the age of 55 years for both sexes, and
female performances tended to decline faster than those of males. Accord-
ing to Reaburn and Dascombe,13 the physiological factors affecting endur-
ance
. performance with increasing age are maximum oxygen consumption
(VO2max), maximal heart rate, stroke volume, lactate threshold, economy of
movement, muscle-fiber type, activity of aerobic enzymes, blood volume,
and skeletal muscle mass.
The decline in endurance
. performance appears primarily because of an
age-related decrease in VO2max. The decrease of. muscle mass with advancing
age plays a role in the age-related decrease in VO2max in master endurance
athletes. On average, the muscle area decreases by about 40 percent between
20 and 80 years old. Both slow- and fast-twitch fibers decline with increasing
age, although the loss of fast-twitch fibers is greater.
Gender and Age Considerations in Triathlon | 49

Table 3.3 Best Total Performance Times for Male and Female Age Groups With
Corresponding Split Times at the Hawaii Ironman Triathlon
Male
Year 2011 a
1996 b
2009 2006 2005 2010 2011 2011 2005 2012
Age group 18–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 >80
(years)
3.8 km swim 51:56 53:16 56:55 1:03:32 1:07:09 1:16:20 1:14:11 1:47:46 1:37:47 1:49:34
(h:min:s)
180 km cycle 4:24:05 4:49:55 5:04:47 4:51:44 5:00:17 5:19:17 5:42:08 5:47:33 6:39:35 7:42:08
(h:min:s)
42 km run 2:44:02 2:45:20 3:04:21 3:24:51 3:34:03 3:25:28 4:14:52 3:52:47 4:55:43 5:41:51
(h:min:s)
Total (h:min:s) 8:03:56 8:28:31 9:11:24 9:26:23 9:47:29 10:08:15 11:19:07 11:45:05 13:27:50 15:38:25
Female
Year 2009c 2010 2012d 2005 2010 2010 2010 2000 2005 NA
Age group 18–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 NA
(years)
3.8 km swim 54:31 1:13:52 1:06:21 1:08:08 1:06:18 1:32:16 1:21:02 1:37:54 1:45:05 NA
(h:min:s)
180 km cycle 4:52:06 5:25:00 5:06:07 5:31:56 5:35:36 6:27:46 6:47:28 7:24:33 7:25:17 NA
(h:min:s)
42 km run 3:03:05 3:17:48 3:09:18 3:47:23 4:00:08 4:05:22 4:59:01 6:07:02 6:19:43 NA
(h:min:s)
Total (h:min:s) 8:54:02 10:02:35 9:26:25 10:35:59 10:51:43 12:17:24 13:16:32 15:19:19 15:54:16 NA

Craig Alexander, 38 years old; bDave Scott, 42 years old; cChrissie Wellington, 32 years old; dNatascha Bad-
a

mann, 45 years old.

Effect of Locomotion Mode


The age-related decline in triathlon performance is specific to the discipline.
For both short- and long-distance triathlons, the age-related decline in cycling
performance is less compared with running and swimming performances14,
15, 16
(see figure 3.5). The question of whether there is better maintenance of
cycling performance or a greater decline in running and swimming perfor-
mances with advancing age can be raised.
One explanation for the different age-related declines in cycling and
running may involve the mechanical power required by those disciplines.
Mechanical power output (P) in running depends on the velocity (V) because
P = k × V (k is a constant), whereas it depends on the third power of velocity
. cycling (P = k × V ). If we assume that the changes in aerobic capacity (e.g.,
in 3

VO2max) with age are directly related to the decline in mechanical power, a
similar reduction in power output for running and cycling with advancing
age would induce a lower reduction in cycling velocity compared with run-
ning velocity. This relationship may explain, in part, why the magnitude of
the decrease in cycling performance with age was less than that for running.
50 } Lepers

1.0 Short distance 1.0 Ironman distance

Performance ratio
Performance ratio

0.9 0.9

0.8 0.8
Swim Swim
0.7 Cycle 0.7 Cycle
Run Run
0.6 0.6

0.5 0.5
18- 25- 30- 35- 40- 45- 50- 55- 60- 65- 70- 18- 25- 30- 35- 40- 45- 50- 55- 60- 65- 70-
24 29 34 39 44 49 54 59 64 69 74 24 29 34 39 44 49 54 59 64 69 74
a Age groups (years) b Age groups (years)

„„Figure 3.5 Age-related decline in swimming, cycling, and running performance for
E5643/Friel/Fig. 3.5a/448532/TimB/R2-kh
(a) short-distance E5643/Friel/Fig.
triathlon and for (b) Ironman-distance triathlon3.5b/448533/TimB/R2-kh
(mean ± SE). For short-
distance triathlon the cycling performance ratio remained significantly (P < 0.01) greater than
swimming from age group 40 onward, and from age group 45 the cycling performance
ratio remained significantly (P < 0.01) greater than running. For Ironman-distance triathlon,
the cycling performance ratio remained significantly (P < 0.01) greater than swimming
from age group 45 onward, except for age group 70, and from age group 50 the cycling
performance ratio remained significantly (P < 0.01) greater than running.
Adapted, by permission, from R. Lepers, F. Sultana, T. Bernard, et al., 2010, “Age-related changes in triathlon
performance,” International Journal of Sports Medicine 31(4): 251-256.

Running involves stretch-shortening cycles (SSC) with eccentric muscle


contractions, whereas cycling is a non-weight-bearing activity that involves
dominant concentric muscle contractions. Fast-twitch fibers, however, atrophy
more than slow-twitch fibers do with age. Because fast-twitch fibers seem
more susceptible to damage than slow-twitch fibers during SSC, the greater
reduction in running performance compared with cycling could be related
to muscle typology changes with age.
But the changes in muscle fiber type distribution (i.e., the percentage of
type I muscle fibers) with advancing age seem less pronounced in well-trained
masters athletes compared with untrained older adults. Moreover, in terms
of muscle damage, although some animal studies have shown that recovery
from eccentric exercise-induced muscle damage is lessened in older subjects
compared with the young, the results from human studies are not as clear.
Results might differ among muscle groups, but there is no evidence that the
muscles of older triathletes are more susceptible to muscle damage than
muscles of young triathletes.
Findings for competitive long-distance runners indicate that the decline in
running times parallels the age-related reductions in maximal oxygen uptake
and lactate threshold. The contributions of these respective physiological fac-
tors to age-related declines in cycling performance, however, are unknown
at present.
Indeed, few studies are available on the effect of age on cycling perfor-
mance. For example, Balmer, Bird, and Davison17characterized the decline
in maximal oxygen consumption and maximal aerobic power with age in
cycling. These authors found that the relative intensity expressed as a per-
Gender and Age Considerations in Triathlon | 51

centage of maximal heart rate or maximal oxygen uptake as well as economy


was not affected by age during a 20-minute time trial. Thus, the possible
lesser decline in cycling performance with advancing age, because of a lesser
reduction in lactate threshold or economy compared with running, needs to
be confirmed by additional studies.
An alternative explanation for the smaller age-related performance decline
in cycling compared with running is the maintenance of a relatively greater
exercise training stimulus in cycling. An overall reduction in the exercise
training stimulus generally occurs with advancing age. In running, the
decline in exercise performance with age has been partly attributed to an
increased incidence of orthopedic injuries, which would limit running
training volume of many older athletes. This factor may influence cycling
performance to a much lesser extent.
Moreover, it has been shown that the protein synthesis rate decreases in
older subjects compared with young subjects, which could limit running
training volume in older triathletes in which muscle damage occurs. With
advancing age, triathletes would likely spend more training time cycling
than running because of changes in physical factors (e.g., increased preva-
lence of injuries). Further prospective studies are necessary to quantify the
changes in training volume in triathletes with age because training history
is an important aspect of performance status at older age.

Comparison Between Short and Long Distances


The triathlon duration exerts an important influence on the age-associated
changes in triathlon performance.15 It has been shown that age-related
changes in swimming performance were not influenced by triathlon dura-
tion; the magnitudes of decreases in swimming performance were similar
for Olympic versus Ironman triathlon (see figure 3.6). Swimming is the first
discipline during a triathlon, so triathletes perform it without suffering from
accumulated fatigue, in contrast to the cycling and running disciplines. On
average, swimming performance declines by about 15 percent at 50 years old
and reaches about 38 percent by 70 years of age.
Previous studies showed no effect of task duration for shorter-distance
swimming events. For example, Tanaka and Seals6 found no significant dif-
ferences in the magnitude of performance decline with age (about 32 percent
at 70 years) among distance events ranging from 100 to 1500 meters in a pool.
The lower performance decline observed by these authors for the same age
(70 years old) compared with triathlon data might be related to differences
between water conditions: calm in swimming pools versus stochastic in open
water for triathlon.
In contrast to swimming, cycling and running with advancing age had a
less pronounced decline at Olympic triathlon than at Ironman triathlon. It
is not clear why task duration in triathlon exerts an influence on age-related
declines in cycling and running performance.
52 } Lepers

Cycle performance ratio Swim performance ratio


1.0
0.9
0.8
0.7
0.6
0.5
20 25 30 35 40 45 50 55 60 65 70

1.0
*
0.9 *
*
0.8 *
0.7
0.6
0.5
20 25 30 35 40 45 50 55 60 65 70
Run performance ratio

1.0
*
0.9 *
0.8 *
*
0.7 *
0.6
0.5
20 25 30 35 40 45 50 55 60 65 70
Total performance ratio

1.0
0.9 *
*
0.8 *
0.7 *
Ironman
0.6 Olympic
0.5
20 25 30 35 40 45 50 55 60 65 70
Age group (years)

„„Figure 3.6 Age-related decline in swimming, cycling, running, and total


performance for short-distance and Ironman
E5643/Friel/Fig. distance triathlons (mean ± SE).
3.6/448535/TimB/R1
There was no distance effect for swimming performance, but age-related
declines in cycling, running, and total performance were less pronounced
for short-distance triathlon. *P < 0.01, significantly different from Ironman
distance triathlon for the same age group.
Adapted, by permission, from R. Lepers, F. Sultana, T. Bernard, et al., 2010, “Age-related changes in
triathlon performance,” International Journal of Sports Medicine 31(4): 251-256.

For the 70 to 74 year age group, the finish time is about 3 hours for the
Olympic triathlon versus about 15 hours for Ironman triathlon. Certainly,
the Ironman triathlon induces greater neuromuscular fatigue in cycling and
running compared with Olympic triathlon. Furthermore, muscle damage
during a 10K run of Olympic distance is limited compared with that occurring
Gender and Age Considerations in Triathlon | 53

during the Ironman marathon. Endurance exercise increases the use of


endogenous fuels to provide energy for working muscles.
It has been suggested that older subjects oxidize more glucose and less
fat during moderate-intensity exercise.18 Compared with young adults, older
adults seem to oxidize less fat during endurance exercise performed at either
the same absolute or relative intensity.19 This shift in substrate use is presum-
ably caused by age-related changes in skeletal muscle, including decreased
skeletal muscle respiratory capacity. Therefore, changes in substrate availability
and utilization with advancing age may explain in part the task duration effect
on triathlon performances, at least in cycling and running.
Additional studies examining the changes in training volume and physio-
logical characteristics of older triathletes are required to understand the
age-associated changes in triathlon performance. Such studies will provide
valuable information for understanding how to maintain physical capacity
and performance with advancing age.

Conclusion
Physiological factors (e.g., percent body fat and oxygen carrying capacity) and
nonphysiological factors (e.g., rate of participation) may be responsible for
gender difference in triathlon performances. The difference in triathlon total
performance between elite males and females (about 10 percent) is similar to
differences seen in other endurance sports such as marathon running. Swim-
ming gender difference in time is generally smaller compared with cycling
and running and can be explained in part by the biological difference in rela-
tive body fat. Female triathletes now have the same opportunities as males
do to train and compete, and an increase in female participation in the future
will probably improve the female performance density that remains, to date,
smaller than that in male races.
Age-related decline in triathlon performance depends on locomotion mode
(swimming versus cycling versus running) and distance (short versus long
distance). Although younger triathletes still have the advantage with regard
to overall performance, masters triathletes have shown in these last decades
a relative improvement of their performances in the three disciplines.20
The question as to whether older triathletes have reached their limits in
triathlon performance should therefore be raised. Most health care provid-
ers agree that racing an Ironman triathlon after 50 years of age is not good
for the body. But most older triathletes say that training for triathlon is one
of the healthiest things that they can do. Where should the limits be set? If
masters triathletes perform at a high level for a long time, it is reasonable to
expect that those destined to maintain that intensity could do so because they
remained largely injury free. A framework for preparticipation evaluation,
training programs, and injury prevention is required to help older triathletes
reach their participation and performance goals injury free, to maximize the
benefits and minimize the risks.
This page intentionally left blank.
Pa r t

II
Technical Execution
and Efficiency
in Each Event
This page intentionally left blank.
CHaPTEr
4
Swimming Biomechanics
for Triathlon
—David Pease, PhD

F or humans, one of the most inefficient ways to move through space is to


swim. In fact, most swimmers are only about 5 percent efficient, whereas at
other activities such as running athletes can be almost 90 percent efficient.1, 2, 3
That finding means that the amount of power that a swimmer is able to create
by pushing and pulling on the water to propel herself is equal to only about
5 percent of the total amount of power that her body is using to generate
those movements.
Because of that inefficiency, swimming is the most challenging of the three
disciplines for most triathletes. But this low efficiency also allows triathletes
to achieve substantial improvements in performance with relatively small
improvements in stroke efficiency.
Triathletes can achieve these improvements in efficiency in many ways.
Common technique faults may lead to inefficient swimming, but these faults
can be addressed. The primary forces that act during swimming—propulsion
and drag—can be manipulated to increase swimming efficiency and thereby
improve overall performance.

Technique
One of the most important aspects in improving efficiency in swimming is to
maintain a smooth stroke while holding a relatively streamlined body posi-
tion. Just how a person does that comes down to the techniques used. This
section describes key aspects of optimizing technique and limiting some of
the errors that triathletes often exhibit.

Center of Mass
Mass can be thought of as the weight of a given object, whether it is a person or
a ball or a mass of water. In conjunction with mass is center of mass. Because

57
58 } Pease

center of mass represents all the parts of the body at once, it moves according
to the movement of various parts of the body (see figure 4.1).
In terms of swimming, the position of the center of mass will influence
how a person sits in the water. When streamlining off a wall after pushing
off, the hands should be outstretched above the head. In this position the
center of mass will move closer to the lungs where most of the buoyancy is
found and put the triathlete in a more balanced position (floating more level).
If the swimmer lowers the arms to the sides, the legs sink more because
the center of mass is moving farther from the lungs and torque is created
around the center of mass. Just to clarify, torque is essentially a force, but
instead of trying to create motion in a straight line, torque is trying to cause
motion about an axis of rotation. Think of the axis of rotation as the axle of
a wheel. Using a hand to spin a wheel generates torque. Although a force
acts through the center of mass of an object, torque is created any time the
line of action of a force doesn’t pass through the center of mass of an object.

„„Figure 4.1 Center of mass position with arms (a) lowered and (b) raised.
Swimming Biomechanics for Triathlon | 59

In the example of the streamlined swimmer, the buoyant force, mostly from
the lungs, acts to lift the upper body relative to the center of mass. Similarly,
the less buoyant mass of the legs creates torque that causes the legs to sink
relative to the center of mass.

Momentum
Essentially momentum is the amount of motion an object has. Momentum is
determined by multiplying the mass of the object by its velocity. Therefore,
a 100-kilogram athlete moving at 2 meters per second and a 50-kilogram
athlete moving at 4 meters per second will have the same amount of motion
or momentum. Momentum is also described in Newton’s first law: When an
object has momentum in a given direction, it will continue moving in that
direction until a force acts on it to change its momentum.
For optimal performance and greatest efficiency in swimming, the swim-
mer wants the amount of forward momentum to be as constant as possible.
The athlete achieves this goal by keeping velocity as consistent as possible. If
velocity is going up and down as the triathlete goes through a stroke cycle,
then he is using more energy. The best way to maintain velocity is to maintain
a consistent level of propulsion and minimize any drag acting on the body.

Propulsion
Propulsion is a force that moves the swimmer in the desired direction, for-
ward through the water. A force generates movement of an object in a given
direction. In the case of swimming propulsion, this direction is the forward
swimming direction. Drag is the opposite. Drag is any force that tries to move
the swimmer backward or at least slow her down.
One of the most confusing aspects of dealing with forces is the concept of
a reaction force. Many have likely heard the saying “For every action there
is an equal and opposite reaction.” This fundamental principle in physics
is derived from Isaac Newton’s third law of motion. In its most basic form
this principle means that when a person pushes on something with a given
force, the object pushes back with the same amount of force. In swimming
the water pushes back on a swimmer’s body with the same amount of force
that the swimmer pushes on the water. This force pushing back on the swim-
mer is actually moving him through the water or slowing him down. This
water force is termed the reaction force because it is a reaction to the force
generated by the swimmer.
The counterintuitive notion of reaction forces leads to some confusion
when describing the propulsive forces generated by a swimmer. We can
find instances in which propulsion is termed either lift generated or drag
generated. In that context the drag is referring to the force generated by the
swimmer, which is directed back toward the feet in the drag direction. The
propulsion is then a reaction to that force. This just means that the force is
60 } Pease

acting in the same direction as the swimmer’s hand is moving and is gener-
ated by a paddlewheel type of motion.
Propulsion described as being generated by lift indicates a force generated
during a sculling or lateral motion of the hand. This movement creates a
propulsive force similar to the way that a propeller creates propulsive force
for a boat. Although the propeller doesn’t move in the same direction as the
boat, it is able to elicit a reaction force from the water that creates propulsion.
Again, this force is the reaction force to the forces generated by the hand.
Substantial discussion has occurred over the last 40 years about how swim-
mers generate propulsion.4–12 The current thinking is that propulsion is a
mixture of the two types, and a majority of the force is likely generated by
the drag type of propulsion.4, 13 This thinking largely dictates how athletes
are coached to perform the techniques used in swimming, in particular the
hand paths that are taught.

Pull
Because it takes advantage of both the lift and drag forces, the S pattern is
generally accepted as the best arm pull pattern (see figure 4.2). The hand
enters the water between the shoulder and the center of the head and then
moves laterally as the arm extends such that the hand ends up just wide

„„Figure 4.2 S pull pattern.


E5643/Friel/Fig. 4.2/448545/TimB/R1
Swimming Biomechanics for Triathlon | 61

of the shoulder. This wide point is often described as the end of the catch
phase of the swimming stroke. From this wide point the insweep begins, in
which the hand moves on a slight angle and gets to the midline of the body
at a point about level with the shoulders. The hand then changes direction
slightly again and sweeps outward until it exits the water next to the body
at about the level of the hip.
To maximize propulsive forces, the fingers should be pointing toward the
bottom of the pool throughout the entire pull. During the insweep phase
this positioning is achieved by bending at the elbow, not at the wrist. If the
wrist is flexed, the elbow tends to drop and point toward the bottom of the
pool. This dropping limits the ability of the forearm to create propulsive
force. This technical flaw, and the way to correct it, is discussed in detail
later in the chapter. Additionally, flexing the wrist during this phase of the
stroke tends to overuse the smaller muscles of the forearm, which are more
prone to fatigue, and therefore reduces the amount of propulsion generated
during a pull.
Two additional concepts often used when talking about technique are
impulse and power.
Impulse is the rate at which a force is applied. It is determined by multiplying
the force (or torque) by the amount of time over which it is acting. In swim-
ming, impulse is used when describing the pull. A swimmer who applies
a relatively low force over a long period can achieve the same impulse as a
swimmer who applies a large force over a short period.
Some athletes use a long stroke length and low stroke rate, whereas others
use a very high stroke rate but shorter stroke length. They can elicit the same
impulse using different techniques, although the athlete with the longer,
slower stroke is generally more efficient and can maintain the stroke for a
longer time.14, 15
Related to impulse is power. Instead of looking at the time over which a
force is applied, power is the product of the force or torque and the velocity at
which it is applied. Again, in swimming terms, if a hand moving at a higher
velocity can apply the same force as a hand that is moving more slowly, then
the faster hand creates more power.

Kick
To this point we have focused on the propulsive forces generated by the hands.
Obviously, most people can generate propulsion with both the hands and the
feet. Particularly for triathletes, the kick acts in two other ways that are highly
beneficial to performance. Fortunately, neither of these functions requires
high kicking rates, thereby helping to save the legs for the later stages of the
race. The first and probably most important function in terms of triathlon
swimming is the generation of vertical forces that help keep the legs higher
in the water and maintain a more streamlined position. This action provides
the same function as a wetsuit, which increases the amount of buoyancy,
62 } Pease

thereby keeping the legs elevated. In nonwetsuit swims, the triathlete needs
to be able to use the legs to fulfill this function.
The final role of the kick is to help maintain balance in the stroke by provid-
ing a counteraction to the movement of the arms. This situation is also present
in running when the contralateral arm swings forward with each forward
stride. In swimming, however, the balance needs to occur around the entry
and extension phases of the stroke. At that point in the stroke the contralateral
leg should be kicking downward. As seen in figure 4.3 this motion promotes
a coordinated roll between the shoulders and hips and keeps the muscles of
the trunk in better alignment, thereby enhancing the stability of the body in
the water and allowing enhanced force delivery to the water.

„„Figure 4.3 Proper balance between pull and kick. Note the right leg
kicking down at point of catch for the left hand.
Swimming Biomechanics for Triathlon | 63

Drag
While discussing ways of increasing propulsion in swimming, to maximize
performance, the triathlete also needs to look at ways of reducing the resis-
tance, or drag, experienced while moving through the water. Triathletes
experience three primary types of drag in the water.

Form Drag
The first main type of drag is known as form drag, or pressure drag. This
type of drag is caused by the frontal area of the swimmer being exposed to
the oncoming water flow.
The easiest way to picture this is to imagine that a coach is taking a picture
of an athlete swimming from directly in front, just below the surface of the
water. The more of the body that appears in the picture, the greater the drag
acting on the body will be. Form drag increases with the square of velocity, so
someone trying to swim twice as fast will experience four times the amount
of drag. Minimizing form drag is obviously highly beneficial.
The primary technique aspect used to minimize form drag is streamlin-
ing (see figure 4.4). This point is related to the lifting role of the kick. If the
legs are allowed to drag low in the water, then the area of the body increases
substantially, causing an increase in drag.
Form drag is reduced when buoyancy is increased. Obviously, an athlete
who is more buoyant rides higher in the water, exposing less frontal area
to the flow of the water. By increasing buoyancy and reducing drag forces,
triathletes are able to use less energy to travel at a given speed through the
water. For that reason, triathletes generally find swimming in a wetsuit
easier than without. The wetsuit provides the extra lift force that is normally
provided by the kick.16

E5643/Friel/Fig. 4.4a/448547/TimB/R1

b
E5643/Friel/Fig. 4.4b/448548/TimB/R1
„„Figure 4.4 (a) Good and (b) poor streamline positions.
64 } Pease

Frictional Drag
The second main type of drag is frictional, or skin, drag. This type of drag
results from the slipperiness of the surface of the swimmer’s body. Generally,
skin is not slippery. Wetsuits and normal swimsuits made of materials more
slippery than skin have been developed in recent years.
Although form drag has a squared relationship to swimming speed, fric-
tional drag has only a linear relationship. The magnitude of frictional drag is
therefore relatively small compared with other types of drag, so the benefits
of reducing it are not as great as those found by reducing other types. Other
than wearing suits designed for drag reduction, swimmers can do little to
reduce frictional drag.

Wave Drag
The final primary type of drag experienced during swimming, and the most
difficult to reduce, is wave drag (see figure 4.5). Wave drag is created when
the swimmer moving through the water creates waves by pushing water up.
To push the water, the swimmer needs to use energy that would otherwise be
used to create propulsion. The bigger the waves are, the greater the amount
of energy is lost.
The biggest problem with wave drag is that it increases with velocity to
the sixth power, which means that if the swimmer doubles her speed, wave
drag increases 64 times! The easiest way to limit wave drag is to develop a

„„Figure 4.5 A wave generated during swimming, leading to the creation of wave drag.
Swimming Biomechanics for Triathlon | 65

smooth swimming technique that consists of a good streamline position in


the water and long, smooth movements that maximize the amount of motion
in the horizontal direction and minimize motion in the vertical.

Common Mistakes and Corrections


So far discussion has focused on the main factors underlying swimming
performance. At this point it may be useful to discuss some further biome-
chanical issues common among triathletes.

Kicking
The problem: The triathlete generates only a slight amount of force through
kicking. Some even say that they move backward when kicking.
The cause: Generally, the cause is lack of ankle plantar flexibility (toe
pointing), as shown in figure 4.6.
The result: Because the primary source of propulsion from the kick results
from drag-type forces, the foot needs to be in a position in which it can push
back on the water to get the propulsive reaction force from the water. The lack
of ankle plantar flexibility often leads to another common problem with triath-
lete swimming technique—initiating the kick by flexing the knee excessively.

E5643/Friel/Fig. 4.6a/448550/TimB/R1

„ Figure 4.6 Examples of (a) good and (b) poor ankle plantar flexibility on the kick.
E5643/Friel/Fig. 4.6b/448551/TimB/R1
66 } Pease

Although flexing the knee does indeed get the foot into the proper position, it
also exposes the back of the leg to the oncoming water flow. The flexed knee
can also lead to driving the knee down as the hip flexes, which exposes the
front of the thigh to the oncoming water. The motion looks similar to some-
one riding a bike. Both of these positions increase resistance and counteract
any benefits of getting the foot into a propulsive position. Additionally, a
substantial amount of additional effort is required to perform this larger
kicking motion, which wastes energy that will be required during the bike
and run legs of the race.
The solution: The best way to improve kicking is to enhance ankle plantar
flexibility and focus on kicking from the hip with only slight flexion at the
knee.

Side-to-Side Imbalance
The problem: One of the most common faults demonstrated by triathletes
during their swimming is a side-to-side imbalance in their movement pat-
terns (see figure 4.7).
The cause: The imbalance is seen most often as a difference in pull widths
of the hands; one hand often pulls much wider away from the midline of
the body than the other. This situation is similar to trying to row a boat with
oars of two different lengths. The result is that the boat continually veers off
to one side.
The result: The same thing happens in swimming. If the arms are pull-
ing at different distances away from the midline of the body, then the torque
around the center of mass will not be balanced and the swimmer will con-
tinually turn to one side (see figure 4.7). This flaw can be especially costly

a b

„„Figure 4.7 Example of (a) a proper-width pull and (b) a too-wide pull. The
combination of these
E5643/Friel/Fig. two creates an imbalance
4.7a/448552/TimB/R1 and poor course
E5643/Friel/Fig. control.
4.7b/448553/TimB/R1
Swimming Biomechanics for Triathlon | 67

in open-water swims where the swimmer cannot follow a black line on the
bottom and make corrections to keep moving in a straight line. In open water
the constant veering off course and resulting course corrections lead to extra
distance swum and a significant amount of wasted energy.
The solution: A simple way to test a triathlete’s stroke for imbalance is to
have the person swim in the middle of a lane in a pool with his eyes closed.
The key is that the swimmer must not peek; the stroke must be allowed to
guide his movement. If the triathlete ends up running into a lane line, an
imbalance is present somewhere in the stroke. If an imbalance is present in
the stroke, the fix is to adjust the path that the hands are travelling. The best
hand path follows an S shape. Observed from the point of view of the swim-
mer, however, the path is much straighter because of body rotation. A good
drill to help monitor hand path is to swim down the middle of the lane and
watch the position of the hand relative to the line on the bottom of the pool.
The hand should track pretty much right down the line. A key to success in
this drill is that the swimmer must be using a significant amount of body
rotation (or roll) around the long axis (head to toe) of the body.

Reduced Body Roll


The problem: A reduced amount of body roll is a common issue with tri-
athletes.
The cause: Insufficient rotation causes several consequences that can limit
performance.
The result: Being flatter in the water increases drag forces compared with
rotating. Underrotating also increases the difficulty of moving the arms
through the recovery phase of the stroke. The arms move closer to the surface
of the water, necessitating greater rotation at the shoulder. The increased rota-
tion at the shoulder can lead to shoulder pain in those who have inadequate
shoulder flexibility.
The solution: Ideally, the body roll should be about 30 to 45 degrees to both
sides (see figure 4.8). This technique not only remedies the problems with flat
body position but also enhances the extension at the front of the stroke after
the hand enters the water, thereby adding to the propulsive impulse when a
high elbow catch is used. To achieve this amount of body roll, a bit of exag-
geration is useful. Therefore, the best drill is the shark fin drill. For this drill,
as the swimmer finishes the pull and the hand comes out of the water, she
places the hand on the hip and pretends to be a shark while rolling all the
way onto her side. While doing this she keeps the other arm extended out in
front and the head aligned with the body, looking down toward the bottom
of the pool. She stays in that position for about six kicks and then repeats on
the other side. As the swimmer gets better at the drill, she should reduce the
number of kicks but maintain the same amount of body roll. When triathletes
start swimming normally, they should begin to feel more natural with the
greater amount of body rotation.
68 } Pease

„„Figure 4.8 Good shoulder rotation. Rotation should be around 30 to


45 degrees.

Dropped Elbow
The problem: The final and most common flaw, not only in triathlon swim-
ming but also in competitive swimming, is the dropped elbow (see figure 4.9).
The cause: Although a high elbow, sometimes seen as early vertical fore-
arm (EVF), has definite advantages in terms of propulsion, it requires specific
strength as well as flexibility at the shoulder joint. Although poor shoulder
flexibility is not usually much of a problem in competitive swimmers, it is
common in triathletes. This weakness limits the ability of many triathletes
to use a high elbow catch effectively.
The result: The dropped elbow position redirects the reaction forces from
the water, making them vertical rather than horizontal, and therefore limits
the amount of propulsion gained.
The solution: Most triathletes should try to get as much of a high elbow as
they can given their limitations. If a line were to be drawn from the shoulder
to the wrist of the pulling arm, the minimum should be that the elbow should
lie on that line at the time of the catch (see figure 4.10). As the pull progresses
Swimming Biomechanics for Triathlon | 69

E5643/Friel/Fig. 4.9/448557/TimB/R1
„ Figure 4.9 Dropped elbow position common to many triathletes.

„ Figure 4.10 Arm position that should be deemed the minimum high elbow catch
E5643/Friel/Fig. 4.10/448558/TimB/R1
position with the elbow lying on the line drawn between the shoulder and wrist.

through the remainder of the insweep, maintaining a high elbow becomes


easier because of the more natural position of the shoulder joint as compared
with that needed during the initial catch. As flexibility and strength increase,
the aim should be for the elbow to be a greater distance above the line between
the shoulder and the wrist. The easiest way to work on fixing this problem is
to forget about the hands for a while. Triathletes should imagine that they are
trying to catch the water with the forearm only. By doing so they will naturally
adopt a higher elbow position (see figure 4.11). Swimming with closed fists
is another way to help achieve this in combination with the forearm catch.
70 } Pease

E5643/Friel/Fig. 4.11/448559/TimB/R1
„ Figure 4.11 Example of a high elbow catch. Although uncommon shoulder
flexibility is required to achieve an even higher elbow catch position, most triathletes
should be able to achieve close to this position with coaching.

Conclusion
This chapter describes the basic biomechanics of triathlon swimming and
offers some useful ideas for correcting common flaws in triathlon swimming.
As with all sport movements, swimming is controlled by some relatively
simple concepts of physics. By studying these factors, triathletes should gain
at least a rudimentary grasp on these issues, which should allow them to
think about how they can change their movements to improve performance.
As we’ve seen in this chapter, many of the common flaws in the swimming
style of triathletes can be easily corrected by thinking about how swimmers
generate propulsion and what positions can optimize those forces. Remember
that the biggest factors in improving swimming performance are to keep the
stroke smooth and consistent to maintain momentum, to keep the stroke
balanced so that the forces from the right and left sides are equal, and to use
techniques that produce the maximum amount of propulsion and the mini-
mum amount of drag. If the triathlete can achieve these few things, he will
be well on his way to improving his swimming performance and therefore
his overall triathlon performance.
CHAPTER
5
Cycling Biomechanics
for Triathlon
—Jeff Broker, PhD, and Sean Langlais, MSc

A mong the three events in a triathlon competition, the bicycle leg stands
alone as the event upon which performance is directly related to the
effective integration of human and machine. This integration attracted the
attention of biomechanists decades ago, partly because of the fascination with
optimal coupling of human to machine but also because of the relative ease
with which cycling can be used to study musculoskeletal performance. In
recent years triathletes have benefited from the sciences targeting the optimal
integration of athlete and machine, specifically when triathlon equipment
options, body positions, and training methods are visited.
This chapter presents relevant scientific findings concerning the bio-
mechanics of cycling for the triathlete. First, the nature of cycling power and
its importance to triathletes is introduced. As cycling power is explored, the
critical elements defining resistance to motion are exposed. The chapter then
shifts to the pedals, where energy derived from muscular effort is delivered
to the bicycle. Pedaling is explored in detail, focusing on pedaling mechan-
ics, its sensitivity to body position, myths surrounding effective pedaling,
and drills to improve pedaling. Next, the aerodynamics of cycling, with a
general focus on the triathlete, is described. Last, perhaps as an ironic transi-
tion into the next chapter on running, we briefly focus our attention on the
fascinating T2 transition, in which triathletes must run effectively and safely
immediately after cycling.

Cycling Power
Cycling power is the rate of performing work, usually expressed in watts (W).
Work represents force applied over a distance. Thus, cycling power represents
the propulsive force moving the bicycle and cyclist forward, through a given
distance, over a specific period.

71
72 } Broker and Langlais

Two variants of a simple power equation can be developed:


power = (force × distance) / time (equation 5.1)
or, by rearranging parentheses,
power = force × (distance / time) (equation 5.2)
The term within the parentheses in equation 5.2, distance per unit time, is
simply cycling velocity. Thus,
cycling power = force × cycling velocity (equation 5.3)
In equation 5.3 cycling power (W) equals the product of the force applied
by the rear wheel against the ground to produce forward motion (in newtons,
N), and cycling velocity (in meters per second, m/sec). A cyclist overcoming
total resistive forces of 7.8 pounds (3.5 kg) (35 N) at a velocity of 19.2 miles
per hour (8.58 m/sec) requires an average power delivery to the rear wheel
of roughly 300 W. To put these numbers into perspective, 1 horsepower (HP)
is equivalent to 746 W; thus, our triathlete would produce roughly 0.4 HP to
ride at 19.2 miles per hour under these conditions.
An alternative form of mechanical work is represented by a torque applied
through an arc of rotation. This represents the rotational variant of work.
Because power equals work performed per unit time, another formula rep-
resenting mechanical power is developed as follows:
power = (torque × angular rotation) / time (equation 5.4)
or, by rearranging parentheses,
power = torque × (angular rotation / time) (equation 5.5)
The rotational variant of power (equation 5.5) applies when cycling power
is measured. By measuring crank torque within the chainring and angular
velocity of the cranks, power at the cranks can be calculated. Alternatively,
torque at the bicycle rear hub, multiplied by rear wheel angular velocity,
provides a measure of cycling power. Notably, the difference between power
measured at the cranks and at the rear hub represents drivetrain energy losses.

Resistance
Cyclists deliver energy to the pedals to overcome resistive elements associ-
ated with bicycle movement.
Resistance to bicycle motion can be expressed in relation to five specific
elements:
• Resistance to bicycle acceleration
• Resistance to climbing (change in height along a grade)
Cycling Biomechanics for Triathlon | 73

• Static rolling resistance


• Dynamic rolling resistance
• Aerodynamic drag
Rider-plus-bicycle mass figures linearly into the first three elements listed,
so a 5 percent increase in rider-plus-bicycle mass (and thus weight) results
in a 5 percent increase in cycling power connected with bicycle acceleration,
climbing hills, and rotating the tires. Herein lies the advantage of lighter
bicycles and, for that matter, lighter riders. The effect of hill climbing on
cycling power is addressed further in chapter 8.
Next, power to overcome dynamic rolling resistance, associated with
deformation of the tire as it contacts the ground, is not dependent on rider
and bicycle mass but increases as the square of cycling velocity. Doubling
cycling velocity quadruples the power needed to overcome dynamic rolling
resistance. Fortunately, the magnitude of this energy loss is relatively small.
Finally, and most important, the power to overcome aerodynamic forces
increases as the third power of cycling velocity. Thus, doubling cycling speed
increases the power needed to overcome aerodynamic forces by a factor of
eight.
When considering these resistive elements, triathletes may want to think
through which of these elements they can control and which they cannot.
Weight reduction clearly plays a role, most notably when the bicycle is being
accelerated and when climbing hills. High-pressure tires reduce resistive
loads at the tire–pavement interface. Body position, bicycle frame and com-
ponent selection, and clothing and helmet features play an important role in
managing aerodynamic resistance.
What cannot be controlled is the magnitude of any grade, the presence
of headwinds (particularly in nondrafting races), and air density (a factor in
aerodynamics and a function of the local temperature, humidity, and altitude).

Pedaling Mechanics
Perhaps second only to bicycle–rider geometry and fit, the term pedaling
mechanics seems synonymous with cycling biomechanics. An interest in what
happens at this critical interface between the cyclist and the machine dates
back more than 100 years, when Sharp (1896) developed the first instrumented
bicycle pedal.1 Since then, instrumented pedals of varying designs have
emerged.2 Our understanding of how a cyclist delivers energy to the bicycle
has benefitted tremendously from these developments.
Pedal forces are typically described in their component terms. A normal
force acts perpendicular to the surface of the pedal, and a tangential force acts
parallel to the pedal surface (in the fore–aft direction). If we know the ori-
entation of the pedal relative to the bicycle crank, these components can be
mathematically resolved into effective and ineffective components. The effective
74 } Broker and Langlais

component drives the crank around the pedal circle, always acting perpen-
dicular to the bicycle crank. The ineffective component performs no useful
work, because it acts parallel to the crank at all times. Some researchers use
these terms to describe pedaling quality.3, 4
Typical normal and tangential force profiles measured during pedaling are
illustrated in figure 5.1. The effective component associated with these forces
is shown in figure 5.2. Note here that crank torque is simply the product of
effective force and crank arm length; thus, the effective force profile shown
in figure 5.2 mirrors the crank torque profile, derived from each pedal.
As indicated, effective force (and thus crank torque) peaks roughly when the
crank is level and forward in the pedal stroke, about 100 degrees past top dead center.

Normal and tangential pedal loads

Normal pedal force


Tangential pedal force
Force (N)

Crank angle (degrees)

„„Figure 5.1 Average normal and tangential components of pedal loading for
cyclists riding at 350 watts andE5643/Friel/Fig.
90 revolutions per minute (rpm) (n = 17). Crank angles
5.1/448560/TimB/R1
of 0 and 360 degrees represent pedal top dead center positions. The left half of the
graph represents the downstroke (0 to 180 degrees), and the right half of the graph
represents the upstroke (180 to 360 degrees).

Effective force
Force (N)

Crank angle (degrees)

„„Figure 5.2 The effective component of pedal loading derived from the normal and
tangential components shownE5643/Friel/Fig.
in figure 5.1,5.2/448561/TimB/R1
again for the 350 W, 90 rpm condition.
This effective component always acts perpendicular to the crank, providing the force
needed to propel the bicycle forward.
Cycling Biomechanics for Triathlon | 75

Effective force is negative during the upstroke, because the force on the
pedal acts downward as the pedal rises. Note as well the low effective force
magnitudes at the top and bottom of the pedal cycle.
A more effective way to visualize how pedal forces vary during the pedal
cycle is to present the data in clock diagram format (figure 5.3). Pedal forces
are represented as force vectors (arrows). Their lengths are proportional to
their magnitudes, and their directions are consistent with the manner in
which they are applied. Clock diagrams dramatically illustrate why effective
forces peak midway through the downstroke; force magnitudes are high and
the forces are directed nearly perpendicular to the crank.
The clock diagram also exposes regions of the pedal stroke where effective
force is low. For example, effective forces and crank torque are low through
the top of the pedal stroke, predominantly because of the low magnitude of
forces during this phase of the cycle.
By contrast, low effective forces, and thus low crank torque, are present
through the bottom of the pedal stroke, not because of low force magnitudes
but because of the orientation of the force vectors, which are largely parallel
to the crank. Lastly, the negative effective force during the upstroke is clearly
explained in the clock diagram, because forces are applied to the pedal in the
opposite direction of crank rotation.
The effective force profile shown in figure 5.2 tells only half the story.
Because the two pedals on a bicycle are connected to each other through
the bottom bracket, net crank torque develops from the combined effect of the
forces applied to each pedal.

„„Figure 5.3 Clock diagram for cycling showing the magnitude and direction of
forces applied to the pedal, as well as pedal orientation. The major torque applied to
the crank during the downstroke is evident,
E5643/Friel/Fig. and forces through the bottom of the
5.3/448562/TimB/R1
pedal cycle are large but contribute little to crank rotation.
76 } Broker and Langlais

Net crank torque is shown in figure 5.4, highlighting substantial variation in


instantaneous torque throughout the pedal cycle. Two distinct peaks in crank
torque are evident, generated during the right- and left-pedal downstrokes.
The left-pedal downstroke peak is larger than the other for this particular
cyclist. Peak net torque is usually less than single-leg peak torque, because
as the downstroke pedal receives its greatest positive torque, the upstroke
pedal simultaneously experiences negative torque. (In essence, loading of the
downstroke pedal acts partially to lift the upstroke leg.)
Also evident in figure 5.4 are distinct minima, occurring just after the
pedals pass through their respective vertical positions. These minima are
explained by low effective forces on both pedals when the cranks are vertical.
The difference in peak torque for the right- and left-pedal downstrokes
shown in figure 5.4 deserves comment. Recall that the net torque pattern
shown represents the combined effects of both pedals. When one of the net
crank torque peaks is higher than the other, the explanation could be that the
effective force on the downstroke pedal (for the left pedal shown in figure
5.4) is larger.
That would be the traditional interpretation. Alternatively, the peak
torque difference may be explained by different force magnitudes during
the upstroke. Thus, the asymmetry could be an upstroke imbalance rather
than a downstroke imbalance. This analysis of asymmetry, although seem-
ing trivial, exposes a common pitfall regarding commercially available
devices that claim the ability to assess pedaling asymmetry. The rider
represented in figure 5.4, in fact, generates nearly identical downstroke effec-
tive force magnitudes yet is heavier during the upstroke on his right side.

Net crank torque


Torque (Nm)

Crank angle (degrees)

„„Figure 5.4 Total crank torque developed from the combined effects of right- and
E5643/Friel/Fig.
left-pedal loading for a cyclist demonstrating 5.4/448563/TimB/R2-kh
pedaling asymmetry. The first peak in
crank torque occurs during the right-pedal downstroke; the second occurs during the
left-pedal downstroke. A first glance suggests a downstroke asymmetry; the left leg
appears to drive the crank more powerfully than the right. But the real asymmetry is
an upstroke effect; the right leg is more active during its upstroke.
Cycling Biomechanics for Triathlon | 77

The source of a net crank torque asymmetry cannot be isolated by measuring


only net crank torques or externally measuring bicycle power output.

Crank Torque and Crank Power


As discussed earlier, pedaling power is mathematically equal to the product
of the net crank torque and instantaneous crank angular velocity. When
instrumented pedals are used to assess cycling biomechanics, the nature
of energy delivered to the pedals can be quantified. Average instantaneous
total power at the pedals for 17 road cyclists riding at 350 W and 90 rpm is
depicted in figure 5.5. Notably, power fluctuates between roughly 110 W and
600 W within a single pedal revolution. These fluctuations characterize the
nature of energy delivery to the pedals.
Interestingly, the magnitude of the oscillations in the instantaneous power
(figure 5.5) qualitatively describes how smooth a given rider pedals. A rider
who develops most of her crank torque during the downstroke of the pedal
cycle will exhibit greater oscillations in instantaneous power than a rider who
distributes crank torque more evenly throughout the pedal circle, particularly
the top and bottom regions of the pedal cycle.

Effective and Ineffective Forces:


Relations With Natural Pedaling Dynamics
In the 1980s the terms effective forces and ineffective forces were introduced by
researchers to describe pedaling quality quantitatively.3, 4 The implication
was that ineffective forces, those parallel to the crank, do not contribute to
the pedaling motion and therefore represent wasted energy. Recent research,
however, suggests that the use of these terms to describe pedaling effective-
ness may not be appropriate.

Total crank power


Power (W)

Crank angle (degrees)

„„Figure 5.5 Average instantaneous crank power associated with riding at 350 W,
90 rpm (n = 17). Notice how crank power oscillates
E5643/Friel/Fig. substantially around the average
5.5/448564/TimB/R1
power level of 350 W. The two minima in total power occur when the cranks are
nearly vertical.
78 } Broker and Langlais

Consider a cyclist sitting on a bicycle with his feet in the pedals but the
cranks are not rotating. Both right and left pedals experience forces from the
weight of the legs acting downward. If the pedals are placed at the top and
bottom of the pedal cycle, these forces would act parallel to the cranks and
would not contribute to crank torque development. The question arises: Are
these forces wasted? Because these forces require no muscular work, being
developed solely by gravitational effects, the answer must be no. In this case,
the pedals and crank serve to maintain the position of the limbs.
Now let’s consider the normal pedaling condition, involving rotation of
the crank as depicted in the clock diagram of figure 5.3. As indicated, large
downward forces are seen through the bottom of the pedal stroke. These
downward forces, often labeled ineffective in that they act to lengthen the
crank and do not assist in rotating the crank, are dominantly generated by the
interaction between the pedal and the leg. The pedal supports the leg under
the influence of gravity and assists in changing the movement direction of
the leg from downward to upward.
Critically, if a cyclist were asked to pedal in such a way that these forces
were minimized or eliminated, significant muscular work would be required,
yet no additional pedaling power would result. These forces acting downward
along the axis of the cranks are highly functional and free of metabolic energy
expenditure.
Kautz and Hull reported a method to separate the naturally occurring
components of pedal loading from the total pedal loads.5 The natural compo-
nent of pedal loading is derived from gravitational and inertial effects acting
on the leg and is thus nonmuscular. When pedal forces are separated into
elements derived from muscular contractions versus those developed from
gravitational and inertial effects, fascinating observations emerge. First, a
significant portion of the measured downward loading at the bottom of the
pedal cycle (figure 5.5) is indeed caused by gravitational and inertial effects,
not muscular actions. During the upstroke, the measured forces seen acting
downward on the pedal are also gravity and inertia based. Thus, the cyclist
is not actively pushing down as the pedal rises.
By contrast, the muscular component of total loading represents more of
what cyclists perceive. Muscular actions drive the pedal downward during the
first half of the pedal cycle, rearward through the bottom of the pedal stroke,
and upward during the upstroke. Despite these muscle actions, however, the
gravitational and inertial-based pedal forces produce the seemingly ineffec-
tive pedal forces. These findings should alter our interpretation of pedaling
technique and pedaling skills training.
In summary, effective pedaling mechanics requires the appropriate recruit-
ment of muscles to generate forces on the pedal acting dominantly perpen-
dicular to the crank. These muscular-based forces will act in concert with
forces arising from gravitational and inertial effects, such that power delivery
to the pedals oscillates dramatically, predictably, and naturally but ultimately
feels and looks smooth.
Cycling Biomechanics for Triathlon | 79

Pedaling Mechanics: Triathletes Compared With


Cyclists of Other Disciplines
Since 1992 more than 150 pedaling mechanics evaluations were performed
at the U.S. Olympic Training Center in Colorado Springs, Colorado.2 These
evaluations included more than 125 cyclists from road, track sprint, track
endurance (pursuit), mountain, and triathlon. Many of the riders were national
team riders at the peak of their cycling careers. The following important
observations were made:2
• Cyclists move their feet in circles during pedaling, but applied forces in
no way appear circular. The clock diagram discussed earlier (figure 5.3)
clearly demonstrates how forces vary during the pedal cycle.
• Cyclists of all abilities generate counterproductive forces on the pedal
during the upstroke during steady-state cycling. Experienced cyclists
generate smaller counterproductive forces during the upstroke because
of active lifting of the leg during this phase. An effective way to visual-
ize the upstroke during cycling is to recognize that the proficient cyclist
lifts her leg actively during the upstroke, but not as fast as the pedal is
rising. The pedal helps lift the leg during the upstroke.
• Counterproductive forces during the upstroke are larger in magnitude
at high cadences and are less negative at high power outputs for a given
cadence. They may even be positive (productive) during sprinting and
climbing.
To compare cyclists across disciplines, every rider tested at the Olympic
Training Center during a 10-year period was evaluated under the same
cycling condition, 90 rpm and 250 W. Many riders were also evaluated at 90
rpm and 350 W.2
Comparison of the disciplines revealed sprinters to be the mashers of the
cycling community, generating large downstroke peak effective forces with
the largest peak-to-peak power oscillations. Road and mountain bike riders
performed greater work in the top quadrant than other riders did, notably at
higher power conditions. Peak-to-peak power oscillations were lowest with
the mountain and road specialists.
Finally, triathletes pedaled similar to the road specialists, but they were
often less proficient at generating work through the top quadrant of the pedal
stroke. Further, triathletes, when faced with an increase in cycling power
(from 250 W to 350 W), generally demonstrated increased downstroke effec-
tive forces, compared with road riders who used more of the pedal cycle to
generate power. The triathletes, although internationally competitive, were
relatively new to competitive cycling, often coming from swimming or run-
ning backgrounds.2 Their limited exposure to competitive cycling and perhaps
an interaction with their running and swimming muscular development may
explain the differences.
80 } Broker and Langlais

Aerodynamics, Rider Position,


and Pedaling Mechanics
Bicycles and cycling accessories are responsible for roughly 20 to 35 percent
of the total aerodynamic drag on a cyclist; the remaining 65 to 80 percent is
attributed to the cyclist’s body. Wind tunnel tests and controlled experiments
using on-board power measuring devices have invariably shown that a cyclist
can dramatically reduce aerodynamic drag by assuming a flat-back, tucked-
head posture with the hands and arms atop aero bars held nearly parallel to
the bicycle top tube (figure 5.6).

© AP Photo/Daniel Roland

„„Figure 5.6 A triathlete in an aerodynamic, flat-back position. Note the horizontal


position of the arms atop aero bars.
Cycling Biomechanics for Triathlon | 81

A simple real-world example highlights the importance of aerodynamic


positioning. Wind tunnel experiments performed in the mid-1990s quantified
aerodynamic drag across rider positions centered on the time-trial and pur-
suit cycling postures.6 One U.S. national team cyclist achieved a 0.28-pound
(1.25 N) decrease in aerodynamic drag at 22.5 mph (10 m/sec), by tucking his
head, narrowing his shoulders and arms atop the aero bars, and dropping
his handlebars less than 1 inch (2.5 cm). This 0.28-pound (127 g) reduction in
drag translates into a 1.26 mph (0.56 m/sec) increase in cycling velocity for
the same power output, providing an advantage of 3 minutes and 30 seconds
over a 40-kilometer race!6
As many triathletes recognize, the flat-back, streamlined posture can wreak
havoc on the ability to pedal effectively. A triathlete’s pedaling mechanics
are shown in figure 5.7, representing his on-the-hoods position as well as his
time-trial position (on aero bars). As the triathlete rotated his torso into an
aerodynamic orientation, the effective force generated on the pedal through
the top of the pedal stroke decreased. Further, effective force magnitudes
during the upstroke (between 180 and 360 degrees in figure 5.7) were more
negative. To offset the losses in torque at the top of the pedal stroke and
through the upstroke, greater effective force (greater torque) was needed
during the downstroke (surrounding 90 degrees in figure 5.7). In essence,
the flat-back position turned this triathlete into a masher.
The musculoskeletal explanation for this phenomenon is that the hip
flexors, responsible for torque generation through the top of the pedal cycle
and during the upstroke, are shortened when the pelvis is rotated forward
to support the flat-back position. Muscles lose some of their force generating
capacity at shortened lengths. Further, in some cyclists, physical interactions
between the thighs and the torso through the top of the pedal stroke can
interfere with power generation.

Effective force, hoods, and aero position

Hoods
Aero bars
Force (N)

Crank angle (degrees)

„„Figure 5.7 Effective force profile for a triathlete exhibiting sensitivity to aerodynamic
positioning. When the rider rotated forward onto aero bars his effective force (and thus
E5643/Friel/Fig. 5.7/448566/TimB/R2-kh
crank torque) through the top of the pedal cycle decreased (0 and 360 degrees), his
effective force during the upstroke became more negative (225 to 315 degrees), and
he needed to apply greater force during the downstroke to make up for the pedaling
weakness.
82 } Broker and Langlais

For triathletes demonstrating sensitivity to aerodynamic positioning an


effective method to recapture pedaling effectiveness involves changing the
geometry of the bicycle. Simply, bicycle geometry with a steeper seat tube
facilitates recapture of the normal hip angle range of motion during pedaling.
This steeper geometry, so intimately linked with present triathlon cycling, is
fundamentally a result of musculoskeletal-based needs, balancing aerody-
namic concerns and muscular capabilities. For more information concern-
ing triathlon-specific bicycle geometry, bike fit, and related aerodynamic
enhancements, see chapter 8.

Drills to Improve Pedaling


Serious cyclists, coaches, and trainers recognize the importance of working
on pedaling technique. But what elements of pedaling technique should
be emphasized, in light of our knowledge concerning the biomechanics of
pedaling? And what methods work best for refining pedaling technique?
The clock diagram in figure 5.3 shows that the top and bottom of the pedal
cycle represent relative dead spots where little torque is developed. A masher
produces almost all of her power during the downstroke, accomplishing little
if anything through the top and bottom. By contrast, the smooth-pedaling
cyclist applies forces at the top and bottom of the pedal cycle partially in the
direction of pedal motion. The resulting total power curve has no dead spots,
or regions where power delivery falls to near zero.
Early-season training should incorporate methods to promote the develop-
ment of these torque patterns. These methods include spinning at a relatively
high cadence on rollers, focusing on the elimination of the fluctuating power
characteristics. One mental image would be to drive the foot into the front of
the shoe through the top of the pedal cycle and to scrape the shoe rearward
through the bottom.
Emphasis on single-limb cycling during indoor and outdoor riding encour-
ages the distribution of pedaling work over a greater part of the pedal cycle.
Single-leg riding on a fixed indoor trainer is one of the more uncomfortable
methods to improve pedaling technique but may be the most effective. Plac-
ing one leg on a box or in a fixed pedal mount allows cyclists to work on right
and left legs separately. Productive torque generation through the top and
bottom of the pedal cycle should be emphasized.
Pedaling technique can also be practiced on the road. Emphasizing crank
torque through the top and bottom of the pedal cycle is recommended. When
cyclists first attempt to pedal in this manner, they immediately notice bicycle
acceleration because they are delivering power to the cranks and regions of
the pedal cycle that were previously power passive. An important lesson
gained from this acceleration is that cyclists do not have to push harder to
go faster. By developing the ability to apply power to the cranks over more
of the pedal cycle, pedaling is smoother, muscular resources are developed
to deliver more power when needed, and cycling performance can improve.
Cycling Biomechanics for Triathlon | 83

Finally, triathletes may suffer from their multidiscipline sport training


when attempting to refine pedaling technique. In a comparison between
highly trained cyclists and highly trained triathletes, Chapman and col-
leagues reported clear differences in muscle recruitment between cyclists and
triathletes during cycling.7 These researchers reported recruitment pattern
differences between trained cyclists and triathletes, pedaling at common
work rates, in all five of the tested lower-extremity muscles (tibialis anterior,
tibialis posterior, peroneus longus, gastrocnemius lateralis, and soleus).
Differences included greater variation between athletes, greater variation
between pedal cycles, more extensive and variable muscle coactivation, and
greater electromyographic amplitudes in triathletes. The results indicate that
control of muscle recruitment is less developed in triathletes than in experi-
enced cyclists, suggesting that multidiscipline training might interfere with
adaptation of the neuromuscular system to cycling. More research is needed
to determine whether multidiscipline interference represents an unavoidable
and permanent aspect of triathlon training and racing, or whether counter-
measures can be developed to reverse or minimize the deleterious effects.

Bike-to-Run Transition
A unique challenge for the triathlete is the need to run, and run effectively,
immediately after finishing the bike leg. The feeling of heavy legs reported
by many triathletes has drawn the attention of physiologists, neuromuscular
control specialists, and biomechanists. Scientific studies focusing on this
transition have grown in number in recent years, designed to explore and
understand the phenomenon, and possibly direct mitigating measures. The
heavy-leg phenomenon is thought not only to affect performance (reduced
running velocity) but also to place triathletes at risk of lower-extremity inju-
ries. Although a section of chapter 6 in this book is devoted to the bike-to-run
transition, we briefly address this transition here from a cycling perspective.
From a neuromuscular standpoint, cycling and running differ in a couple of
important respects. First, and most obvious, running involves short-duration
impacts that place loads exceeding two times body weight on the skeletal
system.8 Cycling, by contrast, involves nonimpact, smoothly fluctuating forces
that rarely exceed body weight. Cycling is clearly nonimpact, particularly
when compared with running.
Cycling is also generally represented muscularly as a concentric activity, in
which the major muscles dominantly shorten in performing the mechanical
work of pedaling. Eccentric muscle actions, in which muscles are active yet
lengthened, occur only in small isolated regions of the pedal cycle.9
By contrast, running involves substantial eccentric–concentric cycles, in
which muscles are significantly stretched, while active, and subsequently
shortened. These stretch–shorten cycles have been shown to enhance muscle
force production as well as contraction efficiency (reduced metabolic cost).
84 } Broker and Langlais

Researchers wonder whether a sudden switch between a smooth and largely


concentric activity to an impact-heavy, eccentric–concentric activity might
cause problems.
Chapman and colleagues at the University of Queensland and the Aus-
tralian Institute of Sport examined joint kinematics (movement patterns),
muscular activity, and metabolic expenditure in experienced and elite tri-
athletes performing a run following a strenuous bout of cycling, compared
with running without a preceding cycling effort.10, 11, 12 These researchers
generally report transient yet significant alterations in neuromuscular control
and energy expenditure, absent significant changes in running mechanics
(kinematics). Notably, clear intersubject differences in response to prerunning
cycling emerged. Some triathletes (including elite) demonstrate minimal or no
effect of cycling on subsequent running muscular activity patterns, whereas
other triathletes exhibit noticeable alterations.10
Sutter and colleagues evaluated the effect of cycling intensity on running
economy and vertical running center of mass excursion (VCOM) in well-
trained triathletes.13 If altering cycling intensity late in the cycling leg of a
triathlon influences running performance, intensity management might be
useful in pacing during . races. Unfortunately, in nine triathletes (six males
and three females, VO2max 63.5 +/− 7.2) running economy and VCOM were
unaffected by prerun cycling intensity. The only measures affected by cycling
intensity were ventilation rates and blood lactate levels.
A recent study with important practical implications introduced plyometric
training into the normal triathlon training regimen, attempting to elicit neu-
romuscular adaptations that might mitigate the cycle−run transition effects.14
Eight elite triathletes who demonstrated altered joint mechanics and muscle
recruitment patterns during running following moderate to intense cycling
were identified. In these eight triathletes, four were subjected to controlled
plyometric training targeting lower-extremity explosiveness, in addition to
their normal triathlon training. The remaining four triathletes trained nor-
mally, without plyometric elements.
Follow-up testing after 8 weeks of training found that running mechan-
ics and muscular recruitment patterns were not altered by prerun cycling
in the triathletes who incorporated plyometrics into their training regimes.
In the four control triathletes, two maintained altered running mechanics
and muscle recruitment in response to cycling. Although the small sample
limits the extrapolation of these findings to triathletes as a whole, the find-
ings, in elite-level triathletes, suggest that plyometric training may be advised
to mitigate the transition-mediated running performance. As is the case in
many areas of triathlon performance science, future research will continue
to unravel this interesting aspect of triathlon racing.
Cycling Biomechanics for Triathlon | 85

Conclusion
In this chapter the focus has been cycling biomechanics. We first focused
on the energy (power) demands on the rider and bicycle system. Triathletes
should be attentive to the sources of drag while riding: rolling resistance,
grade (or hill climbing), bicycle acceleration, and aerodynamic forces. Of
these, managing aerodynamic drag is the most important. A flat-back, time-
trial position on the bicycle should be developed if possible. This positioning
traditionally requires a bicycle with steeper seat tube geometry to prevent
the pedaling stroke from becoming choppy and less efficient. Concerning
grade, riding uphill dramatically increases cycling power, although less so
when bicycles (and triathletes) are lightweight. Power meters are useful in
providing feedback to the rider during climbing and helpful in modulating
physiological effort. This topic will be addressed further in chapter 8.
We then turned our attention to how pedal forces, generated to manage
energy demands, fluctuate in response to both muscular and nonmuscular
influences. Research indicates that triathletes, like their traditional cycling
partners, should abandon the pursuit of true circular pedaling because the
legs and cranks naturally capitalize on the noncircular dynamics of the
coupled legs and bicycle cranks. Instead, triathletes should work to develop
the ability to power the cranks through the top and bottom of the pedal cycle,
specifically when higher cycling power is desired. Indoor one-legged pedaling
drills, high-cadence riding on rollers, and a focus on delivering energy to the
cranks through the top and bottom of the pedal cycle are effective methods
of improving pedaling technique.
Finally, as a logical step into the next chapter, recent findings were presented
concerning the cycle-to-run transition. The triathlete, like the road-racing
time-trial specialist, seeks a unique solution to the optimal human–machine
integration challenge. Aerodynamic positioning without compromised pedal-
ing technique is an achievable goal, but for the triathlete, running performance
off the bicycle must be considered. Recent research suggests that plyometric
training may be useful in mitigating both the perceived and measured dif-
ficulties in running following cycling. Future research will further define and
refine our understanding of the triathlete as a cyclist and even characterize
the triathlete who runs fast, right off the bike.
This page intentionally left blank.
CHAPTER
6
Running Biomechanics
for Triathlon
—George M. Dallam, PhD

R unning form is potentially an important part of successful performance


in triathlon.1, 2, 3, 4, 5, 6 The pattern of movement used when running is
referred to scientifically as running biomechanics. This pattern represents
interplay between the biological lever systems (bones, joints, muscles, and
nervous system) and the physical principles that guide effective movement.

Importance of Running Economy


In the scientific study of running, a long-held notion is that the movement
pattern that people adopt naturally is the most efficient way to run and
consequently should not be changed. This view evolved from early stud-
ies examining the effect of making changes in the stride length and stride
frequency on running economy.1 The early studies often found the same
thing: Changing the stride rate and length away from those used freely (used
naturally) increases the oxygen cost of running over the short term, whereas
typical training without changing technique has no effect on oxygen cost
over the short term.7
This oxygen cost is referred to as running economy. Because other studies
have implicated running economy as one of the major factors responsible for
running performance,1, 5, 8 the assumption is often made that increasing oxygen
cost will make running harder and hinder performance. Consequently, many
scientists who study running economy recommend that runners not attempt
to change how they run naturally.
But some running and triathlon coaches attempt to create new and
improved running technique, or running biomechanics, particularly when
runners appear to overstride, to take steps that appear longer than optimal.
Additional coaching concerns include taking steps that are too short, allowing
body posture to slump during each foot support, creating excessive vertical
movements with each step, twisting the upper body and crossing over the

87
88 } Dallam

arms excessively with each step, leaning forward too much or too little, allow-
ing the pelvis to dip with each step, forcing the low back to arch excessively,
and moving the head about unnecessarily. Of course, coaches commonly
measure the outcome of a change in the running movement more directly than
scientists do. Rather than measuring energy cost, coaches simply measure
performance—how fast the athlete can run any given distance. Unfortunately,
few of the typical form changes recommended by coaches have been directly
studied scientifically, so they remain theoretical at best.
Consequently, the notion of trying to improve how someone runs is conten-
tious and widely debated. Numerous viewpoints and approaches exist. But
some useful scientific evidence does exist concerning various theories about
the development of running technique. Among those the pose method of run-
ning has been studied scientifically and appears to hold some promise.2, 9, 10, 11
This approach to running technique focuses on maximizing the use of
gravity for forward propulsion by creating running movements that more
closely resemble the movements of a wheel. More recently, new studies have
examined the concept of running without shoes as a mechanism by which
running technique may change naturally.12, 13 Such an approach is widely
thought to allow the hardwired running motor pattern that we are born with
to reemerge in those who have altered their running movements by wearing
heavily cushioned shoes that inhibit the normal reflexive responses associated
with natural running biomechanics. There are generally three widely accepted
reasons to change how someone moves in an athletic activity. These include
• improving efficiency (making it easier to move),
• improving performance capability (allowing a person to move faster,
farther, or higher), and
• reducing risk of injury caused by repeating the movement.
The traditional scientific studies of changing running technique primarily
address the first reason, and indirectly at that because they measure running
economy as the outcome variable. More recently, science has begun to address
the other two reasons by examining the influence of changing running
technique on both performance and potential injury risk factors as well.9, 11

Mechanical Efficiency and Performance


The mechanical efficiency of movement refers to the energy cost required
to perform a given level of mechanical work output. For instance, a more
efficient runner creates the same work using less energy than a runner who
is less efficient. Measuring this concept in running is complicated by the fact
that measuring the mechanical work of running (to calculate power output,
for instance) is problematic.14
The simple work formula (force times distance) is hard to apply because
the muscle forces created in running are used to store energy and return it,
Running Biomechanics for Triathlon | 89

rather than simply to generate movement of the limbs directly. Running stores
and returns substantial amounts of the energy used elastically, like a rubber
band, with each step. How much energy is conserved in this way is not now
directly measureable, not known, and probably varies from runner to runner.14
Consequently, although measuring how fast someone runs from point A
to point B is easy, consistently measuring the amount of work required to
do so is not possible, although numerous approaches and formulas exist.14, 15
Consequently, sport scientists typically measure running energy consump-
tion only by the oxygen cost of running, making the assumption that work
will be constant when two individuals are running at a given speed. They
next make inferences about mechanical efficiency (work kilocalories divided
by total energy consumption kilocalories) from there.
A common assumption is that when two runners run at the same speed
and one uses less oxygen to do so, that runner is more mechanically efficient.
But another possibility is that the runner using less oxygen is doing less
mechanical work and is not more efficient when comparing work to oxygen
cost. For instance, this difference in oxygen cost may result from factors such
an anthropometrics, improved neuromuscular activation patterns, or training
status5 and not be related to running mechanics at all.

Improving Running Economy


The most definitive statement that can be made when evaluating someone’s
running economy is that it will improve over long periods through simple
repetition if technique remains constant. This result happens through the
natural process of decreasing coactivation of agonist and antagonist muscle
groups against each other in a process called reciprocal inhibition. When
learning a new motor pattern the early conscious regulation that we create
through the cerebral cortex is gradually replaced by autonomic regulation
in the cerebellum and through spinal reflexes as overlearning occurs. This
process takes considerable time, especially when a runner is replacing an
already well-developed motor pattern with a different but similar one.
The result of reduced coactivation is that the same movement becomes
easier to perform because the opposing muscles around each joint work
against each other less to produce the same outcome in force. Gradual inhi-
bition of cocontraction is the natural outcome of many millions of consistent
repetitions. Although this concept has not been studied directly in running,
the existence of increased oxygen cost during walking because of excessive
cocontraction has been demonstrated in children with cerebral palsy.16
Of course, this concept helps to explain the large body of research that
illustrates that a significant change in running technique results in increased
oxygen costs over the typical period of a research study, 6 to 12 weeks. We
likely increase coactivation of opposing muscles (lose coordination) when
learning a new version of a motor skill, even if the new skill may be more
effective mechanically in producing the performance outcome that we want.
90 } Dallam

An example might be found in learning to pedal at higher cadence. Doing


so initially requires our attention and is usually more energy intensive, yet it
results nearly immediately in a reduced fatigue level in our working muscles.
As a result, the only effective measure of the utility of changing running tech-
nique over short time frames is found by measuring running performance
itself rather than by measuring the energy cost of running. Economy is then
improved primarily through accumulated training repetition with constant
biomechanics at the particular velocities targeted, as well as through adapta-
tion to training interventions such as movement-specific strength training,
higher-speed interval training, and increased altitude training.5

Factors That Affect Economy


Considering the aforementioned limitations in studying running economy,
the aspects of technique that are most commonly associated with a reduced
running energy cost at a given speed include reduced braking forces upon
foot strike, reduced vertical oscillations (up and down movements), and lower
peak vertical and horizontal ground reaction forces with each step taken.1, 4, 5, 17
These differences are generally associated with shorter steps, a higher stride
rate, reduced contact or support time, and a resulting reduction in ground
reaction forces with each step at a given speed in comparison with less eco-
nomical runners.
A range of 90 to 100 strides per minute and a vertical oscillation with each
step of 8 to 10 centimeters have been associated with higher-performing run-
ners as well. It is further theorized that an increased ability to use energy elas-
tically to capture the force of gravity and translate it into a horizontal stimulus
should also be associated with lower energy cost,11 although this hasn’t yet
been demonstrated experimentally. Basically, the kinetic energy associated
with the falling body during each step is captured by stretching muscles and
tendons during landing and translated into an optimal projection angle as the
tissues return to their original length and the body’s momentum continues
to carry the center of mass forward, much like the way that a spring works.
Therefore, an optimal body position and the appropriate level of reflexive
muscle contraction during support are critical to creating efficient forward
propulsion. Runners who have poor body position at landing (foot too far
forward and dorsiflexed, knee too straight, upper body slumped) give away
some of the free energy that they could absorb from gravity. Runners who
project themselves at too high or too low an angle sacrifice distance covered
while airborne from the same projecting force.
Figure 6.1 shows the theoretical relationships between the angle of projec-
tion of the body’s center of mass from the ground and the resultant vertical
oscillations and step distances. Imagine a dot on the side of the head moving
up and down with each step. The darkest line represents the optimal (45-
degree) projection angle, which results in the greatest distance covered
because the vertical forces and horizontal forces at push-off perfectly offset
Running Biomechanics for Triathlon | 91

the downward force of gravity to maximize distance traveled in the air. The
optimal projection angle is associated with a vertical oscillation of about 8 to
10 centimeters for typical step lengths. The higher angle of projection reduces
step length because forces are applied in too vertical a manner—a runner
appears to bounce up with each step. The gray line represents forces applied
too horizontally without an adequate vertical component to offset gravity.
Gravity then forces the body down prematurely, resulting in short, rapid steps.
Imagine a baseball thrown at too low or too high an angle. Either trajectory
reduces the distance of the throw for a given force applied. Keep in mind
that these are not on a time scale, just distance. The oscillations represented
by the gray line represent the slowest step rate because of the time spent to
reach peak height. This might represent a stride rate of 60 to 80 strides per
minute. The dotted line associates with a more optimal stride rate of 90 to 100
strides per minute, and the darkest line associates with an excessive stride
rate of more than 100 strides per minute.

Projection angle, vertical oscillation, and step length


Vertical oscillation (cm)

14 V01 (12 cm)


12 V02 (9 cm)
10 V02 (6 cm)
8
6
4
2
0
0 20 40 60 80 100 120 140 160 180 200
Step length (cm)
„„Figure 6.1 Projection angle, vertical oscillation, and step length.
E5643/Friel/Fig. 6.1/448567/TimB/R2-kh

The Pose Method


It has been demonstrated that a global alteration in running technique, using
pose method instruction, could create the more efficient running mechan-
ics just described (reduced vertical oscillation and increased stride rate) but
also resulted in increased oxygen cost when running on a treadmill over
12 weeks.10 The pose method uses a series of drills designed to develop the
ability to land on the ball of the foot just forward of the center of mass in
a springlike body posture, to remove the foot vertically from the ground
while simultaneously shifting the body weight to the opposite and falling
foot, and to lean forward so as to fall into each step, in turn maximizing
the use of gravity to project the body. As the lead researcher in this study, I
was intrigued by the fact that this increased energy cost occurred in spite of
repeated anecdotal reports of improved running performance among the vast
majority of participants in the pose method treatment group. Unfortunately,
performance measures were not directly used in the study.
Following completion of the study I formally questioned the treatment
group using a Likert scale survey that included questions about performance
92 } Dallam

improvements, chronic injury effects, and desire to continue using the altered
technique. Six of seven treatment group subjects reported improvements in
performance, positive or no effects on existing chronic injuries, and a desire
to continue using the new technique despite knowledge of the results of the
study concerning running economy.
A later pose method experimental study that examined both running
economy while running overland and running performance at 1.5 miles (2.4
km) found no loss in running economy and a 24-second improvement in
1.5-mile running time in the pose method treatment group in comparison
with no meaningful change in the control group.
This result occurred after just 1 week of instruction in the pose method
and concurrently with measureable changes to running kinematics in line
with those discussed previously.2 Considering that the loss of economy in the
first study may have been at least partially a result of testing on the tread-
mill, which has been shown to alter running mechanics in comparison with
testing overland,18 and which created a novel environment for the subjects to
attempt to regulate a newly learned technique, the overall potential for effec-
tive change in running technique and the resulting improvement in running
performance through pose method instruction can be viewed optimistically.

Barefoot Running and Minimalist Shoes


Along these same lines of thinking the science studying barefoot, or unshod,
running clearly demonstrates the immediate potential to change running
biomechanics13, 19 and reduce oxygen cost by simply running without shoes.20
When running without shoes, people immediately adopt19, 20 and eventually
adapt13, 21 running mechanics that parallel those described previously for
running with lower energy cost. They also experience an immediate reduc-
tion in oxygen cost.
Barefoot running results in a shorter step and higher stride rate, a ball-of-
the-foot landing (see figure 6.2), shorter support time, and reduced ground
reaction forces in comparison with shod running at a given speed.13, 19 In
addition, the use of glovelike footwear induces similar immediate changes
in running biomechanics.19 Unfortunately, no published experimental trials
studied the immediate effect of barefoot running changes on performance.
This approach may offer the theoretical advantage, in comparison to more
conventional instruction, of allowing the athlete to adopt an already existing,
and some would argue innate, motor pattern, thereby avoiding the coactiva-
tion issues and increase in oxygen cost described earlier.
Extreme caution, however, must be observed to avoid injury, as often occurs
when a previously habitually shod athlete attempts to adapt too quickly to
the conditions of running without shoes. This caution may be particularly
applicable to those who use glovelike shoes rather than simply run without
shoes, because the shoes remove significant cushioning yet prevent the
normal limitation to excessive barefoot running, which is the accumulation
Running Biomechanics for Triathlon | 93

„„Figure 6.2 Barefoot runner landing on the ball of the foot.

of external skin abrasion and soreness. Wearing glovelike shoes may tempt
athletes to progress distance in a new running style too quickly and has been
associated with case study reports of metatarsal injury.22
In practical terms, barefoot or minimalist shoe running offers potential
as a means of practicing and reinforcing running technique when risk is
minimized.23 To do so, a person can attempt such running in conditions
such as the artificial turf now used on many athletic fields, slowly progress
the running volume while barefoot or in minimalist shoes, and gradually
transition to lighter footwear in normal running conditions. This approach
should naturally reinforce improved technique while minimizing risk for
overuse injuries caused by the adjustment process.
Following is a suggested progression in shoe type, under the assump-
tion that each shoe will be worn to its normal wear duration during normal
94 } Dallam

running conditions. Such a progression should occur over a period of years


rather than weeks or months. Barefoot running, technique instruction, and
minimalist shoe running can then be used to learn and reinforce the tech-
niques applied during regular shoe running. This selection of shoes assumes
a change in running technique to the ball-of-the-foot striking pattern asso-
ciated with barefoot running and anthropometrics that are suitable to the
new technique. At the first sign of problems the runner should retreat to a
previously successful shoe and then consider providing more time for adap-
tation.
1. Start with the current supportive or heavyweight training shoe, ideally
using a version with a minimum heel-to-forefoot height differential.
2. Move to a more neutral training shoe with no overt features designed
to control foot motion.
3. Move to a lighter weight neutral shoe designed for performance training.
4. Move to a racing flat or a well-cushioned minimalist-style shoe such as
the Nike Free.
5. Move to minimalist-style shoe with less cushion.
6. Move to bare feet if desired. If planning to do so, gradually increase
walking time in bare feet.

Triathlete Versus Runner Efficiency


Although triathletes are less skilled in both swimming and cycling than their
comparably trained counterparts in those individual sports,40, 41 the same is
apparently not true for running.42 The first observations mentioned raise the
concern that cross-training, meaning simultaneous training for more than
one sport in a common timeframe, may negatively influence neuromuscular
development and skill level in a single sport. But the greater likelihood is
simply that multisport athletes possess less ability than their elite single-sport
counterparts in the first place.
This result likely occurs as less-than-elite-level swimmers, cyclists, and
runners take up multisport at a later stage in their athletic development while
their more skilled counterparts remain committed to the individual sports of
swimming, cycling, and running. Any comparison between elite single-sport
athletes and multisport athletes will then naturally include a wider variation
in skill level among the multisport athletes.
Because most triathletes come to the sport from a competitive running
background, neuromuscular coordination (or skill) levels is likely more
comparable between triathletes and runners completing similar volumes of
run training.42 Finally, running is the most natural of the three disciplines
that typically make up triathlon; consequently, the role of learning in dif-
ferentiating running skill development is greatly reduced between runners
and triathletes. Despite these explanations for the differences in skill between
Running Biomechanics for Triathlon | 95

single-sport and multisport athletes, multisport athletes can still improve


tremendously by improving their running skill.

Injury Prevention
The rate of musculoskeletal injury that results in lost training time in com-
petitive triathletes at some point in the training and competition process has
been reported to be as high as 75 percent.24 Unfortunately, studies of running
and triathlon injury risk factors invariably fail to examine general movement
technique factors that may relate to injury such as foot strike pattern, vertical
oscillation, or ground reaction force patterns.
Some studies do address what can be more accurately viewed as anatomical
movement limitations that seem to influence injury potential.25 Among these
are limited dorsiflexion capability of the ankle (see figure 6.3), hip immobility
in rotation, leg length differences, excessive joint laxity, excessively high or
low arches,25 and more recently pelvic positioning during support (see figure
6.4).26 Such factors, although certainly elements of the overall biomechanical
profile, do not represent conscious choices in technique. Rather, they are
inherent limitations of an individual’s anatomy and should be addressed
through physical therapy–based methodologies such as functional training.

„„Figure 6.3 Limited ankle dorsiflexion. „„Figure 6.4 Pelvic drop during support.
96 } Dallam

Most studies of running injury risk factors examine training factors such
as total training hours and years of actively competing. A review of such
studies concludes that total training volume and previous injury are the most
important risk factors for injury.27
One scientific review paper has addressed the effects of impact forces in
running.28 The finding is that relatively higher impact forces generally provide
the strongest predictive value for increased injury rate.28 Trials examining
the effects of both pose method instruction2, 10 and a change to barefoot run-
ning13, 20 suggest that these approaches to modifying running technique offer
the promise of reducing injury potential by reducing impact forces. Each
approach has been shown to reduce total ground reaction forces absorbed
with each step taken at a given speed in comparison to heel–toe running9
and running with shoes,12 respectively.
The pose method has also been shown to reduce torque at the knee and
increase torque at the ankle,9 thereby rebalancing total-body force absorption
across the knee and ankle. Although the short-term incidental effect of the
second change is often increased muscle soreness in the calf muscles as the
runner adapts to the new loading pattern, the long-term effect appears to be
a more balanced force distribution pattern that allows a higher total impact
load to be placed on the body while running.
Barefoot running appears to improve muscular response at foot strike,
resulting in more effective ankle and foot pronation and an improvement in
shock absorption.13, 21 This likely results from improved sensory input upon
landing on the ball of the foot without the insulating effect of running shoes.21
Further, the use of modern running shoes has been shown to increase joint
torque at both the knee and hip.29
Note, however, that the recent rush to minimalist footwear and barefoot
running has also produced case study reports of metatarsal stress fracture
in experienced runners.22 This result is not surprising because a lifetime of
walking and running in padded shoes is not likely to prepare someone for
a rapid transition to unshod walking and running. Of course, the pragmatic
solution for a runner who hopes to improve performance and resistance to
injury is likely to be found in a combination of both instruction and footwear
individually applied in relation to both current training level and physical
characteristics. In the same vein, it is reasonable to suggest that runners who
are satisfied with their current performance level and relatively injury free
need not attempt such changes.

Effect of Prior Cycling


on Running Performance
Triathlon competition requires the peculiar need to run following swimming
and cycling in the most typical multisport racing order.18, 30 Therefore, the
athlete runs in a relatively fatigued state.8, 30, 31 Several studies have illustrated
Running Biomechanics for Triathlon | 97

that this running condition results in minor changes in running technique,


as well as losses in economy or performance.8, 30, 31, 32, 33, 34
But an additional study suggests that elite triathletes in World Cup compe-
tition suffer little or no loss in efficiency.35 Three variables are the most likely
to influence this outcome. The first is the legality of drafting other cyclists
during International Triathlon Union competitions such as the World Cup and
World Championship Series events. Many athletes in these events are able
to reduce average power output significantly in comparison with traditional
nondrafting or arbitrary positioning–style cycle segments. Lesser decrements
likely occur in both running efficiency and running performance in the fol-
lowing run segment for those athletes who use their energy wisely during
the cycle segment, and in some cases the cycle segment may even exert a
warm-up effect.18, 30, 35
A second factor is the greater likelihood that elite athletes use specific
combination or brick training to prepare for competition. In this training
approach athletes move back and forth rapidly between training disciplines,
simulating the conditions that they experience in racing. This approach to
training has been shown to improve both transition speed and early running
speed during triathlons.36
A third factor is that high-level cyclists in drafting-legal forms of cycle
racing prefer a higher cadence than that commonly used in nondrafting
cycling triathlons.37 Higher cadences are likely used in elite draft-legal triath-
lon competition as well. Higher-cadence cycling has been shown to increase
stride rate and run speed during early subsequent running,32, 38 and it appears
to result in more economical running than a low-cadence (60 rpm) cycling
approach.38 This last point further illustrates the value of a higher stride
frequency approach to running in the specific context of multisport racing.
Further, one study supports the value of using a triathlon-specific bicycle
with a high seat tube angle. The use of a seat tube with an angle greater
than 76 degrees (see figure 6.5) results in substantial improvements in early

„„Figure 6.5 Seat tube greater than 76 degrees.


98 } Dallam

running speed following cycling as well as both greater overall run speed
and faster combined cycle and run time.39 In theory, using a higher seat tube
angle allows the triathlete to sit more forward on the bike, resulting in a more
relaxed acute hip angle and higher cadence potential while pedaling. Such
changes appear to minimize the negative effects of the cycling position on
subsequent running.

Conclusion
Improved running biomechanics and efficiency appear to offer a significant
avenue by which triathletes can improve overall performance, and those
who desire to improve speed should actively develop them. Current run-
ning technique development approaches such as the pose method and the
use of minimalist shoes or barefoot running show promise as either parallel
or intertwined methods to improve running mechanics and performance,
primarily for those whose performance outcomes are less than satisfactory
or are compromised by injury. These methods should be applied cautiously,
allowing time for successful adaptation. After effective running mechanics
are learned, running economy and efficiency will improve primarily because
of sustained successful running training without injury or further variation
in running mechanics. The specific issue of improved running following
cycling will benefit from high-cadence cycling, the consistent use of combi-
nation cycle–run training, and the use of a triathlon-specific bicycle with a
steep seat tube angle.
Pa r t

III
Environmental Factors
and Equipment
Options
This page intentionally left blank.
CHaPtEr
7
In the Water
—Bruce R. Mason, PhD, and Gina Sacilotto

M any factors affect the swim performance of competitors participating


in triathlon events specifically because of the unpredictable nature
of an open body of water. According to the International Triathlon Union
(ITU), swim distances range from 300 to 4,000 meters. Because of the variety
of uncertainties encountered in an open-water setting, these distances can
become difficult for triathletes to complete. Coaches and triathletes need to
understand several points:
• Differences between swimming in a pool compared with swimming in
an open body of water
• Potential water conditions that triathletes may face during competition
• The use of swimsuits in triathlon events
• The use of wetsuits in triathlon events
• The use of other swim gear in preparation and training for triathlon
competition
This chapter covers these topics and presents ideas for coaches and tri-
athletes to use to make the most of the swim leg of the triathlon.

Pool Versus Open Body of Water


Anyone who has swum in a standard swimming pool and an open body of
water will attest to being exposed to a greater variety of conditions in the
open water. The main difference between swimming pools and open bodies
of water is the lack of control over the environmental factors that triathletes
endure while competing. For example, in pool swimming competitions most
international events are held indoors, therefore almost eliminating weather
elements. In triathlons, however, almost all international events are held
outdoors, therefore exposing competitors to environmental factors such
as extreme air and water temperatures, tide changes, turbulent water flow,
aquatic life, and chemical or bacterial contamination in the water.

101
102 } Mason and Sacilotto

Specifically, within competitive pool swimming, guidelines for the


setup of a competition pool are outlined by the governing body, Federation
Internationale De Natation (FINA). Within a competitive pool event, man-
dated FINA guidelines include the level of lighting, the length and depth
of the pool, the lane widths, lane ropes, starting platforms, lane markings,
specifications for the bulkheads, and the element that affects triathlons the
most, water temperature. In pool swimming competitions, FINA requires
the water temperature to be maintained between 25 and 28 degrees Celsius
(77 to 82 degrees Fahrenheit).
In triathlon events, the ideal temperature that the International Triathlon
Union (ITU) outlines for competition is also between 25 and 28 degrees
Celsius, but this temperature range is impossible to maintain because of
environmental factors, discussed later in the chapter. Realistically, water
temperatures have been known to range between 13 and 32 degrees Celsius
(55 to 90 degrees Fahrenheit).1 Open-water swimming events take place in
an open body of water and more closely resemble a swim leg in a triathlon
than other FINA events. FINA guidelines state that the minimum water
temperature for open-water swimming competitions is 16 degrees Celsius
(60 degrees Fahrenheit). Allowable temperature ranges for competition are
greater in triathlons than in open-water swimming events because triathletes
may possibly use wetsuits in colder water. This issue is discussed later in the
chapter. Similar to open-water events, triathlons have a swim course marked
out with accurately measured and highly visible buoys that are secured to
minimize movement even in severe wave conditions.
Another difference between pools and open bodies of water is how the
race starts. In pool events the start involves a dive from individual blocks
into the athlete’s individual lane. Open-water swimming and the start of a
triathlon involve starting in a small area with all competitors in the same
space of water. The start can be chaotic. Three main start types are involved
in triathlons: in-water starts from a treading-water position, a beach run into
the water, and diving from a pontoon.

In-Water Starts
For in-water starts, the triathletes are positioned in rows, usually book-
ended by two buoys. After the starter declares the beginning of the event,
the triathletes change from a vertical treading-water position to a horizon-
tal position, requiring greater surface area per competitor. This change in
position causes chaos as each athlete tries to make a break and swim in
her own space.
Because all the triathletes are packed closely together while treading
water, the water can become turbulent, possibly causing athletes who are
weak swimmers to be sucked under the water. Novice triathletes should
therefore start from the back and to the side of the starting pack so that
they can learn how to start in the water without the risk of harm. After
In the Water | 103

the triathlete becomes more confident and wants to avoid hindering his
performance time by starting from the back, he should stay to the side of
the pack but move closer to the front. Advanced triathletes tend to start in
the front row toward the middle to create the straightest line possible to
the first course marker.

Beach Starts
When the swim leg of a triathlon is contested in the ocean, competitors often
do a beach run to enter the water. As with an in-water start, all participants
are positioned in rows between two markers, although here the markers are
on the sand. When the starter declares the start of the race, triathletes run
down the beach and into the water. The best time to dive and commence
swimming is when the water is level with naval.2
Because ocean swims involve swimming through the wave breaks, tri-
athletes must learn how to swim out beyond the break. Suggested tech-
niques include diving over the waves, dolphining underneath the waves if
the water is deep enough, or standing side on to the wave so that the water
moves around the body rather than crashes against it, which could cause the
triathlete to be pushed over. In confronting the wave breaks, knowing when
to dive underneath the waves is imperative.
Before the event, triathletes should practice a run in and out to deter-
mine the shore gradient in relation to the water, which can involve gradual
changes in depth as well as an uneven floor. These factors may change how
the triathlete approaches the entry.2 All these start procedures, particularly
with beach starts, may create anxiety, so the triathlete may want to start on
the outer edge or back of the pack and wait a few seconds after the starting
signal to allow more room to swim.3

Diving Starts
The final option for a start in triathlons is a dive from a pontoon into the water.
This type of start procedure is described in the ITU competition guidelines
and therefore can be assumed to be the main form of commencing a triathlon
at the elite level.4 Amateur triathletes seldom have the opportunity to start
from a pontoon in competition.
This starting procedure has two phases. The first is the starting position
selection. Triathletes are usually ordered according to their competition
number, which is usually relative to their ranking in the field. The second
is the starting procedure. One starting official calls, “On your marks,” and
the triathletes move forward to the start line. Then a second starting official
signals the start of the event. To minimize the risk of harm during the actual
event, triathletes should investigate all aspects of the swim leg before the
competition.
104 } Mason and Sacilotto

Water Conditions
In triathlon events, the swim leg is conducted in a variety of settings such as
saltwater oceans and bays and freshwater inland lakes and rivers. Ocean swims
tend to be colder and may have significant turbulent flow and offshore currents.
Ingestion of saltwater may cause participants to become nauseated or seasick
in ocean swims.1 Ocean swims also involve a greater possibility of interaction
with troublesome aquatic life,1 such as jellyfish, stingrays, or even sharks.
With inland swim venues, the water generally remains calm and has
weaker currents. But if a strong wind is present, the water may become choppy
and swimming may be as difficult as it is in the ocean. These elements will
govern not only participation, swim distance, and the use of a wetsuit but
also the type of stroke technique to be used in the race.2 For example, when
water conditions are rough, the triathlete will have to raise her head more
often to locate the course markers.
As a consequence of swimming with the head up more often, the body
position of the swimmer changes because the hips drop lower in the water.
This alteration in swim position increases the amount of surface area exposed
to the oncoming water, which amplifies the amount of resistance that the
triathlete has to encounter. When the water is calm, the swim phase will
generally be faster because the triathletes can minimize the time that their
heads are above the water looking for the course markers and thus maintain
a more streamlined technique.

Water Temperature
According to the ITU, the water temperature for a triathlon event is taken 24
hours beforehand and approximately 1 hour before the start of competition
on race day. The water temperature is taken from the middle of the course
and two other locations around the course at a depth of 60 centimeters. The
lowest temperature is used as the official water temperature.
The water temperature governs whether the competitors are permitted to
wear wetsuits and whether the swim distance will be modified or cancelled.
The water temperatures outlined for the use of a wetsuit are discussed later
in this chapter. Table 7.1 outlines when swim distances will be modified or
cancelled because of the water temperature.
If the measured air temperature is lower than the water temperature, the
water temperatures used can be decreased accordingly (see table 7.2).
These temperature-related modifications are intended to minimize the
potential harm to competitors, because extreme water temperatures mixed
with long durations of exposure are known to influence hyperthermia and
hypothermia. According to the British Triathlon Federation guidelines, a
sufficient number of blankets should be on hand to supply a minimum of
20 percent of the entrants in an event.5 This provision ensures that treat-
ment can be offered to a number of people if they experience hypothermia.
In the Water | 105

Table 7.1 ITU Water Temperature for Modification or Cancellation of the Swim
Component
Original swim
distance (m) Water temperature
°C 16.9–16.0 15.9–15.0 14.9–14.0 13.9–13.0 Below 13.0
°F 62.4–60.6 60.6–59.0 58.8–57.2 57.0–55.4 Below 55.4
750 750 750 750 750 Cancel
1,500 1,500 1,500 1,500 750 Cancel
3,000 3,000 3,000 1,500 Cancel Cancel
4,000 4,000 3,000 1,500 Cancel Cancel

Table 7.2 ITU Water Temperature Modified by Air Temperature


Air temperature (°C)
15 14 13 12 11 10 9 8

20 17.5 17.0 16.5 16.0 15.5 15.0 14.5 14.0


Water temperature (°C)

19 17.0 16.5 16.0 15.5 15.0 14.5 14.0 13.5

18 16.5 16.0 15.5 15.0 14.5 14.0 13.5 13.0

17 16.0 15.5 15.0 14.5 14.0 13.5 13.0 Cancel

16 15.5 15.0 14.5 14.0 13.5 13.0 Cancel Cancel

15 15.0 14.5 14.0 13.5 13.0 Cancel Cancel Cancel

14 14.0 14.0 13.5 13.0 Cancel Cancel Cancel Cancel

Tables 7.1 and 7.2 adapted, by permission, from International Triathlon Union, 2012, ITU competition
rules, 21/02/2012 edition pp. 15, 16. [Online]. Available: http://www.triathlon.org/images/uploads/
itusport_competition_rules_20120215.pdf [December 3, 2012].

Furthermore, all first aid personnel should be trained in handling the symp-
toms of anxiety conditions and cold temperatures.
Other race factors that may be affected are the triathlete’s ability to reduce
the resistive forces encountered while swimming. According to Clarys,6 in
water temperatures of 18 to 24 degrees Celsius (64 to 75 degrees Fahrenheit),
the amount of resistive forces (commonly termed active drag, which is an accu-
mulation of pressure, wave, and frictional drag) that the swimmer encounters
significantly decreases as the temperature increases.6 Therefore, triathletes
participating in events in which water temperatures are on the upper limit
of the range of 18 to 24 degrees Celsius would be able to generate greater
propulsion and reduce the loss of wasted kinetic energy into the water. Note,
however, that several other factors influence the level of resistive forces, such
as anthropometry, swim technique, and swim speed.
106 } Mason and Sacilotto

Adverse Conditions
Along with reducing swim distance because of adverse water conditions,
senior race officials can incorporate race cutoff times or change the venue
of the swim if they deem conditions unsafe. These conditions could include
strong water currents, large waves, breakers, or strong undertows (water
movement under the surface waves). If cutoff times are evoked, they must
be advised to all the entrants at least 7 days before the event.5
According to USA Triathlon, the operation of an event must be in accordance
with the established guidelines of the government agency that controls the body
of water where the competition is to occur. After water quality is established as
suitable for human swimming, the certificate or notification must be kept for
a period of 7 years after the event.7 Similarly, the Canadian triathlon medical
manual mentions that organizers should consult the guidelines for Canadian
recreational water quality in determining an adequate venue for an event.8
The ITU dictates that the technical and medical delegates may enforce
limits on the swim length as well as provisions about wetsuit use in the case
of adverse weather conditions. The final decision should be made 1 hour
before the start of the event.4 The Triathlon Canada medical manual advises
that a swim leg could be cancelled if competitors or potential rescuers are
not able to see the buoys marking the course when sighting from water level.8
When competing in an international triathlon, competitors must have access
to course specifications, which should be distributed to all competitors before
race day. Each triathlete should investigate factors including temperature,
water conditions, and aquatic life in preparation for the event.

Swimsuits
Swimsuits are to be worn at all times throughout all legs of a triathlon.
Where wetsuits are not permitted, the international competitor must wear
an ITU-approved uniform for the swim portion as well as the bike and run
legs of the event. If the competitor chooses to wear a second suit, it must be
worn underneath the approved uniform.4
The uniform can be either a one piece or a two piece. The ITU prefers the
suit to be a one piece. If a two piece is worn, the two garments must overlap
so that the torso is not exposed during any part of the race.9, 10
Elite U23 and junior competitors must complete the entire triathlon in an
unaltered race competition uniform.11 Where wetsuits are permitted, the
wetsuit must be worn over the top of the swimwear and must display only
the manufacturer’s logo, which must not exceed a size of 80 square centi-
meters.10 An example of an approved age-group male and female uniforms
can be seen in figures 7.1 and 7.2.
In competitive pool swimming in recent years, much controversy has arisen
about the use of swimsuits (or “super suits”) because of the performance-
enhancement capabilities of such suits. The competitive swimming govern-
In the Water | 107

„„Figure 7.1 ITU example of approved age group male uniforms.12

E5643/Friel/Fig. 7.1/448575/TimB/R1

„„Figure 7.2 ITU example of approved age group female uniforms.13, 14

E5643/Friel/Fig. 7.2/448576/TimB/R1
ing body, FINA, now has strict criteria for swimsuit manufacturers, and
swimmers must compete in FINA-approved suits, which are listed on the
FINA website.15 Triathlon federations are also amending competition rules
to mirror FINA in the move to allow only textile swimsuits (or speed suits).16
Similar to the swimsuits used in competitive pool swimming, triathlon
swimsuits reduce the water resistance encountered by the triathlete, but the
suits differ in offering an increased amount of movement and durability
that is needed in a triathlon event. These features enable the triathlete to
complete the three legs of the event as well as address the thermoregulatory
108 } Mason and Sacilotto

aspects of the suits to maintain a near normal core body temperature through
the event. The USA Triathlon website has a list of approved swimsuits that
can be worn in any water temperature.17 When unsure of the legality of a
swimsuit, the triathlete should contact the pertinent governing bodies before
purchasing or competing in the suit.

Wetsuits
Whether wetsuits can be used depends on the water temperature taken 24
hours before and on race day. As mentioned earlier in this chapter, the water
temperature must fall within a certain range for the officials to allow the
use of wetsuits. The sole purpose of the suits is to maintain a stable body
temperature in colder water temperatures. Tables 7.3 and 7.4 outline those
temperature ranges.
Note that national federations or unions have specific guidelines for wet-
suit use and that some events may even use different guidelines altogether.
Therefore, before competing, a triathlete should check the event guidelines
to determine whether a wetsuit may be worn.
Back in the mid-1980s it was shown that swim speed was enhanced by
up to 7 percent by a wet suit.18 Toussaint et al.19 also established that speed
increased by 5 percent because of a 14 percent decrease in the triathlete’s
active drag while wearing a wetsuit. A study completed in the mid-1990s
found that wetsuits aided the swim performance of triathletes by 6 percent.20

Table 7.3 ITU Guidelines for Wetsuit Use in Elite U23 and Junior Competitors
Forbidden above Mandatory below Maximum stay
Swim length (m) (°C) (°F) (°C) (°F) in water
300 20 68 14 57 10 min
750 20 68 14 57 20 min
1,500 20 68 14 57 30 min
3,000 22 72 16 61 1 h 15 min
4,000 22 72 16 61 1 h 45 min

Table 7.4 ITU Guidelines for Wetsuit Use in Age-Group Competitors


Forbidden above Mandatory below Maximum stay
Swim length (m) (°C) (°F) (°C) (°F) in water
750 22 72 14 57 30 min
1,500 22 72 14 57 1 h 10 min
3,000 23 73 16 61 1 h 40 min
4,000 24 75 16 61 2 h 15 min

Tables 7.3 and 7.4 adapted, by permission, from International Triathlon Union, 2012, ITU competition
rules, 21/02/2012 edition, p. 15. [Online]. Available: http://www.triathlon.org/images/uploads/
itusport_competition_rules_20120215.pdf [December 3, 2012].
In the Water | 109

These enhancements in swim speed while wearing a wetsuit were


revealed only with triathletes and not with competitive swimmers. The
explanation was that competitive swimmers have a more streamlined
technique and therefore do not gain as much from wearing a wetsuit as
triathletes do. Wearing a wetsuit allows triathletes to expend less energy
in maintaining a horizontal position, thereby conserving more energy
for propulsion.21 This energy conversation is important, particularly in
the longer triathlon events, in storing energy for the bike and run legs.
According to Chatard and Millet20 wearing a wetsuit decreases the energy
cost of a triathlete by 7 to 22 percent. Recent research has reached similar
conclusions.22
For those reasons, wetsuit guidelines were introduced to create a level
playing field among competitors who may not be able to afford the more
advanced suits.18 As was the case with competitive pool swimming swim-
wear, because of the performance enhancement capabilities of previously
designed wetsuits, the ITU now restricts wetsuit thickness. For example,
no suit may be more than 5 millimeters thick because any greater thickness
will further increase the buoyancy of the competitor, providing a potentially
unfair advantage.
Wetsuits also vary in thickness around the body; for example, in some
brands the thickness under the arms is only 1 millimeter. This change in
suit thickness allows the body to continue to thermoregulate during the
longer-distance swims. Because of their typically lean body compositions,1
elite triathletes have been assumed to have relatively poor tolerance to colder
temperatures; therefore, first and foremost, wetsuits are permitted as a health
and safety aid when the water temperature is low.

Other Swim Gear


During the swim leg of the triathlon the competitor is not allowed to use
any floatation or swim-enhancing aids, such as kickboards, fins, snorkel,
or paddles. Swim gear allowed during an event includes swimming caps,
goggles, and nose clips. The ITU outlines that triathletes must wear the swim-
ming cap given to them at the commencement of the event. If the triathlete
chooses to wear another swimming cap, the event cap must be worn over
it. The caps given to the competitors are bright and enable event officials to
monitor the triathletes throughout the swim leg.
Goggles, which are designed, of course, to keep water out of the triathletes’
eyes and allow them to see clearly, are permitted. In training before an event,
triathletes should practice the technique of looking for buoys because stroke
mechanics used in training in a pool differ from those used when competing
in an open body of water.
Nose clips are used by competitors who find it uncomfortable to have water
up their noses or wish to limit airflow to only their mouths. Some triathletes
110 } Mason and Sacilotto

have not sufficiently practiced the mechanism of breathing properly in water


(out through the nose by blowing bubbles in the water and in through the
mouth). A nose clip limits the possibility that water will enter the nose and
become unpleasant for the triathlete, which may cause anxiety during the
swim.
Although equipment is restricted during competition, triathletes can use
several aids during training to assist in technique and power enhancement.
A list of equipment and their application to training can be found in table
7.5.

Table 7.5 Description of Swim Training Equipment


Equipment Description Advantages

Short fins Fins with short blades Promote concentration on muscle strength in
overspeed drills
Long fins Fins with long blades Used in overspeed drills to increase stroke rate
Pull buoy Floatation device placed Focuses on improving the stroke without
between the legs concern for the legs
Drag suit Mesh pants with mesh pockets Resists the athlete, which encourages an
increase in power per stroke
Kickboard Floatation board shaped device Focuses on improving kicking without concern
for the arms
Snorkel Breathing tube Allows the athlete to practice technique without
concern for the breathing movement
Hand Large paddles that fit on the Enhance propulsion through the water if used
paddles hand and can be attached over correctly or suddenly change direction or come
the wrist and at least one finger off if used incorrectly
Finger Small, half-moon-shaped Same as hand paddles but less extreme if
paddles paddles usually attached only to wrong technique is used; also can be used to
the athlete’s middle finger “feel the water”

Conclusion
Because of the unpredictable nature of open-water swimming during the
first leg of a triathlon, several factors need to be considered when preparing
for an event.
First, the triathlete should learn the course—how the course will be set
out, what type of course markers will be used, what type of water the event
will be contested in, whether wetsuits will be allowed, and what type of start
will be used. Answers to these questions will aid in the preparation for an
event because each aspect can be practiced before race day.
Second, the triathlete should learn how to swim in various water condi-
tions, specifically in both oceans and inland bodies of water. An easy way to
prepare for these conditions is to enter open-water events. These events also
allow the triathlete to practice various race starts, adjust stroke technique to
In the Water | 111

the water conditions, and become more confident swimming in an open body
of water. Another benefit of entering these open-water swims is that they are
structured events, so the participants will be monitored and medical care is
close by if the triathlete gets into trouble.
Third, if a wetsuit is permitted, the triathlete should wear one. Wearing
a wetsuit will enable a quicker swim. Before race day the triathlete should
practice removing the wetsuit quickly to enable a fast first transition.
Finally, the triathlete should learn to use training equipment. Joining a
competitive swimming squad will enable the triathlete to learn not only
correct swim technique but also the proper use of training equipment. The
structure and consistency of squad training also will maintain interest in
training for the swim leg of the triathlon. Because the swim leg is the shortest
part of the triathlon, athletes often minimize their training for this section.
Foremost in any competition to be undertaken, the triathlete must consider
safety in the water on the day of the event.
This page intentionally left blank.
CHAPTER
8
On the Bike
—Jeff Broker, PhD, and Sean Langlais, MSc

O ptimum performance during the cycling portion of a triathlon is heavily


influenced by equipment selection and triathlete–bicycle integration. A
triathlete with a top-of-the-line bicycle will not have success if he is poorly
integrated to (fit to) his bicycle. Likewise, at the highest competitive level,
state-of-the-art equipment can separate medalists from nonmedalists. Ulti-
mately, optimal integration of the rider with the bicycle and proper selection
of equipment will provide a competitive advantage, saving precious seconds.
In this chapter we build on the concepts presented in chapter 5 concerning
cycling biomechanics. After briefly visiting the rules concerning the cycling
leg of triathlons, we turn our discussion to bicycle selection and configura-
tion. We begin by discussing concepts of the steeper seat tube of the tri-bike,
highlighting its role in reducing aerodynamic drag without compromising
the triathlete’s physical performance. The discussion then turns to frame
element design, handlebar selection, and wheel considerations. We then
shift our focus to bike fit, outlining guidelines for saddle placement and
handlebar configuration. The chapter concludes by exploring the demands
of hill climbing, bicycle acceleration, the quantitative rationale for drafting
(when legal), and ways to solve the aerodynamic challenge.

Governing Body Rules


All participants of a triathlon must follow the rules set forth by the organiz-
ing body, typically USA Triathlon (USAT) in the Unites States or the Inter-
national Triathlon Union (ITU). Small local or recreational races may not
operate under a governing body and therefore may have their own rules.
Generally, ITU cycling equipment and racing rules follow the guidelines
set forth by the Union Cycliste Internationale (UCI). Much of the following
discussions concerning equipment and drafting are provided with the USAT
and ITU regulations in mind. For further information, consult the ITU and
UCI websites.

113
114 } Broker and Langlais

Bicycle Choice and Geometry Options


The triathlete’s bicycle should provide an appropriate balance among comfort,
handling, pedaling effectiveness, and aerodynamic quality. Unfortunately,
many of these characteristics are conflicted; that is, an overemphasis on one
characteristic may be detrimental to another. Maximal pedaling power, for
example, cannot be achieved in the most aerodynamic cycling position.
Further, the aerodynamic position can be fatiguing to the neck extensor
musculature and places more load on the front wheel, which can negatively
affect handling.
Elite triathletes ride bicycles specifically built for triathlon racing (tri-bikes)
that attempt to balance the characteristics listed previously. These bicycles,
which are themselves aerodynamic, place the rider in a streamlined position
compatible with reasonable power delivery.
Based on pedaling mechanics research (see chapter 5) as well as real-world
observations of elite triathletes, it can safely be concluded that the preferred
operating condition for triathletes involves a flat-back, streamlined riding
position incorporating a more traditional hip angle range of motion. Here
traditional refers to the range of motion specific to non-time-trial riding in
which the hands are on the hoods or dropbars of a standard bicycle. Seat tube
angles associated with this traditional position center around 72 degrees. In a
flat-back aerodynamic cycling posture, the equivalent seat tube angle rises to
78 degrees or more (figure 8.1). Specifically, to achieve a traditional hip angle
operating range in the more aerodynamic position of a tri-bike, the seat tube
must be steepened from 72 to 78 degrees.
Triathlon-specific research concerning seat tube angles has provided a
scientific basis for triathletes to use steeper seat tube angles. Ricard and col-
leagues1 looked at muscle activation patterns across various seat tube angles
while performing Wingate tests. Wingate tests evaluate a cyclist’s peak
anaerobic power, anaerobic capacity, and fatigue profile over a 30-second
maximal cycling effort. They discovered that riders were able to maintain
similar cycling power outputs at steeper seat tube angles (e.g., 82 degrees)
while significantly reducing the muscular activity of the biceps femoris
muscle. They hypothesized that reduced recruitment of the biceps femoris (a
hamstring) may facilitate the triathlete’s transition to running. These results
are consistent with the data of Browning,2 which showed a slight shift of
pedaling demand from the hamstrings to the quadriceps with forward saddle
movement in elite triathletes.
On the topic of seat tube angles and the effects on the T2 transition, Garside
and Doran3 compared triathlete’s self-selected 10-kilometer running perfor-
mances after riding 40 kilometers at two seat tube angles, 73 degrees and 81
degrees. The subjects were nearly 2.3 minutes (on average) faster during the
first 5 kilometers of the run after riding with the steeper seat tube geometry.
The detected physiological differences included increased heart rate, increased
stride length, and increased stride frequency in the steeper seat tube condi-
On the Bike | 115

© Nigel Farrow
„„Figure 8.1 A triathlete riding a tri-bike with an aggressive steep seat tube in an
aerodynamic position. Note how the hip is not excessively flexed as the left pedal
moves through the top of the pedal cycle. Also notice a slightly humped thoracic spine,
possibly associated with pelvic inflexibility.

tion. During the second half of the run, the effects were less noticeable. This
research coincides with the perception of many triathletes that a steeper seat
tube is more beneficial to running performance. For more discussion of the
T2 transition, see chapters 5 and 6.
The draft-illegal competitive triathlete should select a tri-bike. Entry-level
or recreational triathletes, however, may start with a road bike, especially if
cost is an issue. Modifications to traditional road bikes can address the hip
angle range of motion issue to some extent. Seat posts that allow forward
positioning of the saddle can make the effective seat tube angle steeper, cir-
cumventing the need to replace the frame of the bicycle. Simply moving the
saddle forward, however, reduces the distance between the saddle and bottom
bracket, shortening the effective saddle height. Thus, to maintain appropri-
ate saddle height, the saddle should be moved upward as well as forward. A
general guideline used with Olympic pursuit riders was to raise the saddle
1 centimeter for every 2 centimeters of forward adjustment.
Another consideration when modifying a traditional bicycle for triath-
lon use involves handling. Simply moving the saddle forward and slightly
upward moves the rider’s center of mass closer to the front wheel, thus altering
handling characteristics. If bicycle handling is seriously compromised with
116 } Broker and Langlais

a simple saddle position change, forcing a triathlon position atop a standard


bicycle may not be advised. Further, if forward positioning of the saddle
causes interference between the knees and the handlebars during pedaling,
this modification is not appropriate. Assessment of proper aerodynamic
positioning, coupled with an evaluation of pedaling effectiveness and bicycle
handling capability, is best done with a specialist trained in bicycle design
and fitting for triathletes.
ITU rules govern aspects of bicycle frame design.4 The ITU has adopted
many of the UCI rules governing bicycle racing and includes modifications
depending on whether the triathlon is draft legal or draft illegal. ITU rules
state that a vertical line dropped from the nose of the saddle cannot be more
than 5 centimeters forward of the bottom bracket axle.4 This geometrical
relationship limits the forward positioning of the athlete to the preferred
ranges discussed previously. Additional ITU regulations control the overall
length of the bicycle, the front-to-center distance (front axle to bottom bracket
spindle), and the width of the bicycle.4

Frame Materials
Today, most competition bicycle frames are made of carbon fiber, a lightweight
and highly responsive material. Manufacturers can design a carbon fiber
frame to suit the elite or recreational triathlete, seeking to optimize perfor-
mance through superior handling, aerodynamics, vibration attenuation, and
other key elements. The negative features of carbon fiber are price, impact
resistance, and deterioration over time.
Bicycle frames are also made of aluminum, steel, titanium, and other
metals. The most common frame material is aluminum, a cheaper, heavier
substitute for carbon fiber. An aluminum frame bicycle will likely weigh 3
pounds (1.4 kg) more than a carbon fiber frame with similar components.
As with all equipment choices, the decision should be based on the level of
competition, performance, comfort, and price.

Frame Elements
As with saddle position, the ITU regulates the design of a triathlon bike dif-
ferently for draft-legal and draft-illegal racing. In general, the ITU adopts the
UCI time-trial rules for draft-illegal triathlons and uses the UCI road race
rules for draft-legal triathlons. For draft-legal races, the bicycle frame must
be formed around a main triangle of three tubular elements that may be
round, oval, flattened, teardrop shaped, or otherwise in cross-section.4 Bicycle
elements specifically designed to enhance aerodynamics are prohibited; for
example, fairings are not allowed.
Aerodynamic tubular elements are allowed. Notably, a tapered tube sec-
tion with a long dimension (length, parallel to the airstream) that is two to
three times longer than the short dimension (width, perpendicular to the
On the Bike | 117

airstream) will exhibit one-quarter to one-half of the drag of a round tube of


the same width. Prototype “superbikes” developed by the U.S. Olympic team
in 1996 were fashioned from helicopter blades for this reason. (The Olympic
race versions employed carbon fiber aerodynamic frame elements.) Many
commercially available triathlon bikes incorporate such aerodynamic tubing.

Handlebars
The sport of triathlon contributed substantially to the advent of the aerody-
namic handlebar and clip-on handlebar system. In 2003 Kyle attributed the
invention of the modern aero handlebar to a California engineer, Pete Pensey-
ers, who developed custom handlebars to support a record performance
in the Race Across America.5 Manufacturers Scott and Profile introduced
commercially available aerodynamic handlebars shortly thereafter, and by
1988 triathletes were using them in Ironman competitions.5 Simply stated,
aerodynamic handlebars allow a cyclist to ride comfortably in an aerody-
namic position while still being able to control the bicycle. Wind tunnel
tests and on-road assessments with power meters confirm the performance
advantage of aerodynamic handlebars. Triathletes should capitalize on the
performance advantage of aerodynamic handlebars when they are allowed
by race organizers.
As is the case with frames, the ITU regulates the form and configuration
of aerodynamic handlebars.4 Traditional drop handlebars must be used in
draft-legal racing, and clip-on handlebars are permitted provided that they
do not project beyond the brake levers. Gear shifters are not allowed on the
ends of the clip-on handlebars. For draft-illegal races, handlebar systems
including traditional drop bars with clip-on aero bars must not extend beyond
the leading edge of the front wheel. 5

Wheels
Several types of bicycle wheels are available to the triathlete, including simple
rounded-spoked wheels with flat rims, bladed-spoked wheels with aero rims,
disk wheels fashioned from material composites, membrane aero wheels,
and composite molded three- and four-spoked wheels allowed in unlimited
competitions. These wheels differ in stiffness, weight, strength, durability,
and aerodynamic performance. As pointed out by Kyle, almost any wheel can
be designed to support the strength, durability, stiffness, and braking surface
needs, but not all wheel types have low weight and low aerodynamic drag.5
The ITU dictates wheel selection as follows: For draft-legal competitions,
wheel diameters must be between 55 and 70 centimeters, including the tire,
and both wheels must be of equal size and have at least 12 spokes. For draft-
illegal competitions, covers are allowed on the rear wheel only, subject to
race-specific governance (for example, to address high-wind conditions).4
National or local rules may differ; for example, USAT allows the front wheel
118 } Broker and Langlais

to be of different diameter than the rear, but the front wheel must be of spoke
construction, whereas the rear wheel may be spoke or solid.6
Wheel size matters little as far as aerodynamics are concerned because the
aerodynamic advantage of a smaller frontal area for a smaller front wheel is
largely offset by the increase in rotational speed.5 Smaller wheels, however,
are lighter and stronger, so they offer an advantage during climbing and accel-
erating. Smaller wheels may also be superior to larger wheels in crosswinds.
The only apparent scientific disadvantage to small wheels, within reason,
is their higher rolling resistance.5 Wheel size is best determined based on
cost (including the frame changes involved), wheel and tire availability, and
personal handling preference.
The aerodynamic performance of different wheel types is a hotly debated
topic among wheel manufacturers. In general, composite aerodynamic wheels
have lower drag than the best bladed-, oval-, or rounded-spoked wheels.5
When legal in competition, three- and four-spoked wheels are similar in
performance to flat disk wheels, and the three- and four-spoked aero wheels
perform better than flat disks in crosswinds.5
Rolling resistance is the loss of energy at the contact patch where the tire
meets the road. Energy losses occur here because of deformation of the tire–
tube combination. Rolling resistance is affected by the tread design, tread
construction, sidewall construction, and tube configuration. Rolling resistance
is also affected by the road conditions, load (weight) on the wheel, tire pres-
sure, tire size (i.e., 23 versus 27), wheel diameter, steering, and temperature.
Generally, high-pressure tires with thinner and more flexible casings have
lower rolling resistance.

Bike Fit
For triathletes racing in draft-illegal events, the time-trial position has
emerged as the position of performance. Although pedaling mechanics may
be slightly compromised in this position, the advantages gained from supe-
rior aerodynamics are unquestioned. Fitting a bicycle to a triathlete should
seek to strike a balance among comfort, injury prevention, and, of course,
cycling performance.
A rider has only three points of contact with a bicycle: the pedals, the saddle,
and the handlebars. Optimal positioning involves the manipulation of these
three contact points in relation to one another. Fitting a triathlete to a bicycle
is part science and part art. Although many guidelines can be used to direct
appropriate bike fit, no two triathletes are exactly alike; each triathlete has
unique needs for comfort and performance. Following are brief guidelines
developed over the years that are useful in fitting triathletes to bicycles. These
guidelines are similar in many respects to those used with road cyclists, but
they are specifically modified to the forward, flat-back time-trial position
ubiquitously used by triathletes.
On the Bike | 119

Saddle Height
Optimal saddle height in cycling has been studied for decades. Variables
used to study saddle height include power output, caloric expenditure,
muscle activation patterns, joint forces and torques, and even pedal force
effectiveness. In general, oxygen consumption is minimized at a saddle
height (measured in a straight line from the bottom bracket to the middle
top of the saddle) of roughly 100 percent of greater trochanteric height, or
106 to 110 percent of pubic symphysis height. Both of these measurements
are made while the cyclist is standing barefoot. These measurements require
detailed knowledge of anatomy and are not practical for triathletes. Practical
alternatives developed over the years for setting a cyclist’s saddle height,
based on optimization studies and observations of elite cyclists, are briefly
summarized as follows:
Leg extension method: Perhaps the simplest method for quickly setting
saddle height involves the rider sitting atop the bicycle, mounted in a station-
ary trainer. The rider sits comfortably and centered in the saddle, unclips her
shoes from the pedals, and puts the heels of her shoes on top of the pedals.
The cyclist then slowly pedals backward. Saddle height is set such that the
heels stay in contact with the pedals throughout the pedaling motion, just
reaching the pedals at the bottom of the pedal stroke without rocking of
the hips. Clipping into the pedals will provide the additional leg length and
associated knee flexion during pedaling. This saddle height will be slightly
lower than those developed from the methods to follow.
Crotch height method: Here, the cyclist stands in cycling shoes with
his back to a wall and feet roughly 2 inches (5 cm) apart. A thin book or
broomstick is gently pulled up between the legs in a horizontal orientation
until slight resistance in the crotch is encountered. The distance measured
from the floor to the top of the book or broomstick is noted and multiplied
by 1.09. This computation provides the distance from the pedal spindle axis
to the top middle of the saddle when the crank arm is down and parallel
with the seat tube. This method, first developed by Hamley and Thomas in
1967,7 provides an upper limit to saddle height.
LeMond method: Greg LeMond developed a method similar to the crotch
height method in which crotch height is multiplied by 0.883 to estimate
the distance from the center of the bottom bracket to the top middle of the
saddle.8 LeMond further recommended a 3-millimeter reduction from this
final number when using clipless pedals. Pruitt and Matheny noted that
the LeMond method slightly overestimates saddle height because riders
now wear thinner cycling shoes than those used when the method was
developed.9 Also, riders with long feet or considerable soft tissue over their
ischial tuberosities (sit bones) may think that this saddle height is too low.9
This method is also specific to 172.5-millimeter crank arms; a cyclist using
175-millimeter crank arms would subtract 2.5 millimeters from the estimate
computed.
120 } Broker and Langlais

Dynamic knee angle method: This last method requires the cyclist to
ride the bicycle on a fixed fork trainer. While riding, the knee should flex
25 to 30 degrees from its fully extended position when the pedal is at the
bottom of the pedal cycle. Ideally, the knee angle is measured from video
images captured while the cyclist pedals (a dynamic measure). Fortunately,
many bicycle shops now have video equipment for obtaining these impor-
tant dynamic measures. Lacking video equipment, the cyclist can position
her legs at the bottom of the pedal strokes, and this angle can be measured
using a goniometer. Recognize, however, that dynamic pedaling may differ
from the static method described here, so this method is generally not
preferred.
These methods represent estimates to be modified based on individual
preference, comfort, and performance. No one method is the best because
every triathlete will make modifications. Small, incremental modifications to
saddle height may be necessary to identify the optimal height. Also, recent
research at the University of Colorado in Colorado Springs suggests that
cycling kinematics in the laboratory (fixed fork) differ from the kinematics
of on-the-road cycling for the same rider and bicycle.10 Cyclists may alter
their position on the road in response to changes in work rate, fatigue, and
comfort, an important observation when the concept of optimal position-
ing is considered.

Fore-Aft Saddle Position


The fore–aft position of the saddle is critical for the triathlete’s performance
and comfort. The rule of thumb used for fore–aft placement of the saddle for
road cycling places the anterior (most forward) aspect of the knee directly
above the pedal spindle axis when the crank is halfway through its down-
stroke (see figure 8.2a). This approach is referred to as the knee-over-pedal-
spindle (KOPS) method. This position has been developed over the years
to facilitate power delivery to the cranks without overstressing the knee
joint.
For triathletes the KOPS position cannot be achieved with a flat-back,
time-trial posture. As described earlier, the steeper seat tubes used on tri-
bikes induce forward and upward positioning of the saddle relative to the
conventional road-cycling position, increasing the hip angle to facilitate
effective pedaling. This movement of the saddle places the front of the
knee up to 3 centimeters forward of the pedal spindle (see figure 8.2b). This
forward position increases the torque requirements for the knee exten-
sors (quadriceps), placing higher stresses on the knee joint.2 Triathletes
should use higher cadences (lower gears) for the first few weeks after
changing to the more forward position (after switching to the more aerody-
namic position of a tri-bike, for example), to reduce the likelihood of knee
problems.
© AP Photo/Christope Ena
a

© AP Photo/The Canadian Press, Jonathan Hayward

„„Figure 8.2 (a) Road cyclist with the knee directly above the pedal spindle axis
when the crank is halfway through its downstroke; (b) the more forward position for
triathletes.

121
122 } Broker and Langlais

Upper-Body Position With Aero Bars


Reach and reach length are the traditional terms used to describe handlebar
placement relative to the saddle. Bicycle top tube length and stem geometry
are manipulated to achieve proper reach. For road cyclists, whose position-
ing is somewhat applicable to triathletes who compete in draft-legal events,
the cyclist’s nose should be above or slightly behind the stem clamp when
the hands are in the drops position and the elbows are slightly flexed (15
to 20 degrees). The top surface of the handlebars should be 2 to 4 inches (5
to 10 cm) below the top of the saddle, less for shorter riders and more for
taller riders.
For draft-illegal racing, aero bars and clip-on bars are used, which narrow
the triathlete’s arms and shoulders, thus reducing the rider’s frontal area.
The elbows are brought somewhat together atop armrest pads, ideally with-
out compromising bicycle handling (see figure 8.3). This position requires
practice and accommodation.
The height of the aero bars is adjusted to achieve a flat-back position.
The entire back will not be flat, only the upper portion of the back behind
the head and neck. This posture requires forward rotation of the pelvis,
necessitating the forward and upward positioning of the saddle as previ-
ously discussed. In practice, triathletes with limited low-back flexibility
and tight hamstrings have difficulty getting the middle of their backs into
a flat position.
Considerable debate surrounds the best position of the arms in relation
to aero bar usage. Many coaches recommend that the aero bars be config-
ured to have the forearms horizontal and the elbows flexed roughly 90 to
110 degrees (see figure 8.3). This positioning places the upper arms nearly
vertical, a position in which the arms and shoulders can relax. Others rec-
ommend tilting the aero bars up so that the hands are above the elbows,
more in front of the face. In this configuration the upper arms should still
be nearly vertical.
The debate probably arises because wind tunnel tests have demonstrated
considerable variability among athletes and bicycles. Positions that represent
the lowest drag for one triathlete may not be optimal for another. Further,
comfort, power generation, and handling capabilities may differ among
triathletes for the same aerodynamic position.
Testing of U.S. national team cyclists by the U.S. Olympic Committee in
1995 revealed considerable subject-specific aerodynamic features surround-
ing aero bar placement. Many athletes achieved lower drag forces with their
forearms horizontal. One athlete positioned his forearms below horizontal,
with his hands lower than his elbows, yet suffered no aerodynamic con-
sequences. He felt particularly powerful with his hands in this position.
One concept to keep in mind is the degree to which the arms are used
to direct the wind about the triathlete’s body. With the hands together and
the aero bars tipped upward, the triathlete directs the airflow more around
On the Bike | 123

© AP Photo/Daniel Roland

„„Figure 8.3 Triathlete with the elbows together atop armrest pads.

the shoulders and along the outside of the body. With the forearms level
and the hands slightly separated, the midline airflow is directed between
the arms, down between the legs, and across the frame and rear wheel.
The optimal condition for a given triathlete likely varies, and wind tunnel
testing or cycling power meter testing may be required to establish which
position is preferred. The author’s experience is the former position may be
preferred for cyclists with larger legs or less aerodynamic bicycles (nonaero
seat tube, nonaero rear wheel). Here, directing the airflow around the body
may be best. For the rider with smaller thighs or an aero bicycle frame and
rear wheel, the flat forearm position may be preferred.
124 } Broker and Langlais

Terrain, Drafting, and Solving


the Aero Challenge
Bicycle selection and optimal rider–bike integration are important steps that
triathletes should take before they even roll out of the garage. On the road,
triathletes must manage environmental conditions including hills, wind,
and other cyclists.

Power of Climbing and Descending


Many cyclists and triathletes attempt to decrease bicycle weight to manage the
implications of gravitational forces. Obviously, the influence of gravity is real-
ized during hill climbing, when the power to climb must be added to the power
required just to ride. Unfortunately, the energy saved during a descent does
not offset the energy required to climb because of differences in aerodynamic
drag. Therefore, flat courses are faster than hilly courses of the same length.
Briefly, the power necessary to climb is directly related to mass of the rider
and bicycle system and the magnitude of the grade. Figure 8.4 illustrates the
power necessary to climb in relation to grade and velocity (ignoring aerody-
namics and rolling resistance). Power is expressed in watts per pound for ease
of application. For example, the power required for a 150-pound (68 kg) rider to
climb a 10 percent grade at 10 miles per hour (4.4 m/sec) is roughly 300 watts
(2 watts per lb × 150 lb).
Figure 8.4 also demonstrates the penalty paid by heavier riders and by riders
atop heavier bicycles. For our example (10 percent grade at 10 mph), every 5
pounds (2.3 kg) of extra weight increases the power needed to climb by 10 watts.
This relationship has a dramatic effect across a given race population because
differences in rider weight can exceed 50 pounds (23 kg). Not surprisingly, the
best climbers in the world (of road cyclists) are generally slight of build.

Acceleration
Bicycle acceleration occurs when the energy delivered to the rear wheel
exceeds the total resistive forces acting on the rider and bicycle. Accelera-
tion for a given force application is inversely proportional to the total mass
of the bicycle and rider combination. Therefore, lighter riders with lighter
bicycles will accelerate faster under the same propulsive force. Consider
a rider accelerating uniformly from rest over a distance of 200 feet (61 m)
in 10 seconds. At the 200-foot mark this rider will reach a velocity of 27.3
miles per hour (12.2 m/sec). The same propulsive force developed by a
rider on a bike weighing an additional 5 pounds (2.3 kg) will result in the
cyclist’s being 5.25 feet (1.6 m) behind at the 200-foot mark. Clearly, a rider
on a lightweight bike has an advantage, particularly in a race with many
acceleration opportunities.
On the Bike | 125

Power to climb vs. velocity


5
10%

4
Power to climb (W/lb)
7%
3

5% Grade
2
3%
1
1%
0
0 5 10 15 20 25 30
Cycling velocity (mph)

„„Figure 8.4 Power to climb as a function of hill grade (percent) and cycling velocity.
Power is expressed in watts per pound of8.4/448588/TimB/R2-kh
E5643/Friel/Fig. rider weight. Aerodynamic and rolling resistive
forces are not included in the power measure here.

When bicycles accelerate, their wheels increase their rotational velocity


(accelerate) as well. Wheels have rotational inertia, representing their resis-
tance to rotational acceleration. Rotational inertia increases with mass and
increases more so when the mass is distributed farther from the rotation
axis (axle). Light wheels with their mass concentrated closer to the hub are
easiest to accelerate.

Aerodynamics Revisited
Cycling aerodynamics was briefly addressed in chapter 5, and the aerody-
namic performance features of frames, wheels, and rider body positioning
were discussed earlier in this chapter. In many elite ITU triathlons, draft-
ing is legal. Drafting is simply riding within the wake of another rider
ahead (figure 8.5). The effect of drafting on cycling power and thus energy
expenditure is dramatic.
Several methods have been used to study the effects of drafting. A fas-
cinating study by Hagberg and Nicole quantified the benefits of drafting
using oxygen consumption measurements.11 Riders cycled on the road
in various drafting configurations (pacelines and small pelotons) while
boom-mounted breathing systems measured oxygen consumption. Using
this method, Hagberg and Nicole reported metabolic energy savings of 26
percent plus or minus 7 percent connected with paceline drafting in the
second, third, and fourth positions behind a lead rider. Centered at the back
of an eight-rider peloton, a rider enjoys a reduction in energy expenditure
of 39 percent plus or minus 6 percent.
126 } Broker and Langlais

© Zuma Press/Icon SMI


„„Figure 8.5 One triathlete drafting another. The drafting triathlete can reduce power
output by up to 35 percent by drafting effectively.

A modern and simpler method for studying the effects of drafting requires
a crank-based, pedal-based, or hub-based power meter. These systems are
now widely available, providing many cyclists the opportunity to explore
and appreciate the effects of drafting.
Finally, wind tunnel tests provide a unique look at the drafting phenom-
enon. U.S. national team cyclists preparing for the 1996 Olympics were rotated
through a four-rider paceline in a wind tunnel.12 In these optimal conditions,
where bicycle spacing and alignment could be completely controlled, energy
savings in the second, third, and fourth positions behind a lead rider were 37
percent, 45 percent, and 46 percent, respectively. Actual racing advantages,
quantified using power meters with the same athletes and bicycles on velo-
dromes, were 5 to 10 percent less (still providing an advantage of 30 to 40
percent across the three drafting positions).
On the Bike | 127

Given the substantial advantage offered by drafting in cycling, the ITU


and national triathlon federations have set strict guidelines concerning
drafting in draft-illegal races. The ITU has established a draft zone mea-
suring 10 meters in length and 3 meters in width. The long dimension of
this zone begins at the most forward position of the front wheel of the
lead bicycle and projects rearward. A rider cycling within this draft zone
at a constant speed is in violation of the drafting rule. USAT has a similar
rule, except the draft zone is only 7 meters long and 2 meters wide. ITU
and national federation rules permit riders to be in draft zones when pre-
paring to overtake other riders (completed within 20 seconds for the ITU
and within 15 seconds for the USAT); for safety reasons when entering aid
stations, transition areas, or sharp curves; and in race-specific situations in
which exclusions are necessary because of narrow lanes, construction, and
detours.
Finally, using on-board power meters in controlled and systematic test
situations, the effects of various aerodynamic modifications can be explored
for any level of triathlete. Figure 8.6 illustrates a sample power meter–based
assessment of cycling aerodynamics. Ideally, a cyclist rides on an indoor
velodrome or in calm conditions on a flat road, and power measurements are
made across a range of cycling velocities. The data are plotted, adjustments
in rider geometry on the bike or componentry are incrementally made, and
the test is repeated. The goal is to shift the power versus velocity curve to
the right, documenting an ability to cycle at higher speed for a given energy
expenditure or at lower energy expenditure for a given cycling speed.

Aero position modification


600
Baseline position
550
Modified aero position
500
Cycle power (W)

450
400 0.9 mph
increase
350 at 300 W
300
250
200
150
20 22 24 26 28 30 32 34
Cycling velocity (mph)

„„Figure 8.6 A power versus cycling speed characteristic derived from power meter
E5643/Friel/Fig.
data collected at a velodrome. 8.8/448591/TimB/R1
When modifications are made to improve aerodynamics,
the effects of cycling performance are highlighted. (Shown: an increase of 0.9 miles
per hour [0.4 m/s] with no change in cycling power.)
128 } Broker and Langlais

In 1993 Lance Armstrong was tested at the 7-Eleven Velodrome in Colorado


Springs shortly after his Cycling World Championship victory. The goal was
to improve his time-trial position. Through careful and incremental modifica-
tions to handlebar and saddle positions, coupled with a helmet change, Lance
was able to ride at a sustainable power output over 1 mile per hour (1.6 km/h)
faster in his newly modified, more aero position. Ironically, in response to the
findings Lance said that he could ride comfortably in that position because
he had used it years earlier when he was a triathlete.

Conclusion
Triathletes are no different from serious cyclists when it comes to the selec-
tion of cycling equipment. Competitive triathletes recognize the tangible,
scientific advantages offered by superior bicycles and componentry and the
importance of optimal rider and bicycle integration
Triathletes should invest in tri-bikes or modify their existing bicycles to
achieve a more rotated and streamlined time-trial position. Aero handlebars
should be part of the racing package, when allowed by race organizers. The
aerodynamic advantages of the streamlined position are tangible, and a
more forward position of the rider facilitates pedaling while in this flat-back
posture. Modifying road bicycles to place the rider forward must be done
carefully, however, because handling can be compromised when weight is
shifted forward on a standard road frame.
Bike fit is part science and part art. The various methods of setting saddle
position are useful steps to achieving a saddle position within a centimeter
or two of the probable rider-specific ideal. Trial and error involving on-the-
road testing, preferably with a power meter, will dictate further refinements
to saddle position. A comfortable yet powerful position is the goal.
The effects of hill climbing and drafting on cycling power output must be
appreciated. Weight-saving measures benefit cyclists on hills, and drafting,
when legal, can lead to power savings of up to 40 percent. The smart triath-
lete can use this information to improve race strategy, race course and type
selection, and even training program design.
The author’s experience is that triathletes, as athletes, actively seek out
scientific information concerning their sport and freely explore techniques
to improve performance. As scientists we are heartened to recognize that
research findings concerning the cycling leg of triathlons are actively con-
sumed and considered by triathletes and triathlon coaches. As Dr. Chester
Kyle, renowned cycling scientist and aerodynamicist said regarding cycling
science directions, “Improvements in cycling technology and performance
advancements are the result of evolutionary processes, not revolutionary
processes.” Triathletes as cyclists will continue to evolve, closely tracking
emerging scientific findings, and will undoubtedly figure out how to go
faster than ever before.
CHAPTER
9
For the Run
—George M. Dallam, PhD

N umerous factors influence running performance and injury resistance


beyond training and running technique. These items include the surfaces
chosen for training and racing, the types of shoes worn (or not), the use or
disuse of orthotics in shoes, and the conditions of the environment includ-
ing temperature, humidity, presence of pollutants, and altitude. Awareness
and an appropriate decision-making process regarding these variables can
enhance both the training process and the performance outcome.

Running Surfaces
Although the choice of running surfaces used for training may seem either
mundane or irrelevant, the surface is in fact important to a successful train-
ing and racing outcome. In particular, this choice can influence training
performance, race specificity and the transfer of training, and the potential
for overuse and acute injury occurring because of the training process.

Softer Surfaces
A widely held notion is that natural surfaces such as dirt and grass are easier
on the body than harder surfaces such as asphalt (on roadways and recreation
trails) and concrete (on sidewalks and some roadways and trails; see figure
9.1). But during running we tune our muscle activity (create a level of tension
or stiffness) to minimize muscle vibration at impact and presumably to reduce
damage to the muscle fibers.1 In other words, we adapt how we use the body
as a spring by varying muscle tension to absorb force effectively for a variety
of surface characteristics. This modulation of the impact forces means that
there are not large differences in impact between different surfaces.
As a result, it has been hypothesized that any variation in impact forces across
varying surfaces is not significant to injury potential.2 Rather, we simply change
our running mechanics by varying muscle tension and the resulting flexion
of our joints under load to accommodate variations in surface and midsole
hardness. Impact forces on surfaces of varying hardness are thus comparable.3

129
130 } Dallam

„„Figure 9.1 Different running surfaces do not vary significantly in their effects on
the body.

Therefore, in practical terms, a triathlete can generally adapt effectively to


running on a variety of surfaces.

Harder Surfaces
Harder surfaces are less resilient (return less energy), meaning that our
springlike muscle and joint structure must work harder to maintain accept-
able impact forces. In such conditions (running on concrete or a hard track,
for instance) muscle contraction levels and stiffness must be higher to create
the appropriate damping effect.4 An efficient runner loses little energy, and
runners can run faster than they can on softer surfaces such as dirt or natural
grass (see figure 9.2). Faster running, however, causes higher impact forces,
higher peak muscle activations, and greater fatigue.5
For the Run | 131

„„Figure 9.2 Faster running produces higher impact forces, higher peak muscle
activation, and greater fatigue.

Resilient Surfaces
Highly resilient surfaces allow a runner to regain energy with each step, allow-
ing for a reduction in muscle activity at a given speed.6, 7 Think of bouncing
on a trampoline in comparison with jumping up and down on the ground.
Of course, this action requires that our movements match the movements of
the surface to maximize energy return with each step. Too much resilience
can hinder running speed by forcing our exchange of supports to occur too
slowly. For example, have you ever tried to run on a trampoline?
Examples of optimally resilient surfaces that match typical running stride
rates include wooden tracks and the newest versions of synthetic turf, which
are infused with rubber particles. Treadmills often offer a spring-loaded deck,
which can be viewed as a highly resilient surface as well. Such conditions
produce running that is easy on the body and can be the fastest if running
stride rate matches surface resiliency.
Finally, soft energy-absorbing (with minimal return) surfaces such as sand
do not promote the springlike behavior of efficient running. Such surfaces
minimize ground reaction impact forces but also dramatically slow running
speed and increase metabolic effort substantially by reducing elastic energy
return.8
132 } Dallam

Practical Applications
Several practical applications come to mind. First is the need for training
specificity. A triathlete who races on hard surfaces should train adequately on
such surfaces at race speeds to minimize fatigue in racing conditions. From
an injury-prevention perspective, large volumes of slower paced training can
then be carried out on optimally resilient or softer surfaces such as dirt trails,
grass, and artificial turf to minimize accumulated fatigue. Those adjusting
to barefoot running may find the rubberized artificial turf optimal for early
adaptation. Finally, running speed may be optimized by using appropriately
tuned resilient surfaces for speed work.
Additional concern in selecting running surfaces include the regularity and
incline of the surface and the effect of impact forces on foot strike hemolysis,
the loss of red blood cells that results from each impact. Increased hemolysis
further increases the need to replace red blood cells thereby adding additional
risk of anemia. Although extended aerobic running appears to induce hemoly-
sis, the irregularity of natural grass surfaces increases this to a greater degree
than more evenly developed artificial surfaces such as asphalt.9 In addition,
the greater impact forces associated with downhill running increase the rate
of hemolysis in comparison with flat and uphill running.10
Combining downhill and uphill speed work has been demonstrated to
create a greater positive influence on speed improvement than simply run-
ning fast on flat surfaces (see figure 9.3).11 Performing fast race-pace training

„„Figure 9.3 Hill running during training has been shown


to improve running speed.
For the Run | 133

increases the rate of hemolysis and muscle inflammation in comparison


with slower running, although surface types (grass versus asphalt) no longer
appear to matter.12 Finally, running on surfaces that are canted sideways has
been shown to induce distinct muscle activation changes,13 likely resulting in
mediolateral (side-to-side) changes in both posture and running biomechanics
if pursued habitually or without balanced compensation (running equally
on cants in the opposing direction).
The main applications of this information are found in the need to assess
iron status in those who pursue a complex approach to run training including
a variety of levels of running speed and the use of uphill and downhill train-
ing, as well as either to avoid or to counterbalance canted-surface running.

Running Shoes
The paradigm that exists around the use of specialized running shoes reflects
the belief system that doing so will prevent chronic running injuries and
improve performance. But the scientific evidence can be viewed as suggest-
ing the opposite.14, 15, 16, 17
A 2009 review of the available scientific literature addressing the prescrip-
tion of running shoes to prevent overuse injuries concludes that this practice
is not evidence based.15 Further, the available scientific evidence on running
without shoes illustrates a lower level of running injury among populations
who are unshod in comparison with those who wear shoes habitually, sug-
gesting that successful adaptations to barefoot running may be more effective
in the prevention of injury than the wearing of specialty running shoes.16

Studies on Running Barefoot


In a review of the literature on running barefoot published in 2001,17 the
author concludes that the use of specialized running shoes can be a negative
factor in the development of some injuries including ankle sprain and plantar
fasciitis. He states that ankle sprains may be increased either by decreasing
awareness of foot position or by increasing the twisting torque on the ankle
during a stumble, and that plantar fasciitis and other chronic injuries of the
lower limb may be increased by modifying the transfer of shock to muscles
and supporting structures. Further, he adds that running without shoes offers
concern for puncture wounds, bruising, thermal injury, and overuse injury
during the adaptation period. Finally, he observes that running shoes play
an important protective role on some courses, in extreme weather, and with
certain pathologies of the lower limb.17 Consequently, the decision to wear or
not wear specialty running shoes is neither clear cut nor simple. Triathletes
interested in gaining the potential benefits of running barefoot or in mini-
malist shoes should consider multiple factors including their own running
history, biomechanics, and typical running surfaces used. In addition, any
change away from conventional running shoes should be made gradually.
134 } Dallam

In regard to the performance-enhancing potential of running without


shoes, clear experimental data illustrate that running economy is improved
in the range of 2 to 4 percent as a result of both reduced weight on the foot
and altered running biomechanics that are likely to increase elastic energy
return18 (see figure 9.4). Running shoes, by contrast, absorb energy that is
not returned to the musculoskeletal system for forward propulsion and are
therefore less economical than running without shoes.18

„„Figure 9.4 Running barefoot may improve running economy.


For the Run | 135

Advantages of Training With Shoes


In the real world of racing and training, the potential for improved per-
formance from running with shoes in comparison with running without
shoes might come from several avenues including
• enhanced energy return versus absorption,
• improved stance stability because of the external friction character-
istics of the shoe,
• reductions in impact forces and stiffness during stance to reduce
muscular fatigue over time, and
• reduced anxiety in relation to landing impacts in terrain with uneven
surfaces or rocks.
Reviews of the available literature on the energetics of running with
shoes suggest that effective mechanisms to enhance energy return to
meaningful levels are not yet in existence in running shoe technology.19
Rather, running shoes have been designed primarily to absorb energy.19
Consequently, the best approach for performance improvement by this
mechanism that is currently available is most likely to be found by using
shoes with minimal cushioning and the lowest weight possible to minimize
energy absorption.
Further, relatively stiff midsoles in the ball of the foot appear to mini-
mize energy loss through flexion of the metatarsal phalangeal joints (the
toes).20 Cross country spikes, for example, accomplish this by virtue of the
relatively stiff plate necessary under the ball of the foot to provide a base
for the installation of the spikes.
The enhancement of external friction characteristics is effectively accom-
plished by the selection of shoe outsoles appropriate for the nature of the
surface run on, including minimal or no lugging for road shoes in optimal
conditions, increased lugging for loose dirt and slippery road conditions,
and spikes for conditions such as grass and ice.
The effective reduction of impact forces is accomplished primarily by
alterations in running biomechanics and muscle forces and is less affected
by running shoe cushioning.21, 22 Consequently, the use of highly cushioned
shoes is more likely to create a performance disadvantage when running
on harder surfaces. A preferred strategy is to adapt to running on such
surfaces in the shoes intended for racing.
All runners have a level of protection to which they have both adapted
and habituated that likely provides them with the confidence to run at the
speeds to which they are adapted. Protection may range from a barefoot
condition to the wearing of heavily cushioned shoes, and the conditions of
running may vary from the most compliant rubberized turf to rocky or hard
terrain covered with small obstacles. Although the concept has not been
scientifically studied, exceeding our current habituated behavior is likely to
136 } Dallam

induce anxiety and alter running mechanics in a way that reduces speed.
To embrace this concept, imagine the effect of running over broken glass.

Recommendations
Overall, the shoes most likely to provide the best performance will be as
lightweight as possible, relatively stiffer in the area of the ball of the foot,
equipped with the appropriate outsole to suit conditions, and designed not
to inhibit the normal actions of barefoot running. In multisport competition
such shoes should also be relatively easy to pull on bare feet while stand-
ing. Of course, such a shoe must be progressively adapted to in training and
adequately cushioned to provide the psychological confidence at foot strike
necessary for the full expression of individual running technique at racing
speed. For some relatively small number of people in a modern population,
this guideline does not rule out the concept of running without shoes in
certain conditions.
A conservative application of this knowledge, in consideration of the lack of
scientific data that directly reflect modern conditions in developed societies,
being that the vast majority of triathletes have grown up wearing shoes and
running in specialty running shoes, is that any transition to lighter weight,
less cushioned, less supportive footwear should be done with extreme cau-
tion and a slow patient progression.

Orthotics
Orthotics are custom-made insole appliances often prescribed by physicians
to assist in overcoming running- and walking-related chronic injuries. Such
appliances range from minimalistic, soft support systems to full-foot semi-
rigid systems. The intent of such devices is to support the foot through its
longitudinal and transverse arches in either their current condition or an
idealized condition.
The original intent of orthotic devices was to limit the process of pronation,
which occurs in the ankle and foot as a means of absorbing shock upon foot
strike, to align the rest of the body properly during locomotion. Excessive
pronation was thought to predispose a runner to a variety of lower-body
injuries. The occurrence of such injuries rose nearly exponentially during the
increased participation of people in recreational running during the running
boom of the 1970s and 1980s.23
But we have since learned that pronation is a necessary function and that
attempts to control it through shoe construction and orthotic devices may
unproductive over time for many runners.24 The more conservative approach
advocated recently is to prescribe orthotics to those for whom there are clear
rationales for treating specific injuries such as plantar fasciitis.25 A review of
the variety of potential applications of orthotics found them to be generally
For the Run | 137

favorable, particularly for pain management.26 Further, the use of orthotics


in a prophylactic manner in a military population undergoing basic training,
a situation known to produce extremely high injury rates, reduced injury
occurrence in the treatment group to a third of that in the control group
among recruits considered at increased risk for injury.27
Another rationale for the use of orthotics in sport and physical activity is the
reduction of energy use. In other words, if the orthotic device aligns skeletal
movements in a productive way, muscle activations and oxygen consump-
tion should be reduced.28 This theory, however, has not been supported in
the scientific literature. A mid-1980s study of running economy comparing
conditions when unshod, when shod, and when shod with orthotics found no
improvement with orthotics and higher oxygen cost in general when wearing
shoes. The same study found less angular displacement of the knee in the
barefoot condition as well.29
In summary, orthotics should not be considered as performance-enhancing
devices for running. Rather, their use should proceed from individual diag-
nosis and custom fitting after changes in training, running biomechanics,
and footwear have proved inadequate to resolve specific overuse injury issues
known to be treatable with orthotics. Finally, the use of orthotics may also
be warranted in short-term training situations in which the normal rates of
progression are likely to be violated.

Environmental Conditions and Running


Triathletes race across a variety of environmental conditions in temperature,
humidity, altitude, and pollution level. Extremes of temperature and humidity
negatively influence endurance-performance capability and exercise safety,
but triathletes can reduce concerns and improve performance in such condi-
tions by using acclimatization or acclimation methods in training. Increased
altitude exposure also reduces performance capability, an effect that can be
negated to some degree by using training acclimatization and acclimations.
Further, such training strategies may offer performance benefits even when
competing in optimal conditions. Pollutants, on the other hand, serve only
to inhibit performance capability and exacerbate breathing difficulties, so
avoidance strategies become important.

Temperature and Humidity


Increasing ambient temperature beyond some optimal range inhibits running
performance. Two recent studies examining historical race data show that
marathon racing performance begins to degenerate at 5 degrees Celsius (41
degrees Fahrenheit) and continues to degenerate further through the available
data at 25 degrees Celsius (77 degrees Fahrenheit). This loss in performance
occurs to a progressively greater degree with lower-performing runners.30, 31
138 } Dallam

Even greater performance decrements at higher temperatures are certain


to occur with the additional concern for heat-induced illness including heat
stroke and heat exhaustion. In triathlon, in which the running portion may
occur at temperatures above 35 degrees Celsius (95 degree Fahrenheit),32 heat
injuries are a predominant reason that athletes end up in the medical tent.
In multisport this phenomenon is further exaggerated by prior dehydration
that occurs during the swimming and cycling legs of the event.33
A study that examined the occurrence of heat-related injuries in both an
early season and a later season triathlon event illustrates the large reduction
in heat-related injuries that occur after changing general environmental
conditions later in the racing season have allowed heat acclimatization to
occur.32 Although most heat illness prevention strategies focus on taking
in fluids during the race, wearing optimal clothing, and slowing the pace
appropriately for the conditions,33 prior heat acclimatization or acclimation
may be even more critical to successful performance in the heat.
Acclimatization refers to the process of adapting to environmental conditions
by living in them naturally. This strategy is problematic for most athletes
because they must move to a new training location for a period of days to
weeks and suffer a loss in performance capability while completing the adap-
tive process. In addition, they may not even experience positive adaptations
beyond that which occurs through training alone.34 Further, the process is
difficult to control in terms of progressively increasing exposures.
Acclimation refers to the process of increasing tolerance using manmade
conditions. This approach allows the minimum necessary progressive
exposures so as to minimize the detrimental effect of the process on current
training. In addition, acclimation can be carried out in the triathlete’s home
location. A recent review of such practices concludes that heat acclimation
can be effective in 5 or fewer days, may be manipulated to produce a more
significant response than natural acclimatization, and is more likely to be
beneficial even in already well-trained athletes.35
Practical strategies for implementing heat and humidity acclimation in
running include using treadmill running with controlled heat and humid-
ity levels indoors, using heavier clothing in normal ambient conditions, and
using passive exposures through devices such as hot tubs or saunas. The
advantage of the third approach is that normal training intensities can be
maintained by training in more optimal conditions while heat acclimation
proceeds passively. Concern for adequate fluid replacement and the mini-
mization of increases in body core temperature should always be exercised.

Altitude
Increased altitude and the resulting reduction in barometric pressure and
oxygen availability produce decrements in maximal oxygen uptake at as little
as 300 meters (984 feet) above sea level in well-trained athletes. This reduc-
tion in oxygen uptake increases linearly with increasing altitude through
For the Run | 139

the available measurements at 2,800 meters (9,200 feet).36 With the increas-
ing amount of competition in multisport events, such as off-road triathlons,
occurring at altitudes well above sea level, strategies for minimizing the
deleterious effect of increased altitude become important for many athletes.
Prior acclimatization to altitude clearly improves altitude performance by
reducing this deficit.37 Further, it appears that even as few as seven intermit-
tent exposures to increased altitude can significantly reduce the performance
decrement.38
The effects of altitude acclimatization on performance at sea level are less
clear cut.39 Acclimatization to moderate altitudes of 2,000 to 2,500 meters is
likely to increase maximal oxygen carrying capacity. But running power
output or training velocities may be hampered by reduced oxygenation levels,
particularly when efforts extend beyond 1 to 2 minutes. Over an extended
period this effect is likely to offset any gains made in oxygen carrying capac-
ity when competing at sea level. This outcome appears to result from a loss
in peripheral muscle adaptations that facilitate faster running at sea level if
high-quality training at altitude is not addressed.
With the advent of what is called the high–low training model, whereby
athletes live at moderate altitudes and incorporate training in conditions that
allow for sea-level work output, this problem has been largely rectified. A
general review of the literature addressing the high–low altitude training
model suggests that an average performance improvement of 1 to 2 percent,
maintainable for up to 3 weeks upon return to sea level, is possible beyond
that which can be accomplished with an optimal training program alone.40
Success using such a model has been demonstrated at both 3,000 meters and
5,000 meters of running.41
Criteria suggested necessary to make this model effective in marathoners
are as follows: 20 or more hours a day of natural environmental exposure at
altitudes ranging from 2,100 to 2,500 meters per day for 4 weeks combined
with appropriate higher-velocity low-altitude or hyperoxic training. Further,
athletes attempting to adapt successfully to such training should ensure
that their iron stores are adequate.42 Producing adequate training velocities
for the low-altitude training has been accomplished by using supplemental
oxygen, travel to lower altitudes, and shortened training intervals whereby
sea-level speeds can be maintained.37, 39, 40, 42 Also proposed is a more com-
plex scheme that combines low-altitude training with high supra-threshold
training completed in short interval fashion.43
For all practical purposes several methods of implementing this model
exist. These include habitual residence at moderate altitude combined with
supplemental oxygen training on treadmills, habitual or part-year residence
at moderate altitudes with interspersed travel to lower altitudes for high-
quality training, constant shifting of training locations between moder-
ate altitudes and sea-level locations in 3- to 4-week periods, and periodic
moderate-altitude camps typically preceding periods of intensive training
at sea level before major competitions.
140 } Dallam

In summary, increased altitude compromises endurance performance,


although the compromise can be minimized by prior acclimatization. Lon-
ger-term acclimatization to altitude can also be used to improve sea-level
performance as long as exposures are adequate and a high–low training
model is used.

Pollution
Before the Beijing Olympic Games, the United States Olympic Committee
conducted a scientific symposium that had the primary purpose of helping
coaches and athletes prepare for competition in one of the most polluted
cities in the world in recognition of the importance of this topic. The
pernicious effect of both acute and chronic forms of air pollution results in
both performance decrements44 and the potential for destructive changes to
respiratory health.45, 46
Considering that exercise can increase the rate of ventilation 10-fold or more
in comparison with resting ventilation, while simultaneously encouraging an
oral breathing route that delivers largely unfiltered air to the lungs, it is not
surprising that air pollutants have a powerful effect on human physiology.
Recently, an increase in exercise-induced bronchospasm among nonasthmatic
athletes has been identified, further illustrating the chronic effects of high
levels of ventilation and the inhalation of airborne irritants and chemicals.47
Current treatments emphasize a pharmacological approach,47 but additional
preventative strategies should be considered as well.
The primary polluting gas of concern is carbon monoxide, a by-product of
the incineration of gasoline and tobacco products as well as many industrial
processes. Carbon monoxide directly binds to hemoglobin, competing suc-
cessfully with oxygen to reduce oxygenation. Consequently, one important
strategy is to avoid training where people smoke and drive cars, as well as
in industrial areas. A Canadian study conducted in Toronto concluded that
carbon monoxide levels were lowest before 7:00 a.m. and after 8:00 p.m. and
highest during morning rush hour. The study further suggested the pre-
vention strategy of exercising indoors during periods of high pollution or
when conducting the most intense training to minimize carbon monoxide
exposure48 (see figure 9.5).
Of further concern is the airborne fine-particulate matter associated with
incineration or dust in the air. Exercise has been shown to increase the deposi-
tion of this material in the lung by 4.5 fold.49 Avoiding this problem requires
the same strategies mentioned previously as well as wearing a mouth filter
or adapting to nasal breathing during exercise training.
Addressing the problems created by air pollution for triathletes requires
a multidimensional avoidance strategy:
1. Avoid training in areas of high vehicular traffic.
2. Avoid training outdoors during periods of high vehicular traffic.
For the Run | 141

3. Conduct high-ventilation training in the areas available where pollution


is lowest, often indoors.
4. Use filter masks, perform nasal breathing, or reduce ventilation as able
if exposed to high concentrations of particulate matter in the air, as in a
dust storm or smoke.

„„Figure 9.5 Consider running indoors on a treadmill to avoid pollution exposure.


142 } Dallam

Conclusion
A variety of environmental factors and equipment options influence both
performance and injury resistance in running, including temperature,
humidity, altitude, and pollution levels along with the selection of running
surfaces and shoes.
Although most triathletes can successfully adapt to a variety of running
surfaces without increasing injury risk, to develop speed, they should seek
optimal running surfaces that are resilient but do not absorb too much force.
Applying conventional wisdom regarding the utility of dirt trail running to
limit fatigue comes at the cost of the slower running that this approach will
necessitate. The incorporation of uphill and downhill running along with
running at higher speeds requires attention to maintaining serum iron levels.
Running on banked surfaces requires altered muscle-activation patterns that
likely will change running biomechanics over time. These surfaces should
be avoided or counterbalanced as able.
Ideal running shoes can increase performance and improve injury resis-
tance. Such shoes will be lightweight, minimally cushioned, and lacking
the gait control features present in many modern running shoes. The use of
such shoes, however, requires a slow and patient progression from current
running footwear. Orthotic use should be limited to diagnostic prescription
for specific running-related injuries when training and biomechanical modi-
fications have proved unsuccessful.
Heat acclimation strategies should be used to improve performance and
deter heat injury when competing in hot and humid conditions. Altitude
acclimatization should be used before competitions at altitudes higher than
the triathlete’s altitude of residence. A high–low altitude training model can
maximize performance at sea level.
Airborne gaseous pollutants and particulate-matter intake should be
minimized through avoidance strategies to ensure maximum performance
potential and minimize the potential to develop respiratory problems associ-
ated with exercise.
CHAPTER
10
Triathlon Training
Technologies
—Hunter Allen

T echnological advancements in recent years have made lives easier and


simpler and have even advanced knowledge about training. For technol-
ogy to be useful to triathletes and coaches, it needs to help them become faster
and stronger and increase their stamina. If technology is truly a breakthrough
tool, it will help triathletes achieve their goals in a shorter time and with
less effort. Many technological tools can do this, and triathletes and coaches
should learn to use and integrate these tools into their triathlon training and
racing. This chapter introduces the most current and useful training tools in
the marketplace today.

Improving Training Through Technology


One of the most important advances in technology with all electronic train-
ing devices (GPS, power meters, heart rate monitors, and so on) is that each
now records a complete history of every workout second by second. These
recordings have fundamentally changed how triathletes train, giving them
the ability to review every training run, ride, and race and to analyze the
data for signs of improvement. Triathletes can learn many things through the
scientific method, which incorporates observation, hypothesis testing, data
collection, and conclusions about what is effective and what is ineffective.
Training with these devices can help triathletes learn their strengths and
weaknesses, their relative fitness as related to themselves and to others, the
effectiveness of their training regimens, the quickness of their response to
training, the way in which they should pace energy expenditure, and ways
to plan and predict peak performance. Of course, triathletes can learn many
other things along the way—these are only the highlights.

143
144 } Allen

All coaches and triathletes are looking for improvement. In doing so, they
constantly consider what they can change to improve their training—whether
they should increase their. volume, adjust the training frequency for one of
the disciplines, train the VO2max energy system more or less, and more.
The primary question is always, “Should the triathlete change something?”
If a technology tool isn’t answering that fundamental question, then it’s not
an effective tool.
Keep this fundamental question in mind when examining each of the
tools used in triathlon. Of course, a triathlete may already possess an effec-
tive tool but not have enough knowledge to use it effectively. The coach and
triathlete must learn how to do this by doing research using a book like this
or possibly by hiring a coach to shorten the learning process.

Heart Rate Monitors


Heart rate monitors (HRM) are ubiquitous in the endurance sports world.
They are relatively inexpensive, give the user some basic feedback on exertion
level, and allow the user to understand his response to a training dose and
establish training zones, which can be used to help direct training.

Understanding the Limits of Heart Rate Monitors


Heart rate is only a response to a training dose. A heart rate monitor does
not quantify training load or work, and understanding that distinction is
important. Heart rate is the dependent variable in the equation, and many
factors affect it, such as heat, humidity, fatigue level, number of hours slept
the night before, hydration status, and even level of excitement. If heart rate
reaches 160 beats per minute (bpm), it could mean that a dog chased the
runner down the road or that the runner was doing a 20-minute threshold
interval. It could also mean that the triathlete is just standing at the start line
and anxious about beginning the race.
The other key understanding about heart rate is that it lags behind effort.
Heart rate can lag up to 30 seconds behind the current effort, and this presents
a couple of problems:
• First, it means that triathletes are constantly chasing the perfect heart
rate number—overshooting it, easing off, going hard again to get it back
where it should be, and repeating this process over and over, which cre-
ates an inconsistent rhythm and suboptimal effort.
• Second, for shorter efforts, heart rate is largely irrelevant because in many
cases effort is already over before heart rate even begins rising. A peak
heart rate could be achieved after the effort is finished.
Although heart rate is not the perfect training tool because it tells only half
the story, even knowing half the story is better than not knowing any of it.
Triathlon Training Technologies | 145

Establishing a Training Zone


A testing protocol to establish training zones is an important first step. The
testing protocol must be a maximal sustained effort long enough for heart
rate to stabilize and for the noise, or small fluctuations, to be eliminated
during the test itself. This test also needs to be done on a relatively flat course
or steady climb to avoid downhills that help triathletes rest and lower their
heart rate. The test must be done at the maximal, sustained effort for at least
20 minutes; 30 minutes would be even better because a longer time will better
account for the initial effort.
The average heart rate of the last 10 or 20 minutes of the test will be a good
number to represent threshold heart rate and establish training zones. Each
of the many training zone systems has its positives and negatives. One of
the best is Joe Friel’s zones, first introduced in The Triathlete’s Training Bible.1
Friel’s cycling heart rate zones are listed in table 10.1.
Thousands of cyclists and triathletes have used these zones because Friel
recognized how big the difference is between the perceived exertions of the
athlete and heart rate as the athlete approaches threshold. The difference
between riding or running just below threshold, just at threshold, and then
pushing to just above it is a large one from the perceived exertion standpoint,
but not from the pure heart rate number standpoint. The difference might be
only five or six beats, but there is a world of difference in how it feels. Physio-
logical changes occur in that narrow window as well.
Clearly separating these areas is an important and necessary component in
learning to train in the proper zone. Note also that a triathlete will likely have
to deal with two sets of zones, one for cycling and one for running. Generally,
in running the heart rate will be higher because more of the upper body is
engaged in the activity than in cycling. In most circumstances a triathlete’s
running threshold heart rate is about four or five beats above her cycling
threshold heart rate, but triathletes need to do a running test along with a
cycling test to quantify the difference between how the heart responds to
both sports. Joe Friel’s running heart rate zones are in table 10.2.2

Using a Monitor for Pacing


Pacing is an important part of triathlon, and using a HRM for this purpose is
a practical and simple way to help accomplish goals. Many triathlon coaches
use heart rate as a governor in Ironman distance races to prevent overexertion
during the bike leg and to compensate for heat and hydration issues. This
approach may or may not be a good idea, depending on the individual tri-
athlete. Based on the fitness level of the triathlete, the percentage of threshold
heart rate should be adjusted for the distance of the race.
For example, a strong age-group triathlete will be able to hold 87 to 90
percent of threshold heart rate (not max HR) for a half-Ironman quite easily
Table 10.1 Cycling Heart Rate Zones
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
<109 109–122 123–128 129–136 137–140 141–145 146+
<110 110–123 124–129 130–137 138–141 142–146 147+
<110 110–124 125–130 131–138 139–142 143–147 148+
<111 111–125 126–130 131–139 140–143 144–147 148+
<112 112–125 126–131 132–140 141–144 145–148 149+
<113 113–126 127–132 133–141 142–145 146–149 150+
<113 113–127 128–133 134–142 143–145 146–150 151+
<114 114–128 129–134 135–143 144–147 148–151 152+
<115 115–129 130–135 136–144 145–148 149–152 153+
<116 116–130 131–136 137–145 146–149 150–154 155+
<117 117–131 132–137 138–146 147–150 151–155 156+
<118 118–132 133–138 139–147 148–151 152–156 157+
<119 119–133 134–139 140–148 149–152 153–157 158+
<120 120–134 135–140 141–149 150–153 154–158 159+
<121 121–134 135–141 142–150 151–154 155–159 160+
<122 122–135 136–142 143–151 152–155 156–160 161+
<123 123–136 137–142 143–152 153–156 157–161 162+
<124 124–137 138–143 144–153 154–157 158–162 163+
<125 125–138 139–144 145–154 155–158 159–163 164+
<126 126–138 139–145 146–155 156–159 160–164 165+
<127 127–140 141–146 147–156 157–160 161–165 166+
<128 128–141 142–147 148–157 158–161 162–167 168+
<129 129–142 143–148 149–158 159–162 163–168 169+
<130 130–143 144–148 149–159 160–163 164–169 170+
<130 130–143 144–150 151–160 161–164 165–170 171+
<131 131–144 145–151 152–161 162–165 166–171 172+
<132 132–145 146–152 153–162 163–166 167–172 173+
<133 133–146 147–153 154–163 164–167 168–173 174+
<134 134–147 148–154 155–164 165–168 169–174 175+
<135 135–148 149–154 155–165 166–169 170–175 176+
<136 136–149 150–155 156–166 167–170 171–176 177+
<137 137–150 151–156 157–167 168–171 172–177 178+
<138 138–151 152–157 158–168 169–172 173–178 179+
<139 139–151 152–158 159–169 170–173 174–179 180+
<140 140–152 153–160 161–170 171–174 175–180 181+
<141 141–153 154–160 161–171 172–175 176–181 182+
<142 142–154 155–161 162–172 173–176 177–182 183+
<143 143–155 156–162 163–173 174–177 178–183 184+
<144 144–156 157–163 164–174 175–178 179–184 185+
<145 145–157 158–164 165–175 176–179 180–185 186+
<146 146–158 159–165 166–176 177–180 181–186 187+

146
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
<147 147–159 160–166 167–177 178–181 182–187 188+
<148 148–160 161–166 167–178 179–182 183–188 189+
<149 149–160 161–167 168–179 180–183 184–190 191+
<150 150–161 162–168 169–180 181–184 185–191 192+
<151 151–162 163–170 171–181 182–185 186–192 193+
<152 152–163 164–171 172–182 183–186 187–193 194+
<153 153–164 165–172 173–183 184–187 188–194 195+
<154 154–165 166–172 173–184 185–188 189–195 196+
<155 155–166 167–173 174–185 186–189 190–196 197+
<156 156–167 168–174 175–186 187–190 191–197 198+
<157 157–168 169–175 176–187 188–191 192–198 199+
<158 158–169 170–176 177–188 189–192 193–199 200+
<159 159–170 171–177 178–189 190–193 194–200 201+
<160 160–170 171–178 179–190 191–194 195–201 202+
<161 161–171 172–178 179–191 192–195 196–202 203+
<162 162–172 173–179 180–192 193–196 197–203 204+
<163 163–173 174–180 181–193 194–197 198–204 205+
<164 164–174 175–181 182–194 195–198 199–205 206+

Table 10.2 Running Heart Rate Zones


Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
<120 120–126 127–133 134–139 140–143 144–149 150+
<120 120–127 128–134 135–140 141–144 145–150 151+
<121 121–129 130–135 136–141 142–145 146–151 152+
<122 122–130 131–136 137–142 143–146 147–152 153+
<123 123–131 132–137 138–143 144–147 148–153 154+
<124 124–132 133–138 139–144 145–148 149–154 155+
<125 125–133 134–139 140–145 146–149 150–155 156+
<125 125–134 135–140 141–146 147–150 151–156 157+
<126 126–135 136–141 142–147 148–151 152–157 158+
<127 127–135 136–142 143–148 149–152 153–158 159+
<128 128–136 137–143 144–149 150–153 154–158 159+
<129 129–137 138–144 145–150 151–154 155–159 160+
<130 130–138 139–145 146–151 152–155 156–160 161+
<131 131–139 140–146 147–152 153–156 157–161 162+
<132 132–140 141–147 148–153 154–157 158–162 163+
<132 132–141 142–148 149–154 155–158 159–164 165+
<133 133–142 143–149 150–155 156–159 160–165 166+
<134 134–143 144–150 151–156 157–160 161–166 167+
<135 135–143 144–151 152–157 158–161 162–167 168+
(continued)

147
Table 10.2, continued
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
<136 136–144 145–152 153–158 159–162 163–168 169+
<137 137–145 146–153 154–159 160–163 164–169 170+
<137 137–146 147–154 155–160 161–164 165–170 171+
<138 138–147 148–155 156–161 162–165 166–171 172+
<139 139–148 149–155 156–162 163–166 167–172 173+
<140 140–149 150–156 157–163 164–167 168–174 175+
<141 141–150 151–157 158–164 165–168 169–175 176+
<142 142–151 152–158 159–165 166–169 170–176 177+
<142 142–152 153–159 160–166 167–170 171–177 178+
<143 143–153 154–160 161–167 168–171 172–178 179+
<144 144–154 155–161 162–168 169–172 173–179 180+
<145 145–155 156–162 163–169 170–173 174–179 180+
<146 146–156 157–163 164–170 171–174 175–180 181+
<146 146–156 157–164 165–171 172–175 176–182 183+
<147 147–157 158–165 166–172 173–176 177–183 184+
<148 148–157 158–166 167–173 174–177 178–184 185+
<149 149–158 159–167 168–174 175–178 179–185 186+
<150 150–159 160–168 169–175 176–179 180–186 187+
<151 151–160 161–169 170–176 177–180 181–187 188+
<152 152–161 162–170 171–177 178–181 182–188 189+
<153 153–162 163–171 172–178 179–182 183–189 190+
<154 154–163 164–172 173–179 180–183 184–190 191+
<155 155–164 165–173 174–180 181–184 185–192 193+
<155 155–165 166–174 175–181 182–185 186–193 194+
<156 156–166 167–175 176–182 183–186 187–194 195+
<157 157–167 168–176 177–183 184–187 188–195 196+
<158 158–168 169–177 178–184 185–188 189–196 197+
<159 159–169 170–178 179–185 186–189 190–197 198+
<160 160–170 171–179 180–186 187–190 191–198 199+
<160 160–170 171–179 180–187 188–191 192–199 200+
<161 161–171 172–180 181–188 189–192 193–200 201+
<162 162–172 173–181 182–189 190–193 194–201 202+
<163 163–173 174–182 183–190 191–194 195–201 202+
<164 164–174 175–183 184–191 192–195 196–202 203+
<165 165–175 176–184 185–192 193–196 197–203 204+
<166 166–176 177–185 186–193 194–197 198–204 205+
<166 166–177 178–186 187–194 195–198 199–205 206+
<167 167–178 179–187 188–195 196–199 200–206 207+
<168 168–178 179–188 189–196 197–198 199–207 208+
<169 169–179 180–189 190–197 198–201 202–208 209+
<170 170–180 181–190 191–198 199–202 203–209 210+
<171 171–181 182–191 192–199 200–203 204–210 211+

Tables 10.1 and 10.2 adapted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd
ed. (Boulder, Co: VeloPress), 54, 55.

148
Triathlon Training Technologies | 149

and still have plenty of energy for the run. An elite professional might hold
91 to 95 percent of threshold heart rate, whereas a beginner will need to be
more conservative and keep heart rate lower at around 80 to 85 percent. Of
course, many other factors could affect heart rate, so the triathlete should
do plenty of training with a HRM to understand this critical relationship.
Triathletes and coaches looking for a more reliable tool with less variability
should consider a power meter.

Power Meters
A power meter is fast becoming as ubiquitous a training device as the heart
rate monitor in the triathlon world, and it has certainly proven its worth over
the past 10 years. Power meters measure the amount of cycling work done
by the triathlete in watts. Wattage is force (how hard the triathlete pushes
on the pedals) multiplied by angular velocity (how fast the triathlete pedals,
or cadence). Therefore, the triathlete can create 1,000 watts by pushing very
hard in a big gear at a low cadence or by pedaling very quickly but not very
forcefully. Wattage is the dose in the training dose-response system, and heart
rate is the response. Wattage measures the actual work completed by the rider.
A power meter a near perfect training tool because wattage is not depen-
dent on the weather, amount of sleep, or fatigue. Three hundred watts is 300
watts, whether in the rain, snow, or at the end of a 100-mile (160 km) ride.
Three hundred watts as a work rate is the same for everyone regardless of
size, weight, and fitness. One of the greatest benefits of measuring wattage
is that it tells triathletes exactly their ability to complete work and is easily
comparable with others.
Among the reasons to use a power meter are the following:
• Tracking fitness changes. A power meter allows triathletes to know with
certainty whether fitness is improving and when they have reached a peak.
Measuring improvement by tracking speed or heart rate on a bike leaves
too many variables to influence it, such as wind, temperature, road surface,
elevation profile, packs of riders, and more. With power data, the triathlete
knows for certain whether he is improving.
• Race analysis. A power meter can help analyze race performance. A
coach can easily see when the triathlete burned a match or used too much
energy in parts of the race that weren’t decisive. Did she make a tactical error
in a race but not realize it? Did she pace herself properly?
• Pinpointing strengths and weaknesses. With a power meter, a triathlete
can analyze performance and training to find out what his natural talents
are and whether he should stick with hilly events or flatter ones. By learning
his strengths and weaknesses, the triathlete can determine which event is
best suited to him and identify any holes in his training so that he can excel
in many types of events.
150 } Allen

• Improved interaction between triathlete and coach. A coach can


instantly see what the triathlete is doing in races and training rides and make
suggestions to achieve further improvements. A power meter doesn’t lie!
• Optimize training time. Triathletes who train with a power meter can
concentrate on the workload and find extra motivation to improve their
efforts. For example, the triathlete who watches average watts drop near the
end of a 5-minute effort will likely find additional motivation to achieve a
5-minute wattage goal.
• Pacing of efforts. Power meters allow triathletes to pace their effort
better in interval workouts, hill climbs, and time trials. By knowing threshold
power, triathletes can hold to it like glue in a time trial or hill climb, so they
will know that they went as hard as they possibly could.
• Mobile testing lab. A power meter allows coaches and triathletes to test
on a regular basis, to see quantitatively what areas have improved and what
still need work. By testing regularly, the triathlete can better understand
potential for improvement while avoiding overtraining. Training is testing,
and testing is training. Every training session can be a peak performance.
• Coordinate sport nutrition for best performance. Knowing how much
work (in kilojoules) triathletes do in training allows postexercise meals to be
planned so that energy intake meets energy expenditure. As a result, they
can recover faster and be able to train harder sooner.
One of the downfalls of training with a power meter is becoming obsessed
with the numbers and forgetting that a personal best is not possible on every
ride. The power meter does not lie, so workouts can be mentally challeng-
ing when fitness is not as high as desired. Triathletes tend to be critical of
themselves, so having a “truth meter” on the bike might not be the best idea
for hyper-self-critical triathletes.

Beginning to Use a Power Meter


What should a coach or triathlete do first after purchasing a power meter?
Collect data. The first mission is just to ride with the meter on the bike and
download every ride. Every ride is important, and every download is key to
understanding. The triathlete should not change training for the first week
or two and just follow a normal routine while collecting the data.
These downloads will soon begin to give some valuable information on
current training: how much time is spent in the new power training zones,
at what wattage level the triathlete consistently pedals, what her preferred
cadence is, how many kilojoules it takes before she becomes significantly
fatigued, and more. In this first step, the goal is to learn some basic things
about the triathlete as a cyclist that will apply later in training and help her
improve.
Triathlon Training Technologies | 151

Learning the Functional Threshold Power


After becoming accustomed to the device, testing begins. The second step in
this process is to learn the triathlete’s functional threshold power (FTP). FTP
is defined as the highest power that a rider can maintain in a quasi-steady
state without fatiguing for approximately 1 hour. When power exceeds
FTP, fatigue will occur much sooner, whereas power just below FTP can be
maintained considerably longer. Therefore, FTP is the best average watts for
a 1-hour time trial.
There are many ways to find a triathlete’s FTP, but the best way is just to
go out and hammer all out for an hour and see what can be done. This effort
is a painful but necessary part of the process. Triathletes will need to test
for FTP many times in the future, because this is one of the best ways to see
how much they have improved over time.
An alternative field test is to do 20 minutes as hard as possible and then
subtract 5 percent. This will give a close approximation of FTP.

Establishing Power Training Levels


The third step is to establish power training levels. After FTP is known, the
triathlete can determine power training levels. The power training levels
are anchored on the triathlete’s FTP; level 4 (in the middle of the 7 levels) is
defined at 91 to 105 percent of FTP.3 With FTP as 100 percent, the rest of the
training levels can be determined (see table 10.3). Understanding at what level
triathletes are training is critical to creating the training response that they
and coaches are looking for. If triathletes want to improve their anaerobic
capacity, then intervals should be between 120 and 150 percent of FTP. At
this level they will know that they are training at the correct level to induce
training adaptation.

Table 10.3 Power Training Levels


Level Name/purpose % of threshold power % of threshold HR RPE
1 Active recovery ≤55% ≤68% <2
2 Endurance 56.75% 69.83% 2–3
3 Tempo 76.90% 84.94% 3–4
4 Lactate threshold 91–105% 95–105% 4–5
.
5 VO2max 106–120% >106% 6–7
6 Anaerobic capacity 121–150% N/A >7
7 Neuromuscular power N/A N/A (maximal)

Adapted, by permission, from H. Allen and A. Coggan, 2010, Training and racing with a power
meter, 2nd ed., (Boulder, Co: VeloPress), 48.
152 } Allen

Creating a Power Profile


After establishing training zones, the triathlete can create a power profile.
Figuring out the triathlete’s relative strengths and weaknesses is the fourth
step. In a project with Dr. Andrew Coggan the power profiling chart (table
10.3) was created, initially to help riders compare themselves to the best in
the world and with their racing peers. Once created, however, it became an
important tool in figuring out the riding style of triathletes and the areas
where they need help. Are they better sprinters or time trialists? By compar-
ing wattages with their power profiles, coaches and triathletes can tell exactly
what their strengths and weaknesses are.
Unfortunately, figuring out power profiles involves more testing. To estab-
lish a power profile, triathletes need to test their best 5-second, 1-minute, and
5-minute wattages. These . represent neuromuscular power (level 7), anaerobic
capacity (level 6), and VO2max (level 5) wattages respectively. To produce
the best numbers, the testing should be done on a day when the triathlete is
fresh and not tired from a long block of training. The importance of learn-
ing a triathlete’s power profile cannot be understated. Relative strengths
and weaknesses will determine the training plan going forward to attack
and improve weaknesses and to monitor . the improvement in strengths. For
example, a triathlete with a strong VO2max (level 5) but poor FTP (level 4)
has the potential to improve FTP but has just not trained it sufficiently yet.
Understanding a triathlete’s power profile will direct the training that best
suits upcoming events and addressees the triathlete’s specific needs.

Using a Power Meter as a Pacing Tool


One of the most useful things that a triathlete can do with a power meter is
to use it as a pacing device. Pacing is incredibly important in triathlons of
any distance, and most successful triathletes use a power meter to optimize
pacing. Pacing in the swim, on the bike, and in the run is critical, and the
longer the races are, the more critical pacing becomes to success.
Pacing is a skill learned through trial-and-error experiences on the training
grounds every day and at races. Pacing on the bike can also be learned using
a power meter to give an objective, real-time view. The reason that pacing for
triathlon is difficult is that for races at the Olympic distance and longer, the
perceived exertion at the correct pace is much lower than triathletes know
they can do if they were racing in only one event instead of three.
The rating of perceived exertion (RPE) is an important feeling that every
triathlete needs to associate with power output. One of the best ways found
for triathletes to learn this is by doing progressively longer intervals at dif-
ferent intensities, becoming keenly aware of the feelings associated with each
level of intensity.
The key to pacing for triathlons is learning the proper wattage output to
maintain for the bike leg and still be able to have a good run leg and then
Triathlon Training Technologies | 153

calibrating that with the correct perceived exertion. The problem with this
calibration is that RPE changes daily and is affected by sleep, stress, hydra-
tion level, and more, just as heart rate is affected by those factors. On race
day, RPE will likely be lower than the actual effort. Here a power meter can
play a significant role in a triathlete’s race.
Table 10.4 helps illustrate the appropriate intensities and pace guidelines for
various distances of triathlon races when using a power meter or analyzing
pacing skills in postrace analysis.

Table 10.4 Pacing Guidelines for Triathlon Events


Percentage of FTP Corresponding
Type of Intensity factor as a percentage of Coggan
triathlon Distance (fraction of NP) average power training level
Sprint 10 km (6.2 mi) 1.03–1.07 100–103% 4
Olympic 40 km (24.8 mi) 0.95–1.00 95–100% 4
Half-Ironman 90 km (56 mi) 0.83–0.87 80–85% 3
Ironman 180 km (112 mi) 0.70–0.76 68–78% 3
Double Ironman 361 km (224 mi) 0.55–0.67 56–70% 2

Pacing at the Proper Percentage of FTP


Triathletes relatively new to triathlon or competing in their first Ironman-
distance event are advised to be conservative and maintain a pace between 68
and 71 percent of their FTP. This pace will solidly be in their endurance zone
and allow them to ride comfortably and conserve plenty of energy for the run.
Triathletes who are more experienced and who have been training consistently
for 3 years or more could consider riding between 70 and 74 percent of FTP;
this pace should be just intense enough. They will ride quickly and still have
energy for the run. The most experienced triathletes and those going for a
personal best or for a top placing can safely ride between 75 and 80 percent of
FTP and still be able to run hard. Elite triathletes need to be solidly between
80 and 84 percent of their threshold power to be in contention for the win.

Speed Distance Devices


Speed distance devices are another relatively new technology training tool.
The latest of these are global positioning system (GPS) watches and swim-
ming accelerometers. GPS watches are excellent tools for triathletes because
they display pace on the screen while the triathlete is running so that she
can know immediately whether she is running at too fast or slow a pace and
then adjust quickly. Along with providing instantaneous pace data, a GPS
device records workouts and races in a second-by-second fashion, allowing
triathletes to understand whether they are improving by analyzing the data
in a training analysis software package such as TrainingPeaks.4
154 } Allen

Running Devices
Just as with the heart rate monitor and the power meter, triathletes needs
to learn their threshold pace for running (FTP, or functional threshold pace)
and then use pace zones to train optimally. For runners, the threshold test
is a 15-kilometer run, or about 45 minutes of all-out running, to find FTP.
Running is much harder on the musculosketal system than cycling is and
hence requires only a solid 45-minute test rather than the full hour that a
cycling threshold test requires. After the threshold pace is known, athletes
can use that number to create running pace zones. A couple of common zone
methodologies are available. One of the easiest to understand is Joe Friel’s
pace zones,2 which are shown in table 10.5.

Table 10.5 Running Pace Zones in Minutes and Seconds Per Mile Based on
Recent 5K or 10K Race Times
5K time 10K time Zone 1 Zone 2 Zone 3 Zone 4–5a Zone 5b Zone 5c
14:15–14:44 30:00–30:59 6:38+ 5:52–6:37 5:27–5:51 5:09–5:26 4:37–4:58 4:36–max
14:45–15:14 31:00–31:59 6:50+ 6:02–6:49 5:37–6:01 5:18–5:36 4:45–5:06 4:44–max
15:15–15:44 32:00–32:59 7:02+ 6:13–7:01 5:47–6:12 5:27–5:46 4:53–5:15 4:52–max
15:45–16:09 33:00–33:59 7:13+ 6:23–7:12 5:56–6:22 5:36–5:55 5:01–5:24 5:00–max
16:10–16:44 34:00–34:59 7:25+ 6:33–7:24 6:06–6:32 5:45–6:05 5:10–5:33 5:09–max
16:45–17:06 35:00–35:59 7:36+ 6:43–7:35 6:15–6:42 5:54–6:14 5:18–5:41 5:17–max
17:07–17:34 36:00–36:59 7:48+ 6:54–7:47 6:25–6:53 6:03–6:24 5:26–5:50 5:25–max
17:35–18:04 37:00–37:59 8:00+ 7:04–7:59 6:34–7:03 6:12–6:33 5:34–5:59 5:33–max
18:05–18:29 38:00–38:59 8:11+ 7:14–8:10 6:44–7:13 6:21–6:43 5:42–6:08 5:41–max
18:30–18:59 39:00–39:59 8:23+ 7:24–8:22 6:53–7:23 6:30–6:52 5:50–6:16 5:49–max
19:00–19:29 40:00–40:59 8:34+ 7:35–8:33 7:03–7:34 6:39–7:02 5:58–6:25 5:57–max
19:30–19:54 41:00–41:59 8:46+ 7:45–8:45 7:12–7:44 6:48–7:11 6:06–6:34 6:05–max
19:55–20:24 42:00–42:59 8:58+ 7:55–8:57 7:22–7:54 6:57–7:21 6:14–6:43 6:13–max
20:25–20:49 43:00–43:59 9:09+ 8:05–9:08 7:31–8:04 7:06–7:30 6:22–6:51 6:21–max
20:50–21:19 44:00–44:59 9:21+ 8:16–9:20 7:41–8:15 7:15–7:40 6:31–7:00 6:30–max
21:20–21:49 45:00–45:59 9:32+ 8:26–9:31 7:51–8:25 7:24–7:50 6:39–7:09 6:38–max
21:50–22:14 46:00–46:59 9:44+ 8:36–9:43 8:00–8:35 7:33–7:59 6:47–7:17 6:46–max
22:15–22:41 47:00–47:59 9:56+ 8:47–9:55 8:10–8:46 7:42–8:09 6:55–7:26 6:54–max
22:42–23:09 48:00–48:59 10:07+ 8:57–10:06 8:19–8:56 7:51–8:18 7:03–7:35 7:02–max
23:10–23:37 49:00–49:59 10:19+ 9:07–10:18 8:29–9:06 8:00–8:28 7:11–7:44 7:10–max
23:38–24:04 50:00–50:59 10:31+ 9:17–10:30 8:38–9:16 8:09–8:37 7:19–7:52 7:18–max
24:05–24:34 51:00–51:59 10:42+ 9:28–10:41 8:48–9:27 8:18–8:47 7:27–8:01 7:26–max
24:35–24:59 52:00–52:59 10:54+ 9:38–10:53 8:57–9:37 8:27–8:56 7:35–8:10 7:34–max
25:00–25:24 53:00–53:59 11:05+ 9:48–11:04 9:07–9:47 8:36–9:06 7:43–8:19 7:42–max
25:25–25:54 54:00–54:59 11:17+ 9:58–11:16 9:16–9:57 8:45–9:15 7:52–8:27 7:51–max
25:55–26:29 55:00–55:59 11:29+ 10:09–11:28 9:26–10:08 8:54–9:25 8:00–8:36 7:59–max
26:30–26:49 56:00–56:59 11:40+ 10:19–11:39 9:36–10:18 9:03–9:35 8:08–8:45 8:07–max
26:50–17:19 57:00–57:59 11:52+ 10:29–11:51 9:45–10:28 9:12–9:44 8:16–8:53 8:15–max
27:20–27:44 58:00–58:59 12:03+ 10:39–12:02 9:55–10:38 9:21–9:54 8:24–9:02 8:23–max
27:45–28:14 59:00–59:59 12:15+ 10:50–12:14 10:04–10:49 9:30–10:03 8:32–9:11 8:31–max
28:15–28:44 60:00–60:59 12:27+ 11:00–12:26 10:14–10:59 9:39–10:13 8:40–9:20 8:39–max
Reprinted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd ed. (Boulder, Co:
VeloPress), 49.
Triathlon Training Technologies | 155

Training with a GPS device is similar to training with a power meter


because the athlete needs to test, assess strengths and weaknesses, under-
stand training zones, develop a training plan around those zones, and finally
analyze the data to confirm workout adherence and improvement.

Swimming Accelerometers
Swimming accelerometers are a new entry into the training with technology
market and have merit. An accelerometer is a watch that uses motion sensors
to record data accurately for every lap that the triathlete swims. The device
makes these data easily available for analysis both on the watch screen and
by downloading to a computer.
The accelerometer records total training distance, training load, and calories
burned in every session. It also captures the structure of every session exactly
as the athlete swims it so that coaches get a complete record of every workout
performed in the pool. Accelerometers can monitor stroke rate (strokes per
minute) and facilitate study of how it varies with distance, training pace, and
fatigue. The ability to monitor stroke count on every lap is probably one of
the most helpful features of the watch for postworkout analysis. Split times
taken on every lap can be used to assess pacing skills.
Some accelerometers have easy-to-use data-transfer and analysis software
including a full online training log or diary to record training sessions and data
along with full integration with other training analysis software programs,
such as TrainingPeaks WKO+. For triathletes who use a power meter on the
bike and a GPS or foot pod when running, an accelerometer completes the pic-
ture, allowing them to measure complete training load in all three disciplines.

Training Analysis Software


The software takes a training gizmo and turns it into a training tool. Without
downloading and analyzing the data, a triathlete has nothing more than an
expensive watch or bicycle speedometer. Training analysis allows the tri-
athlete to review individual workouts, look at data over the past season, and
develop a clear understanding of the optimal training load.
Each of the training devices mentioned in this chapter comes with its own
basic training software that generally downloads and displays one day’s worth
of data in graphic form. Each workout is important to understand, and the
analysis should assure that the proper training zones were addressed, that
intervals were completed, and that the correct muscular stress was created
to facilitate the greatest training adaptation. By parsing out the individual
intervals within the analysis software, a triathlete can begin to understand
the training pattern and response to that pattern.
In figure 10.1 a triathlete has completed a set of intervals on the bike that
addressed . the level 4 power training zone (lactate threshold) and the level
5 zone (VO2max), both of which are excellent zones to train for triathletes.
156 } Allen

360
351 Duration: 20:31:00 Pa:HR: 7.26% Min Max Avg
342 Work: 230 kJ Distance: 2.173 mi Power: 0 238 187 watts
333
324 TSS: 37.2 (1.044) Elev. gain: 1094 ft Heart rate: 79 162 127 bpm
315 Norm power: 188 Elev. loss: 0 ft Cadence: 4 86 62 rpm
306
297 VI: 1 Grade: 9.6% (1096 ft) Speed: 3.5 9.7 6.3 mph
288 Pw:HR: 26.4%
279
270
261
252
243
234
225
216
207
198
189
180
171
162
153
144
135
126
117
108
99
90
81
72
63
54
45
36
27
18
9
0
0:00
5:00
10:00
15:00
20:00
25:00
30:00
35:00
40:00
45:00
50:00
55:00
1:00:00
1:05:00
1:10:00
1:15:00
1:20:00
1:25:00
1:30:00
1:35:00
1:40:00
1:45:00
1:50:00
1:55:00
2:00:00
„„Figure 10.1 Analysis E5643/Friel/Fig.
of a set of intervals completed by a triathlete on a bike. This
10.1/448603/TimB/R1
triathlete
. completed two 20-minute intervals at threshold and then did two 5-minute
VO2max intervals to prepare for an upcoming Ironman-distance event.

By using analysis software such as TrainingPeaks WKO+, the triathlete


is able to review data over a longer period. WKO+ software houses charts
detailing performance improvements, which can be more informative than
individual workout results.
One of the most critical charts for a triathlete is the performance manage-
ment chart. This chart explains the relationship between chronic training
load (CTL), or fitness; acute training load (ATL), or fatigue; and training
stress balance (TSB), or performance. Critical to peaking or creating a peak
of fitness is the balance of fitness and fatigue. On the day that a triathlete
wants to peak, fitness should be the highest possible and fatigue should be
low enough to create a personal best performance. (See chapter 23 for more
on this tapering and peaking concept.)
In figure 10.2 the triathlete has properly peaked for two Ironman-distance
events by timing the taper correctly, performing a rational and progres-
sive recovery after the first event, and executing a sane buildup to the next
peak.
Triathlon Training Technologies | 157

min/mi TSB TSSM watts


10:12 90 208 240
9:41 81 197.6 228
9:11 72 187.2 216
8:40 63 176.8 204
8:10 54 166.4 192
7:39 45 156 180
7:08 36 145.6 168
6:38 27 135.2 156
6:07 18 124.8 144
5:37 9 114.4 132
5:06 0 104 120
4:35 –9 98.6 108
4:05 –18 83.2 96
3:34 –27 72.8 84
3:04 -36 62.4 72
2:33 –45 52 60
2:02 –54 41.6 48
1:32 –63 31.2 36
1:01 –72 20.8 24
0:31 –81 10.4 12
0:00 –90 0 0
1/1/2008
1/19/2008
2/6/2008
2/24/2008
3/13/2008
3/31/2008
4/18/2008
5/6/2008
5/24/2008
6/11/2008
6/29/2008
7/17/2008
8/4/2008
8/22/2008
9/9/2008
9/27/2008
10/15/2008
11/2/2008
11/20/2008
12/8/2008
12/26/2008
TSB (TSB) CTL (TSSM) 20 minute pace (min/mi) 120 minute pace (min/mi)
ATL (TSSM) ATL (TSSM) 20 minute power (watts) 120 minute power (watts)

„„Figure 10.2 Performance management chart showing proper peaking for two
Ironman-distance events. E5643/Friel/Fig.
The darker line10.2/448604/TimB/R1
shows the fitness (CTL), the lighter line shows
fatigue (ATL), and the shadow demonstrates the ability to perform (TSB). The solid
line is the triathlete’s best 10 runs for the year. The dashed line represents the best 10
rides for the year.

Mobile Training Applications


Mobile applications currently provide limited analysis features for triathletes
and coaches compared with more robust analysis programs, but they do show
merit for improving communication between triathletes and coaches, which
is a key component of making appropriate training decisions. Improvements
in these apps will likely address these limitations.

Conclusion
Training with technology has become an essential part of triathlon success,
and triathletes who monitor their training dose and response have an edge
on those who do not. Although technology adds complexity to training,
capturing workouts daily and downloading them into training analysis
software is highly beneficial to performance. Optimizing training time
158 } Allen

is one of the greatest benefits of training with technology. Because most


triathletes have tight time constraints, technology has to be considered to
create the greatest opportunity for personal bests on the bike, in the run,
and during the swim.
Pa r t

IV
Physiological Function
in Triathlon Training
This page intentionally left blank.
CHaPTEr
11
Aerobic Capacity
—Ross Tucker, PhD

A erobic capacity is recognized as one of the crucial physiological attributes


for success in endurance sport, to the point that it has perhaps been
overvalued as a predictor of performance . and a marker for training status.
Maximal aerobic capacity, expressed as VO2max, is measured relatively easily
in laboratories and is deemed a good indication . of overall cardiovascular
fitness. When applied to a large population, VO2max is also a strong predic-
tor of performance, but this predictive power diminishes significantly to
the point of being dissociated from performance in a homogenous group of
triathletes, because other factors related to the metabolic, neurological, and
muscular systems contribute to performance.
The result of this preoccupation with measurement, and the ever-increasing
.
understanding of exercise physiology, means that the concept of VO2max is
often misunderstood and overapplied. In this chapter we describe the key
concepts of aerobic capacity, including measurement, interpretation, and
implications for training and performance.

Historical Perspectives .
and Measurement of VO2max
Maximal oxygen consumption is measured using progressive or incremental
exercise trials to exhaustion. Triathletes typically run or cycle at increasing
workloads until volitional fatigue occurs, while inspired and expired air are
measured. The progressive increase in exercise intensity demands increas-
ing levels of muscle activation and metabolic work. The result is that oxygen
consumption increases as a function of intensity. The highest achieved work-
load as well as the highest oxygen consumption measured are recorded and
used to obtain information on the maximal performance and physiological
level of the athlete.
Nobel laureate A.V. Hill introduced
. exercise physiology, and thus coaches
and athletes, to the concept of VO2max in 1925.1 He was able to measure oxygen
consumption during running using a method known as the Douglas bag

161
162 } Tucker

method. Briefly, he ran at a range of progressively increasing speeds while


breathing into a large apparatus resembling a backpack connected to his face
by a tube and mask. He found that as running speed increased, so too did
oxygen use, until he could no longer increase his running speed. He noted
that “the oxygen requirement increases continuously as the speed increases,
attaining enormous values at the highest speeds: The actual oxygen intake,
however, reaches a maximum beyond which no effort can drive it.” Hill sur-
mised that the reason for this maximum was “limitations of the circulatory
and respiratory system.”1
Later, Taylor et
. al. described the presence of a plateau when they observed
an increase in VO2max of less than . 150 milliliters between two successive
workloads, and the concept of a VO2max plateau was created.2 Beyond this
point, at higher exercise intensities, any further energy requirement would
have to be met from oxygen-independent sources, and the theorized anaerobic
limit to exercise would be exceeded.
The question over whether exercise performance is limited by the attain-
ment of this plateau in oxygen consumption, or whether the peak in oxygen
consumption is in fact limited by other factors, is the subject of much debate
in the scientific literature3 and is not our primary focus here. But two impor-
tant practical points should be considered, both with . implications for the
understanding of measurement and application of VO2max.
First, the presence
. of a plateau is actually relatively rare during labora-
tory tests of VO2max (see figure 11.1) and may be the exception rather than
the norm. For example, Lucia et al. found that less than half of elite cyclists
displayed a plateau4 and concluded, “In a good number of highly trained
humans, the main factor limiting maximal endurance might not necessar-
ily be oxygen-dependent.”4 The prevalence of the plateau has been found to
be even lower in Olympic-level runners, and was found in only 12 out of 71
subjects in another study.
.
5

Second, tests for VO2max terminate despite submaximal levels of muscle


activation. That is, muscle fibers remain inactive at the point of volitional
exhaustion, which is significant because these inactive fibers could be
recruited to continue exercise, with a resultant increase in oxygen consump-
tion. That this does not occur has led to the theory that exercise is terminated
by
. the brain at submaximal levels of muscle activation and that the measured
VO2 at this point is a consequence, rather than a determinant, of the underly-
ing physiology and performance.6 For the same reason, the highest. oxygen
consumption measured in a test to fatigue is sometimes called the VO2peak,
to distinguish it from a max where the plateau would supposedly be reached.3
. Regardless of the model or theoretical explanation, the measurement of
VO2max can provide important information on training status and, poten-
tially, performance levels in triathletes because the peak speed or power
output achieved as well as the oxygen measurement are both performance out-
comes. Thus, as part of a controlled training program, repeated measurements
can be conducted to identify training responses and predict performance.
Aerobic Capacity | 163

.
„„Figure 11.1 Testing for VO2max.

.
Protocols for Measurement of VO2max
.
A number of protocols exist for the measurement of VO2max. The selection
of protocol has important implications for the measured aerobic capacity
and, perhaps more important,
. the maximal power output or running speed
that is associated with VO2max. For example, Lucia et al. used a protocol for
elite cyclists that began at a power output of only 25 watts and increased by
25 watts every minute until exhaustion,4 whereas Hawley and Noakes used
a protocol that began at 3.3 watts per kilogram, increased by 50 watts after
2.5 minutes, and increased again by 25 watts every 2.5 minutes.7 For cyclists
who reach 400 watts, both protocols would be expected to last approximately
20 minutes, but the former begins at a significantly lower power output and
may be more suitable for sedentary people, whereas the latter entails spend-
ing far longer at a given power output, which may facilitate measurement
of steady-state physiology during the trial. Hawley and Noakes, using this
164 } Tucker
.
protocol, established an accurate equation for the prediction of VO2max using
peak power output:
.
Estimated VO2max (L/min) = 0.01141 × PPO + 0.435.7
Padilla et al.8 have used an intermittent protocol on world-class cyclists,
using a starting power output of 110 watts, increasing by 35 watts every 4
minutes, and allowing a 1-minute recovery interval after each stage. This
protocol is designed to last approximately 1 hour for elite cyclists.
. Clearly,
this has implications for the measurement of work rates at VO2max, as well
as the metabolic demand of exercise, and should be borne in mind by coaches
and scientists wishing to assess aerobic capacity and performance.
An additional factor to consider for triathlon is the exercise modality used
for
. testing. In the same way that heart rate is affected by exercise modality,
9

VO2max measured during running is greater than that measured during


cycling and swimming, because of the involvement of a larger muscle mass
and the weight-bearing nature of running.
. (Cross-country skiing is generally
considered to produce the highest VO2max.) But this result is highly depen-
dent on the state of training, because this influences . the ability to achieve
the high workloads necessary to produce a higher VO2max. For example, a
cyclist performing a running protocol would be expected to underperform
relative to his performance using a cycling protocol, and vice-versa for
runners.
Triathletes may thus have context-dependent aerobic capacity, depending
on the method of assessment and state of training within each of the three
disciplines at the time of testing. This has important implications for the
prescription of training zones based on aerobic capacity.
.
VO2max Benchmarks
.
Table 11.1 depicts the measured VO2max levels in elite triathletes, as well as
moderately trained and sedentary people in various sports. Large differences
clearly exist between high-performing triathletes and sedentary people. As
noted later in
. this chapter, these differences are partly a result of differences
in starting VO2max levels (genetically influenced) and partly because of the
response to training, which has a genetic component.
Differences in training status make direct
. comparisons between studies
and modalities difficult. In triathletes, VO2max values have been found to
range between 39 to 49 milliliters per kilogram per minute during tethered
swimming, 57 to 61 milliliters per kilogram per minute during cycling, and
61 to 85 milliliters per kilogram per minute during running.10 These values
are marginally lower than values measured
. in single-event endurance spe-
cialists. Males tend to have higher V.O2max values than females, even when
11

expressed relative to body weight. VO2max increases into adulthood, and


adolescents have significantly lower aerobic capacity than adults do.
Aerobic Capacity | 165
.
Table 11.1 Varying Levels of VO2max
.
Athlete and level VO2 max (ml ∙ kg−1 ∙ min−1) Reference
Elite male triathletes measured during 78.5 ± 3.6 12
running
Elite male triathletes measured during 75.9 ± 5.2 12
cycling
Elite junior (<18 yr) triathletes tested 67.9 ± 5.9 (males) 13
during running 56.1 ± 2.4 (females)
Junior regional level triathletes 62.7 ± 2.5 14
Elite Tour de France cyclists 72 ± 1.8 with a range from 62.5 to 82.5 15
Elite Tour de France champion 71.5, measured 3 months post Tour 16
Olympic-level runners (middle and long 79.1 ± 0.7 (males) 5
distance) 66.1 ± 1.2 (females)
Elite junior swimmers (<18 yr) 61.6 ± 3.6 (males) 13
52.1 ± 3.6 (females)
Sedentary people during running 30.6 ± 7.9 17, 18, 19, 20

. Of significance is the observation that a relatively large range of measured


VO2max values can . occur within the elite population, as shown by Lucia et
al. in which the VO2max of elite cyclists differed by 20 milliliters per kilo-
15

gram per minute (over 30 percent). As mentioned previously, other factors


contribute to performance, such as running economy (see chapter 12) and the
ability
. to sustain a high relative intensity during competition. The result is that
VO2max is a poor predictor of performance within a relatively homogenous
or narrow performance range of individuals.

Aerobic Capacity, Aerobic Metabolism,


and Training Responses
Apart from providing an indicator of cardiorespiratory and performance
capacities, the measurement of aerobic metabolism during progressive exer-
cise has numerous other important applications. One is the determination of
running economy or cycling efficiency (described in chapter 12), and another
is to allow the determination of the metabolic cost and fuel utilization during
exercise.
As exercise intensity increases, oxygen consumption rises because of the
increasing metabolic demand. The oxidation of carbohydrate and fat meets
this energy demand and can be estimated based on the measurement of the
respiratory exchange ratio, or RER. (Note that this is different from respira-
tory quotient, or RQ, in that RER is measured at the mouth using expired
gases, whereas RQ refers to the measurement taken at the tissue level.) RER
is calculated as the ratio of oxygen consumed to carbon dioxide produced
and, because of the metabolic reactions involved in the oxidation of carbo-
hydrate and fat, provides an accurate estimate of the relative contribution of
166 } Tucker

each to energy. The exclusive oxidation of carbohydrates results in an RER


of 1, whereas the exclusive oxidation of long-chain fatty acids results in an
RER of 0.7.
Therefore, when exercise is performed below the anaerobic threshold, RER
will range between 0.7 and 1.0. At some point RER rises above 1, reflecting
the liberation of carbon dioxide from the body’s bicarbonate pool because of
acidosis, and is interpreted as an indication of anaerobic metabolism, the point
at which the anaerobic threshold has been crossed (see chapter 13). Beyond
this point, the estimation of energy use is fraught with difficulty, requiring
assumptions or more invasive measurement. As a result, economy and fuel
utilization are typically measured at submaximal levels.
Returning to RER, the contribution of fat and carbohydrate to energy
metabolism can be estimated using table 11.2. Here, each RER value cor-
responds to a percentage of energy contributed by both carbohydrate and
protein. (Note that these do not factor in protein contributions to energy,
which may be a factor for ultraendurance activity.)
For example, at an RER of 0.76, 18.4 percent and 81.6 percent of the energy
consumed are from carbohydrate and fat sources, respectively. This can be
converted into a caloric equivalent, in which the energy consumption in
kilocalories per liter of oxygen is estimated. (At an RER of 0.76, for example,
each liter of oxygen contributes 4.751 kilocalories of energy.)
As exercise intensity rises, a larger proportion of energy demand is met
by carbohydrate until, at an RER of 1.0, energy is entirely provided by car-
bohydrate sources. One of the benefits of endurance exercise training is to

Table 11.2 Respiratory Exchange Ratios Showing Fat


and Carbohydrate Contributions to Energy Metabolism
RER Cho % Fat % kCal/L O2
0.7 0 100.0 4.686
0.72 4.8 95.2 4.702
0.74 11.6 88.4 4.727
0.76 18.4 81.6 4.751
0.8 25.2 74.8 4.776
0.82 38.8 61.2 4.801
0.84 45.6 54.4 4.825
0.86 52.4 47.6 4.850
0.88 59.2 40.8 4.899
0.90 66.0 34.0 4.924
0.92 72.8 27.2 4.948
0.94 79.6 20.4 4.973
0.96 86.4 13.6 4.998
0.98 93.2 6.8 5.022
1.0 100.0 0 5.047
Aerobic Capacity | 167

improve the body’s capacity to oxidize fat as a fuel, which allows higher
exercise workloads to be achieved before this point is reached. The metabolic
adaptations allowing increased fat oxidation are explained in more detail
in chapter 14, but the result is that highly trained triathletes use glycogen
at lower rates, which has implications for fatigue and performance (chapter
15). Indeed, the capacity to use fat is one of the crucial attributes that enables
ultradistance triathlon performance.

Effects of Training on Aerobic Capacity


.
Some dispute has arisen about the capacity to increase VO2max with train-
ing because some studies have shown a large range of responses among
individuals. These results may have occurred because of differences in the
training status of athletes before the controlled training program, as well as
genetic differences among people.17
Three large-scale
. studies have looked at how an aerobic training program
can improve VO2max in previously sedentary people.17,19,20 Collectively, the
Heritage studies, the DREW study, and . SSTRIDE studies have found that a
5-month training program increases VO2max by 15.2 percent plus or minus 9.7
percent,17,19,20 and that further improvements are likely with a longer training
period and possibly the introduction of higher intensity training.
But the trainability of aerobic capacity varies significantly among individu-
als.
. For example, approximately one in seven people (14 percent) improved
VO2max by less than 8 percent compared with baseline, whereas 8 percent
of the population improved by almost 30 percent.17

Genetic Factor
These disparate responses are strongly associated with genes. Using a
genomewide association approach, it was found that 21 single-nucleotide
polymorphisms,
. or SNPs (gene variants), account for 49 percent of the train-
ability of VO2max.17 Those who carried 9 or fewer of the 21 SNPs were found
to improve by less than 10 percent,
. whereas those who carried 19 or more of
these alleles improved their VO2max by 30 percent.17 Clearly, the presence of
certain SNPs has a strong influence on the response to training and points
to the importance of genetic factors, and thus talent identification. It. seems
reasonable to suggest that those who show a greater improvement in VO2max
may also show the largest improvements in performance with training.
.
Effect of Training on VO2max
.
For this reason it is important to recognize that the measurement of VO2max
involves a performance test. Because oxygen consumption is a function of the
exercise work rate (running speed or cycling power output) achieved, then
it makes sense that an effective training program that improves endurance
168 } Tucker
.
performance will also result in an increase in VO2max as a result of reaching
higher exercise intensities after training.
.
The 16 percent improvement in VO2max in three large cross-sectional
studies is largely because of an increase in performance capacity, which is a
function of cardiorespiratory factors as well as muscle, metabolic, and neu-
rological adaptations to training. The measurement of aerobic capacity with
training is therefore not simply a function of changing cardiorespiratory
function; other systems also contribute to performance.
Because of this complex interaction between performance and aerobic
capacity,
. changes in performance can often be dissociated from changes in
VO2max. One reason may be that training improves economy or efficiency, as
we shall discuss in chapter 12. The result is that at the same workload, oxygen
consumption will be reduced. Further, any increases in performance . because
of training may not necessarily be reflected as an increase in VO2max. For
coaches and triathletes, the value of the test should be extended beyond. the
measurement of aerobic capacity to the exercise intensity at which VO2max
is measured, because this is a performance measure.
.
Training to Improve VO2max
.
In terms of the most effective training method for improving VO2max, recent
research has found that aerobic, high-intensity interval training is more effec-
tive than low- and moderate-intensity training in trained athletes.21 In this
study, four methods of training were compared:
1. Long, slow distance running—continuous running at 70 percent of HRmax
for 45 minutes
2. Lactate threshold running—approximately 25 minutes of running at 85
percent of HRmax
3. Short interval running—47 repetitions of 15 seconds of running at 90 to
95 percent of HRmax with 15 seconds of slow running recovery
4. Long interval running—4 repetitions of 4 minutes at 90 to 95 percent of
HRmax with 3 minutes of recovery
All training sessions were performed 3 days a week for 8 weeks. The largest
increases in aerobic capacity were achieved with the higher intensity
. train-
ing (the short and long interval sessions), which increased VO2max by 5.5
percent and 7.2 percent respectively,
. probably because of increases in stroke
volume (see figure 11.2). VO2max was unchanged by the long, slow distance
and lactate threshold running.
A similar finding has been made for cycling exercise, performance, and
metabolic changes associated with training,22 in which higher intensity inter-
val training produces similar effects to traditional endurance training with
less than a third of the training time (see chapter 14). This suggests that the
addition of higher intensity work, or possibly even the replacement of lower
Aerobic Capacity | 169

„„Figure 11.2 Running high-intensity intervals.

volume training, may have a timesaving effect, as well as being physiologi-


cally more effective.
But this result does not negate the need for longer duration endurance
training, because other benefits may accrue during lower intensity training
(running economy was shown to improve similarly in all four groups, for
example), and practical reasons may come into play as well. The risk of injury
is higher during sprints, which may make them unfeasible for triathletes
wishing to improve performance and become more competitive.

Conclusion
The measurement of oxygen use as an indication of performance is relatively
simple and has some important implications for the identification of inherent
ability and for tracking the physiological changes that occur
. with training.
The most commonly used index is aerobic capacity, or VO2max, . which can
provide valuable information for coaches and scientists. But VO2max should
be interpreted in the context of performance, rather than being viewed as the
limiting determinant of exercise ability. Other indices include the measure-
ment of fuel substrate utilization and energy use during exercise. Aerobic
170 } Tucker

function during triathlon is an important component of performance, but it


should not be viewed in isolation from other key attributes, such as economy
and anaerobic threshold, which are discussed subsequently.
CHAPTER
12
Economy
—Ross Tucker, PhD

E conomy refers to how efficiently oxygen is used at submaximal intensities,


and it provides an indirect measure of the energetic cost of swimming,
cycling, or running. In running and swimming, economy is measured as
the volume of oxygen used to cover a given distance at submaximal intensi-
ties, whereas for cycling economy can be expressed as mechanical efficiency
because it is possible to make accurate measurements of useful work (power
output) as well as total energy use by means of respiratory gas analysis.
Therefore, in the same way that a car is economical if it uses less fuel at a given
speed, humans are classified as economical when oxygen consumption (and
therefore energy consumption at submaximal speeds) at a given workload is
lower than it is for other athletes.

Measuring Economy
Running economy is normally measured as the volume of oxygen used
per kilogram per kilometer. For example, if an athlete with a mass of 70
kilograms uses 4 liters of oxygen per minute when running at a speed of 18
kilometers per hour (33:20 10K pace), economy is calculated by first convert-
ing the volume of oxygen used to a value relative to body mass (64 milliliters
per kilogram per minute) and then expressing it as the total oxygen used
to cover 1 kilometer. In this case, the athlete would have an economy of 190
milliliters per kilogram per kilometer, which is a typical value for an elite
athlete whose running economy ranges between 180 and 210 milliliters per
kilogram per kilometer.1 A slower athlete would use much more oxygen to
accomplish the same pace.
Swimming economy is similarly measured as the oxygen consumption
required to cover a given distance, usually a meter. This value for oxygen
consumption can easily be converted to an energy cost in kilojoules, and
the energetic cost can be expressed per meter swum. For example, studies
of elite swimmers have found that at a speed of 1.5 meters per second, the
energy cost of swimming is 1.23, 1.47, 1.55, and 1.87 kilojoules per meter for
freestyle, backstroke, butterfly, and breaststroke, respectively.2 Freestyle is

171
172 } Tucker

thus the most economical stroke, approximately 50 percent more economical


than breaststroke, the least economical stroke.
For cycling, it is possible to calculate gross or delta efficiency because,
unlike for running, power output can be accurately measured. Cycling
efficiency is then calculated as the ratio of the useful work done (the power
output) to the total energy expended. For example, consider a cyclist whose
oxygen consumption is measured as 4 liters per minute while cycling at
300 watts on a cycle ergometer. The total energy cost of producing this
power output can be calculated. To do this, we need to measure the respi-
ratory exchange ratio, or RER, which provides an indication of the relative
contribution of carbohydrate and fat to energy supply (see chapter 11). In
our example, assume that RER is measured as 0.85, which implies that
approximately 50 percent of the energy comes from fat and 50 percent from
carbohydrate sources.
At that RER, the caloric equivalent, defined as the energy consumed per
liter of oxygen, is 4.862 kilocalories. Knowing that the cyclist is consuming
4 liters of oxygen per minute, we can calculate that energy consumption is
equal to 19.45 kilocalories per minute. This measure is easily converted into
joules per second, or watts, and we can then calculate that the total energy
consumption of our cyclist is 1,335 joules per second, or 1,335 watts. Therefore,
because the cyclist is producing 300 watts of useful work, the gross efficiency
of the cyclist is calculated as 22.1 percent.

Significance of Economy
Economy is increasingly recognized as a crucial determinant of swimming,
running, and cycling performance. Despite its recognized importance, how-
ever, economy remains a relatively poorly understood physiological attribute.
In a 2007 review of running economy published in Sports Medicine, exercise
physiologists Carl Foster and Alejandro Lucia described running economy
as “the forgotten factor in elite performance.”1
In an endurance activity such as triathlon it is clearly beneficial to use less
oxygen and energy at a given intensity because performance is regulated in
part by the ability of the body to supply and use oxygen and provide energy
for work over the long duration of endurance exercise. Studies of elite cyclists
in the Tour de France reveal that
. the best performing cyclists often have typi-
cal or even lower levels for VO2max (see figure 12.1) but are distinguished
instead by their exceptionally high cycling efficiency.3 That is, they use less
oxygen at a given workload than their peers do and are able to convert a
greater proportion of total energy into useful work or power output.
. Logically, of course, those who are most economical would also have lower
VO2max values. By definition, the most economical athletes will use the least
oxygen at a given workload, so even at peak intensities oxygen consumption
will
. be significantly lower. For that reason in elite athletes economy and
VO2max typically show an inverse relationship.3, 4
Economy | 173

© Pierre Teyssot/Action Plus/Icon SMI


. „Figure 12.1 The best cyclists at the Tour de France typically have lower levels for
„
VO2max and use oxygen more efficiently.

Differences in running economy are also used to explain differences


in running performances and to account for . the observation that within
a performance-matched group . of runners, VO2max is a poor predictor of
performance. But because VO2max can be easily measured and perhaps
more important because it can be improved with training, it has received
disproportionate attention from exercise physiologists, coaches, and athletes,
when devoting more attention to running economy may unlock better results,
particularly for triathletes

Training Influences on Economy


Despite the recognized importance of economy on performance, data are
scarce on how and why training alters economy in each of the three triath-
lon disciplines. In isolation, both running and cycling economy, which have
been extensively studied, appear to have the potential to be improved with
training, although the magnitude of the possible improvement is debated in
the scientific literature.
But because of the multidisciplinary nature of triathlon, it may be that
there is more scope to train to overcome the normal impairments in economy
associated with training the different disciplines, in particular the transi-
tion between cycling and running. That is, it has been found that running
174 } Tucker

economy is impaired during running after cycling compared with running


in isolation.5, 6 This impairment, however, appears to be related to the abil-
ity level and experience of the triathletes; elite triathletes display little to no
change in running economy after cycling.7
Training may thus attenuate the normal impairment in running economy
after cycling, and this topic is discussed subsequently. But first we briefly
consider the three disciplines independently.

Swimming Economy
Research has shown that both physiological and biomechanical factors influ-
ence swimming economy. These factors include body length, body mass, body
surface area, buoyancy (a function of body composition), passive torque, and
differences in swimming technique and stroke length. (See reference 8 for
discussion.) Clearly, some of these factors cannot be altered by training and
are a function of body size and dimensions, emphasizing the importance of
talent identification and selection.
Swimming technique is clearly a crucial component of swimming economy.
The metabolic power needed to swim at a certain velocity depends on pro-
pelling efficiency, gross efficiency, and drag, all of which can be influenced
through equipment and correct swimming technique, particularly a focus
on body position or angle in the water8 (figure 12.2). Coaches inherently rec-
ognize that efficient technique is crucial for triathlon success. The ability to
swim with lower oxygen use, longer strokes, and reduced drag points to the

„„Figure 12.2 Correct swimming technique, especially body angle, is critical to


improved economy.
Economy | 175

value of focusing on correct technique for triathlon. For more on effective


and optimal swimming technique, refer to chapter 4.

Cycling Efficiency
There has been much debate about the potential to improve cycling efficiency
with training. A recent review by Hopker9 concluded that in well-controlled
and rigorously designed studies, training has a positive influence on gross
efficiency, defined as the ratio of power output to energy expenditure. (Note
that this measure differs from delta efficiency, which is the ratio of the change
in power output to the change in energy expenditure, and thus includes
metabolic rate at rest, whereas gross efficiency does not.)
The specific mechanisms by which training may improve cycling efficiency
have yet to be determined but may include muscle fiber type alterations,
changes to muscle fiber shortening velocities, changes within the mitochon-
dria, and neural characteristics.9, 10, 11, 12, 13, 14 Recent research has also found a
correlation between maximal muscle strength and efficiency, particularly in
older cyclists, so it appears that training interventions can improve cycling
efficiency.11
In terms of cycling-specific training, changes in gross efficiency are sig-
nificantly correlated with both the volume and the intensity of training, but it
appears that high-intensity training is most strongly related to improvements
in gross efficiency. For example, as little as 6 weeks of high-intensity train-
ing (above OBLA) increases gross efficiency in endurance-trained cyclists by
approximately 1.5 percent.14 This margin is similar to that found when com-
paring the gross efficiency of professional and elite cyclists,15 the efficiency
of trained and untrained cyclists,12 and the change in efficiency that has been
found to occur during a competitive cycling season.13
Strength is another factor identified as predictive for cycling efficiency.
Younger cyclists have been found to be more efficient compared with older
cyclists,10, 11 but this difference disappears if the older cyclists perform strength
training,11 which suggests that one of the key predictors of cycling efficiency,
at least in older athletes, is maximal muscle strength.
In this study,11 the cycling efficiency in master cyclists (over the age of 50)
improved within 3 weeks of beginning a strength training program consisting
of three weekly sessions of leg extension exercises performed at 70 percent
of a predetermined one-repetition maximum. Ten sets of 10 extensions were
performed, with a 3-minute recovery between sets, and this was sufficient
to improve cycling efficiency from 19.66 percent to 22.86 percent. In young
cyclists (26 years old), the same strength training program produced a small
but nonsignificant improvement in cycling efficiency.
In terms of the effect on performance, mathematical methods have been
used to estimate that a 1 percent improvement in efficiency would result in
a 63-second improvement during a 40-kilometer time trial,16 although this
potential performance advantage has yet to be assessed in controlled studies.
176 } Tucker

Running Economy
Training is thought to improve running economy primarily because of neuro-
muscular adaptations including improved muscle power and more efficient
use of elastic energy stored during the stretch-shortening cycle.17 Running-
specific training improves running economy because of learned patterns of
motor control and improved coordination, and runners are most economical
at the speeds that they are more familiar with as a result of training.
Recent studies have shown that resistance and plyometric training pro-
duce significant improvements in running economy within relatively short
durations, ranging from 6 weeks to 14 weeks (as reviewed by reference 17).
Plyometric training, including explosive exercise such as hopping, jumping,
and bounding, is thought to be effective because it improves the function
of the neuromuscular system during running18, 19 (see figure 12.3). Improve-
ments include increased levels of preactivation (muscle activity before ground
contact), increased stiffness of the leg, and enhanced function of the stretch-
shortening cycle, in which an eccentric contraction is followed immediately
by a concentric contraction.18, 19 These changes combine to produce measurably
lower ground contact times and changes in EMG activity, and are thought to be
the primary mechanism for improved running economy because of training.
Another intervention that has received much attention is flexibility train-
ing. Theories differ about how stretching and improving flexibility may
influence running economy. One theory holds that stretching will reduce the
visco-elastic properties of muscles and tendons, making it easier to move the
limbs. On the other hand, improved flexibility may reduce the ability of the
muscle–tendon to store and harness elastic energy, which may reduce running
economy.17 Running economy has been negatively correlated with flexibility—
economical runners tend to have reduced flexibility.20, 21 But these studies are
correlative only, and intervention studies are required to determine whether
changes in flexibility affect running economy in the same predicted manner.
Only three controlled studies have examined this question. Two found that
flexibility training had no influence on economy,22, 23 whereas a third showed
that running economy was in fact improved by a stretching program.24 But
in this last study, the volunteers all had tight hip extensor and flexor muscles
to begin with, and economy was measured immediately after stretching.
Subsequent reviews have suggested that stretching may acutely improve
running economy but that a stretching program has no long-term effect on
it. Further research studies are required to examine this theory. Evidence is
presently insufficient to warrant the incorporation of stretching as a means
to improve economy, although there are other reasons, of course, to advise
regular stretching.
For comprehensive reviews on running economy and the factors influenc-
ing it, readers are referred to review articles by Foster and Lucia1 and Bonacci
et al.17
Economy | 177

„„Figure 12.3 Plyometric training improves the function


of the neuromuscular system and economy.

Neuromuscular Function in Triathletes—


the Effect of Multidiscipline Training
One of the challenges in triathlon training is the requirement to alternate
between disciplines. This switching has implications for economy, in particu-
lar the neuromuscular function that may underpin it. For example, trained
triathletes display muscle activation patterns that are more similar to novice
cyclists than trained cyclists.25
This result occurs even when the cycling training loads are matched, so it
is not simply a case that triathletes perform relatively less cycling training.
Specifically, triathletes show more variation between pedal strokes and more
muscle coactivation (quadriceps and hamstrings being active simultaneously,
for example) than trained cyclists, even though cycling loads are similar.
This finding suggests that multidisciplinary training interferes with
learning the patterns of muscle activation during cycling. This conclusion
is supported by studies of motor learning, which have found that when
two activities are practiced within a short time of one another, learning the
second task is biased by the previous task and that training adaptations can
178 } Tucker

be partly overwritten by interference. The same patterns are not observed for
running, which may be a function of great variability of running kinematics
and muscle activation patterns to begin with but may also reflect the difficulty
of conclusively finding these differences in research studies.
Note that the studies investigating differences between triathletes and
cyclists or runners are typically underpowered (small sample number), short
in duration, and unable to control for training volume and intensity precisely.
Therefore, stronger evidence for these theories remains absent.
Practically, one implication of this observation of interference is that
improvements in economy may be slowed by a multidisciplinary approach.
Therefore, a triathlete who is seeking improved performance in one discipline
through improved economy or technique may need to consider prolonged
periods of single-discipline focus, when she can attempt to address this dis-
cipline without interference from the other two. This approach is speculative
and has the obvious downside of reducing focus and training time on two
disciplines, although it may be considered in specific cases.

Effect of Prior Cycling


on Running Economy
In support of the concept that multidiscipline training exerts a negative effect
on economy, it has been found that triathletes show reduced running economy
when running after cycling compared with running only.5, 6 This result seems
to be associated partly with changes in muscle activation patterns and the
resultant changes in running kinematics, but it may also be influenced by
metabolic and whole body physiological responses that occur during cycling,
such as glycogen depletion, ventilatory muscle fatigue, dehydration, and leg
muscle fatigue during the cycling leg.5, 6
Recent work has shown that this impairment in economy is not present in
elite triathletes. They use the same volume of oxygen during a run after a short
low-intensity cycle, a longer high-intensity cycle, and during an isolated run
and have the same patterns of muscle activation in all the conditions. There-
fore, cycling did not influence running economy or neuromuscular control.26
This study is significant for triathlon performance and training because the
ability to run with maximum efficiency after the cycle leg is crucial for per-
formance. The extent of the change appears related to experience and ability
level because the best performing triathletes show the smallest impairments
in running economy off the bike.26, 27
No doubt this result occurs in part because of the training level of the ath-
letes, which suggests that training to improve the ability to run economically
immediately after cycling is crucial for triathlon success. But some athletes
may have some innate characteristics, and it has been suggested that tri-
athletes who are able to preserve their neuromuscular control and running
economy after cycling may be those who could be most successful at the sport.
Economy | 179

Conclusion
Economy is a crucial component for success, particularly in triathlon, in
which multidiscipline training may compromise this variable in any given
discipline. Although in part genetically influenced, economy can clearly be
improved by a combination of specific training, strength training, plyometrics,
and high-intensity training, and this should be a focus for those wishing to
achieve a higher level of performance. When combined with high aerobic
capacity, economy is a powerful predictor of endurance performance. There-
fore, technical aspects of training for each discipline should be prioritized
for optimal performance.
This page intentionally left blank.
CHAPTER
13
Anaerobic Threshold
—Ross Tucker, PhD

T he concept of the anaerobic threshold dates back over 50 years, to research


published by Wasserman and McIlroy.1 Because of differences in termi-
nology, confusion regarding the identification and definition of thresholds,
and the recognition that there may be more than one threshold, uncertainty
surrounds the precise meaning of the anaerobic threshold, and more impor-
tant, its physiological. significance. As was the case for aerobic capacity and
the measurement of VO2max, a tendency to rely on measurable physiological
changes often leads coaches and athletes to overvalue these concepts. Aerobic
capacity, economy, and anaerobic threshold are three sides of a triangle. The
elite triathlete possesses all three.
. The ability to sustain a high relative workload, as a percentage of either
VO2max or peak exercise work rate, for long periods is crucial to success in
triathlon. We have seen that within a group of similarly performing endur-
ance athletes, aerobic capacity is often a poor predictor of success. One of the
reasons for this is the crucial importance of being able to sustain high relative
exercise intensities in competition. Regardless of the definition, the threshold
concept is a key success factor for triathlon. In this chapter, we describe the
concept, measurement, and significance of the anaerobic threshold.

Defining Anaerobic Threshold


The anaerobic threshold (AT) is defined as either the oxygen consumption or,
more practically, the exercise intensity beyond which any further contribu-
tions to total energy supply come from oxygen-independent sources.1, 2 That
is, the AT is the maximum exercise intensity that the triathlete can maintain
without producing a significant increase in lactate above resting levels.
Recall from chapter 11 that energy demand is a function of exercise inten-
sity and is met by the oxidation of fat and carbohydrate. As exercise intensity
increases, the demand for energy requires that oxygen-independent, or so-
called anaerobic, metabolic pathways contribute to the total energy supply.
As a result, measurable physiological changes occur, some of which are used
as markers to identify the AT intensity.

181
182 } Tucker

Marker 1: Lactate
One of these markers is an exponential increase in the concentration of lac-
tate in the blood, used most widely to identify the AT. Lactate is a molecule
formed as the end-product of glycogenolysis and glycolysis, and it is often
implicated as a limiting factor for exercise, despite evidence suggesting that
it has beneficial effects and is in fact a crucial substrate for energy production
and a signaling molecule (as discussed later).
Lactate has also been wrongly blamed for the acidosis that occurs when
exercise intensity increases beyond the so-called anaerobic threshold. In
reality, lactate is only indirectly involved in the decrease in pH that occurs
during high-intensity exercise, and the acidosis actually occurs because the
formation of lactate involves the transfer of hydrogen ions to coenzymes. But
lactate remains implicated as the key marker of the anaerobic threshold, and
for this reason AT is also called the lactate threshold.

Marker 2: Increase in the Respiratory


Exchange Ratio
Another physiological variable used to detect the anaerobic threshold is an
increase in the respiratory exchange ratio (RER) above 1.0.2, 3 This is the point
at which the volume of carbon dioxide produced exceeds oxygen used as a
result of liberation of carbon dioxide from the body’s bicarbonate buffer pool.
(See chapter 11 for a more detailed explanation of RER.) The excess CO2 is
breathed out, and the result is the increase in RER above 1.
In the absence of direct lactate testing using blood samples, many labora-
tory studies infer the anaerobic threshold using this method, which has been
shown to give comparable results to blood measurement and has the benefit
of being noninvasive.2, 3

Marker 3: Onset of Blood Lactate Accumulation


Onset of blood lactate accumulation (OBLA) is the point at which the lactate
concentration begins to increase exponentially, using mathematical definitions
to identify this point.3 This definition is much the same as that for anaerobic
or lactate threshold, as described earlier. But in the original research this point
happened when blood lactate concentrations were approximately 4 millimoles
per liter, so the concept of OBLA being the point at which blood lactate levels
reaches 4 millimoles per liter has persisted, although this was not originally
intended as the definition. As a result, OBLA differs from anaerobic threshold in
some studies, although it was and still is intended to convey the same concept.

Marker 4: Ventilatory Threshold


In addition to the anaerobic threshold, the ventilatory threshold (VT) has also
been identified. (There are in fact two VTs.) VT has numerous mathematical
Anaerobic Threshold | 183

and physiological definitions,3, 4 but it is generally identified as the point at


which the ventilation increases at a more rapid rate as a function of workload.
That is, it is the ventilatory equivalent of lactate in the LT definition, and the
workloads eliciting the AT and VT are generally similar, but differences in the
method used to identify them,5 as well as a significant subjective component,
mean that the workloads at which they occur are slightly different.
This difference has implications for the prescription of training zones
based on LT or VT because they are a function of the method used to measure
them. Coaches and triathletes implementing one or more of these markers
to prescribe training volume and monitor performance and adaptations to
training thus need to understand that they are defined somewhat arbitrarily.
Comparisons within an athlete for tracking changes over time should be
made using the same threshold or marker, as well as the same method of
determining the threshold.

Other Markers
Other functional applications of the threshold concept include that of func-
tional threshold power, or FTP, which has been defined as the highest aver-
age power output or pace that an athlete can sustain for 1 hour. FTP is thus
similar in concept to AT, which is defined as the highest intensity that does
not cause a rise in lactate concentration, because the requirement to sustain
power output for 1 hour means that FTP is at or near the level where this
physiological change would occur.
In our laboratory, we have found that the average power output achieved
for a maximal 40-kilometer cycling time-trial is also a good approximation
of the AT because for well-trained cyclists a 40-kilometer cycling time trial
lasts approximately an hour. These methods are attractive, because they
allow the workrate or intensity to be measured without the requirement for
time-consuming, expensive, and invasive specialized testing.
Finally, for all the preceding intensities, the workrate eliciting the thresh-
old can be identified as either speed or power output, as well as the heart
rate. Heart rate is, of course, susceptible to numerous factors such as diet,
temperature, diurnal variations, stress, and hydration status, but it may,
under controlled conditions, provide an alternative means of prescribing and
monitoring training intensities.

Aerobic Versus Anaerobic Metabolism


and Lactate
The metabolic processes responsible for anaerobic metabolism are complex
but worth considering briefly. The metabolic pathway responsible for the
production of lactate is glycogenolysis, an 11-step pathway involving the
sequential breakdown of glycogen to pyruvate.
184 } Tucker

Pyruvate is then converted to acetyl CoA through a reaction catalyzed by


the enzyme pyruvate dehydrogenase (PDH). When this occurs, acetyl CoA is
next converted to citrate before undergoing a series of reactions in a pathway
known as the TCA, or Krebs, cycle, followed by oxidative phosphorylation of
the resultant electron carriers, the final one of which is oxygen. As a result of
this series of reactions, adenosine triphosphate (ATP), the energy currency
of the cell, is generated in what are known as the oxidative pathways (refer
to chapter 1 for more about ATP).
But pyruvate can also be converted to lactate by the enzyme lactate dehy-
drogenase (LDH), and this reaction is the final step in the glycolytic pathway.
This occurs because the flux, or rate of metabolic reactions, through glycolysis
exceeds the capacity of the PDH to catalyze the conversion of pyruvate to
acetyl-CoA. The result is that pyruvate concentrations rise and LDH converts
it to lactate.
This process has an important metabolic function, because the formation
of lactate allows NADH, the coenzyme, to be converted back to NAD+, which
then facilitates the production of ATP molecules earlier in the glycolytic path-
way. Because of lactate formation, glycolysis can continue at sufficient rates
to provide ATP to muscle, albeit for a limited time only.

Rate of Glycolysis
The driving factor for the formation of lactate, rather than acetyl CoA, is
the flux or rate of reactions occurring in glycolysis. The formation of pyru-
vate is driven in part by the energy charge of the cell (a depletion in ATP
or NADH or an increase in ADP or NAD+ stimulates glycolytic enzymes,
producing more pyruvate), and by sympathetic nervous system hormones
such as adrenaline, which increase the breakdown of glycogen, ultimately
leading to more pyruvate formation. Because the sympathetic response and a
decrease in the energy charge of the cell occur as intensity increases, the rate
of glycolysis and thus pyruvate formation is a function of exercise intensity.
PDH, however, represents a rate-limiting enzyme in the pathway, and
excess pyruvate is converted instead to lactate. Indeed, one of the key meta-
bolic effects of endurance training is that PDH levels increase significantly,
which means that the capacity to convert pyruvate to acetyl CoA increases,
with the result that lactate levels at the same exercise intensity are reduced.
Put differently, people with greater oxidative capacity, either because of train-
ing adaptations or genetic factors, will achieve a higher steady-state exercise
intensity before lactate levels begin to rise—that is, their exercise workrate
at AT will increase. For that reason AT is often a good predictor of triathlon
or endurance performance.
With respect to training effects on lactate production, it is known that
endurance training shifts the AT to higher exercise intensities. This result
occurs primarily because the oxidative capacity is improved, and we describe
Anaerobic Threshold | 185

the mechanisms for this in chapter 14. The outcome is that endurance-trained
triathletes are able to swim, cycle, and run at a higher relative speed or power
output, which has performance implications (described subsequently).

Lactate Shuttle
A final consideration when monitoring lactate is that the level measured in
the blood is a function of both the production of lactate in the cell and the
reuptake for oxidation in other tissues. This reuptake and oxidation form part
of the lactate shuttle, which was described by Brooks.6, 7 The lactate shuttle
concept is that lactate is formed continuously under aerobic conditions and
then oxidized in other tissues as a source of energy. Brooks describes how
lactate is shuttled between glycolytic and oxidative muscle fibers within a
working muscle bed, between working skeletal muscle and the heart, brain,
liver, and kidneys.
Effectively, this model proposes that rather than being viewed as alterna-
tive pathways, glycolytic and oxidative pathways should be viewed as linked
because lactate is the product of one (glycolysis) and the substrate of another
(oxidative pathways), albeit in different cells as a result of shuttling between
compartments.6, 7
Lactate transfer occurs through monocarboxylate transporters, which
shuttle lactate into and out of cells. Endurance training enhances the func-
tion of these transporters, the result being that lactate can be removed from
the blood more rapidly.7 This is another reason why highly trained athletes
often have lower lactate levels than less trained athletes at a given exercise
intensity, not solely because of reduced production but also because training
enhances the reuptake and oxidation of lactate.
Lactate levels at steady-state intensity and AT intensities are thus a predic-
tor of exercise performance because they are markers of the cellular ability
to use oxidative pathways rather than glycolytic pathways for the formation
of ATP. This is in turn influenced by the hormonal and metabolic responses
to exercise, including the sympathetic responses to exercise and the capacity
of oxidative enzymes, which enable higher exercise intensities to be reached
before higher lactate, and thus hydrogen ion concentrations, occur.

Influence of Exercise Intensity


on Energy Supply
The predominant pathway responsible for providing energy is strongly
influenced by exercise intensity, which is inversely related to the exercise
duration. That is, shorter sprint activities are performed at a high intensity,
whereas ultraendurance events including Ironman triathlons are performed
at lower intensities, so they use different energy pathways.
186 } Tucker

Energy Pathway for Sprints:


Phosphocreatine System
For exercise lasting less than 20 seconds, the predominant energy pathway
responsible for ATP formation is the phosphocreatine system, which is able
to form ATP at rapid rates from another high-energy compound, phospho-
creatine. This system has limited capacity, however, and for exercise lasting
longer than a few seconds, other sources of ATP must be used. One such
source is glycolysis, described previously, which produces ATP and pyruvate.
This pathway is responsible for energy supply for exercise lasting minutes.
Beyond this, oxidative pathways assume most of the ATP production in
the form of carbohydrate oxidation and fat oxidation. Note that there are
not distinct intensities at which one pathway is switched on while another
is switched off—all contribute to energy but in vastly different amounts
depending on the demand, which is a function of intensity.

Energy Pathway for Triathlon:


Oxidative Metabolism
For triathlon events, the predominant source of energy, even for sprint triath-
lons, is oxidative metabolism. But the ability to exercise above the AT level
is crucially important, not only because the stochastic nature of competi-
tion demands it but also because evidence from training studies indicates
that training at higher intensities improves performance and economy (see
chapter 12).
In terms of competition, research on the cycling leg of Olympic-distance
triathlon events in the World Cup has shown that most of the time (51 percent
plus or minus 9 percent) is spent below VT1 (which is an approximation for
AT), whereas 17 percent plus or minus 6 percent of the 40-kilometer leg was
spent at workloads higher than maximal aerobic power (MAP).8 These periods
of higher intensity are crucial to racing success, because they influence the
triathlete’s ability to respond to surges, to bridge gaps following the swim
leg, and to negotiate variable terrain and climbs.
Over longer distances, as expected, the relative intensity is further reduced.
Laursen et al. found that highly trained Ironman triathletes completed the
cycle leg of an Ironman at a power output that was considerably below the
power output measured at the ventilatory threshold.9

Need for High-Intensity Training


In terms of training, despite the observation that the most time is spent below
AT, evidence suggests a crucial role for training at higher intensities. As
described in chapter 12, high-intensity training has been found to improve
cycling efficiency.10
Anaerobic Threshold | 187

Further, short-duration, high-intensity sprint training produces physio-


logical adaptations similar to those produced by low-intensity, higher
volume
. training.11 In a study, four to six repeats of 30 seconds at 250 percent
of VO2max, corresponding to the training intensity usually associated with
predominantly glycolytic pathways, induced increases in muscle oxidative
capacity, muscle buffering capacity, and glycogen content that were similar
to those achieved by a group of cyclists
. who performed 90 to 120 minutes of
continuous cycling at 65 percent of VO2max.11

Triathlon Performance
and Anaerobic Threshold
The ability to sustain high relative power outputs is clearly important for
endurance performance. The definition for AT implicitly includes perfor-
mance, because it is defined as the maximal intensity that can be sustained
before an exponential increase in blood lactate levels can occur. For this reason
FTP is a relatively good approximation of AT intensity.
Regardless of the mechanism for the resultant fatigue when this intensity
is exceeded, athletes who are able to sustain higher speeds at this exercise
intensity will be successful. Research from Van Schuylenbergh has found
that the best predictors of triathlon performance are running speed and
swimming speed at the maximal lactate steady state, or anaerobic thresh-
old.12
Schabort et al. have further shown that blood lactate measured during
steady-state cycling at a workload of 4 watts per kilogram or during run-
ning at 15 kilometers
. per hour can be combined with peak treadmill run-
ning velocity and VO2max during cycling to predict triathlon performance.13
Using stepwise multiple regression analysis, they produced an equation to
predict performance in Olympic-distance triathlon using two of the previ-
ously mentioned variables:
Race time (in seconds) =
−129 × PTRS (km/h) + 122 × lactate at 4 W/kg (mM) + 9,456.
For example, a high-level triathlete would typically achieve a PTRS speed
of 22 kilometers per hour and may have a lactate concentration of 4 millimoles
at a power output of 4 watts per kilogram. This athlete’s predicted race time
would be 7,106 seconds, or 1 hour and 58 minutes. An increase in the PTRS (an
indication of improved running ability, a function of numerous physiological
adaptations) or a decrease in lactate (which indicates improved aerobic and
anaerobic function) would reduce the predicted time.
Of course, the prediction will be influenced by the method used to deter-
mine the input characteristics, as described in chapter 11.
188 } Tucker

Finally, Hue et al. found that the lactate concentration at the end of
a 30-minute laboratory cycling bout and the distance covered during a
20-minute performance run immediately following this cycling bout were
the only factors that could be correlated with triathlon race performance in
elite triathletes.14
Clearly, the measurement of submaximal performance variables, when
combined with measurements of lactate concentrations in controlled condi-
tions, provides an accurate prediction of triathlon performance. The explana-
tion for this was described previously, namely that the lactate concentration is
a marker for the metabolic and hormonal capacity and response to exercise.
Genetic factors and the influence of training on these variables means that
laboratory or field measurements of lactate can provide accurate predictions
of performance.
But this is true only when laboratory conditions and other physiological
variables are tightly controlled. For example, Swart et al. have described that
the monitoring of lactate as a means to track training status and performance
over time is unreliable and potentially misleading.15 This inaccuracy is the
result of multiple confounding factors, such as accuracy of portable lactate
analyzers, carbohydrate depletion, mode of exercise (particularly important
for triathlon), ambient temperature, muscle damage, and overtraining, all
of which affect the normal relationship between lactate concentration and
exercise intensity, and thus the AT. They concluded, “Changes in blood lac-
tate concentration should be interpreted with caution as the changes do not
track training status or exercise intensity with sufficient precision to have a
practical application.”13
Similarly, Pyne et al. have shown that in elite swimmers the swimming
velocity at lactate threshold improves with training, as does lactate toler-
ance, but those changes were not associated with changes in swimming
performance.16 This finding challenges the notion that coaches or athletes
should regularly monitor lactate to gauge the effectiveness of a training
intervention.

Conclusion
The concept of thresholds is crucial for triathlon performance. Despite some
confusion about variable definitions for the many possible thresholds and
their measurement, the ability of the triathlete to sustain a high relative
workrate is crucial to success and is thus a focus of training. The key is to
recognize the concept and to ensure that the prescription of training and
monitoring of intensities is consistent over time.
Endurance training produces many physiological changes, but in terms
of the effect on performance, one of the most significant is the ability
to exercise at a higher intensity before the AT intensity is reached. This
Anaerobic Threshold | 189

improvement occurs primarily because of muscle adaptations to train-


ing and enables the triathlete to sustain a higher power output or pace
for long periods. For this reason, AT has been identified as a predictor of
performance. The specific muscle adaptations responsible are considered
in the next chapter.
This page intentionally left blank.
CHAPTER
14
Muscle Types and
Triathlon Performance
—Ross Tucker, PhD

O ne of the most persistent and perhaps misapplied concepts in exercise


physiology is the theory of muscle fiber types and function. Coaches and
athletes recognize that some people possess greater speed or explosiveness
than others and that training produces different responses among individu-
als. Muscle biopsy methods early on established that not all muscle was equal
in terms of the speed and force produced or fatigue characteristics of the
muscle, and athletes were quickly separated into somewhat oversimplified,
polarized groups of fast-twitch athletes (sprinters), and slow-twitch athletes
(endurance athletes).
The development of improved molecular methods has allowed much better
understanding of muscle fiber types, function, and characteristics, making
it clear that the binary approach is indeed an oversimplification and that
muscle fiber type and function are more complex than was initially believed.
Despite unanswered questions, research has provided an answer to the
age-old question from both athletes and coaches: Can we change the fiber-
type composition of our muscles through training? Based on the available
evidence, the short and somewhat disappointing answer is, Not really.1, 2 In
the words of Andersen and Aagard, “Both coaches and scientists know that
it is not possible to turn a donkey into a race horse by means of exercise and
training. Hard work will, at most, turn the donkey into a fast and explosive
donkey.”2
But a longer answer to the posed question is available, including various
nuances that point to the significant physiological and metabolic changes that
occur within muscle because of training. In this chapter we introduce the
concept of muscle fiber types and discuss how training affects not only the
general fiber type classification of muscle but also muscle metabolic proper-
ties, leading to improved performance.

191
192 } Tucker

Muscle Fiber Type Physiology


and Characteristics
The ability of a muscle to contract at high speeds and forces is largely
explained by the composition of the individual fibers of that muscle. Muscle
fibers are distinguished by contractile and metabolic factors, and the type
of myosin is the key difference between the three recognized muscle fiber
types. Myosin, one of several contractile proteins, makes up the thick fila-
ment of the muscle fiber.
The heavy chain of the myosin molecule exists in three forms, and the
differences between these isoforms confer to skeletal muscle its unique prop-
erties, particularly its speed of contraction.1 The three isoforms are MyHC I,
MyHC IIA, and MyHC IIx, and from these isoforms the muscle fiber types are
named.2 That is, the three recognized pure fiber types are type I fibers, type
IIa fibers, and type IIx fibers. There are also hybrids of these isoforms, muscle
fibers that coexpress MyHC I and MyHC IIA or MyHC IIA and MyHC IIx.2
In terms of the characteristics of the muscle fiber types, numerous func-
tional differences exist among the three identified types (see table 14.1). These
differences lead to the more common functional classification of muscle fibers
into type I slow oxidative fibers and type II fibers, in which type II fibers
comprise three subtypes: IIa, IIx, and IIb.
This method of classification more comprehensively reflects the differences
among fiber types, not only according to contraction speed and MyHC type
but also by metabolic properties. These properties are summarized in table
14.1 (adapted from Zierath and Hawley3) and discussed subsequently.

Table 14.1 Muscle Fiber Type Characteristics


ST oxidative fibers (type I) FT oxidative fibers (type IIa) FT glycolytic fibers (type IIx)
Myosin heavy chain I Myosin heavy chain IIa Myosin heavy chain IIx
High aerobic (oxidative) Moderate aerobic (oxidative) Low aerobic (oxidative)
capacity and fatigue capacity and fatigue capacity and fatigue
resistance resistance resistance
Low anaerobic (glycolytic) High anaerobic (glycolytic) High anaerobic (glycolytic)
capacity and motor unit capacity and motor unit capacity and motor unit
strength strength strength
Slow contractile speed Fast contractile speed Fast contractile speed
10 to 180 fibers per motor 300 to 800 fibers per motor 300 to 800 fibers per motor
neuron neuron neuron
Low sarcoplasmic reticulum High sarcoplasmic reticulum High sarcoplasmic reticulum
development development development
High mitochondrial density Moderate mitochondrial Low mitochondrial density
density
High capillary density Moderate capillary density Low capillary density
55% in untrained people 30% in untrained people 15% in untrained people
Muscle Types and Triathlon Performance | 193

These functional differences center on the speed of contraction in the three


muscle fiber types. According to the literature, muscle fibers can be ranked
from slowest to fastest as follows: MyHC I, MyHC I/IIa hybrid, MyHC IIa,
MyHC IIa/IIx hybrid, and then MyHC IIx.4, 5 The difference in contractile
speed has been measured at approximately 1:2 for MyHC I: MyHC II-contain-
ing fibers,6 meaning that fast-twitch fibers take half the time to reach peak
tension compared with slow-twitch fibers.
But other significant differences exist, most notably those relating to the
metabolic differences among fiber types.7 Here the characteristics of the
type I fibers that make them less suitable for explosive, short-duration sprint
events make these fibers more suited to endurance activity. Their increased
capacity for oxidative metabolism, for example, confers an endurance advan-
tage, because it enables more efficient energy production through oxidative
pathways, as described in chapter 13.

Influence of Muscle Fiber Composition


on Athletic Performance
Functionally, the result of the characteristics listed in table 14.1 means that
an athlete who possesses a large proportion of type II (FT) muscle fibers
will be able to achieve higher muscle force and power output during fast
movements. These differences in force don’t appear at slow speeds because
the type I fibers have sufficient time to build up force. The differences are
present only at higher speeds, particularly during acceleration and sprinting-
type activities. For this reason, athletes who have a large proportion of type
II muscle fibers are most suited to sports in which acceleration and explosive
force production is a requirement for success.
In contrast, triathlon, which is an endurance activity even during sprint
triathlons (see chapter 13 for the metabolic explanation), would be more
suitable for people who have a higher percentage of type I, or slow-twitch,
muscle fibers because aerobic characteristics (chapter 11), economy (chapter
12), and fatigue resistance (chapter 15) are the key factors for triathlon success.
The research generally bears this out. For example, one study of elite
middle- and long-distance runners revealed that they had a higher propor-
tion of type I fibers in the vastus medialis than did wrestlers, kayakers, and
powerlifters.8 Research also generally shows that successful endurance ath-
letes possess higher ST proportions than successful sprinters do9, 10, 11 and that
elite endurance athletes have a higher proportion of ST fibers than subelite
endurance athletes do.12
But this principle, although true in general, does not translate into the
simple dichotomy of athlete that is often thought to exist. That is, successful
endurance athletes do not always have the greatest proportion of ST muscle
fibers. For example, large ranges exist among people who are similarly trained
194 } Tucker

within the same endurance activities. Also, successful endurance performance


does not require an exceedingly high percentage of type I muscle fibers.
For example, Fink et al. showed that two performance-matched marathon
runners (2 hours and 18 minutes) differed enormously with respect to the
proportion of ST fibers—50 percent compared with 98 percent.9 These large
ranges are seen elsewhere,8 and a linear relationship clearly does not exist
between proportion of ST fibers and endurance performance, despite the
general finding that elite endurance athletes have more ST fibers than subelite
or untrained individuals do.9
Ultimately, although certain fiber-type compositions are considered advan-
tageous for certain events, the relationship does not appear sufficiently close
to allow the conclusion that fiber-type composition
. is the key factor driving
performance.13 Rather, factors such as VO2max, economy, and thresholds
appear to be more crucial determinants of performance.

Effect of Training on Muscle Fiber Type


Having described that elite endurance athletes possess a higher percent-
age of ST fibers than subelite and untrained people, it is compelling to ask
whether these high ST proportions are the result of training or whether these
individuals are selected because of a genetic difference that persists despite
training? Considerable debate has swirled around this question,14, 15, 16 and
the exact capacity to alter fiber type ratios by converting one type to another
remains unknown.
As described earlier, there is limited plasticity in the muscle fiber type
composition, but endurance training seems to lead to some changes in muscle
fiber composition because of its effects on various signaling pathways. That
is, fast-twitch glycolytic fibers may be converted to FT oxidative fibers and to
ST fibers as a result of calcium-signaling pathways that are activated specifi-
cally by endurance training.3 In rats the experimental activation of the PPAR-δ
receptor leads to muscle fiber transformation,17 which further suggests that
fiber conversions may be possible through the exercise-stimulated activation
of these and other receptors that are known to be activated by exercise and
to modulate gene expression.
Other animal studies have shown that exposing a muscle with predomi-
nantly fast muscles fibers to low-frequency electrical stimulation, similar to
what is received by slow muscle fibers, leads to a gradual shift toward MyHC
I isoforms. Complete removal of the motor nerve, as well as paralysis caused
by spinal cord injury, can shift fiber type from FT to ST.2 These situations are
hardly physiological, however, and are not relevant to a triathlete or coach.
Therefore, it is more important to consider training interventions that have
studied changes in muscle fiber-type composition.
In 1973, in the first study to examine the effects of training on muscle fiber
type, Gollnick et al. had six untrained volunteers cycle for 1 hour per day, 4
Muscle Types and Triathlon Performance | 195

days a week, for 5 months.18 They found that limited conversion of fast-twitch
to slow-twitch muscle fibers occurred (32 percent to 36 percent) and concluded
that endurance training did not alter fiber type.
Subsequent studies have found similar percentage changes, although some
of these have been statistically significant,19 probably because of differences
in individual responses to training and different sample sizes used in the
studies. The magnitude of changes, however, is consistently small, suggesting
that the conversion of FT to ST fibers is limited.
A more important consideration is that these early studies were able to
distinguish only between FT and ST fibers, and not between the subtypes of
FT fibers. Research has subsequently shown that the conversion of type IIb to
type IIa fibers does occur in greater magnitudes with endurance training, and,
to a lesser extent, that type IIa fibers can be converted to type I fibers. These
changes are often put forward as part of the explanation for the observed,
albeit slight, improvements in cycling efficiency or running economy that
occur with endurance training (see chapter 12).
Ultimately, the capacity to change muscle fiber composition and to convert
among fiber types may be limited, at least in magnitude. But the metabolic
properties of muscle have enormous plasticity, and the result is that endur-
ance training can induce large performance differences, not necessarily by
altering fiber-type composition, but through its effects on the function and
performance of existing muscle fibers, which are considered next.

Effect of Training on Muscle Physiology


and Metabolism
Endurance exercise training has been shown to produce profound changes
in muscle structure, function, and metabolism. These changes are reviewed
comprehensively elsewhere.14 Early research established that prolonged
endurance training increased the activities of all the major enzymes of
metabolism, including hexokinase, phosphofructokinase (PFK), and lactate
dehydrogenase (LDH) in the glycolytic pathways, and pyruvate dehydroge-
nase, malate dehydrogenase, 3-hydroxyacyl-CoA dehydrogenase (HADH),
NADH dehydrogenase, succinate dehydrogenase, cytochrome oxidase, and
oxoglutarate dehydrogenase (OGDH) in the oxidative pathways.14

Role of Mitochondria
The primary factor responsible for these increases, particularly in the oxi-
dative enzyme pathways, is an exercise-induced increase in mitochondrial
mass caused by contractile activity. (The increase in mitochondrial enzyme
activity is present only in the legs of runners and cyclists, for example.) The
increase in mitochondrial mass is the result of an increase in both the size
and number of mitochondria in the muscle.14 The result is that more pyruvate
196 } Tucker

can be oxidized, which in turn reduces the accumulation of lactate at a given


exercise intensity (see chapter 13 for explanation). These changes also mean
that endurance-trained muscle has increased capacity to oxidize fatty acids,
with a resultant glycogen-sparing effect and its benefits for reducing fatigue
(see chapter 15).
These metabolic adaptations occur rapidly after the onset of training.
Within 5 days of starting training, lactate concentrations and glycogen deple-
tion are reduced, despite the fact that oxidative capacity and mitochondrial
changes occur only later (31 days).14 Also, detraining produces a more rapid
loss of oxidative potential than glycolytic potential.

Metabolic Benefits of Training


The recent discovery that higher intensity training may produce greater
improvements in aerobic capacity than traditional endurance training (see
chapter 11) also extends to the metabolic benefits associated with training.21,
22
Specifically, sprint interval training consisting of four to six repeats of all-
out 30-second sprinting with 4 to 5 minutes of recovery produced similar
metabolic adaptations compared with higher volume, lower intensity endur-
ance training.21, 22
That is, all the mitochondrial markers for oxidative metabolism were
increased (including pyruvate dehydrogenase and HADH), glycogen use
was decreased, and fat oxidation increased regardless of training method.
The difference in time commitment between these two training methods is
enormous—an average of 90 minutes per week for sprint training compared
with approximately 5 hours per week for endurance training, and a 90 percent
reduction in training volume (measured as energy consumption in kilojoules).

Intense Training Leads to Muscle Adaptation


This finding, along with those showing that aerobic capacity and economy
are both effectively increased with high-intensity interval training20 (chapters
11 and 12), is important because it shows that high-intensity interval train-
ing is an effective strategy to induce rapid adaptations in skeletal muscle
and exercise performance that are comparable with the changes induced by
endurance training.
But these findings do not dispel the need for traditional endurance train-
ing. Higher volume training unquestionably has a place, notably to produce
improved function of the cardiovascular system, neuromuscular adapta-
tions, and adaptation in joints, tendons, and muscles. Also, these studies are
typically performed on untrained or moderately trained athletes, who have
greater potential than competitive or highly trained athletes to improve as
a result of training.
It is debatable whether high-intensity interval training alone can elevate
physiological and metabolic function, and hence performance, to the level
Muscle Types and Triathlon Performance | 197

that a competitive triathlete may aspire to. But these findings do suggest
that for triathlon, for which time constraints are of particular concern, the
addition of high-intensity interval training, at appropriate times, has physio-
logical benefits usually thought to be obtained solely through high-volume
endurance training.

Conclusion
Muscle fiber type appears to play a role in the ability of people to excel in a
particular type of sport. Those with a high proportion of FT fibers are likely
to succeed in explosive events, whereas endurance success requires a higher
proportion of ST fibers. But this relationship is by no means rigid, and other
physiological variables appear to exert more influence over performance than
muscle fiber type does. The ability to convert one fiber type to another appears
to be limited, although conversions from FT glycolytic to FT oxidative and,
to a lesser extent, to ST fibers, have been shown to occur.
The more crucial recognition is that endurance training results in signifi-
cant adaptations that increase the metabolic efficiency and oxidative capacity
of the muscle. The result is that fuel substrate utilization is altered, the meta-
bolic implications of a given workrate are favorably changed, and fatigue may
be delayed. The type of training that produces these responses has recently
been questioned, because higher intensity intervals are now known to be as
effective as traditional endurance training. For practical reasons, however, the
coach must balance the various training stressors to maximize performance.
This page intentionally left blank.
CHAPTER
15
Fatigue Resistance
and Recovery
—Ross Tucker, PhD

A n overriding purpose of training is fatigue resistance. All training is


designed with the ultimate purpose of improving triathlon performance,
which is to say, avoiding or tolerating the onset of fatigue. Of course, some
sessions are designed to assist with technical aspects of the swim, bike, and
run, and these are directly concerned with improving technique, but they
too indirectly contribute to increasing the intensity achievable before the
debilitating effects of fatigue are felt.
Fatigue is an ill-defined phenomenon, having physiological, psychologi-
cal, and emotional qualities, and numerous factors contribute. As a result,
fatigue is hotly debated and somewhat polarizing in the scientific literature.
Coaches and most athletes are well aware of when fatigue happens, why (in
the broader picture) it happens, and how to train to improve performance,
which is essentially either delaying fatigue or achieving a higher speed before
fatigue kicks in.
In this chapter we address some of the theoretical explanations for fatigue
and explain how training alters fatigue and thus performance.

Context-Specific Fatigue
To appreciate the influence of training on fatigue, we first need to appreciate
the physiology of fatigue. The key concept is that fatigue is contextual. It is a
multifactorial phenomenon, and the result is that the physiology of fatigue
is usually dependent on the point of view.
The parable of blind men trying to describe what an elephant looks like
by touching only one part of it is appropriate to how we tend to view and
understand fatigue. A blind man who touches the leg describes an elephant
as a pillar; the man who feels the tail describes it as a rope; to the man who
feels the trunk, it is like a tree branch; the man who touches the tusk says that
the elephant must be a solid pipe. All are correct, of course, in a narrow way,

199
200 } Tucker

but all are completely wrong. The men’s ability to describe the elephant is
limited by their inability to see it from afar and to touch different areas of it.
But this story is not different from how fatigue has been viewed and
explained in the scientific literature. As a result, an integrated model for
fatigue, accounting for a wide range of observations and exercise situations,
remains somewhat elusive.
To illustrate the contextual nature of fatigue, recall that in chapter 13 we
described the predominant energy pathways used for exercise of different
durations and intensities. Clearly, an athlete in the final 2 miles of a 26.2-mile
(42 km) run after 12 hours in an Ironman triathlon experiences an entirely
different metabolic, thermoregulatory, cardiovascular, and hormonal chal-
lenge compared with an Olympic-distance triathlete sprinting the final 500
meters of an event lasting less than 2 hours. As a result, the physiological
underpinning of fatigue will differ significantly.
In the Ironman event, a hot day produces a different physiological expla-
nation of fatigue compared with a cool day. Attributing fatigue to the same
physiological changes in these athletes would be entirely incorrect. The key
point is that training to manage fatigue requires that all the physiological fac-
tors be recognized and managed in training. In so doing, the first paradigm
that must be challenged is the theory that fatigue is a failure of physiology.

Failure Concept of Fatigue


Fatigue, in a real-world, practical sense, cannot be viewed as a failure of
physiology, although science has often explained it that way. For example,
the theory exists that fatigue occurs because a critical level of hyperthermia
is reached.1, 2, 3 Another theory holds that fatigue occurs because of a limita-
tion in oxygen supply, often as a result of failing to pump sufficient blood
to the working muscles.4 “Hitting the wall” in marathon running or longer-
distance triathlons is the conceptualization of one of many failure models
for fatigue—in this case, the failure to maintain normal blood sugar levels
(euglycemia) because of the depletion of liver glycogen.
In all these explanations, fatigue occurs because of the loss of “homeostasis
in one or more homeostatic systems.”5, 6 But that model is incomplete, the result
of laboratory studies that have traditionally approached fatigue using a model
in which exercise is performed to exhaustion at a fixed work rate. Here, the
exercising volunteer is made to exercise literally to the point of exhaustion,
and the cause of exercise failure is then inferred through measurement of
the physiological milieu at that point of exhaustion.
For example, cycling trials to exhaustion at lower workloads have led
researchers to deduce that fatigue occurs because of insufficient energy
supply or glycogen depletion in either muscle or liver. Cycling trials to
exhaustion in hot conditions resulted in a rise in body temperature to 104
degrees Fahrenheit (40 degrees Celsius)—normal is 99 degrees Fahrenheit
Fatigue Resistance and Recovery | 201

(37 degrees Celsius)—and the conclusion was that fatigue occurred because
of the high body temperature.1, 2 And, of course, like the man who describes
the elephant as a pillar while touching the leg, the researchers are not incor-
rect; their interpretation is just incomplete.

Fatigue: From Failure to Regulation


A newer paradigm has recognized the concept of self-paced exercise and
regulation, which introduces the concept of pacing strategy to how we under-
stand fatigue. Here, fatigue is a sensation, not a distinct event that happens
only at the point of homeostatic failure.5, 6 That is, rather than being the result
of failure to maintain homeostasis, fatigue is relative and regulated specifi-
cally to prevent this failure.7 And training, in terms of its effect on fatigue,
acts by improving homeostatic regulation, so that both the physiology and
the sensation of fatigue are altered from the first moment of exercise.
To understand this concept fully and to understand how training alters
the physiology of fatigue, we consider three common examples: heat, energy
supply, and oxygen availability.

Effects of Heat
Heat imposes an enormous physiological challenge on athletes. Heat is a
common challenge in triathlons, particularly in the Ironman distances,
during which the run leg takes place during the hottest part of the day, so it
is a relevant and practical illustration of the physiology of fatigue.
People had previously believed that exercise in hot conditions was nega-
tively affected because our bodies could not deliver enough blood to the
skin for cooling while still matching the demand for muscle blood flow.8
The theory was that the requirement from the muscles to supply oxygen
and energy and remove carbon dioxide and other by-products would be in
conflict with the demand from the skin during exercise in the heat, creating
internal competition for blood supply.
But in 1979 Nadel et al. showed that this theoretical problem was not a
factor because the body was more than capable of meeting both demands
simultaneously by diverting blood from the renal and splanchnic circula-
tions to match the demand.9 This conclusion was subsequently confirmed
by Savard et al.10, and it became clear that other factors were responsible for
impaired performance in the heat.

Effect of Heat on the Brain and Muscle Function


Researchers noticed that exercising volunteers in the laboratory in hot condi-
tions often lost coordination and had impaired motor and cognitive function
at the point of complete exhaustion.10, 11 That finding implicated the brain, and
when it was observed that exhaustion almost always happened at the same
202 } Tucker

temperature of approximately 40 degrees Celsius (104 degrees Fahrenheit),


researchers proposed that there was an upper limit to body temperature
beyond which exercise would not continue.1, 2
As technology improved, so did our ability to study how the brain and
muscle function were affected by high temperatures. One of the key stud-
ies came in 2001 from Denmark, where Nybo and Nielsen were able to use
electromyography to measure muscle activation levels after exercise in hot
and cold environments. They found that when cyclists had fatigued at a body
temperature of 104 degrees Fahrenheit (40 degrees Celsius), they were able to
activate less muscle than when their body temperature was only 100 degrees
Fahrenheit (37 degrees Celsius). The hot brain therefore failed to maintain
the required level of muscle recruitment, so the conclusion was that the high
body temperature caused fatigue because it prevented the central drive for
the recruitment of muscle fibers.1
Again, this conclusion is not necessarily incorrect. The study, in fact, pro-
vides crucial information about what happens at the very end of exercise. It
is incomplete, however, and perhaps inappropriate, for an obvious practical
reason. We rarely exercise at the same speed until we literally have to stop
exercise altogether. In training and in races, we slow down well in advance
of complete termination of exercise, so studies of self-paced exercise were
needed to complete the picture. Those came in 2002, in the form of numerous
research studies that began to accumulate evidence for physiological regula-
tion in advance of hyperthermia.12, 13, 14

Adjusting for Heat


Marino et al. found that athletes doing an 8-kilometer time trial in hot con-
ditions started slower than they did in cool conditions even though their
body temperatures were no different between conditions and were not close
to the theoretical ceiling that had been proposed by previous research on
the “critical hyperthermia hypothesis.”14 The key was that the bigger ath-
letes slowed down more, right from the start of the time trial, which says
that it is not a direct effect of body temperature but rather an anticipatory
reduction in speed that affects performance. The fact that bigger runners
are more affected by this anticipatory reduction is probably because they
have higher rates of heat storage, suggesting that the brain monitors not
the actual temperature but how the rate of heat gain is changing, and then
slows us down.
That was followed by a study by Tucker et al., in which cyclists slowed
down in hot conditions during a 20-kilometer cycling time trial from the first
5 kilometers onward, although they too had submaximal core temperatures
that were no different from those measured in a cool environment.15 This
reduction in power output was associated with a reduction in EMG activity
in the quadriceps muscles, and it was concluded that the centrally regulated
activation of muscle fibers was decreased in advance of an excessive rise
Fatigue Resistance and Recovery | 203

in body temperature so that the required 20-kilometer time trial could be


completed in the fastest possible time without either premature fatigue or
potentially harmful changes in homeostasis.15
Nybo and Nielsen had shown that if the body does overheat, fatigue
occurs as a distinct event and that a dramatic reduction in muscle activa-
tion is the cause.1 But the physiology of fatigue happens well in advance of
this point during self-paced exercise, such as a race or training session.14,
15
Here, muscle activation is reduced in anticipation of that critically high
body temperature.
Equally significant is the increase in muscle activation levels and thus
power output at the end of exercise. This result is typical. Every athlete who
approaches the finish line of a race has experienced finding the necessary
energy to increase pace for the final spurt. This spurt is a significant obser-
vation, because it shows that during exercise, the body maintains a muscle
activation reserve and can recruit those muscles to enable the sprint at the
end of exercise.
For example, Tucker et al. showed a significant increase in skeletal muscle
activation and power output at the end of 20-kilometer cycling time trials in
the heat. The increase in EMG activity and power output happened despite
body temperatures that were significantly elevated compared with those
measured earlier in the trial, showing that a direct effect of high body tem-
perature was not responsible for a reduction in EMG activity as Nybo and
Nielsen had previously suggested. The maintenance of a reserve is a crucial
component of anticipatory regulation.
How is this anticipatory reduction achieved? This discussion is complex
and not directly relevant here, but it is described in detail elsewhere.5, 7 But
it has been proposed that the brain is constantly making calculations based
on changes that happen throughout exercise. If the pace is too high on a hot
day, then the brain activates less muscle; the pace drops, and the rate of heat
storage and body temperature are returned to acceptable levels. Central to
this occurrence is the rating of perceived exertion, or RPE, a subjective but
crucial component, because it explains why fatigue is now viewed as a sensa-
tion, not an event. In this RPE-regulated model, athletes compare how they
feel to how they expect themselves to feel at any moment during exercise
and adjust their intensity accordingly. This modulation is likely a result of
both conscious processes (the conscious awareness of RPE) and subconscious
processes, including the underlying physiology responsible for, in this case,
thermal homeostasis, as well as the generation of RPE and the expectation
of RPE during exercise.

Energy Stores
The same is true during longer duration endurance exercise during which
the depletion of energy stores is a potential limiting factor. Glycogen deple-
tion in either the muscle or liver is known to result in fatigue. The levels of
204 } Tucker

these fuel stores, however, appear to be regulated during self-paced exer-


cise, and this would prevent the premature fatigue associated with deple-
tion and the even more dire physiological consequences of total glycogen
depletion.16, 17
According to this theory, the RPE would be elevated for a given work-
load,7, 18 and the pace would be adjusted so that the RPE, and therefore the
rate of fuel use, would be reduced specifically to ensure that this limit is not
reached. Of course, in ultradistance events, this failure does occur. Athletes
do hit the wall, and glycogen depletion does occur, as does the attainment of
critically high body temperatures. Athletes who are sufficiently motivated or
who are unable to reduce the workload sufficiently (even reducing to walking
speed on a run may be insufficient on a hot day or during an ultradistance
event) remain in danger of reaching these physiological limits.

Oxygen Supply
In the case of an Olympic- or sprint-distance.triathlete, who may be running
at an intensity approaching 90 percent of VO2max during the final stages
of a race, the physiological systems contributing to fatigue involve oxygen
content of the arterial blood, oxygen delivery to muscle, or the removal or
accumulation of metabolic by-products such as hydrogen ions, ADP, and the
depletion of ATP.
In these situations, where exercise intensity is greater, it is less clear what
system or systems are responsible for the sensation of fatigue. Recently,
however, Amman proposed that adjustments in the recruitment of muscle
and thus exercise intensity were made based on afferent feedback from the
muscle, specifically to ensure that excessive peripheral fatigue (loss of muscle
function) does not occur.19, 20, 21 That is, muscle function was protected during
higher intensity exercise because of inputs that include chemical changes in
muscle and oxygen content of the arterial blood.
In all these exercise situations, workload is reduced through a reduction in
muscle activation before the known or theorized point of failure is reached,
and the rating of perceived exertion is central to this regulation. When the
pace is fixed, fatigue is detected as an increase in the RPE—holding that pace
becomes more difficult. If the pace is free to vary, then fatigue is seen as a
slowing down at the same overall effort.
Crucially, we are aware only of our subjective judgment of effort. We do
not have a warning light like a car to alert us to the fact that we are run-
ning low on fuel or oxygen or that our body temperature is increasing. But
we do have a subjective perception of discomfort or exertion, which may
be viewed as a subconscious warning light, and this subjective perception
of effort is what allows us to adjust our pace and in turn our physiological
cost of exercise.
Fatigue Resistance and Recovery | 205

Training and Fatigue:


A Three-Pronged Approach
The key to understanding the influence of these theories on training is to
appreciate several points:
1. Fatigue should be viewed not as a distinct event at which the athlete
fails but rather as a drop in output pace or power for the same effort.
Alternatively, fatigue occurs when the perception of effort is elevated for
a given workrate. This notion has clear implications for the management
of training quantity and quality, as described subsequently.
2. Training will alter the physiology that determines perceived exertion.
That is, the signals that are responsible for this regulation of perceived
exertion, the sensation of fatigue, will change with training, so that the
stimulus to increase the RPE occurs later and the anticipatory regulation
is attenuated.
3. Our interpretation of the brain’s signal of fatigue is crucial to perfor-
mance, and although we cannot completely override the brain’s regula-
tion, we can recognize that a reserve exists and begin to approach our
limits more closely.

Effect of Training on Exercising in the Heat


In the case of exercise in the heat, both heat loss and heat production at the
same exercise workload change favorably as a result of endurance training.
That is, heat loss capacity improves while heat production decreases.
Athletes increase sweat capacity as more sweat glands are recruited and
because heat loss ability is greatly enhanced by the plasma volume expan-
sion that occurs because of training. (This mechanism largely explains
acclimatization to heat.22) In addition, heat production declines because
any given workload corresponds to a reduced relative intensity (as a per-
centage of maximum), which has been shown to be a key predictor of body
temperature.22
The result is that the rate of heat storage is reduced at a given workload
and core body temperature rises at a lower rate because of training or
acclimatization to heat. This outcome is reflected in the subjective rating of
perceived exertion, which has been found to increase at a lower rate because
of repeated exposures to heat. Six consecutive exercise bouts to volitional
exhaustion in hot environments can produce full acclimatization. In prac-
tice, the time required is likely longer because the physiological demand
of exercising to exhaustion on 6 consecutive days is both unrealistic and
disruptive to normal training. Full acclimatization may therefore require up
to 2 weeks of planned training. Ultimately, the exercising athlete is able to
206 } Tucker

sustain a higher pace, experience less fatigue, and delay the point at which
a critical level of hyperthermia may occur.

Effect of Training on Energy Use


Similarly, for energy use, training has numerous benefits. As described previ-
ously, one of the key adaptations to training is an increase in mitochondria,
which results in an increase in the capacity to oxidize fat as a source of fuel
and a resultant glycogen-sparing effect. As for heat, the fact that a trained
athlete competes or trains at a lower relative exercise intensity (as a percentage
of maximum) means that the sympathetic response to exercise is reduced, and
this further favors the preferential use of fat and sparing of muscle glycogen.
As a result, a trained athlete oxidizes significantly more fat at a given
intensity than an untrained athlete does. In addition, glycogen storage capac-
ity increases with training, resulting in a larger fuel store with which to
begin exercise. The brain is therefore monitoring an entirely different set of
physiological signals, so the point of glycogen depletion is delayed because
of metabolic adaptations to training.
Again, it has been shown that the RPE increases at a rate that is influenced
by the level of glycogen in the muscle; low levels are associated with a rapid
rise in RPE.18 Therefore, altering the rate of glycogen utilization, or optimizing
fuel use, is reflected as a change in the rate at which the RPE, or sensation of
fatigue, rises during exercise. In both instances, it is easy to appreciate that
the changes with training will completely alter the signals that the brain
monitors during training and racing.

Importance of RPE
As for the significance of fatigue being a sensation,23 this point is crucial. It
means that athletes can monitor physiology by monitoring the subjective
ratings of perceived exertion during training sessions. Scientists may balk at
the idea that a subjective rating of perceived exertion can outperform sophis-
ticated technology such as heart rate variability and chemical tests, but the
global trend towards fatigue monitoring does indeed reflect the realization
of the importance of the RPE. The astute triathlete or coach who recognizes
the physiological value of the RPE is able to control exercise intensity and
volume to manage and avoid fatigue that occurs with overtraining.

Conclusion
Our understanding of fatigue has evolved in recent years to explain more
broadly what is known to occur during self-paced exercise. A complex regula-
tory process integrates input such as body temperature, oxygen availability,
energy supply, and metabolic status from various physiological systems and
then enables the adjustment of exercise intensity by means of alterations in
Fatigue Resistance and Recovery | 207

muscle activation levels to ensure that performance is optimized and that


homeostasis is maintained.
The effect of training is to alter the input to the brain because training
enables higher workloads to be achieved before potentially limiting per-
turbations can occur in these systems. For example, fat oxidation increases,
glycogen depletion is delayed, and body temperature rises less rapidly, thus
delaying the critical level of hyperthermia at which the athlete would be
forced to stop exercise.
Although this theory does not fundamentally change the practice of
coaching or making training decisions (the triathlete must still perform the
same training to induce the same adaptations; there are no shortcuts here),
it does inform a novel approach to the management of performance, pacing
strategy, and fatigue, particularly in physiologically stressful conditions such
as hot and humid environments, high altitude, and limited energy supply. In
these situations, the pacing strategy and the coach and triathlete’s ability to
understand how pacing strategy is regulated is crucial to success.
This page intentionally left blank.
Pa r t

V
Training Modes
and Methods
for Triathletes
This page intentionally left blank.
CHaPTEr
16
Warm-Up and
Cool-Down
—David Warden

I n the complexities of training, triathletes and coaches often forget that


the mass of the human body is subject to the same physical laws as all
other matter. Specifically, the effect of temperature on any solid or liquid is
to change its physical state. When a solid is heated, energy is added to the
system, increasing the vibration of the particles. When cooled, the particles
slow down. The blood, muscles, and other tissues of triathletes respond to
this phenomenon as any physical matter does, and the effect on performance
can be significant. Although triathletes of all abilities practice warm-ups
and cool-downs, implementation varies. Triathletes who implement specific
warm-up and cool-down protocols based on certain variables are likely to
see an improvement in performance and recovery.

Warm-Up for Performance


At the most basic level, the benefits of warm-up could be summed up as
reducing resistance in the triathlete’s body. Reduced resistance is a natural
consequence of the heated state of any material. In early examinations,1, 2 a
20 percent reduction of resistance was reported in some joints following a
light warm-up, as was a decrease in muscle fiber stiffness. But the physical,
physiological, and metabolic consequences of warming up are much broader,
and some of the benefits are unrelated to temperature change.
The two forms of warming up are passive and active. Passive warm-up is
the act of heating the body using an external source, such as a shower, hot
tub, or heating pad. Active warm-up, during which the triathlete duplicates
the movement of an activity, such as by jogging, is more common. Although
the passive method contributes to an elevated temperature and the associ-
ated benefit of energy conservation in the triathlete, it does not introduce the
physiological and metabolic benefits of an active warm-up and will not be
discussed in this chapter.

211
212 } Warden

The focus of this section is warm-up before an event, but the applications
are similar for a warm-up before the main set of a workout. The principles
laid out can be applied to any workout that includes intervals of moderate
intensity or simulated racing. Additionally, like all effective training, any
warm-up incorporated into a competitive event should be frequently prac-
ticed in training.

Benefits of Warm-Up
The benefits of warm-up are diverse. They include
. increased blood flow, an
increase in oxygen uptake, an elevated baseline VO2, reduced muscle stiffness
and soreness, injury prevention, and even reported psychological benefits.

Increased Blood Flow and Oxygen Uptake


The immediate results of the heated state of blood, muscles, and tissues include
higher core and muscle temperatures, which in turn improve and increase blood
flow. This increased blood flow and its associated oxygen content promote a
potential increase in oxygen uptake,3 a significant predictor for endurance per-
formance. In addition, as early as 1909 researchers demonstrated that hemoglo-
bin gives off twice as much oxygen when its temperature is raised by 5 degrees
Celsius.4 With muscle temperature increased by as much as 6 degrees Celsius
from warming up,15 the temperature of hemoglobin will likely rise by a similar
amount. Although
. oxygen delivery to the muscles is only one component in
determining VO2max, an increase in the supply of oxygen to the muscles is key.
.
Elevated Baseline VO2
The warm-up has additional performance benefits possibly related to tempera-
ture change and the resulting metabolic consequences of a warm-up (figure
16.1). A primary advantage is that the triathlete
. enters the main workout or
competition with an elevated baseline VO2.5, 6 This “excited” state of oxygen
consumption may allow the triathlete to maintain aerobic workload early in
the activity, lowering the risk of early anaerobic work, using less glycogen
for fuel, and increasing time to exhaustion.

Decreased Muscle Stiffness


Additionally, general muscle movement reduces muscle stiffness by breaking
the bond between actin and myosin filaments.7 These filamentous protein
molecules continue to bond at rest, which can lead to muscle stiffness. Taking a
muscle group through a range of motion tends to break these bonds, decrease
stiffness, and maximize range of motion.

Psychological Consequences
The warm-up may even offer psychological benefits. Athletes who visualize
a warm-up report improved performance.8 In interviews, 75 Canadian
Warm-Up and Cool-Down | 213

40°

39°

38°

37°

Temperature (celsius)
36°

35°

34°
Ambient VO2 % of
33° temperature t/min max VO2

20° C 3.96 76
32°

31°

30°
Rest

0 10 20 30
Minutes

„„Figure 16.1 Muscle temperature before and after warm-up.


E5643/Friel/Fig. 16.1/448619/TimB/R2-kh
Adapted, by permission, from B. Saltin, A.P. Gagge, and J.A. Stolwijk, 1968, “Muscle temperature during
submaximal exercise in man,” Journal of Applied Physiology 25(6): 679-688.

Olympians identified a preperformance routine as a distinguishing


characteristic of an Olympic-level athlete. They also found that athletes who
maintain a regular prerace checklist, including a warm-up, appear to be more
likely to duplicate the success of previous performances.

Muscle Soreness and Injury Prevention


Although most of this chapter focuses on warm-up in relation to performance,
evidence also shows that warm-up has postexercise advantages. These benefits
include reduced muscle soreness in the days following exercise and a reduc-
tion in overall potential soft-tissue injury. Even 10 minutes of slow walking
before eccentric exercise reduces muscle soreness 48 hours later.9
Additionally, although there is conflicting evidence that warm-up prevents
injury, research suggests that active warming of the muscles increases elas-
ticity and range of motion,10, 11 thus reducing the chance that the muscle will
stretch beyond capacity during intense exercise.

Risks of Warm-Up
Despite the documented performance advantages of warm-up, some risks
are involved. These possibilities include an increase in core temperature, a
decrease in glycogen stores, and a general increase in fatigue before the event.
In endurance sports, the triathlete’s capacity to store heat becomes a limit-
ing factor in performance,12 particularly for longer exercise sessions during
214 } Warden

which the ambient temperature may continue to rise throughout the event.
A warm-up that increases the triathlete’s initial core temperature carries the
risk that the athlete may start the event at an elevated core temperature,13 thus
reducing the overall capacity to store heat later in the exercise and reducing
performance. In cool environments, a warm-up is unlikely to contribute to
overheating.
Although exercise of less than 3 hours’ duration carries little risk of
glycogen depletion for a moderately trained triathlete, the conservation of
glycogen stores must be carefully considered for longer events. A 30-minute
warm-up for a moderately trained male could reasonably consume hundreds
of calories of glycogen. Although this risk can be easily mitigated through
ingesting additional calories during the warm-up and controlling intensity,
a triathlete who enters long-term endurance exercise with reduced glycogen
stores risks a decrease in overall performance.
Perhaps the most ironic risk of warm-up is the potential to introduce fatigue
to the triathlete before the onset of desired peak performance. A warm-up
of unnecessary duration or intensity crosses the line between warm-up and
intense exercise, thus prematurely placing the triathlete in a fatigued state.14
This potential for fatigue is relative to the triathlete’s fitness. Risk is signifi-
cantly reduced in trained triathletes, and warm-ups should be practiced and
experimented with in training.

Warm-Up Duration
Because the primary benefits of warm-up . are the elevation of blood and
muscle temperature, increased baseline VO2, and increased range of motion,
the warm-up time should be of sufficient length to introduce these three ben-
efits but short enough to avoid the risks of diminished heat storage capacity,
glycogen depletion, and general fatigue.
Although muscle temperature does not continue to rise significantly after
20 minutes of exercise,15 core temperature, which is the primary factor in the
heat-related decrease in performance, does continue to rise with increased
exercise duration, giving incentive to the.triathlete to keep the warm-up short.
In fact, healthy males reached a steady VO2 after just a 10-minute warm-up.16
Finally, trained cyclists achieved a higher power output with a 16-minute
warm-up when compared with a 50-minute warm-up.17
Therefore, for endurance exercise, a warm-up of up to 20 minutes appears
to promote the physical and metabolic benefits that increase performance
without the negative fatiguing effects that accumulate with an increase in
warm-up duration.
Despite the documented performance advantages of warm-up, under some
circumstances it should be adjusted or eliminated from the workout or event.
Based on an understanding of the benefits and risks of warm-up, the warm-
up duration should be varied based on three factors:
Warm-Up and Cool-Down | 215

• Duration of the exercise or event


• Fitness of the individual
• Ambient temperature of the exercise environment

Adjusting Warm-Up Duration Based on Event Length


As the length of an endurance event increases, the need for warm-up declines,
not only because of the warm-up risks previously discussed, which increase
with a longer event, but also because of the low intensity associated with
increased exercise or race duration. For the purposes of this discussion, long
refers to an endurance event during which complete glycogen depletion would
be possible if additional carbohydrate is not ingested throughout the event,
perhaps 3 hours for a moderately trained male or an event that lasts longer
than an Olympic-distance triathlon. The triathlete can take advantage of the
lower intensity associated with long events by treating the first 20 minutes
as the warm-up, which would provide similar benefits without introducing
the risks.
For short events, the requirement for early intensity is great, as is the
need to call on close-to-peak power at the onset of the event, particularly for
elite or draft-legal triathletes who must stay with the lead swimming pack.
This early intensity further increases the risk of prematurely achieving an
anaerobic
. state. The need for increased muscle temperature, elevated baseline
VO2, and range of motion are immediate. Additionally, the risk of glycogen
depletion for a short endurance event of less than 3 hours is low, as is heat
storage capacity for early- to mid-morning performance. Therefore, a warm-
up is recommended for shorter endurance events.
For triathletes, a specific recommendation is to warm up for sprint- and
Olympic-distance events and to avoid additional activity for half- and full-
Ironman events.

Adjusting Warm-Up Duration Based on the Triathlete’s Fitness


The triathlete’s fitness must also be considered in warm-up. The more fit the
individual is, the longer the warm-up that she can tolerate without an effect
on performance. When the risk of undue fatigue is considered in association
with warm-up, the onset and degree of fatigue from activity varies from one
triathlete to another based on fitness. A 20-minute warm-up for a 90-minute
event is a significant share of the activity for many triathletes. Anecdotal
observation of elite triathletes often reveals a warm-up of 60 minutes or
longer for a 60-minute event (half marathon) without an apparent decrease
in performance.
Although no evidence shows that this longer warm-up promotes increased
physical or metabolic benefit, the high fitness level of the individuals allows
them to perform that extended activity without the risks that would apply to
untrained people. The decision about duration of warm-up should be made
216 } Warden

and applied in training to verify tolerance and the consequences to race-day


performance.

Adjusting Warm-Up Duration Based on Temperature


Although less a consideration than fitness and event duration, the ambient
temperature should also be considered when structuring a warm-up. The
warmer the temperature is, the greater the risk that the warm-up will contrib-
ute to higher core body temperature. Although the objective of the warm-up
should still be to increase muscle temperature, rarely does increasing core
body temperature offer an advantage.
Conversely, on cool or cold days, more than 20 minutes may be needed to
warm the muscles sufficiently, so a longer warm-up would be necessary to
reach the desired muscle temperature. Additionally, a cool environment does
not carry the same risk of elevated core body temperature during either the
warm-up or the subsequent event.
Although little specific literature addresses the effect of exercise on muscle
temperature in varying ambient environments, a warm-up of longer than
20 minutes may be justified for events in cool conditions, perhaps 30 to 40
minutes for short endurance events, to ensure that blood and muscle tem-
perature have been properly elevated. For warm days, when the temperature
could affect performance later in the event, the warm-up should be reduced
to less than 20 minutes.

Warm-up Intensity and Specificity


.
Researchers . reported a stable elevated baseline VO2 after 10 minutes at 90
percent of VO2max.16 But this intensity is fairly high, and the duration may not
be adequate to raise blood and muscle temperature.
. Other research supports
an intensity of between 60 and 70 percent of VO2max output,11, 14 which can be
tolerated well for 20 minutes and is intense enough to promote the physical
and metabolic benefits of warm-up without risk of undue fatigue. Although
the consequences of warm-up on fatigue .vary by fitness level, because this
intensity (60 to 70 percent of individual VO2max) is relative to the athlete’s
fitness level, a 20-minute warm-up at this intensity presents low risk of pre-
mature fatigue before events of less than 3 hours in length.
The intensity should include brief bursts of higher-than-race-intensity
speed, or at least the maximum intensity estimated for the event. These 10-
to 30-second intervals should be intense enough to prep the body for the
upcoming race-intensity stress, activate the neuromuscular systems, and
put the muscles through a greater range of motion. The number of intervals
should be three to five, each 10 to 30 seconds in duration, should include
long recoveries of several minutes, and should . be performed at somewhere
between 100 percent and 150 percent of VO2max output, depending on the
triathlete’s fitness level, race length, and ambient temperature. A higher fitness
Warm-Up and Cool-Down | 217

level, shorter event, and cooler day justify a longer and more intense interval.
See table 16.1 for a specific example of a race-day warm-up.
Ideally, the warm-up should include activities specific to the event.
Although doing so is relatively simple for an individual running, cycling, or
swimming event, it becomes complicated for a triathlete. A preferred triathlon
warm-up would include 5 to 10 minutes of activity in all three disciplines, in
reverse order (run, bike, swim).
Although performing a run and swim warm-up before the event is often
easy, a bike warm-up can be impractical and even impossible in the final
30 minutes before a triathlon. Research is limited on the warm-up effects
of one activity on performance in another, but it is the author’s opinion that
the bike warm-up can be eliminated from the triathlete’s routine on race day
and replaced with run and swim warm-up activity. Although this approach
contradicts the activity-specific principle of warm-up, many of the physical
and metabolic benefits of warm-up would theoretically transfer from one
activity to another.

Table 16.1 Suggested Warm-Up for Sprint- and Olympic-Distance Triathlon*


Activity Time (min)
.
Run at velocities between 60 and 70% of VO2max velocity** 10
.
Run three to five intervals of 10 to 30 seconds at 100 to 150% of VO2max velocity** 10
with long easy recoveries
Return to transition area, check equipment, put on wetsuit 10
.
Continue warm-up in water at 60 to 70% of VO2max velocity** or perform 10- to 5
30-second intervals at anticipated swim start intensity
*Cool-day warm-up for events at which maximum anticipated competitive temperature is less than 80
degrees Fahrenheit (27 degrees Celsius). For warmer days, reduce warm-up time by 50%.
.
**See chapter 20 for determining VO2max swim and run velocity.

Warm-Up Timing
Unfortunately, the benefits of warm-up dissipate with inactivity,
. but they
can be maintained for some time after warm-up. Baseline VO2 returns to its
normal profile. after 20 to 45 minutes of inactivity. But the greatest increase
18

in baseline VO2 after warm-up occurs when the activity begins within 10
minutes of the end of the warm-up (figure 16.2).
Muscle temperature appears to be more tolerant of inactivity. Temperature
drops by less than 1 degree Celsius 15 minutes after exercise and remains
elevated by over 2 degrees relative to preexercise for over an hour after exer-
cise at an ambient temperature of 22 degrees Celsius.19 Muscle stiffness and
range of motion can change quickly during prolonged inactivity.
In any competitive athletic event, timing the warm-up
. to end within 10
minutes of the start to gain the benefit of elevated VO2 is challenging. For a tri-
athlete, precise timing can be exceptionally difficult, depending on the logistics
218 } Warden

and layout of the race. For this reason, a triathlete should consider a “broken”
warm-up consisting of 15 to 20 minutes of running with associated intervals,
followed by a final check and setup of the transition area, and then possibly
putting on a wetsuit and reporting to the swim venue to continue the warm-up.
The time from the end of the run warm-up to the start of the swim warm-
up will likely be more than the recommended 10 minutes. Therefore, the
triathlete must continue the warm-up in the water at the swim venue. to
mitigate any muscle temperature loss and to reinstate a high baseline VO2
before the start.. The swim warm-up should be consistent at a relatively high
percentage of VO2 velocity (60 to 70 percent), both to regain the metabolic
benefits potentially lost from inactivity and to maintain or increase blood
and muscle temperature.
Even this broken warm-up will not be possible or feasible at some events,
such as a dry-land start or in extremely cold water in which muscle tempera-
ture could decrease even with activity. In such cases, the triathlete should
consider a warm-up of only running to realize the remaining psychological

4.5 4.5
4.0 4.0
3.5 3.5
3.0 3.0
VO2 L·min–1

VO2 L·min–1

2.5 2.5
2.0 2.0
1.5 1.5
Control Control
1.0 1.0
After 10 minutes of recovery After 20 minutes of recovery
0.5 0.5
0.0 0.0
–120 –60 0 60 120 180 240 300 360 –120 –60 0 60 120 180 240 300 360
Time (seconds) Time (seconds)
a b

4.5 4.5
E5643/Friel/Fig. 16.2a/448621/TimB/R2-kh
4.0 E5643/Friel/Fig. 16.2b/453503/TimB/R2-kh
4.0
3.5 3.5
3.0 3.0
VO2 L·min–1
VO2 L·min–1

2.5 2.5
2.0 2.0
1.5 1.5
Control Control
1.0 1.0
After 30 minutes of recovery After 45 minutes of recovery
0.5 0.5
0.0 0.0
–120 –60 0 60 120 180 240 300 360 –120 –60 0 60 120 180 240 300 360
Time (seconds) Time (seconds)
c d
.
„„Figure 16.2 Baseline and activity-driven VO2max after 10, 20, 30, and 45 minutes
between warm-up
E5643/Friel/Fig. and activity.
16.2c/453504/TimB/R2-kh E5643/Friel/Fig. 16.2d/453505/TimB/R2-kh
Reprinted, by permission,
. from M. Burnley, J.H. Doust, and A.M. Jones, 2006. “Time required for the restoration of normal
heavy exercise VO2 kinetics following prior heavy exercise,” Journal of Applied Physiology 101(5): 1320-1327.
Warm-Up and Cool-Down | 219

and postexercise benefits, as well as the physical benefits of increasing blood


and muscle temperature, breaking the bond between actin and myosin fila-
ments, and increasing range of motion. These warm-up effects appear to
be somewhat tolerant of brief inactivity and can still provide an increase in
performance.

Cool-Down for Recovery


The term cool-down has been incorporated into exercise science with multiple
definitions. For the purpose of this chapter, cool-down will be defined as
two distinct postexercise protocols: active cool-down and passive cool-down.

Active Cool-Down
Active cool-down is the gradual reduction in exercise intensity at the end
of a training session. The triathlete remains active, although at a fraction of
the previous exercise intensity. The most common form of cool-down is the
equivalent intensity of a jog or even a walk for approximately 10 minutes at
the end of an exercise session.
Despite the lack of evidence-based benefits of active cool-down, no research
indicates that it is harmful or detrimental to performance. Triathletes who
perform more than one workout a day may benefit from a 20-minute active
cool-down at very low intensity. Active cool-down is traditionally believed
to have multiple benefits, including the dissipation of lactic acid, reduced
muscle soreness, and reduced likelihood of dizziness or fainting.

Clearing of Lactic Acid


Research
. does show that an active cool-down of 5 minutes at 32 percent of
VO2max (a brisk walk or light jog) is more effective than 5 minutes of com-
pete rest in reducing lactate in the blood after anaerobic activity.20 The same
research also indicates an increase in peak short-term power in repetitive
exercise bouts following an active cool-down, which was associated with
reduced lactate in the blood from cool-down.
But the lactate buildup in this study resulted from peak power of only
6 seconds, not the much smaller buildup of lactate that would be expected
from endurance exercise. In addition, lactate, once viewed as the enemy and
a primary cause of fatigue, is not a contributing factor in fatigue for aerobic
activity.21 Lactic acid accumulates only in brief high-intensity efforts of less
than 1 minute. Although the dissipation of large accumulations of lactate
after anaerobic exercise may provide some benefit when calling upon addi-
tional anaerobic power, lactate dissipation is a concern neither during nor
after exercise for aerobic activity. For the triathlete, who rarely incorporates
maximum-effort anaerobic training of less than 1 minute, large accumula-
tion of lactate is rare, as is the need to call on repetitive bouts of peak power.
220 } Warden

Cool-Down and Muscle Soreness


Lactic acid is also anecdotally blamed for muscle soreness felt a day or two
after intense exercise (delayed-onset muscle soreness, or DOMS). The theory
is that the lingering lactic acid adds to muscle soreness. Not only is this accu-
mulation unlikely for aerobic activity, as related previously, but the soreness
theory has been widely dismissed, in part because of a study that showed
that runners who reported less muscle soreness actually had higher con-
centrations of lactate during a 45-minute run in a controlled environment.22
A previously cited study concluded, “There was no evidence of an effect of
cool-down on either measure of soreness or tenderness.”9 DOMS appears to
have much more to do with eccentric muscle contraction leading to structural
damage, not lingering lactic acid.
An active cool-down study more applicable to triathletes, who often train
twice a day, was based on a 30-minute time trial followed by 20 minutes of
one of five different recovery methods (variations on passive, compression,
and active cool-down). The subjects then rested and repeated the time trial.
The only recovery method to improve performance in the second time trial
was an active cool-down at 80 watts, suggesting that a 20-minute cool-down
at a light spin may be beneficial for triathletes who train twice a day.23

Prevention of Fainting
Perhaps the most likely benefit of active cool-down is to prevent postexercise
dizziness by allowing the “muscle pump” to stay active and prevent the pool-
ing of blood in the extremities.24 The alternating contraction and relaxation
of skeletal muscle creates a pumping action in the veins to aid with venous
return, which reduces the risk of light-headedness or fainting after exercise.

Passive Cool-Down
Passive cool-down is the process of cooling the body using external, inac-
tive methods, such as ice baths, hydrotherapy, cryotherapy, or contrast baths.
Intense or sustained exercise causes microtrauma, or tiny tears in muscle
fibers. Although fitness is ultimately increased by introducing measured
stress and even minor damage (followed by rest) to muscles and tissue, the
ability to speed the recovery process allows the triathlete to return quickly
to additional intense or sustained exercise.
Passive-cool down is thought to minimize exercise-induced damage and
speed recovery by reducing swelling and inflammation, decreasing meta-
bolic activity, slowing down physiological processes, and constricting blood
vessels to flush, or squeeze, waste products out of the affected tissues. With
contrast baths, repetitive cooling and rewarming the body is thought to
increase blood flow and circulation by pumping the blood in and out of the
tissues because of the constriction and expansion of blood vessels from the
variations in temperature.
Warm-Up and Cool-Down | 221

Scientific research supporting passive recovery is rich. In one study, cyclists


performed 5 consecutive days of 105 minutes of moderately intense exercise,
including a 9-minute time trial.25 The exercise was followed by 14 minutes
of either cold-water immersion, hot-water immersion, contrast bath, or inac-
tivity. Both the cold-water bath and contrast bath recovery showed a slight
improvement in time-trial performance over the 5-day period compared
with hot water or inactivity. Another study looked at performance 24 and
48 hours after 3 days of intense exercise and found that cold-water recovery
resulted in a faster return to baseline performance compared with contrast
baths.26 With the literature leaning toward cold-water recovery over contrast
baths and the consideration of the complexity of contrast baths compared
with cold water only, an effective and practical approach for the triathlete is
to use cold-water recovery.
Although the studies supporting passive cool-down vary in terms of water
temperature and duration, a common protocol would be water temperature
between 10 and 15 degrees Celsius (50 to 60 degrees Fahrenheit) with submer-
sion for 10 to 20 minutes. For triathletes, this would rarely include the need for
whole-body submersion and could be limited to the lower body, for example,
by returning to the water start and standing waist deep in water after a race.
Finally, stretching after exercise could be considered part of a cool-down
protocol. The benefits of stretching and flexibility and ways to incorporate
stretching into a triathlon training program are covered in chapter 17.

Conclusion
These protocols require a relatively small investment of time from the triath-
lete while returning potentially significant physiological
. and even psycho-
logical benefits. For example, an increase in baseline VO2 and oxygen uptake
would normally take months or years of regular training to develop. Yet a
brief warm-up can provide an immediate, although temporary, increase in
these critical components of endurance performance. Likewise, investment
in passive cool-down allows a quicker path to recovery, effectively extending
the triathlete’s training week by reducing recovery time.
Like all training techniques, warm-up and cool-down should be practiced
during training and low-priority races before implementing them on race
day. When implemented properly, active warm-up can play a significant role
in improving a triathlete’s performance on race day, and passive cool-down
can speed the ability to return to intense training sessions so that new fitness
can be gained.
This page intentionally left blank.
CHAPTER
17
Flexibility
and Core Strength
—David Warden

T heoretically, flexibility, or range of motion, should have a direct positive


correlation with improved athletic performance and injury prevention.
Intuitively, a triathlete’s ability to maximize movement should result in
increased power, particularly for cyclic disciplines such as swimming, cycling,
and running. Additionally, because soft-tissue injury is caused almost exclu-
sively by pushing a muscle or joint beyond its strength or range of motion,
increased flexibility would theoretically reduce the risk of injury. Empirical
data on flexibility, however, present a complex and sometimes contradictory
picture.
Often developed concurrently with flexibility is core strength, or postural
stability. A popular concept among triathletes, its role in injury prevention
appears positive. Although flexibility may lengthen the lever, core strength
appears to play a crucial role in stabilizing the lever.
Research suggests that a triathlete’s performance can be improved when
specific (but not all) joint and muscle groups increase in flexibility. Although
stretching appears to offer little in terms in terms of injury prevention, it does
offer other limited benefits. Additionally, a core strength program appears to
provide significant benefit in terms of injury prevention.

Flexibility and Performance


To illustrate the inconsistent relationship between flexibility and athletic per-
formance, the flexibility of over 200 elite athletes, including 90 Olympians, was
compared with the total-body flexibility of nonathletes.1 Elite male athletes
in sports such as basketball, soccer, judo, rowing, and fencing were found
to be less flexible than, or equal in flexibility to, the general public, whereas
swimmers, cyclists, and tennis players were found to be more flexible. Elite
female athletes in swimming, but not in cycling, showed increased flexibility.

223
224 } Warden
Bob Seebohar

The researchers concluded that only swimming appears to benefit from an


increase in flexibility across genders.
Although increased flexibility appears to lead to increased performance
in some sports, the trend is not universal. Variations also occur by gender.
Because the focus of this book is triathlon, let’s review how flexibility affects
performance in swimming, cycling, and running.

Swimming
Whether evidence is considered anecdotally or empirically, swimming
performance and flexibility appear to be directly related. Unlike cycling or
running, during which the legs are rarely required to extend to their fullest
reach, outstanding swimming requires maximum length and reach of the
arms to lengthen the lever.
In freestyle swimming, limited shoulder flexibility results in a low elbow
during the recovery phase of the stroke. (Figure 17.1 shows good shoulder
flexibility.) The effect of drag in swimming is so great that ankle flexibility
is required to produce a streamlined lower leg and pointed foot. Addition-
ally, the ability to whip the foot across a wide arc during a flutter kick adds
propulsion to the swim. A 2009 study found a direct correlation between
ankle flexibility, flutter kick speed, and 50-meter swim times,2 and knee flex-
ibility, or the ability to hyperextend the knee slightly, is directly associated
with swim volume.3
Although swimmers would theoretically benefit from flexibility in multiple
joints and muscles in the upper body, the primary areas of focus should be

„„Figure 17.1 Good shoulder flexibility.


Flexibility and Core Strength | 225

ankles (tibialis), lats (latissimus dorsi), shoulders, pectorals (pectoralis), and


knees. See the section Sample Flexibility Exercises for good stretches to try.

Cycling
Unlike swimming and running, empirical literature on the relationship
between cycling and flexibility is scarce. Among coaches and elite athletes,
there is a common anecdotal belief that flexibility of the legs will allow an
increase in power. For example, tight hamstrings could theoretically restrict
the downstroke of the pedal cycle. Or, even more likely, they could reduce the
speed with which the leg can reach the bottom of the stroke. This same logic
applies to the soleus and gastrocnemius muscles of the lower leg. As refer-
enced earlier, a correlation exists between flexibility and elite male cyclists,
which provides a compelling case for flexibility in cycling.
Most research on flexibility and athletic performance focuses on the rela-
tionship between flexibility and propulsive force. In cycling, however, tri-
athletes have another unique benefit from flexibility unrelated to propulsion.
The ability to maintain an aggressive aerodynamic position for 30 minutes
in a sprint-distance triathlon to over 6 hours for a long-course event has a
direct effect on cycling speed. Flexibility in the back, shoulders, and neck
allows an aerodynamic and comfortable ride. In addition, inflexibility in
the aerodynamic position, which leads to discomfort or pain, could reduce
overall power output when cycling.
The primary areas on which to focus to improve flexibility in triathlon
cycling would be the shoulders, hamstrings, and lower back. See the section
Sample Flexibility Exercises for good stretches to try.

Running
Perhaps no discipline has more conflicting data on performance and flexibil-
ity than running. As with swimming and cycling, at first glance it appears
logical that increasing the range through which force can be applied should
increase propulsion. But multiple studies support the theory that the unique
biomechanics of running may require a decrease in range of motion for
maximum economy and performance.

Less Flexible, More Economical


For example, a review of 34 international runners4 measured age, height,
body mass, and trunk flexibility (sit-and-reach test) and correlated those
measurements with running economy. Although no correlation was found
within that group between aerobic running economy and age, height, or body
weight, an inverse relationship was found between economy and flexibility.
The least flexible runners were also the most economical.
The researchers theorized that “stiffer musculotendinous structures
reduce the aerobic demand of submaximal running by facilitating a greater
226 } Warden

elastic energy return during the shortening phase of the stretch-shortening


cycle.”5 This study was repeated in 2009 using subjects’ 10K running times,
and again an inverse relationship was found between the sit-and-reach test
and running economy.6
This inverse relationship between running economy and flexibility extends
to more than just the trunk. For example, runners with decreased range of
motion in standing hip rotation and dorsiflexion of the foot (toes pointed
upward) also have an increase in running economy.7 The theory is that the
stiff joints led to “increasing storage and return of elastic energy and mini-
mizing the need for muscle-stabilizing activity.”8
Other research on the effect of flexibility on running performance is more
ambivalent. Male and female runners adding 2 hours a week of stretching to
their regimens for 10 weeks were compared with male and female runners
who did not add the stretching routine.9 Although the stretching group was
able to increase their sit-and-reach test by approximately 3 centimeters over
their nonstretching counterparts, the stretching group exhibited neither a
decrease nor an increase in running economy.

Static Stretching Before Running


The previous studies focused on the effect of general, long-term flexibility
on running economy, but we also want to know about the effect on perfor-
mance of static stretching just before running. Multiple studies10, 11, 12 have
looked at prerun static stretching and come to the conclusion that although
prerun stretching does temporarily increase range of motion, it has no effect
on subsequent running economy.
The argument that flexibility can improve running performance is mostly
speculative, but that speculation is logical. As previously mentioned, if
flexibility increases the range of force (produces a longer lever), then the
action is more energy efficient. Other theories also support the link between
flexibility and performance. For instance, because run speed is a result of
both stride rate and stride length, flexibility could allow the triathlete to
increase the length of the stride. Tight hip flexors could cause a reduction
in stride length, but an increase in stride length could theoretically com-
pensate for any elastic muscular energy efficiency lost in those more flexible
muscles.
Good running form also requires a sustained forward drive on the part
of the end of the foot. The longer the forefoot can drive forward before
leaving the ground, the longer the overall stride length will be. If the toes
leave the ground too early, forward thrust and distance are reduced or may
even be channeled upward instead of forward. For the foot to maintain a
longer plantar flex (toes pointed down) in the foot strike cycle, it must have
adequate range of motion. Conversely, inflexible ankles could result in pre-
mature separation from the ground and reduced stride length, as illustrated
in figure 17.2.
Flexibility and Core Strength | 227

Loss of stride

Early toe-off

3
2

Path of early toe-off

Path of late toe-off


1

33 cm 5 cm
25 cm

„ Figure 17.2 Foot strike cycle illustrating inflexible ankles resulting in premature
E5643/Friel/Fig 17.2/448624/alw/pulled-R1
separation from the ground and reduced stride length.
Reprinted, by permission, from D.E. Martin and P.N. Coe, 1997, Better training for distance runners, 2nd ed. (Champaign,
IL: Human Kinetics), 27.

Upper-Body Range of Motion


Can upper-body range of motion affect stride length in running? A compelling
theory by Pritchard is that if a runner’s foot crosses over the midline, the total
distance run is longer.13 Any path taken by the foot or leg that deviates from
a straight line in the direction of desired travel could add several centimeters
per stride, and those centimeters would add up. Over the course of a 10K run,
for example, 2 extra centimeters per stride at 620 strides per kilometer would
add 124 meters of running.
Pritchard postulates that although there could be multiple causes for cross-
over, one cause is upper-body torque (UBT). UBT occurs when an inflexible
shoulder or pectoral pulls the upper body back along with the swing of the
arm, causing the opposing leg to cross over in a compensating movement.
A triathlete who implements a performance-motivated flexibility program
should therefore focus on the plantar flexion of the ankles, shoulders, pec-
toralis, and hip flexors. See the section Sample Flexibility Exercises for good
stretches to try.

Considerations for the Triathlete


The conflicting information on flexibility and performance is particularly
challenging for a triathlete, because the three disciplines compete with one
another in terms of ideal flexibility. How does the triathlete decide between
increased dorsiflexion, which improves the kick in swimming but causes a
228 } Warden

reduction in running economy? Will the decreased hip rotation correlated


with better running harm swim performance? Would it be better to keep
trunk flexibility limited to maximize running and accept the potential trad-
eoff of reduced cycling power or aerodynamics?
Although significant research supports decreased range of motion in the
lower body for increased run performance, equally compelling evidence sup-
ports increased range of motion to improve performance in all three triathlon
disciplines. When all is carefully considered, the best choice for triathletes is
to incorporate a flexibility program into their training, targeting the joints and
muscle groups linked to performance in swimming, cycling, and running.
Finally, if all other risks and rewards are relatively equal, the decision for
a triathlete on flexibility could reasonably be based on just one consideration:
potential injury prevention.

Sample Flexibility Exercises


The following exercises may help the triathlete improve flexibility in the
ankles, upper back, shoulders, chest, knees, hamstrings, lower back, and hip
flexors. Follow the guidelines in table 17.1 for static stretching.

Table 17.1 Static Stretching Guidelines


Frequency Stretching 2 or 3 days per week is effective in improving joint range of motion. The
greatest gains occur with daily flexibility exercise.
Intensity Stretch to the point of feeling tightness or slight discomfort.
Time Hold a static stretch for 10 to 30 seconds (recommended for most adults).
Volume A reasonable target is to perform 60 seconds of total stretching time for each
flexibility exercise, perhaps 3 × 20 seconds per muscle.
Pattern Two to four repetitions of each flexibility exercise is recommended. Flexibility
exercise is most effective when the muscle is warmed through activity.

Adapted from American College of Sports Medicine, 2011, “Position stand. Quantity and quality of
exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness
in apparently healthy adults: Guidance for prescribing exercise,” Medicine & Science in Sports &
Exercise 43(7): 1334-1359.
Flexibility and Core Strength | 229

Kneeling Ankle Stretch Kneel on both knees with the feet close together,
toes pointed, and the tops of the feet resting on the floor. Keeping the knees
together as much as possible, lean back over the feet, using your hands to sup-
port your weight (figure 17.3).

„„Figure 17.3 Kneeling ankle stretch.


230 } Warden

Ankle Motion Sit with the legs straight and feet together. Alternate pointing
the toes (figure 17.4a), flexing the feet (figure 17.4b), and rotating the feet in
a circle (figure 17.4c).

„„Figure 17.4 Ankle motion: (a) point toes; (b) flex feet; (c) rotate feet.
Flexibility and Core Strength | 231

Standing Lat Stretch Stand with the feet shoulder-width apart and keep the
knees soft. Lift the arms overhead. Bend the left arm and lower the left hand
behind your head. With your right hand, grasp the left elbow and gently lean
to the right, feeling the stretch up the left side (figure 17.5). Switch sides and
repeat.

„„Figure 17.5 Standing lat stretch.

Kneeling Physioball Lat Stretch Kneel in front of a large physioball. Place


your forearms on the ball and press your torso toward the ground (figure 17.6),
keeping the back straight and abdominals engaged.

„„Figure 17.6 Kneeling physioball lat stretch.


232 } Warden

Doorway Shoulder Stretch Stand in a doorway or other sturdy frame and


place the left leg forward in lunge position. Place both forearms on the frame
and lean into it. Depending on your flexibility, the arms can be lower (elbows
aligned with waist), medium (elbows aligned with shoulders; figure 17.7), or
high (elbows aligned with ears).

„„Figure 17.7 Doorway shoulder stretch.

Up-and-Over Shoulder Stretch Stand with the feet shoulder-width apart


and keep the knees soft. Reach the right arm overhead, bend the right elbow,
and lower the right hand behind the back. If flexibility allows, grasp the right
hand with the left hand (figure 17.8). Switch sides and repeat.

„„Figure 17.8 Up-and-over shoulder stretch.


Flexibility and Core Strength | 233

Lying Full-Body Stretch Lie on your back with your legs straight and arms
reaching overhead. Point your toes and stretch your arms overhead as much
as possible (figure 17.9).

„„Figure 17.9 Lying full-body stretch.

Pec Stretch Stand with the feet close together, keep the knees soft, and posi-
tion your left side beside a wall, doorway, or other sturdy frame. Place your left
hand on the wall at shoulder height and slightly behind your shoulder to feel a
stretch in the left side of the chest. Turn your head to the right and gently press
right to deepen the stretch (figure 17.10). Switch sides and repeat.

„„Figure 17.10 Pec stretch.


234 } Warden

Corner Pec Stretch Stand in a corner, facing the corner, and have your feet
together and legs straight. With the elbows at shoulder height, place your
forearms on opposite walls. Lean into the corner to feel the stretch in the chest
(figure 17.11).

„„Figure 17.11 Corner pec stretch.

Knee Stretch Sit with your back against a wall, the left leg bent, and the right
leg straight with the ankle propped on a small box or foam block. Gently press
the right knee toward the floor (figure 17.12), being careful not to hyperextend
the knee. Switch sides and repeat.

„„Figure 17.12 Knee stretch.


Flexibility and Core Strength | 235

Sitting Hamstring Stretch Sit with the right leg straight, the foot flexed,
and the left leg bent, resting the left foot against the inside of the right thigh.
Reach with both hands toward the right foot, keeping the back straight (figure
17.13). Grasp your ankle, if flexibility permits. Switch sides and repeat.

„„Figure 17.13 Sitting hamstring stretch.

Standing Hamstring Stretch Stand with the feet about hip-width apart and
keep the knees soft. Bend at the hips, keep the back flat, and reach with both
hands toward the feet (figure 17.14).

„„Figure 17.14 Standing hamstring stretch.


236 } Warden

Chair Lower-Back Stretch Sit in a chair with the feet flat on floor about
shoulder-width apart. Lean forward and reach with both hands toward the floor
between the feet (figure 17.15).

„„Figure 17.15 Chair lower-back stretch.

Child’s Pose Kneel on the floor on both knees, keeping the knees and feet
as close together as possible, depending on your flexibility. Reach the arms
overhead and lean forward, pressing the buttocks back toward the ankles and
the hands on the floor in front of the head (figure 17.16).

„„Figure 17.16 Child’s pose.


Flexibility and Core Strength | 237

Standing Lower-Back Stretch Stand with the feet wide apart and keep
the knees soft. Link your fingers behind your head. As you inhale, turn at the
waist to the left. Exhale as you lower your torso toward your left knee. Inhale,
reaching with the hands to the left of your left foot (figure 17.17). Return to the
starting position and repeat to the other side.

„„Figure 17.17 Standing lower-back stretch.

Runner’s Lunge Hip Flexor Stretch Stand with the feet together. With the
left foot, take a big step forward into a lunge position, keeping the left knee
over the left ankle and the right leg straight. If flexibility allows, place both
hands on the floor on either side of the left foot (figure 17.18). Press the hips
down slightly to stretch the hip flexors. Return to the starting position and
repeat with the right leg.

„„Figure 17.18 Runner’s lunge hip flexor stretch.


238 } Warden

Kneeling Hip Flexor Stretch Kneel on the ground on both knees. With the
left foot, step forward into a lunge position, keeping the right knee and shin
on the floor. Place the hands on the hips and gently press the hips forward to
stretch the hip flexor (figure 17.19). Switch legs and repeat.

„„Figure 17.19 Kneeling hip flexor stretch.


Flexibility and Core Strength | 239

Roles in Injury Prevention


In 2003 the University of Heidelberg surveyed 656 active Ironman triathletes.14
Seventy-five percent reported experiencing at least one injury during their
times as triathletes. Because this number included all injuries, including
bruises and abrasions, the rate does not seem unexpectedly high. But when
limiting the scope to include only injuries to muscles, tendons, or ligaments,
the rate was an alarming 62 percent. Considering that regular runners have
a muscular tendon injury rate of 65 percent,15 triathletes, despite lower run
volume, are at about the same risk of injury as pure runners.
Although the Heidelberg study covered injury rate over the entire career
of the triathlete, a British study found a 37 percent injury rate over just 8
weeks of triathlon training.16 Because of this extraordinarily high rate of
injury among triathletes, the prevention of injury and the resulting increase
in the ability to train could play a major role in the improvement of a tri-
athlete’s performance. The triathlete should consider potential injury a real
threat, and look for methods to reduce its likelihood, which may include
flexibility.

Flexibility Linked to Injury Prevention


Although many factors influence rates of injury in triathlons, including age,
training volume, and intensity, evidence shows that under certain conditions
flexibility could play a role in injury prevention. For example, over 200 college
athletes in various sports were tested for flexibility in the iliopsoas, iliotibial
band, hamstring, rectus femoris, and gastrocsoleus muscles.17
At the end of the season, a direct correlation was found between injuries
sustained during the season and preseason tightness in males, but not in
females. But because this study included athletes from disciplines other
than swimming, cycling, and running, it may not be sufficient evidence for
preseason flexibility in the triathlete in and of itself.
Further research about flexibility and injury prevention also indicated
a correlation between stretching and injury.18 Although stretching before
exercise reduced injury, the link was limited to female cyclists only and the
benefit was isolated to groin and buttocks conditions.
Plausible reasoning stands behind the theory that an increase in flexibility
should reduce injury. If the fibers of the muscles are compared with the fibers
of a rope, it is easy to accept the analogy that a more flexible and pliable rope
would have fewer fiber tears when stretched. If soft-tissue injury were caused
exclusively by a combination of lack of strength and lack of flexibility, it would
make sense that an increase in flexibility would decrease injury.
240 } Warden

Flexibility Not Linked to Injury Prevention


Ample research, however, indicates that stretching offers no benefit in reduc-
ing injury. Although the assumption that flexibility was linked to injury
prevention had been ubiquitous among athletic professionals for decades,
it came into question as early as 1998 with this statement from the Ameri-
can College of Sports Medicine: “There is a lack of randomized, controlled
clinical trials defining the benefit of flexibility exercises in the prevention
of musculoskeletal injuries.”19 As a result, a flurry of studies on stretching
and flexibility took place over the next decade.
Perhaps one of the best studies was presented at the American Academy
of Orthopaedic Surgeons’ 2011 Annual Meeting.20 Half of 1,400 runners
were introduced to three stretches for the quadriceps, hamstrings, and calf
muscle groups. The stretch time was 3 to 5 minutes immediately preced-
ing running. The other half performed no stretching before running. The
researchers concluded that there was no difference in injury risk for the
two groups.
Reinforcing this theory is a review of 361 studies (a meta-study) related to
stretching and injury prevention.21 This panel of researchers also concluded,
based on the combined data from these hundreds of studies, that there was
no link between stretching and an increase or decrease in injury. Unlike
the study of 1,400 runners, which focused on stretching before exercise, this
meta-analysis reviewed stretching both before and after exercise.

Additional Flexibility Considerations


These studies would seem to be a devastating blow to the flexibility camp,
but some additional points need to be considered. First, note that no evidence
in these studies indicated an increase in injury linked to stretching. Although
stretching does not seem to offer much benefit regarding injury prevention,
it seems to present little risk. Additionally, because some previously cited
research does support preseason flexibility as a way to mitigate injury, the
scale would seem to be tipped in favor of at least preseason flexibility to
combat injury.
Although the focus of this chapter has been the relatively narrow effect of
flexibility on performance and injury, flexibility can provide other benefits
as well. In his book The Science of Flexibility, author Michael J. Alter argues
that the benefits of a flexibility program include a wide range of research-
based outcomes, including relaxation, relief of low-back pain, and enhanced
sleep, all of which significantly benefit triathletes.22
Additionally, the studies that refute flexibility for injury prevention also
tend to be limited in scope in terms of muscle groups, focusing largely
on the hamstrings, quadriceps, and gastrocsoleus muscles. In some cir-
Flexibility and Core Strength | 241

cumstances flexibility in other muscle groups plays a critical role in the


prevention of injury.
For example, an established treatment of iliotibial band syndrome (ITBS),
a common injury among runners and cyclists, is an increase in flexibility
of the iliotibial band.23 Ober’s test, which tests shortening of the iliotibial
band, is a reliable method to confirm the cause of knee pain linked to ITBS.
Stretching is also a key consideration in both prevention and treatment of
Achilles tendinopathy.24
Although increased risk of injury does not appear to result from either
stretching or not stretching, there does appear to be a risk in rapidly
changing an existing stretching program. Runners who usually stretched
and were then assigned to a no-stretch group had a 40 percent increased
risk of injury, and nonstretchers who were assigned to stretching had a 30
percent increased risk of injury.20 In other words, no difference in injury
risk occurred in either the stretch or no-stretch group, but risk significantly
increased if the runner abruptly adopted or abandoned an established
stretching routine.
Similar to flexibility relative to performance, flexibility relative to injury
prevention is equivocal in the data establishing benefits and risks. It is the
author’s opinion, based on available research, that the benefits (not limited
to injury prevention) of a general flexibility program outweigh any risks.
The most promising and logical strategy to reduce injury through flex-
ibility is to increase slowly and then maintain general flexibility through
stretching frequency and duration. The stretching should take place after,
not before, exercise.

Core Strength
Core strength, or postural stability, is a mature concept. In the 1920s Joseph
and Clara Pilates opened a studio in New York City. Their method, which
they originally called contrology, focused on core postural muscles that
help keep the human body balanced and provide support for the spine.
A triathlete’s core has since been commonly defined as the transversus
abdominis, multifidus, internal oblique, paraspinal, and pelvic floor muscles
(figure 17.20). A broader interpretation of the core is any muscle in the region
between the buttocks and chest, which would include some muscles below
the pelvic floor, such as the gluteus medius.
Although the link between postural stability and performance was
assumed for decades, it was only recently revealed that the cocontraction
of the transversus abdominis and multifidus muscles occurred before any
movement of the limbs.25 In other words, all movement starts at the core.
242 } Warden

External
oblique

Transversus
abdominis

Internal
oblique

Rectus
abdominis

E5643/Friel/Fig 17.20a/448645/alw/pulled-R2-kh

Erector spinae:
Spinalis
Longissimus
Iliocostalis

Multifidus Quadratus
lumborum

Gluteus Gluteus
minimus medius

„„Figure 17.20 Core muscles: (a) front; (b) back.

E5643/Friel/Fig 17.20b/451110/alw/pulled-R2-kh

Core Strength and Injury Prevention


Most research on injury prevention associated with core stability or strength
is linked to the reduction of chronic low-back pain unrelated to exercise. But
some research indicates that core strength prevents athletic injury as well.
Subjects with patellofemoral (knee) pain demonstrated 26 percent less hip-
abduction strength and 36 percent less hip external-rotation strength than
similar age-matched controls who did not report knee pain.26 The collapse
Flexibility and Core Strength | 243

of the hip when running, because of poor lower-core strength, is thought


to place the knee in a misaligned position, adding stress and increasing the
risk of injury.
A similar study was done among college track athletes. Researchers found
that athletes who did not sustain an injury over a season were significantly
stronger in hip abduction and external rotation, and the researchers further
concluded that hip external-rotation strength was the only useful predictor
of injury status among this group of athletes.27
Another common running and cycling condition, ITBS, was found to be
directly related to weakness in the gluteus medius.28 In this study, females
with ITBS were able to lift only 7.82 percent of their body weight through
hip adduction compared with 10.19 percent for noninjured female runners.
Injured males scored even worse, lifting only 6.86 percent of their body weight
compared with 9.73 percent for noninjured males. The injured runners were
then enrolled in a 6-week, standardized rehabilitation protocol that directed
special attention to strengthening the gluteus medius.
After rehabilitation, the females demonstrated an average increase in
hip abductor torque of 34.9 percent in the injured limb, and the males had
an average increase of 51.4 percent. After 6 weeks of rehabilitation, 22 of 24
athletes were pain free with all exercises and able to return to running. At a
6-month follow-up there were no reports of recurrence.
Further research reviewed the recurrence of hamstring injury in relation
to a complete core strengthening routine.29 Twenty-four athletes who were
suffering from an acute hamstring strain, including 10 triathletes and run-
ners, were placed into either a core strengthening program or a hamstring
strengthening and stretching program. The core strengthening routine con-
sisted of a full regimen of core exercises, including side steps, crossover steps,
multiple bridges, and push-up stabilization with trunk rotation, allowing the
complete range of core muscles to be exercised.
In the first 2 weeks after returning to their sport, the reinjury rate was
significantly greater in the strengthening and stretching group, in which
6 of 11 athletes had reinjured the hamstring. None of the 13 athletes in the
core strengthening group had reinjured the hamstring after 2 weeks. After
1 year, the reinjury rate was again significantly greater in the strengthening
and stretching group, in which 70 percent of the athletes had restrained the
hamstring, but only 1 athlete in the core strengthening group had reinjured
the hamstring.
Even further evidence indicates that core strength is associated with other
athletic maladies that would, on the surface, appear completely unrelated. One
example is exercise-related transient abdominal pain (ETAP). These painful
side stitches, characterized by a sharp, localized abdominal pain, plague run-
ners of all abilities. People with poor posture (weak core) were more likely
to experience ETAP, and the level of pain is directly correlated to posture.30
The worse the posture is, the more severe the ETAP pain is.
244 } Warden

Although the link between core strength and hamstring, knee, IT band,
and ETAP may seem unrelated at first, it is consistent with the theory that
all movement starts at the core. Thus, the core exercises are strongly recom-
mended for triathletes of all abilities to mitigate a wide variety of potential
injuries.

Core Strength and Performance


The research on core strength is almost exclusively linked to injury preven-
tion. Few significant findings correlate core strength to performance other
than finding that all movement starts at the core,25 which does not neces-
sarily imply a link between core strength and performance. One study did
find an impressive 30-second improvement in 5K run performance.31 The
overwhelming evidence suggests no link between core strength and per-
formance, including research on swimming and running, which found no
performance improvement after incorporating a core stability program,32, 33
and additional research that found no link between core strength and run
performance or general functional movement.34
Certainly, a reduction in injuries would have a positive effect on endur-
ance performance by allowing the athlete to perform a greater volume of
high-quality training. But a reasonable expectation is that core strength
would positively affect the biomechanics of a triathlete’s swim, bike, and
run disciplines and thus improve performance. The techniques described
in part II of this book could not be executed properly without core strength.
For example, the proper running posture described in chapter 6 could
not be maintained indefinitely without core strength to support the posi-
tion. Without core strength, an ultimate breakdown in form is likely during
this last and arguably most difficult portion of a triathlon. Because of the
unique nature of swimming, in which the triathlete has relatively little
resistance for the limbs to utilize, the core theoretically becomes a pri-
mary mover. Even without a strong empirical link between core strength
and performance, the argument for core strength to improve performance
through biomechanical efficiency is reasonable.

Sample Core Strength Program


These six exercises will help a triathlete strengthen the core muscles. Follow
the guidelines in table 17.2. Perform the exercises three times a week.
Flexibility and Core Strength | 245

Table 17.2 Recommended Core Exercises


Perform three times per week.
Exercise Duration Repetitions
Plank (bridge) 20 to 60 seconds 4
Side plank (bridge) 20 to 60 seconds 4 per side
Side step 60 seconds 3 per direction
Grapevine step 60 seconds 3 per direction
Hip abductor 60 seconds 3 per side
Push-up stabilization with trunk rotation n/a 2 × 15 each side

Plank or Bridge Use the abdominal and hip muscles to hold the body in a
facedown, straight plank position, having the elbows and feet as the only points
of contact (figure 17.21).

„„Figure 17.21 Plank or bridge.

Side Plank or Side Bridge Use the abdominal and hip muscles to hold the
body in a straight plank position with the right side toward the floor (figure
17.22). One elbow and one foot are the only points of contact. Perform on
both sides.

„„Figure 17.22 Side plank or side bridge.


246 } Warden

Side Step Wrap a resistance band around the ankles or shins. Step to the
side, keeping the feet low to the ground (figure 17.23). Step in both directions.

„„Figure 17.23 Side step.

Grapevine Step Step laterally with the trail leg going over the lead leg (figure
17.24) and then behind the lead leg. Step in both directions.

„„Figure 17.24 Grapevine step.


Flexibility and Core Strength | 247

Hip Abductor Lie on one side with the lower leg bent and the upper leg
straight. Lift the upper leg up and down (figure 17.25). Perform on both sides.

„„Figure 17.25 Hip abductor.

Push-Up Stabilization With Trunk Rotation Start in push-up position. Lift


the left hand and rotate the chest to the left, lifting the left hand to point to
the ceiling (figure 17.26). Pause and return to the starting position. Repeat on
the other side.

„„Figure 17.26 Push-up stabilization with trunk rotation.


248 } Warden

Conclusion
Although limited empirical evidence supports the notion that stretching
and flexibility prevent injury, the evidence shows that a consistent stretching
program does not offer any risk and may provide at least some benefit. Tri-
athletes motivated by injury prevention would be better served by adopting
a core strengthening program, which has a well-established link to injury
prevention. In terms of endurance performance, swimming speed is closely
correlated to flexibility, but running performance appears to be inversely
related to lower-body flexibility. The relationship between cycling perfor-
mance and flexibility is ambiguous, as is the link between core strength and
performance.
CHAPTER
18
Strength Training
—David Warden

D r. Tudor O. Bompa, considered the father of modern periodization, was


once asked about the biggest training mistakes that triathletes make.1
He responded that the two biggest mistakes were not using evidence-based
training methods and not implementing a year-round strength training pro-
gram. With this endorsement, justifying the exclusion of strength training
for the serious triathlete would be difficult.
In his breakthrough book Periodization: Theory and Methodology of Training,
which introduced the method of periodization, Dr. Bompa lays out three criti-
cal biomotor abilities: strength (force), endurance, and speed.2 Although each
discipline requires a different balance of these abilities, some element of all
three is necessary to maximize performance in any sport. Furthermore, Dr.
Bompa states that the development of strength “should be the prime concern
of anyone who attempts to improve an athlete’s performance.”3
For the triathlete especially, strength is paramount. For example, running
performance is often simplified as a calculation of only two factors—stride
frequency and stride length. True run performance, however, is the result
of three factors—stride frequency, stride length, and force,4 force being the
ability to accelerate a mass in a desired direction.
Weyland et al. found that when comparing slower runners with faster run-
ners, the faster runners indeed had a stride length 1.69 times greater than the
slower runners did, a stride frequency 1.16 times greater, and a force applied
to the running surface 1.26 times greater. Force was more closely associated
with run speed than stride frequency was. The researchers concluded that
force played the primary role in reducing contact time with the ground. The
more force that a runner applied to the ground, the shorter the time was
between the end of the contact period of one foot and the beginning of the
contact period of the opposite foot.
Additionally, strength is one of two abilities required to develop the
advanced ability of muscular endurance, which is the combination of
strength and endurance. Although some strength is developed as a residual
of endurance training, true force can be developed only through a specific
and dedicated strength training program.

249
250 } Warden
Bob Seebohar

Strength Training and Performance


Unlike endurance and speed training, strength training can be seen by tri-
athletes as contrary to the training principle of specificity. The nature and
scope of strength training, particularly gym-based training, mimics the
movement of the triathlon disciplines only superficially. Triathlon is a steady-
state sport and, with the exception of draft-legal ITU-style racing, exhibits
little need for bursts of brief power and explosive force. Rather, the triathlete
will perform best with a constant and measured intensity consistent with
the distance of the event.
For that reason, strength training has a poor reputation among triathletes.
Performing it is sometimes seen as stealing time from precious sport-specific
training, or even worse, as the cause of weight gain that leads to reduced
economy and performance.

Strength Training Benefits


Despite the lack of sport-specific movement and the theoretically unnecessary
development of power and mass, a considerable body of research indicates
that strength training can improve a triathlete’s endurance performance in
cycling and running in particular. The two primary benefits from strength
training appear to be increases in both endurance and economy.
In a frequently-cited study,5 researchers added three strength training
sessions for 10 weeks to the regimens of athletes trained in both cycling and
running. As expected, leg strength increased significantly, but thigh girth
remained unchanged. .
More important, cycling time to exhaustion at 80 percent of VO2max
increased from 71 to 81 minutes, an exercise duration and intensity highly
applicable to triathletes. Short-term endurance performance (from 4 to 8
minutes), which would add power to the final push in a triathlon event,
also improved 11 to 13 percent in both cycling and running. Additionally,
12 weeks of strength training. resulted in a 33 percent increase in cycling to
exhaustion at 75 percent of VO2max, also an intensity particularly relevant
to triathletes.6
Another study found that concurrent strength and endurance training
resulted
. in a significant improvement in 5K run time unrelated to an increase
in VO2max.7 This improvement would be quite useful in the final minutes
of triathlon, specifically in sprint-distance triathlon events.
Specific to endurance running, researchers found that adding strength
training concurrently with run training significantly improved endurance
running.8 The strength regimen was 8 weeks long, but this research divided
the group into three types of strength training: heavy (high load), explosive
(high power), and muscular endurance (high duration). All three types of
strength training resulted in improved run endurance, and the muscular
endurance training resulted in the highest gains. This finding suggests that
Strength Training | 251

although any type of strength training can improve endurance, the best
results will come from the most sport-specific strength training.
Much of this research supports the positive relationship between strength
and endurance performance for durations of 4 to 80 minutes. But what will
the effect of strength training be on exercise of 2 hours’ duration or longer?
Although the research on endurance of this duration is limited, significant
science supports a positive correlation between strength training and both
cycling and running economy. Any improvement in economy (usually mea-
sured as the relative oxygen cost in milligrams of oxygen per kilogram per
minute to run at a given velocity) is almost certain to improve endurance,
particularly at longer distances.
For example, eight well-trained runners completed an 8-week strength
training program on 3 days per week, in addition to their regular train-
ing.9 They were compared with nine control runners who did not include
strength training in their regimens. The strength training group improved
running economy by an impressive 5.0 percent, which likely contributed
to a 21.3 percent improvement in time to exhaustion at maximum aerobic
threshold.
Similar results were seen with a group of 15 triathletes assigned to either
an endurance-only group or a group that did endurance training and two
sessions of strength training per week for 14 weeks.10 The endurance-plus-
strength group had a significant improvement in running economy.
Perhaps even more compelling evidence than these individual studies
comes from two systematic reviews of multiple studies on strength training
and endurance performance. After reviewing several studies and using a
scale to determine the quality and consistency of the research, the researchers
concluded that “resistance training likely has a positive effect on endurance
running performance or running economy,”11 and “replacing a portion of a
cyclist’s endurance training with resistance training will result in improved
time trial performance and maximal power.”12

How Strength Training Improves Endurance


and Economy
How could strength training contribute to this increase in endurance and
economy? Strength training appears to promote a combination of biome-
chanical and metabolic changes that directly affect endurance and economy.
Because run speed is a result of stride length, stride frequency, and force,
a triathlete who can maintain stride length late in the run has an increased
likelihood of maintaining run speed. This notion is borne out in a study
in which 18 very quick endurance runners were placed into periodized
strength and nonstrength groups for a period of 8 weeks.13 At the end of
that period, the runners performed a workout of three intervals, and the
researchers measured the stride length of the first and third intervals. The
252 } Warden

periodized strength group had no significant loss of stride length during


the third interval, whereas the nonstrength group lost almost 3 percent of
stride length in the final interval.
Additionally, researchers from Ohio University used an MRI to measure
the contrast shift (activation) of muscles before and after a 9-week strength
training program.14 After the strength training, the MRI revealed less con-
trast shift for a given amount of work. These results suggest that resistance
training results in the use of less muscle to lift a given load. This finding
has significant potential application for triathletes. Less use of muscle for a
given amount of work would not only delay muscular fatigue and reduce
muscle glycogen use but also reduce oxygen consumption for that given
load and thus help explain the increase in economy seen after performing
a strength training program.
At a metabolic level, strength training has been shown to introduce
remarkable changes. Healthy males were tested for levels of phosphocreatine
(an essential molecule in the formation of ATP), .lactate, and glycogen after
approximately 1 hour of exercise at 72 percent of VO2max.15 They were tested
again for those levels postexercise after 4 to 12 weeks of strength training
three times a week. The results were a 39 percent increase in phosphocreatine,
a 37 percent decrease in lactate, and a 37 percent increase in glycogen levels
after exercise, when compared with the period before strength training. In
a previously. cited study,6 the 33 percent increase in time to exhaustion at 75
percent of VO2max after 12 weeks of strength training was attributed to a
12 percent increase in lactate threshold.

Controversy Regarding Elite Male Triathletes’


Need to Strength Train
Despite this notable body of evidence supporting strength training in endur-
ance triathletes, many triathletes and coaches believe that strength training
may not benefit performance in elite male triathletes. Additionally, they often
express concern regarding weight gain and the limited time available for
task-specific endurance and speed training.
Strong research supports strength training for endurance performance
specifically for female athletes,16 junior athletes,17 and masters athletes.18
Because female, junior, and masters athletes maintain less muscle mass than
young men do, elite male triathletes could theoretically benefit less (or not
at all) from strength training.
Additional arguments against strength training for elite male endurance
triathletes include the difficulty of fitting additional work into a schedule
that often already exceeds 20 hours a week. This considerable volume of
specificity may eclipse any benefit from strength training. Finally, some stud-
ies refute the performance benefit of strength training in both cycling19 and
running.20 These studies concluded that female cyclists had no improvement
Strength Training | 253

in a 1-hour cycle test after 12 weeks of twice weekly .strength training and
that middle-aged runners had no improvement in VO2max, stride length,
or stride frequency.

Most Evidence Shows Strength Training


to Be Helpful to Cycling and Running
The evidence supporting strength training for endurance performance is
therefore far from unequivocal. The majority of evidence, however, suggests
a positive relationship. Even the limited studies refuting strength training for
endurance performance do not find a decrease in performance. Rather, they
simply find no performance increase, which means that there is little risk to
the triathlete in a trial of strength training. These negative studies represent a
small portion of the literature, whereas significantly more research indicates
that strength training improves endurance performance.
More specifically, the concerns regarding elite triathletes—weight gain
and sacrificing training time—are themselves refuted in multiple studies.
For example, previously cited studies supporting the benefits of strength
training11, 12 included advanced cyclists who rode more than 90 miles (145
km) a week and advanced runners who ran more than 30 miles (48 km) per
week. Further research concluded that strength training can lead to enhanced
long-term (greater than 30 minutes) endurance capacity, in both well-trained
and highly trained top-level endurance athletes.21
Concerns regarding strength training and weight gain are certainly justi-
fied. A triathlete’s mass is a critical factor in performance, both physiologically
and even aerodynamically. But weight gain by itself becomes irrelevant when
measuring economy or power-to-weight ratio. Because running economy is
the measure of oxygen cost for a given velocity, when economy improves any
concurrent increase in the mass of the triathlete is immaterial.
In elite cycling, a gain of 1 kilogram of body weight can be alarming but
becomes insignificant if the triathlete’s power has increased by 2 percent
because the total power-to-weight ratio has increased. Rather than focus-
ing on simple weight gain, a triathlete or coach should avoid uneconomical
weight gain. Note also that although many studies reported an increase in
limb girth and weight with strength training, other studies reported no
weight gain from the regimen.7
Finally, objections to redirecting precious training time to strength training,
particularly for time-constrained age-group triathletes, seem to be addressed
in the scientific literature. Most of the studies cited in this chapter had the
subjects add strength training to an existing endurance program. But other
studies showed the benefits of strength training when athletes replaced as
much as 32 percent of their endurance training with strength training.5, 9 In
fact, strength training was most effective when replacing existing endurance
training rather than being used as an additional approach.12
254 } Warden

Effect of Strength Training on Swimming


The research discussed in this chapter has hereto focused on the benefits of
strength training on cycling and running performance. Not surprisingly, the
three disciplines of triathlon may have different responses to strength train-
ing. In swimming, swim coaches at all levels have used dry-land strength
training extensively. Despite this consistent use by swim professionals, the
empirical data supporting dry-land strength training are weak.

Greater Strength, Yes; Better Swimming, No


In one study collegiate swimmers added strength training on 3 days per week
to their existing 6 days per week of in-water training.22 Despite an increase of
25 to 35 percent in dry-land strength, no change in distance per stroke was
observed compared with the control group, and no significant differences
were found in any of the swim power and swimming performance tests.
Similar results were found with another group of collegiate swimmers.
Again, although dry-land power improved, no difference was seen in stroke
length or performance time in a 400-yard (366 m) swim time trial between
the swimmers who did and did not perform dry-land strength training.23
Furthermore, despite claiming six gold and two bronze medals at the 2004
Olympics in Athens, Michael Phelps reported in a 2004 interview, “I've never
lifted a single weight in my life. It's all from training in the water, period.”24
Although Phelps did begin dry-land training in 2005 before an even more
impressive 2008 Olympic performance, the benefit was reported as limited to
his push-off from the wall,25 a benefit not relevant to open-water triathletes.
A study from the University of Colorado may have described it best: “In
contrast to running and cycling, traditional dry-land resistance training or
combined-swim-and-resistance training does not appear to enhance swim-
ming performance in untrained individuals or competitive swimmers, despite
substantially increasing upper-body strength.”26
The anecdotal evidence supporting dry-land strength training, stemming
from its extensive use by swim professionals, is difficult to ignore, but the
empirical evidence is equally convincing. Note that no evidence suggests that
dry-land strength training impedes performance. The studies cited earlier did
not find any detrimental effects of strength training on swim performance.
When discussing swimming and strength training, it is essential to
distinguish the research regarding dry-land strength training from that
regarding in-water strength training. The latter has positive results on swim
performance. In-water strength training is discussed later in this chapter.

Dry-Land Strength Training Not Needed


It is the author’s opinion that the relative importance of flexibility and strength
are related to the technique required to perform the task. For example, cycling
certainly requires a certain amount of skill, but it is relatively easy compared
Strength Training | 255

with the technique required in swimming. An inverse relationship may


exist between the complexity of a sport and the role of strength, and a direct
relationship may be present between complexity and the role of flexibility.
Because of the complexity of swimming, dry-land training for swimming
appears to be unhelpful. In previously cited research, the authors suggested
that “the lack of a positive transfer between dry-land strength gains and
swimming propulsive force may be due to the specificity of (swim) training.”22
Other research postulated that “resistance training may be a valuable adjunct
to the exercise programmes followed by endurance runners or cyclists, but
not swimmers; these latter athletes need more specific forms of resistance
training to realise performance improvement.”27
The hypothesis of the relationship between strength and flexibility and the
three triathlon disciplines is supported in the literature cited in this chapter
and in chapter 17, but further research is required to confirm this supposition.

Gym-Based Strength Training


With the benefits of strength training well established, how do triathletes
incorporate it into their endurance training? The most successful approach
is a program that is both sport specific and periodized. Sport-specific means
that the strength training focuses on the prime movers, or multijoint exercises
that mimic the movements of the discipline. Periodized means the strength
training follows a preparatory-to-specific path. For example, strength training
was found to benefit stride length in runners only when the strength train-
ing was periodized.13 In this particular study, the runners spent 4 weeks in
a prep phase of easy lifting to adapt the muscles to the additional upcoming
stress, 8 weeks in a specific period in which the strength training mimicked
the increased intensity of running, and finally 4 weeks in a period of strength
maintenance.
Sport-specific strength training is, of course, unique to every sport. For
triathlon, table 18.1 lists the recommended exercises and the associated dis-
cipline that receives the greatest benefit.
In his book The Triathlete’s Training Bible,28 author Joe Friel proposes four
distinct periodized strength training phases: anatomical adaptation (AA),
maximum transition (MT), maximum strength (MS), and strength mainte-
nance (SM). This chapter also proposes an additional strength training phase,
the competition maintenance (CM) phase.

Table 18.1 Multijoint Exercises and Applicable Discipline


Multijoint exercise Applicable disciplines
Squat or leg press Cycling, running
Bent-arm lat pull-down Swimming
Chest press or push-up Swimming
Seated row Cycling (particularly climbing)
256 } Warden

Anatomical Adaptation
Consistent with the theory of endurance periodization, anatomical adaptation
is a preparatory phase introduced at the beginning of an overall triathlon
training program or season, usually in the early season preparation phase last-
ing into the early weeks of base training. Long, slow repetitions both promote
slow-twitch muscles and prepare the body for the increased demand in load as
the strength program progresses. Use caution when introducing this training,
particularly for the first time, and increase the load slowly over the AA period.
Failure to introduce this new type of training can result in severe soreness that
can disrupt days of training or even cause injury, requiring an even longer
recovery time. Table 18.2 lists the suggested implementation of the AA phase.

Table 18.2 Anatomical Adaptation (AA) Phase


Total AA sessions 8 to 12
Sessions per week 2 or 3
Load (% of 1RM) 40 to 60
Sets per session 3 to 5
Reps per set 20 to 30
Recovery time 60 to 90 seconds
Speed of lift Slow

Triathlon exercises (in order of completion)


1. Squat or leg press

2. Bent-arm lat pull-down

3. Squat or leg press

4. Chest press or push-up


5. Personal weakness (hamstring curl, knee
extension, heel raise)
6. Seated row

7. Core (see chapter 17)


Adapted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd ed.
(Boulder, CO: VeloPress), 243.

Maximum Transition
The MT phase is designed to prepare the body for the heavier loads of the
critical MS phase. Use caution when introducing this phase and the increase
in load. Note that only certain exercises (in bold in table 18.3) move to the
MT phase; all others remain in the AA phase. This approach promotes long,
lean, fat-burning muscles that will not be required to generate as much force
as the key prime movers. See table 18.3 for the MT phase.
Strength Training | 257

Table 18.3 Maximum Transition (MT) Phase


Total MT sessions 3 to 5
Sessions per week 2 or 3
Sets per session 3
Reps per set 10 to 15
Recovery time 90 to 180 seconds
Speed of lift Slow

Triathlon exercises (in order of completion) Load goal


Squat: 1.3 to 1.7 × body weight
Squat or leg press*
Leg press: 2.5 to 2.9 × body weight
Seated row* 0.5 to 0.8 × body weight
Personal weakness (hamstring curl, knee
40 to 60% 1RM
extension, heel raise)
Bent-arm lat pull-down* 0.3 to 0.8 × body weight
Core (see chapter 17)

*Carried over from the AA phase.


Adapted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd ed. (Boulder,
CO: VeloPress), 244.

Maximum Strength
The MS phase is the primary purpose of gym-based strength training. This
phase is when power is primarily developed for the triathlete. Although power
is rarely called on in a steady-state event, it is crucial in draft-legal events,
when climbing, or when finishing a race. The work done in the AA and MT
phases prepares the triathletes for these heavy MS loads to solidify the ability
to generate force and power. Like the MT phase, only specific exercises move
to the MS phase; all others remain in the AA phase.
Be aware that endurance and speed training may suffer during this phase
because the MT phase introduces a significant level of fatigue. To minimize
this effect, strength training should be periodized and coordinated with
endurance training so that the MT phase ends by the first part of the base
training phase. This approach allows the later stages of base training and
further phases to focus on sport-specific improvement without the heavy legs
introduced in the MT phase.
Additionally, scientists found that the fatigue introduced from strength
training can be mitigated by ordering the strength sessions to take place after
an endurance set, which allows a high- quality endurance set and improves
results from the strength session.29 See table 18.4 for details about the recom-
mended MS phase.
258 } Warden

Table 18.4 Maximum Strength (MS) Phase


Total MS sessions 8 to 12
Sessions per week 2
Sets per session 3 to 6
Reps per set 3 to 6
Recovery time 120 to 240 seconds
Speed of lift Slow

Triathlon exercises (in order of completion) Load goal


1. Squat or leg press Squat: 1.3 to 1.7 × body weight
Leg press: 2.5 to 2.9 × body weight
2. Seated row 0.5 to 0.8 × body weight
3. Personal weakness (hamstring curl, knee
40 to 60% 1RM
extension, heel raise)
4. Bent-arm lat pull-down 0.3 to 0.8 × body weight
5. Core (see chapter 17)

Adapted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd ed.
(Boulder, CO: VeloPress), 246.

Strength Maintenance
Like any other fitness or skill developed, strength must be maintained. The
SM phase is designed to sustain the new abilities to generate force while not
interfering with critical sport-specific endurance and speed training beyond
the base phase. Note in table 18.5 that the standing, bent-arm lat pull-down

Table 18.5 Strength Maintenance (SM) Phase


Total SM sessions Until race or competition phase
Sessions per week 1
Load (% of 1RM) 60, 80 (last set)
Sets per session 2 or 3
Reps per set 6 to 12
Recovery time 60 to 120 seconds
Speed of lift Moderate

Triathlon exercises (in order of completion)


1. Squat or leg press

2. Seated row
3. Personal weakness (hamstring curl, knee
extension, heel raise)
4. Bent-arm lat pull-down

5. Core (see chapter 17)

Adapted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd ed.
(Boulder, CO: VeloPress), 247.
Strength Training | 259

reverts to AA phase repetition frequency in the SM phase, whereas the hip


extension and seated row alone, in bold, continue with the lower reps of the
SM phase.

Competition Maintenance
Many coaches recommend the termination of strength training at the onset
of the race phase to avoid the buildup of additional fatigue. Continuing
limited strength training into the race phase, however, will result in better
performance than abruptly ending strength training.30 This plan is particu-
larly relevant for female and masters triathletes, who may lose muscle mass
faster than younger men do.
The CM phase is designed to provide limited strength training benefit
without introducing resistance-induced weakness in competition. Because
of this, a CM session should be completed at least 72 hours before competi-
tion. Table 18.6 outlines the recommended CM phase. Note that all exercises
follow the same load and repetition goals in the CM phase.

Table 18.6 Competition Maintenance (CM) Phase


Total CM sessions Through race or competition phase
Sessions per week 1
Load (% of 1RM) 80 to 85%
Sets per session 2
Reps per set 5
Recovery time 60 to 120 seconds
Speed of lift Moderate

Triathlon exercises (in order of completion)


1. Squat or leg press

2. Seated row
3. Personal weakness (hamstring curl, knee
extension, heel raise)
4. Bent-arm lat pull-down

5. Core (see chapter 17)

Load Goals
Naturally, the repetitions recommended here would be easy to achieve with
little resistance. Appropriate load must be administered to stress the body
sufficiently to gain new fitness. Load goals are generally determined either
as a percentage of body weight or as a percentage of 1 repetition maximum
(1RM).
260 } Warden

Several formulas and protocols can be used to determine 1RM. One author
proposed the following protocol for 1RM testing:31 Begin with a warm-up
of 5 to 10 repetitions at 40 percent to 60 percent of the triathlete’s estimated
maximum. After a brief rest period, the load is increased to 60 percent to 80
percent of the triathlete’s estimated maximum, and the triathlete attempts
to complete 3 to 5 repetitions. At this point small increases in weight are
added to the load, and a 1RM lift is attempted. The goal is to determine the
triathlete’s 1RM in three to five trials. The triathlete rests for 3 to 5 minutes
before each 1RM attempt.
A more practical approach, or at least an approach to estimate the 1RM
that initiates the preceding method, is a formula that states that a 1RM can
be calculated based on the resistance and number of repetitions to exhaus-
tion, as follows:32
1RM = 100 × weight / (102.78 − 2.78 × reps)
For example, an athlete is able to squat 200 pounds (91 kg) 15 times. Using
the preceding formula, his 1RM would be estimated as follows:
1RM = 100 × 200 lb / (102.78 − 2.78 × 15 reps), or a 1RM estimate of 327
pounds (148 kg)
Strength Training | 261

Strength Training Exercises


Here we describe the exercises listed in tables 18.1 through 18.6. Follow the
specific guidelines for each phase regarding number of sets, repetitions, and
load.
Squat Stand with your feet shoulder-width apart and toes out slightly. If
using a barbell, hold it across your upper back in a closed grip. If using dumb-
bells, hold them to your sides, palms turned in. Maintaining a natural curve in
the lower back, slowly squat, moving the buttocks back as though sitting on
a chair, making sure that the knees do not move over the toes (figure 18.1).
When the thighs are parallel to the floor, straighten the legs to return to the
starting position.

„„Figure 18.1 Squat.


262 } Warden

Leg Press Sit in an incline leg press machine or sled with your back flat against
the seat back, your feet shoulder-width apart on the platform, and your legs
straight. Slowly bend your knees to lower the weight, bringing the knees to a
90-degree angle (figure 18.2). Do not lock your knees. Straighten your legs to
return to the starting position.

„„Figure 18.2 Leg press.


Strength Training | 263

Bent-Arm Lat Pull-Down Stand facing a cable machine with your feet
shoulder-width apart or sit on the seat of the cable machine, facing it, feet
flat on the floor. Using a wide grip, grasp the bar with the palms facing away
from your body and your arms extended overhead. Bend the arms to pull the
bar down to the upper chest (figure 18.3). Extend the arms to return to the
starting position.

„„Figure 18.3 Bent-arm lat pull-down.

Chest Press Lie on a weight bench or stability ball with your feet flat on the
floor. Using a closed, overhand grip, grasp the barbell with your hands about
shoulder-width apart. Slowly lower the barbell toward your chest (figure 18.4).
Push the barbell up to return to starting position. If using dumbbells, hold a
dumbbell in each hand with the palms turned toward the feet and lower the
dumbbells until they touch your chest.

„„Figure 18.4 Chest press.


264 } Warden

Push-Up Lie face down on the floor with your hands under your shoulders
and toes pointed. Push up with your hands until your body is in a straight line
from the ankles to the top of your head. Bend the elbows to lower your torso
toward the floor (figure 18.5). Then push up again.

„„Figure 18.5 Push-up.

Seated Row Sit on the bench facing a pulley machine. With palms facing each
other, grab the handles of the machine and have your arms straight. Keeping
your back still, pull the handles toward your torso (figure 18.6), pulling your
shoulder blades together. Pause before straightening the arms to return to the
starting position.

„„Figure 18.6 Seated row.


Strength Training | 265

Hamstring Curl Lie face down on a hamstring curl machine with your ankles
under the pads. Slowly bend your knees to pull the pads toward your buttocks
(figure 18.7). Return to the starting position by slowly straightening your legs.

„„Figure 18.7 Hamstring curl.

Knee Extension Sit on a leg extension machine with your ankles under the
pads and knees bent. Slowly straighten your knees to lift the weight (figure
18.8). Return the weight to the starting position by bending your knees.

„„Figure 18.8 Knee extension.


266 } Warden

Heel Raise Stand on a small plyometric box with your heels off the edge
of the box. Hold on to a sturdy object for balance, if you wish. Press through
the toes to lift your heels (figure 18.9), keeping the knees soft. Slowly return
to the starting position.

„„Figure 18.9 Heel raise.


Strength Training | 267

Distance-Specific Strength Training


Is the approach to strength training the same for the various triathlon dis-
tances? The cycling and running improvements in force, endurance, and
economy that occur with strength training are certainly desirable for tri-
athletes participating or specializing in any distance. But incorporating the
recommended strength training frequency and volume to an already busy
week of training can be challenging. Recovering from strength training in an
8-hour training week is much easier than recovering from the same strength
training that is part of a training week of 15 hours or more.
Therefore, the long-distance triathlete, who will often exceed 15 hours of
training in the latter stages of a training season, needs to map out an annual
training plan that ensures that the MS phase of strength training is complete
before the heaviest volume of training begins, usually in base training. This
plan is practical even for short-distance training, but it is crucial for the long-
distance triathlete, who must complete the strength training periodization
plan and move to a once-weekly strength maintenance program as soon as
possible.

Sport-Specific Strength Training


If gym-based resistance training shows promise of improving endurance
performance, then combining strength training within the specificity of a
workout for a particular discipline might be even better. Here, power workouts
using swim paddles, plyometrics, and hills can play a role in a triathlete’s
training. As with gym-based resistance training, the purpose is to apply high
levels of quick force for a portion of the workout. Unlike dry-land resistance
training, in-water swim resistance training has robust support from scientific
studies. One report found that the measured use of swim paddles may help a
swimmer develop a feel for stroke length (a predictor of swim performance),
because paddles were shown to increase stroke length.33 A swimmer who
becomes accustomed to the velocity and feel of distance per stroke when
using paddles may be able to translate that skill to open-hand swimming.
Other in-water resistance devices and programs have been shown to
improve performance. A device similar to hand paddles improved stroke
length and 200-meter race time by 2.3 seconds.34 Additionally, an in-water
bungee resistance device tethered to the swimmer improved distance per
stroke and velocity more than either an assisted device (pulling the swim-
mer) or using no device at all.35
Endurance runners have long claimed that hill work (a form of plyomet-
rics) has improved their performance, and some evidence backs that up.
An increase in running economy of 2 percent occurred after just 6 weeks of
plyometrics in speeds up to 12.88 kilometers per hour,36 and three plyometric
sessions of 30 minutes per week improved running economy by 4.1 percent
during even faster speeds of 18.0 kilometers per hour.37
268 } Warden

The addition of force to an endurance workout should include brief intervals


of 30 to 60 seconds at a low cadence with high force, using short recoveries of
only 15 to 30 seconds. This kind of workout is best done in hills for cycling
and running. When swimming with paddles to introduce force, the reduced
load-bearing fatigue of swimming would allow the triathlete to swim at a
constant state for 5 to 30 minutes, depending on paddle size.

Conclusion
With the overwhelming endorsement and evidence supporting strength
training for triathletes, strength training should be treated as the fourth
discipline. Triathletes can expect results with predictability similar to that of
core endurance training. As with flexibility, the benefits of strength training
are not universal. Swim performance lacks a strong link to dry-land train-
ing. But the overall established benefits of increased ability to generate force,
economy, and endurance from a periodized gym-based and sport-specific
strength program would be a welcome gain for triathletes of all distances
and experience.
CHAPTER
19
General and
Specific Training
—David Warden

A common observation from the casual triathlon spectator is that relative


to more popular athletic events, triathlon lacks an element of surprise.
No one can dispute that triathlons are exciting, rich with storyline and the
conquering of remarkable individual challenges. But at any given event with
a prize purse, the top 1 percent is relatively consistent. In part because of the
chaotic nature and interplay of team sports, in part because of the individu-
alistic nature of triathlon, endurance performance is somewhat predictable.
But more important, whether genetically or through hard work (or both),
endurance performance results almost exclusively from, and can be predicted
by, the general abilities of aerobic capacity and endurance, the ability to
resist fatigue (muscular endurance), economy (skills), lactate threshold, and
metabolic efficiency. Although an interdependency and a cause-and-effect
relationship certainly exists among these abilities, this chapter seeks to isolate
and address many of these general predictors of endurance performance and
distinguish them from more specific training.

Aerobic Base Building


This book addresses numerous essential training components and issues
(devices, skills, intervals, nutrition, and so on) to help triathletes maximize
success. Although it would be difficult to support the claim that any one
training component is more important than another, if we were to make the
attempt, aerobic endurance training would be an excellent candidate.
The human body uses two kinds of energy producing systems: aerobic
(with oxygen) and anaerobic (without oxygen). Although the definition of
energy system use is unequivocal, defining aerobic exercise can be difficult. If
aerobic exercise were to be defined as exercise intensities that use the aerobic
energy system, it would then include virtually all exercise intensities, because
even at maximal intensities a combination of the two energy systems is used.

269
270 } Warden
Bob Seebohar

If it were to be defined as exercise that uses the aerobic energy system only
slightly more than the anaerobic system, triathletes would still find themselves
exercising at unsustainable intensities.
Aerobic exercise could potentially be quantified as the intensity at which
sufficient oxygen is available for all carbohydrate to be converted into energy.
But this technical threshold is inconvenient for a triathlete to measure without
real-time access to a lab. For the purposes of this chapter, therefore, aerobic
exercise will. refer to deeply aerobic exercise at intensities of less than 60
percent of VO2max.

Need for a Highly Trained Aerobic System


An analysis of the energy system contribution at various run distances
found that the aerobic–anaerobic usage ratio of 400-meter running was
approximately 40–60 but that at 800 meters it changed to 60–40.1 Therefore,
the crossover point for the primary energy system usage from anaerobic to
aerobic is probably somewhere around the intensity required for 600-meter
performance, or perhaps 75 to 90 seconds. Thus, for an exercise event that
lasts beyond 90 seconds, total energy output is mostly aerobic. Because all
triathlon distances cover far more distance and time than that and the aero-
bic to anaerobic ratio can reach as much as 95–5, we can see how much an
endurance triathlete needs to depend on a highly trained aerobic system.
This dependency on the aerobic system is evident in the structures of the
training programs of elite endurance athletes. For example, elite Norwegian
cross-country skiers record the highest levels of maximal oxygen uptake,2 and
71 percent of those skiers’ training is performed at less than 2.0 millimoles
blood lactate, a deeply aerobic intensity. Furthermore, 7 percent is between 2
and 4 millimoles, still below the traditional lactate threshold intensity.3 Note
that these elite athletes essentially trained either very easy or very hard,
avoiding moderate intensities.
Almost 2,000 miles (3,200 km) away, Spanish researchers in Madrid went
a step further and attempted to correlate time spent in aerobic intensities to
performance with elite-level runners. With fascinating consistency, these
researchers found that the elite athletes spent 71 percent of their training time
at low
. intensity, although they defined low intensity as less than 60 percent
of VO2max or less than 70 percent of maximum heart rate, as opposed to
specifying lactate levels.4 Additionally, this research found a relationship
between total training time at low intensities and 10K run performance, .
in which the total time spent at intensities less than 60 percent of VO2max
resulted in better run performance.
Another impressive comparison reviewed two training groups that spent
approximately 80 percent or 65 percent, respectively, of total training in a
low-aerobic intensity and 12 percent or 25 percent, respectively, at a tempo
pace over 21 weeks.5 Both groups also spent a similar
. remaining percentage
of time at high intensities above 90 percent of VO2max. The group spending
General and Specific Training | 271

80 percent of their time at low intensity improved their 10K run performance
by 157 seconds plus or minus 13 seconds, compared with an improvement in
the group spending 65 percent of their time at low intensity of 121.5 seconds
plus or minus 7.1 seconds.
Simply accepting and mimicking the highly aerobic training programs of
elite athletes would be convenient. But the serious triathlete or coach should
also ask these questions: What are the adaptations taking place with aerobic
training that lead to this increase in performance? How can we confirm that
an aerobic base has been achieved?

Adaptations From Aerobic Training


An impressive array of metabolic and physiologic adaptations from aerobic
training can be found in table 19.1, which lists some of the differences between
untrained and trained men. Although each of these adaptations is crucial,
triathletes may be most interested
. in the increase in glycogen storage, per-
centage of muscle fiber type, VO2max, and fat utilization.

Table 19.1 Metabolic and Physiologic Adaptations to Endurance Training


Percentage difference of trained
Variable vs. untrained men
Glycogen +41
Max lactate +36
Slow-twitch muscle fibers +20
Fast-twitch muscle fibers −50
Max cardiac output +75
Max stroke volume +50
.
VO2max +107
Body fat percentage −27
Muscle glycogen utilization* −41
Increase in fat utilization* +63
*Data from B.F. Hurley, P.M. Nemeth, W.H. Martin, et al., 1986, “Muscle triglyceride utili-
zation during exercise: Effect on training,” Journal of Applied Physiology 60(2): 562-567.
Adapted from W.D. McArdle, F.I. Katch, and V.L. Katch, 2010, Exercise physiology:
Nutrition, energy, and human performance, 7th ed. (Baltimore: Lippincott Williams &
Wilkins), 457.

Fat Utilization as an Indicator of Fitness


Although the ability to use fat as a primary source of fuel is not the sole
indicator of aerobic fitness, it is widely used by triathletes and coaches as a
relatively inexpensive and accessible way to determine the level of aerobic
fitness. Metabolic testing cannot provide the measurements of aerobic fit-
ness variables listed in table 19.1, but it is assumed that if a metabolic profile
272 } Warden

shows a high level of fat utilization through broad levels of intensity, the other
variables must also be strong.
For example, slow-twitch muscle fibers (with high fat oxidation) are unlikely
to be well developed if a metabolic profile indicates low fat utilization. Figures
19.1 and 19.2 contrast the fat utilization of two similarly built cyclists at differ-
ent levels of fitness. Note that the well-trained cyclist maintains a high level
of fat utilization (6 or more kilocalories per minute) deeply into high intensity,
whereas the untrained cyclist peaks quickly (3 kilocalories per minute) at a
lower rate of fat usage and it continues to drop as intensity increases.

Metabolic profile, well-trained cyclist


8
7
6
Fat Kcal/min

5
4
3
2
1
0
119 120 120 126 128 131 131 136 143 146 152 160
Heart rate

„„Figure 19.1 Metabolic profile of a well-trained cyclist.


E5643/Friel/Fig. 19.1/448669/TimB/R2-kh

Metabolic profile, untrained cyclist


8
7
6
Fat Kcal/min

5
4
3
2
1
0
116 117 121 124 127 131 134 139 145 150
Heart rate

„„Figure 19.2 Metabolic profile of an untrained cyclist.


E5643/Friel/Fig. 19.2/448670/TimB/R1
General and Specific Training | 273

The implications of substrate utilization (fat versus carbohydrate) are


particularly significant in half- and full-Ironman training. Even a simple
calculation shows that, assuming similar total energy expenditure, the tri-
athlete represented in figure 19.1 will use more than twice as much fat for
fuel as the triathlete represented in figure 19.2. Similar results
. were found at
relatively high intensities of 75 percent to 85 percent of VO2max; endurance-
trained men used 58 percent more fat for energy compared with untrained
men.6
The finding of a dramatic increase in fat utilization found in table 19.1
took place over just 12 weeks of endurance training. This fits well with a
triathlete’s annual training plan, and perhaps suggests spending the initial
mesocycles (the base periods) at approximately 90 percent . of total volume
in intensities below 70 percent of velocity or power at VO2max. Even during
the build phases, elite athletes focus on aerobic training at 70 percent of total
training time in deeply aerobic intensities. See chapter 22 for more about
periodization training.

Anaerobic Base Training


No serious discussion of aerobic base building can be complete without a
review of anaerobic base building. The typical triathlete, however, rarely
needs to call on true anaerobic power. Not only is it rarely needed, it is rarely
recommended. As previously defined, true anaerobic intensity can last only
approximately 75 seconds, and the consequence to the triathlete of going
anaerobic in an endurance event can be dire. What many triathletes think of
as anaerobic is really highly aerobic intensity, such as climbing for several
minutes or going very hard for the last mile of a race.
Still, anaerobic intensity has application in triathlon. Triathletes train-
ing for draft-legal triathlon must be able to maintain proximity to the lead
swimmers and cyclists, because participation in the lead peloton is a require-
ment for success. They must be able to call on anaerobic intensities to avoid
being dropped. Triathletes participating in sprint-distance events, in which
the margin of victory is often just a few seconds, may have to depend on
the anaerobic system in the final few hundred meters of the event. Finally,
anaerobic interval training. of approximately 30 seconds for several weeks has
been shown to increase VO2max, which would improve performance across
all triathlon distances as well as help maintain fitness during the low-volume
competitive phase of a training plan (see chapter 20). .
Other than limited interval training to increase and maintain VO2max, the
decision of how much anaerobic training to include in a training program
should ultimately be based on specificity. If the triathlete expects to call on
anaerobic intensity when racing, it should be included in training.
274 } Warden

General Versus Race-Specific Preparation


Triathletes training for all distances share a set of fitness goals. Despite
the significant difference in event duration from the shortest to the longest
triathlon, some general abilities build the foundation for peak performance
for any steady-state endurance event lasting more than even a few minutes.
At the same time, after those general abilities have been developed, the
skill requirements for a 60-minute event are quite different from those needed
for a 14-hour event. As the competition period nears, the triathlete must alter
training to simulate the stresses and intensities expected for the particular
distance.

General Preparation
Although the history and theory of periodization are covered in chapter 22,
the application of its method and mode in triathlon training fits well within
this section. What is race specific? What is not?
A common misconception among triathletes is that general preparation in
the early phases of a training plan is exclusively long, slow, distance train-
ing. Consistent with the research on aerobic training mentioned previously,
70 percent or more of all training should indeed be deeply aerobic. But what
of the other 30 percent? General preparation does indeed require long, slow
distance, but it does not exclude high-intensity training.
General preparation should focus on increasing the working . capacity of the
triathlete by increasing abilities such as aerobic capacity (VO2max), aerobic
endurance, economy, metabolic efficiency, flexibility, speed, force, and mus-
cular endurance. These abilities are essential for any triathlete competing at
any distance, because they provide the foundation for the subsequent and
specific phases of training.
Achieving these abilities in the general phase often requires the triathlete
to train at intensities or with techniques that appear contrary to his selected
triathlon distance. An Ironman triathlete might question the need to perform
intense intervals at wattage three times what he expects to perform when
racing. A sprint-distance specialist may question the need to perform a 2-hour
bike ride, which is four times her expected cycling duration when racing. But
because all endurance triathletes need these abilities they are considered an
essential part of general preparation, as illustrated in table 19.2.
Additionally, modern triathletes face competition periods that include
triathlons of all distances. Although not recommended for mastery of event-
specific racing, because of personal restrictions and choice many triathletes
find themselves racing both sprint and Ironman in the same season. In such
situations, developing general abilities becomes paramount for the multidis-
tance, multidiscipline triathlete.
See table 19.2 for a list of general abilities applicable to all triathlon distances
and their associated training methods.
Table 19.2 General Triathlon Abilities and Methods
Ability Definition Benefit Training methods
.
Aerobic Ability to use oxygen to High VO2max. The more Interval training
. above, at,
capacity produce energy. . Often oxygen that can be delivered or near VO2max making
measured as VO2max, to the muscles, the more up 15% of total training
a significant predictor intense the exercise that volume (see chapter 20).
of performance at any can be maintained.
. A
triathlon distance. higher VO2max allows
higher velocity .at a lower
percentage of VO2max.
Aerobic Ability to sustain Develops a high oxygen Long, slow endurance
endurance prolonged exercise by transport system, stroke training at
. approximately
maintaining oxygen volume, and cardiac output. 60% of VO2max making
delivery to working Promotes quick recovery. up more than 70% of total
muscles. training volume.
Economy A measure of how Use less oxygen and Long, slow endurance
efficiently a person uses therefore less energy for a training at
. approximately
oxygen at a given output given output or velocity. 60% of VO2max making
or velocity. up more than 70% of total
training volume.
Drills that promote stride
and stroke length, high
cadence, foot strike, and
general technique.
Strength training.
Metabolic Ability to oxidize a Preserves carbohydrate Long, slow endurance
efficiency high level of fat as a stores. training at
. approximately
percentage of overall 60% of VO2max making
energy use. up more than 70% of total
training volume.
Nutrition.
Flexibility Ability to move joints Increase the range through Routine stretching after
and soft tissues through which force is applied and exercise (see chapter 17).
a wide range of motion. support injury prevention
(see chapter 17).
Speed Ability to quickly perform Improve cadence and Cadence drills.
a technique properly. reaction time. Overspeed training.
Force Capacity for muscles to Overcome difficulty in hills, Intervals with low
overcome resistance. improve economy and, cadence (hills, paddles,
support injury prevention plyometrics).
(see chapter 18). Strength training (see
chapter 18).
Power Ability to apply force Limited use but applicable to Strength and interval
quickly. draft-legal races or the final training. at intensities
moments of a race. above VO2max.
Muscular Ability for muscles Resist fatigue and maintain Intervals of 2 to 12
endurance to overcome fatigue race-pace velocity minutes at highly aerobic
and apply force over throughout the race. intensities (see chapter
prolonged exercise. Particularly helpful for time- 20).
trial type performance and
the end of a triathlon.

275
276 } Warden

Specific Race Preparation


As the triathlete nears the competition phase, training must become more race
specific. The goal of the specific training period is to mimic the conditions and
stress of racing to the point that race day feels like an intense, routine workout.
As the triathlete transitions from general to race-specific preparation, the
combined duration and intensity should begin to simulate the stress expected
on race day. The abilities developed in general preparation are not abandoned;
rather, they are gradually replaced by specific training. In many cases, the
methods used to promote general abilities are identical to the methods used
to promote specific abilities.
For example, Ironman triathletes may spend 70 percent of their time at
a deeply aerobic intensity in the general phase and then continue with that
level of intensity in the specific phase. Sprint triathletes may invest time
performing muscular endurance in general preparation and maintain the
same frequency and intensity into their specific training. Table 19.3 lists some
examples of possible race-specific workouts.

Specific Tactical Training


Specific training is not limited to aerobic exercise. Tactics and strategy must be
implemented and practiced early so that they become second nature on race
day. Triathlon training requires as much planning around the characteristics
of the racecourse as around the characteristics of the triathlete.
Tactical training should include sessions that use the same nutrition and
equipment to be implemented on race day. The terrain and environment
must also be considered. Although hill training would be considered a
general ability (to develop force), if the run course for the race includes hills,
hill training should be frequent in the specific phase. If high temperatures
are expected for the event, the triathlete should train in a similar environ-
ment.
Humans also respond to diurnal cycles, and the triathlete who performs
all run training at 6:00 a.m. may be in for an unpleasant surprise when she
hits the run at noon on race day. Although more relevant for sprint and Olym-
pic training, expediting transitions is an important task for which to train
specifically. A triathlete who has not used the specific period to train for all
expected nutritional, equipment, environmental, and strategic contingencies
cannot expect peak performance.
Table 19.3 Specific Triathlon Training Sessions
Distance Swim workout Bike workout Run workout Training race
Sprint 10 × 100 meters 2 × 15 minutes at 95 10 minutes off an Sprint-
leaving every 2 to 100% of threshold intense bike session distance race
minutes power, 5 minutes of at goal race pace
750-meter time trial recovery 5K time trial
Open-water 30-minute time trial 4 × 1 mile (1.6 km)
sessions at 95 to 100% of on 1 minute of rest
threshold power
Olympic 3 × 500 meters on 3 × 20 minutes at 90 20 minutes off an Sprint-
30 seconds of rest to 95% of threshold intense bike session distance race
1,500-meter time power, 10 minutes of at goal race pace
trial recovery 2 × 20 minutes
Open-water 5 × 12 minutes at 90 at goal race pace
sessions to 95% of threshold with 10 minutes of
power, 3 minutes of recovery
recovery 4 × 1 mile (1.6 km)
on 1 minute of rest
Half- 10 × 200 meters 3 × 30 minutes at 80 45 minutes off an Half-marathon
Ironman leaving every 4 to 85% of threshold intense bike session race
minutes power, 10 minutes of at goal race pace Cycling stage
2,000-meter time recovery 3 × 20 minutes
. at race
trial 3-hour ride with 80% of VO2max
Open-water tempo finish at 80 velocity on 5
sessions to 85% of threshold minutes of rest
power 2-hour long easy run

Ironman 4 × 1000 meters on 3 × 60 minutes at 75 1 hour off a long Half-Ironman


1 of minute rest to 80% of threshold bike session at goal race
4,000-meter time power, 15 minutes of race pace
trial recovery at 60% of 3 × 20 minutes
Open-water threshold power at
. 70 to 75% of
sessions 6-hour ride building VO2max velocity on
from 65% to a finish 5 minutes of rest
at 75% of threshold 2-hour long easy run
power
Draft-legal With paddles, 10 × 10 × 30 seconds at Follow Olympic- Cycling
100 meters sprint on 175% of peak power distance run training criterion
1 minute of rest (see chapter 20) on Olympic-
50-meter sprints on 30 seconds of rest distance race
a diving start Follow Olympic-
1,500-meter time distance bike
trial training
Open-water
sessions
General ability workouts continue into the specific phase, but the frequency and volume change to
support the specificity of the upcoming competition phase.

277
278 } Warden

Muscular Endurance Training


Muscular endurance plays a special role for the triathlete. Muscular endurance
is the ability of a skeletal muscle or group of muscles to continue contracting
and applying limited force over a long period; it is the combination of force
and endurance. Although a triathlete’s upper limit of aerobic capacity and
endurance, lactate threshold, and metabolic efficiency will not change during
a triathlon, muscular fatigue is likely the first ability to betray the triathlete
and cause him to slow down, most often during the run. Therefore, muscular
endurance becomes one of the more important general abilities to develop.
To develop muscular endurance, the triathlete cannot simply develop force
and endurance independently. Although these two abilities must indeed be
developed as a prerequisite, true muscular endurance is born from the simul-
taneous application of force and endurance. This training is most commonly
performed at intervals of highly aerobic intensities as discussed in chapter 20.

Skill Training
Identifying skill training as a general or specific ability would be difficult. Skill
work is certainly a general ability, because it is a required base from which
to build new levels of muscle memory and fitness before moving on to more
specific training. Skill work is also required in the specific phase of training,
because desired race intensities cannot be obtained without the supporting
technique. Additionally, general aerobic economy is largely a result of efficient
movement permitted by the development of a skill. Therefore, skill training
is an essential part of the entire training season, from aerobic base building
to general preparation and through the specific phase before competition.

Swim Skill Training


For the triathlete entering the sport as an adult, the swim can be the most
daunting, discouraging, and depressing part of training. Fortunately, age-
group triathletes have had a long history of success in mastering the swim.
Unlike cycling and running, in which a triathlete can gain substantial velo-
city from years of muscling through, achievement in swimming can never be
realized by brute force and determination alone. Swimming success requires
skill work.
A substantial swim literature reflects various philosophies. Researchers
Toussaint and Beek proposed a formula7 that swim velocity could be deter-
mined by
1. drag,
2. power input (aerobic capacity),
3. efficiency (mechanical), and
4. power output (force).
General and Specific Training | 279

In other words, half of the components that lead to great swimming are
technique based. At the risk of oversimplifying the hundreds of books and
philosophies on swim technique, the common principle that they share is to
reduce drag and increase propelling efficiency. Table 19.4 attempts to sum-
marize the most common and effective drills used to support these two
pillars of swimming.

Table 19.4 Swim Drills to Reduce Drag and Increase Propelling Efficiency
Drill Reduce drag Increase propelling efficiency
Swim with snorkel, buoy, or fins X
Swim golf a
X X
High elbow X

Catch-up drill X
Kick drills b
X X
V-line X

Buoy press X

Fist drill X
Kick, count, stroke drill X X
Finger spread c, d X

Swim on side X
a
Alberty, M.R., Potdevin, F.P., Dekerle, J., Pelayo, P.P., and Sidney, M.C. 2010. Effect of stroke rate reduc-
tion on swimming technique during paced exercise. Journal of Strength and Conditioning Research,
25(2), 392–397.
b
Deschodt, V.J., Arsac, L.M., and Rouard, A.H. 1999. Relative contribution of arms and legs in humans
to propulsion in 25-m sprint front-crawl swimming. European Journal of Applied Physiology, 80(3),
192–199.
c
Marinho, D.A., Barbosa, T.M., Reis, V.M., Kjendlie, P.L., Alves, F.B., Vilas-Boas, J.P., and Rouboa, A.I.
2010. Swimming propulsion forces are enhanced by a small finger spread. Journal of Applied Bio-
mechanics, 26(1), 87–92.
Minetti, A.E., Machtsiras, G., and Masters, J.C. 2009. The optimum finger spacing in human swim-
d

ming. Journal of Biomechanics, 42(13), 2188–2190.

Swim With Snorkel, Buoy, or Fins Use a snorkel, buoy, or fins while swim-
ming. Focus on specific arm phases of each stroke to isolate and observe the
phases without the distraction of breathing or kicking. Swim aids can be used
in conjunction with all drills.
Swim Golf For each lap, have someone record the lap time in seconds and
add the stroke count. Try to reduce stroke rate and distance per stroke to achieve
a lower score.
From Alberty, M.R., Potdevin, F.P., Dekerle, J., Pelayo, P.P., and Sidney, M.C. 2010. Effect of stroke rate reduction on
swimming technique during paced exercise. Journal of Strength and Conditioning Research, 25(2), 392–397.

High Elbow Swim with an exaggerated high elbow.


280 } Warden

Catch-Up Drill Do not begin the catch phase until the opposite hand has
entered the water (figure 19.3). This drill promotes quadrant swimming.

„ Figure 19.3 Catch-up drill.


E5643/Friel/Fig. 19.3/448674/TimB/R1

Kick Drills Swim with a kickboard or swim on your back with your arms
extended and hands together. This drill increases total velocity by making arm
strokes more efficient and reduces drag by keeping the legs high.
From Deschodt, V.J., Arsac, L.M., and Rouard, A.H. 1999. Relative contribution of arms and legs in humans to propulsion in
25-m sprint front-crawl swimming. European Journal of Applied Physiology, 80(3), 192–199.

V-Line Swim with the hands slightly outside the shoulders when extended
(figure 19.4) to avoid crossing over the midline of the body.

„ Figure 19.4 V-line drill.

E5643/Friel/Fig. 19.4/448679/TimB/R2-kh
General and Specific Training | 281

Buoy Press Using a snorkel with this drill is best. Consciously press the chest
while kicking with your face down and in the water. Hold your arms at your
sides or extend one arm.
Fist Drill Swim with your fists clenched or while holding tennis balls. This can
be done with both fists clenched at the same time or one hand at a time to
enhance the contrast. Focus on using a high elbow.
Kick, Count, Stroke Drill With one arm extended, kick on your side for a
count of 4, 5, 6, or 7. Take one to three full arm strokes to rotate to your other
side for a new count. Focus on the rotation.
Finger Spread Swim with your fingers slightly spread.
From Marinho, D.A., Barbosa, T.M., Reis, V.M., Kjendlie, P.L., Alves, F.B., Vilas-Boas, J.P., and Rouboa, A.I. 2010. Swimming
propulsion forces are enhanced by a small finger spread. Journal of Applied Biomechanics, 26(1), 87–92; and Minetti, A.E.,
Machtsiras, G., and Masters, J.C. 2009. The optimum finger spacing in human swimming. Journal of Biomechanics, 42(13),
2188–2190.

Swim on Side Swim with your navel facing the wall, with or without one arm
extended (figure 19.5). Wearing fins for this drill is best.

„ Figure 19.5 Swim on side drill.

Bike Skill Training


E5643/Friel/Fig. 19.5/448681/TimB/R1

Although not as complex as swimming, cycling can be improved through


specific skill training. Like swimming and running performance, cycling
performance is ultimately a product of stroke rate and distance per stroke. In
a relatively fixed position, the cyclist can either increase distance per stroke
through force (big gear) or increase the efficiency of the stroke. Triathletes
should therefore include cycling drills that promote a high cadence and pedal
stroke efficiency.

Cadence
Even a cursory review of professional cyclists reveals that they use a cadence
within a few revolutions of 90 rpm unless they are climbing. A formal review
of professional cyclists competing at the Giro d’Italia, Tour de France, and
Vuelta a España revealed a mean cadence of 89.3 plus or minus 1.0 during the
group stages and 92.4 plus or minus 1.3 during time trials.8 Although many
studies prove that a high cadence reduces gross efficiency9, 10 (comparing
282 } Warden

total work performed to total energy expenditure), this drawback appears to


be compensated by advantages including better run times off the bike, more
work done in the recovery phase of the stroke, increased pedaling efficiency,
and more fat oxidation.
Additional benefits of a high cadence include improved subsequent run
performance,11 an increase in pedaling efficiency, and recruitment of more
muscle groups in the recovery phase of the stroke.12 Furthermore, a lower
cadence requires more force and therefore more recruitment of fast-twitch
muscle fibers. Long-term cycling with a low cadence theoretically changes the
triathlete’s metabolism to use more carbohydrate. In contrast, a high cadence
promotes slow-twitch development. Although perhaps not a concern for
short-course triathletes, long-course nutrition could be affected by decreased
fat utilization from a low cadence.
But the preceding does not mean that a triathlete should spend all her time
drilling at a 90 or more rpm. The development and maintenance of force and
power developed through low cadence can translate into improved hill work,
time-trial performance, and aerobic capacity.13, 14 In summary, the triathlete’s
bread and butter should be a high cadence, but force should be developed
and maintained year-round.
To drill for increased cadence,
• practice 30 right or left foot cycles in 20 seconds (equals 90 rpm),
• practice cadences of 100 or more rpm for several minutes at a time, and
• use a cycling computer with a cadence function year round to obtain
real-time feedback.

Pedal Stroke Efficiency


Like any machine, a cyclist increases efficiency by administering power
uniformly through the range of motion of a lever. The cyclist must overcome
the tendency to rely on the downstroke of one leg to assist the recovery of
the opposite leg during the pedal stroke. This “free ride” from the recovery
leg is lost power.
To overcome this common cycling deficiency, the triathlete must prac-
tice drills that encourage the hip flexors and hamstrings to lift during the
recovery phase, thus changing the recovery phase to an active phase. This
ability can be accomplished through a combination of the following drills
and equipment:
• Single-leg drill (indoor only). With one leg resting on a stool, the oppo-
site leg is isolated and cycles independently. The focus should be on the
upstroke and a smooth transition from downstroke to upstroke.
• Scraping drill. As the downstroke finishes, imagine scraping mud off the
bottom of the shoe and pulling the foot back toward the rear of the bicycle.
General and Specific Training | 283

• PowerCranks are an independent crank system that forces each leg to


complete the stroke independently. These devices are best added to a
separate bike because they replace your existing crank system.
• Various power meters including the CompuTrainer bike trainer system
provides real-time feedback on pedaling efficiency through the SpinScan
software interface, including the amount of power generated during each
point in the pedal cycle. The Polar CS600X and LOOK Kéo Power power
meters display a pedaling index, the ratio between the minimum and
maximum forces of a single pedaling cycle to determine how round the
distribution of power is.

Run Skill Training


The primary goal of the runner should be to increase running economy or
the oxygen cost for a given velocity through a quick cadence, reduced braking
and contact time from a midfoot strike, and a proud posture. Here are some
drills to improve running economy.
Cadence Practice 30 right or left foot strikes in 20 seconds (equals 90 rpm).
Landers, G.J., Blanksby, B.A., and Ackland, T.R. 2011. Cadence, stride rate and stride length during triathlon competition.
International Journal of Exercise Science, 4(1), article 6. Available at http://digitalcommons.wku.edu/ijes/vol4/iss1/6.

High Knees With a slight forward lean, stay tall while quickly lifting and driving
down the knees. Cadence is rapid. Forward movement is not relevant. Striking
at the heel is impossible if the drill is done properly.
Hands on Head Alternate between normal running and running with the
hands on the head. Any lateral movement in your running gait (inefficiency)
will be amplified.
Glute Kicks With the thigh pointing directly down and a slight lean, kick the
legs up to the gluteal muscles. Forward movement is not relevant. The purpose
is the quick action toward the gluteal muscles.
Ballerina Overemphasize the forefoot strike on the toes with high cadence. This
running form is not optimal. The purpose is to discourage rear or heel strikes
in favor of a midfoot strike.
From Hasegawa, H., Yamauchi, T., and Kraemer, W.J. 2007. Foot strike patterns of runners at the 15-km point during an
elite-level half marathon. Journal of Strength and Conditioning Research, 21(3), 888–893; and Ardigo, L.P., Lafortuna, C.,
Minetti, A.E., Mognoni, P., and Saibene, F. 1995. Metabolic and mechanical aspects of foot landing type, forefoot and
rearfoot strike, in human running. Acta Physiologica Scandinavica, 155(1), 17–22.

Knee Drive Lead the body with the knees, driving forward.
.
Strides Perform 6 × 30 left-foot strikes at VO2max velocity and then walk
back for 90 seconds to rest. Focus on using a high cadence and lifting the heel
directly up to the gluteal muscles.
284 } Warden

Conclusion
Although the triathlete may be tempted to despair at the assertion that triath-
lon performance is largely predictable, this realization should be considered
largely good news. Armed with an understanding of some of the primary
predictors of endurance performance, including aerobic capacity and endur-
ance, the ability to resist fatigue (muscular endurance), and economy (skills),
the triathlete can modify training to develop these primary and other general
abilities that will coalesce for race-day success.
CHAPTER
20
Interval Training
—David Warden

T he interval may be a triathlete’s most effective tool. Interval training


involves repeated short to long periods of high-intensity exercise, broken
up with recovery periods of light exercise or rest. Intervals allow triathletes
to perform more work at higher intensities than they could if they attempted
those same high intensities without a rest interval.
In an early study of intervals, athletes were able to double the total amount
of work done at high intensity when the rest interval was doubled.1 When
the rest interval was tripled, the total amount of work done at high intensity
had no measurable limit. People performing run intervals at a duration of
50 percent of the time to exhaustion, with rest periods of the same duration,
were able to spend more than twice as much time at the same velocity as
they did when running one set to exhaustion.2

Benefits of Interval Training


Intense interval training not only allows the individual to do more total
work than when . doing uninterrupted intensities but also allows the body
to maintain VO2max for . a longer period of the workout. Run interval train-
ing at 100 percent
. of VO2max velocity, interspersed with. rest periods at 50
percent of VO2max velocity, allows runners to stay at VO2max even during
a portion. of the rest interval, resulting in nearly three times the amount of
time at VO2max compared with continuous running at the same intensity.3
This ability
. to complete more work at higher intensities and to spend more
time at VO2max contributes directly to a triathlete’s increase in fitness. Theo-
retically, by more than doubling the amount of time spent at high intensities,
positive physiological changes would follow more rapidly than from less total
time spent at high intensities.

Performance Improvements
Empirically, intervals have been proved to improve performance over endur-
ance training alone. Although the interval work intensities in these studies

285
286 } Warden
Bob Seebohar

.
vary
. from as low as 80 percent of VO2max output to as high as 160 percent of
VO2max output, considerable improvement from interval training has been
measured in several predictors of performance as well as actual performance
results. .
An interval study. found significant improvement in overall VO2max, . the
running velocity at VO2max, the time that the runner could spend at VO2max
velocity, and overall 3,000-meter run time.4 This improvement over multiple
measures of performance occurred after 10 weeks of two sessions of interval
training per week. No improvement occurred in the control group, which
performed only endurance training.
Fortunately for triathletes, the literature supporting positive results from
high-intensity training is common across all three disciplines. Thirty-nine
well-trained cyclists and triathletes found that two sessions of interval
training per week, in addition to their endurance training, improved their
40-kilmeter time trial performance by 4.4 to 5.8 percent, an improvement of
approximately 3 minutes.5 The cyclists and triathletes who maintained only
endurance training had no improvement in their time trial.
When competitive young swimmers were split into two groups, high
volume and high intensity, the high-intensity group spent 50 percent less
total time swimming (30 minutes versus 60 minutes of main set) but swam
at higher intensity intervals.6 The result for these high-intensity swimmers
was a 2.8 percent improvement in 2,000-meter swim performance, compared
with a 1.8 percent improvement for the high-volume swimmers.

Physiological Changes
Although the ultimate goal of any elite triathlete is to reduce the total amount
of time to complete a specific distance, the physiological changes associated
with the improved time trial have broad implications. In a previously cited
study, the cyclists’ and triathletes’ measures of peak oxygen consumption
before interval training were already an enviable 64.5 plus or minus 5.2 (as
measured in milliliters per kilogram per minute).5 The interval training not
only improved 40-kilometer time trial performance (a factor in Olympic-
distance triathlon success) but also increased peak oxygen consumption
between 5.4 and 8.1 percent from the already elite preinterval levels. .
Similar increases of 5.5 milliliters per kilogram per minute in peak VO2 after
6 weeks of interval training
.
7
were found in moderately trained athletes. This
increase in general VO2max from interval training in both elite and . moderate
athletes has implications across all triathlon distances because VO2max is a
general predictor of endurance performance.

Enjoyment of Interval Training


Perhaps as important as the performance increase itself, interval training
may play an important role in training adherence and overall enjoyment of
Interval Training | 287

training. Runners who performed either 50 minutes of moderate-intensity


running, without intensity variation, or 50 minutes of high-intensity intervals
(total interval work equaled 18 minutes) were asked to rate their level of enjoy-
ment of the workout using the Physical Activities Enjoyment Scale (PACES).8
Despite no difference between the two groups in average heart rate and total
work completed within the 50-minute run, the high-intensity interval group
scored 9 points higher on the enjoyment scale.
The result of this increase in pleasure is likely not because of the increased
intensity but because of the intensity spread over the intervals. Multiple
research9, 10 supports the concept that an increase in the intensity of continu-
ous exercise generally leads to a decrease in enjoyment. Intervals allow the
total workout intensity to be higher with more total intense work completed,
and breaking up the intensity with rest intervals allows more total pleasure
in the workout. A more enjoyable training regimen results in an increased
chance of exercise adherence and therefore improved performance.11

Interval Volume
If interval training is so effective, then why not either include it in every
workout or use it for a high percentage of total training volume? Although
a triathlete may be tempted to add high volumes of intensity, either as con-
tinuous or interval sessions, because of their effectiveness, doing so carries
significant risks, including injury, compromised recovery, metabolic inef-
ficiency, and overtraining.
People generally agree that the total amount of exercise is associated with
increased risk of injury. Total amount of exercise, or exercise volume, is a
combination of exercise frequency, duration, and intensity. An increase in
any of these elements, particularly intensity, can increase the risk of injury
to the athlete.
Additionally, an increase in intensity often occurs at the expense of aerobic
work. The benefits of an aerobic base will not be discussed in detail in this
chapter, but they include fat oxidation, development of slow-twitch muscle
fibers, and an increase in the total number of red blood cells, to name just a few,
and they develop only after considerable aerobic duration. A disproportion-
ate level of interval training will likely reduce the volume of this important
low-intensity training. Aerobic training volume should always make up the
bulk of a triathlete’s training.
In fact, an aerobic base facilitates the ability to perform intervals more
effectively. Aerobic fitness translates into faster recovery from interval
sets because of increased aerobic response, improved lactate removal,
and enhanced phosphocreatine regeneration.12 In other words, a triath-
lete desiring quality anaerobic intervals must maintain an aerobic base to
support an effective recovery interval and therefore an effective overall
interval session.
288 } Warden

Perhaps the greatest consequence of high-intensity interval training is


the risk of overtraining. Although overtraining is a complex condition that
includes physiological, physical, and emotional elements, reduction in hemato-
crit, hemoglobin, and red blood cell count are all associated with overtraining.
Unfortunately, interval training can play a direct role in the reduction of
these essential oxygen-carrying cells and molecules. In one study, hematocrit,
hemoglobin, and red blood cell count were all significantly reduced in just
3 weeks of interval training 5 days per week.13 Serum ferritin, an indirect
measure of iron in the blood, remained low even after the interval training
was completed, despite an iron supplementation throughout the training at
260 percent of the U.S. recommended daily allowance. Although this weekly
interval frequency (five) is abnormally high, it illustrates just how quickly
a triathlete’s iron stores can be depleted from interval training. Like severe
injury, overtraining because of insufficient iron stores can cause months of
training interruption.
Because of these collective disadvantages of too much intensity, an optimal
training program includes approximately 15 percent of total training dura-
tion at highly aerobic or anaerobic intensities, as demonstrated by multiple
studies .14, 15, 16, 17

Interval Structure
The interval structure is composed of the work interval duration and repe-
titions, total interval duration, intensity, and rest period. For example, an
interval set of 5 × 5 minutes on 1-minute rest would have a total interval
duration of 25 minutes (5 × 5 minutes), a work interval of 5 minutes, a work
interval repetition of 5, and a rest period of 1 minute.
Interval structure depends on the interval type and purpose. Intervals
can be categorized into two types. General
. intervals are designed to increase
broad fitness parameters such as VO2max and muscular endurance. These
are typically high-intensity, anaerobic, or highly aerobic short-duration work
intervals at outputs between 15 seconds and 6 minutes. The purpose of these
intervals is to increase a triathlete’s base speed, which can translate into a
faster steady-state aerobic output.18 Triathletes training for any distance would
benefit from this increase in base thresholds by incorporating general intervals
into the early period of an annual training plan.
Specific intervals are intervals performed at or close to the pace specific to
the triathlete’s race distance. For example, the sprint-distance triathlete may
perform specific intervals at 100 percent of his 1-hour threshold intensity,
whereas a long-distance triathlete’s specific intervals may be performed at
only 80 percent of 1-hour threshold intensity. These intervals are designed
to prepare the triathlete for the particular stress and duration of an endur-
ance event. They are highly aerobic work intervals lasting several minutes to
hours. As the triathlete approaches her competition period, the number and
Interval Training | 289

volume of intervals done at specific race intensity should replace the more
intense general intervals.

General Intervals
.
General intervals appear to increase VO2max and muscular endurance best
at work interval durations anywhere from 15 seconds to 6 minutes,5, 17, 18, 19, 20
but not necessarily all durations within that window. For example, benefit in
work interval duration was specifically found at either 15 seconds or 4 minutes
at 90 to 95 percent of maximum heart rate.19 Similarly, work intervals of 30
seconds or 4 minutes are better at increasing muscular endurance compared
with intervals at 1, 2, and 8 minutes because intensities increased inversely
to the work interval duration.18 .
Considerable improvement in muscular endurance and VO2max was found
at work interval durations of approximately 2 to 3 minutes,
.
5, 20
or, more specifi-
cally, interval durations at 60 percent of the time that VO2max output could
be maintained, typically around 4 to 6 minutes for a well-trained triathlete.
Other interval
. research showed that increased performance used intensi-
ties below VO2max output of 4 to 5 minutes in duration.17, 18, 19 , 21 This finding
supports an interval training
. program that includes either very short intense
work intervals above VO2max output of . 15 to 30 seconds (supramaximal),
intervals at intensities associated with VO2max of 2 to 3 minutes (maximal),
or intervals of 4 to 5 minutes at intense but submaximal outputs. Figure 20.1
quantifies the physiological effects of incorporating these various interval
intensities into a training program.

Physiological effects of high-intensity interval training


6%

5%

4%
Submaximal, 90% VO2max
3%
Maximal, 100% VO2max

2% Supramaximal, 130% VO2max

1%

0%
Maximum oxygen Anerobic Economy
uptake (VO2max) threshold

„„Figure 20.1 Physiological effects of high-intensity interval training.


E5643/Friel/Fig. 20.1/448682/TimB/R3-kh
Adapted from C.C. Paton and W.G. Hopkins, 2004, “Effects of high-intensity training on performance and physiology of
endurance athletes,” Sportscience 8: 25-40.
290 } Warden

General Interval Intensity


Naturally, the intensity of the interval increases as the duration of the work
interval or rest period decreases. Work intervals in the 15- to 30-second range
show positive results at intensities of 90 to 95 percent of maximum heart rate
for running19 and 175 percent of peak power output (defined as peak 30-second
power in a progressive exercise test) for cycling.18
Intensity for running or cycling working intervals
. of 2 to 3 minutes also
appears to be best at output associated with VO2max.5, 20 For longer intervals
of 3 to 6 minutes, submaximal intensities of 80 to 85 percent of 30-second
peak power for cycling all had positive
. results .5, 20, 21
A protocol for determining VO2max output and 30-second . peak power
can be found in the sidebar Determining Peak Power and VO2max Velocity
or Output.
Intuitively, a triathlete might assume. that shorter, more intense anaerobic
intervals would be best for increasing VO2max and the longer highly aerobic
intervals would be best for improving muscular endurance. The evidence
reviewed
. here seems to suggest that both work interval durations can increase
VO2max and muscular endurance, although the literature leans toward the
longer 2- to 6-minute work intervals as the superior duration. But . although
both 15- to 30-second and 3- to 4-minute intervals increase VO2max, the
increase is more significant with 2- to 4-minute intervals.4, 5, 19, 22
Additionally, a review of 22 studies on interval training revealed that the
greatest benefit across a. spectrum of fitness measures occurred with inten-
sities associated with VO2max lasting several minutes.23 When comparing

.
Determining Peak Power and VO2max Velocity or Output
Peak 30-second power protocol (bike with power meter)
1. Warm up for 5 minutes at a self-selected pace.
2. Begin at a workload of 100 watts.
3. Increase the workload by 15 watts every 30 seconds until exhaustion.
4. Peak power is the highest 30-second power output completed during the
incremental test.
.
V O2max velocity or output (all disciplines)
1. Warm up for 20 minutes at a. comfortable pace and include several
10-second bursts at estimated VO2max velocity or output.
2. Perform a 5-minute time trial at maximum maintainable pace.
.
3. VO2max velocity or output is the average velocity or power output during
the time trial.
Interval Training | 291

work interval intensity and durations of submaximal (greater than 10 min-


utes), maximal (2 to 10 minutes), and supramaximal (less than 2 minutes),
this review found overwhelming benefit from the maximal intervals over
supramaximal intervals .in measures of submaximal endurance, maximal
endurance, peak power, VO2max, anaerobic threshold, and economy. There-
fore, a triathlete should consider a training mix of both short and longer
general intervals, with the frequency skewed toward intervals of 2 to 6
minutes.

General Interval Duration


Work interval repetitions should be inversely related to the work interval
duration. As the work interval decreases, the number of repetitions increases.
When considering how to structure the repetitions of an interval workout,
the total interval duration (work interval duration times repetitions) should
be considered.
For supramaximal intervals, repetitions of 10 to 20 at 30 or 60 seconds have
been shown to be effective,23 and additional research shows improvement in
15-second intervals.19 The consistent total interval duration of these studies
was 4 to 6 minutes, although one study was a clear outlier at 20 minutes of
total interval duration24 (20 repetitions times 1-minute work interval at 100
percent of peak power).
But that study found that 1-minute intervals resulted . in improvement
only in peak power and ventilatory thresholds, not in VO2max. The other
studies of shorter, more intense 30-second intervals (i.e., 175 percent of peak
power as well as the research about 15-second intervals . at up to 95 percent
of maximum heart rate) all found increases in either VO2max or time-trial
performance, measurements much more desirable than peak power in a
steady-state endurance athlete. Therefore, the triathlete should structure the
workout to include these general supramaximal work interval durations of
15 to 30 seconds and total interval duration of 4 to 6. minutes.
Research on interval intensities associated with VO2max (maximal inter-
vals) supports a total interval duration of 15 to 45 minutes made up of indi-
vidual work intervals ranging from 2.5 to 5 minutes. Although no empirical
evidence is available on an upper limit to the total interval duration, the
literature suggests that effective performance can result from
. a total interval
duration of up to 45 minutes at intensities close to or at VO2max.
This total volume of intensity in a given workout appears to have an effec-
tive balance of risk and reward; the risk of overtraining or injury is relatively
low, but the documented fitness reward is high. The triathlete should structure
the. workout to include work interval durations of 2 to 6 minutes at or close
to VO2max output
. and total interval duration of 15 to 45 minutes. Intensities
should
. be at V O2
max output closer to the 15-minute range and just under
VO2max output closer to 45 minutes.
292 } Warden

General Interval Rest Period


The rest period between work intervals is the key to intervals and the primary
reason that triathletes can perform more total high-intensity work than they
can when training at uninterrupted intensities. Interval training must be
implemented in a way that allows sufficient recovery to promote increased
total work but does not permit excessive rest that would compromise fitness
gains.
General intervals in this chapter have been divided into either . very short
work intervals of 15 to 30 seconds at intensities well above VO2max output
(supramaximal),
. work intervals of 2 to 3 minutes at outputs associated
. with
VO2max (maximal), or intervals of 3 to 6 minutes at outputs below VO2max
(submaximal). Studies on rest periods for these three work interval intensi-
ties are consistent, but the structure changes based on work interval intensity
and repetitions. .
For example, Billat et al. found positive results in total time at V. O2max
from a 1:1 ratio of work interval duration to rest interval duration at VO2max
velocity, but the work and rest interval durations were only 30 seconds at
this maximal intensity.3 Similar results were found with a 1:1 ratio at 15 sec-
onds.19 At supramaximal intensities, the rest interval must become longer.
A 1:9 ratio at 30-second intensities
. at 175 percent of peak power resulted in
a significant . increase in VO2max. Running velocities of 30 seconds at 130
5

percent of VO2max velocity also showed a 1:9 ratio.4


The 1:1 ratio is appropriate for maximal intervals of less than 30 seconds;
work of 2 to 6 minutes requires a change in that ratio. A 1:2 work-to-recovery
rest interval promoted an increase in performance at maximal intensity inter-
vals of 2 to 3 minutes,5, 20 whereas a rest interval of 1 minute for submaximal
intervals of 4 to 6 minutes is consistently used in testing that resulted in
improved performance.17, 18, 21
Perhaps the most interesting research on rest intervals was done with the
work interval intensity, duration, and repetitions identical and the only differ-
ence being that the interval set was either an approximate 1:1 or 1:2 work-to-
rest ratio.5 Intuitively, the assumption would be that the same amount of work
in a shorter period would be more likely to improve performance. . Although
both rest intervals promoted nearly identical improvement in VO2max, the
group of athletes using the 1:2 work-to-rest interval ratio actually did more
work in a subsequent time-to-exhaustion test. In the case of rest intervals,
shorter is not always better. Triathletes can safely use a 1:2 work-to-rest ratio
at maximal intensities of 2 to 3 minutes.
The term rest interval should not imply completely inactive (passive) rest.
Several studies 25, 26, 27 have shown that an active rest interval (anything from
brisk walking . or light spinning to easy running) at intensities of 30 to 65
percent of VO2max velocities is superior in clearing lactate when compared
with complete rest.
Interval Training | 293

General Interval Frequency


The frequency of interval training for improved performance appears to be
well established at two to three times per week. Interval training of two times
per week is superior to one time per week, and the magnitude of improve-
ment is related to frequency.28
But like many aspects of triathlon
. training, more is not always better. One
study found no difference in VO2max improvement when comparing inter-
val training at two or four sessions per week.29 Significant improvement in
multiple markers of performance was present in as little as four sessions in
6 weeks,21 two sessions per week ,5, 17, 18, 20 or three times per week.19, 22
Considering the evidence that more than three interval sessions per week
is no more effective than four per week and the evidence that five sessions
per week can rapidly lead to potential overtraining, it appears that triathletes
should aim for two or three high-intensity general interval sessions per week,
or perhaps one session per discipline per week of either a supramaximal,
maximal, or submaximal interval set. An example in a given week might
include a swim interval session of 12 × 25 meters on 20 to 30 seconds of rest
(supramaximal), a run interval session of 6 × 800 meters on 6 minutes of
rest (maximal), and a cycling interval session of 6 × 6 minutes on 1 minute
of rest (submaximal).
Theoretically, a triathlete could use interval training to work on sport-spe-
cific weaknesses. For example, a triathlete who is strong in run performance
but weak in swim performance could modify the weekly training plan to
include two run sessions and one cycling session per week.

Periodization of General Intervals


In the research cited in this chapter, positive results occurred in as little as
3 weeks,18 although other studies lasted 4 to 8 weeks. But when comparing
interval training of 7 or 13 weeks, the trend was to see greater improvement
at 13 weeks.29 In a periodized training plan, this finding would suggest that
the triathlete do 2 to 3 months of general interval training (perhaps during
the base phase; see table 20.1) before moving into more race-specific interval
training and doing interval intensities at race pace in later phases.
This plan does not mean that general high intensity would be eliminated
from the training plan as the triathlete moves into more race-specific train-
ing or actual racing. Researchers experimented with high-intensity intervals
during the competitive phase in highly trained cyclists. The results were an
increase in endurance performance when continuing high-intensity interval
training during the competitive phase compared with a control group that
did not participate in high-intensity training.30
As expected, the interval duration and repetitions were short at 5 × 30
seconds on 30 seconds of rest on 2 or 3 days per week, enough to promote a
significant increase in performance without adding additional fatigue in the
294 } Warden

Table 20.1 General Interval Structure, Base Training Period


Rest period Frequency per
Work Total work (work-to- week (total, not
Interval type Intensity interval duration rest ratio) per discipline)
Supramaximal 175% of peak power 15 to 30 4 to 6 1:9 ≤1
(cycling) . seconds minutes
130% of VO2max
velocity (running,
swimming)
90 to 95% of max heart
rate
.
Maximal 100% of VO2max output 2 to 3 15 to 30 1:2 1 or 2
minutes minutes
Submaximal 80 to 85% of peak power 4 to 6 30 to 45 4:1 to 6:1 ≤1
(cycling). minutes minutes
90% of VO2max velocity
or output (cycling,
running, swimming)
Supramaximal intervals of 5 × 30 seconds with 30 seconds of rest 2 or 3 days per week can be continued
into the competition phase.

competition phase of training, and used the philosophy of 2 to 3 months of


general intervals followed by limited interval training replaced by specific
intervals.

Specific Intervals
Unlike general high-intensity intervals, race-specific intervals have had little
attention directed to them in the scientific literature. Although the interval
data on the individual disciplines of swimming, cycling, and running are
mature and rich, the relatively immature sport of triathlon has been little
studied by the scientific community.
Fortunately, we have the principle of specificity to guide the triathlete.
Although
. general intervals will place the triathlete at an increased baseline of
VO2max and muscular endurance (predictors of performance at any triathlon
distance), specific intervals must be the primary method of interval training
as the triathlete nears the competition phase. The goal is to introduce the
specific intensities and stress that the triathlete will encounter during the race.
Coaches and triathletes widely believe that although interval training must
begin to mimic the actual race intensity as the competition phase approaches,
the maintenance of muscular endurance (or the ability to generate force for
long periods) is critical for all triathlon distances. Intervals that promote
muscular endurance should be continued from the general phases of training
(as submaximal intervals) even into the more specific phase.
Although part VIII of this book will cover event-specific training, including
interval training for sprint, Olympic, half-Ironman, and Ironman distances,
some brief examples of specific interval training are listed in table 20.2.
Interval Training | 295

Table 20.2 Specific Interval Structure Examples, Base or Build Phases


Frequency
Race Work Total work per week per
distance Intensity interval duration Rest period discipline
Sprint At goal race pace 6 to 12 30 to 60 2 to 3 1 or 2
minutes minutes minutes
Olympic At goal race pace 6 to 12 30 to 60 2 to 3 1 or 2
minutes minutes minutes
Half-Ironman At goal race pace 10 to 30 30 minutes 2 to 5 1 or 2
minutes to 3 hours minutes
Ironman At goal race pace 10 to 60 30 minutes 2 to 10 1 or 2
minutes to 6 hours minutes
.
All distances 80 to 90% of VO2max 6 to 12 30 to 60 2 to 3 1 or 2
(muscular velocity or output minutes minutes minutes
endurance (cycling, running,
maintenance) swimming)

Conclusion
.
The impressive increase in VO2max from interval training is reason enough
for triathletes to implement this type of training, but interval training offers
additional benefits. Triathletes rarely perform only one distance during their
triathlon careers. Many find themselves participating in triathlons lasting
from 60 minutes to 17 hours.
To be prepared for events that have this significant spread of duration and
intensities, the triathlete can maintain base fitness for an event of any length
through general interval training and then switch to specific intervals before
the particular event. Thus, the triathlete can maintain a dynamic annual train-
ing plan and be prepared to enjoy all triathlon events. Finally, although the
performance benefits of interval training are well established, the additional
benefit of increased enjoyment and associated increase in training compliance
add an unexpected advantage from interval training.
This page intentionally left blank.
Pa r t

VI
Training Strategies
in Triathlon
This page intentionally left blank.
CHaPTEr
21
Duration, Frequency,
and Intensity
—Stephen J. McGregor, PhD

W hen contemplating the overall training program, triathletes and coaches


alike are confronted with a profound question: How much, how often,
and how hard should I train? For triathletes in particular, a complex set of
variables complicates the answer to this question since triathletes must attain,
or maintain, proficiency in three disciplines, as opposed to other endurance
athletes who typically need to train for only one. Because of this, the answer
to the question of how the athlete should structure his or her training is not
necessarily straightforward.
This requirement to spend time training in each discipline dictates the need
for a relatively high frequency of training that ultimately results in a high
volume of total training time. At the same time, the intensity (or difficulty)
of the training must be modulated to stimulate fitness adaptations but allow
recovery and rejuvenation. Therefore, the demands of training present special
challenges to triathletes that are unique among endurance sports.
Generally, three principles guide training for any kind of sport: overload,
specificity, and reversibility.1, 2
• The principle of overload states that for a tissue or system to adapt, it must
be exposed to stress that is greater than that to which it is accustomed.
• The principle of specificity asserts that for a tissue or system to adapt to
a specific requirement, it needs to be stressed in a manner similar to that
requirement. A simplistic example might be the intuitive notion that the
legs can be trained for endurance by running or cycling, but they will
adapt to each discipline in a specific fashion.
• Finally, the principle of reversibility maintains that after a stress is
removed, the adaptation that was elicited will be lost; in simple terms,
Use it or lose it.

299
300 } McGregor

Each of these principles should be considered with regard to training


for triathlon, but in this chapter we focus primarily on overload. That said,
specificity and reversibility are part of the subtext underpinning the overall
approach to training. An approach designed to manipulate intensity will
depend to some extent on the nature of the event (e.g., sprint, Olympic,
Ironman) and will therefore be determined by the principle of specificity.
Similarly, the principle of reversibility should be considered when devising
the sequence of individual discipline training sessions, because if overload
is neglected in a particular discipline (e.g., reducing swim overload to
accommodate run overload), the adaptations in that discipline may suffer,
relatively speaking.
Overload can be manipulated in one of three ways: frequency of stress,
duration of stress, and intensity of stress. Another factor to consider is
volume, which is a function of duration and frequency. Further, each of these
main factors can be applied in different ways. For example, intensity could
be manipulated by increasing the load in strength training, the speed in
running, or the power in cycling. Similarly, frequency can be manipulated
in terms of sessions per day, per week, per month, or a combination thereof.
In the case of individual disciplines, we will consider intensity overload in
swimming primarily in terms of pace, in cycling in terms of power and heart
rate, and in running in terms of pace and heart rate.

Frequency
The frequency of training is an important parameter to consider because it
contributes to total training volume. But in the context of triathlon, the chal-
lenge is maintaining sufficient frequency of training stress for three individual
disciplines that collectively contribute to overall sport success.
Few studies have examined the direct effect of training frequency, par-
ticularly in highly trained athletes. The response of untrained people to
various training frequencies has been well established for decades, though.
Optimal results are achieved with three to five training sessions per week,
and performing more than five sessions does not increase cardiovascular
adaptations.3, 4, 5, 6 Because serious triathletes will likely be training more than
five times per week, these guidelines are of little value.
. On the other hand, in recently trained people, training adaptations
(VO2max) can be maintained with as little as 2 to 4 days per week of train-
ing.7, 8 Because training adaptations, even within the endurance domain,
are somewhat specific to the activity being practiced, the observation that
optimal adaptations occur within 2 to 4 days per week, with diminishing
returns above that, may be applicable to each individual discipline within
triathlon. Therefore, a training regimen that incorporates 3 or 4 workouts per
discipline per week would result in 9 to 12 training sessions per week, which,
anecdotally, is a common weekly training frequency range for triathletes.
Duration, Frequency, and Intensity | 301

Duration and Volume


The duration of time for individual training sessions depends on a number
of factors, such as event specificity (e.g., Olympic versus Ironman), periodiza-
tion, and injury prevention. The latter factor is an important consideration
because injuries are common in triathlon, particularly as a result of run-
ning.3, 4, 5, 6 For athletes focused on Ironman-distance events, though, large
volumes of training will be necessary, consisting of some training for long
duration.
There is no evidence, though, regarding the individual contribution of
training duration on adaptations in trained triathletes, so factors such as
those mentioned previously need to be considered for this training variable.
Training volume is a function of frequency and duration, and because three
disciplines need to be trained, global training frequency will likely be high
for most triathletes, as mentioned previously, in the range of 9 to 12 training
sessions per week or more.
Because a relatively high training frequency is a necessary component of
triathlon training, changes or manipulations of training volume will come
primarily from manipulations of training duration. Total training volume
will be dictated by a number of factors but will depend largely on event
specificity. Recently, Knechtle et al.9 compared participants in Triple-Ironman
Germany and Ironman Switzerland. The authors reported that the Triple-
Ironman athletes (n = 64) trained on average approximately 19 hours per
week whereas the Ironman athletes (n = 71) trained approximately 14 hours
per week. In another recent study, Neal et al.10 reported that 10 masters-level
triathletes (42 years of age) trained between 8 and 11 hours per week over
the course of 6 months.
In both of the aforementioned studies, the majority of the training volume
(approximately 50 to 60 percent) was allocated to cycling. The second greatest
volume was devoted to running, and the least to swimming. These were not
elite triathletes, so it’s not clear whether these training volumes are optimal,
but these data can serve as benchmarks for age-group triathletes. Anecdotally,
for elite or professional-level triathletes, training volumes commonly exceed
20 hours per week (unpublished observations).
Although it is important to consider the training variables frequency, dura-
tion, and volume, in many cases the answer to the question of how much to do
regarding these variables is simply to do as much as possible. In other words,
because training for the ultraendurance sport of triathlon requires proficiency
in three individual disciplines, most triathletes, particularly age-group tri-
athletes, will be relatively time constrained due to nonsport obligations and
will be unable to realize sufficiently high training volumes to elicit optimal
adaptations. Therefore, manipulating frequency and volume will likely result
in suboptimal adaptations, which leaves intensity as the primary training
variable that the triathlete can modulate.
302 } McGregor

Intensity
Possibly the most controversial training variable to consider is intensity. Par-
ticularly in triathlon, debate continues about the optimal intensity and even
the role of intensity with regard to training for the triathlon. If the debate
with regard to training intensity in endurance sport were divided into two
camps, these camps would be the high-intensity training (HIT) camp11, 12, 13, 14
and the high-volume, and hence, low-intensity camp.15, 16

High-Intensity Training (HIT)


Advocates of the HIT approach argue that intense training performed above
the lactate threshold will elicit adaptations that result in improved perfor-
mance that cannot be obtained through standard low-intensity endurance
training, particularly in highly trained athletes.13, 17, 18 Several investigations
have demonstrated dramatic improvements in endurance performance from
HIT in previously untrained or recreationally active individuals,19, 20, 21, 22 but
evidence indicates that such an approach can be beneficial for highly trained
athletes as well.14, 17, 23, 24, 25, 26, 27
Several investigators have used 4- or 5-minute efforts at 80 to 85 percent
of peak power for 4 to 6 weeks and demonstrated improvements in 40-kilo-
meter time-trial cycling performance in highly trained athletes.24, 25, 27 Other
investigations have used shorter efforts (30 to 60 seconds) at higher intensi-
ties (100 to 175 percent of peak power) and demonstrated improvements in
endurance performance as well.23, 25, 27 Although relatively high intensity,
these efforts are performed within the aerobic domain, or slightly above it,
so it is not surprising that improvements in endurance performance would
be observed with such efforts. Iaia and Bangsbo have performed extensive
investigation of training that falls in more of the anaerobic domain, what has
been coined speed endurance training.
The term speed work can impart different connotations, depending on the
context or circumstances. Speed work should be considered in two primary
ways:
• anaerobic training performed with the objective of improving the rate
of anaerobic energy provision or anaerobic capacity or
• neuromuscular training performed at powers corresponding to anaerobic
metabolism but more with the objective of increasing neurological or
biomechanical aspects of movement patterns to improve performance.
Iaia et al.11 have added criterion definitions for several components of
speed work training. First, they define speed training as bouts that consist of
short, maximal efforts (e.g., 2 to 10 seconds) interspersed with relatively long
recovery periods (e.g., 50 to 100 seconds). Speed training is used to develop
maximal-speed generation and neurological training.
Duration, Frequency, and Intensity | 303

Metabolic Responses to Anaerobic Training


Despite the fact that maximal efforts of 30 seconds in duration generally are
considered anaerobic, repeated efforts of this duration or shorter become
increasingly aerobic. Investigators have shown that repeated maximal efforts
of 5 to 30 seconds in duration become primarily aerobic largely because of the
reduced contribution from glycolytic and glycogenolytic metabolism.30, 31, 32, 33 In
fact, after only three maximal efforts of 30 seconds in duration with 4 minutes
of recovery, the total anaerobic contribution during the third 30-second effort
is reduced to less than half and aerobic metabolism accounts for approximately
60 percent of total energy provision, which makes them, by definition, aerobic.34
Aside from the obvious potential benefits to anaerobic performance, several
investigators have reported aerobic adaptations to various types of training
programs that are based on anaerobic efforts.11, 20, 35, 36 So, although anaerobic
training may not seem appropriate for an endurance sport such as triathlon,
regardless of the distance, some foundational evidence supports the potential
benefits of this type of training to aerobic performance. An argument can be
made that anaerobic training may actually be a stimulus for aerobic adaptations,
but it is likely not an optimal stimulus.
Although maximal efforts of short duration. become mostly aerobic, short-
duration efforts are not long enough to elicit VO2max or are too short to elicit
large volumes of accumulated aerobic work. Therefore, anaerobic training may
have a place in the overall training scheme, but it should likely be used judiciously.

Speed endurance training is defined as any other anaerobic training and


is further divided into speed production and speed maintenance training.
In speed production training, the effort lasts less than 40 seconds and is fol-
lowed by relatively long recovery periods, generally five times the work period.
Speed maintenance training consists of efforts lasting 5 to 90 seconds but
has shorter recovery periods of less than three times the work period. The
distinction between speed production and speed maintenance is the relatively
higher intensities that can be maintained during speed production work
because of the proportionally longer recoveries. With speed maintenance
work, intensities will be lower, particularly because fatigue ensues with the
shorter recovery periods.
As previously stated, the purposes of speed training can be targeted either
at neuromuscular learning, with the intention of improving movement pat-
terns that result in greater speed development, or at metabolic processes
that illicit increased ability to generate energy from anaerobic sources faster
(i.e., rate) or in greater amounts (i.e., capacity). This type of training has been
shown to improve performance as well as some underlying physiological
304 } McGregor

parameters that were not previously investigated in relation to endurance


performance.11, 12, 28, 29

High-Volume, Low-Intensity Training


In contrast to the HIT approach to intensity application, an intensity and
volume distribution that is heavily weighted to the lower-intensity spec-
trum is advocated by some authors. In particular, Seiler has argued for an
approximate 80–20 percentage distribution, in which 80 percent of training
is performed below the lactate threshold and 20 percent above.15, 16 He cites
as support for this contention that, anecdotally, this distribution seems to be
the one that many high-level endurance athletes in various disciplines have
come to use through the process of trial and error.
Additionally, he cites evidence from Esteve-Lanao et al.,15 who examined
two groups of subelite distance runners, one that trained with an 80–20 dis-
tribution and the other that trained with a 67–33 distribution. Over a 5-month
period, the 80–20 group improved significantly more than the 67–33 group
in a standardized performance test (−157 plus or minus 13 seconds versus
−122 plus or minus 7 seconds).
Other data in support of the high-volume, low-intensity approach come
from Fiskerstrand and Seiler,37 who examined training data for elite rowers
over the course of 30 years. Evidence showed that over this time, total train-
ing volume had increased, with a 20 percent increase in volume of training
below .the LT but a decrease in volume of training above the LT. At the same
time, VO2max increased by approximately 10 percent and rowing ergometer
performance improved by 10 percent. Note that rowers primarily compete
in the 2,000-meter event, which requires approximately
. 6 to 7 minutes to
complete, a time that corresponds closely to the VO2max,38, 39 yet the majority
of training is performed below the LT.

Training Adaptations Similar Across Event Durations


This finding demonstrates a principle that some astute practitioners have
been aware of for decades—aerobic performance relies on similar training
adaptations regardless of the duration of the event. In particular, Arthur
Lydiard, the legendary coach of numerous Olympic medalists in running
from the 800 meters to the marathon, argued that he trained all his runners
regardless of discipline in much the same way. From a speech in 1990 he said,
Even my middle distance runners like Snell and Davies also did the same
mileage. Because what does the middle distance runner and distance
runner require? They require the high oxygen uptake level as the governing
factor in their performance level. They need speed and they need anaerobic
development. So, this is why I trained them all the same, except for the last
10 weeks when we decided which event they were better suited for and
changed the training to co-ordinate their efforts for that particular event.”40
Duration, Frequency, and Intensity | 305

Training Volume Distribution


To illustrate this point further, in figure 21.1 data are presented from three
athletes with regard to their training volume distribution. In figure 21.1a,
the run training volume distribution for an Ironman World Championships
(IMWC) age-group podium winner is presented. This data were collected for
every run training session by GPS speed distance devices over the course of
4 months leading up the IMWC.
40
36
32
28
24
Percent

20
16
12
8
4
a
0
1 3 5 7 9
55
50
45
40 E5643/Friel/Fig. 21.1a/448685/TimB/R2-kh
35
Percent

30
25
20
15
10
b 5
0
1 3 5 7 9
27
24
21
18 E5643/Friel/Fig. 21.1b/448686/TimB/R2-kh
Percent

15
12
9
6
3
c
0
1 3 5 7 9

„„Figure 21.1 Running pace zone distribution generated from speed and distance
data obtained from global positioning system (GPS) devices in (a) Ironman World
Championships (IMWC) age-group competitor for 4 months before IMWC; (b) subelite
E5643/Friel/Fig. 21.1c/448687/TimB/R2-kh
national-caliber runner for one season; and (c) elite amateur triathlete for one season
306 } McGregor

With regard to the runner presented in figure 21.1a, more than 90 percent
of training volume was performed below the functional threshold pace (FTP),
or maximal lactate steady-state pace (MLSS). Similarly obtained data from
a subelite national-caliber distance runner are presented in figure 21.1b. In
this case, the runner exhibits an extreme distribution in which more than 95
percent of training was below the FTP–MLSS.
Finally, in figure 21.1c, data from an elite amateur triathlete show a distri-
bution pattern different from the other two. In this case, 81 percent of run
volume was performed below the FTP−MLSS, but apparently this athlete
performed more training closer to the FTP−MLSS than the other two athletes
did. In comparison with the other two athletes, this athlete was the only one
to improve run FT pace over the course of the training year.

Pace Zone Distribution


The data are divided into a pace zone distribution that this author has pre-
sented previously,41 in which the FTP, which is analogous to the MLSS, occurs
in zone 6. The pace zone determination approach used for these runners is
presented in figure 21.2, in which the FTP is determined through testing or
based on extrapolation from recent performances such as the 3K, 5K, or 10K,
and then the other zones are
. centered on associated physiological responses
or adaptations (e.g., Z8 = VO2max, Z10 = speed). 41

Recent race (Min: Sec) Target training pace (Min: Sec/Mile)


3K 5K 10K PZl 2: L Aero 3: M Aero 4: H Aero 6: Thresh 8: VO2max 10: Speed
18:10-17:41 33:55-32:01 66:04-66:21 50 14:41-13:21 13:20-12:15 12:14-11:37 11:01-10:41 10:08-9:55 9:38-2:40
17:40-17:13 32:00-31:05 66:20-64:37 49 14:22-13:02 13:01-11:57 11:55-11:19 10:44-10:24 9:51-9:41 9:22-2:40

10:21-10:13 18:30-18:13 33:23-37:49 23 9:10-8:00 7:59-7:11 7:10-6:43 6:26-6:15 5:50-5:44 5:26-2:40


10:12-10:03 18:12-17:57 37:48-37:14 22 9:03-7:53 7:52-7:04 7:03-6:37 6:21-6:10 5:44-5:38 5:20-2:40
10:02-9:54 17:56-17:41 37:13-36:41 21 8:56-7:46 7:45-6:58 6:57-6:31 6:15-6:04 5:40-5:34 5:16-2:40

„„Figure 21.2 Pace zone index (PZI) chart extract for pace zone determination in
E5643/Friel/Fig.
running. The PZI can be extrapolated 21.2/448688/TimB/R3-kh
from performances in races such as the 5K. In this
case, a 5K performance of 18:00 would correspond to a PZI of 22. The corresponding
pace zones to the right of that value would then be applied. The FTP would lie within
zone 6: threshold.
Data from McGregor, S.J, and Fitzgerald, M. 2010. The Runner’s Edge. Champaign, IL: Human Kinetics.

Cycling Power Distribution


In figure 21.3 data collected from portable power meters in cycling are
presented. In figure 21.3a, an example comes from an elite espoir (i.e., U23)
bicycle racer for an entire season in which he placed in the top five in his
national professional time-trial championship and competed in the espoir
world championships. Data are collected from an onboard power meter for
all training sessions and most races and are presented in a power distribution
scheme developed by Allen and Coggan42 in which the functional threshold
power (FTP) occurs in the zone labeled TH.
60

50

Percent of total training time


40

30

20

10

a 0
AR E TE TH VO2 AC

60
E5643/Friel/Fig. 21.3a/448689/TimB/R3-kh

50
Percent of total training time

40

30

20

10

b 0
AR E TE TH VO2 AC

60
E5643/Friel/Fig. 21.3b/448690/TimB/R3-kh

50
Percent of total training time

40

30

20

10

c 0
AR E TE TH VO2 AC

„„Figure 21.3 E5643/Friel/Fig.


Cycling power21.3c/448691/TimB/R3-kh
zone distribution generated from portable
onboard power meter data obtained for (a) an elite espoir professional cyclist
for one season; (b) an elite amateur national-caliber cyclist for one season;
and (c) an elite amateur age-group national champion cyclist for one season.

307
308 } McGregor

As with the FTP for running, the FTP in cycling corresponds approxi-
mately to the maximal lactate steady state effort (MLSS), but in this case is
power based, as opposed to pace based for running. With regard to training
structure, despite the inclusion of high-intensity efforts produced during
elite-level road racing included in the dataset, 87 percent of total training and
racing volume is performed below the FTP−MLSS.
In figure 21.3b data are presented from an elite amateur cyclist who placed
fourth in his national elite time trial. In this case 88 percent of total training
volume, including races and training, was performed below the FTP−MLSS. In
the case of both of these athletes, performance at the FTP improved approxi-
mately 5 percent over the course of the season.
Finally, in figure 21.3c power meter data are presented from a masters-
level national champion cyclist who won several elite category time trials.
Data include all racing and training sessions, and although the total training
volume is approximately two-thirds of the other elite-level cyclists, over the
course of the season, 88 percent of all training was performed below the FTP.
Note that despite the apparent overwhelming weight of sub-FTP training
distribution, these athletes either raced or performed supra-FTP efforts on a
frequent, almost weekly, basis. So, HIT training was a component of the train-
ing program, but a relatively small component. These anecdotal observations
coming from quantitative data collection methods in runners, triathletes,
and cyclists support the notion that a large majority of training is performed
below the FTP or MLSS, regardless of discipline.
Further, the difference among individuals in the distribution pattern of
volume in the various training zones presented in figures 21.1 and 21.3 is
somewhat striking, yet the relative proportion of training below FT in gen-
eral terms is essentially the same among individuals. This begs the question,
How important are differences in intensity between subthreshold exercise
intensities for optimal development? The data from triathletes and runners,
though, might indicate that a 90 percent distribution is the upper limit for
sub-FTP training because improvements were not observed in athletes who
trained with this distribution. With the proliferation of electronic training
devices (e.g., portable power meters and GPS), more information in this regard
will likely be accumulated in the future and shed more light on this issue.

Conclusion
The basic training variables frequency, duration, volume, and intensity, which
contribute to the principle of overload, are foundational to training for tri-
athlon, yet there is no clear evidence about the optimal application of these
variables to elicit the maximal training effect and performance. Nevertheless,
several general guidelines regarding these variables can be offered:
Duration, Frequency, and Intensity | 309

• In simple terms, because most triathletes will be time constrained, the


frequency and volume of training should be as high as practical.
• A training intensity distribution consisting of approximately 80 to 90
percent of total training performed below the FTP–MLSS is advisable.
• Within the confines of this distribution, HIT training is likely required
as a component of the overall training program to address the 10 to 20
percent of training that should be performed above the FTP–MLSS.
• Intensive training should probably be more aerobic in nature as opposed
to anaerobic to assist with the overall aerobic development of the triath-
lete.
This page intentionally left blank.
CHAPTER
22
Periodization
—Stephen J. McGregor, PhD

M ost triathletes anticipate and accept, even relish, that they need to
train a great deal to be successful in their sport. But simply training
a lot with no particular plan or objective is unlikely to lead to success.
A key element to success in triathlon, or most endurance sports for that
matter, is an organized approach to training that allows the athlete to
gain fitness and ultimately optimize performance for target events of the
season without becoming overstressed. So, at the start of each season, the
triathlete is faced with the daunting task of organizing the yearly training
plan in a way that will result in the best possible performance at the target
events of the season. The approach that most triathletes typically take is
called periodization.
In general terms, periodization is the systematic planning of training
over a relatively long time frame (e.g., year, several years, and so on), in
which the larger time scale is subdivided into smaller periods or cycles.
A pictorial example of one of the earliest documented approaches to peri-
odization can be seen in figure 22.1. In this example, the training year
is divided into small periods consisting of preparatory, competitions, and
transition phases, which are further divided into general preparatory, specific
preparatory, precompetition, main competition, and transition phases. As we
will see later in this chapter, this traditional approach to periodization is
made up of one monocycle and gross manipulations of volume and inten-
sity designed to elicit optimal performance at one target competition late
in the year during the competition phase. This periodization approach has
been modified extensively since its original introduction in 1965, but many
coaches and triathletes still use the basic elements.

311
312 } McGregor

100%

General prep. Specific prep. Precomp. Main comp. Trans.

Preparation Competitions Trans.

„„Figure 22.1 Original periodization model presented by Matveyev. The main


competition year is subdivided into several phases of training. The solid line represents
the volume of training, and the dotted line represents the intensity of training.
From L. Matveyev, 1965, Periodization of sports training (Moscow: Fizkultura i Sport).
E5643/Friel/Fig 22.1/448692/alw/pulled-R2-kh

Rationale for Periodization


Much of the rationale for periodization theory is based on the work of Hans
Selye and his model of General Adaptation Syndrome (GAS)1, 2, 3 (see figure
22.2). From this model, it has been proposed that any stress on the system can
result in an adaptation whereby the body becomes more fit, or resistant to this

Training
stimulus
Level of performance

Original C
performance A New level of
level performance

Resistance or
Alarm adaptation phase
phase
Overtraining or
B exhaustion

Time

„„Figure 22.2 Illustration of Selye’s general adaptation syndrome (GAS) theory.


A = typical training thatE5643/Friel/Fig. 22.2/448693/TimB/R1
results in positive adaptation; B = overtraining that results in
detrimental adaptation; C = overreaching or supercompensation that results in superior
adaptation to that from typical training.
Reprinted, by permission, from T.O. Bompa and G.G. Haff, 2009, Periodization: Theory and methodology of training,
5th ed. (Champaign, IL: Human Kinetics), 14. Adapted, by permission, from A.C. Fry, 1998, The role of training
intensity in resistance exercise overtraining and overreaching. In Overtraining in sport, edited by R.B. Kreider, A.C.
Fry, and M.L. O’Toole (Champaign, IL: Human Kinetics), 114.
Periodization | 313

stress. In general, this fitness is simply a resistance or tolerance to stress—any


stress to which the system is exposed, not just training stress.
An important point is that a recovery period is required for the adaptation
to the stress and increased fitness to occur. People often say that the adapta-
tion to training stress does not occur during the training session itself but
during the subsequent rest or recovery period. In a larger context, this same
principle applies in that if a training stress is imposed in a continuous fashion,
the system may adapt, but if the stress is not removed, the system will become
overstressed and break down.
Therefore, the periodization approach allows for the application of stress
and recovery in a systematic fashion to optimize the adaptation and minimize,
or avoid, the potential for overstressing the system and causing a breakdown
(i.e., overtraining or underperformance).
Specifically with regard to training, the rationale for periodization is several
fold. First, it is generally accepted that periods of training stress followed by
recovery result in a supercompensation that produces improved performance.1,
4, 5
The degree of supercompensation resulting from training depends largely
on the magnitude of the overload imposed;6 hence, periodization allows for
progressive increases in training load, interspersed by recovery. Second, peri-
odization is a systematic approach to this pattern of overload and recovery that
allows for planning of training, adaptation, and competition. Further, concep-
tually, periodization facilitates the visualization of short-term training objec-
tives and benchmarks within the context of the longer-term goals that are the
primary objectives of the season. Therefore, the periodization plan serves as a
temporal guide for the application of training loads in a sequential fashion such
that optimal performance may be achieved for the primary target events.1, 3, 4, 7,
8, 9, 10, 11
Finally, another purpose of periodization may simply be to relieve train-
ing monotony, which in turn may result in greater physiological adaptations.1

Periodization Cycles
In periodization theory, cycles are the subdivisions by which the larger
timeframes are broken down. An unfortunate problem in the periodization
literature is the lack of standardization, and therefore some discrepancies in
this nomenclature exist. For instance, in some cases, the yearly or seasonal
plan would be referred to as the macrocycle, the 8-week increments would
be mesocycles, and the 7-day increments would be microcycles.4, 11, 12, 13 Others
might consider a macrocycle one of several divisions of the yearly or seasonal
plan, and they would not use mesocycles.1, 14 For purposes of consistency, in
this text we will use the macrocycle to describe the largest timeframe for
a given training objective, typically a year or season. Mesocycles are the
subdivisions of the macrocycle that typically focus on a particular attribute
of overall training objective (e.g., base, specialization, competition, peak). In
this context, an attribute is any skill or characteristic that can be affected by
training. As such, the term can be broadly used to describe physiological
314 } McGregor
.
characteristics (e.g., anaerobic capacity or VO2max) or technical skills and
abilities (e.g., swimming technique or running economy). Microcycles are
the smallest division of organized training apart from the actual training
sessions themselves.

Traditional Periodization in Triathlon


Probably the most prolific application of periodization in triathlon comes from
the popular text by Friel.7 In this work, Friel presents a linear model of tradi-
tional periodization (figure 22.3). Using this approach, if each mesocycle is
condensed by approximately half, the periodization scheme could be applied
for a period as short as 20 to 22 weeks, but as presented it would typically
be applied for a year. An in-depth analysis of the suggested components of
each mesocycle and microcycle is beyond the scope of this chapter, but the
reader is encouraged to seek out Friel7 as a reference and an exemplar of linear
periodization approach that is widely used in triathlon and other endurance
sports.1, 7, 15 In the text, Friel provides this template as a generalized approach
that is founded on the commonly accepted traditional model of periodiza-
tion for endurance sports, but he acknowledges that other approaches may
be effective as well. Possible alternative approaches are presented later in
this chapter.
Other examples of traditional periodization applied to triathlon found in
the literature include Rowbottom et al.,5 in which the authors examined highly
trained triathletes over the course of 9 months. During this period, the tri-
athletes doubled their training load. This approach of continually increasing
the training load over the course of 9 months consisting of a 5-month prepa-
ratory mesocycle (July to November) and a 4-month competition mesocycle
(December to March) would be considered a traditional linear periodization
approach.5, 16

Macrocycle Training year

Preparation Competition Transition

Specific Pre-
Mesocycles General preparation Competition Transition
preparation comp

Prep Base Build Peak Race Transition

Microcycles 1 2 3 4 5 6 7 8 Weeks 9-42 43 44 45 46 47 48 49 50 51 52

„„Figure 22.3 A traditional periodization plan as presented by Friel that is composed


of one macrocycle. The single macrocycle22.3/448694/TimB/R1
E5643/Friel/Fig. is divided into several mesocycles, which are
further divided into microcycles. This approach could be used for a yearly monocycle
approach or for a multicycle approach in which two or more of these macrocycles
would be planned within a calendar year.
Reprinted, by permission, from J. Friel, 2009, The triathlete’s training bible, 3rd ed. (Boulder, CO: VeloPress), 38.
Periodization | 315

Note that 6-week microcycles were used with a 1-week regeneration period
before testing and beginning the next microcycle. Numerous physiological
parameters were examined, but the only significant change occurred in run
speed at anaerobic threshold, which increased from 15.6 to 16.6 kilometers
per hour (approximately a 6 percent increase) from the start to the finish of
the 9-month period.
Several arguments have been raised against the traditional periodization
concept.4, 17, 18, 19 In particular, competing frequently at a consistently high
level, as elite and professional athletes do, may conflict with the periodization
scheme.20 Another problem rests in coordinating the periodization of three
disciplines that may have conflicting training approaches (e.g., LSD for run-
ning versus intervals for swimming).21, 22, 23, 24, 25 So, for triathletes who must
develop proficiency in three disciplines with disparate approaches, a singular
overarching traditional periodization scheme may be problematic to develop.

Alternative Periodization Approaches


In response to the aforementioned criticisms of the traditional periodization
approach, alternative methodologies have been developed. An alternative
termed block periodization is primarily advocated by Issurin.4, 18, 26 Another is
undulating, or nonlinear, periodization.

Block Periodization
Block periodization is an approach whereby the annual cycle is divided into
three mesocycles that are cyclically repeated in a fashion that depends on
the nature of the competitive season. The three mesocycles (accumulation,
transmutation, and realization) are joined together in a sequential fashion
much like in traditional periodization, but it is argued that during these
mesocycles, the triathlete can focus on a more limited number of attributes,
or abilities, than in the traditional mixed approach. Therefore, the triathlete
can achieve intensified training and development of these attributes using
the block periodization approach.
During the accumulation mesocycle, volume is typically higher and inten-
sity is low, as is typical with the early mesocycles of a traditional periodization
approach. This mesocycle is said to last only 2 to 6 weeks, though, before the
triathlete moves into the more intense transmutation mesocycle. During this
mesocycle, intensity is high. This period is considered the most stressful, but
it lasts only 2 to 4 weeks, depending on the overall structure of the annual
plan. The final realization phase consists of drills to prepare specifically for
competition using a sport-specific program, which also allows for recovery
from the fatigue of the preceding mesocycle.4 One main advantage to this
approach as proposed by Issurin can be seen in figure 22.4, in which peri-
odization is structured around several competitions throughout the year,
culminating with the targeted primary objective at the end of the annual
316 } McGregor

plan.4 This plan shows how the pattern of the three mesocycles is repeated
sequentially, enabling relative rest, a high fitness level for competitions, and
repeated intensified training in the interim.
A potential limitation to the block periodization approach, which is identi-
fied by Issurin himself, lies in the fact that during each mesocycle, particular
attributes are targeted while nontargeted attributes may be ignored. For
example, the triathlete who focuses on anaerobic capacity during a meso-
cycle may ignore aerobic endurance. So, the issue of detraining is of concern
if a previously targeted attribute is ignored and fitness of that nontargeted
attribute is lost in a subsequent mesocycle. Therefore, Issurin argues that the
coach must be aware of the duration of the training residuals of each attribute
so as not to leave attributes with short-lived training residuals untargeted
too long. Training residuals are either the delayed adaptations to training or
persistent adaptations, which can remain for various periods after a train-
ing overload has been removed. For example, Issurin states that both aerobic
endurance and maximal strength adaptations persist for approximately 30
days, whereas anaerobic glycolytic endurance will dissipate after only 18 days
and alactic speed will last approximately 5 days.18

Targeted
mesocycles mesocycles reference points

event
Accumulation Transmutation Realization Competitions,

5
4
3
2
1
mesocycles

I II III IV V VI
Stages

Preparation period Competition period

„„Figure 22.4 Presentation of a block periodization annual plan with competitions


placed at the end of each block and with 22.4/448695/TimB/R1
E5643/Friel/Fig. the plan culminating with the sixth primary
targeted competition.
Reprinted, by permission, from V.B. Issurin, 2010, “New horizons for the methodology and physiology of training
periodization,” Sports Medicine 40(3): 189–206.
Periodization | 317

At the same time, Issurin argues that the training residuals are a central
aspect of the block periodization approach, in that the relatively short meso-
cycle approach relies on training residuals to maintain fitness of particular
attributes over the course of the rapid fluctuations in training load and
focus. This approach is claimed to have been responsible for the success of
the former USSR canoe–kayak team (6 Olympic and 17 world championship
medals from 1988 to 1990) as well as Alexander Popov, multiple Olympic and
world champion swimmer.4, 27

Undulating, or Nonlinear, Periodization


Another periodization approach that has been developed in response to per-
ceived problems with the traditional approach is termed nonlinear periodization.
In the traditional approach, each period is designed for a specific purpose
(e.g., preparation, base, build, and so on) that should be followed in a sequen-
tial fashion, but the nonlinear approach does not adhere to these strictures.
Few studies address this approach as it applies to endurance sport, and
fewer still are specifically related to triathlon, but in the field of strength
training, numerous peer-reviewed studies have been done.28, 29, 30, 31, 32, 33, 34, 35
Although it may seem counterintuitive to extrapolate the validity of strength
training studies to an endurance sport such as triathlon, performance mod-
eling studies have shown that global training adaptations can be modeled
similarly in sports as disparate as the hammer throw and triathlon based on
quantification of training loads.36, 37 Therefore, until more data are available
specifically regarding triathlon, the aforementioned strength training studies
will have to suffice.
Anecdotally, in terms of endurance sport, the nonlinear approach contrasts
traditional periodization by not adhering to a more segregated distribution
of training loads whereby the early base mesocycles consist of low intensity
and high volume and progress to specialized mesocycles consisting of lower
volume and high intensity. Instead, nonlinear periodization consists of a
more balanced application of intensities throughout the entire macrocycle.
The various training modalities (e.g., LSD, aerobic intervals, anaerobic train-
ing) are applied more consistently throughout the entire macrocycle and are
periodized over the course of days as opposed to segregated by mesocycles.
A primary argument in favor of this approach is along the lines of the block
periodization rationale—that more frequent competitions spread throughout
the year require a more consistent, higher fitness level over the course of the
macrocycle. Numerous practitioners have likely been applying this approach
over the years not as the application of a grand theoretical design, but as a
matter of necessity, because competition schedules have dictated it.
In fact, a popular training manual for cycling, written by Greg LeMond
more than two decades ago,38 recommended such an approach, as espoused
by Cyrille Guimard and Paul Koechli, to accommodate the greater frequency
of cycling competitions relative to other endurance sports.18 Further, Haff
318 } McGregor

has argued that the nonlinear approach is synonymous with the traditional
model in that both approaches use day-to-day variations in training load to
optimize adaptation and avoid overtraining.17 So, the hallmark of nonlinear
periodization that distinguishes it from a traditional approach with regard
to endurance sports is simply the lack of a sequential series of mesocycles
that progress from low intensity and high volume to high intensity and low
volume, and a more consistent distribution of training intensities throughout
the year or macrocycle.

Evaluating Periodization
Through Performance Modeling
When devising the overall periodization plan, it is difficult for the triathlete
to objectively see the changes and adaptations that may be occurring on
multiple time scales. So, how can the triathlete or coach see the forest for the
trees, if you will, when evaluating the periodization plan?
One technique that has shown some promise in simplifying the complexi-
ties of the overall periodized program is mathematical performance modeling.
Modeling the effects of training on performance has been of interest in the
scientific literature since Calvert et al.39 first proposed an impulse–response
(IR) modeling approach more than three decades ago. They devised their
approach based on the ideas that an acute training bout (impulse) elicited
two antagonistic responses: an initial negative component (i.e., fatigue) that
detracted from performance and a delayed positive component (i.e., fitness)
that ultimately contributed to improved performance after dissipation of the
negative component.
These concepts of reduced performance because of incurring fatigue
through training and increased performance with reestablishment of homeo-
stasis and subsequent supercompensation are essentially synonymous with
those of periodization. Therefore, the performance modeling approach devel-
oped by Calvert et al.39 (see figure 22.5) could be thought of as mathematical
approach to modeling the result of periodization.
Now understandably, for nonmathematicians such an equation will
likely be intimidating, but in simple terms, the concept is straightforward.

t–1 t–1
p1 = p0 + ka 
s=0
e–(t – s)/aws – kf 
s=0
e–(t – s)/fws

Initial Long-term Short-term


performance accumulated accumulated
level training = training =
fitness fatigue

„„Figure 22.5 Performance modeling approach.


E5643/Friel/Fig. 22.5/448696/TimB/R1
Periodization | 319

A performance at any given time in the future (pt) is determined by the initial
performance level at day 0, the positive effects of training that can be thought
of as fitness, and the negative effects of training that can be thought of as
fatigue. So, the effects of training can be mathematically reduced to fitness
and fatigue components that essentially dictate performance at a given time.
In their initial work, Calvert et al. quantified training using a heart rate metric
termed TRIMPS, or training impulse.39 Again, though, for practitioners such
as coaches and triathletes, this mathematical approach may be too esoteric
and complex, so a simplified approach has been developed specifically for
the cycling and running subdisciplines of triathlon.
In essence, all derivations of the mathematical model can be simplified to
Performance = fitness − fatigue
Therefore, Coggan devised an approach that eliminates much of the com-
plex math and can be performed using simple spreadsheet calculations40, 41
(better yet, this approach has been incorporated into user-friendly consumer
software packages). By removing several of the components of the original
model (e.g., ka and kf), Coggan argued that performance on a given day depends
on the training stress balance (TSB), which is determined by an interaction
between the positive chronic training load (CTL) and the negative acute
training load (ATL). The CTL is essentially synonymous with fitness and the
ATL with fatigue. Therefore, performance on a given day is proportional to
TSB = CTL − ATL
In other words, a triathlete’s TSB is equal to the chronic training that he or
she has been exposed to over a long period of time minus the acute training
that he or she has undergone in recent days. Conceptually, every triathlete
should intuitively understand this idea; the purpose of training is to accumu-
late fitness by incurring a chronic training load over a long period (months
or years). At the same time, training imparts some negative consequences in
the near term that result in fatigue and reduced performance. The balance
of these two factors dictates performance, and managing them appropriately
results in optimal performance.

Quantifying Daily Training Loads


Again, the trick to periodization for the triathlete is being able to visualize
and manage training loads over multiple time scales. Fitness will accumu-
late over months and, in terms of residuals, may persist for approximately
a month or more. Fatigue, on the other hand, will accumulate immediately
and typically dissipate within a matter of days to a week.
To determine the chronic and acute training loads, Coggan devised a
training stress score (TSS) to quantify the daily training load using cycling
power meters in terms of volume and intensity. A benchmark TSS score was
320 } McGregor

determined to be 100 based on a maximal effort that was performed for 1


hour, which corresponded to the functional threshold (FT).
The CTL and ATL are determined from daily TSS based on exponentially
weighted rolling averages of 42 days for CTL and 7 days for ATL. The 42- and
7-day periods reflect the time scales on which each respective training load
dissipates. The practical importance of these time scales and their use will
be demonstrated in reference to an elite runner later in this chapter, but the
general concept of the relationship among TSB, CTL, and ATL can be seen in
figure 22.6, which presents performance modeling of an age-group triathlete.
TSB TSSM
20 62
18 TSB = 19 58.9
16 55.8
14 52.7
12 49.6
10 46.5
8 43.4
6 40.3
4 37.2
2 34.1
0 CTL= 35 31
–2 27.9
–4 24.8
–6 21.7
–8 18.6
–10 ATL= 16 15.5
–12 12.4
–14 9.3
–16 6.2
–18 TSB (TSB) ATL (TSSM) CTL (TSSM) 3.1
–20 0
12/25 1/8 1/22 2/12 2/26 3/5 3/19 4/2 4/16 4/30

TSB TSSM
20 62
18 TSB = 19 ATL= 61 58.9
16 55.8
14 52.7
12 49.6
10 46.5
8 43.4
6 40.3
4 CTL= 46 37.2
2 34.1
0 31
–2 27.9
–4 24.8
–6 21.7
–8 18.6
–10 ATL= 16 15.5
–12 12.4
–14 9.3
–16 TSB = –15 6.2
–18 TSB (TSB) ATL (TSSM) CTL (TSSM) 3.1
–20 0
12/25 1/8 1/22 2/12 2/26 3/5 3/19 4/2 4/16 4/30

„„Figure 22.6 Performance modeling chart for an age-group triathlete


over the first 4 months of a season for running. Top: positive TSB of 19 results
E5643/Friel/Fig. 22.6/448697/TimB/R3-kh
from a CTL of 35 and ATL of 16. Bottom: negative TSB of −15 results from a
CTL of 46 and ATL of 61.
Periodization | 321

We can see changes in TSB, CTL, and ATL over the course of the early
season. In January (top), a positive TSB is shown, whereas in March (bottom),
a negative TSB is shown. This relationship also demonstrates a concept that
may not be otherwise readily apparent—when fitness (CTL) is relatively stable,
as in this case, changes in performance (TSB) are inversely related to changes
in fatigue (ATL). In simple terms, then, when fatigue is high, performance is
low, and when fatigue is low, performance is high.

Determining the Optimal Level of Fitness


Based on Training Load
Coggan’s approach to performance modeling was devised for cycling, but as
previously mentioned, McGregor et al.20, 42 (figure 22.7) used a derivative of the
approach and analyzed an Olympic middle-distance runner’s training and
performances over a 7-year span. The authors found that this performance
modeling approach provided valuable insight regarding the overall peri-
odization structure that could be of value to the coach or athlete who used it
on a regular basis. In particular, the optimal level of fitness (CTL) based on
training load that elicited best performances was determined.

70 1000

60 900
50 800
40
700
30
600 Mercier score
TSS/d

20
500
10
400
0
300
–10

–20 200

–30 100

–40 0
1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209
217
225
233
241
249
257
265
273
281
289
297
305
313
321
329
337
345
353
361

„„Figure 22.7 Plot of impulse response (IR) model parameters versus performance
E5643/Friel/Fig. 22.7/448698/TimB/R1
for 2000 Olympic year. Top line = chronic training load (CTL); bottom line = model
response output (TSB); triangles = Mercier score (MS). The shaded area highlights
the range of CTL values for which the top 10 percent of performances occurred for
this athlete during the 7 years of study. The square indicates when the personal best
performances for this athlete occurred in 2000, and the circle indicates performances
during the Olympics.
Adapted, by permission, from S.J. McGregor, R.K. Weese, and I.K. Ratz, 2009, “Performance modeling in an
Olympic 1500-m finalist: A practical approach,” Journal of Strength and Conditioning Research 23(9): 2515-2523.
322 } McGregor

McGregor et al. converted performances in the 800-meter through 3,000-


meter events to Mercier scores, which is a system for comparison of different
athletic events that are all based on world record performances and given a
maximum value of approximately 1,020:
Briefly, they are the end-result of a linear fit to the weighted average of
the 5th, 10th, 20th, 50th, and 100th world-ranked performances in each
event over the past 4 years. The performances from more recent years are
given a higher weighting, which can tend to skew the comparisons if one
of the events had a weak year.43
This approach allowed the authors to use different performances from 800
meters through 3,000 meters for performance evaluation.20 They reported a
parabolic relationship between fitness (CTL) and performances expressed as
Mercier scores whereby best performances were observed when CTL was
relatively high, but as CTL increased, at a certain point (approximately 55),
performance declined (figure 22.8).20
Taken together, these data showed that this middle-distance runner needed
to have a CTL of at least 50 TSS/d for optimal performance. At the same time
a CTL of more than 55 TSS/d caused performance to decline. But as the athlete
tapered over the course of the season, as can be seen in figure 22.7, the CTL
dropped below 50 TSS/d and performances declined relative to earlier in the
season when CTL was higher.

CTL optimal =
optimal performance

1000.00

900.00
Mercier score

800.00

CTL low = unfit = CTL high = fatigued =


700.00
performance poor performance
decline

600.00
0.00 20.00 40.00 60.00
CTL (AU)

„„Figure 22.8 Quadratic relationship between Mercier score (MS) and chronic training
load (CTL). A quadratic relationship was identified when model parameter CTL values
E5643/Friel/Fig. 22.8/448699/TimB/R2-kh
for the day of competition performances (n = 141) were fit to MS calculated for the
performance (P = 0.001).
Adapted, by permission, from S.J. McGregor, R.K. Weese, and I.K. Ratz, 2009, “Performance modeling in an
Olympic 1500-m finalist: A practical approach,” Journal of Strength and Conditioning Research 23(9): 2515-2523.
Periodization | 323

Even if a statistical analysis is not performed, by following the curve


(figure 22.7) it should be evident that CTL (fitness) is declining rapidly at the
end of the season and performances will soon decline. Therefore, graphical
presentation of these modeling data can provide the triathlete with the big
picture perspective of training adaptations and the effects on performance.42
Aside from CTL values, triathletes and coaches also need to consider the
TSB, which is arguably of most importance to target events because the TSB
is the output function of the modeling equation, which should be directly
related to performance. Coggan has presented data indicating that personal
best performances for efforts longer than 5 minutes in cycling are obtained
when TSB is slightly positive. As with CTL, however, above a certain level of
positive TSB, performances decline as well (figure 22.9).40

25

20

15
Percent

10

0
0

–5

10

15

20

25

30

30
–3

–2

–1

–1

to

to
to

to

to

to

to

to

>
<

to

to

to

–5

0
0

10

15

20

25
5

–1
–2

–2

–1

„„Figure 22.9 Training stress balances (TSBs) for personal best (PB) performances of
cycling efforts of a duration of 5 minutes
E5643/Friel/Fig. or longer. The highest frequencies of PBs for
22.9/448700/TimB/R1
these performances occurred with a TSB between 5 and 10; above 15, performances
decline substantially.
Reprinted, by permission, from H. Allen and A. Coggan, 2010, Training and racing with a power meter, 2nd ed. (Boulder,
CO: VeloPress), 158.

Balancing Performance With Fitness:


How Much Training Is Too Much?
Balancing a positive TSB (performance) with a high level of CTL (fitness) is
a difficult task. It is self-evident that low fitness (low CTL) is not conducive
to optimal performance, but triathletes and coaches often ask two questions:
How much training is enough, and how much is too much?
324 } McGregor

Table 22.1 Chronic Training Load (CTL) Guidelines for Triathletes in Cycling and
Running Based on Experience Level and Distance Focus
Combined TSS/d
Triathlete level Cycling TSS/d Run TSS/d (cycle and run)
Novice 30 to 50 10 to 15 40 to 65
Experienced ager 50 to 70 20 to 30 70 to 100
Elite short course 50 to 70 20 to 30 70 to 100
Elite long course 70 to 90 30 to 50 100 to 140

Anecdotally, some guidelines for CTL values for individual disciplines in


triathlon are provided in table 22.1. Note that a novice triathlete who targets
an arbitrarily high CTL may be misplacing her priorities. Particularly in run-
ning for those without a running background, seeking a high CTL will likely
lead to injury. The principles presented in chapter 21 regarding progressive
overload should be observed as the triathlete attempts to increase fitness.
Along those lines, anecdotal observations are that a CTL increase of 5 TSS/d
per week in individual disciplines is reasonable, but in triathlon the combined
stress of a more rapid increase in CTL can lead to injury or underperformance.
It may be tempting to express CTL as a singular combined value for both
disciplines or graphically on one chart, but CTL for individual disciplines are
not additive. For example, neither a cyclist who exhibits a CTL of 120 TSS/d
in cycling alone nor a runner who exhibits a CTL of 100 TSS/d in running
alone will perform on par in triathlon with an elite triathlete who exhibits a
CTL of 70 TSS/d in cycling and 40 TSS/d in running despite the comparable
combined CTLs. Similarly, the triathlete would not compete on par with either
of the individual sport specialists in their respective sports. The combined
CTLs are not additive across sports.
That being said, training load likely has some upper limit that can be
tolerated, and this limit may be reflected in the combined CTL value, so it
warrants attention in that respect. On the other hand, tracking performance
modeling for the individual disciplines separately can show the triathlete
how fitness and performance in the individual discipline is affected by the
overall periodization plan.
On a related note, the modeling of Fitz-Clark et al.44 supports the most
commonly used approach to periodization in endurance sport, that being
a traditional linear approach for optimal performance at one target event.
Therefore, some of the alternative approaches (e.g., block periodization or
nonlinear periodization) may be necessary for athletes who compete in fre-
quent competitions of equal relative importance or where a larger number
of consistent results may be more desirable than one high-profile result at a
given point in time (e.g., world championships). On the other hand, a more
traditional linear approach may be preferable if one primary performance is
of utmost importance.
Periodization | 325

Conclusion
Periodization is the most common approach to overall training plan design in
endurance sports such as triathlon. The literature does not provide evidence
of a clearly superior way to periodize the overall training program, but the
traditional linear, monocyclical schema is likely the approach most often
used. Because the traditional linear approach has been so widely used and
was advocated by such successful practitioners as Lydiard21 before formal-
ized scientific study on the topic had been pursued, it may be close to an
optimal approach.
In particular, for a sport such as long-course triathlon that depends heavily
on aerobic endurance of the athlete, emphasizing this development over other
factors such as anaerobic capacity, strength, or short-term maximal power
production is likely important. This pattern of development lends itself to
a traditional linear approach. On the other hand, short-course athletes who
may need to compete more frequently and may require more speed to be
competitive may benefit from using an alternative approach such as block
periodization or nonlinear periodization.
Performance modeling may be used to assist the triathlete in seeing the big
picture regarding the overall periodization plan. This information supports
the notion that optimal performance comes from the triathlete possessing a
high CTL and a slightly positive TSB, which is synonymous with high fitness
and good performance at target events. These characteristics would be evident
from a heavy training load and an appropriate taper before target competi-
tions, and performance modeling could be used to envision this approach
in an objective fashion. Finally, performance modeling can be used to assess
the effects of the overall periodization plan on the individual disciplines.
This page intentionally left blank.
CHAPTER
23
Tapering and Peaking
for Races
—Stephen J. McGregor, PhD

T he taper is an intentional reduction in the athlete’s training load in the


final days to weeks before an objective competition with the aim of opti-
mizing performance at targeted competitions.1, 2 Further, Mujika3 has noted
the distinction made previously by Houmard4 that a taper is not simply a
reduction in training load, but a progressive reduction in load over time,
culminating with the objective competition.
A further clarification should be added that the taper typically follows an
extended period of fitness building by the application of large training loads
and that the taper will allow the dissipation of accumulated fatigue while a
state of supercompensated fitness is present. This combination will promote
optimal performance within a relatively narrow time window. This window
of optimal performance is typically referred to as the peak. Therefore, the
taper is the reduction in training load that will lead to a peak in fitness, but
the two terms are often used interchangeably. From the aforementioned
points we can derive several criteria that distinguish a taper from a simple
reduction in training:
1. An intentional, planned reduction in training load
2. A progressive reduction as opposed to a constant reduction in training
load
3. A period of reduced training load that follows an extended period of
persistently heavy training to maximize fitness
4. A reduction in training load that is finite in duration and relatively short
term
Each of these criteria is specifically addressed in detail in this chapter so
that recommendations for taper optimization in triathlon can be presented.
For more than 25 years, numerous studies have investigated the physio-
logical and performance effects of reduced training and the taper.1, 3, 5, 6, 7

327
328 } McGregor

From this data, it has been determined that a reasonable expected perfor-
mance gain from a taper would be approximately 2 to 3 percent, with a range
from 0.5 to 6 percent being possible.3, 7 At the highest level of sport, Mujika et
al. have shown that performance improvements on the order of 2 to 3 percent
in swimming were greater than the difference between the gold medal and
fourth place in the 2000 Sydney Olympics.8
For elite athletes, the taper can mean the difference between winning
outright versus not medaling at international competitions. Do not expect
performance improvements greater than 8 percent in any discipline because in
the studies performed in endurance sports (not including time to exhaustion
or incremental protocols), the greatest reported performance improvements
are about 6 to 8 percent across various disciplines.5, 9, 10, 11 Therefore, although
the performance improvements elicited by the taper might appear modest,
in highly trained athletes they could have significant effect at the elite level
of competition.

Elements of the Prerace Taper


Because the precompetition taper is a programmed reduction in overall train-
ing load and because the physiological consequences of reduced training (i.e.,
detraining) are well documented,12 ,13, 14 the nature of the reduction in training
load is an important consideration. Because reductions in central cardio-
vascular parameters as well as peripheral muscle metabolic parameters are
to be expected with a period of reduced training, the question becomes, How
best can the triathlete reduce the overall training load while maintaining the
capacity for optimal performance? Which parameters—intensity, volume, or
frequency—should be maintained or increased, and which should be reduced?

Intensity
Conventional wisdom might suggest that an optimal taper would include a
reduction in intensity, but much evidence indicates the contrary. Numerous
investigators have reported that physiological adaptations can be maintained
and performances improved during periods of reduced overall training loads
if high-intensity training is maintained or increased.1, 7, 15, 16, 17
In runners, time to exhaustion at 1,500-meter pace improved by 22 percent
with a high-intensity, low-volume taper.18 In recently trained subjects who
reduced training intensity by 33 percent (high intensity) or 66 percent (low
intensity) over an extended period, performance in a 5-minute test effort
did not change in the high-intensity group, but it declined 30 percent in
the low-intensity group.19 Within a group of 24 distance runners, those who
reduced volume 85 percent while maintaining interval intensity work during
run training improved 5K run performance by 3 percent whereas a control
group did not.19 Thus, evidence consistently shows that intensity should be
maintained and not reduced during a taper.
Tapering and Peaking for Races | 329

Volume
If intensity appears to be a critical parameter to maintain during the taper,
then volume appears to be a critical parameter that should be reduced during
the taper. Total training volume is a function of frequency and duration,
but in most cases, training volume is manipulated primarily by duration as
opposed to frequency.
Numerous studies have provided evidence in support of the argument that
large reductions in training volume can be implemented during the taper
and performance improved as long as intensity is maintained. Bosquet et
al. performed a meta-analysis of 27 studies in the literature that met various
stringent criteria required for inclusion. They found that optimal performance
improvements were obtained with tapers that included a 41 to 60 percent
reduction in training volume with no change in intensity and frequency.1
Their analysis showed that when all disciplines (swimming, running, bicy-
cling) were included in the analysis, the positive effect of the taper dropped at
training volume reductions greater than 61 percent and less than 40 percent.
Therefore, in general, substantial reductions in training volume seem to be a
necessary component of a properly executed taper.
An interesting aspect of the Bosquet et al. study was apparent when the
data were examined by individual discipline. For example, in the case of
both cycling and swimming, the greatest positive effects during the taper
were still observed when training volume was reduced between 41 to 60
percent. On the other hand, in running, the strongest and only significant
effects were observed when the training volume was reduced 21 to 40 percent.
Although the authors did not recommend a smaller reduction in training
volume for running versus the other disciplines, this approach appears to
merit consideration.
One explanation for the discrepancy between disciplines may be that,
in general, training volumes are typically lower for running than in other
disciplines. Because total training volumes are lower, corresponding accu-
mulated fitness will be lower, resulting in lower end-taper fitness after a
reduction in training volume. Some have argued that fitness level does not
affect the optimal approach to the taper,3, 7 but a slightly different approach
to training volume reduction should possibly be considered by triathletes for
running because larger volume reductions are clearly of benefit for cycling
and swimming.1

Frequency
If the critical nature of maintaining intensity and reducing volume during
the taper are well established, the importance of the role of frequency during
the taper is less clear. Numerous studies have demonstrated that both physio-
logical parameters and performance can be maintained or improved in the
face of reduced training frequencies over a period of 2 to 4 weeks.10, 20, 21, 22, 23
330 } McGregor

Houmard has suggested that swimmers maintain daily high-intensity work


during the taper, that ideally they should not reduce frequency by more than
20 percent, and that reductions of 50 percent or more would be detrimental.24
One factor that seems to be an issue with frequency reductions during the
taper is that of feel for the activity. This aspect is likely of greater importance
at higher levels of ability and competition.
Insight regarding this issue can be gleaned from the work of Neufer et
al.,25 in which they reduced training frequency to 1 or 3 days per week in
competitive collegiate swimmers. In the group that trained 1 day per week,
after 4 weeks swim stroke rate was significantly higher and stroke distance
was significantly lower than before reduced training. At the same time, the
group that trained 3 days per week maintained those parameters. So, aside
from the physiological characteristics contributing to performance, the tech-
nical characteristics should be considered as well.

Taper Duration
Probably the best evidence in support of an optimal duration of the taper
comes from the meta-analysis performed by Bosquet et al.1 As previously
mentioned, the authors examined data from 168 studies of competitive athletes
that met certain criteria, which ultimately led to the inclusion of 27 studies in
the analysis. Their analysis showed that when all disciplines were combined,
the overall effect of tapering was greatest when the duration was 2 weeks
and was reduced for shorter and longer durations.1
When the data were examined by individual discipline, a greater margin
for error seemed to occur with a swimming taper than with a run taper. In
other words, the positive effects of a taper could be observed in swimming
for durations as short as 8 to 14 days up to more than 22 days. On the other
hand, positive effects of the taper could be observed in running only up to 14
days, after which the positive effects dissipated.1 For cycling, positive effects
from a taper could be observed for durations less than 7 days, but they were
significant and much stronger for 8 to 14 days. No data were available for
durations longer than 14 days, so it is not clear whether the positive effects
of the taper are more long lasting as in swimming or dissipate faster as in
running. But in all three disciplines optimal positive effects are observed at
2 weeks, so this duration is recommended across disciplines.

Shape of the Taper


Another question that should be considered when planning a taper is, What
should the shape of the taper be? In other words, should the training load be
reduced over time in a progressive linear fashion, a stepwise, square-wave
fashion, or a progressive exponential fashion?
If a taper is considered a reduction of training load over a given period,
the reduction can be applied in three main ways:
Tapering and Peaking for Races | 331

• First, a linear taper is a reduction whereby the training load immediately


preceding the taper would be considered 100 percent and the final desired
training load would be a given proportion of 100 percent (e.g., 50 percent).
The training load would be reduced in a linear fashion over the course
of the taper to end up at the final reduced level.
• Second, a step taper is an approach in which the load is reduced by the
given amount (e.g., 50 percent) at one time and remains at the reduced
amount for the remainder of the taper.
• Finally, in an exponential taper the load is reduced by an exponential factor,
which will lead to relatively larger initial reductions in load and smaller
reductions as the taper progresses. The exponential taper is further
divided into fast-decay and slow-decay varieties. Conceptually, these
different taper approaches can be seen in figure 23.1. In particular, note
how the fast decay tapers (fast exponential and fast variable exponential)
initially decrease more rapidly than the slow decay taper.
In general, researchers believe that the progressive taper (e.g., linear or
exponential) is more beneficial for performance improvements compared with
the step taper.1, 5, 26, 27 Further, Banister et al.26 and Zardakas et al.27 reported that
the exponential approach with a fast decay is preferable to either exponential
slow decay or linear approaches. Additional work is necessary in this area,
but the most beneficial type of taper to use appears to be an exponential,
fast-decay approach.

120.00 Slow exponential (50%)


Fast variable exponential (50%)
100.00 Fast exponential (Mujika)
Linear
Percentage of typical volume

80.00

60.00

40.00

20.00

0.00
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Days of taper

„„Figure 23.1 Tapering approaches. The figure presents the main types of taper
approaches (linear, exponential fast decay, exponential slow decay) that result in an
E5643/Friel/Fig. 23.1/448702/TimB/R1
approximate 50 percent reduction in training volume at the end of 14 days. Also, for
comparison, an exponential fast decay is presented that corresponds closely to what
Mujika has proposed.5
332 } McGregor

Mujika5 has presented a conceptual comparison of the various taper


approaches and compares an exponential slow-decay taper that results in
an approximate 70 percent reduction in training load over 14 days versus an
exponential fast-decay taper that results in an approximate 90 percent reduc-
tion over the same period. Because the optimal volume reduction appears to
be in the 40 to 60 percent range,1 this author suggests an alternative approach
that results in an approximate 50 percent reduction in volume over 14 days
regardless of the approach.
To achieve this, a variable exponent that changes each day must be applied.
This method results in a proportionally greater reduction in volume early
in the taper and a lesser reduction later in the taper. Figure 23.1 shows a
comparison of the three tapering approaches (linear, exponential slow decay,
variable exponential fast decay) in which all result in an approximate 50
percent reduction in volume over 14 days.
Table 23.1 presents the daily volumes as a percentage of typical volume
resulting from these two exponential approaches and the percentage differ-
ence between a variable fast decay versus slow decay. The fast decay results
in a greater initial reduction of training load, but in the final days before
the competition, the volumes are similar between the variable fast and slow
approaches.

Table 23.1 Comparison of Taper Approaches to Reduce Volume by


Approximately 50 Percent Over 14 Days
Variable Difference
Exponential slow exponential fast between variable
Days of taper decay decay fast versus slow*
100.00 100.00
1 94.50 87.75 6.74
2 89.30 78.51 10.79
3 84.38 71.38 13.00
4 79.74 65.81 13.93
5 75.35 61.39 13.97
6 71.21 57.84 13.37
7 67.29 54.96 12.33
8 63.59 52.60 10.99
9 60.09 50.66 9.43
10 56.78 49.05 7.73
11 53.66 47.71 5.95
12 50.71 46.59 4.11
13 47.92 45.65 2.27
14 45.28 44.85 0.43
*The right-hand column presents the difference in volume percentage on each day
between the two approaches.
Tapering and Peaking for Races | 333

Triathlon-Specific Studies
Numerous studies have used triathletes or a multisport design, and these may
provide special insight into aspects of the taper that might be particular to tri-
athlon. In a study of 16 highly trained triathletes, half of whom supplemented
with an antioxidant compound, Margaritis et al. examined the effects of a
2-week taper in which training load was reduced between 32 and 46 percent
following a 4-week period of standardized overloaded training. As a test of
performance, the athletes completed a pretaper and posttaper duathlon (5K
run, 20K bike, 5K run). Both supplemented and control athletes significantly
improved
. duathlon performance after the taper by an average of 2.6 percent,
and VO2max increased 3.0 percent.28
Vollaard et al.29 examined the effects of a 2-week taper following a 2-week
period of training overload in highly trained triathletes. As a test of perfor-
mance, the athletes performed weekly standard 45-minute cycling efforts at
70 percent of maximal aerobic power followed by a 1-minute time trial. The
overload training period consisted of a 40 percent increase in training load,
and the taper used a 40 percent reduction in load. Because of the taper, time-
trial performance improved 4.9 percent.
Eleven Ironman triathletes were examined26, 27 over a 3-month training
period that was broken up by two taper periods of 10 and 13 days. During
each taper period the triathletes were divided into two groups to examine
separate taper effects. Each of the progressive tapers resulted in a performance
improvement in a 5K run ranging from 2 to 6 percent, whereas the step taper
did not improve performance. In another study 16 experienced triathletes
who performed a 2-week taper following a period of intensified training
improved 3-kilometer running performance by 7 percent.30, 31
Collectively, these results demonstrate that in trained triathletes, a pro-
gressive taper over the course of 1 or 2 weeks will result in performance
improvements in running and cycling. How these improvements would be
extrapolated to real-world competition, especially in events of much longer
duration (e.g., half-Ironman and Ironman distance) is difficult to ascertain with
certainty. One limitation in the tapering literature with regard to triathlon
is the relative lack of data for events lasting several hours, so this avenue of
research offers interesting possibilities for the future.
Despite this, Pyne et al. argue that the results of existing tapering literature
indicate that the duration of the event is of little consequence and that general
tapering principles and results are broadly applicable across various disci-
plines and durations.7 Therefore, current evidence may be valid for extrapo-
lation to efforts of longer duration such as half-Ironman or Ironman events.

Performance Modeling
Experimental approaches and observational reports are of great value, but
their utility is limited because of both the small number of high-level athletes
334 } McGregor

who can be recruited into studies and the constraints placed on high-level
athletes by their competitive schedules. Therefore, mathematical performance
modeling studies have provided unique insights into aspects of tapering that
might otherwise not be possible.
Much of the performance modeling literature26, 32, 33, 34, 35, 36, 37, 38, 39 is based
on the work of Calvert et al.’s impulse–response model.40 This modeling
approach was tested in highly trained athletes over the course of a calendar
year.37 The authors present individual data of one of the athletes, an Olympic
finalist, in graphical form. As can be seen in a figure 23.2 from the Hellard et
al. paper, an extended period of reduced training load occurred starting with
the training block at week 33, a full 23 weeks before the primary objective of
the world championships (WC). Performances approximately equivalent to
those observed at the WC occurred 14 weeks earlier in the season in week 45.

World championships
102

100
Performance (%)

98

96

94

92

90
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65
Weeks

Training period Rest period

100
90
80
Training load (%)

70
60
50
40
30
20
10
0
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65
Weeks

„„Figure 23.2 Modeling of training load based on the Banister model in an Olympic
finalist swimmer.
E5643/Friel/Fig. 23.2/448704/TimB/R2-kh
From P. Hellard, M. Avalos, L. Lacoste, et al., 2006, “Assessing the limitations of the Banister model in monitoring training,”
Journal of Sports Sciences 24(5): 509-520. Reprinted by permission of Taylor & Francis Ltd.
Tapering and Peaking for Races | 335

These performances resulted from the taper at the end of the training block
preceding the WC block. These performances indicate that multiple peaks of
fitness can be elicited by multiple tapers in one competitive season.
The scenario reported by Hellard et al.37 is analogous to the situation
presented by McGregor et al. in data collected for an Olympic 1,500-meter
finalist over the course of 7 years.39 Using a simplified modeling approach,
originally proposed by Coggan36, 41 in cycling, McGregor et al. calculated
daily training load measures termed training stress score (TSS) and generated
estimates of fitness and fatigue termed chronic training load (CTL) and acute
training load (ATL), respectively. Because conceptually, the physiological side
of performance rests on the interaction of fitness and fatigue, and the balance
of these two parameters dictates performance, the training stress balance
(TSB) was developed to represent the simple relationship.
TSB = CTL − ATL
This modeling approach made conceptualization of the fitness and fatigue
components of performance, which are typically manifested on different tim-
escales, more readily attainable by coaches and athletes without a scientific
background. Further background regarding the development of this approach
is presented in chapter 22 of this text and outside references.39, 41
From this analysis, a similar scenario to that of the Olympic swimmer from
the Hellard et al. study was observed. As can be seen in figure 23.3a, training
load in TSS is highest at the beginning of the calendar year and then declines
over the course of the year even though the primary objective of the season,
the Olympics, occurred at the end of the season (arrow).39
Because of the relatively longer time constant of fitness (about 42 days), the
CTL component remains elevated until approximately midseason, despite the
persistent reductions in training load. As the CTL declines into the shaded
region, performances are best (figure 23.3b; box). As training load is reduced
further, CTL continues to drop out of the shaded region and performances
decline. This pattern of tapering resulted in the best performances occurring
at the period of the season that preceded the primary objective, the Olympics.
The authors argued that after fitness declined below an acceptable level,
performances declined as a result.
These results provide evidence for the arguments that
• reductions in training load for the taper should be relatively short in
duration, on the order of 2 to 4 weeks, and
• multiple optimal peaks of performance are difficult to achieve during
one season.
This contention is further supported by the modeling work of Fitz-Clarke et
al.,42 who used influence curves to model optimal training design and tapers
that would lead to best performances in a single year. Their modeling again
showed that optimal performances would result from the total cessation of
120.00

100.00

80.00
TSS

60.00

40.00

20.00

a 0.00

70 E5643/Friel/Fig. 23.3a/448705/TimB/R1 1000

60 900
50 800
40
700
Mercier score

30
600
TSS/d

20
500
10
400
0
300
–10

–20 200

–30 100

–40 0
1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
201
209
217
225
233
241
249
257
265
273
281
289
297
305
313
321
329
337
345
353
361

„„Figure 23.3 (a) DailyE5643/Friel/Fig. 23.3b/448706/TimB/R1


TSS training load during an Olympic year in which the runner
made the 1,500-meter Olympic finals (arrow). (b) Performance modeling of the athlete
presented in (a). The top line represents long-term fitness (CTL), and the bottom line
represents predicted performance trends (TSB). Season-best performances occurred
in the area surrounded by the box. Olympic performances occurred within the circle.

336
Tapering and Peaking for Races | 337

training for 16 days before the competition of interest. They also determined
that optimal performance can be achieved for only one competition in a
year because such a performance requires a reduced period of training that
adversely affects subsequent performances.
Therefore, when planning the season, the ideal approach is to plan only
one taper. Performance at all other competitions is somewhat compromised
because training must be maintained at a higher level than what would
produce optimized performance. In practical terms, however, as mentioned
in chapter 22, modern competition schedules may not permit the luxury of
planning for optimal performance in only one event at the end of a competi-
tive season. Therefore, practical circumstances typically require a compromise
when planning overall season periodization and tapering strategies.

Conclusion
From the available evidence presented in this chapter, several broadly appli-
cable recommendations can be made with regard to taper optimization.
• In general, the taper should be composed of an approximate 40 to 60
percent reduction in training volume that comes with a minimal reduc-
tion in training frequency and intensity.
• Differences in optimal volume reduction may exist between triathlon
disciplines, and smaller reductions in running volume (approximately
20 to 40 percent) may be warranted.
• Training intensities should be maintained during the taper, and fre-
quency should be reduced no more than 20 percent.
• The duration of the taper should minimally be 2 weeks and preferably
not more than 3, and the training load should be reduced in a progres-
sive, exponential fashion.
• Further, the exponential reduction in training load should follow a fast-
decay as opposed to a slow-decay pattern.
Although the area of tapering for performance in endurance sports has
drawn substantial interest, much remains to be learned. In particular, for
events lasting several hours, the optimized taper described in this chapter may
require modification. Specifically with regard to triathlon, the aforementioned
recommendations would certainly hold for sprint- and Olympic-distance
events, but for long-course events (e.g. half-Ironman and Ironman), further
investigation is warranted. Several studies1, 3, 5, 7 have argued that event dura-
tion does not seem to be a factor in taper optimization, but concrete evidence
in this regard for events such as the Ironman distance is still lacking.
Therefore, these recommendations should be used only as guidelines.
Coach and athlete discretion are an important consideration, particularly
with increasing caliber of the triathlete. Nevertheless, in general, if these
338 } McGregor

guidelines are followed, a triathlete and coach should feel confident that a
performance improvement of 2 to 3 percent relative to pretaper levels can
be expected with proper execution of the taper in triathlons from sprint- to
Olympic-distance events and possibly for longer events as well.
CHAPTER
24
Physiology of
Overtraining
—Matt Fitzgerald

E xercise imposes stress on the body. Elevated heart rate, increased blood
flow to the extremities, heightened release of hormones including epi-
nephrine and cortisol, even conscious perception of effort—all these acute
responses to exercise combine with many others to create an overall stress
state that enables the body to meet the demands of sustained exertion.
The physiological effects of such exertion do not cease when the workout
ends but continue for some time afterward. The stress of exercise triggers
processes of recovery, such as muscle tissue repair, which restore full function
to the systems affected by the stress of exertion, and processes of adaptation,
such as mitochondrial biogenesis, which enhance the body’s ability to tolerate
the stress of the next workout.
The most readily observed sign of the body’s successful adaptation to
repeated exercise stress is improved performance. The normal pattern in
a systematic program of conditioning such as triathlon training is one of
gradually increasing ability to sustain speed over distance. But exceptions
occur. Triathletes at times experience an unexpected decline in performance
despite executing the sort of incrementally increasing training workload that
normally improves performance.
Several factors can cause this sort of unexpected decline in performance.
Most are organic factors, such as an iron-deficiency anemia. Sometimes,
however, performance declines expectedly in the absence of any clear organic
cause.
Overtraining, or overtraining syndrome, is generally characterized as an
unexpected decline in performance that is not caused by an organic factor
such as anemia.1 The name derives from the idea that, in such cases, excessive
training itself is the primary cause of the performance decline. Like other
stressors, exercise in excessive amounts can overwhelm the capacity of the
body to adapt to it functionally. When this happens, the very workouts that
once enhanced performance begin to worsen it.

339
340 } Fitzgerald

Overtraining Versus Overreaching


All unexplained declines in performance are not automatically classified as
overtraining syndrome. Only severe and prolonged cases earn this designa-
tion. Experts distinguish three levels of severity: functional overreaching,
nonfunctional overreaching, and overtraining.

Functional Overreaching
Functional overreaching is a brief period of performance decline, lasting no
longer than a few days, that is a normal intermediate outcome of intensive
training. In colloquial terms, it is the 1 or 2 flat days that predictably follow
an especially challenging individual workout or short block of training.
In a well-designed training program, the predictable onset of functional
overreaching is quickly followed by a planned period of reduced training,
typically lasting 1 to 7 days. In most instances such relative rest not only
restores performance to preoverreaching levels but also yields a higher level
of performance. The term functional overreaching is suggestive of the idea that
this type of performance decline is, when properly managed, ultimately
beneficial to the triathlete.
Scientists have developed a variety of popular software tools that tri-
athletes can use to quantify their fatigue level and recovery needs so that flat
days can be anticipated and functional overreaching can be easily managed.
No algorithm, however, can make the body’s response to training entirely
predictable. Nearly all triathletes share the experience of sometimes feeling
unexpectedly flat after a day or two of easy training and, conversely, feeling
unexpectedly sharp despite fatigue from recent hard workouts.
In yet unpublished research, exercise physiologists Fabio Nakamura of the
State University of Londrina, Brazil, and Samuele Marcora of the University
of Kent, England, have documented large degrees of day-to-day variation
in individual performance in standard endurance tests despite unchanging
levels of fitness and preexisting fatigue. They speculate that daily variations
in mood and motivation may partly account for the “good day–bad day”
phenomenon, as they call it. Whatever the explanation, the important thing
for triathletes to understand is that the occasional inexplicable bad day cannot
be avoided in training and is in itself nothing to worry about.

Nonfunctional Overreaching
Nonfunctional overreaching is a state of performance decline that lasts longer
than a few days but less than 2 weeks if responded to quickly with relative
rest. It is qualitatively the same as functional overreaching in the sense that
it is defined by an unexpected loss of performance. The difference is a matter
of degree in two dimensions; the loss of performance is both more severe
and longer lasting.
Physiology of Overtraining | 341

Nonfunctional overreaching is often colloquially referred to by triathletes


as a stale patch in training. The condition typically occurs when two key
errors occur in succession. First, a triathlete or her coach creates an overam-
bitious training schedule that places the triathlete in a state of functional
overreaching more quickly than anticipated and many days before the next
period of scheduled recovery. The triathlete or coach then compounds this
error by failing to respond to the situation with unscheduled relative rest,
instead sticking to the plan. Consequently, the fatigue deficit increases and
functional overreaching becomes nonfunctional overreaching.

Overtraining Syndrome
Overtraining syndrome (OTS) is defined as a state of severe performance
decline that persists longer than 2 weeks even despite relative or total rest.1 If
nonfunctional overreaching results when a triathlete in a state of functional
overreaching fails to respond quickly with rest, overtraining syndrome occurs
when an triathlete in a state of nonfunctional overreaching continues to ignore
his severe performance decline and persists in heavy training. The onset of
overtraining syndrome is never sudden or unforeseeable, in other words.
The threshold of 2 weeks is somewhat arbitrary, having originated as a
general observation agreed upon by a panel of experts that convened at Oxford
University to “clarify the diagnostic criteria” of overtraining.2 But it is useful
nevertheless. Within the first 2 weeks of onset, nonfunctional overreaching and
overtraining syndrome look much the same. Only time can distinguish them.

Differences Between Overreaching


and Overtraining
Nonfunctional overreaching is resolved within 2 weeks if the triathlete
rests. Most cases of overtraining syndrome last for many months, and some
cases continue for years, despite rest. When a triathlete experiencing a loss
of performance is unable to bounce back after 2 weeks of rest, it is usually
possible to predict that she is unlikely to recover even with continued rest
for many more weeks.
Although the signs and symptoms of functional overreaching are limited
for the most part to performance decline and the accompanying fatigue,
overtraining syndrome is more than merely a severe and prolonged state of
performance decline; it is also characterized by a host of other symptoms,
most of them psychological in nature. These will be discussed in more detail
in the next section.
Another key difference between functional and nonfunctional overreach-
ing, on the one hand, and overtraining syndrome, on the other, is that OTS
occurs only in elite-level triathletes. Brazilian researchers recently found no
symptoms or predisposing factors of OTS in a large sample of recreational
triathletes training at fitness centers.3
342 } Fitzgerald

The reason that full-blown overtraining syndrome is confined to the elite


population of triathletes appears to be that an extremely high level of fitness
and physical resilience is required to train through a period of declining
performance and fatigue long enough for the syndrome to manifest fully.
Subelite triathletes are likely to be forced to rest by injury or illness before
reaching the level of maladaptation to exercise that defines OTS.
Motivation may be a factor as well. Subelite triathletes who have less at
stake in their sport than elite triathletes are generally less willing to continue
training intensively despite mounting fatigue and worsening performance
decline.

Signs of Overtraining
Overtraining can be recognized by four major signs: decline in performance,
fatigue, mood disturbances, and certain physiological changes. None of these
signs alone indicates a problem, but when all of them appear together a state
of overtraining is likely to be diagnosed.

Decline in Performance
The defining characteristic of overtraining syndrome is, again, a significant
decline in performance that cannot be corrected with less than 2 weeks of
relative rest.
As mentioned previously, in most cases of overtraining syndrome many
months of relative rest are needed to restore performance fully. Significant
rest is required because the decline in performance that is associated with
overreaching is caused by normal physiological fatigue, whereas the decline
in performance that is associated with overtraining syndrome is linked to
a malfunctioning of the neuroendocrine system caused by excessive expo-
sure to physiological fatigue. Recovering from overtraining syndrome is not
simply a matter of removing the stressor of intensive training but of healing
a systemic injury brought on by that stressor.

Fatigue
A second universal symptom of overtraining syndrome is fatigue. OTS suf-
ferers feel both chronically fatigued in everyday life and abnormally fatigued
during exercise. The relationship between exercise intensity and perceived
exertion changes, so that any given intensity of exercise feels more challeng-
ing to the triathlete than normal. The underperformance and the fatigue that
are always observed in overtrained triathletes are obviously linked. That is,
the fatigued state of the triathlete has a causal relationship to compromised
performance. What is less clear is the source of the fatigue.
Fatigue in triathletes may be either central or peripheral in nature. Central
fatigue is an inability of the brain to drive muscle contractions at normal
Physiology of Overtraining | 343

levels. Peripheral fatigue is an inability of the muscles to respond normally


to simulation from the motor centers of the brain. Some evidence indicates
that fatigue in overtraining is at least partly central in nature. For example,
studies have demonstrated differences in patterns of central drive to the
muscles between healthy and overtrained Olympic weightlifters.4
To the degree that overtraining is caused by fatigue centered in the brain,
another unanswered question is whether that fatigue is wholly or partly
psychological. Plenty of research demonstrates that endurance performance is
compromised in situations in.which perceived exertion is elevated. Therefore,
the reduced performance in VO2max tests that is exhibited by overtrained tri-
athletes is probably to some extent voluntary. In other words, triathletes quit
such tests at lower levels of performance than normal because they have reached
a maximum tolerable level of perceived exertion more quickly than normal.
Researchers have been unable to identify an underlying mechanism of
overtraining syndrome to explain the fatigue and performance decline of OTS.
A number of candidates have been considered. An increased blood level of
creatine kinase, a biomarker of muscle damage, has been observed in many
overtraining sufferers, but it is not seen in all cases; furthermore, creatine
kinase levels cannot be used to distinguish overtraining from garden-variety
acute postworkout muscle damage.
Low levels of glutamine, an important marker of immune function, are
also often seen in the overtrained, but a recent scientific review declared the
findings in this area inconclusive.5 Researchers have also looked at the ratio
of cortisol and testosterone, which yields a window into nervous system
function, but the results of these investigations have been contradictory; some
studies find a depressed ratio in subjects with unexplained performance
decline, and others find no abnormalities in the cortisol–testosterone ratio.

Mood and Physiological Symptoms


Although the defining characteristics of prolonged performance decline and
persistent fatigue are universal in overtraining syndrome cases, the condition
is also associated with a long list of other symptoms that are present in many
but not all cases. These indicators include mood symptoms such as depres-
sion, irritability, loss of motivation, and insomnia, and physiological symp-
toms such as changes in resting heart rate and decreased lactate production
during exercise. Overtraining has also been shown to reduce. performance
in submaximal exercise tests including lactate threshold and VO2max. Either
central or peripheral fatigue could cause these declines.
In 2003, Paula Robson of the University of Cape Town, South Africa, argued
for what she called the interleukin-6 hypothesis of overtraining.6 Interleukin-6
is a critical regulatory protein that is released in large amounts from damaged
and glycogen-depleted muscle cells during exercise. Elevated interleukin-6
levels in the brain are also known to cause fatigue in exercise and contribute
to postexercise tissue inflammation.
344 } Fitzgerald

Robson suggested that repeated release of high levels of IL-6 in triathletes


during and after exercise may precipitate a state of chronic, systemic inflam-
mation that could explain all the signs and symptoms of overtraining syn-
drome. But in the several years since this hypothesis was put forward, it has
received little confirmation.
Numerous other potential markers of overtraining have also been studied,
including oxidative stress responses to exercise, blood viscosity, and sali-
vary IgA levels. Although overtraining involves disturbances to the neural,
immune, endocrine, and metabolic systems, scientists have been unable to
find a single definitive physiological symptom of overtraining. Although there
are seemingly obvious cases of overtraining in which triathletes experience
major performance declines that persist despite prolonged rest, by and large
these individuals are difficult to distinguish physiologically from triathletes
who are not overtrained.

Diagnosing Overtraining
The lack of a definitive physiological marker of overtraining syndrome also
means that no definitive diagnostic test exists for the condition. But tests
can be performed to rule out other possible causes (such as iron-deficiency
anemia) of the symptoms that a triathlete presents.
Because overtraining syndrome is defined by a performance decline,
performance testing would be the ideal method of diagnosing OTS if it were
practicable, but in most cases it is not. Performance testing would only work to
identify a performance decrement characteristic of OTS if a baseline test result
were available to compare with the result of a postonset test. Furthermore, as
the authors of a 1995 scientific paper observed, “Using performance testing to
diagnose overtraining is difficult because of the problem with standardizing
procedures, the rarity of valid sport-specific tests, and the lack of generaliza-
tion between laboratory test and field test performances.”7
In addition to backing into a diagnosis of overtraining through a process
of elimination of organic causes of performance loss, clinicians rely on stan-
dardized interviews and questionnaires to diagnose overtraining syndrome.
The diagnosis is generally made when the triathlete reports most of the fol-
lowing signs and symptoms:
• Loss of performance
• Lack of energy and vigor
• Unusual fatigue and lethargy during training
• “Heavy” legs
• Muscle soreness
• Depression
• Moodiness and irritability
• Loss of motivation
Physiology of Overtraining | 345

• Compulsive need to exercise (possibly despite low motivation)


• Loss of enjoyment for sport
• Insomnia
• Lack of appetite
• Increased resting heart rate
• Increased number of upper respiratory tract infections
In a 1998 review, Canadian exercise physiologist Roy Shepherd wrote,
“Overtraining remains more easily detected by decreases in physical perfor-
mance and alterations in mood state than by changes in immune or physio-
logical functions.”8 Nearly 20 years later the situation is largely unchanged.

Causes of Overtraining
We might assume, based on its name, that overtraining is caused simply
by overtraining, or excessive training over a long period combined with
inadequate rest. But science has not pinpointed the mechanisms by which
training overload precipitates the signs and symptoms of overtraining with
sufficient clarity to conclude that overtraining is in fact caused solely by too
much exercise and not enough rest.
If overtraining were the sole cause of overtraining syndrome, then a certain
training load would always cause overtraining syndrome in a particular
triathlete. But the individual triathlete’s maximum training tolerance clearly
varies by circumstance. All sources of stress, whether physiologic or psycho-
logical, are processed by the body in similar ways, so that sources of stress
outside of training (e.g., job stress, relationship stress, even psychological
stress within the realm of sport) can combine with training to create a total
allostatic load that results in overtraining signs and symptoms.
Overtraining syndrome is mainly a maladaptation to the stress of heavy
training loads. For that reason, overtraining syndrome is never seen in tri-
athletes shouldering light and moderate training loads. But other stressors
in the life of the triathlete contribute. This fact suggests that reducing other
stressors and improving the ability to manage stressors beyond training
may constitute effective ways to reduce the risk of overtraining syn-
drome.
In 2000, scientists associated with the British Olympic Association put
forth the idea of renaming overtraining syndrome “unexplained under-
performance syndrome.”9 These researchers objected to the prevailing term
because it implied a single known causation.
Although the British team made a good point, the new name failed to
catch on, perhaps because the word unexplained almost seemed to suggest
unexplainable. Overtraining may not be the sole cause of what most experts
persist in calling overtraining syndrome, but it’s close enough to qualify the
popular name as a perfectly legitimate shorthand descriptor.
346 } Fitzgerald

Overtraining as a Psychological Disorder


By definition, overtrained triathletes are unable to perform at normal levels.
As we’ve seen, however, no clear physiological explanation for underperfor-
mance can be found in most cases. Although overtrained triathletes typically
exhibit certain physiological abnormalities such as high levels of creatine
kinase and low levels of glutamine, such markers are inconsistent, often
indistinguishable from the same markers in triathletes who are merely in a
state of normal fatigue, and thus insufficient to explain a the severe decline
in performance capacity that characterizes OTS. For that reason, and because
changes in mood and other psychological variables are the most pronounced
symptoms of overtraining, some researchers characterize OTS as primarily
a psychological disorder.
In 2002 Lawrence Armstrong and Jaci VanHeest at the University of Con-
necticut published a paper that explored the similarities between overtrain-
ing syndrome and clinical depression.10 They noted that the two conditions
share many symptoms, including loss of motivation, lethargy, and fatigue,
and affect the same brain structures, neurotransmitters, endocrine pathways,
and immune responses. Armstrong and VanHeest proposed that these simi-
larities were not coincidental, but that the two conditions shared an etiology.
Specifically, these researchers hypothesized that overtraining syndrome
begins when a highly motivated triathlete suffers a disappointing perfor-
mance, as every triathlete does from time to time. Assuming that the dis-
appointing performance was caused by inadequate training, the triathlete
responds to this disappointment by increasing his training workload.
If this assumption is incorrect, the triathlete’s response will lead to further
performance decrements and possibly to further counterproductive increases
in training. Over time a vicious cycle emerges, in which both the performance
declines and the physical stress resulting from excessive training contribute
to the development of the specific alterations in brain chemistry that are (or
were) believed to underlie clinical depression.
The specific alteration in brain chemistry that has received the most notice
in research on depression and overtraining syndrome is reduced transmission
of the neurotransmitter serotonin. A 2006 study compared serotonin levels
in overtrained and healthy triathletes and found no difference in serotonin
levels between the two groups.11
By contrast, a 2010 study found that elite triathletes with overtraining syn-
drome (or unexplained underperformance syndrome, as these researchers
called it) had higher levels of sensitivity in the serotonin receptors of the brain
(which would reduce brain serotonin levels) than did healthy elite triathletes.12
Clearly, more research is needed to clarify whether alterations in serotonin
activity are consistently associated with the overtrained state and whether
patterns of serotonin activity in overtrained triathletes and depressed indi-
viduals are similar. In the meantime, the so-called serotonin hypothesis of
Physiology of Overtraining | 347

depression has been seriously challenged by recent neuropsychology research.


Experts in that field increasingly believe that the true nature and etiology of
depression may be far more complex than a single neurochemical imbalance.
In this regard, ironically, depression has a clear parallel with overtraining
syndrome, which has defied every effort at reduction to a tidy explanation.
Regardless of what is happening in the physiology of the brains of over-
trained triathletes, the syndrome has an obvious psychological component.
Although the idea that OTS is fundamentally a type of depression has not
acquired many adherents in the scientific community, the consensus within
that community is that psychological factors—especially the motivation to
persist in hard training despite declining performance—are evident at the
origin of the condition and that its most salient symptoms are also psycho-
logical in nature.

Prevention and Treatment of Overtraining


Overtraining is easy to prevent in theory, because its onset is always gradual
without being subtle. Every triathlete is sensitive to the telltale signs of fatigue
and declining performance, which are always mild at first and then slowly
become more severe, allowing the triathlete to address the developing condi-
tion long before a point of no return is passed. Triathletes in heavy training
need only monitor their performance, fatigue level, and mood consistently
and take measures that include but are not necessarily limited to relative rest
when warning signs appear.
Performance monitoring should be as systematic as possible. Triathletes
should keep a detailed training log that includes split times for swim work-
outs, ideally wattage data from cycling sessions, and pace data from runs.
Scheduling standardized test workouts once every few weeks in each disci-
pline will aid in systematizing this performance-monitoring process.
As mentioned previously, a moderate, short-term decline in performance
over the course of training for an event is not necessarily a negative phe-
nomenon and can even be beneficial in the case of functional overreaching.
Triathletes, therefore, should not automatically panic every time they take a
step backward in their training. As long as any performance decline, includ-
ing an expected decline following a block of intensified training, is followed
quickly with either planned relative rest or unscheduled rest to promote
regeneration, nonfunctional overreaching and overtraining can be avoided.

Energy Levels and Mood State Monitoring


Monitoring subjective variables including energy levels and mood state may
also help triathletes interpret any performance declines that occur or even
anticipate performance declines and avoid them with unscheduled relative
rest. In 2002 Owen Anderson developed a simple questionnaire for such
348 } Fitzgerald

self-monitoring that triathletes are encouraged to use daily throughout train-


ing.13 The questionnaire consists of the following six statements:
• I slept well last night.
• I am looking forward to today’s workout.
• I am optimistic about my future performance.
• I feel vigorous and energetic.
• My appetite is great.
• I have little muscle soreness.
Triathletes are asked to rate each statement on the following scale:
1: strongly disagree
2: disagree
3: neutral
4: agree
5: strongly agree
According to Anderson, a total score of 20 or more indicates that the
triathlete is well recovered and ready to continue with planned training. A
score below 20 indicates that the triathlete needs to train lightly to regener-
ate further.

Difficulty of Prevention
Because competitive triathletes are often resistant to making unplanned
reductions to their training workload, preventing overtraining syndrome can
be more difficult in practice than it is in theory. When full-blown overtraining
syndrome has developed, prolonged rest is the only option.
In a 2006 paper published online, performance enhancement specialist
François Gazzano recommended a treatment protocol of 1 week of complete
rest followed by 6 to 12 weeks of a very gradual increase in training. Most clini-
cians agree, however, that each OTS case is highly individual. Although some
overtrained triathletes can recover within 12 weeks on a planned schedule,
others need much more time, and all overtrained triathletes must be willing
to step back from their planned schedules of gradual return to full training
in response to their symptoms. The recovering overtrained triathlete must
always remember that not listening to her body got her into this mess and
only listening to her body will get her out of it.

Conclusion
Effective triathlon training depends on balancing work and rest. Triathletes
often work too much and rest too little. The consequences of such an imbal-
ance include fatigue and stagnating or declining performance. Although the
Physiology of Overtraining | 349

severe downward spiral of overtraining is relatively uncommon among age-


group athletes, the less severe state of nonfunctional overreaching thwarts
thousands of triathletes’ race ambitions each year.
Use the information in this chapter to avoid nonfunctional overreaching
and overtraining. Be aware of the early signs of a work–rest imbalance to
spot incipient problems early, address them appropriately, and stay on track
toward race goals.
This page intentionally left blank.
Pa r t

VII
Training Base Building
for Triathlon
This page intentionally left blank.
CHaPTEr
25
Swim Base Building
—Jim Vance

J oe Friel once said, “As much as 80 percent of race-day fitness comes from
the base period.”1 Considering the approach that a training year or season
should be periodized from general preparation training to specific prepara-
tion, it makes sense that general skills and abilities would be the base or
foundation from which most of performance for the season would derive.
General preparation phase and base period being synonymous, the skills
and energy systems that best generally prepare the triathletes for the season
define this phase.
Swimming performance is highly skill dependent. This attribute makes
it unique in the world of endurance sport. To improve performance in this
discipline, coaches and triathletes primarily need to develop skills rather
than endurance.
How long this phase should last varies from triathlete to triathlete based
on goals, stage of skill development, initial fitness level, race schedule, injury
history, and even training climate. Just as coaches and triathletes set goals
for a season, setting training objectives or goals for the base period will help
define the general preparation approach to all training sessions in this period.
The time of the season devoted to base training usually follows a period of
rest and regeneration from a long prior season. Base training may also occur
at a midpoint of the year during a multiple-peak season or when the coach
or triathlete recognizes the need to reestablish basic physical attributes that
have significantly eroded.

Swimming Fitness Goals


During Base Period
Before beginning any training program, coaches and triathletes must have
clear goals in place, both for performances for the season and for training,
which will help in making all training decisions for the base period.
The first and most important step after setting goals for the base period
is to assess what the limiting factors are for the triathlete. These limitations

353
354 } Vance

are any attributes that would prevent the triathlete from reaching those
performance goals, either basic aerobic fitness, anaerobic fitness, muscular
endurance, or specific skills and abilities. When these attributes are known,
a coach can more easily determine how much time is needed in the base
period. Triathletes new to the sport may spend an entire season training only
basic skills and fitness. Because these triathletes may be lacking in physical
attributes, simply training consistently with basic sessions can bring about
steady improvement for many months, possibly the entire season.
Note that many studies have shown that body type plays a significant role
in swim performance. Some aspects of body type may be beyond the control
of the triathlete. Those that are manageable, such as body mass, could well
be a swimming performance limiter to address away from the pool.2, 3, 4, 5
For an elite or advanced age-group triathlete coming into the base period
with an ambitious race schedule, this phase may last only a few weeks, per-
haps to be repeated later in the season. It could also be extended if the coach
or triathlete believes that addressing aerobic fitness for a longer period is
needed to prepare better for a long season.

Focusing on Technique and Skills


Swim technique is an important aspect of swim performance that must
always be considered when making training decisions. If the triathlete lacks
the technical abilities to perform at a high level in a triathlon swim, fitness-
focused swim training as the central basis of performance training will likely
be a waste of time that yields little improvement and only reinforces poor
technique. A review of chapter 4 can clarify proper technique.
The skills that a coach or triathlete might focus on include lowering stroke
counts, increasing stroke rate, mastering bilateral breathing ability, increas-
ing distance per stroke, or some other individual marker or movement that
needs attention. This point is discussed later in the chapter.
If the triathlete is highly skilled, endurance fitness becomes the likely limit-
ing factor in performance. The limitation, identified by an assessment of past
performances, could be aerobic, muscular, or anaerobic endurance fitness. For
the latter, the focus could simply be to improve the triathlete’s ability to stay
with the lead pack at the start of a race. The goal could also be to prepare to
race a 3,800-meter Ironman swim by developing better aerobic or muscular
endurance. Generally, as the level of competition increases, the role of swim-
ming endurance fitness becomes more important. But, as mentioned earlier,
skill must never be compromised in an attempt to enhance fitness.

Determining the Balance of Training


The following are questions for a coach to ask about the triathlete to determine
the balance needed in the base period for training focused on skill develop-
ment versus training focused on endurance fitness:
Swim Base Building | 355

• What are the triathlete’s seasonal performance goals?


• How important is the swim leg for this triathlete as compared with bike
and run abilities and relative to the performance goals?
• Where does the triathlete typically rank overall in the swim results of a
triathlon? Could this be improved easily?
• How many years of experience does this triathlete have in the sport of
triathlon?
• How many years of experience does this triathlete have in the sport of
swimming?
• How much swim training has this triathlete done in past seasons?
• What have the base periods of previous years looked like from a swim-
ming perspective?
• What is the range of motion in the key joints, especially the shoulder and
ankle, which are important for swim technique?
• What injury history does this triathlete have?
After the coach or self-coached triathlete has determined whether the
triathlete’s base period should have a skill development focus, an endurance
fitness focus, or a balanced approach, the structuring of the mesocycle can
begin.

Base Period Mesocycle


The next consideration for a coach or triathlete is determining how much
time is available to prepare for the major competitions of the season. Gener-
ally, the more time that is available for general preparation, the less risk a
coach takes in training decisions and the more potential the triathlete has
for improvement. Some experts suggest 8 to 12 weeks for triathletes who
have not been training seriously for a period of several weeks, or a somewhat
shorter time if the triathlete is starting at a high fitness level.6, 7 Mesocycles
are discussed in more detail in chapter 22.
For the triathlete who primarily needs skill development, frequency of
swimming becomes arguably the most important aspect of training. Fre-
quency is entirely an individual matter based on the triathlete’s prior skill
development. A minimum of three 60-minute sessions per week is recom-
mended for skilled swimmers. Sessions should be more frequent and shorter
in duration for the novice.8
Skill development sessions do not have to be complex. Studies show that
simple efficiency improvements, such as better glide and reduced passive
drag or even better pacing, can significantly improve a triathlete’s perfor-
mance.9, 10, 11 As stated previously, however, these sessions do need to be
consistent and frequent.
356 } Vance

For the triathlete who needs greater endurance fitness, addressing the
clear fitness weaknesses while maintaining or continuing to develop general
fitness are the key considerations. The way in which coaches and triathletes
choose to structure this period can be as varied as the triathletes themselves.
What does an endurance fitness approach to the base period look like?
The energy systems that coaches or triathletes choose to focus on at this
time are unique to the individual and are based on experience. Common
goals and procedures of general preparation include improving aerobic
capacity, developing anaerobic power, maintaining aerobic and anaerobic
endurance, increasing specific joint flexibility, and even improving some
stroke mechanics.6
Strength training, or dry-land resistance training, is a common aspect of
training for many triathletes. Although evidence of improvement may be
found in running and cycling, ample evidence indicates that the benefits of
strength training do not transfer well to swim performance.12, 13, 14 Therefore,
improvements in performance from strength training may translate well to
cycling or running but should not necessarily be expected in a skill-dependent
sport like swimming.
Again, considering the individual triathlete’s goals and background will
help determine the optimum frequency, training intensity, and distances
when planning the base mesocycle.

Base Period Microcycle


In a microcycle intended for a skill-focused triathlete, some key considerations
should determine days in the water devoted to skill development. Microcycles
are discussed in more detail in chapter 22.
The first consideration is whether group training with a coach on deck
will be part of the program. These sessions are commonly called masters
groups or swim clubs. Although some triathletes may resist or be afraid of
these types of training groups, studies have shown that an effective coach
on deck can greatly reduce swim errors and help triathletes better maintain
the trained improvements.15, 16
Another major consideration is the structure of other training sessions, such
as bike, run, and strength training in relationship to swim skill development
sessions. Excessive pool time may well fatigue the triathlete and reduce the
effectiveness of the skill development session. The central nervous system
(CNS) needs to be in an optimal state for triathletes who are seeking skill
development.17 If the triathlete is doing more than one swim session per day,
swimming after a long day of work, or swimming after other training ses-
sions, the time of day and timing between these sessions must be considered
if the optimal CNS state is desired.18
When focusing on endurance fitness development, the microcycle consid-
erations are mostly limited to the highest intensity sessions as the basis for
Swim Base Building | 357

scheduling other training sessions. Just as with skill development, the CNS
has many demands placed on it in maximal speed work, explosive strength
work, and other high-intensity exercises, so triathletes need to come into these
sessions with as little fatigue as possible to maximize their gains.17 The more
intense the sessions are, the more critical weekly planning becomes because
24 to 48 hours are needed to recover effectively from overload endurance
sessions.7
Given the technical and high-intensity training demands on the CNS, the
structuring of a microcycle should start with the key workouts for accom-
plishing the goals of the period. The restorative sessions are then interspersed
between them to allow adequate recovery. The coach or self-coached triathlete
must continually monitor the response to the mix of training dose to ensure
positive adaptation.

Base Period Individual Sessions


When constructing an individual training session, whether for technical
skill development or endurance fitness development, the most neurologically
demanding aspects of the session should come early in the session, after
an adequate warm-up but before the triathlete reaches a state of significant
fatigue. In practical terms, the most technically demanding and physically
intense efforts must occur early in the session and the longer, less intense or
aerobic efforts can occur later.
Some of the most overlooked aspects of swim skill are those specific to
open-water swimming. Open-water skills sessions throughout the base period
can often produce the critical performance jumps necessary to achieve goals.
This point is especially true for drafting skills that can conserve energy
throughout the race.4, 19, 20, 21, 22
Simply developing the ability to swim straight in open water has produced
marginal performance improvements.23 A program, whether for beginner or
advanced triathletes, that considers open-water skills is likely to yield posi-
tive performance results. Workouts and drills for simulating and improving
open-water skills are listed later in this chapter.
Swimming skill development is key for many, if not all, swimmers and is
typically found to be the biggest challenge facing novice triathletes. At least
one study, however, showed that all triathletes who engaged in an aerobic
swim training program saw fitness performance improvements, so training
focused on skill development should not come solely at the expense of swim-
ming endurance fitness training.24 Conversely, focusing solely on endurance
fitness and not attempting to fine-tune technical skill risks a reduction in
race-day performance for even the best of swimmers in triathlon.
The goals of the individual session should consider the state of the tri-
athlete’s CNS; the time of day; and the needs, goals, and background of the
individual triathlete.
358 } Vance

Workout Creation and Monitoring


for Adaptation
Workouts for skill development and endurance fitness enhancement can vary
greatly but are often mutually beneficial. For example, an ability to apply a
greater force and a longer distance per stroke has both skill and endurance
fitness components. Focusing the purposes of the workouts and designing
them to meet the goals of the triathlete, both for skill and endurance fitness,
is one of the biggest challenges in triathlon.
After goals and objectives have been established for the season and the base
phase, coaches and triathletes need to monitor progress toward those goals.
This approach can help produce positive adaptation and prevent performance
plateaus. Coaches and triathletes should determine a test set that represents
the demands of the goal races and goal performances. Begin the base phase
with this test and revisit it often throughout the period.

Monitoring Progress
If as much as 80 percent of race-day readiness is established in the base
phase, then monitoring progress early in the phase to gauge progress toward
race-day performance will provide feedback about the effectiveness of the
program. If general preparation can yield significant improvement toward
a specific goal, then confidence and motivation can be enhanced. After the
general preparation is complete, continued monitoring with the same testing
protocol during the ensuing mesocycles will continue to provide feedback
about the effectiveness of the specific preparation.
After small changes in technique are observed or after consistent technique
training has been performed, testing the triathlete and tracking the test results
can help quantify skill improvement and training session effectiveness.
Perhaps the greatest argument for consistent monitoring and testing of
triathletes is the prevention of performance plateaus. If a coach sees a plateau
in performance before the triathlete is aware of it, consistent improvement is
more likely to occur. A self-coached triathlete can use the test to aid in making
training decisions. This process will keep the triathlete mentally fresh and
motivated because he will note continued success.
Because of testing, when a coach or triathlete sees that training is no longer
yielding the desired results, a change can be made quickly. Given the limited
training time available, the more skilled that the triathlete is at making this
determination, the more likely she is to experience positive performance gains.

Drills for Skill Improvement


Because fitness goals vary widely from triathlete to triathlete, the discussion
of workouts here is limited to tools and drills for skill improvement.
Swim Base Building | 359

The most effective technical skill development sessions require mental


engagement. The more mentally engaged the triathlete is, the greater the
opportunity is for positive training adaptations.25 To be effective, the drills
and workouts described require all the mental engagement that the coach
or triathlete can summon. The following are intended to help the coach and
triathlete recognize skill deficiencies while promoting improvement.
Swim Golf This drill provides objective feedback regarding technique changes.
The triathlete swims a 50 (meters or yards) counting strokes (both hands). At
the end of the 50, the stroke total is added to the time in seconds. The result is
the score for that 50. Fatigue needs to be minimized because it affects results,
so long recoveries should be used between the 50s and a consistent, moderate
effort needs to be employed.
The ideal scenario is to lower the score by both time and stroke count over
the course of a few such 50s. This drill provides the opportunity for triathletes
to experiment with stroke counts or with other variables such as bilateral or
single-side breathing to discover the effect on time. The drill helps the triathlete
understand individual tendencies while experimenting with technique.
Tennis Balls In this drill triathletes simply swim with a tennis ball in each hand
while focusing especially on the catch phase of the stroke. The tennis ball swim
is somewhat better than closed-fist swim drills, which are rather common. With
a tennis ball in the hand the triathlete is unable to cheat by slightly exposing
more palm.
Bilateral Breathing Triathletes of all abilities tend to have a dominant breath-
ing side. This predisposition ultimately means an imbalance of stroke mechanics
and force application between the left and right sides throughout the stroke.
Swimming easily and aerobically with bilateral breathing is an effective way
to develop a balanced stroke, improve coordination, and reinforce technical
changes. Bilateral breathing is a great workout tool for triathletes to build both
technical skill and fitness.
Paddles Paddles have been discussed in chapters 7 and 18. The main pur-
pose for using paddles is the development of force in the pull. Paddles may
also correct technique both at the entry point of the hand into the water and
in maintaining a high elbow for better leverage of the arm (see figure 25.1).
Paddles are one of the best tools for advanced triathletes looking to improve
endurance fitness and for triathletes at all levels needing technical improve-
ments.
Snorkel A snorkel allows a triathlete to focus on technical aspects that she is
attempting to improve without the distraction of turning the head to breathe.
A snorkel gives the triathlete more time to view her own stroke underwater
(see figure 25.2). As with bilateral breathing, using a snorkel helps to balance
out both sides so that breathing does not occur continuously to one side.
„„Figure 25.1 Swimmer using paddles.

„„Figure 25.2 Swimmer using a snorkel.

360
Swim Base Building | 361

Alternative Strokes By training with strokes other than freestyle, such as the
butterfly and backstroke, triathletes can gain better body awareness, develop a
feel for the water, and improve concentration and mental engagement in training
sessions. In the case of butterfly, considerable endurance fitness improvements
may occur because of the physical demands of the stroke.
Open-Water Pack Swimming Coaches and triathletes with access to a
50-meter pool can remove lane lines and place buoys near the corners of the
pool. A small pack of triathletes then swims circuits around the buoys to simu-
late the conditions common to triathlon such as drafting, positioning, turns,
and contact.
Drafting If a large pool with removed lane lines isn’t available, triathletes can
swim tightly in a line, switching leaders every lap or 100, thus encouraging
them to stay on the feet of those in front of them. This drill allows them to deal
with kicking feet while maintaining a close position for drafting.
Pontoon Starts Although pontoon starts are not common among age-
group triathletes, they are common among the youth elite, junior elite, and
elite racing in the United States and across the ITU circuit. Using the pool deck
ledge as a simulated pontoon, triathletes dive start into an interval. By being
closely bunched together, they develop their dive skill while becoming more
confident on the starting pontoon. Safe pool depth must be considered before
employing this drill.
Swimming Blind Having triathletes push off the wall with their eyes closed
and swim until they touch a lane rope helps identify a tendency to drift either
left or right in open water.

Conclusion
After determining goals and limiters, coaches and triathletes can begin
to assess how to plan the general preparation or base phase of training to
emphasize technical skill, endurance fitness, or both. While stressing mental
engagement, a coach can monitor the triathlete for adaptation and change
the program as needed to meet individual needs.
This page intentionally left blank.
CHAPTER
26
Bike Base Building
—Neal Henderson, MSc

T he base training phase is the single most important training period for
endurance athletes preparing for competition or for those simply look-
ing to improve their level. The majority of physiological improvements and
endurance development occurs during this phase of training. Developing
cycling-specific fitness during the base period for triathlon requires neu-
romuscular development, appropriate progression of endurance building,
race-specific demands training, and work on the triathlete’s weaker areas
in cycling.

Intensity and Duration


During Base Building
The intensity and duration for the majority of training that occurs while
building endurance during the base period is best described as relatively
low intensity with increasing duration. The greatest percentage of training
performed by elite-level endurance athletes consists of low intensity base-
building workouts.1 Attempts to shortcut the base-building phase of train-
ing by increasing the intensity or decreasing the length of the base phase
are counterproductive in the long run. At least one study showed that the
intensification of endurance training, even for amateur endurance athletes
training less than 10 hours per week, resulted in decreased performance
compared with endurance training that was predominantly aerobic.1
These terms do have a relative basis in that a low-intensity base-level
ride for a top male professional triathlete might be performed at 250 to 300
watts and a long ride for an athlete training for an Ironman distance race
could be upward of 120 miles (190 km). For a novice triathlete, a similarly
stressful ride might be performed at 60 to 80 watts for 20 miles (32 km). In
both cases, the stress of the training session on the respective athlete might
be similar.

363
364 } Henderson

Identifying Appropriate Training Intensity


With respect to intensity, it is a good idea to perform some sort of field
or laboratory-based testing to identify appropriate training intensity for
aerobic endurance base building. Laboratory testing can be performed to
identify the upper limit of effort than can be sustained for about 1 hour
in trained athletes. This test is typically referred to as a lactate profile test,
and the point identified is called the lactate threshold. A lactate profile test
typically consists of 6 to 10 stages of progressively harder exercise. Mea-
surements of heart rate, perceived effort (or rating of perceived exertion,
or RPE), and blood lactate concentration are taken at the end of each 3- to
5-minute-long stage.
A similar test can be performed using shorter stages and expired gas
measurements (metabolic analysis, measuring oxygen and carbon dioxide
fractions of expired air) to estimate the threshold. The laboratory-identified
lactate threshold point is strongly correlated with the amount of power that
can be generated for 1 hour by trained cyclists.2
Field tests are typically performed as maximum effort sustained for 20
to 60 minutes. Triathletes are encouraged to perform both laboratory and
field testing several times throughout the season. The most important times
to test would be at the beginning of the base phase to ensure that training
is performed at the appropriate intensity and after the base phase as the
triathlete is ready to begin the build for a more intense phase of training.
Performing objective testing like this allows the tracking of improvements
in fitness and performance.
Periodic field testing also gives critical information that can be used to
help identify proper pacing, nutrition, and hydration strategies. These field
tests can be performed either as all-out maximum performance tests for
a specific duration or for a defined distance or course. Triathletes should
perform their field tests on courses and in conditions that they are likely
to encounter during their major competitions.3
Although the maximum power that can be generated during 1 hour of
cycling is the gold standard and definition of threshold power, the task is
difficult to complete effectively in training. Fortunately, because the rela-
tionship between power and duration for aerobic efforts shows that for each
doubling of time the drop-off in power is approximately 8 percent, shorter
field tests can yield good estimates of threshold power.
Most triathletes are capable of putting out a maximum effort for dura-
tions shorter than 1 hour in training. Therefore, a better test to estimate
threshold power in a field test is to warm up appropriately and then perform
a 5-minute effort at maximum power. The power achieved in a 5-minute
maximum test is typically a good . indication of the power at.the triathlete’s
aerobic maximum, or power at VO2max. In a laboratory test, VO2max is typi-
cally measured with 1-minute-long stages with increasing power output
Bike Base Building | 365

at each stage until failure while measuring oxygen and carbon dioxide as
well as the total volume of air expired to measure the absolute amount of
oxygen that the triathlete consumes during maximal exercise.
The final stage of power achieved during the lab test is strongly correlated
with the amount of power that can be sustained for 3 to 8 minutes of. all-out
exercise, which is why the 5-minute field test is a good analog to VO2max
power. After a brief 5-minute active recovery, the triathlete then performs
a maximum 20-minute effort to evaluate maximum sustainable power, or
simply threshold power. This method of testing has been popularized by
Dr. Andrew Coggan and his partner Hunter Allen.4 The 5-minute maximum
effort must precede the 20-minute effort, and the athlete must perform
absolute maximum efforts for both tests to obtain reliable data.
Most triathletes who perform the testing correctly will see that the
5-minute power value is 125 to 135 percent of the 20-minute power value. A
triathlete who achieves less than a 20 percent difference between 5-minute
and 20-minute power likely held back during the 5-minute test to achieve a
higher 20-minute value. In those instances, decrease the 20-minute power
value by 5 to 10 percent to reflect threshold power accurately. A single
20-minute effort without any prior maximal effort will typically yield a
power output that is 15 to 20 percent above the athlete’s actual threshold
power. Always look at the actual pacing used by the triathlete to achieve
the result. This area can often provide critical feedback toward pacing for
future tests and competition.
Another way to perform field testing is to constrain effort to a given vari-
able such as heart rate. Knowing the heart rate associated with the triathlete’s
threshold from laboratory testing can give an excellent upper ceiling for
both purely aerobic levels of effort and efforts that would be associated with
threshold power or pace. Long-distance triathletes would gain more useful
information on progression in race-specific power and pace by constraining
the heart rate during a longer field test of 20 to 40 miles (32 to 64 km) at 15
beats below the threshold heart rate.
A short-course triathlete, on the other hand, would gain more relevant
field-testing information from performing a 10- to 15-mile (16 to 24 km) test
effort while holding the heart rate constant at his threshold heart rate value.
In either case, the environmental conditions should be constant from one
test to another, especially air temperature and wind speed. Also, the com-
ponents, wheels, helmet, and clothing are variables that should be constant
from one field test to the next to ensure comparable data and information.
In many cases, endurance athletes consider the appropriate training
intensity for building their endurance foundation too easy. In fact, the
correct intensity is simply appropriately easy. One of the adaptations that
occur with low-intensity aerobic training is improving the body’s ability to
use fat as a fuel source during exercise, which allows carbohydrate stores
to be maintained.
366 } Henderson

Other adaptations that occur are an increase in blood delivery to the


active muscle through increased capillary density and improved cardiac
output, which is increased by improvements in the stroke volume of the
heart, or the amount of blood that is pumped with each beat.5 Another benefit
of slowly building endurance training volume with lower intensity is that
it is a safe and healthy way to increase endurance while avoiding overuse
injury and burnout.
Besides the specific methods of identifying training intensity such as heart
rate, RPE, and power output, a simple method is effective—the talk test. When
performing training at an aerobic base-building intensity, triathletes should
be able to talk relatively comfortably, to string together nearly a complete
sentence before needing to take a break for a breath.
For most people, the heart rate associated with this level of effort is 20 to
40 beats per minute lower than their threshold heart rate, which is typically
between 85 and 90 percent of their sport-specific maximum heart rate and
approximately 55 to 75 percent of their threshold power, the amount of power
that they can sustain for 1 hour of all-out cycling. In terms of perceived effort,
this effort will typically be between 2 and 4 on a 0 to 10 scale or between 11
and 15 on Borg’s original 6 to 20 scale.

Rationale Behind Low Intensity


and Long Duration
The aerobic endurance adaptations that result from low-intensity training
take weeks, months, and years to develop to their maximum. For that reason,
those who train for endurance sports performance need many years to reach
their potential. Triathletes who try to shortchange the low-intensity aerobic
training phase often end up sustaining overuse injuries and prematurely
burning out. Keep in mind that not all training during the base phase has
to be low intensity, but most. of it should be. Adaptations to higher intensity
training such as threshold, VO2max, and neuromuscular power occur much
quicker, so those forms of training can be placed later in the training plan,
often only weeks before competition.

Workout Volume and Structure


Triathletes should spend 50 percent of their total weekly training volume
(hours) on the bike during the base phase. Typically, this volume of training
is achieved with three to five cycling workouts per week. One to two work-
outs will be steady-state base endurance training, one workout will address
peak neuromuscular development, and then one or two rides will address
race-specific demands or personal weaknesses in cycling.
Bike Base Building | 367

Steady-State Training
Steady-state training sessions predominantly consist of longer rides per-
formed nearly exclusively at the aerobic base intensity (55 to 75 percent of
lactate threshold power, or 60 to 70 percent of maximum heart rate). The goal
of one workout each week would be to ride progressively longer, increasing
by 10 to 20 percent each week. Every third or fourth week typically would be
a recovery week that is significantly reduced in duration, normally a 30 to 60
percent reduction from peak volume. Triathletes should select appropriate
training routes that are not too hilly and do not require a lot of starts and
stops so that they can spend as much time as possible at the proper intensity.
Indoor stationary bike trainer workouts can also be effective for building up
aerobic base training ride volume.

Neuromuscular Development
Neuromuscular development is primarily developed through either sprinting
or big-gear work. Also, the combination of these two can be effective. Neuro-
muscular training in cycling is all about recruiting the maximum amount of
muscle fibers to produce peak torque and power. Peak power development
in cycling typically occurs between 110 and 130 rpm and can be sustained
for only 3 to 5 seconds. In trying to develop peak power the triathlete should
select an appropriate starting gear at which the cadence is near 90 to 100
rpm to begin. Efforts can be performed both seated in the saddle as well as
standing out of the saddle. Starting each sprint at a variety of cadences from
low cadence in a big gear to high cadence in an easy gear helps develop the
ability to produce high power output across a range of torque levels. This
kind of effort is extremely important for off-road and draft-legal triathletes
to perform. Triathletes who focus on long-distance nondrafting events have
less need to incorporate neuromuscular workouts into their training, but such
workouts can serve as a form of strength or resistance training for those with
limitations to conventional strength training.
Triathletes should have complete rest after each sprint effort, typically
at least 5 minutes. Between sprints, they should continue to pedal at an
aerobic base training intensity. Triathletes can also perform neuromuscular
development training by pushing a big gear at a slower cadence. One way
to do this is to do a standing start, what track riders do at the start of their
events. Using a gear such as a 53 × 15 or 53 × 16, they begin with 10 to 15
seconds of effort out of the saddle, trying to get up to the fastest speed pos-
sible. As with the high-speed sprint efforts, full recovery between efforts is
critical. Most neuromuscular development workouts consist of only 6 to 12
total efforts during a 1-hour training session. This is the kind of workout in
which quality trumps quantity.
368 } Henderson

Race-Specific Demands
Each race that a triathlete competes in might have certain features that make
it unique, such as a steep or long uphill. A hallmark workout for triathletes
preparing for these types of races might include two to four repeats of a
similar climb in progressively harder gears and with variations in standing
and seated climbing during the repeats. For those preparing for an off-road
triathlon, performing mountain bike rides on terrain similar to the upcoming
event at least once a week is important. In other races, a perfectly flat course
that is subject to significant wind might be a special challenge. Even in the
base phase of training triathletes should be including a workout each week
or every other week that will address race-specific demands for the season’s
highest priority events. Being prepared both physically and mentally for
the challenges of race day will give the triathlete confidence and proper
preparation heading into a big race. Overdoing it is possible, however, so
race-specific preparation workouts during the base phase of training should
be performed only once each week or even just every other week. A progres-
sion of the training sessions focused on race-specific demands should occur
throughout the base phase. Typically, these workouts should initially contain
only 20 to 30 percent of full race-specific demands in a session. The amount
of race-specific intensity will increase throughout the base phase, and for
long-course triathletes a goal of achieving 75 to 100 percent of the race-specific
demands in training is appropriate. Short-course racers often perform even
more than 100 percent of race-specific demands in a single session to ensure
appropriate capacity to perform the task in the race and still be able to run
well afterward. Decreasing the rest between race-specific efforts and increas-
ing the actual intensity are also effective means of increasing training stress
without adding volume to any given race-specific workout.

Inherent Weaknesses
As much as we enjoy and are proficient at certain kinds of efforts and train-
ing sessions, we need to address areas in which we may be weak or not well
prepared. A long-course triathlete who enjoys steady-state training needs to
add some variable paced efforts to training. Training sessions that address
weaknesses can be incorporated in many ways, but keeping these sessions
specific and to the point is important. The goal of these kinds of sessions is
not necessarily to turn a weakness into a true strength, but to decrease any
possible deficit that the weakness may be causing. An honest evaluation of
past training and racing results relative to the triathlete’s peers will typically
highlight the areas in which he is weak. Neuromuscular and race-specific
training may possibly overlap with race-weakness training sessions. In these
cases, the triathlete must be cautious to avoid performing more than one
neuromuscular or race-specific intensity workout each week.
Bike Base Building | 369

Workout Guidelines
Most riders are better off riding alone during their aerobic steady-state base
building rides, because group riding almost always ends up incorporating
more intense training. Performing the base building endurance training
sessions at excessive intensity is counterproductive. Doing so will increase
fatigue over time and not allow the body to achieve general aerobic fitness
adaptations such as improved capacity to use fat as fuel.
The goal of aerobic endurance rides in training is to build the volume and
duration of these rides gradually to exceed the longest goal race distance
that a triathlete may be competing in during the season. Another goal of
long endurance training rides is simply to increase confidence in being able
to maintain a consistent strong effort for several hours. The long endurance
rides are also an excellent time to practice hydration and fueling strategies
that will be used during races.

Long Rides
Triathletes focused on Olympic-distance events should build up to at least
a 40-mile (64 km) ride and even up to 50 miles (80 km) to build bike-specific
endurance that will allow them not only to cycle at good intensity, speed,
and power but also to be able to have the best possible run off the bike.
For Half-Ironman (90 km cycle leg) races, triathletes should build up to
80- to 100-mile (130 to 160 km) rides at least two or three times before the
race. Ironman-distance competitors should perform at least one epic ride
of 130 to 160 miles (210 to 260 km) before the race and have several other
rides of 100 to 120 miles (160 to 190 km) under their belts during the base
phase.
These milestone long rides build not only endurance but also the psycho-
logical confidence that comes from being able to complete much more
than the body will have to do on race day. These rides are also important
in evaluating whether a given bike position can be maintained for the
duration of the event, and they provide a good testing ground for on-bike
nutrition and hydration experimentation.
Triathletes who live in climates that are not conducive to performing
long rides outdoors during the winter can use stationary trainers and
cross-training as a means of building up steady-state endurance volume.
Athletes who live in cold and snowy areas can mix stationary bike train-
ing and cross-training with activities such as cross-country skiing and
snowshoe hiking to put in long hours of steady-state endurance training
without being exclusively on the bike. Even triathletes, the ultimate cross-
trainers, can benefit from additional cross-training, especially during the
base phase.
370 } Henderson

Weekly Progression
The weekly progression of training volume on the bike usually can be toler-
ated at greater levels with less risk of injury than can be done with swim-
ming and especially running volume. In most cycling workouts, triathletes
can increase the volume at least 15 to 25 percent per week on a weekly long
ride. (In swimming or running, the recommended increase in the volume of
sustained long sessions is only 5 to 10 percent.)
Because cycling doesn’t involve the impact and eccentric muscle contrac-
tions of running and because the bigger muscles of the legs are less likely
to suffer the overuse injuries that can affect the smaller muscles and mobile
joints of the shoulders, which are stressed with swimming, triathletes can
be more aggressive in increasing cycling volume compared with the two
other disciplines.

Sample Weekly Schedule


The following is a typical weekly workout schedule to build endurance for a
triathlete preparing for a hilly Half-Ironman (90 km cycle leg) race who is a
good uphill rider but not as strong on the flats or downhills, especially tight
turns on downhills.
1. Long-distance ride: Ride 80 miles (130 km) on rolling terrain primarily
at base intensity. Include six 10-second seated sprint efforts in the time-
trial position on flat sections of the course to work on recruiting aero
position strength and power.
2. Threshold interval ride: Ride 35 to 40 miles (55 to 65 km) on a course
that includes at least one hill that is .5 mile (.8 km) long. After an appro-
priate warm-up, perform six efforts at 90 percent of threshold power on
the flat for .5 mile before the hill and then perform at 100 to 110 percent
of threshold up the .5-mile hill. Rest with easy active recovery spinning
for 5 to 10 minutes between intervals.
3. Descending and cornering practice: Warm up well and then practice
four to six progressively faster 90-degree turns to both the right and the
left. Focus on looking ahead, picking the fastest line through the turn,
and steering the bike primarily by shifting body weight. The total ride
might be only 10 to 15 miles (16 to 24 km) for this type of workout.

Conclusion
Building an effective cycling base is critical for success in triathlon. Spend-
ing significant portions of time developing a cycling base is the only way to
build the endurance necessary to be a competitive triathlete. The foundation
of fitness that all other phases of training will be built on depends on the
Bike Base Building | 371

cycling base. Being able to ride beyond the race distance of the goal event is
important both physiologically and psychologically. Increasing neuromus-
cular power on the bike and building confidence with race-specific workouts
will set up triathletes for later success. Reducing any weaknesses during the
base phase is also smart, because triathletes will then be able to focus on
improving the most important aspects of their fitness and performance for
their competitive events.
This page intentionally left blank.
CHAPTER
27
Run Base Building
—George M. Dallam, PhD

T he classic concept of building a base in running refers to the idea of


developing running performance by running increasingly larger volumes
at a largely aerobic effort level or pace. It is widely, although not uniformly,
believed that developing the ability to run very long distances at slow pace
regularly is an important initial component of being able to run moderately
long race distances at a faster pace later.
This concept is often referred to as laying the foundation on which higher
intensity training will be built. This idea of base training fits into the use of
a periodization training approach most commonly referred to as the general
preparation phase of each macrocycle. Later periodization phases typically
consist of a more diversified training program that includes a range of train-
ing intensities designed to be more specific to the stresses of triathlon or
running racing. Such an approach is now broadly advocated in preparation
for successful performance in endurance events.

Conflicting Evidence Regarding


Base Building
But the utility of the base training or general preparation phase of peri-
odization, and even periodization itself, has not been widely examined in
carefully controlled scientific studies in running or triathlon.1 In addition,
the few studies that examine the concept indirectly in other sports are not
supportive.2 Research in running and other endurance sports that examine
the current practices of elite athletes suggests that a mix of running intensi-
ties (about 20 percent of work at higher intensity and about 80 percent of
work at aerobic threshold and below) is a far more common training model
throughout most of the competitive year.3, 4, 5, 6
In addition, spending more relative training time at and below the
aerobic threshold intensity in training also appears to be related to greater
competitive success in running, assuming that a minimum amount of time

373
374 } Dallam

is spent training in the higher intensity zones.3 This minimum might be


thought of as one or two sessions per microcycle that focus on both basic
speed and speed endurance. A shift in this direction has also been docu-
mented in rowers who improved to an internationally competitive status
over an extended period.7 Of course, these observations may be explained
by the greater capacity for total training volume likely to occur in higher
ability endurance athletes who have a greater concentration of slow-twitch
muscle fibers and the time necessary for recovery to respond favor-
ably.
Although not directly studied, beginning triathletes may be well served
by using a base training approach initially to build training tolerance
gradually before entertaining a full range of training intensities. The most
commonly cited reasons for doing so are preventing injury and preparing
the body and mind for the rigors of greater intensity training as described
previously. This approach also may be useful in the initial build-up for
recreational triathletes who take considerable time off from training during
the winter.
Elite and high-performing age-group triathletes, however, generally
train nearly year round. The question of greatest importance for them is
whether it is productive to cycle through periods of classical base training,
systematically or not.

Base Training Versus


Higher-Intensity Training
In a study examining this concept in highly competitive cross-country
skiers, athletes who had not made significant progression following the
use of a classic periodization approach, including a base training general
preparation phase, over 1 year, were placed in a group that reduced the
volume of low-intensity training and increased the volume of higher inten-
sity training throughout the next training year. They essentially replaced
base training with a more race-specific form of race preparation.2
The addition of an increased volume of higher intensity training and
a reduction in overall training volume resulted in substantially larger
improvements in the treatment group, both in comparison with themselves
the previous year and in comparison with the initially higher performing
control group who continued to follow a classical periodization approach
through the second year.2
Further, a study has demonstrated in running that a training program
using a variety of intensities evokes a greater response than one using higher
volumes of primarily aerobic intensity training.8 Although these findings
do not definitively negate the value of using base training in the classical
sense, they do bring into question the choice of method of periodization.
Run Base Building | 375

Tenets of Periodization
The basic tenets of periodization include the use of a systematic variation
of volume and intensity of training in cyclical phases. In so doing, a variety
of training intensities are combined to create a synergistic effect on perfor-
mance capability at the target race distance. Typical elements of an endur-
ance training periodization plan might include speed work, race-pace work,
and endurance work, and, with the recent acknowledgement of the value of
strength and peak power,9 resistance training and plyometrics.
The cyclical phases, referred to as mesocycles, follow each other, and each
has a specific focus. These most typically include a general preparation of
high-volume aerobic running followed by increasingly more race-specific
phases including the faster paced forms of training.
Arthur Lydiard’s approach, which has been the basis for many current run
training programs, follows a high volume of base training with successive
mesocycles of strength focus using hill work and fartleks and then speed
emphasis using flatter, faster running and interval training, specific prepa-
ration cycles before racing. But even in Lydiard’s day, some athletes varied
between different forms of training stress (long runs, hill runs, and speed runs,
for instance) on a regular basis over a relatively short cycle of training through-
out the training year. This approach has been referred to as mixed training.
Modern periodization approaches have evolved in at least two common
forms that reflect these earlier ideas. These forms are referred to as linear
periodization, which is similar to the Lydiard approach in having a differ-
ent training emphasis in each phase, and undulating periodization, which is
similar to the mixed training approach by regularly including all emphasis
areas. Unfortunately, the relative effectiveness of each type of periodization
has not been well studied in endurance sports, so any attempt to make infer-
ences must come from another area, that being resistance training.

Comparing Linear to Undulating


Periodization
The two common forms of periodization, linear and undulating, have been
fairly broadly studied recently in resistance training.10, 11, 12, 13, 14, 15, 16 Linear peri-
odization, the classic approach, proceeds as follows. Each successive training
mesocycle focuses on a single level of training intensity over multiple days
of a microcycle, proceeding from relatively low intensity to relatively high
intensity over successive mesocycles.
By contrast, undulating periodization proceeds without mesocycles,
including only microcycles (typically 7 to 10 days) that incorporate a cyclical
variation of differing intensity levels on differing training days within the
microcycle. Examining the differences in these approaches helps address the
376 } Dallam

underlying question of whether the systematic and sequential use of limited-


focus training cycles is an effective approach.
The scientific literature on linear versus undulating periodization in resis-
tance training, when equated for total training stimulus and addressed as
whole, produces the following observations:
• both forms of periodization result in significant improvements among
athletes in varying states of training;10, 11, 12, 13, 14, 15, 16
• no differences in strength are found after shorter periods (9 weeks) in
untrained males11 and untrained females;12
• linear periodization has been shown to produce greater increases in
strength over longer periods (12 weeks) in recreationally trained males10
as well as no difference in premenopausal women;17
• undulating periodization has been shown to produce statistically greater
increases in strength among trained males over longer time frames
(12 weeks)15 and the inference of greater strength increases in trained
males,13,14 although neither of the last two results was statistically sig-
nificant; and
• reverse linear periodization (working from high intensity to low intensity)
has been shown to produce greater increases in muscular endurance than
either linear periodization or undulating periodization.15
The first inference that we might draw from these data is that a complex
training approach using multiple intensities is uniformly successful in cre-
ating improvement. Second, however, neither periodization organizational
scheme is clearly more effective in the early phases of training, although an
undulating approach may hold more promise for those who are already well
trained. One theory concerning the reduced efficacy of a linear approach to
periodization in comparison to an undulating one among well-trained athletes
is that successive training cycles do not maintain previously acquired physio-
logical abilities adequately, ultimately limiting performance improvement if
the quantities under consideration are symbiotic in nature. An example would
be the interaction of strength and endurance capacity.9
Last, when the target performance is more endurance focused than is pos-
sible in daily training efforts, it may be more effective to work from higher
intensity training cycles toward lower intensity training cycles versus the
opposite, which is more traditional.
Anecdotally, we know that many Ironman triathletes have used this
approach, training with more emphasis on a variety of intensities at shorter
distances and racing early in the year and then performing a block of training
that is primarily endurance focused as the build-up to a specific Ironman event.
Consequently, the use of base training periods or mesocycles with a more
single-minded focus on the development of increased endurance and move-
ment strength capabilities seems to have the following applications in a peri-
odization approach to training:
Run Base Building | 377

1. The early season development of both beginning triathletes and those


who have taken a significant break from the training process
2. As a method of specific race preparation when the triathlete is attempt-
ing to race at distances significantly longer than typical daily training
volumes
3. In triathletes who have become psychologically and physically fatigued
following extended periods using an undulating periodization approach
with a focus on multiple intensities of training

Building a Base Through


a Periodized Run Training Program
Several widely accepted, although not scientifically validated, concepts sur-
round the design of base training mesocycles. Following is a brief rundown
on each of them.

Focusing on Aerobic Training Intensities


The first is that focusing primarily on aerobic training intensities offers the
greatest potential to adapting successfully to progressively increasing train-
ing stress later. This is thought to be a function of the relatively lower level of
autonomic nervous system fatigue5, 6 and lower glycogen usage18 associated
with such training. Controlling training intensity so that it remains at an
aerobic level can be accomplished by identifying the pace at which the aerobic
threshold is reached and then progressively increasing the intensity of work.
The aerobic or first lactate threshold is identified by the first significant
increase in blood lactate above the baseline (typically 1 or 2 millimoles) with
each progressive increase in running speed during a progressive running
test. This velocity, or the heart rate associated with it, can be identified using
any of the following methods of determining the aerobic threshold:
1. Use heart rate monitoring to determine when 80 percent of the heart rate
maximum has been reached. Determining relative training intensity reli-
ably by heart rate requires that the maximum heart rate be measured,
not estimated.
2. Use the predicted pace for the marathon distance and beyond. Predicted
pace for various distances can be estimated using a variety of mathemati-
cal models.
3. Use ratings of perceived exertion (RPE of 12 to 14 on the Borg scale of 6
to 20) generated from a graded exercise test.
4. Identify the velocity at the aerobic or first lactate threshold by measuring
lactate values during a graded exercise test or by using the simple talk
test whereby the triathlete runs at the fastest speed at which he can still
talk normally.
378 } Dallam

Gradually Increasing the Training Load


The second premise is that training load should be increased by no more
than 10 percent in any given microcycle to reduce the potential for injury.
This concept has been contradicted, however, by a recent study illustrating
no reduction in injury among novice runners following a longer progres-
sion based on the 10 percent rule in comparison with a control group that
progressed faster.19
One interpretation of this data is that the act of progressing training without
a periodic break is the real culprit in creating injuries because of accumulated
fatigue, not the rate of progression, because neither group appeared to do so.
Of course, if this interpretation is true, the faster progressing group should
have produced more injuries sooner. This was the case in the previously
mentioned study. Injury rates were compared at the end of the training
period (8 weeks versus 13 weeks) and were nearly identical at 20 percent of
those participating in each group, implying that the faster-progressing group
reached that proportion of injuries sooner.19

Adding a Restoration Cycle


This idea leads to a third training concept—that several progressive training
cycles during which training load has been increased should be followed
by a reduced training load cycle, sometimes called a restoration cycle, to
disburse the accumulated fatigue of the prior progressions and allow more
successful adaptation.
The recently developed technological capacity to measure work output
directly in cycling20 has permitted the observation of this phenomenon in
real-world training scenarios. Further, the concept has been modeled suc-
cessfully in endurance training.21 Consequently, when athletes increase their
current training load over their recent historical average training load, which
represents their adaptive status or fitness, they induce calculable fatigue
(current load minus historical load) and a reduction in performance capac-
ity. With successive increases in the training load beyond current capacity,
fatigue increases further. Eventually this increasing or accumulated fatigue
will result in injury or overtraining.
This understanding leads to the 3-to-1 rule: After several increases in
training load over progressive microcycles of training, triathletes should
proactively reduce training load to reduce accumulated fatigue and allow
further progression before they incur an injury or become overtrained. In
other words, after 3 weeks of increases in training load, the load should
be reduced for 1 week. In theory, such an approach is likely to reduce both
injury probability and the potential for overtraining. In application, training
loads are reduced substantially (up to 50 percent) over a period of days to a
full microcycle.
Run Base Building | 379

Incorporating Some Intensity


The last widely accepted belief regarding base training in running is that some
elements of the broad spectrum of training intensity should be used for the
purpose of maintaining the specific adaptations associated with them. Com-
monly, this is accomplished using alactates, or strides, as they are referred to
in running. Strides are short, 15- to 20-second fast running efforts designed
both to maintain running speed and to develop running technique at higher
speeds. In addition, it has been observed that it is useful to maintain the
steady-state speed associated with racing by periodically using a relatively
shorter tempo or race-pace effort.

Sample Four-Week Mesocycle


Table 27.1 shows an example of a 4-week mesocycle in a base training plan for
a typical competitive age-group triathlete, currently adapted to 18 miles (30
km) per week of running and focused on sprint- and international-distance
competitions. The cycle is designed to increase endurance using the principles
described previously while maintaining high-end speed, race speed endur-
ance, and hill-running strength.

Table 27.1 Four-Week Base Training Mesocycle


Week Day Training
1 1 5 miles (8 km) aerobic running on a hilly course, 8 × 100-meter strides
5 miles (8 km) aerobic running with 4 × 100-meter strides followed by 20-minute
2
tempo effort
3 8 miles (13 km) aerobic running
Total distance: 18 miles (29 km)
2 1 5.5 miles (9 km) aerobic running on a hilly course, 8 × 100-meter strides
5.5 miles (9 km) aerobic running with 4 × 100-meter strides followed by 20-minute
2
tempo effort
3 9 miles (14.5 km) aerobic running
Total distance: 20 miles (32.5 km)
3 1 6 miles (9.5 km) aerobic running on a hilly course, 8 × 100-meter strides
6 miles (9.5 km) aerobic running with 4 × 100-meter strides followed by 20-minute
2
tempo effort
3 10 miles (16 km) aerobic running
Total distance: 22 miles (35 km)
4 1 3 miles (5 km) aerobic running on a hilly course, 4 × 100-meter strides
3 miles (5 km) aerobic running with 4 × 100-meter strides followed by 10-minute
2
tempo effort
3 5 miles (8 km) aerobic running
Total distance: 11 miles (18 km) (restoration)
380 } Dallam

Base Training Considerations


Several concepts are significant to maximizing the effectiveness of a base
training period. These include evaluating the adaptive process in some way,
quantifying training load in a manner that allows intelligent load progression
and successful adaptation to occur, and incorporating additional elements
of a complex training approach such as strength, strength endurance, and
skill training.

Evaluating Adaptation
When training is aimed at increasing endurance first and foremost, this
quality should be measured systematically to see whether the program is
successful. Of course, the best measurements are part of the training process
itself. An easy way to evaluate improving endurance without the need for
time trials or racing is to measure the association between heart rate and pace
in some commonly repeated element of the training process.
For instance, in the previous example, if the triathlete measures the heart
rate and average pace of the tempo run and conducts it during each microcycle
in the same conditions and location, evaluating the success of the adaptive
process of the overall cycle will be relatively easy. When heart rate is reduced
at a given pace or the pace is faster without a concomitant and proportionate
increase in heart rate, then successful adaptation is taking place. When this
is not the case, a reevaluation of the process should occur.
Of course, this evaluation helps in answering another fundamental ques-
tion about base training: How long should this be carried out? An easy answer
is that base training and progressive increases in training load should be
carried out until successful adaption is no longer possible. Typically, this
occurs over periods from about 4 weeks to as much as 24 weeks, depending
on training background. At that point training tolerance has been reached,
so moving to a more complex undulating approach or the next successive
phase in a linear periodization approach is appropriate.
Further, a triathlete who has a method for quantifying total training load
now has the most valid means possible of establishing an optimal training
load, which is something just below that at which successful adaptation
stopped occurring. This optimal load should be taken into consideration
when designing successive training cycles that will contain a greater volume
of higher intensity training. Typically, the total volume of training should be
reduced to maintain a training load within the triathlete’s current optimum
range.

Measuring the Training Load


Quantifying loads in cycling is possible using power-measuring devices,
but such devices are not yet available for running, for reasons discussed in
Run Base Building | 381

chapter 7. But triathletes can create a rudimentary system for measuring


running load if both pace and distance can be measured or calculated. Load
is definable as the total work accomplished. The work rate of running is
indirectly represented by time of running multiplied by the average pace of
running. With most current GPS systems, the time and average pace in miles
per hour of a run can be easily determined and used in a simple spreadsheet-
based calculation. An example follows:
58 minutes total run time × 6.5 miles per hour average pace
= 377 stress units
Such a simple system for quantifying the outcome of work performed does
not account for the change in work rate needed to facilitate running up or
down hills. Consequently, an assumption is that training will be carried out on
similar courses to produce reasonably meaningful microcycle-to-microcycle
comparisons of training load.
After training load can be measured accurately and reliably, the basic
principles of progression can be applied. The most fundamental progression
principles are as follows.
Training load should increase gradually until successful adaptation is no
longer occurring. This point represents a ceiling for any given person beyond
which further training is counterproductive. This point has been reached
when a reduction in work capacity occurs rather than an improvement.
Training load progression should be attempted using small adjustments,
ideally of about 5 to 10 percent of the total load. Small relative increases in
load reduce the likelihood of injury and increase the probability of success-
ful adaptation.
Progressions in training load should occur only when successful adaptation
to the previous load has occurred. For instance, following an increase in total
microcycle running load, successful adaptation has occurred when average
running velocity has become faster at any given heart rate or level of effort.
When this is evident, training load can be further increased.
Temporary reductions in training load over several days, on the order of
30 to 50 percent of the present load, should occur systematically following
successive increases in load. This plan permits the shedding of accumulated
fatigue and increases the possibility that further progressions will be suc-
cessful in creating adaptation.

Strength Training
For those who engage in the practice of strength training for triathlon, base or
general training periods are probably the most useful time to make progres-
sive increases in the resistance training load as well, following the principles
already described. This approach has been demonstrated to minimize the loss
in stride length acutely during intensive running training22 and may help to
382 } Dallam

offset losses in speed and range of motion occurring chronically from run-
ning large volumes at slower speeds. As with the endurance training process,
the efficacy of the approach should be regularly assessed by evaluating effort
and force levels across key resistance training sets.

Strength Endurance Training


Strength endurance training refers to the idea of increasing the resistance
to forward motion to a small degree in swimming, cycling, or running by
use of means that will minimally alter movement mechanics yet still allow
sustained aerobic efforts. In running, both inclines23 and weighted vests can
be used for this purpose. The process creates increased muscle activations
at the hip with each step23 while reducing step rate slightly. Using inclines or
weighted vests is analogous to using paddles in swimming or using a larger
gear in cycling.
Such training has been hypothesized to recruit muscle fibers normally
active only during higher intensity training, allowing a successful conver-
sion of such fibers to a more oxidative or endurance-oriented profile that is
more resistant to fatigue. Base training periods offer an opportune time to
include and progress strength endurance running training in the overall
training plan. The goal is to prepare the body for the rigors of both specific
hill-focused training as well as higher speed training later.

Skill Development
Base training periods also offer an opportune time to focus on furthering skill
development in running. This is often accomplished by including drill sets
following training session warm-up periods and preceding the bulk of the
aerobic running. Such a practice will ideally improve the runner’s awareness
and the use of intended mechanics throughout the remainder of the train-
ing session. High volumes of relatively slower running allow this process to
occur with minimum distraction.
Running skill development methods likely to be successful include the use
of limited barefoot running, the implementation of a learning method such
as the pose method, or the use of functional training designed to improve
total-body mobility.

Conclusion
The effectiveness of using traditional, narrowly focused base training periods
in running is not clearly established in the published sport science research.
Considering this finding, the decision to use a base training approach should
be based on individual training considerations. Base training is likely to offer
the greatest benefits to triathletes who are initiating a run training process
Run Base Building | 383

for the first time or are returning to run training following extended periods
of nonrunning.
For triathletes who train and race year round, periodic base training peri-
ods can be used to decompress from the physical and psychological stress
created by more intensive complex training approaches. But during the bulk
of the training year, a complex approach to training emphasizing a variety
of speeds and distances is more likely to stimulate improvement over time.
To minimize any potentially deleterious effects of base training periods
on the broader spectrum of running abilities, triathletes should incorporate
maintenance-level work for basic running speed using strides and maintain
steady-state speed endurance using tempo runs.
Further, strength training should be included in base training periods with
a focus on progressive strength development. Adherence to basic progression
principles including the use of aerobic intensities for most training (about 80
percent or more), slow progressions in total load using the 10 percent rule,
and systematic periods of restoration training using the three-to-one rule may
reduce the likelihood of overuse injury or overtraining. Finally, quantification
of training load offers a useful method to guide the progression as well as to
identify the optimal training load for successive training cycles.
This page intentionally left blank.
P a rt

VIII
Multisport
Event-Specific Training
and Racing Tactics
This page intentionally left blank.
CHaPTER
28
Sprint
—Gale Bernhardt

W hen triathletes enter a triathlon, both they and their coaches should
have strategy and tactics in mind before race day. Strategy is the big-
picture look at the event. If a triathlete is entering the Local Excellent Olympic-
Distance Triathlon as her first triathlon, her strategy for training and race day
will be different from that used by a seasoned triathlete looking to score a
spot on the podium. Strategy is a long-term view of triathlon, how the sport
fits into lifestyle, and the overall means for achieving goals.
Training and racing tactics are more fluid and change with current cir-
cumstances such as weather, opponents, level of fitness, and performance
strengths, to name a few. Tactics are a blend of art, science, and individual
triathlete capability.

Exertion Levels for Training and Racing


Before discussing how fast triathletes should be training and suggesting
durations of each session, the best approach is to define exertion levels or
training and racing zones. Unfortunately, not all exercise experts use the
same system to define exercise intensity. To further the confusion, swim
coaches use a different language to define exercise intensity than triathlon,
bike, and run coaches do.
Endurance sport, triathlon, and exercise experts define specific exercise
zones by drawing lines along the physiological function continuum discussed
in part IV. The layperson might expect that these lines would be clearly
defined. They are not.
When reading more information on triathlon, swim, bike, and run training,
triathletes will find that the systems contain similarities. For the purposes
of this part of the book, we will define seven training and racing zones, as
seen in table 28.1.

387
388
Table 28.1 Reference Scale for Rating of Perceived Exertion and Training Zones
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
Rating of perceived 6 to 9 10 to 12 13 or 14 15 or 16 17 18 or 19 20
exertion (RPE; origi-
nal Borg scale)
Rating of perceived 1 or 2 3 or 4 4 or 5 6 7 8 or 9 10
exertion (RPE; new
Borg scale)
Swim pace Work on form, no T-pace + 10 s T-pace + 5 s T-pace T-pace T-pace − 5 s per 100 As fast as
clock watching per 100 per 100 possible
Percentage of lactate 55 or less 56 to 75 76 to 90 91 to 99 100 to 105 106 to 120 121 or more
threshold power
Percentage of lactate 80 or less 81 to 88 89 to 93 94 to 99 100 to 102 103 to 105 106 or more
threshold heart rate
(bike)
Percentage of lactate 84 or less 85 to 91 92 to 95 96 to 99 100 to 102 103 to 106 107 or more
threshold heart rate
(run)
Breathing and Gentle rhythmic Breathing rate and Aware of Starting to Breathing is deep and Heavy, labored Maximal exer-
perceived exertion breathing. Pace is pace increase slight- breathing a little breathe hard. forceful. Many notice a breathing. Pace is tion in breath-
using runners as the easy and relaxed. ly. Many notice a harder. Pace is Pace is fast andsecond significant change noticeably challeng- ing. Pace is
example Intensity is a jog or change with slightly moderate. Hold- beginning to be-in breathing pattern. Pace ing but sustainable sprinting effort,
trot. deeper breathing, al- ing a conversa- come uncomfort- is all-out sustainable for for 15 to 30 minutes causing high
though still comfort- tion is more able, approach- 1 to 1.5 hours. Mental at race effort. Dis- discomfort that
able. Running pace difficult. ing race pace focus is required, pace comfort is high but is unsustainable
remains comfortable for all-out 1- to
is moderately uncomfort- manageable. for more than 1
and conversation 1.5-hour run. able, and conversation is minute.
possible. undesirable.
Purpose and cross- Easy, aerobic, Aerobic, extensive Tempo, inten- Subthreshold, Lactate threshold endur- Aerobic capacity, Anaerobic
reference of terms recovery. endurance, aerobic sive endurance. muscular endur- ance, anaerobic threshold speed endurance, capacity, power.
commonly used to threshold endurance Ironman- ance, threshold endurance, superthresh- anaerobic endur-
describe each zone (Note: Some coach- distance race endurance, an- old, muscular endurance. ance.
es call this region pace for inter- aerobic threshold International-distance Sprint-distance race
lactate threshold.) mediate and endurance. race pace is typically in pace is typically in
advanced ath- zone 4 or 5a for advanced zones 4 to 5b. Ex-
letes is typically athletes. perienced athletes
within zones 1 may be in zone 5c
to 3. for limited time.
Adapted, with permission, from G. Bernhardt, 2007, Training plans for multisport athletes (Boulder, CO: VeloPress), 22-23, and G. Bernhardt, 2009, Training plans for cyclists
(Boulder, CO: VeloPress), 28-29.
Sprint | 389

As triathletes investigate further, they will find that the various systems
used by fitness experts are quite similar. Some of the training zone terminol-
ogy is cross-referenced in table 28.1. The table can serve as a starting point
to examine various training intensity charts used for heart rate, power, pace,
and rating of perceived exertion (RPE) training.
Notice that a running pace column is not included. Multiple online calcula-
tors are available to estimate running pace based on a number of run tests,
including races. As with the training zones, all run calculators are slightly
different. Triathletes can use the running perceived exertion column to cross-
reference pace charts.

Using Technology to Determine Pace


The world is filled with technological devices that can give triathletes instan-
taneous numbers for pace, power, and heart rate. Although many people value
the usefulness of these devices, triathletes should never lose the ability to be
in tune with their bodies and use perceived exertion for training and racing.
Many successful high-level triathletes are so in tune with their bodies that
they use perceived exertion alone to determine race pace.
Although perceived exertion works well for many triathletes, some,
including those racing long distances, can often benefit from using heart rate,
power, pace, or some combination of those tools. Using those tools prevents
triathletes from beginning a race too fast and fading, making the end of the
race a survival march rather than a run.

Sprint Training
Sprint races are typically 400 to 750 meters (450 to 800 yards) of swimming,
20 to 25 kilometers (12 to 15 miles) of cycling, and 4 to 5 kilometers (2.5 to 3.1
miles) of running. Slightly shorter or longer races can be found, but this range
gives an idea of typical race distances. Sprint races are the perfect distance
for people to begin to develop passion for triathlon. This distance is also ideal
for triathletes who have a need for speed. Racing all out as fast as they can
go is exhilarating for many people.
Depending on the specific race distance and the triathlete’s athletic ability,
sprint-distance races typically take 1 to 1 1/2 hours to complete. Triathletes
with poor fitness can begin training and be ready to race in about 12 weeks.
Triathletes who have fitness from a single sport, such as running, can be
ready to race in about 6 to 8 weeks.1

Training Strategies
The number one strategy is to optimize the triathlete’s fitness given current
fitness level, daily time available to train, and the number of weeks available
before race day.
390 } Bernhardt

Triathletes at a Lower Level of Fitness


When triathletes have a low level of fitness, training needs to emphasize
building endurance, improving the aerobic system, strengthening tendons
and ligaments, and minimizing the risk of injury. This phase of training is
often referred to as general preparation by coaches and physiologists around
the world and in various sports. Triathletes do this training at intensity zone
1 or 2.
If the triathlete has a low level of fitness and less than 12 weeks until race
day, most of the training should be kept at zone 1 or zone 2 intensity. For tri-
athletes who have low fitness but many weeks before race day, training can
move from zone 1 to zone 2 intensity up to zone 5c. The progression of the
training load will depend on the triathlete’s profile, including single-sport
history, triathlon history, training time available, the person’s response to a
given training load, and his race goals.

Slightly Experienced Triathletes


Triathletes who have completed at least one sprint race in the previous
season and are looking to improve race performance can include more
volume and intensity, or speed work, in training. The weekly training
times in table 28.2 can be increased to accommodate the triathlete with
more fitness. With a good base of zone 1 to zone 3 endurance in the 12 to
16 weeks before race day, these triathletes can add workouts at threshold
pace or zone 4 or 5a intensity.
In the 8 to 12 weeks before the race, the triathlete can begin adding
threshold intensity work in one workout in each discipline. The intensity
can be structured intervals, or it can be accumulated time, driven by terrain.
In either case, the triathlete begins with around 15 minutes of accumulated
intensity in a single workout. For example, if the triathlete’s long run for
the week will be 45 to 60 minutes and that workout will include threshold-
paced running, she should accumulate about 15 minutes at threshold pace.
A convenient way to increase intensity and work on sport-specific strength
is to run the uphill portions of a hilly course at threshold intensity.
Obviously, because threshold pace on hills is not equivalent to pace on a
flat course or a track, triathletes will need to use perceived exertion, heart
rate, or a combination of both. If the racecourse is hilly or on trails, the tri-
athlete should use similar terrain in workouts to be prepared physically and
mentally for race-day demands.
If triathletes are estimating time accumulated at threshold, downloading
device data after the run can establish whether the time accumulated was
high, low, or spot-on. The data can be used to help hone pacing skills.
If the triathlete decides to keep the long run aerobic and do threshold
interval work in one of the shorter runs, a 3-to-1 or 4-to-1 ratio for work
Sprint | 391

to rest is appropriate.2 For example, if the workout is 45 minutes long, the


triathlete should warm up for 15 to 20 minutes. After the warm-up, he does
five 3-minute efforts at threshold pace, recovering for only 1 minute at zone
1 intensity between efforts. The work-to-rest ratio doesn’t have to be exactly
3-to-1 or 4-to-1. Sufficient time should remain at the end of the workout for
cool-down running at zone 1.
Sample Four-Week Progression for Running and Cycling Depending on
how much time remains between the beginning of threshold training and
race day, the triathlete should build accumulated time at threshold up to 20 to
30 minutes. The intervals can be manipulated to build endurance at threshold
pace. For example, a 4-week progression might look like the following:
5 × 3 minutes at zone 4 or 5a (1-minute recovery interval at zone 1)
5 × 4 minutes at zone 4 or 5a (1-minute recovery interval at zone 1)
4 × 5 minutes at zone 4 or 5a (1.5-minute recovery interval at zone 1)
3 × 6 minutes at zone 4 or 5a (2-minute recovery interval at zone 1)
3 × 8 minutes at zone 4 or 5a (2-minute recovery interval at zone 1)
Of course, the triathlete can include a steady 15 or 20 minutes at threshold
pace. To break the work into smaller segments, triathletes can include some
recovery, though incomplete, to maintain a higher average pace for the speed
portion of the workout.
For cycling, a similar strategy and structure can be used for the intervals.
The triathlete begins with around 15 to 20 minutes of time accumulated at
zone 4 or 5a intensity and builds accumulated time at threshold to 30 to 40
minutes.
Sample Four-Week Progression for Swimming In the pool, the triathlete
can use intensity, pace, or a combination of both. Because intermediate tri-
athletes have endurance, it is easy to conduct a 500-meter time trial to bench-
mark current fitness. From the time trial, an average pace per 100 meters can
be determined. For example, if the 500- meter time-trial result is 8:50, the
average 100-meter pace is 1:46. The triathlete should aim to improve that pace
by swimming repeats slightly faster than time-trial pace.
A 4-week progression is shown. The triathlete must be sure to do a 10- to
15-minute warm-up set before swimming fast. In the set, “send-off” means
that the triathlete pushes off the wall each time the pace clock reads that time.
Recovery time is the time between the end of the swim and the next send-off.
Recovery in this case will be variable.
5 × 100 on a 2:15 send-off, aiming to swim all 100s faster than 1:46
5 × 100 on a 2:10 send-off, aiming to swim all 100s faster than 1:46
5 × 100 on a 2:05 send-off, aiming to swim all 100s faster than 1:46
5 × 100 on a 2:00 send-off, aiming to swim all 100s faster than 1:46
392 } Bernhardt

Triathletes With Higher Levels of Fitness


For racers with higher levels of fitness, a reasonable question is whether the
triathlete is already a strong runner. Should one fast workout be dedicated
to running, or would it be better to maintain running speed and dedicate a
second workout to improving one of the other disciplines?
Doing more than one fast workout per discipline each week is a matter of
personal preference and race strategy. Triathletes looking to improve overall
race performance may be better off maintaining the strong discipline and
emphasizing the other two disciplines, although this strategy may not be
reasonable if the triathlete is prone to injury in one of the disciplines. For
example, if the triathlete is a weak runner but has a chronic injury area, such
as a knee, then adding another day of speed work in running may not be a
good choice.
Another pertinent question is whether intermediate triathletes can do
more than three fast workouts each week. A rule of thumb is that two to four
fast, or otherwise stressful, workouts per week is enough. Most intermediate
sprint-distance triathletes who train 4 to 6 hours per week find that four fast
sessions are enough to bring about improvement.
Quality days should be fast and recovery days easy. The body makes per-
formance gains after stress and recovery.

Advanced Triathletes
The advanced sprint triathlete is looking for performance improvements.
These highly fit and experienced triathletes have race data. They know
personal open times (discipline done alone, not part of a triathlon) for a 500-
meter swim, 12- to 15-mile bike time trial, and a 3.1-mile run. They also have
triathlon race data for various sprint-distance triathlons.
Advanced triathletes keep fit year-round, doing some type of endurance
sport 5 or 6 days per week. These triathletes have completed a progression
of training before race day that permits training at zone 5b paces in the 4 to 6
weeks before race day. Their training progression may resemble the following:
10 to 20 weeks before race day: zone 1 to 3 intensities.
6 to 13 weeks before race day: zone 4 or 5a. Using pace or power numbers
to set interval speeds is preferable.
2 to 7 weeks before race day: zone 5b intensity. Specific race-pace data
should be used to determine interval speeds in training sessions.
A wide variety of workouts can be used to improve speed. Work-to-rest
ratios for intervals are 1-to-1 or greater. To allow more rest, a ratio such as
1-to-2 or 1-to-3 can be used. For single-discipline workouts, paces can begin
at 6 to 10 percent faster than open race paces. For brick workouts, paces can
begin at 6 to 10 percent faster than triathlon paces.
Sprint | 393

Training becomes more intentional and specific. If goal races are on hilly
terrain, training is conducted on hills. If race-specific hills are not available
on natural terrain, hills can be done on a treadmill for running or simulated
on an indoor bike trainer.
A 30- to 60-minute swim session, a 2-hour bike ride, and a 1-hour run at an
aerobic pace are all considered easy and normal for the advanced triathlete.
In some cases, advanced triathletes participate in multiple races in a 6-
to 8-week period. Depending on the volume of training completed, some
sprint triathletes reduce training volume and keep some intensity in the
plan between races. Typically, triathletes use the 2 to 3 days after a race to
go easy in zones 1 and 2. If the triathlete will race the next weekend, the few
workouts before race day should be used for sharpening by keeping some
race-pace efforts in training. If there are 2 weeks between races, more inten-
sity will usually be included immediately after the recovery days and on the
weekend between races.
Discipline workouts are only part of the advanced triathlete’s training.
Training to do well at racing includes attention to diet, mental skills training,
and possibly strength training.

Sample 15-Week Plan


The sample 15-week plan (table 28.2) gives a snapshot of training progres-
sion for sprint race training. Approximate training hours are given so that
triathletes can see a model of varying weekly work volume and volume
reductions to allow recovery. Workout types, based on intensity descriptions
given in table 28.1, can help guide the workouts that triathletes decide to put
into a given training week. The primary focus for the week is a reminder that
training is a progression and that triathletes cannot work on every item of
fitness in every workout or even within every week. Finally, other comments
are given to suggest plan modifications based on fitness levels.
Table 28.2 Sample 15-Week Training Plan For Sprint Triathlon
Approximate Types of
Week training hours workouts Primary focus for week Other
1 3 Aerobic, tech- Work on proper technique Those with low fitness remain in
nique, neuromus- and aerobic endurance; zones 1 and 2. Those with higher
cular speed. begin to introduce short levels of fitness can include work
segments (30 seconds in zone 3.
or less) of accelerating
speed.
2 3.5 Aerobic, tech- Work on proper technique Those with low fitness remain in
nique, neuromus- and aerobic endurance; zones 1 and 2. Those with higher
cular speed. begin to introduce short levels of fitness can include work
segments (30 seconds in zone 3.
or less) of accelerating
speed.
3 2 Keep workouts Recovery. Aerobic time trials (TT) can be
similar to the included at the end of the week.
previous 2 weeks Duration of each TT depends on
but reduce training the triathlete’s current fitness.
volume. Restrict intensity to around the top
of zone 2 or low zone 3 within a
narrow heart rate range of roughly
three to five beats for cycling and
running. For swimming, use the
average pace achieved for 3 ×
100 meters with 20 seconds of
recovery. Highly experienced tri-
athletes can make the TTs all-out
efforts.
4 4 Aerobic, intensive Continue focus from the Include zone 3 in some work-
endurance, tech- last block and build long outs in an unstructured manner,
nique, neuromus- workout endurance. restricting total time in zone 3 by
cular speed. ability. Experienced and highly
fit triathletes can begin adding
lactate threshold work this week.
5 4.5 Aerobic, intensive Continue focus from the Include zone 3 in some work-
endurance, tech- last block and build long outs in an unstructured manner,
nique, neuromus- workout endurance. restricting total time in zone 3 by
cular speed. ability. Experienced and highly fit
triathletes can include threshold
work this week.
6 2.5 Keep workouts Recovery. Consider introducing a brick
similar to the (bike-to-run) workout and practic-
previous 2 weeks ing transitions.
but reduce training
volume.
7 4.5 Aerobic, intensive Build longest workouts Long swim workout roughly twice
endurance inter- to the limit of this training that of race distance. Long run
vals, technique, block. builds to between 45 and 60
neuromuscular minutes. Long ride builds to 90
speed. minutes.

394
Approximate Types of
Week training hours workouts Primary focus for week Other
8 5 Aerobic, intensive Build longest workouts Long swim workout roughly twice
endurance inter- to the limit of this training that of race distance. Long run
vals, technique, block. builds to between 45 and 60
neuromuscular minutes. Long ride builds to 90
speed. minutes.
9 3 Keep workouts Recovery. Repeat the aerobic TT to measure
similar to the progress.
previous 2 weeks
but reduce training
volume.
10 5 Aerobic, threshold Work on lactate threshold At this point in the training, it is
endurance inter- speed with the interval tempting to make every workout
vals, technique, work. Include some lim- fast, pushing the limits. Refrain
neuromuscular ited time above threshold from doing this. Most triathletes
speed, unstruc- (zones 5b and 5c) within respond well to two to four chal-
tured and limited select workouts. lenging or key workouts per week.
speed endurance. The remaining workouts should
be aerobic, recovery, or technique
focused.
11 5 Aerobic, threshold Work on lactate threshold
endurance inter- speed with the interval
vals, technique, work. Include some time
neuromuscular above threshold (zones 5b
speed, unstruc- and 5c) within select work-
tured and limited outs, limiting this time.
speed endurance.
12 3 Keep workouts Recovery. Consider doing a shortened
similar to the distance race rehearsal workout
previous 2 weeks this week.
but reduce training
volume.
13 4.5 Aerobic, race Intervals or long workouts If brick workouts have not be done
pace, technique, should include race-pace yet, include at least one this week.
neuromuscular work.
speed.
14 4 Aerobic, race Intervals or long workouts Final race rehearsal this week.
pace, technique, should include race-pace
neuromuscular work.
speed.
15 1.5 + race Keep workouts Include intensity segments An expected range of race paces
similar to the at race pace or slightly can be plotted based on previous
previous 2 weeks faster. workout paces.
but reduce training
volume.

395
396 } Bernhardt

Racing Tactics
The tactics a triathlete should use for race day are directly related to that
triathlete’s fitness level and experience in the sport of triathlon. Triathletes
doing their first race who have low levels of fitness should focus on comfort-
able completion of the event. Triathletes with high levels of fitness who are
gunning for a personal best performance, and perhaps a podium spot, will
use different tactics than the new racer.

Warm-Up
Beginner triathletes who have built just enough fitness to complete the event
do not need to do an extensive warm-up. Most of the time, these triathletes will
not warm up before the race. They will warm up during the swim, moving
from relaxed zone 1 intensity to the highest swimming intensity they used
during training. Most of the time this means beginning the swim in zone 1
and finishing it at zone 2 or zone 3 speed. The bike and run portions of the
race will likely begin at zone 2 intensity and move to the highest intensity
used during training.
Triathletes with some fitness and race experience may decide against a
warm-up. The main difference between these triathletes and the beginners
is that the race intensity tends to be higher during the last half of each event
and for the overall race. Some of these experienced triathletes will prefer a
very short warm-up of 5 to 10 minutes of easy bike spinning to check the
gears, followed by a run of 2 to 5 minutes. These athletes may or may not
include a short 100-meter warm-up swim.
Because the advanced triathlete’s race is fast from the start, a prerace
warm-up routine is necessary. These athletes may ride the bike for 20 to 30
minutes before the start on race day. They will likely also do a run of 10 to 15
minutes. In both the bike and run warm-ups, a few race-pace efforts should be
included. These efforts are typically 60 seconds or less in length and include
generous recovery times of at least 2 to 3 minutes.
A swim warm-up is also part of the race-day routine. Swimming 300 to
500 meters is not uncommon. A few 25-meter segments at race pace should
be included in the warm-up. If swimming is not an option for warm-up,
swim cords can be used.

Pacing
A negative-split tactic is the best way to enjoy a successful sprint event for
newer triathletes. Plan to begin the race at a pace that seems too easy. Tri-
athletes should plan to finish the race at the fastest pace practiced in training
or slightly faster.
Experienced triathletes’ race-day speeds should be similar to the speeds
used in training. Because the experienced triathlete had some 4 to 8 weeks that
Sprint | 397

included threshold work, a good portion of the race can be done at threshold
intensity. Triathletes still green at racing should aim for a negative-split pace
in each discipline. They should begin each discipline in zone 1 or 2 and push
to zones 3 to 5a at the halfway point.
As soon as the gun goes off, faster triathletes are hitting zone 4 to 5a pace.
Advanced triathletes use discretionary zone 5b and perhaps zone 5c. These
anaerobic efforts are used tactically to advance overall race placement. In the
swim these efforts are used to drop drafting competitors or bridge a gap to
the feet of a competitor for purposes of drafting.
A negative-split strategy may be used by beginning the first half of each
discipline at zone 4 and performing the second half of each discipline at zone
5b pace as much as possible. Although the triathlete may have a prerace strat-
egy to perform a negative-split race, the competition may demand the use of
higher race paces at times not favored by the triathlete. The best triathletes
will be ready and willing to wield fitness at any time, regardless of time or
location within the event.
In the most hotly contested races, triathletes must be ready for a final kick
sprint to the finish line. The most competitive athletes, with the best competi-
tion, will need to train for this final kick.

Transitions
In sprint-distance racing, fast transitions are more important than they are
at longer distances. Races can be easily won or lost in the transitions. When
exiting the swim, running to the first transition (T1) is necessary. If the run
to T1 is long or hilly, this element needs to be included in training.
Transition setup before the race has cycling shoes attached to the pedals.
The shoes and pedals are held parallel to the ground by attaching lightweight
rubber bands to the shoes and securing the shoes to the frame. This setup
keeps shoes from wildly rotating while athletes run out of transition. Rotat-
ing shoes can catch on the ground, jamming and causing the shoe to release
from the pedal. Obviously, stopping to retrieve a lost shoe is a time waster.
The fastest triathletes use a flying mount, similar to those used in cyclo-
cross, to jump onto the bike and begin pedaling with their feet on top of the
shoes. The light rubber bands break away. The feet can then be slipped into
the shoes when the triathlete has settled into a comfortable pace.
Notice that socks are not mentioned. Fast transitions for short races do
not allow the use of socks. This topic will be further discussed in chapter 33.
As in T1, the fastest triathletes perform a flying dismount before reaching
the dismount line for the second transition (T2). To be ready to dismount,
triathletes remove their feet from the cycling shoes and pedal with their feet
on top of the shoes for the last 100 to 200 meters.
At the dismount line, these triathletes hit the pavement running. Slipping
into running shoes sans socks, they head out on the run course. The fastest
and most fit triathletes are capable of running the entire 5K in zones 5a and
398 } Bernhardt

5b. Those with less fitness are better off beginning the run in zone 3 or 4
and doing a negative-split effort by doing the last half of the run in zone 5b.

Weather and Altitude


In a perfect world, the weather on race day would be not too hot, not too cold,
but just right. But because our world isn’t perfect, triathletes need to be pre-
pared for weather extremes. Although race directors often cancel or modify
a race if weather is extreme, the decision of whether or not to race remains
with the individual triathlete.
If a triathlete expects to race in hot and humid conditions on race day,
doing some acclimatization is wise. A rule of thumb is that 10 to 14 days are
needed to acclimatize to hot conditions. If triathletes can conduct training in
a hot environment, doing an aerobic workout in the heat is wiser than doing
a track workout. In all cases, triathletes should expect pace to decrease in any
hot conditions, even after acclimatization.
If a triathlete lives in a cool location but is preparing for a hot race, wear-
ing extra clothing for workouts to stimulate sweating is a proven tactic. But
triathletes should not wear rubber sweat suits or other clothing that is not
breathable.
Early spring and late fall races may be downright cold. Sipping on a hot
beverage from a thermos before the swim can help triathletes keep warm at
the start line and through a sprint race. If a triathlete gets cold easily, slipping
on a jacket for the bike ride is worth the extra time.
If the triathlete is racing at altitude but lives at sea level, arriving at altitude
45 hours or more before the race will help performance in endurance events,
according to at least one researcher.3

Conclusion
Although a relatively new triathlete and a highly experienced triathlete differ
in many ways, the primary differences are whether the focus is on improv-
ing technique or speed, how much overall training volume is done, and how
much of the volume is performed at higher intensities.
The best triathletes customize their training plans to fit their particular
needs. New triathletes are at risk of injury or illness if they use a training
plan built for highly experienced triathletes. Highly experienced triathletes
who use a training plan that does not challenge them physically or mentally
will not improve. The recommended approach is to determine individual
needs and then plan accordingly.
CHAPTER
29
Olympic
—Gale Bernhardt

T he Olympic- or international-distance triathlon standard is a 1.5-kilometer


swim, a 40-kilometer bike ride, and a 10-kilometer run. The distances
might be familiar to some single-sport triathletes. The 1,500-meter swim is
a classic long-distance swim event. In cycling, performance in a 40K time
trial is measured and compared by triathletes around the world. Of course,
10K events on and off the track are popular for runners. All three events are
individual sports in the Olympics, so the name “Olympic” came to be used
for this distance.
The sport of triathlon made its debut as an Olympic sport at the 2000
Games in Sydney, Australia. Professional triathletes representing their
countries’ colors race at venues around the globe, competing for valuable
ranking points. Those points make them eligible for a spot on the starting
line of the Olympic Games. Currently, the number of competitors is limited
to 55 males and 55 females, and no country may have more than 3 triathletes
on the starting line.
Although most age-group or nonprofessional triathletes race in a non-
drafting format, the pros race in a draft-legal format, meaning that they can
ride close to other riders on the bicycle leg, using the aerodynamic benefit
of the group to their advantage. Race venues tend to be spectator-friendly
multilap courses. Triathletes must be strong swimmers to make it into the
early or bigger cycling packs. Triathletes in the first pack usually attempt to
stay away from the main pack during the race, while those in chase packs
aim to catch up. Whether triathletes are in the lead pack or the chase pack,
they need to meter their work efforts so that they can perform a strong run
after the bike ride.
Obviously, the training and racing tactics for a draft-legal format are dif-
ferent from those for a nondrafting event. This chapter focuses on training
and race tactics for nondrafting events.
Race results of Olympic-distance events show that faster racers on flat
courses turn in total finish times near the 2-hour mark. Other racers on the
same course may be turning in finish times near the 5-hour mark. Because

399
400 } Bernhardt

of the longer course, the spread in finish times is greater than it is in sprint-
distance races.

Exertion Levels for Training and Racing


As the race distance increases, relative individual race speed decreases.
The fastest average pace that any individual triathlete can sustain for an
Olympic-distance event will be faster than the pace that he can achieve for a
half-Ironman race yet slower than the pace that he will perform for a sprint-
distance race.
Triathletes unfamiliar with training and racing intensities can review the
information in table 29.1.

Olympic Training
The primary concern of most triathletes is comfortable completion of the
event. They want to build enough endurance so that they can keep moving
for 3 1/2 to 4 1/2 hours. For example, if the triathlete can swim at a pace of 2
minutes and 45 seconds per 100 meters, swim time will be 41 minutes. If she
can ride 24.8 miles (39.9 km) at 15 miles per hour (24.1 km/h), her bike time
will be about 1:40. By finishing the event at a 12-minute-per-mile (7:27/km)
pace in the run, she will have a run split time of 1:15. Adding the individual
sport times and including some transition time, her total race time will be
3:45 to 4:00.1

Enhancing Training
Serious Olympic-distance racers should have a base level of zone 1 to 3 (see
table 29.1) fitness in swimming, cycling, and running 12 to 20 weeks before
race day. The main difference between the sprint and the Olympic triathlete
is volume of training time. The Olympic triathlete has more time available
to train and recover from training than the sprint triathlete does.
Notice the point about recovery. To get faster, triathletes need to include
more intensity in training and stay healthy. To do this, they need adequate
recovery time. Some triathletes do well by increasing training time, intensity,
or both for 3 or 4 weeks in a row before they take a recovery week. Recovery
weeks typically include a 20 to 50 percent reduction in training volume but
maintain some intensity in several workouts.2
Other triathletes see better progress when they build volume, intensity, or
some combination of both over 2 weeks and then take a recovery week. The
person who does the most training or the most intensity isn’t the winner.
For swimming, the Olympic-distance triathlete aims to improve 1,500-
meter pace or reduce the energy required to swim 1,500 meters. If he doesn’t
already know his pace for a 1,500-meter swim, a broken 900 meters is a good
Table 29.1 Reference Scale for Rating of Perceived Exertion and Training Zones
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
Rating of perceived 6 to 9 10 to 12 13 or 14 15 or 16 17 18 or 19 20
exertion (RPE;
original Borg scale)
Rating of perceived 1 or 2 3 or 4 4 or 5 6 7 8 or 9 10
exertion (RPE; new
Borg scale)
Swim pace Work on form, no T-pace + 10 s T-pace + 5 s T-pace T-pace T-pace − 5 s per 100 As fast as
clock watching per 100 per 100 possible
Percentage of 55 or less 56 to 75 76 to 90 91 to 99 100 to 105 106 to 120 121 or more
lactate threshold
power
Percentage of 80 or less 81 to 88 89 to 93 94 to 99 100 to 102 103 to 105 106 or more
lactate threshold
heart rate (bike)
Percentage of 84 or less 85 to 91 92 to 95 96 to 99 100 to 102 103 to 106 107 or more
lactate threshold
heart rate (run)
Breathing and Gentle rhythmic Breathing rate Aware of Starting to breathe Breathing is deep and Heavy, labored Maximal
perceived exertion breathing. Pace is and pace increase breathing a little hard. Pace is fast forceful. Many notice breathing. Pace exertion in
using runners as easy and relaxed. slightly. Many harder. Pace is and beginning a second significant is noticeably breathing.
the example Intensity is a jog notice a change moderate. Holding to become change in breathing challenging but Pace is
or trot. with slightly deeper a conversation is uncomfortable, pattern. Pace is all- sustainable for 15 to sprinting effort,
breathing, although more difficult. approaching race out sustainable for 1 30 minutes at race causing high
still comfortable. pace for all-out 1- to 1.5 hours. Mental effort. Discomfort is discomfort
Running to 1.5-hour run. focus is required, high but manageable. that is
pace remains pace is moderately unsustainable
comfortable and uncomfortable, and for more than 1
conversation conversation is minute.
possible. undesirable.
Purpose and Easy, aerobic, Aerobic, extensive Tempo, intensive Subthreshold, Lactate threshold Aerobic capacity, Anaerobic
cross-reference of recovery. endurance, endurance. muscular endurance, anaerobic speed endurance, capacity,
terms commonly aerobic threshold Ironman-distance endurance, threshold endurance, anaerobic power.
used to describe endurance race pace for threshold superthreshold, endurance.
each zone (Note: Some intermediate and endurance, muscular endurance. Sprint-distance race
coaches call this advanced athletes anaerobic International-distance pace is typically
region lactate is typically within threshold race pace is typically in zones 4 to 5b.
threshold.) zones 1 to 3. endurance. in zone 4 or 5a for Experienced athletes
advanced athletes. may be in zone 5c for
limited time.
Adapted, with permission, from G. Bernhardt, 2007, Training plans for multisport athletes (Boulder, CO: VeloPress), 22-23, and G. Bernhardt, 2009, Training plans for cyclists

401
(Boulder, CO: VeloPress), 28-29.
402 } Bernhardt

place to start. To do this, the triathlete swims 3 × 300 meters fast, recover-
ing for 30 seconds after each swim. He watches the clock and gets a time for
each 300 meters, aiming to have a difference of less than 15 seconds among
the efforts. The average time for the three swims is the average 300-meter
pace. That number divided by three determines the threshold pace (T-pace)
per 100 meters.3
To work on endurance for the 1,500 meters, the triathlete should swim
broken sets of 1,000 to 1,500 meters using T-pace and keeping recovery
between swims at a minimum. For example, he can swim 10 to 12 × 100 meters
on a send-off of T-pace plus 10 seconds. He should aim to swim the first half
of the set at T-pace and the last half of the set at T-pace minus about 2 seconds.
For cycling and running, the sprint triathlete can include zone 4 to zone
5a intervals in training that total 20 to 30 minutes (table 29.1). Olympic tri-
athletes can build this threshold training intensity to total 40 to 60 minutes
in both cycling and running. The amount of time accumulated depends on
the individual triathlete, recovery, and time limitations.
Most Olympic-distance triathletes do not have access to power meters.
Bike training will have to be done using heart rate and perceived exertion.
Running, however, is a different story. Nearly everyone has access to a
track. Triathletes can use running pace from a recent 10K event to design
intervals based on pace and distance, rather than heart rate and interval
duration.
Unless doing a bike-to-run workout, a brick, the triathlete should use
an open 10K running pace for run workouts. For a brick workout, the pace
should be based on running performance in a race. Triathlon 10K paces are
often 10 to 20 seconds per mile (6 to 12 seconds per kilometer) slower than
open or standalone 10K runs.
If the goal race or races include hills, the triathlete should complete some
of the threshold training with hill repeats or by riding a hilly course. The
triathlete who lives in a flat area can simulate hills by riding into a stiff head-
wind, using a tougher gear than normal, or doing some combination of both.
Triathletes can consider doing some training on the racecourse if access
is available.

Approach of Advanced Triathletes


The advanced triathlete is experienced and looking to maximize performance.
In most cases, she is aiming for a podium spot or, at minimum, improving
her placement in the overall field of racers. Competing at races that include
a high-quality field is desirable.
Advanced racers may be looking for a position on their country’s team to
travel to the International Triathlon Union (ITU) World Championships. Each
country has a slightly different qualification process. To find the process for
their country, triathletes can visit the website of their national federation. A
list of the world federations can be found on the ITU website.4
Olympic | 403

Whether they are looking to compete on the world stage or not, advanced
triathletes are interested in performance, specifically, triathlon performance.
They are competing 24 weeks a year or more. They may participate in
other sports such as cyclo-cross or Nordic skiing or compete in swimming,
cycling, or running single-sport events, but these activities are intended only
to complement the triathlon season. These triathletes are serious about the
sport of triathlon.
Racing at top age-group levels outside of local, small-town events requires
three to four workout sessions per discipline per week. On some days, there-
fore, triathletes will do two workout sessions per day.
Optimizing personal performance in the sport requires analyzing triathlon
performance in comparison with the competition. Triathletes should look at
results and note where they fall in relation to others:
• Where does swim time rank? Is the triathlete the slowest or the fastest
swimmer in the group? By how much time? Seconds or several minutes?
Does rank change depending on the course (pool, lake, or ocean swim)?
• Where does bike time rank? Do the same evaluation for cycling. Is the
triathlete at the top of the category or at the bottom? Does rank change
if the course is flat, hilly, hot, cold, or windy?
• Where does run time rank? Evaluate the run rank in the same way.
• Does the triathlete’s head get in the way? Does the triathlete become
excessively nervous at races? Do negative thoughts creep into his mind?
Can he visualize great performances?
• Does the triathlete have nutrition problems on race day? Does he run
low on energy? Does he sweat excessively? Does his gastrointestinal
tract cause problems?
The receding list, although not comprehensive, covers the major categories.
When coaches and triathletes are looking to optimize performance, they
need to look at all the variables that influence race-day outcomes. Identifying
opportunities for improvement allows the coach and triathlete to prioritize
training strategy.
A triathlete who is equal to the competition in cycling and running but
loses time in swimming obviously needs to improve the swim. If the swim
and run are in the top 10 percent of the category but bike time is in the lower
50 percent, cycling performance needs to be improved.
Depending on the individual needs for the triathlete’s performance
improvement, a variety of strategies can be used:
• Single-sport focus. The triathlete might consider spending the 4 to 6
months before the triathlon season training primarily on the discipline that
is holding her back. If that is the swim, she could consider swimming 5 days
per week, adding a stroke coach, swimming with a masters group, doing some
swim competitions, or doing some combination of those options. If cycling
404 } Bernhardt

is holding the triathlete back, she could consider riding with a fast group,
doing some bike racing or mountain biking, or using a power meter to train.
Or, if her running performance is inadequate, she could train for several 5K
and 10K events. In all cases, the other two disciplines take a back seat to the
developing discipline workouts.
• Conduct select workouts at much higher intensities than race effort.
Although most of the race is spent at lactate threshold effort, the triathlete
can do some training at intensities well above threshold and for extended
periods. This goal can be accomplished with structured intervals or by includ-
ing single-sport races in training. Certainly, well-placed group workouts can
be advantageous, if used wisely.
• Optimize recovery. A triathlete who is running ragged because of life
commitments may need more recovery than usual. For some triathletes, a
7-day reduced-volume recovery cycle is perfect. Other triathletes do well with
only 5 days. Still others need 10 days. Optimizing individual performance
means optimizing everything, including recovery.
• Track and analyze training data. Is the current training program pro-
ducing the desired results? How can a coach or triathlete know? What mix
of volume and intensity improves performance? What mix drives the triath-
lete to illness or injury? How long after a specific workout block should a
triathlete evaluate training success? Today’s workout does not mean instant
performance improvement tomorrow or the next day.
Success takes time and seldom occurs by accident.

Sample 12-Week Plan


The 12-week plan shown in table 29.2 gives triathletes an idea of the overall
progression of training. New triathletes can be ready to race within 12 weeks,
but they will not be capable of racing at their personal potential. More than
12 weeks are needed to build the depth of fitness necessary to race at an
advanced level.
Advanced triathletes who have years of endurance fitness built and multiple
Olympic races on their race resumes will use the final 12 weeks before the
race to push the envelope of threshold speed, looking for small improvements.

Racing Tactics
Race-day tactics will vary depending on the individual triathlete’s fitness
and previous race experience. New triathletes with limited fitness should
aim to remain aerobic during most of the race. Advanced triathletes will
push lactate threshold pace and sometimes exceed lactate threshold intensity
during the event.
Table 29.2 Sample 12-Week Training Plan for Olympic Triathlon
Approximate
training
Week hours Types of workouts Primary focus for week Other
1 3 to 7 Aerobic, technique, Work on proper technique. Those with low fitness should
neuromuscular speed. Those with limited endurance remain in zones 1 and 2.
should build aerobic Those with higher levels of
endurance. Advanced fitness can include zone 3 in
triathletes should build zone long workouts and perhaps
3 endurance. Introduce short some zone 3 intervals.
segments (30 seconds or Strength training once or
less) of accelerating speed. twice per week is optional.
2 4 to 7 Same as week 1. Same as week 1. Those with low fitness should
aim for a long ride (about 1
hour) and a long run (30 to 45
minutes). Those with higher
fitness can aim for long rides
(2 to 3 hours) and long runs
(45 to 60 minutes).
3 4 to 8 Same as week 1. Same as week 1. Same as week 2.
4 2 to 5 Keep workouts similar to Recovery. Bricks or time trials (TT) can
the previous 2 weeks but be included at the end of the
reduce training volume. week. Duration of each TT
depends on the triathlete’s
current fitness.
5 to 7 4 to 10 Aerobic, intensive Those with low fitness should Triathletes with limited fitness
endurance (for less fit continue to build long work- should build endurance
triathletes), threshold outs, aiming for long rides to and intensive endurance.
endurance (for advanced about 2 hours and long runs Advanced triathletes are at
triathletes), technique, to about 60 minutes. For ad- or near peak endurance and
neuromuscular speed. vanced triathletes, long rides can focus on building speed
are 3 hours and long runs are and endurance at threshold.
60 to 75 minutes.
8 2 to 5 Keep workouts similar to Recovery. Bricks or TTs can be included
the previous 2 weeks but at the end of the week.
reduce training volume. Duration of each TT depends
on the triathlete’s current
fitness.
9 and 4 to 8 Aerobic maintenance, Triathletes with low fitness The last high-volume training
10 intensive endurance should continue to build week is week 10.
(for less fit triathletes), long rides to approximately
threshold endurance (for 3 hours and long runs to 60
advanced triathletes), to 75 minutes. Advanced
technique, neuromuscular triathletes should focus on
speed. workouts that challenge
threshold speed, and some
may include limited aerobic
capacity intervals.
11 4 to 6 Aerobic maintenance, All triathletes should reduce Training volume this week is
intensive endurance training volume while keeping less than week 10 to allow
(for less fit triathletes), relative intensity in some recovery heading into race
threshold endurance (for workouts. day.
advanced triathletes),
technique, neuromuscular
speed.
12 2 or 3 plus Recovery, threshold Training volume decreases to Triathletes should aim to race
race endurance intervals, create a rested state on race at intensities that they have
technique, neuromuscular day. Limited intensity remains practiced in training.
speed endurance. in the plan.

405
406 } Bernhardt

Pacing
Pacing strategy begins with the starting line of the swim. If the race is in open
water and the start is a mass start (not organized in starting waves based
on seed times), triathletes need to line up in a location that suits their swim-
ming ability. Strong swimmers are in the front and closest to the buoy line.
Slower swimmers who are more timid in open water should be at the back
of the group and on the outside, away from the buoy line. Strong open-water
swimmers will swim fast and aggressively along the shortest line. Although
body contact is usually not intentional, those at the front should expect to
have body contact from other swimmers. All competitors need to recognize
that appropriate self-seeding makes the event more enjoyable for everyone.
For each discipline, triathletes should race at efforts similar to the paces
held in training.
After completing the swim, intermediate-level triathletes should run at a
mostly aerobic pace to their bikes. Depending on their depth of fitness and
race experience, these triathletes should ride the first half of the bike course
at mostly zone 1 to 2 or zone 2 to 3 intensity. At the halfway point, they can
increase speed to include zone 3 to 5a, holding the higher intensity until a
mile or so before transition. Heading into transition, they should decrease
the force on the pedals and keep cadence high.
After a fast transition, intermediate triathletes should run the 10K at efforts
similar to those used in the bike ride and similar to paces used in training.
Serious triathletes’ race effort quickly goes to zone 3 or higher and is seldom
below zone 3 for the entire event. Much of the time in each discipline is spent
at lactate threshold; triathletes call this redlining the entire race. They will
perform at anaerobic intensity at times to crest a hill, shake a competitor, or
keep a competitor in sight.
All triathletes should begin the run at a slower pace and plan to finish
at a faster pace. They need to avoid getting caught up in the frenzy of the
transition area and starting the run at an unreasonably fast pace. The tactic
of starting out slower often leads to a negative split, meaning that the second
half of the run is faster than the first. Negative splits have proved to produce
faster run performances in both elite5 and highly trained triathletes.6 A good
place to start is to run the first kilometer at a pace roughly 5 percent slower
than open 10K pace.
Triathletes who are aiming to be on the podium or finally slip past the
rival who has gone faster in previous races must be willing to pay the uncom-
fortable price of high speed. Mental toughness and the ability to tolerate
the discomfort of high paces are essential to success at high levels in sport,
including this one.
Olympic | 407

Weather
As race distances get longer, weather plays a more prominent role in perfor-
mance. When preparing for the event, triathletes should research weather
patterns for the time of the year of the race. Early spring and late fall races
can be cold. Sipping on a hot beverage from a carafe before the swim can
help triathletes keep warm at the starting line. If a triathlete gets cold easily,
slipping on a jacket for the bike ride is worth the extra time.
Midsummer events may include heat and humidity that can slow race pace
significantly. The body uses a great deal of energy to cool down, particularly
when triathletes attempt to produce high paces in hot, humid conditions. The
thermal stress takes a toll, and triathletes are forced to slow down.
If a triathlete expects to compete in hot and humid conditions on race day,
doing some acclimatization is wise. A rule of thumb is that 10 to 14 days are
needed to acclimatize to hot conditions. If triathletes can conduct training in
a hot environment, doing aerobic workouts is wiser than doing track work-
outs. In all cases, they should expect pace to decrease in any hot conditions,
even after acclimatization.
If a triathlete lives in a cool location but is preparing for a hot race, wearing
extra clothing for workouts to stimulate sweating is a proven tactic. But wear-
ing rubber sweat suits or other clothing that is not breathable is not advised.

Altitude
If the triathlete is racing at altitude but lives at sea level, arriving at altitude
45 hours or more before the race will help performance in endurance events,
according to at least one researcher.7

Conclusion
As race distance increases, the need for endurance conditioning increases. The
first step is to perform long workouts and train consistently. As triathletes gain
fitness and race experience, training volume can be held relatively constant.
More advanced racers can emphasize muscular endurance, lactate threshold
endurance, and even aerobic capacity.
As triathletes move to faster racing, the commitment to the sport of triathlon
increases and so does the level of tactical training. Successful racing at the
top levels is a direct result of carefully planned training.
This page intentionally left blank.
CHAPTER
30
Half-Ironman
—Gale Bernhardt

H alf-Ironman-distance events are typically composed of a 1.2-mile (1.9


km) swim, a 56-mile (90.1 km) bike ride, and a 13.1-mile (21.1 km) run.
Added together, the distances of the individual disciplines come to 70.3 miles
(113.1 km).
Covered more in depth in chapter 31, the World Triathlon Corporation
(WTC) owns the Ironman brand and trademark. Any race director may host
a half-Ironman-distance event. The races that carry the “Ironman 70.3” title
are WTC-owned and branded half-Ironman-distance events.
Depending on the racecourse and race-day conditions, professional men
may race a half-Ironman event in less than 4 hours and professional women
may finish in less than 4 1/2 hours. Top age-group triathletes often come in
under the 5-hour mark.
If courses become more difficult because of terrain or weather, race times
become slower. A sampling of average race times for 35 races around the
world yields a range from 5:09 to 6:33.1
A difference of 1 1/2 hours in average finish time is a big spread. When
estimating race-day finish times in each discipline, the estimate should be
based on current, proven athletic ability, not wished-for athletic ability.

Exertion Levels for Training and Racing


Experienced racers know that as the race distance increases, the pace in each
discipline decreases. Triathletes unfamiliar with training and racing intensi-
ties can review table 30.1.

Half-Ironman Training
Having successfully raced one or two half-Ironman events, the intermediate
triathlete is looking to improve time or placement in the field. If the triathlete
races the same event from one year to the next, determining whether training
has indeed had a positive effect on race results is easy. If the triathlete decides

409
410
Table 30.1 Reference Scale for Rating of Perceived Exertion and Training Zones
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
Rating of 6 to 9 10 to 12 13 or 14 15 or 16 17 18 or 19 20
perceived exertion
(RPE; original
Borg scale)
Rating of 1 or 2 3 or 4 4 or 5 6 7 8 or 9 10
perceived exertion
(RPE; new Borg
scale)
Swim pace Work on form, no T-pace + 10 s T-pace + 5 s T-pace T-pace T-pace − 5 s per 100 As fast as
clock watching per 100 per 100 possible
Percentage of 55 or less 56 to 75 76 to 90 91 to 99 100 to 105 106 to 120 121 or more
lactate threshold
power
Percentage of 80 or less 81 to 88 89 to 93 94 to 99 100 to 102 103 to 105 106 or more
lactate threshold
heart rate (bike)
Percentage of 84 or less 85 to 91 92 to 95 96 to 99 100 to 102 103 to 106 107 or more
lactate threshold
heart rate (run)
Breathing and Gentle rhythmic Breathing rate and Aware of Starting to Breathing is deep and Heavy, labored Maximal
perceived exertion breathing. Pace pace increase slight- breathing a little breathe hard. forceful. Many notice a breathing. Pace exertion in
using runners as is easy and ly. Many notice a harder. Pace is Pace is fast second significant change is noticeably breathing.
the example relaxed. Intensity change with slightly moderate. Holding and beginning in breathing pattern. Pace challenging but Pace is
is a jog or trot. deeper breathing, al- a conversation is to become is all-out sustainable for sustainable for 15 to sprinting effort,
though still comfort- more difficult. uncomfortable, 1 to 1.5 hours. Mental 30 minutes at race causing high
able. Running pace approaching race focus is required, pace effort. Discomfort is discomfort that
remains comfortable pace for all-out 1- is moderately uncomfort- high but manageable. is unsustainable
and conversation to 1.5-hour run. able, and conversation is for more than 1
possible. undesirable. minute.
Purpose and Easy, aerobic, Aerobic, extensive Tempo, intensive Subthreshold, Lactate threshold Aerobic capacity, Anaerobic
cross-reference of recovery. endurance, aerobic endurance. muscular endurance, anaerobic speed endurance, capacity, power.
terms commonly threshold endurance Ironman-distance endurance, threshold endurance, anaerobic endurance.
used to describe (Note: Some race pace for threshold superthreshold, muscular Sprint-distance race
each zone coaches call this intermediate and endurance, endurance. pace is typically in
region lactate advanced athletes anaerobic International-distance zones 4 to 5b. Ex-
threshold.) is typically within threshold race pace is typically perienced athletes
zones 1 to 3. endurance. in zone 4 or 5a for may be in zone 5c for
advanced athletes. limited time.
Adapted, with permission, from G. Bernhardt, 2007, Training plans for multisport athletes (Boulder, CO: VeloPress), 22-23, and G. Bernhardt, 2009, Training plans for cyclists (Boulder,
CO: VeloPress), 28-29.
Half-Ironman | 411

to race a different event, a measure of success is improved placement within


category or percentage of field.
For example, if the triathlete placed in the bottom third of his category in
his first half-Ironman race and later placed at the 50 percent mark or higher,
his performance improved. Because of course difficulty, as mentioned previ-
ously, time at the second event might be slower than time at the first event.
Although new half-Ironman triathletes are completing some 50 to 100
percent of predicted race time within a 2-day training block, the intermediate
triathlete is completing 90 to 110 percent of predicted race time in a 2-day block.

Training Plan for Intermediate Triathletes


The intermediate triathlete’s biggest training week might look something
like this:
Monday: Strength training or swim or bike for 60 minutes
Tuesday: Swim 60 minutes, run 45 minutes
Wednesday: Bike 60 to 90 minutes
Thursday: Swim 45 to 60 minutes, run 45 minutes
Friday: Day off or run 30 minutes
Saturday: Swim 45 to 60 minutes, bike 4 hours
Sunday: Run 90 to 120 minutes
Total training time: 11 1/2 to 13 1/2 hours
For the intermediate triathlete, both weekly volume and weekly intensity
increase. Depending on individual ability and race limiters, the intermediate
triathlete should begin by scheduling around 20 percent of the weekly train-
ing volume at zone 3 intensity (see table 30.1). For example, if the beginner
triathlete trained 10 hours, 2 hours of that training time would be at zone 3
intensity. The 13 hours of intermediate triathlete training would have around
2 hours and 40 minutes of zone 3 intensity at the 20 percent load.
A second strategy for the intermediate triathlete is to keep weekly training
hours lower but include more intensity. Although some triathletes might be
able to handle 30 percent of weekly training volume at zone 3, others risk ill-
ness or injury. Constantly monitoring training load and triathlete response is
critical to success. The aim should not be to include the most intensity; rather,
the goal should be to include the least amount of volume and intensity pos-
sible to bring about continuous improvement.

Approach of Advanced Triathletes


Advanced triathletes have successfully completed a minimum of two half-
Ironman races, and their preparation for a key race consumes a minimum
of 6 months. These triathletes may complete multiple half-Ironman events
in a single season of racing.
412 } Bernhardt

Goals for advanced triathletes vary. Some may be seeking to stand on the
podium or qualify for a championship race. Others are looking to optimize
personal performance, not necessarily to win any awards. Whatever their
specific goals, most advanced triathletes are data driven.
These triathletes review personal race performances from past triathlons.
Swim, bike, and run paces are collected for baseline performances. Transi-
tion times are reviewed from each event and compared with the best per-
formances in the race. The goal of training is to improve race performance
from baseline numbers.
The triathlete who is looking for a podium performance or championship
qualification is more likely to select specific race venues and travel to events.
For example, ocean swims or non-wetsuit swims favor strong swimmers.
Hilly bike courses favor strong cyclists and discourage drafting. Some tri-
athletes may prefer a hilly ride followed by a flat run, whereas others excel
if both the bike and run courses are hilly.
Advanced triathletes use course selection in training. They review course
profiles and replicate sections of the course in training.

Lactate Threshold Training


One training strategy is to do a lactate threshold training block, keeping
training volume relatively low, before beginning to build event-specific
endurance. If event-specific endurance begins building 12 to 15 weeks before
race day, the lactate threshold training block is implemented 18 to 23 weeks
before race day. The goal is to raise lactate threshold pace and power to give
the triathlete a bigger aerobic engine.
The lactate threshold training block may include Olympic-distance races.
When the triathlete is within the 12 to 15 weeks of race day, she may include
Olympic-distance races at the end of a reduced-volume training week. These
relatively shorter races help maintain speed.
Compared with the intermediate racer, the advanced racer devotes more
time to triathlon training. These triathletes also have the capability to include
more intensity in training. More intensity comes in the form of doing both
a higher percentage of training load at higher intensities and more training
at intensities above zone 3.

Training Plan for Advanced Triathletes


The golden question is, “What is the optimal percentage of intensity?” To
date, no research has quantified the optimal load for triathletes. A study by
Seiler and Espen examined the training load of elite cyclists and runners
and concluded that roughly an 80:20 ratio of low- to high-intensity training
yielded excellent long-term results in triathletes who train daily.2
Analysis becomes more difficult when a triathlete is training in three dis-
ciplines. A reasonable approach is to begin with an intensity load near the
20 percent mark and make adjustments from there.
Half-Ironman | 413

If an intermediate triathlete’s biggest training week is 11 to 13 hours, the


advanced triathlete’s load will be slightly more. An advanced triathlete’s
training week might be the following:3
Monday: Strength training or swim or bike for 60 minutes
Tuesday: Swim 60 minutes, run 60 minutes
Wednesday: Bike 75 to 90 minutes
Thursday: Swim 60 minutes, run 30 to 45 minutes
Friday: Bike 75 minutes
Saturday: Swim 60 to 90 minutes, run 2 hours
Sunday: Bike 5 hours
Total training time: 15 to 16 hours
If both intermediate and advanced triathletes are doing an intensity load of
20 percent of training volume, the advanced triathlete is doing more training
hours at higher intensity simply because the overall load is larger.
The process is not quite as simple as assigning a 20 percent intensity load
to each discipline. The triathlete must analyze strengths and performance
limiters to decide how best to apply an intensity load.
Among elite swimmers, cyclists, and runners, cyclists and swimmers
accumulate more weekly training volume than runners do. Although swim-
mers commonly perform two 2-hour sessions per day and cyclists regularly
complete 5-hour training rides, neither of these training loads (hours of train-
ing sessions) typically happens in running. Running poses more injury risk
from loads of high volume and high intensity.
The challenge is to find the appropriate blend of volume, frequency, inten-
sity, and the other training parameters. Only after considering the triathlete’s
discipline history, injury history, and race goals can the optimal volume and
intensity load, per discipline, be planned.

Sample 12-Week plan


The 12-week training plan (table 30.2) displays an overall roadmap for train-
ing progression for intermediate to advanced half-Ironman racers. These
triathletes have a deep base of fitness when they reach the final few weeks
before race day. Before the final 12 weeks preceding race day, these triathletes
are already swimming at least 1.2 miles (1.9 km) once or twice per week.
They are riding 3 or 4 days per week and doing a long ride of about 2 hours.
They are running 3 or 4 days per week, including a long run of 60 to 90
minutes.
This base of fitness allows intermediate and advanced triathletes to push
the intensity of workouts in the final 3 to 4 months before the race. Pushing
intensive and lactate threshold endurance intensities before a solid aerobic
base has been built risks illness and injury.
Table 30.2 Sample 12-Week Training Plan for Half-Ironman Triathlon
Approximate Types of Primary focus
Week training hours workouts for week Other
1 7 to 10 Aerobic, The primary focus is The most important aspect of
technique, building endurance half-Ironman racing is building
intensive for event completion. event and discipline endurance.
endurance. The secondary focus Advanced triathletes should
is adding workouts include some threshold endurance
with zone 3 intensity. in workouts.
2 7 to 10 Same as week Same as week 1. Most triathletes adapt well
1. with two to four key workouts
(breakthrough, or hard, workouts)
each week. Depending on
individual race limiters, key
workouts may be in only one
discipline or spread among all
disciplines.
3 7 to 10 Same as week Same as week 1. Although the first 4 weeks of this
1. plan show 2 weeks of high volume
followed by 1 week of recovery
(3:1), some athletes do better on a
format of 2:1 or even 1:1.
4 6 to 8 Keep workouts Recovery. Bricks or time trials (TT) can be
similar to the included at the end of the week.
previous 2 Although important, lightning-fast
weeks but transitions are not as important for
reduce training half-Ironman racing as for sprint-
volume. and Olympic-distance racing.
5 8 to 12 Aerobic, The primary focus Strong swimmers can often
technique, remains building long swim 2 days per week and have
intensive workout endurance. a successful race. Advanced
endurance. By the end of next racers may swim 4 or 5 days per
week, the long swim week. Intermediate cyclists should
should be a minimum be increasing the distance and
of 1.2 miles (1.9 km). volume of zone 3 intensity in long
The long ride should rides. Advanced athletes should
be 3 to 4 hours. The be including some zone 4 to 5a
long run is 1 1/2 to 2 work in a short interval ride or
hours. perhaps within the long ride.
6 8 to 12 Same as week Same as week 5. Although some fatigue is expected
5. during big training blocks,
exhaustion must be avoided.
Nonessential workouts should
be eliminated if fatigue becomes
excessive.
7 8 to 12 Same as week Same as week 5. Individual discipline training
5. distances are near race-day
distances for advanced triathletes.
8 6 to 8 Keep workouts Recovery. Some triathletes make great
similar to the gains by reducing training volume
previous 2 for 5 days, but others need 10.
weeks but Adequate time must be allowed for
reduce training recovery while training time and
volume. intensity are optimized.

414
Half-Ironman | 415

Approximate Types of Primary focus


Week training hours workouts for week Other
9 8 to 11 Aerobic, Build final race Advanced triathletes who respond
technique, endurance. If well to high volume should
intensive endurance has been continue to build long ride time
endurance. built in previous this week and next. Run time may
weeks, the final remain constant.
volume of intensity
peaks this week.
10 8 to 11 Aerobic, Depending on If week 9 was not the peak in
technique, individual triathlete training volume, the peak should
intensive needs, training occur no later than week 10.
endurance. volume may begin
decreasing this
week.
11 6 to 10 Aerobic, Decrease training Dress rehearsal for race day
technique, volume but maintain should be completed no later than
intensive some intensity. this week.
endurance.
12 2 to 3 plus race Recovery, Training volume For intermediate and advanced
sharpening. decreases to allow a triathletes, a good amount of zone
rested state on race 3 can be included on race day.
day. Limited intensity Advanced triathletes will include
remains in the plan. zones 4 and 5a, depending on
the individual. Certainly, the entire
race cannot be executed above
zone 3.

Notice that the types of workouts remain constant throughout the plan.
Intermediate and advanced triathletes are aiming for three things: to build
race endurance, to increase aerobic speed in all disciplines, and to increase
lactate threshold heart rate so that the aerobic engine is larger.

Racing Tactics
When possible, intermediate and advanced triathletes should train on por-
tions of the course for key races. When training on the course is not possible,
these triathletes should try to see part or all of the course by driving it. After
seeing the course, the triathlete can more easily develop a mental image of
success specific to that course.

Pacing
Many top racers do not use technology when racing—no heart rate monitor,
no power meter, often not even a watch. The pace that they need to travel
416 } Bernhardt

during a race is so automatic that no technology is needed. Other racers are


most successful if they have technology to help them during the race. Some
racers use technology to limit pace early in the event. Others use technology
to aim for specific power or heart rate outputs on the bike and specific pace
and heart rate on the run.
Triathletes can decide for themselves whether to use technology to optimize
race performance. Triathletes who do not want to monitor themselves with
devices during races yet need the data can wear the devices but tape over the
display. Data can be downloaded after the event for analysis.
Triathletes who do take advantage of technology on race day should set
race-day goals based on training benchmark data. They can set a range
of times from “best race possible in perfect conditions” to “wind and bad
weather.” Setting a race-day time range rather than a specific time for each
discipline makes race times more realistic. This strategy also prevents taking
on an all-or-nothing mind-set.
Advanced racers may need a short warm-up before the event. These tri-
athletes will do little of the race at zone 1 intensity and most of it in zones 2 to
4. Although time in zone 4 does not make up most of the event, the advanced
triathlete can push some zone 4 intensity. Obviously, racing an entire 6-hour
event at lactate threshold is not possible, so using some discretion is important.

Weather
Average half-Ironman triathletes will be racing for 5 to 6 1/2 hours. Over this
length of time, changing weather can play a significant role in race time. Tri-
athletes need to be mentally prepared to adjust race-time goals if conditions
become hot or windy. They must be willing to slow their pace to optimize
race-day performance.
If a triathlete expects to race in hot and humid conditions on race day,
doing some acclimatization is wise. A rule of thumb is that 10 to 14 days are
needed to acclimatize to hot weather. If triathletes can conduct training in
a hot environment, doing aerobic workouts in the heat is wiser than doing
track workouts. In all cases, triathletes should expect pace to decrease in hot
conditions, even after acclimatization.
If a triathlete lives in a cool location but is preparing for a hot race, wearing
extra clothing for workouts to stimulate sweating is a proven tactic. Wearing
rubber sweat suits or other clothing that is not breathable is not advised.
On the other end of the weather spectrum, early spring and late fall races
can be cold. Sipping on a hot beverage from a carafe before the swim can
help triathletes keep warm at the starting line. If a triathlete gets cold easily,
slipping on arm warmers and a vest or a jacket for the bike ride may be worth
the extra time.
Half-Ironman | 417

Altitude
If the triathlete lives at sea level but is looking to race at altitude, arriving at
altitude 45 hours or more before the race will help performance in endurance
events, according at least one researcher.4

Conclusion
Racing the half-Ironman distance requires more training volume than racing
the Olympic-distance event. Controlling pace early in the race is critical for
success. Monitoring nutrition and hydration during this long event is also
more important than it is for shorter races.
Practicing pacing, nutrition, and hydration plans during long training
events creates a greater likelihood of success and reduces stress on race day.
This page intentionally left blank.
CHAPTER
31
Ironman
—Gale Bernhardt

F or some triathletes, Ironman-distance events are the pinnacle of the sport.


Certainly, television coverage has helped make these events extremely
popular. The human-interest stories featured throughout the program give
many a reason to be inspired.
An Ironman event is composed of a 2.4-mile (3.9 km) swim, a 112-mile
(180.2 km) bike ride, and a 26.2-mile (42.2 km) run. The name “Ironman” often
refers to the championship race held in Hawaii each year, but it may refer to
Ironman-branded races. The word “Ironman” and the famous M-dot logo are
trademarks owned by the World Triathlon Corporation (WTC), a privately
owned company. Races that are Ironman-distance events may not carry the
word “Ironman” in the title or any advertising that might give triathletes the
impression that these events are associated with WTC.
Ironman-branded events are the only events that are qualifying races for
the world championship event in Hawaii. Triathletes may also gain entry into
the world championship event through a lottery selection process.
The event distance of 140.6 miles (226.3 km) yields average finish times of
12 1/2 to 13 hours.1 Racers must complete the event in less than 17 hours to
be considered official finishers.
Because of the event distances and logistics, most Ironman-distance events
have cutoff times for each segment of the race. In nearly all events, racers must
complete the swim in less than 2 hours and 20 minutes. For events beginning
at 7:00 a.m., the bike ride must be finished by 5:30 p.m. The marathon must
be completed by midnight.
If triathletes are unable to meet the cutoff times, their race will end before
they complete the full distance.

Exertion Levels for Training and Racing


For any individual triathlete, Ironman-distance race pace is slower than the
pace used for the shorter race distances covered in chapters 28 through 30. To
hone in on the right intensity to optimize training, triathletes can use table
31.1.

419
420
Table 31.1 Reference Scale for Rating of Perceived Exertion and Training Zones
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
Rating of per- 6 to 9 10 to 12 13 or 14 15 or 16 17 18 or 19 20
ceived exertion
(RPE; original Borg
scale)
Rating of 1 or 2 3 or 4 4 or 5 6 7 8 or 9 10
perceived exertion
(RPE; new Borg
scale)
Swim pace Work on form, no T-pace + 10 s T-pace + 5 s T-pace T-pace T-pace − 5 s per 100 As fast as
clock watching per 100 per 100 possible
Percentage of 55 or less 56 to 75 76 to 90 91 to 99 100 to 105 106 to 120 121 or more
lactate threshold
power
Percentage of 80 or less 81 to 88 89 to 93 94 to 99 100 to 102 103 to 105 106 or more
lactate threshold
heart rate (bike)
Percentage of 84 or less 85 to 91 92 to 95 96 to 99 100 to 102 103 to 106 107 or more
lactate threshold
heart rate (run)
Breathing and Gentle rhythmic Breathing rate Aware of Starting to breathe Breathing is deep and Heavy, labored Maximal
perceived exertion breathing. Pace and pace increase breathing a little hard. Pace is fast forceful. Many notice breathing. Pace exertion in
using runners as is easy and slightly. Many harder. Pace is and beginning a second significant is noticeably breathing. Pace
the example relaxed. Intensity notice a change moderate. Holding to become change in breathing challenging but is sprinting
is a jog or trot. with slightly deeper a conversation is uncomfortable, pattern. Pace is all-out sustainable for 15 to effort, causing
breathing, although more difficult. approaching race sustainable for 1 to 1.5 30 minutes at race high discomfort
still comfortable. pace for all-out 1- hours. Mental focus effort. Discomfort is that is
Running pace re- to 1.5-hour run. is required, pace is high but manageable. unsustainable
mains comfortable moderately uncomfort- for more than 1
and conversation able, and conversation minute.
possible. is undesirable.
Purpose and Easy, aerobic, Aerobic, extensive Tempo, intensive Subthreshold, Lactate threshold Aerobic capacity, Anaerobic
cross-reference of recovery. endurance, endurance. muscular endurance, anaerobic speed endurance, capacity, power.
terms commonly aerobic threshold Ironman-distance endurance, threshold endurance, anaerobic endurance.
used to describe endurance race pace for threshold superthreshold, Sprint-distance race
each zone (Note: Some intermediate and endurance, muscular endurance. pace is typically
coaches call this advanced athletes anaerobic International-distance in zones 4 to 5b.
region lactate is typically within threshold race pace is typically Experienced athletes
threshold.) zones 1 to 3. endurance. in zone 4 or 5a for may be in zone 5c for
advanced athletes. limited time.
Adapted, with permission, from G. Bernhardt, 2007, Training plans for multisport athletes (Boulder, CO: VeloPress), 22-23, and G. Bernhardt, 2009, Training plans for cyclists
(Boulder, CO: VeloPress), 28-29.
Ironman | 421

Cross-Training Approach
for Intermediate Triathletes
The intermediate Ironman triathlete has successfully completed at least one
Ironman-distance race. The goal for the next event is to improve average pace
or placement within the field. These triathletes have been training for years
and are training specifically for long-distance racing 6 to 10 months before
an Ironman-distance event. These triathletes may or may not be building
weekly volume throughout that period, depending on their level of fitness
and performance limiters.
Six months before the race, this triathlete has begun a specific preparation
phase. Within this phase, the triathlete is doing workouts intended to improve
performance on race day.
The first approach for an intermediate triathlete is to perform weight
training, long rides of 2 to 3 hours, and 60- to 90-minute runs. Intensity is
in the form of zone 3 interval work in the specific preparation phase. Those
intervals are typically placed in shorter workouts midweek but can be placed
in a long workout on a weekend.
This triathlete should be working in the weight room to build strength and
muscular balance, eventually translating that work to power in swimming,
cycling, and running. In one or two phases of strength training the gym
work makes up two of the triathlete’s key sessions for the week. Because of
this focus, swim, bike, and run training stays mostly aerobic.
As the triathlete moves closer to race day, strength training is reduced to one
session of maintenance work per week or every other week. The other strength
days are replaced by running or cycling workouts. As the strength training
load decreases, intensity in swimming, cycling, and running increases.
This format is suitable for triathletes who live in northern latitudes and
compete in early summer races. Quality riding outdoors may be limited
because of snow and cold temperatures. Doing three or four sessions per
week on an indoor bike trainer becomes excessively boring. Replacing one or
two bike workouts with a strength training workout helps prevent boredom
and in most cases helps build strength for cycling.
For this triathlete outdoor runs may be easier, but because of slick condi-
tions, workouts with intensity above zone 3 may be a problem. In this case,
zone 3 training should be moved indoors to a treadmill.
This triathlete may replace long rides or runs with activities such as snow-
shoeing or cross-country skiing in the early phases of training. This type of
aerobic cross-training allows the triathlete to maintain aerobic base fitness
and enjoy the outdoors.
422 } Bernhardt

Lactate Threshold Approach


for Intermediate Triathletes
A second approach for intermediate triathletes is to do a block of lactate
threshold training (zone 4 to 5a) during the specific preparation phase. This
work begins 6 to 12 months before the race and lasts 3 to 6 months, allowing
enough time for a 1- to 2-week recovery period before beginning the final
12 to 14 weeks of precompetitive preparation leading to the Ironman race.
For this intermediate triathlete, the threshold training block includes
threshold intervals similar to those used by triathletes training for sprint-
and Olympic-distance races. This training block may include sprint- and
Olympic-distance races as well.
For both intermediate and advanced triathletes, the specific preparation
phase looks the same. Training volume is increasing, as is the distance or
duration of long runs and rides. Using a 4:1, 3:1, or 2:1 ratio of higher volume
weeks followed by a rest week helps keeps the triathlete healthy and promotes
adaptation from the previous training load. The best ratio is the one that
allows the triathlete to progress quickly while minimizing injury and illness.

Biggest Training Week for


the Intermediate Triathlete
Approximately 3 or 4 weeks before race day, the intermediate triathlete has
his biggest training week. Triathletes handling a bigger load and more train-
ing volume may want to taper volume for a longer period than a triathlete
with less fitness. This triathlete’s biggest week may look something like the
following:2
Monday: Strength training maintenance for 60 minutes or off day
Tuesday: Swim 90 minutes, run 60 to 75 minutes
Wednesday: Bike 60 minutes
Thursday: Swim 60 minutes, run 30 minutes
Friday: Bike 60 minutes
Saturday: Swim 1 hour (can switch Tuesday and Saturday), run 3 to 3 1/4
hours
Sunday: Bike 6 hours
Total training time: 16 to 17 1/2 hours
Notice that the long swim is placed on Tuesday. Some triathletes may
need to split long workouts to maintain quality or to meet personal schedule
requirements. Also, notice that the long run is on Saturday and the long ride
is on Sunday. Depending on the triathlete’s personal schedule and perfor-
Ironman | 423

mance limiters, these long workouts can be reversed or may be scheduled


more than 24 hours apart.
For example, triathletes may prefer to do a group ride on Saturday if they
are motivated by the group and enjoy the companionship. Another consid-
eration for workout placement is triathlete durability and injury history. Tri-
athletes who have had running injuries may be best served by doing the long
run on Saturday, Friday, or Thursday on relatively fresh legs and completing
the long ride on Sunday.

Training Strategies
Triathletes looking to place workouts more specific to the swim, bike, and
run order will want to place the long swim on Saturday morning, in open
water if possible, and ride immediately after the swim. The long run will
then come on Sunday.
As for intensity, the intermediate triathlete will spend a lot of time in zone
3. Some consider zone 3 the beginning of lactate threshold training, but others
believe that lactate threshold begins at low zone 4. Researchers found that
elite athletes spend 80 percent or more of their training time below lactate
threshold.3 Intermediate and advanced triathletes can use this 20 percent
guideline to improve performance. Because zone 3 isn’t as taxing as zone 4 to
5a, intermediate triathletes may find that they can spend more than 20 percent
of their training volume at zone 3 intensity. Accurate journaling and record
keeping is needed to monitor the training load. The more comprehensive
the data collection is, the easier it is for the triathlete to design an appropri-
ate load. The goal is to improve continuously and not lose training time to
having to recover from illness or injury.

Training for the Advanced Triathlete


The advanced triathlete has raced several Ironman events. More than likely,
this triathlete is aiming to qualify for Kona, the Ironman World Champion-
ship event. At a minimum, this triathlete is looking to optimize training to
obtain a personal best performance.
This triathlete has data collected on race pace and may have power data
as well. Although a power meter is not required, it does help make training
more accurate.
If the triathlete is looking to qualify for Kona, data analysis needs to be done
to determine what is holding the triathlete back from placing higher in her
category. Several questions need to be asked relative to others in her category.
Is the triathlete a weak swimmer?
Is the triathlete a weak cyclist?
Is the triathlete a weak runner?
424 } Bernhardt

Does the triathlete place better in her category on a hilly course?


Does the triathlete place better in her category in hot weather?
After the triathlete has identified where performance needs improvement
relative to others in the category, training to improve performance in the
weak or limiting discipline begins 6 to 10 months before the Ironman event.
Additionally, a triathlete who is attempting to qualify for Kona should select
a race based on the estimated ability to do well at on that course at that time
of year.
Advanced triathletes typically train 12 to 15 hours on a normal basis and
perform a recovery week of 9 to 10 hours during the specific preparation
phase. During this time, these triathletes may be doing workouts in zone 4
to 5a as well as a limited amount of time above 5a.
Although the rule of thumb to perform 20 percent of volume at intensity
is a place to begin, most triathletes will agree that 1 hour of zone 3 intensity
is easier on the body than 1 hour of zone 4 intensity. To date, no comprehen-
sive studies have examined training loads of advanced age-group triathletes.
Advanced triathletes may be capable of handling 30 to 40 percent of the
training load at intensities above zone 2. Generally, less than 10 percent of
the advanced triathlete’s volume is zone 4 and above.
Because advanced triathletes are training 12 to 15 hours per week in the
specific preparation phase, they may be training 15 to 20 hours per week
during the final 12 to 14 weeks before the Ironman. They will likely ride
more than one century (160 km) ride during this time, whereas intermediate
triathletes may have ridden only a single century ride in the same period.

Sample Big Training Week


for an Advanced Triathlete
The following is an example of an advanced triathlete’s big training week:2
Monday: Strength training maintenance or yoga or an easy spin on the
bike for 60 minutes
Tuesday: Swim 75 minutes, run 30 to 60 minutes
Wednesday: Bike 60 minutes, run 30 to 45 minutes
Thursday: Swim 60 minutes, run 30 minutes
Friday: Bike 1 to 2 hours
Saturday: Swim 1 1/4 to 1 1/2 hours, run 2 to 3 hours
Sunday: Bike 5 to 7 hours (some triathletes may add an additional swim
session on Friday or Sunday)
Total training time: 15 to 18 3/4 hours
Ironman | 425

It is fair to say that not all triathletes respond positively to intensity above
zone 3, but others are bike racing in addition to preparing for an Ironman
event. Some advanced triathletes have a lifestyle that allows them to train
multiple times per day—training 18 to 20 hours per week, enjoying naps, and
having recovery massages. Others do not have a lifestyle that can support this
level of training volume, or their bodies do not respond well to high volume.
In any case, the training strategy should consider the individual triath-
lete’s goals, performance limiters, lifestyle, and response to volume of train-
ing load and volume of intensity. The art of self-coaching requires constant
fine-tuning to the training plan to make adjustments for how the triathlete
is responding to the load.

Sample 13-Week Plan


The 13-week training plan (table 31.2) reveals a suggested progression for
training for intermediate to advanced Ironman racers. These triathletes have
a solid base of fitness before these final weeks. Before the final 13 weeks
before race day, these triathletes are already swimming at least 1.75 to 2.4
miles (2.8 to 4.0 km) once or twice per week. They are riding 3 or 4 days per
week, including a long ride of at least 3 hours. They are running 3 or 4 days
per week, including a long run of 1 1/4 to 1 3/4 hours.
This base of fitness allows intermediate and advanced triathletes to push
the intensity, or the volume of intensity, of workouts in the final 3 months
before the race. Some triathletes work on increasing the pace at zone 3 inten-
sity, whereas others work on increasing the volume of time tolerable at zone
3. Finally, some will attempt to work on both.
There is no single approach to improving race-day pace. Some strategies,
such as increasing discipline volume, volume of zone 3 intensity, and volume
of work done above zone 3, carry more risk for injury and illness. Optimiz-
ing pace, given individual responses to training loads, is part of the art of
assembling a training plan.

Racing Tactics
A good training plan has triathletes well prepared and rested before race day.
Often they get to the starting line with more eagerness and enthusiasm than
patience. Ironman must be raced with the self-control to meter pace through-
out the event. Holding back early in the race helps triathletes finish strong.
One way for intermediate Ironman racers to control pace is to limit effort
to zone 1 and 2 intensity for the first half of the distance in each discipline—
swim, bike, and run. Intensity can be increased to allow some zone 3 in the
second half of the distance in each discipline. The goal is not to maximize
zone 3 but to aim for a negative-split race in which the second half of the
distance traveled in each discipline is done faster than the first half.
Table 31.2 Sample 13-Week Training Plan for Ironman Triathlon
Approximate Types of Primary focus for
Week training hours workouts week Other
1 12 to 15 Aerobic, The primary focus is The most important aspect of
technique, building endurance for Ironman-distance racing is building
intensive event completion. The event and discipline endurance.
endurance. secondary focus is Advanced triathletes should include
adding workouts with threshold endurance in some
zone 3 intensity. workouts.
2 12 to 15 Same as week Same as week 1. Most triathletes adapt well with two
1. to four key workouts each week
(breakthrough, or hard, workouts).
Intermediate racers with a strong
swimming history can swim 2 days
per week and have a successful
race. Advanced racers may swim
4 or 5 days per week. Intermediate
cyclists should increase the distance
and volume of zone 3 intensity in
long rides. Advanced triathletes may
include some zone 4 to 5a work in a
short interval ride or perhaps within
the long ride.
3 12 to 15 Same as week Same as week 1. Although the first 4 weeks of this
1. plan show 2 weeks of high volume
followed by 1 week of recovery
(3:1), some triathletes do better on a
format of 2:1 or even 1:1.
4 6 to 10 Keep workouts Recovery. Bricks, time trials (TT), or Olympic-
similar to the distance races can be included
previous 2 at the end of the week. If racing
weeks but is included, recovery from the
reduce training race must be planned in the week
volume. following the race.
5 14 to 17 Aerobic, The primary focus If advanced triathletes include half-
technique, remains building long Ironman racing in the final weeks
intensive workout endurance. By heading into the Ironman event,
endurance. the end of next week, these races are often placed in
long swim workouts weeks 5 through 7.
should be around 2.4
miles (3.9 km). The long
ride should be 5 hours.
The long run should be
2 to 2 3/4 hours.
6 14 to 17 Same as week Same as week 5. Although some fatigue is expected
5. during big training blocks,
exhaustion must be avoided.
Nonessential workouts should
be eliminated if fatigue becomes
excessive.

426
Approximate Types of Primary focus for
Week training hours workouts week Other
7 6 to 10 Keep workouts Recovery. Some triathletes make great gains
similar to the by reducing training volume for 5
previous 2 days, but others need 10. Adequate
weeks but time must be allowed for recovery
reduce training while training time and intensity are
volume. optimized.
8 16 to 20 Aerobic, Build final race Advanced triathletes who respond
technique, endurance. If well to high volume should continue
intensive endurance has been to build long ride time this week and
endurance. built in previous weeks, next. Run time may remain constant.
the final volume of
intensity peaks this
week or next.
9 16 to 20 Aerobic, Build final race Advanced triathletes who respond
technique, endurance. If well to high volume should continue
intensive endurance has been to build long ride time this week and
endurance. built in previous weeks, next. Run time may remain constant.
the final volume of
intensity peaks this
week.
10 6 to 10 Keep workouts Recovery. Some triathletes make great gains
similar to the by reducing training volume for 5
previous 2 days, but others need 10. Adequate
weeks but time must be allowed for recovery
reduce training while training time and intensity are
volume. optimized.
11 14 to 16 Aerobic, Training volume begins Advanced triathletes may decrease
technique, decreasing from the training volume in each
intensive maximum loads no later discipline during different weeks
endurance. than this week. before race day.
12 9 to 12 Aerobic, Decrease training Dress rehearsal for race day
technique, volume but maintain (clothing, nutrition, hydration, race
intensive some intensity. pace) is completed no later than this
endurance. week.
13 3 to 5 plus race Recovery, Training volume For intermediate and advanced
sharpening. decreases to allow a triathletes, a good amount of zone
rested state on race 3 can be included on race day.
day. Limited intensity Advanced triathletes will include
remains in the plan. zone 4, depending on the individual.
Certainly, the entire race cannot be
executed above zone 3.

427
428 } Bernhardt

Although a negative-split pace is desirable, achieving it is not always pos-


sible on some courses. If hills come early in the ride or the run, a negative-
split effort needs to be employed. Data from long workouts will provide
race-day guidelines for heart rate, power, and pace. Triathletes know from
data collected what to expect on race day. Although selecting all the best
long workouts and making those paces the race-day goals is tempting,
smart triathletes make a range of race goals based on an optimal race and
a suboptimal race depending on weather.
As discussed in the nutrition chapters, drinking too much or too little
water during an Ironman event is a problem. Some triathletes do well allow-
ing thirst and hunger to dictate hydration and fueling. For others, a guideline
for fluid and fuel consumption rates per hour of racing works better. In either
case, triathletes need to have a plan for race day and practice it in training.
A well-prepared triathlete with a race-day pacing plan has a better chance
of successfully finishing the race than a well-prepared triathlete without one.
The advanced triathlete is keenly aware of pace and is aiming for a neg-
ative-split exertion on race day, although exertion levels are slightly higher
than they are for the intermediate triathlete. Advanced racers are aiming
to do most of the race at zone 2 to 3 intensity, and they may briefly go into
zone 4. These periods may be related to catching a competitor or surging
to drop one.
If possible, advanced triathletes aiming for a Kona slot should have spot-
ters on course to let them know where they are relative to the competition.
Triathletes can become remarkably inspired to push the pace if they know
that they are in contention for a Kona slot.
For the advanced triathlete aiming for a personal best performance, the
entire race is aimed at producing power and pace numbers achieved in
training.

Weather
The average Ironman triathlete will race for 12 1/2 to 13 hours. In this time,
weather can change and play a significant role in race time. Triathletes need to
be mentally prepared to adjust race time goals if conditions are hot or windy.
Triathletes must be willing to slow the pace to optimize race-day performance.
A triathlete who expects to race in hot and humid conditions on race day
should do some acclimatization. A rule of thumb is that 10 to 14 days are
needed to acclimatize to hot weather. If the triathlete can conduct training
in a hot environment, doing aerobic workouts in the heat is wiser than doing
track workouts. In all cases, triathletes should expect pace to decrease in any
hot conditions, even after acclimatization.
If a triathlete lives in a cool location but is preparing for a hot race, wearing
extra clothing for workouts to stimulate sweating is a proven tactic. Wearing
rubber sweat suits or other clothing that is not breathable is not advised.
Ironman | 429

On the other end of the weather spectrum, early spring and late fall races
can be cold. Sipping on a hot beverage from a carafe before the swim can
help the triathlete keep warm at the starting line. If a triathlete gets cold
easily, slipping on arm warmers and a vest or a jacket for the bike ride may
be worth the extra time.

Conclusion
Ironman training and racing become more complicated as the triathlete’s
goals become higher. Because of the volume of training required to complete
or compete in an Ironman event, close attention must always be given to the
triathlete’s response to the training load. Healthy and happy triathletes always
race better than those who are injured, ill, or burned out.
This page intentionally left blank.
CHAPTER
32
Duathlon
—Gale Bernhardt

T he sport of duathlon is a great way for run–bike specialists to shine.


Rather than beginning with a swim, duathlon begins with a run. A bike
ride comes next, and a second run completes the event. The sport of duathlon
carries some distance variability. The International Triathlon Union World
Championship event distance standards are the following:1
Sprint: 5K run, 20K bike, 2.5K run
Standard: 10K run, 40K bike, 5K run
Long distance: 10K run, 150K bike, 30K run
Some nations have national championship events that are different dis-
tances than those contested at world championships. For example, the USA
Triathlon National championship distance is often a 5K run, a 30K to 40K
bike, and a 5K run.2
For this chapter, the 10K run, 40K bike, and 5K run distance will be used
for training and racing discussion.
Table 32.1 gives some sample times from the finishers of a recent U.S.
national championship.3
Because the data come from a national championship event, the finish
times represent a faster overall average than most local events. Note some
key items in the data:
• Duathletes over the age of 70 are competing at national championships
for a slot on a world championship team.
• For top finishers, the difference between the first 5K time and the second
5K time is around 1 minute (roughly 20 seconds per mile [12 seconds per
kilometer]). The midpack to final finishers had about a 2-minute differ-
ence (close to 40 seconds per mile [25 seconds per kilometer]) between
the two run times.
• The difference in finish times between elites and top age-group athletes
is 3 to 4 minutes. The elites travel 3 to 4.7 percent faster than age-group
athletes do.

431
432 } Bernhardt

• Top age-group athletes are about 10 to 12 years older than elite athletes.
Speed loss per year is in the range of 0.3 to 0.4 percent for the top per-
formers in this data set.
For all duathletes, one training goal is to minimize the degradation of pace
between the first run and the second run.

Table 32.1 Split Times for Competitors at a 5K–35K–5K Triathlon


Pace in Pace in Pace in
minutes miles minutes
per mile 35K per hour 5K per mile Overall
5K run (min/km) T1 bike (km/h) T2 run (min/km) time
First place elite 16:15 5:15 0:30 48:20 26.9 0:33 17:03 5:30 1:22:38
male (age 28) (3:15) (43.3) (3:25)
First place elite 18:38 6:01 0:42 54:33 23.9 0:34 19:44 6:22 1:34:10
female (age 27) (3:44) (38.5) (3:57)
First place age- 16:21 5:17 0:32 51:59 25.4 0:41 17:24 5:37 1:26:14
group male (3:16) (40.9) (3:29)
(age 40)
First place age- 19:15 6:13 0:54 55:34 23.4 0:43 20:38 6:40 1:37:00
group female (3:51) (37.7) (4:08)
(age 38)
Midpack age- 20:21 6:34 0:45 1:06:00 19.2 1:10 22:48 7:21 1:50:56
group finisher (4:04) (30.9) (4:34)
(age 40)
Official final 50:07 16:10 2:37 2:00:32 10.8 2:12 52:08 16:49 3:47:34
finisher (age 74) (10:02) (17.4) (10:26)

Exertion Levels for Training and Racing


Individual duathletes can benefit by personalizing their training at the correct
intensity. To assist in optimizing training, duathletes can reference table 32.2.

Enhancing Training
Intermediate racers have completed at least one duathlon and are looking to
get faster. Although these duathletes are interested in more structured train-
ing and speed, typically they are not interested in increasing training hours.
Improved performance beyond the entry level can be accomplished in about
3 months, even when keeping training time less than 6 1/2 hours per week.4
One strategy for training is to do at least one combo (run-to-bike) or brick
(bike-to-run) workout per week. The structure of the workout can have both
disciplines performed at an aerobic pace. Aerobic workouts, those in zone 1
to 2, help duathletes relax, let them settle in to a steady pace, and allow them
to practice T1 or T2 (see table 32.2).
Table 32.2 Reference Scale for Rating of Perceived Exertion and Training Zones
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5a Zone 5b Zone 5c
Rating of 6 to 9 10 to 12 13 or 14 15 or 16 17 18 or 19 20
perceived exertion
(RPE; original Borg
scale)
Rating of 1 or 2 3 or 4 4 or 5 6 7 8 or 9 10
perceived exertion
(RPE; new Borg
scale)
Swim pace Work on form, T-pace + 10 s per 100 T-pace + 5 s per T-pace T-pace T-pace − 5 s per 100 As fast as
no clock 100 possible
watching
Percentage of 55 or less 56 to 75 76 to 90 91 to 99 100 to 105 106 to 120 121 or more
lactate threshold
power
Percentage of 80 or less 81 to 88 89 to 93 94 to 99 100 to 102 103 to 105 106 or more
lactate threshold
heart rate (bike)
Percentage of 84 or less 85 to 91 92 to 95 96 to 99 100 to 102 103 to 106 107 or more
lactate threshold
heart rate (run)
Breathing and Gentle Breathing rate and Aware of Starting to Breathing is deep and Heavy, labored breath- Maximal exer-
perceived exertion rhythmic pace increase slightly. breathing a little
breathe hard. forceful. Many notice a ing. Pace is noticeably tion in breath-
using runners as breathing. Many notice a change harder. Pace Pace is fast and second significant change challenging but sus- ing. Pace is
the example Pace is easy with slightly deeper is moderate. beginning to be- in breathing pattern. Pace is tainable for 15 to 30 sprinting effort,
and relaxed. breathing, although Holding a come uncomfort- all-out sustainable for 1 to minutes at race effort. causing high
Intensity is a still comfortable. conversation is able, approaching 1.5 hours. Mental focus is Discomfort is high but discomfort that
jog or trot. Running pace remains more difficult. race pace for required, pace is moder- manageable. is unsustain-
comfortable and all-out 1- to 1.5- ately uncomfortable, and able for more
conversation possible. hour run. conversation is undesirable. than 1 minute.
Purpose and Easy, Aerobic, extensive Tempo, intensive Subthreshold, Lactate threshold endur- Aerobic capacity, Anaerobic
cross-reference of aerobic, endurance, aerobic endurance. muscular endur- ance, anaerobic threshold speed endurance, capacity, power.
terms commonly recovery. threshold endurance Ironman-distance ance, threshold endurance, superthreshold, anaerobic endurance.
used to describe (Note: Some coaches race pace for endurance, an- muscular endurance. Sprint-distance race
each zone call this region lactate intermediate aerobic threshold International-distance race pace is typically in
threshold.) and advanced endurance. pace is typically in zone 4 zones 4 to 5b. Experi-
athletes is or 5a for advanced athletes. enced athletes may be
typically within in zone 5c for limited
zones 1 to 3. time.
Adapted, with permission, from G. Bernhardt, 2007, Training plans for multisport athletes (Boulder, CO: VeloPress), 22-23, and G. Bernhardt, 2009, Training plans for cyclists
(Boulder, CO: VeloPress), 28-29.

433
434 } Bernhardt

For the combo workouts, two more strategies include a run at threshold
pace followed by an easy ride or an easy run followed by a ride that includes
threshold work. In both cases the threshold work can be broken into intervals
or done for a steady time or distance.
When doing combo or brick workouts, intermediate duathletes should also
be working on fast transitions during some of the workouts. For workouts
that include fast transitions, duathletes may want to consider investing in
special platform pedals, described in chapter 33.
A second strategy for intermediate duathletes is to keep most of the run and
bike workouts separate, doing a brick or combo workout every 2 to 3 weeks.
For some duathletes, keeping the focus on single-discipline workouts may
yield better improvements. Duathletes may need to experiment to determine
which strategy is best.

Biggest Training Week


For the duathlete looking to minimize training and maximize results,
training around 6 hours per week will work. The biggest training week
for this duathlete 8 weeks before race day might look something like the
following:4
Monday: Off day
Tuesday: Run 30 minutes, mostly zone 1 or 2, and include pick-ups
Wednesday: Ride 45 minutes, mostly zone 1 or 2, and include skill drills
Thursday: Run 45 minutes and include threshold time totaling 15 to 20
minutes
Friday: Ride 45 minutes, mostly zone 1 or 2
Saturday: Long run for 90 minutes, mostly zone 1 or 2
Sunday: Long ride for 2 hours on a hilly course, zones 1 to 5a
Total training time: 6 1/4 hours

Additional Training Strategies


Many duathletes are familiar with doing intervals or steady efforts at lac-
tate threshold. Duathletes can also build threshold fitness by accumulating
time in zones 4 to 5a during an endurance ride and using the hills to drive
intensity. Riding hills not only improves lactate threshold but also provides
a discipline-specific strength workout. Obviously, the sample week does not
include a combo or brick workout.
With the remaining weeks approaching race day, volume can be held
constant during the week and the long weekend run can be reduced to 60
minutes. With volume held constant during nonrecovery weeks, threshold
intervals can be done in cycling and running to improve race pace.
Duathlon | 435

As a starting point for planning, the duathlete should observe the rule of
doing two to four key workouts per week to provide both physical and mental
stress. Second, the total volume of zone 4 to 5a work should be no more than
20 percent of total training volume.5 The plan should be conservative at first
until it is known what kind of load the duathlete can handle. When individual
training response is documented, the training load can be adjusted up or
down to achieve the best results.

Advanced Training
The advanced duathlete is experienced and is aiming for a personal best per-
formance, a place on the podium, or perhaps a spot on a world championship
team. Depending on predicted race finish time, most top age-group athletes
will be racing at all intensities.
Advanced racers have a strong base of fitness and have performed 8 to 9
weeks of lactate threshold training, including unstructured threshold train-
ing followed by specific intervals.
Threshold training is followed by 6 to 10 weeks of training that includes
anaerobic endurance and anaerobic capacity sessions. The blend of thresh-
old training and anaerobic endurance training is not an exact science and
is driven by the ability of the duathlete to respond to the training stimulus.
Like the intermediate duathlete, the advanced duathlete can approach
training in single-discipline sessions, combo workouts, or brick workouts.
The rules of thumb about doing two to four key workouts per week and limit-
ing intensity to 20 percent of volume still apply for the advanced duathlete.
Intensity for the advanced duathlete is defined as zone 4 and above.
Although the duathlete is running a 10K, doing a bike ride, and running
a 5K, the duathlete should be training to develop the overall endurance to
run a 15K. Any formulas used to calculate running paces for track workouts
can be aimed at improving both 10K and 5K running speeds.
If duathletes examine the results from past world championship events on
the International Triathlon Union website, they will find that running speed
fades only about 10 to 20 seconds per mile (6 to 12 seconds per kilometer)
from the pace of the first 10K to the pace for the final 5K of the race. Young
adult age-group athletes are fast, running paces between 5:00 and 7:00 per
mile (between 3:06 and 4:20 per kilometer).
For the lactate threshold bike training, intervals should be aimed at lactate
threshold heart rate (zone 4 to 5a) or power improvement using a 3:1 or 4:1
work-to-rest ratio along with steady efforts at threshold. Steady efforts can
be solo or repeats that are 10, 15, or 20 minutes long.
Advanced duathletes can also benefit from 3- to 5-minute intervals on the
bike in zone 5b, followed by rest intervals equal to or double the work time.
In the final 4 to 6 weeks before race day, the advanced duathlete should
include several combos, bricks, or shortened race simulation days at the
436 } Bernhardt

track or use a device that yields pace running. The advanced duathlete must
have the capability to minimize the difference between pace on the first and
second runs.

Sample 12-Week Plan


The sample 12-week plan (table 32.3) gives general suggestions for the 12
weeks of training leading into race day. The plan assumes that duathletes
have accomplished the preparatory training leading into this training block
for 6 months.

Table 32.3 Sample 12-Week Training Plan for Duathlon


Approximate
training Primary focus
Week hours Types of workouts for week Other
1 3 to 7 Aerobic, technique, Intermediate duathletes Focus key workouts
neuromuscular speed, should focus on on success limiters
lactate threshold. improving speed at until the final 6 weeks.
lactate threshold. Strength training once
Advanced athletes or twice per week is
may work on aerobic optional. Core strength
capacity in running or training a minimum
cycling. of once a week is
recommended.
2 4 to 8 Same as week 1. Same as week 1. Duathletes often run 4
to 6 days per week.
3 4 to 10 Same as week 1. Same as week 1. The number of weeks
to build intensity or
training volume before
inserting a recovery
week depends on the
individual duathlete.
Some duathletes
need only 4 or 5 days
of reduced training
volume, but others need
7 to 10 days.
4 4 to 6 Keep workouts similar to Recovery. Combos, bricks, or
the previous 2 weeks but time trials (TT) can be
reduce training volume. included at the end of
the week. Duration of
each TT depends on
the duathlete’s current
fitness.
Duathlon | 437

Approximate
training Primary focus
Week hours Types of workouts for week Other
5 to 7 6 to 12 Aerobic, technique, Intermediate duathletes Both intermediate and
neuromuscular speed, should keep the focus advanced duathletes
lactate threshold. on improving speed need the ability to run
at lactate threshold. If a fast 5K after a 10K
speed has peaked, they run and a 40K ride. At
can increase duration minimum, duathletes
at threshold speed want to minimize pace
appropriate for the race degradation between
distances. Advanced the first and second
duathletes should runs.
continue to work on
aerobic capacity.
8 4 to 6 Keep workouts similar to Recovery. Executing workouts on
the previous 2 weeks but courses similar to the
reduce training volume. key race is important.
9 to 6 to 12 Same as weeks 5 to 7. Same as weeks 5 to 7. The last high-volume
10 training week is week
10. Some duathletes
need more taper before
race day than others.
For these duathletes,
the last high-volume
week may be week 9.
11 4 to 6 Same as previous All duathletes reduce Race-day gear should
weeks. training volume be chosen no later than
compared with the this week, including
previous week while clothing and gear
keeping relative selection on the bike.
intensity included in
some workouts.
12 2 to 3 plus Recovery, short Training volume Duathletes should aim
race segments of intervals decreases significantly to race at intensities that
done at race pace or to allow a rested state have been practiced in
slightly faster. on race day. training.

Racing Tactics
Duathletes should temper pace on the 10K, aiming for lactate threshold pace.
Aiming to ride the 40K bike leg at mostly zone 3 to 5a intensity is possible for
well-trained duathletes. They should finish strong on the final run, aiming
for the final 5K to be no more than 10 to 20 seconds slower per mile (6 to 12
seconds per kilometer) than the 10K.
438 } Bernhardt

Because of the combo, brick, and race simulation training sessions, duath-
letes should be well aware of race pace and rate of perceived exertion. The
first 10K should be run at the duathlete’s self-selected race pace. Intermediate
duathletes should aim for a bike ride that is a negative-split effort by riding
the first half at zone 3 to 4 and the last half at zone 3 to 5a, spending more
time at zone 4 to 5a. Advanced duathletes will ride a good portion of the first
half of the bike at zone 4 intensity and the last half at zone 4 to 5b, keeping in
mind that the 5K must be within 10 to 20 seconds per mile (6 to 12 seconds
per kilometer) of the 10K.
In some cases, the course will determine where intensity is placed within
the ride. If the course has most of the hills in the first half, a solid amount
of intensity will be placed earlier in the ride. If course is rolling, intensity
may be placed on the uphill portions and partial recovery can occur in the
downhill sections.
In most races, the result is determined in the last half of the bike ride and
the second 5K. Advanced duathletes can target other racers, whether in the
same category or not. These targets can be used as an incentive to pass the
next racer and keep pace high.

Weather
At race distances as long as a 10K run, a 40K ride, and a 5K run, weather
can play a role in performance. When preparing for the event, duathletes
should research weather patterns for the same time of the year as the race.
Early spring and late fall races can be cold. Sipping on a hot beverage from a
carafe before the 10K can help duathletes keep warm before the race begins.
Keeping a warm-up suit on until the last moment also helps. If the race is on
a chilly day the duathlete may choose to wear tights and arm warmers or a
long-sleeved top.
If the race begins in cool conditions but temperatures are expected to
increase, some duathletes don’t mind starting the race a bit underdressed
knowing that they will warm up during the race.
Midsummer events can be contested in heat and humidity that can slow
race pace significantly. The body uses a great deal of energy to cool down,
particularly if duathletes attempt to produce high paces in hot, humid condi-
tions. Thermal stress takes a toll, so duathletes are forced to slow the pace.
If a duathlete expects to race in hot and humid conditions on race day,
doing some acclimatization is wise. A rule of thumb is that 10 to 14 days are
needed to acclimatize to hot weather. If the duathlete can conduct training
in a hot environment, doing aerobic workouts in the heat is wiser than doing
track workouts. In all cases, duathletes should expect pace to decrease in hot
conditions, even after acclimatization.
If a duathlete lives in a cool location but is preparing for a hot race, wearing
extra clothing for workouts to stimulate sweating is a proven tactic. Wearing
rubber sweat suits or other clothing that is not breathable is not advised.
Duathlon | 439

Altitude
If a duathlete is racing at altitude but living at sea level, arriving at altitude
45 hours or more before the race will help performance in endurance events,
according to at least one researcher.6

Conclusion
Duathlon is a fantastic sport for many athletes. Because the swim is not part
of the event, duathletes typically devote less time to training than triathletes
do. A good portion of the time savings comes from eliminating the commute
to the pool. Extremely efficient running and cycling can be done from the
duathlete’s own doorstep. Convenience can be a big attraction.
Duathletes who take a structured approach to training can make the most
of limited training time. They need to begin with the appropriate foundation
training to prepare muscles, tendons, and ligaments for the fast training to
be done in the final weeks before the race.
Underprepared duathletes risk injury if they decide to train at fast paces
every day of the week. This approach does not allow the body time to recover
and advance to a new level of fitness.
For athletes of any level, fast performances relative to competitors in the
same category can be achieved with strategic training and tactical racing.
This page intentionally left blank.
CHAPTER
33
Combination
Workout Training
—Gale Bernhardt

A lthough most triathlon training focus goes toward the primary disci-
plines of swimming, cycling, and running, the transitions that occur
between disciplines are also important. Two main aspects of transition need
to be considered. First is the body’s ability to change from discipline to dis-
cipline, to change from swimming to cycling and from cycling to running.
Each discipline puts different demands on the body.
The second consideration is the logistics of each transition, which includes
equipment organization, sequence of movement, and speed of transition. For
advanced triathletes racing short-course events, optimizing transition time
may mean the difference between a podium position and no podium posi-
tion. The speed of transitions is less critical in long-distance Ironman racing,
but triathletes have no reason to be sloppy.
Practicing transitions is important for triathletes who are looking to
improve performance. The workout most commonly used for this purpose
is the bike-to-run workout to practice the second triathlon transition (T2),
commonly called a brick. Triathletes can also practice the swim-to-bike
transition (T1) in workouts. This workout, along with a run-to-bike workout
for duathletes, is often called a combo.

Transition Considerations
Elite World Cup racers are the fastest at transitions because their racing suc-
cess depends on it. Although this book does not cover International Triathlon
Union (ITU)–style draft-legal racing, video clips of these racers give coaches
and triathletes an idea of what the world’s fastest transitions look like. One
can find videos on the ITU website at www.triathlon.org and by searching
the Internet.

441
442 } Bernhardt

Why are these racers so fast at T1? A fast T1 can mean the difference
between making the first bike pack, making the chase pack, or being out
alone. Chasing alone while everyone else is working within a peloton is a
significant disadvantage on the bike. Peloton riders can usually take a break
during a race, drafting while others are working. Chasing alone is a solo time
trial effort that allows no rest.

Mastering T1 Challenges
Several issues during T1 encourage triathletes to be smart. Those who are
the fittest and smartest are the fastest. The challenges begin immediately
after exiting the swim. Triathletes run to their personal T1 spot, wanting to
minimize the time spent standing in one location. They must do two things:
minimize transition time with equipment choices and organize the sequence
of activities so that any time not spent swimming, running to transition, or
riding the bike is near zero.
More tips for T1:
• Choosing a triathlon suit or swimsuit. Some triathletes complete the
entire race in a swimsuit, but most prefer a triathlon suit for speed and com-
fort. A tri suit has long legs like bike shorts, but the chamois pad is very light,
providing some comfort on the bike but soaking up negligible water during
the swim leg. Not changing from a swimsuit to cycling clothes in T1 makes
the transition faster. Long-distance racers, however, may prefer a change of
clothes because they want more comfort than a light chamois can provide
during the long period on the bike.
• Wearing a wetsuit. If the swim temperatures require a wetsuit, preparing
for T1 begins the moment when triathletes exit the water. While running to
the T1 area, they remove cap and goggles. They unzip the wetsuit and peel
it down to the waist to allow a fast stripping of the wetsuit at the individual
T1 spot. Fit and leg length of the wetsuit likely will determine the speed at
which the triathlete can remove it. To improve speed, the leg length can be
cut to be shorter, allowing larger holes for ankle and foot removal at faster
speed. Triathletes should also consider whether the length of the swim and
water temperature provide a significant advantage to wearing a wetsuit given
the possible increased T1 time.
• Wearing socks. Some short-course racers and long-distance triathletes
may want to put on socks to reduce the possibility of blisters. Faster Olympic-
distance racers nearly always ride without putting on socks because of the
time savings. Some sport-lubrication or powder products can help improve
transition times and prevent blisters if they are placed in the shoes or on the
feet before the race start.
• Clipping cycling shoes into the pedals. Running in cycling shoes in a
transition area for T1 or T2 can be awkward. Some cleats are slippery on an
Combination Workout Training | 443

asphalt transition surface. Barefoot is much safer and faster, provided the
ground is free of hazards.
• Fastening shoes to the bike frame to keep them parallel to the ground
while running to the mount or dismount line. Triathletes can use thin rubber
bands to attach the heel of the cycling shoes to the frame (see figure 33.1). This
technique keeps the cranks from turning as the triathlete runs with the bike.
The risk of turning cranks is that shoes can become hooked on the ground
and be forced out of the pedals.
• Mount the bicycle. The fastest triathletes typically use a flying mount.
They run with the bicycle and jump onto the seat while the bike is in motion.
Cyclo-cross racers also use this technique. The benefit is a fast mount and
quick acceleration to race speed. Triathletes should practice this technique
regularly to make the mount fast but smooth, without hard landings or loss
of bike control.
• Pedal with feet on top of shoes. After a fast transition, triathletes want
to get the bike moving and up to race speed as quickly as possible. To accom-
plish this, racers pedal with their feet on top of the cycling shoes until it is
convenient to slip their feet into the shoes with minimal coasting. Of course,
triathlon-specific cycling shoes with one Velcro strap are easier to fasten than
shoes with multiple straps.

„„Figure 33.1 Shoes fastened to the bike frame with rubber bands.
444 } Bernhardt

• Wear sunglasses under helmet straps. Sunglasses can be put on in tran-


sition or attached to the bike frame with light tape to be put on after the bike
is rolling at race pace. If put on in transition, sunglasses usually are stored
in the cycling helmet to keep them from being lost or knocked into traffic.
No matter when sunglasses are finally put over the eyes, placing the frame
under the helmet straps so that the helmet can be removed in T2 without
removing the sunglasses is the fastest and best process for T2.

Mastering T2 Challenges
As in T1, triathletes want to minimize any time spent in transition during
T2. By this time in the race, fatigue is beginning to accumulate. Most of the
time, the race time is over the halfway mark. Tired legs are now asked to
change from a circular pedaling motion to a stepping, running motion. This
awkward change in activity can be made more comfortable with practice.
The best time to practice T2 is during brick workouts. Several tips can help
triathletes practice T2 during brick workouts and on race day:
• Drawing near T2. As triathletes get within a quarter mile (.4 km) or so of
the transition, they pull their feet out of their cycling shoes and pedal with their
feet on top of the shoes, as they do when exiting T1. Triathletes looking for a
fast dismount can do a flying dismount like that used by cyclo-cross racers, as
opposed to a slower dismount that requires coming to a full stop and straddling
the bike frame at the dismount line. For the flying dismount, one foot stands
on the pedal as the other leg swings over the bike frame to the same side of the
bike as the support leg. The free leg leads out front for a forward first step off
the bike. The triathlete dismounts the bike at a full run from a moving bicycle.
• Wearing socks. Running without socks should be practiced in training
before doing it in a race. Hot spots often develop into blisters when running
sans socks. If the triathlete knows where the hot spots are, sport-lubrication
or powder products can be put in the foot bed of the shoe to prevent or mini-
mize blistering. Lubrication works for some triathletes, and powder products
prevent blisters for others. Triathletes should practice to see what works.
• Wearing shoes and a race belt. With running shoes stacked on top of a
hat and a race number belt, triathletes can quickly slip their feet into running
shoes if the shoes have elastic laces. Triathletes grab the race number belt
and a hat, and then begin running. Both items can be put on while running.

Making a Smooth Duathlon Transition


Many duathletes prefer to use a platform pedal attachment (see figure 33.2).
The attachment runs the entire length of a running shoe so that more of the
pedaling force can be used to propel the bicycle. A toe clip secures the shoe
to the platform. The bottom of the platform has holes drilled for cleat attach-
ment, similar to a cycling shoe. This device allows duathletes to transition
from run to bike and back to run without a shoe change.
Combination Workout Training | 445

Photo E5643_450596, Figure 33.2, ID:


450596, still to come.

„„Figure 33.2 Platform pedal attachment.

Combination Workouts
Combination workouts bring two or more disciplines into a single workout,
either for convenience or for specific race preparation. The most common
combination workouts are swim to bike, bike to run (usually called a brick),
and run to bike, depending on the goals of the triathlete and time of year.

Swim-to-Bike Workouts
A small segment of the triathlon population experiences some lightheaded-
ness when transitioning from the prone position of swimming to the standing
position of running, as triathletes do when moving from the swim to the first
transition. Another small segment of the triathlon population experiences
unusual leg fatigue going from swimming to running and then cycling.
For these triathletes, one strategy is to set up a bike on a trainer on the pool
deck.
Triathletes can begin with an easy swim of 500 meters or so and then
transition to the trainer for an easy spin of around 10 minutes. They repeat
this sequence two to four times in a single workout.
If the triathlete is not adapting or feels so lightheaded that passing out is a
possibility, a doctor should be consulted to be certain that no medical issues
446 } Bernhardt

are present. Depending on the severity of the problem, triathletes may want
to be checked out before doing any swim-to-bike workouts.
As triathletes adapt to the easy swim-to-bike workouts on the pool deck,
they should increase intensity by following a fast swim segment with an easy
ride. The second round should be an easy swim followed by a faster ride. As
adaptation to the transition between swimming and cycling continues, the
triathlete can increase the intensity of both the swim and the ride.
Many triathletes do swim-to-bike workouts as a matter of convenience,
particularly on weekends. Many do a pool workout and then head straight
to a bike workout. With workouts sequenced in this manner, they can decide
which workout or workouts should include intensity. As triathletes approach
race day, they may want a swim-to-bike workout as a dress rehearsal for
race day.

Bike-to-Run Workouts
Swim-to-bike and run-to-bike workouts are often called combination, or
combo, workouts. The bike-to-run workout is often called a brick. Although
the history of the word is not clear, one theory is that the name was given to
the workout because when triathletes go from fast cycling to running, their
legs feel like bricks.
To help triathletes adapt to the change of body movement and muscle
recruitment from cycling to running, and the feeling that this change pro-
duces, aerobic brick workouts are a good place to start. Some prefer to do
brick workouts every week throughout the training plan, but others limit
brick workouts to once per month, perhaps as a workout during a recovery
week. Others limit brick workouts to certain macrocycles. No standard has
been set about how often to perform brick workouts, and some triathletes
appear to make this adaptation better than others do.
In one study on elite international Olympic-distance racers, the intensity of
cycling did not have an adverse effect on neuromuscular control and running
economy.1 Even moderately trained triathletes experienced little influence on
running muscle recruitment after cycling.2 These studies may lead the reader
to believe that experience in the sport of triathlon eliminates any effect of
cycling on running economy and muscle recruitment, but that is not true.
A third study found that despite years of training, some elite triathletes do
experience changes in leg movement and muscle recruitment in running
after cycling.3 The effects of cycling on neuromuscular control and running
economy appear to vary among people.
When deciding how many bricks to include in a program, triathletes
should consider their experience level, goal race distance, and race results.
Slower sprint- and Olympic-distance racers are more likely to do short brick
workouts. For faster sprint- and Olympic-distance racers, brick workouts
are often in the range of 50 to 100 percent of race distance. For half-Ironman
racers, bricks are often 25 to 50 percent of race distance. For Ironman racers,
Combination Workout Training | 447

bricks become less important because the need for blazing fast transitions is
not an issue except for the top triathletes.
For Ironman racers, the benefit-to-risk considerations of long brick workouts
need to be evaluated. For example, how much value is gained from doing a
60-mile (100 km) bike ride followed by a 10- to 13-mile (16 to 20 km) run? Would
this triathlete be better served by entering a half-Ironman race and using
that race as part of the training strategy? Is the triathlete prone to running
injuries? What is expected to be gained from the brick workout? Individual
athlete strengths and weaknesses need to be considered when making train-
ing decisions. The bias should be toward conservative undertraining so that
the triathlete remains injury free and mentally sharp.
Intermediate and advanced sprint- and Olympic-distance racers often
complete brick workouts every 3 to 4 weeks. These workouts are done at the
same intensity as other workouts in the macrocycle. The intensity portion of
the brick can be structured in multiple ways:
• Aerobic ride followed by an aerobic run.
• Aerobic ride followed by a run that includes some portion at current
training-cycle intensity. This run can be a steady effort or broken into
intervals.
• Ride that includes some portion at current training-cycle intensity. This
ride can be a steady effort or broken into intervals and is followed by
an aerobic run.
• Ride followed by a run in which both disciplines include some portion
at intensity.

Run-to-Bike Workouts
Duathlon T1 is easier to practice than triathlon T1 for most triathletes. Any
yard or garage can be turned into a mock T1 area. The duathlete can go for
the assigned run, return home, complete the transition, and head out on a
bike ride.
The intensity for any run-to-bike workout should match the intensity of the
rest of the workouts in that macrocycle. As workout intensity increases with
an approaching race day, race-pace run-to-bike workouts can be included in
the mix. Examples include the following:
• Run 5 kilometers, doing the last 1.5 kilometers at race pace. Immediately
transition to an easy ride of 10 kilometers.
• Run 2.5 kilometers at aerobic intensity. Transition to a 15-kilometer
negative-split ride. Begin at aerobic intensity for 7.5 kilometers and then
ride the last 7.5 kilometers at close to race intensity. Faster duathletes can
finish at zone 3 to 5a intensity and build from zone 3 to 5b in the second
half of the ride.
448 } Bernhardt

• Run 5 kilometers, doing the last 1.5 kilometers at race intensity. Imme-
diately transition to a ride of 15 kilometers. Make the first 7.5 kilometers
at race intensity and finish at aerobic intensity.
The design of the workout should have intent for the duathlete. That intent
may be transition practice, muscle recruitment when changing disciplines
at an easy pace, or race-pace rehearsal. New and intermediate duathletes
may consider making the workout distances less than race distances. Top
duathletes may want the distances to be the same as race distances. They
may perform only a portion of the workout at race pace so that they save the
best performance for race day.

Conclusion
Whatever the level of the duathlete or the triathlete or the distance of goal
races, an organized process for transition is helpful. The multisport athletes
needs to be organized, smooth, and fast on race day. Rehearsing transitions at
race intensity before race day helps eliminate errors and increases confidence
when race day arrives.
Triathletes and duathletes may use only a few of the fast transition tips in
this chapter to optimize T1 and T2 speeds. The more competitive the racer
is, the more focus the racer will put on fast transitions.
The savvy racer will include combo and brick workouts to improve race
time, not just to do a fun workout. When including these workouts in a pro-
gram, the triathlete should be sure to have a specific purpose and goal for
the workout.
Pa r t

IX
Sports Medicine
for Triathletes
This page intentionally left blank.
CHaPTEr
34
Triathlete Body
Maintenance
and Medical Care
—John Post, MD

P icture a triathlete who has an injury or a nagging pain that’s been slowing
her down but who has an important race in the near future. How might
she continue to train so that she doesn’t worsen the pain but at the same time
can prepare properly to arrive on race day at her peak of fitness? Many tri-
athletes have done this balancing act at some point. This chapter focuses on
preventing injuries and, if injury occurs, on knowing when to pull the trigger
and involve the medical system. A triathlete who has a medical resource team
in place before an injury can often get an accurate assessment of the problem
and be back in training in a timely fashion.

Off-Season Recovery and Planning


Many annual training plans (ATPs) commence at the conclusion of the previ-
ous season with a rest period away from swim, bike, and run. This period is
essential to cleanse the palate, so to speak. Psychologically, it ends one season
and prepares the triathlete for the next, rekindling the fires of desire for what
will likely be a long haul in the near future.
This period is definitely not one of inactivity. The triathlete is encouraged
to engage in other activities and sports that he enjoys such as yoga, racquet-
ball, basketball, and so on. Such activities veer away from the drudgery of
a long swim or bike ride, thus helping to rejuvenate passion for the sport.
Many people believe that 2 to 4 weeks is adequate to refresh and recharge.
During this time the foundation for the upcoming training year can be laid
out in a structured fashion that first asks, Where am I physically? Triathletes
can use this time to assess their health and fitness (see figure 34.1). In fact,
such an assessment could possibly be one of the most important tasks of the

451
452 } Post

off-season. Like the information exchanged in the doctor’s office, the informa-
tion that the triathlete produces here is important, because many coaching
decisions and plans are based on it. Completeness is rewarded. Data should
include any previous musculoskeletal surgery; triathlon-induced injuries,
both recent and past, including cause if determined and remedy; accidents;
and other areas of possible concern.
Armed with this information, the triathlete and coach can create the
skeleton of a training plan for the upcoming year to decrease the potential
for injury.

Review of previous triathlon-related injuries


• Injuries related to a previous season or sport
• Cause of each injury if known
• Remedy for each injury
• Plan to avoid these injuries in the future

Personal habits
• Adequate sleep
• Appropriate diet
• Weight concerns
• Spousal support

Health maintenance
• Tetanus up to date
• Colonoscopy
• Flu shot
• Skin review
• Other issues

Equipment and training site safety review


• Bike fit
• New batteries in lights
• Reflective clothing
• Day and night training location and surface evaluation

Injury resource team members identified


• Personal physician
• Bike injury expert
• Running injury expert

„„Figure 34.1 Personal health and fitness assessment.


Triathlete Body Maintenance and Medical Care | 453

If the triathlete understands from day 1 that reduction in the chance of


injury is among the highest priorities, then deviation from the training plan
will be minimized. In triathlon, the high volume of repetitive training makes
it critical to use health assessments to lay a solid foundation for the season.
The serious triathlete keeps a detailed daily logbook, particularly when
it comes to aches and pains, and most particularly when it comes to change.
Often the solution to a physical problem can be found in those pages. The
problem could involve a change in running shoes, the addition of hill work,
pushing too hard on successive days, training hard when ill, altering cleat
position, or other seemingly minor changes that are the keys to unlocking the
causative factors of an injury. Coaches and others often incorrectly assume
that a triathlete possesses a basic level of understanding about injuries, a
misconception that can hinder diagnosis and recovery from an injury.
The whole point of training is the management of load, of the progressive
increase of stress in a manner that does not produce unmanageable overload.
The pages of the logbook contain clues about stress imbalance that might
lead to injury. In some settings, complaining of pain or problems is seen as
a sign of weakness. That circumstance is definitely not the case in triathlon,
in which good information leads to good decisions.
Well before this stage the triathlete has defined a series of goals for the
season, be they a physical challenge, weight loss or maintenance, fighting the
effects of aging, or any number of other objectives. And now, like an athletic
cartographer, the triathlete and coach can construct a map to triathlon suc-
cess, giving the competitor the highest probability of toeing the start line
in the first race of the season prepared for success with the lowest risk for
injury. Decisions are made carefully by keeping previous injury patterns as
part of the equation.
Triathlon can be risky business. As with any endurance sport, the triath-
lete’s level of dedication (or perhaps level of addiction, because that term
may be more accurate) may progress from fitness to overtraining to injury.
Many bypass overtraining, instead proceeding straight to injury. If the risk
is at one level for the single-sport competitor, we should consider the greater
risk that applies to the multisport athlete. Triathlon participation is not a solo
endeavor, so the triathlete’s support system, including family and workmates,
needs to be part of the analysis.
This personal health and fitness assessment can be done by triathletes of
all levels at any time during the year. They should envision it as an ongoing
process. A person who is taking a car on a long journey would perform just
such an evaluation of the tire pressure, oil status, any recent nagging noises,
and any changes in performance.
A similar review is done by the triathlete, perhaps if only on a subconscious
level. This assessment is particularly important if there has been an annoying
ache in the Achilles area, neck pain at the end of a swim workout, or worse. Part
of the triathlete’s query would include organ systems that had been a source
of difficulty in the past. What am I likely to stress now? How is my training,
454 } Post

focused on my race limiters, likely to exacerbate prior difficulties or injuries? In


short, triathletes need to make honest, unbiased decisions about their abilities.
Being honest with oneself is hard. Minimizing or ignoring facts in this
setting is incredibly easy. What active role are triathletes and coaches taking
to correct the identified issues in the personal health and fitness assessment?
Note that many of those in coaching or teaching have never learned how to
put an injury management plan in place, making it difficult for them to pass
this skill on to the triathlete. Specific guidance is required for the triathlete to
• identify limiters early in the personal health and fitness assessment,
• become educated about the symptoms and causes of the current issue,
• seek clinical care to obtain an accurate diagnosis,
• follow some type of plan to remedy the condition just as she would follow
a training plan, and
• understand that the plan comes not from the coach but from the clinical
entity.

Injury Resource Team


In Joe Friel’s The Triathlete’s Training Bible1 a good bit of time is spent defining
and setting reasonable goals. We can’t all win Hawaii. The thinking triathlete
has created a seasonal road map that will allow him to meet certain racing
performance criteria, ideally without incurring a short- or long-term injury.
Part of this plan includes triathlete conduct (and a little bit of luck) in which
measures are taken to avoid injury.
This process is continuous. The triathlete might strive to be the most care-
ful bike rider in the group, the one least likely to crash or be struck by traffic.
She is the one who pays attention to interval changes in footwear, watches the
ever-changing weather, observes a schedule of bike maintenance and repair,
and makes careful choices of when and where to ride.
But accidents and injuries happen even to the careful, prepared triathlete.
Ankles get twisted on long runs, or shoulders may ache after a tough session
in the pool. When the pain is not enough to consider altering the workout,
medical intervention is likely not warranted.
Currently, medical professionals use the visual analog pain scale, or VAS,
to quantify pain. It’s a 0 to 10 scale, in which 0 is no pain, 1 or 2 is annoying
pain, and 10 is the worst pain imaginable. If pain is a 3 or less, triathletes
would probably benefit from an alteration or reduction in training without a
visit to the medical team. But when the acuity of pain is higher and doesn’t
resolve with rest, seeking medical attention promptly is probably wise.
But if a person is in pain, where does he turn? A seasoned triathlete has a
stable of resources, much like a golf bag full of clubs. The appropriate resource
is chosen for the appropriate shot. Many consider it wise to have the follow-
ing people as a part of an inner circle.
Triathlete Body Maintenance and Medical Care | 455

• Local running shoe shop pro. He has seen more running-related com-
plaints—and come up with reasonable solutions—than any physician, and
he has a stake in the triathlete’s frequent return for more shoes and running-
related products. If the search for a solution begins here, the triathlete may
not need to seek medical attention at all. And besides, this consultation is
probably free.
• Local bike shop. This is the place where triathletes may have purchased
their bikes and other cycling gear. Many of the employees here have excep-
tional knowledge based on experience. For example, a nagging or chronic
injury is often related to a poorly fitted bike. Many shops have professional
bike fitters who can make adjustments and easily resolve such issues. The
mechanics have not only turned wrenches on bikes for years but also race
on the weekends. They will patiently answer the simplest of the triathlete’s
equipment and bike-fit questions.
• Family physician. Even triathletes get the flu and have hypertension,
diabetes, asthma, and the like. The primary care physician’s office is a great
place to keep the engine tuned up, be reminded that a tetanus booster is
needed, and much more.
• Sports physician. Many doctors include sports medicine as a small part
of their practice. You might reflexively think of an orthopedic surgeon for
those needs, and frequently, but not always, you’d be correct. But the trend in
21st century medicine is toward specialization, and a different choice would
often be better. Increasingly, primary care doctors, physiatrists, and others
emphasize the sports side. For example, Andy Pruitt, the director of the
Boulder Center for Sports Medicine and world-renowned source for solving
bike-related issues at the highest level (Lance Armstrong, Floyd Landis, Chris
Carmichael, and Bobby Julich), is a doctor of education.2 Such professionals
have designed their practices with the athlete in mind.
• Physical therapist, athletic trainer, or massage therapist. States have
various licensing requirements for these specialists. They also have a host
of access variations. In some states, for example, a physical therapist can
see patients only on physician referral. That said, after a condition has been
diagnosed, these hands-on professionals may be an effective resource for
a triathlete’s return to normal activity levels. They may also play a role in
triathlete recovery.
Triathletes normally experience some degree of pain while exercising and
in the early recovery period. Usually the pain is just muscle soreness from
a difficult workout effort. But if they begin to have specific pain in a joint or
a muscle group, backing off a bit and reducing the load for a few days may
be the appropriate action. Better to forego a full workout or two than risk
potential peril.
If resolution of the pain does not occur, it may be time to see one of the
members of the injury resource team. If these resource people have been
456 } Post

identified before they are needed, activation at this juncture is a simple


matter. One of the critical aspects of a smart training plan is the ability to get
in specific volumes of intense quality training. If triathletes have underlying
problems that keep them going easy day after day, they are not able to train
optimally. Such training limiters need to be addressed.

Strength Training for Injury Prevention


Although chapter 18 covers strength and power training and chapter 36 deals
with injury recovery, here we look at both from the point of view of the physi-
cian seeing the triathlete as a patient. In other words, what, if anything, can
triathletes do (or not do) in their training to reduce or eliminate the potential
for injury? We know that this process will never be foolproof. Even with the
best planning and experience, unexpected outcomes occur.
Does strength training have a role in diminishing injuries in modern
triathlon? Besides enhancing performance, increasing strength does play
a role for many triathletes in decreasing their potential for becoming an
injury statistic. The intelligent triathlete makes time in the training week to
round out, or balance, the musculoskeletal system because doing so not only
improves the probability of success on the racecourse but also diminishes
the potential for injury.
Not all that long ago people thought that doing anything with weights was
both foolish and arguably detrimental to the endurance athlete. The common
misunderstanding in the past was that strength-related efforts would lead to
bulking up—the Schwarzenegger look—and that although it might produce
a striking appearance at the beach, it ultimately meant a slower, heavier,
muscle-bound racer.
Six-time Hawaii Ironman champion Dave “The Man” Scott was among
the first to understand the relationships among strength, performance, and
injury. He did so quietly, amassing his first five wins before publishing
Dave Scott’s Triathlon Training in 1986.3 He noted, “Most of the top athletes
are faster runners than I am, yet I have been able to pass them at the end
of triathlons, much to the surprise of many spectators” (p. 137). He goes on
to point out, “Strength training . . . strengthens ligaments and tendons and
joints and reduces injury potential” (p. 139). Tim Noakes, MD, in Lore of Run-
ning4 states, “There is clear evidence to suggest that acute muscle injuries can
be prevented by strengthening muscles and eliminating muscle imbalances
between opposing muscles” (p. 783).
The National Strength and Conditioning Association (NSCA) defines
strength as the ability of a muscle or muscle group to exert maximum force.5 It
is expressed as a one-repetition maximum, or 1RM. Similarly, muscle strength
is the maximum amount of force that a muscle can exert against some form
of resistance in a single effort.
Triathlete Body Maintenance and Medical Care | 457

Triathlon: A High-Risk Sport


Generally, although there are some notable exceptions, much of the current
literature supports strength training for triathletes not only for performance
gain but also for the purpose of its contribution to a lower injury rate. This
point is of greater importance than many realize. Burns reported a 50 percent
injury rate in 131 triathletes and noted that overuse accounted for 68 percent
of preseason and 78 percent of in-season injuries reported.6
Unexpectedly, increasing years of triathlon experience was the most sig-
nificant predictor of preseason injury risk. High preseason mileage increased
the risk of injury during the season. In theory, shouldn’t the overall triathlete
population have a lower injury rate than a comparison group of single-sport
endurance racers? If a triathlete comes from a cycling background, aren’t the
running and swimming workouts just cross-training? Well, maybe not. Some
information does indeed point to a lower rate of overuse problems,7 but many
other studies show that the injury rate is much higher. Some report injury
rates as high as 90 percent over a 5-year period.8
Some suggest that triathletes set themselves up for injury by participating
in three endurance sports to excess. Thus, triathletes may tip the scale toward
injury in all three areas instead of just one. In other words, if a running
population sees metatarsal stress fractures, Achilles tendon strains, and IT
band friction syndrome problems, and the folks at the swim club experience
acromioclavicular and shoulder joint issues, they usually don’t overlap. Not
so in triathlon, which includes three endurance sports that usually require
the athlete to perform longer and more frequent workouts compared with
the single-sport participant. Using this logic, people suggest that triathlon
is a high-risk sport.

Risk Factors
Risk factors for injury in triathlon have been identified as
• total weekly training distance,
• weekly cycling distance,
• cycling training pace,
• weekly swimming distance,
• number of weekly running workouts (not distance), and
• total number of workouts per week.
If the reader predicted that of all triathlon injuries, the rate of running
injuries would be highest, that guess would be correct. Most, 58 to 64 per-
cent, of these issues were running related, 16 to 34 percent were of cycling
origin, and the remaining small percentage was related to swimming. The
predominant areas of injury were the low back, knee, and Achilles. The final
458 } Post

recommendation was this: “Triathletes should spend extra time strengthening


those parts of their bodies in functional ways, i.e., during movement patterns
which mimic those occurring naturally in their sports.”9

Flexibility for Injury Prevention


When warming up before exercise, is being able to do a split that dem-
onstrates a high degree of flexibility a desirable quality in all triathletes?
Probably not. Flexibility has been defined as “the total achievable excursion
(within the limits of pain) of a body joint through its range of motion.”10 The
athlete today should know that flexibility is “an individual variable, joint
specific, inherited characteristic that decreases with age, varies by gender and
ethnic group, and bears little relation to body proportion or limb length.”10
To the benefit of the triathlon community, flexibility can be altered with
appropriate exercise.
So why would a triathlete want to be more flexible, within certain bound-
aries? Many positives are associated with this practice, usually completed
postworkout (infrequently before) or during a strength session. Some incor-
porate stretching into their workday routine by finding time intermittently
throughout their day to practice. Flexibility can be too much of a good thing
because the excessively flexible triathlete is at risk for joint dysfunction.11
The triathlete who has always been called double-jointed or was born with
loose ligaments can be at risk. Joint dislocations to the shoulder, kneecap,
and fingers are more common among such people, and although they make
up only a small fraction of those who participate in triathlon, they account
for a large proportion of those who have joint stability issues.
Various types of stretching have been proposed including ballistic, static,
proprioceptive neuromuscular facilitation (PNF), and others. No single
winner has emerged. Many good flexibility programs have been proposed,
including an inventive regimen by Matt Fitzgerald in The Complete Triathlon
Book.12
In short, although we have considered stretching and flexibility a required
part of exercise, recent studies have shown that “increasing range of motion
beyond function through stretching is not beneficial and can cause injury
and decrease performance.”13

Massage
The role of massage in triathlon is unclear. Athletes are told that massage
speeds recovery, but no reliable scientific data supports this claim. But many
say that it does feel good. They are told that massage increases or maintains
soft-tissue mobility.
Reportedly, therapeutic massage functions as a collection of movements of
both the deep and superficial layers of tissue to lengthen muscle units that
Triathlete Body Maintenance and Medical Care | 459

have become contracted from lack of use, overuse, injury, or illness. Possibly
by increasing blood and lymph circulation, massage can reduce anxiety and
produce a sense of overall relaxation.
For the purposes of this text the word massage incorporates the classic
understanding of the term, or the manipulation of superficial and deeper
layers of muscle and connective tissue, as well as the Graston technique
(inducing microtrauma with instruments)14 and active release technique
(ART), the soft-tissue, movement-based massage technique,15, 16 and others.
Massage has been around for over a thousand years. More than 80 varieties
of massage are recognized. A quick census today would show that the most
commonly practiced specific technique is the well-known Swedish massage,
which has been around for almost 200 years. It is classified as a relaxation
sort of massage, and many types are known to Western practitioners.
The hands, feet, forearms, fingers, and other body parts can be used to
perform it. Although some recommend massage preexercise, most use it for
postexercise recovery and treatment of painful conditions. In a recovery situ-
ation, racers set their own schedules (often dependent on cost), be it weekly
on an off-training day, biweekly, or even monthly. Many athletes perform
self-massage. Injuries to the iliotibial band and quadriceps soreness respond
well to treatment with foam rollers or other such devices.
Some believe that among the benefits are muscular relaxation and elimi-
nation of areas of spasm or trigger points, which they think reduces the
potential for future injury. Others report that massage makes it possible for
them to train harder because they believe it removes waste products from
their muscles that would ordinarily persist.
In the athletic realm, the term sports massage is understood by many to
incorporate a rehabilitative posture as might be found with strain and counter-
strain, myofacial release, trigger point release, active release, and Graston.
Supporters claim that techniques like ART can cure them of tendinitis, back
pain, headaches, shoulder pain, carpal tunnel syndrome, and the like. These
claims are difficult to validate. The ART website (www.activerelease.com)
contains a plethora of ways to receive the treatment and courses to take, but
it offers no research, no medical studies, and no peer-reviewed literature
demonstrating that the technique is both safe and effective, achieves long-
term success, and is based in science.
That said, the technique suggested is remarkably similar to the time-
honored anchor and stretch or pin and stretch that massage and physical
therapists and others use for deep soft-tissue work. Pin and stretch has been
shown to be effective in scar tissue mobilization.
A similar finding persists at the Graston website, which describes the
technique, supplies a possible explanation of its mechanism, and includes
a list of providers.14 But no scientific basis for the treatment is presented.
Numerous athletes provide testimonials and stories of successes. So who
are we to believe? When describing the Graston technique in Science Based
460 } Post

Medicine, Harriet Hall, MD, states, “The fact that lots of people use it and think
it works does not constitute evidence that it actually works. Lots of people
used bloodletting and thought it worked. Lots of people believe in astrology.”17
Or, does the triathlete follow the teaching of Tom Holland, author of The
12-Week Triathlete,18 who writes “While there are many claims about the sup-
posed benefits of massage that I may question, I do believe that massage is
extremely beneficial to athletes in general and triathletes in particular” (p.
128)? All those who have had a massage but have no knowledge of the specif-
ics of the musculoskeletal system know that they felt better afterward. If they
had sore legs from that interval workout or low-back pain from a long, hard
bike ride, the massage made them feel better whether we (they) understand
how it works on a cellular level or not.
But to the contrary, if a triathlete is in a situation in which a surgical proce-
dure is recommended for some problem, what steps would he take to ensure
that the procedure and the surgeon are safe and effective? Should we not
expect the same from ART, Graston, and other practices? In short, massage
is just one tool available to the triathlete. When used appropriately, it may
play a role in both injury prevention and treatment. Further research may
give us the tools needed to continue this discussion.

Conclusion
Health may be defined as the functional level or metabolic efficiency of a
person. Health means being free from injury, pain, or illness. This definition
relates well to this chapter because the more important aspects of triathlon
health maintenance have been reviewed with a particular focus on informa-
tion that is of practical value to the triathlete. Possibly the most important
lesson is that triathletes themselves play a key role in the preemptive main-
tenance of their health, both body and spirit. Planning training with this in
mind and having a medical resource team in place when things don’t go
according to plan contribute to the ultimate success of the season.
CHAPTER
35
Triathlon Injuries and
Preventive Measures
—Nathan Koch, PT, ATC

T he incidence of overuse injury is clearly prevalent in the sport of triathlon.


To identify and treat overuse injury in a particular sport, it is critical to
understand injury rates and causation. A study in the Journal of Orthopedic
Sports Physical Therapy surveyed the training and injury patterns of 131 tri-
athletes over a 10-week prospective period during the triathlon competition
season and included a retrospective 6-month analysis of training history
and prior overuse injuries. The study revealed that 50 percent of triathletes
sustained an injury in the 6-month preseason at an injury exposure rate of
2.5 per 1,000 training hours. Thirty-seven percent were injured during the
10-week competition season at an injury exposure rate of 4.6 per 1,000 train-
ing hours. Overuse accounted for 68 percent of preseason injuries reported
and 78 percent of competition season injuries reported. Increasing years of
triathlon experience was the most significant predictor of preseason injury
risk. A previous history of injury and high preseason running mileage
increased the risk of injury during the competition season.1
In January 2010 the Journal of Strength and Conditioning published a study
comparing the incidence of overuse injury between Ironman- (IM) and
Olympic-distance (OD) triathletes (see table 35.1). The number of overuse
injuries sustained over a 5-year period did not differ between OD and IM
triathletes. Fewer OD triathletes (16.7 versus 36.8 percent) reported that their
injury recurred. Although OD triathletes sustained fewer running injuries
than IM triathletes did (1.6 ± 0.5 versus 1.9 ± 0.3), more athletes subsequently
stopped running because of pain (41.7 versus 15.8 percent) and for longer peri-
ods (33.5 ± 43.0 versus 16.7 ± 16.6 days). In OD, the number of overuse injuries
sustained inversely correlated with percentage of training time, number of
intense run sessions, and bike hill repetitions. The IM overuse injury number
correlated with the number of intensive sessions done (speed run and speed
bike sessions).2 In other words, higher intensity training reveals itself as a
high-risk and high-reward activity.

461
462 } Koch

Table 35.1 Comparing Overuse Injuries Between Ironman-Distance


and Olympic-Distance Triathletes
OD athletes IM athletes
Injury recurrence 16.7% 36.8%
Running injuries sustained 1.6 ± 0.5 1.9 ± 0.3
Stopped because of pain 41.7% 15.8%
Number of days stopped because of pain 33.5 ± 43.0 16.7 ± 16.6

The entire musculoskeletal system plays a role in the optimal performance


of the triathlete. If a link in the chain of movement is broken, the injury does
not necessarily manifest itself immediately or even in the general vicinity of
the breakdown. When attempting to diagnosis overuse injury in triathlon
and create prevention programs, the entire chain of movement of the body
must be addressed. Although each individual triathlete may have specific
musculoskeletal and biomechanical concerns, some issues are unique to each
discipline as well.

Swimming-Specific Injuries and Prevention


The most common swimming-specific injuries involve the shoulders, but
injuries to the spine and neck are not uncommon.

Shoulder Injuries
Shoulder injuries are the most common injury among swimmers, typically
resulting from increases in training intensity, volume, and distance. Because
the shoulder is a ball-and-socket (glenohumeral) joint with a shallow socket
and relatively large ball, it allows tremendous range of motion. This degree
of mobility permits the shoulder to perform more movement than virtually
any other joint, but it also makes the shoulder prone to breakdown under
repeated stress. The average competitive swimmer swims approximately
60,000 to 80,000 meters per week. With a typical count of 8 to 10 strokes per
25-meter lap, each shoulder performs 30,000 rotations each week.3 Although
this training volume may be more than the typical triathlete performs, an
understanding of how much movement is occurring in the shoulder is helpful
in creating a picture of why it is so susceptible to overuse injury.
The most common swimming-related injury is described as impingement
syndrome (generalized diagnosis that may include rotator cuff tendinitis or
tendinosis, bursitis, long head of the biceps tendinitis or tendinosis, and so
on) and presents as pain with overhead activity, pain at water entry or catch
phase, and inability to sleep on the affected side.
Typically, these injuries are biomechanical in nature and can be addressed
by modifying stroke pattern or in some cases by correcting the athlete’s bio-
mechanical restrictions. When dealing with an injured shoulder the triathlete
Triathlon Injuries and Preventive Measures | 463

should avoid the use of paddles and kickboards. Contributors to shoulder


injury in the swim stroke are hand entry crossing midline, thumb down entry,
asymmetric body roll, and unilateral breathing. Besides obtaining proper
swim stroke instruction, the triathlete needs to increase spinal mobility and
maintain good rotator cuff and scapular stabilizer strength. The following
three exercises before or after each swim session can be helpful.

Thoracic Extension Over Half Foam Place the foam roll perpendicular to
your midback. Extend your spine over the foam roll and lift both arms above
your head (figure 35.1). Hold for 1 minute and repeat three to five times, moving
the foam roller to higher segments of your thoracic spine.

„„Figure 35.1 Thoracic extension over half foam.

Rotator Cuff Strengthening Position 1 (figure 35.2a): Squeeze the towel roll
down with the elbow and keep the elbow at 90 degrees at all times. Position
2 (figure 35.2b): Lift your hand while squeezing the towel roll and not moving
the shoulder back. Slowly lower back to the starting position.

„„Figure 35.2 Rotator cuff strengthening: (a) position 1; (b) position 2.


464 } Koch

Ts on a Physioball Place your hips and core onto a physioball and place your
feet on the ground for balance. Look at the ground to protect your neck. Keep
your shoulders from rising up toward your ears. Concentrate on squeezing your
shoulder blades together (figure 35.3). Keep your thumbs facing the ceiling.

„„Figure 35.3 Ts on a physioball.

Spine and Neck Injuries


Although less common in swimming, a spinal injury may be more debilitating
than an injury to the shoulder. Typically, spinal injuries become chronic in
nature because they are adjusted or compensated for by the athlete’s unknow-
ingly changing the stroke pattern (i.e., turning out of one side only or rotat-
ing the trunk instead of the neck). Posture and positioning of the neck and
shoulders during swimming, much like in sitting, plays a significant role in
the amount of stress placed on the spine. In most cases, spinal injuries result-
ing from swimming are related to the aggravation of an already pathological
spine by prolonged spinal positioning, meaning that the athlete has had a
known or unknown spinal injury in the past.
Neck or cervical injuries can be described as radiculopathy, radiculitis,
herniated or bulging disc, or degenerative disc disease with symptoms that
may include radiating pain, numbness or tingling into the arm, scapular
(shoulder blade) pain with turning or extending the head, a noticeable loss
of head-turning ability, or sudden arm weakness or muscle atrophy. Low-
back or lumbar injuries such as radiculopathy, radiculitis, spinal stenosis,
herniated or bulging disc, degenerative disc disease, and sciatica may also
Triathlon Injuries and Preventive Measures | 465

include radiating pain, numbness or tingling in the lower extremity during


or after swim, pain in the buttocks while sitting after a swim workout, or
sudden leg or ankle weakness.
Preventing spinal injury in swimmers requires particular attention to
the triathlete’s symmetry of spinal rotation, mobility into extension, and
overall posture. Exercises should focus on moving the spine with full range
of motion. Exercises are particularly important for triathletes who have day
jobs that require prolonged sitting or standing in one place. The following
recommendations and exercises will help keep the spine mobile and its
movement symmetrical.

Neck Rotation Stretch Grab the right side of your head with your left hand
and gently pull your head toward your left shoulder (figure 35.4a). Keep both
shoulders down and relaxed. You should feel a stretch on the right side of the
neck. You should never feel pain, numbness, or tingling. For a static stretch,
hold the stretch for 1 minute and switch sides (figure 35.4b). Perform three
stretches on each side.
If you perform stretches prior to swimming, perform dynamic stretches.
Repeat the stretch multiple times without holding the stretch at the end range.

a b

„„Figure 35.4 Neck rotation stretch: (a) stretch the right side of the neck;
(b) stretch the left side of the neck.
466 } Koch

Seated Trunk Rotation on a Chair or Physioball Sit on a physioball with


the hips, knees, and feet in line and the back straight. Lift your arms so that
your hands are behind or to the sides of your head. Engage your abdominal
muscles and core as you rotate your trunk as far as possible (figure 35.5). Do
not allow the hips, knees, or feet to move. Feel the stretch around your midback
and shoulders. Hold the stretch for 1 minute then stretch to the other side.
Perform three stretches on each side.

„„Figure 35.5 Seated trunk rotation on a chair or physioball.


Triathlon Injuries and Preventive Measures | 467

Standing Back Extension With Arms Overhead Stand and lift your arms
overhead, keeping the elbows straight and arms in. Gently extend your spine
back, extending the neck and head last (figure 35.6). Think about moving
each vertebrae individually from the bottom to the top. Feel the stretch in the
spine, abdominal muscles, and shoulder blades. You should never feel pain,
numbness, tingling, or dizziness. Hold the stretch for 1 minute then release.
Stretch three times.

„„Figure 35.6 Standing back extension with arms overhead.


468 } Koch

Cycling-Specific Injuries and Prevention


Cycling is probably the most scientific of the three disciplines. A symbiotic
relationship requires a perfect match of an athlete’s unique biomechanical
characteristics with a specific geometry to reach optimal performance. In
other words, the bike must allow the body to function without excessive
musculoskeletal resistance or wasted energy. When the body is pushed
beyond its biomechanical limit, breakdown may occur and result in injury.
In December 2010 the American Journal of Sports Medicine published research
on the incidence of cycling-related injury in professional cyclists. The research-
ers from Norway interviewed 109 professional cyclists about overuse inju-
ries that they had sustained in the previous 12 months. Results of the study
revealed that 45 percent of overuse injuries were located in the lower back and
23 percent in the knee. Twenty-three time-loss injuries were registered—57
percent in the knee, 22 percent in the lower back, and 13 percent in the lower
leg. Few cyclists had missed competitions because of pain in the lower back
(6 percent) or anterior knee (9 percent).4 Note that lower-back injuries were
the most common, although knee injuries resulted in the largest amount of
training and racing time lost.
Although less common than low-back pain in cycling, cervical (neck) and
thoracic (midback) injuries in cycling are frequent and typically related to
the position of the head and neck over long periods. With the cervical spine
maximally extended and the thoracic spine maximally flexed, tremendous
stress is created at the transition, or breaking, point between the two. If the
athlete has abnormal mobility or has sustained an injury in the past, this
position can cause pain or create an overuse injury.

Neck and Midback Pain


In a 2007 study on the incidence of neck pain in multisport athletes published
in the Journal of Neurosurgery: Spine, 164 athletes responded to a medical
questionnaire. Approximately 64 percent of responding athletes reported
that their neck pain was sports related. Total years in the sport and number
of previous sports-related injuries were predictive.5
The most critical preventative measure that a triathlete can take with regard
to neck and midback pain in cycling is proper positioning with a bike fit done
by an experienced professional. Much like sitting at the computer all day, sit-
ting on the bike causes postural stiffness and tension in the spine, creating
potential for overuse injury. Prevention exercises for the neck and midback
should focus on moving the spine in the direction opposite that which it
remains in for long periods on the bike. Wall angels are good preventative
exercises for the neck and midback. Thoracic extension over half foam (figure
35.1) and Ts on a physioball (figure 35.3) from the section on swimming injury
prevention are also effective exercises.
Triathlon Injuries and Preventive Measures | 469

Wall Angels Position 1 (figure 35.7a): Start with the back, hips, head, and
arms against a wall. Position 2 (figure 35.7b): Raise the arms above the head.
Slowly lower the arms down the wall, emphasizing squeezing the shoulder
blades back and together. Complete two or three sets of 10 to 15 repetitions.

a b

„„Figure 35.7 Wall angels: (a) position 1; (b) position 2.

Low-Back Pain
Low-back (lumbar) pain is an extremely frequent complaint in cycling, yet it is
easily preventable. During cycling the lumbar spine is in a flexed or forward
bend position, and the cyclist must be able to tolerate this position over long
periods. To function optimally on the bike, the spinal, gluteal, and hamstring
muscles and spinal and hip joints must be flexible enough to maintain this
position with ease. In addition, muscles must be able to contract sufficiently
to stabilize the pelvis and spine. Good strength and flexibility prevent exces-
sive rotation and rocking on the saddle that can lead to overuse injury. The
ability to stabilize becomes more important when positioning is aggressive
or when power is increased.
Much like the cervical spine position on the bike, the lumbar spine posi-
tion also mimics the sitting posture at a desk or in front of a computer. These
470 } Koch

prolonged postures create significant imbalances in muscle activation and


restricted joint and soft-tissue mobility. The most critical preventative measure
that a triathlete can take with regard to low-back injury is proper positioning
on the bike facilitated by an experienced professional. Low-back exercises
should focus on increasing spinal extension mobility and strengthening the
back extensors and gluteal muscles. The following are preventative exercises
for the lumbar spine.

Press-Ups Lie face down. Push onto your elbows while keeping your hips
against the floor or table (figure 35.8). Hold this position for 30 to 60 seconds
or do 15 to 20 repetitions.

„„Figure 35.8 Press-ups.

Back Extension Over Physioball Lie over a physioball with your hips centered
on the ball and your feet on the ground or supported against a wall. With your
arms across your chest or your fingertips behind your ears, extend your torso
up until your body forms a straight line with your legs (figure 35.9). Return to
the starting position and repeat 30 times.

„„Figure 35.9 Back extension over physioball.


Triathlon Injuries and Preventive Measures | 471

Bridge With Long Arc Quad Position 1 (figure 35.10a): Lie on your back
with your knees bent and your feet flat on the ground. Tighten your abdominal
muscles and squeeze your buttocks. Push your hips up so that they are level
with your torso. Position 2 (figure 35.10b): While keeping your hips level and
your thighs even with each other, gradually extend one leg so that the leg is
straight. Do not allow the hips to drop. Alternate legs and repeat 10 to 15 times
on each leg. Complete two or three sets. Use your arms to help you balance but
not to lift you up. To increase the difficulty, cross your arms over your chest.

„„Figure 35.10 Bridge with long arc quad: (a) position 1; (b) position 2.
472 } Koch

Single-Leg Bridge Position 1 (figure 35.11a): Lie on your back with your
knees bent and your feet flat on the ground. Bring one knee toward your chest.
Position 2 (figure 35.11b): Lift your hips off the ground until your torso forms
a straight line with your leg. Keep your knee bent. Slowly lower your hips to
the ground, keeping your hips level. Alternate legs. Repeat 15 to 20 times on
each leg. To increase the difficulty, straighten your lifted leg.

„„Figure 35.11 Single-leg bridge: (a) position 1; (b) position 2.


Triathlon Injuries and Preventive Measures | 473

Pelvis and Hip Injuries


Pelvis and hip injuries in cycling, as in all sports, are becoming more common
because medical professionals are gaining more knowledge and diagnosis is
becoming faster and easier. On the bike, the skeletal and soft-tissue portion
of the pelvis has four key contact points on the saddle that provide a stable
platform from which the lower extremities can push and create momentum.
The muscles surrounding the pelvis must be strong enough to act as stabiliz-
ers as momentum and forces increase. As the thigh gets closer to the chest
at the top of the pedal stroke (extreme hip flexion), the lack of hip resistance
becomes crucial to a stable pelvis, optimal power transfer, and avoidance of
overuse injury.
Overuse hip injuries in cycling are preventable and are typically related
to the millions of revolutions performed in this extreme hip-flexed position.
Some cycling-related injuries to the pelvis and hip are discussed in the fol-
lowing paragraphs.
Femoral acetabular impingement (FAI) is a bony impingement of the femur
on the acetabulum (hip socket located on the pelvis). The two basic types of
bony impingement are CAM and pincer. CAM refers to a deformity of the
femur at the head and neck junction. Pincer refers to an overgrowth of the
acetabulum (socket). CAM and pincer deformity can occur independently
or in combination. Pain is typically located in the groin or in front of the hip
and is aggravated by a hyperflexed hip position (knee closer to the chest).
Athletica pubalgia (i.e., sports hernia) encompasses a variety of pelvic
injuries involving the abdominal and pelvic musculature outside the ball-
and-socket hip joint and on both sides of the pubic symphysis. Pain is typi-
cally located in the lower abdominal area and increases with hip extension
or under abdominal strain.
Other injuries commonly seen in the hip are tendinitis and tendinosis of
the hip flexor or hamstring tendons, bursitis, and proximal iliotibial band
friction syndrome (ITBFS). These soft-tissue injuries typically resolve quickly,
within 1 to 3 weeks, if properly cared for. Pain is usually localized to the
tendon, is not deep, and increases with increased activity. These injuries may
be irritated by sitting after workouts.
Preventing hip injury in a triathlete requires close attention to bike posi-
tioning as well as a specific exercise prescription. On the bike, the triathlete
should consider reducing crank arm length, using an oblong or oval ring,
emphasizing a flat ankle position at the top of the pedal stroke, and possibly
raising the front end of the bike. Investigating saddle geometry that can best
support the pelvis may be necessary. Proper positioning of the pelvis on the
saddle, the amount of weight through the pelvis, and hip position during
the pedal stroke will directly affect stability and subsequently the incidence
of overuse injury.
Hip injury prevention exercises should focus on increasing hip mobility
and increasing lateral and rotational hip strength.
474 } Koch

Monster Walk Place an exercise band around both ankles. Bend the knees
and sit back while keeping body weight balanced on both feet. The shoulders
stay directly over the knees and ankles throughout the exercise. Step to the side,
reaching with the midfoot, and create resistance with the band (figure 35.12).
Slowly follow the lead leg with the trail leg, working to control the resistance
of the band. Do not allow the feet to touch. Keep the toes pointing straight
ahead. Continue taking controlled steps with the abdominals and legs engaged
while maintaining proper alignment and tension.

„„Figure 35.12 Monster walk.


Triathlon Injuries and Preventive Measures | 475

Sidesteps Place an exercise band around both ankles. Stand tall with the
abdominals engaged and knees extended. Maintaining this upright posture,
step to the side, working to control the resistance of the band and reaching with
the midfoot (figure 35.13). Keep the toes pointing straight ahead, abdominals
engaged, hips level, and knees extended.

„„Figure 35.13 Sidesteps.


476 } Koch

Hip Flexor Stretch Kneel with your trunk in a door frame, corner of a wall,
or other sturdy frame. Keep your head, upper back, and gluteal muscles tight
to the frame. Reaching your arms overhead, posteriorly tilt your pelvis to bring
your lower back toward the frame (figure 35.14). For a greater stretch, reach
your knee farther back.

„„Figure 35.14 Hip flexor stretch.


Triathlon Injuries and Preventive Measures | 477

Scorpions Lie face down on the ground. Lift one leg toward the ceiling,
keeping the knee bent. Next, keeping your chest close to the ground, rotate
your hips so that your lifted leg crosses over your body (figure 35.15). Return
to the starting position and repeat on the other leg. Complete 10 to 15 repe-
titions on each side.

„„Figure 35.15 Scorpions.

Knee Injuries
The knee could be viewed as the center of activity for the lower body in
cycling. In other words, the knee will go where the hip, foot, and ankle tell
it to go. The knee functions primarily as a hinge joint and is powered by the
prime movers in cycling, the quadriceps and the hamstrings. The kneecap and
the femur create the patellofemoral joint. The patella, or kneecap, is a shallow
saucerlike bone that floats when the knee is fully extended and gradually
presses harder into a groove in the femur as the knee bends.
Hence, if conditions are present during the cycling motion that offset the
tracking of the patella in the femoral groove, an overuse injury is likely to
occur. Musculoskeletal conditions affecting biomechanics such as genu varus
(bow legged), genu valgus (knock kneed), coxa varus (hip bowing), coxa
valgus (hip angled inward), excessive or lack of femoral or tibial rotation,
iliotibial band (ITB) tightness, leg-length discrepancy, and pelvic obliquity
can create abnormal patellofemoral tracking and result in injury.
478 } Koch

Anterior knee pain is pain in the front of the knee that is typically aggra-
vated by actively extending the knee under load in an open (foot off the
ground as a kicking motion) or closed (foot on the ground as in a squat)
position. On the bike, pain typically occurs during the power phase of the
pedal stroke (at approximately the 3 o’clock position) that increases under
greater loads (seated climbing).
Lateral knee pain or iliotibial band friction syndrome (ITBFS) causes pain
at the outside aspect of the knee. This pain is most common at the bottom of
the pedal stroke as the knee extends. Therefore, it can be exacerbated by a
saddle that is positioned too high.
Medial knee pain can occur as pes anserine tendinitis or tendinosis or
bursitis. Pain is located at the medial or inside aspect of the knee below the
kneecap. Swelling and tenderness are common at the location of the pes anser-
ine. The pes anserine is an anatomical term used to describe the position of
attachment for three tendons: sartorius, gracilis, and semitendinous. These
tendons help control rotation between the upper leg and lower leg. During
cycling, if rotation of the lower leg is excessive in relation to the upper leg,
this condition may occur. This problem is sometimes related to type of pedal
and cleat position.
Because the knee is so heavily influenced by hip and ankle mechanics
during the pedal stroke, assessing the entire chain of movement is necessary
to prevent injury. First, make sure that both knees are functioning symmetri-
cally in the fore–aft pedal axle position and the knee flexion position at dead
bottom center.
A comprehensive review study in Sports Medicine examined the effects
of bicycle saddle height on knee injury risk and cycling performance. Most
evidence suggests that a relatively small adjustment in saddle height (5 per-
cent) affects knee joint motion by 35 percent.6 Cleat alignment also heavily
influences the position of the knee, and the alignment can be adjusted to
reduce pressure or strain. A common recommendation is that pedals with
float may assist in this reduction.
A research study in the Journal of Biomechanical Engineering concluded
that everting the foot (turning it out) may be beneficial in either preventing
or alleviating patellofemoral pain syndrome in cycling.7 In some cases, such
as a leg-length discrepancy, the knee may function differently on the left
compared with the right. In this instance bike positioning becomes more
complicated and requires a skilled medical bike fit to determine the cause and
the amount of discrepancy. Assuming that a shim or lift between the shoe
and cleat will solve all leg-length or asymmetrical issues is not safe. Although
bike fit is critical to preventing knee injury, the triathlete can accomplish a
lot by focusing on her own body through increasing the mobility of the hip
and ITB and strengthening the hamstrings and gluteal muscles.
Triathlon Injuries and Preventive Measures | 479

Foam Rolling for the ITB Lie with the side of your thigh on a foam roller
(figure 35.16). Roll up and down the iliotibial band in a slow, controlled fashion.
The motion should stay above the knee and below the greater trochanter (hip
bone). Repeat 10 to 15 times on each side and do two sets on each leg.

„„Figure 35.16 Foam rolling for the ITB.

Hamstring Curls With Physioball Lie flat on the ground with your heels
centered on a physioball and your toes pointed at the ceiling. Dig your heels
into the physioball, tighten the abdominals, and lift the hips (figure 35.17).
Keeping the hips off the ground and pressure on the heels, pull the physioball
toward your hips and then push it back out.

„„Figure 35.17 Hamstring curls with physioball.


480 } Koch

Dynamic Gluteal Muscles Stand on one leg. Squat down and extend the
opposite leg back, crossing it behind the squatting leg. (Think of the finish posi-
tion in bowling.) Return to the starting position and alternate legs. Complete
30 to 40 repetitions on each leg.

„„Figure 35.18 Dynamic gluteal muscles.


Triathlon Injuries and Preventive Measures | 481

Foot and Ankle Injuries


Foot and ankle injuries in cycling are relatively uncommon, but they can be
difficult to overcome. The foot and ankle create a small lever arm that should
remain stable throughout the pedal stroke. Motion at the ankle should be
minimal to provide a stable platform that allows the powerful quadriceps
and gluteal muscles to transfer force to the pedals. The more the ankle moves,
the more susceptible to injury it is.
Achilles tendinitis and tendinosis cause pain in the Achilles tendon that
increases with increased intensity and is typically tender to the touch. Tendi-
nosis refers to a chronic or degenerative condition of the tendon. This typically
occurs in the cyclist because of a leg-length discrepancy, a shoe with a heel
counter that is too high, or cleats that are too far forward.
Morton’s neuroma, or in the early stages, hot spots, occurs in the forefoot
(between the second and third toes or between the third and fourth toes).
This inflammation of the nerve or nerve sheath is located between the bones
of the foot, and it is typically very tender to the touch. Pain, numbness, and
tingling get worse with increased cycling volume and training, with pro-
longed seated climbing, and when walking barefoot.
Metatarsalgia describes pain located at the forefoot that commonly is
directly under the first or second metatarsal heads (ball of the foot). Callous
formation will typically be found at the location of pain, and the athlete will
overpronate and have increased pain with weight-bearing activity. Metatar-
salgia usually occurs with increases in training intensity or volume, or with
changes in footwear.
Prevention of foot and ankle injuries requires attention to pedal selection,
cleat placement, and shoe preference. Although little research has been done
on the optimal length, width, and shape of the actual pedal, the surface area
of the pedal and cleat may influence foot pressures and subsequently result
in overuse injury. Simply put, triathletes with bigger and wider feet may
require a larger platform to reduce foot pain.
Cleat placement has recently been debated in the cycling world. Current
efforts are to push the cleat position back from the metatarsalphalengeal loca-
tion (forefoot) to the tarsometatarsal location (midfoot). This more rearward
position of the cleat on the shoe can affect the foot and ankle positively by
reducing pressure on the forefoot and by reducing the lever arm at the ankle.
This positioning may result in less gastrocsoleus (calf) muscle activity and
thus may reduce or eliminate forefoot numbing, reduce stress on the Achil-
les, and conserve the triathlete’s calf muscles for running. At the same time,
this more rearward position may negatively affect climbing performance.
Besides cleat position, actual shoe construction may contribute to injury.
Proper width, length, height of heel counter, strap location, and material of
the shoe insert are important considerations in selection of the proper cycling
shoe. A research study done in Foot and Ankle International found that cycling
482 } Koch

shoes made with carbon fiber produced peak plantar pressures 18 percent
higher than those of plastic design (121 kPa versus 103 kPa). Because most
competitive shoes are made of carbon fiber, the addition of an insole, arch
support, or metatarsal pad that will conform to the contours of the foot may
be necessary.8
Foot and ankle injury prevention exercises during cycling should focus
on maintaining gastrocsoleus muscle length, addressing tightness or restric-
tions in the soft tissue, and strengthening the small muscles in the foot that
help stabilize the foot.

Gastrocsoleus Stretch With Midfoot Support Stand with your hands on a


wall and assume a lunge position. Align the outside of your back foot straight
forward. Place a folded towel under the arch of your back foot and push your
heel to the ground (figure 35.19). Hold the stretch for 30 to 60 seconds. Next,
bend the knee of your stretching leg and repeat the stretch. This positioning
will place more stretch on the soleus muscle.

„„Figure 35.19 Gastrocsoleus stretch with midfoot support.


Triathlon Injuries and Preventive Measures | 483

Foam Rolling the Gastrocsoleus Sit on the ground with your legs straight
and ankles crossed. Place the foam roller under your lower leg. Slowly roll up
and down the length of the calf muscle (figure 35.20). Stay above the Achilles
tendon and roll up underneath the knee. Roll 10 to 15 times on each leg and
do two sets for each leg.

„„Figure 35.20 Foam rolling the gastrocsoleus.

Arch Crunches Place your foot on a towel and crunch your toes (figure
35.21), pulling the towel toward you. Continue this motion for 2 to 3 minutes
and switch feet.

„„Figure 35.21 Arch crunches.


484 } Koch

Running-Specific Injuries and Prevention


Running injuries account for the most frequent and most debilitating injuries
sustained by triathletes. Runners in general exhibit a higher incidence of
overuse injury than cyclists or swimmers do. This discrepancy is typically
attributed to the fact that running is a weight-bearing activity that results in
forces acting on the body up to three times the athlete’s body weight. This
axial force coupled with the tremendous amount of eccentric (muscle working
as it lengthens) muscle activity creates a stressful environment for the body
and is the epitome of high risk–high reward training.
How likely is a runner to become injured during training? Where on the
body do most injuries occur, and what are the significant factors? In a com-
prehensive review of 17 research studies on running injuries published in
the British Journal of Sports Medicine in 2007, the incidence of lower-extremity
running injuries in long-distance runners ranged from 19.4 percent to 79.3
percent. The predominant site of injury was around the knee (7.2 to 50 percent).
The most significant results were that a long training distance per week in
male runners and a history of previous injuries in males and females were
significant risk factors for injuries.9 Because of the wide range in running
injury incidence reported, a specific study about the incidence of running
injuries in triathletes would be a likely next step in the research.
For now, keeping a training log, having a coach, and listening to the body’s
response to training are the first steps to successful injury prevention. Addi-
tionally, assessing running technique and footwear under the expertise of
an experienced coach or physical therapist is important. Much like the golf
swing, run technique can be successful in many variations as long as the basic
principles are observed. Additionally, exercise training in the form of resis-
tance and plyometric training has resulted in enhanced running economy,
possibly leading to decreased injury incidence.10

Lumbar Spine Injuries


Preventing lumbar spine injuries in the runner requires a review of the
training program to assess body readiness for intense workouts, assessment
of running gait and foot strike, and exercises that focus on spinal and hip
mobility and core stability strengthening. Recent research suggests that by
increasing hip extension mobility, the athlete can reduce anterior pelvic tilt
and subsequently reduce compression and stress on the lumbar spine.11
Triathlon Injuries and Preventive Measures | 485

Inchworm Position 1 (figure 35.22a): Start in push-up position with the hands
under the shoulders and the toes on the ground. Position 2 (figure 35.22b):
Keeping the hands stationary, walk the feet toward the hands, lifting the hips
in the air. Do not allow the knees to bend. Once in the uppermost position,
keep the feet stationary and walk the hands forward to return to the starting
position. For added stretch, drop the heels to the ground while lifting the hips
in the air. Repeat 15 to 20 times while moving across the room.

„„Figure 35.22 Inchworm: (a) position 1; (b) position 2.


486 } Koch

Superman on BOSU Lie on a BOSU ball, rounded side up, and have the hips
centered on top. Keeping the arms forward and knees locked, lift both arms
and both legs at the same time to Superman position (figure 35.23). Focus
on squeezing the shoulder blades and the gluteal muscles. Hold for 30 to 60
seconds and complete two or three sets of 10 to 15 repetitions.

„„Figure 35.23 Superman on BOSU.

Pike With Physioball Begin in a push-up position with your feet on a phys-
ioball. Keeping the arms stationary and the legs straight, lift your hips and roll
the physioball toward you until your body is in an inverted pike position (figure
35.24). Return to the starting position and repeat 30 to 40 times.

„„Figure 35.24 Pike with physioball.


Triathlon Injuries and Preventive Measures | 487

Pelvis and Hip Injuries


Pelvis and hip injuries in running differ from those that occur during cycling.
The pelvis is more posteriorly rotated with the hip joint in extreme flexion on
the bike, whereas in running the pelvis is typically tilted anteriorly and the
hip joint must be able to flex and extend through the running gait cycle. As
alluded to earlier, the ability of the hip joint to achieve hip extension during
the swing phase of running is critical to efficiency, performance, and injury
prevention.
Most pelvic and hip injuries occur because of poor running technique (lack
of hip extension, excessive rotational or side-to-side movement, overstriding,
heel striking), decreased muscle length (hip flexor or hamstrings), gluteal or
adductor weakness, postural deficits (scoliosis, LLD), abnormal foot and ankle
mechanics, training errors, or nutritional deficits. The number of female ath-
letes with pelvic and hip pain has increased from less than 1 percent to 15.1
percent in the last two decades, and the most common activity is running.12
These pelvic and hip injuries are discussed in the following paragraphs.
Femoral acetabular impingement (FAI), as in cycling, has bony impinge-
ments, CAM, or pincer, which commonly has symptoms of pinching sensation
in the front of the hip or groin, groin pain, and deep ache in the hip with sit-
ting or stretching. Pain is typically worse at the beginning of a run and may
improve during the run, unlike the pain associated with a stress fracture.
Athletica pubalgia (i.e., sports hernia) encompasses a variety of pelvic
injuries involving the abdominal and pelvic musculature outside the ball-and-
socket hip joint and on both sides of the pubic symphysis. Athletes present
with inguinal (groin) or lower abdominal pain that increases with running
or physical activity. This condition can be challenging to diagnosis and is
often misdiagnosed as a tendinitis or muscle strain.
Sacral, pelvic, or femoral stress fractures occur with significant increases
in running volume or intensity. In addition to training errors, nutritional con-
cerns and changes in footwear or orthotics may be related to stress fractures.
More commonly affecting females, stress fractures present with generalized
pain around the location of the fracture, and pain may increase during weight-
bearing activity. Athletes cannot typically run through this injury.
Trochanteric bursitis is inflammation of the bursa sac, a sac of fluid that
provides lubrication between a bone and a tendon. Pain is typically on the
outside of the hip. The hip is tender to the touch and may be swollen.
Hamstring tendinitis or tendinosis in the posterior hip or buttocks region
is inflammation or degeneration of the hamstring tendon where it attaches to
the ischial tuberosity (sit bone). This injury occurs with increases in training
intensity and is the result of increased strain on the tendon as stride length
increases.
Piriformis syndrome refers to a painful condition located in the buttocks
where the piriformis muscle is in spasm, injured, or irritated. Compression of
the sciatic nerve may occur. Pain typically is located in the buttocks, although
488 } Koch

it may be referred into the back of the thigh (hamstring), and it increases with
sitting, squatting, and athletic activity.
As with the lumbar spine, preventing pelvic and hip injuries in the runner
requires a review of the training program to assess body readiness for intense
workouts, assessment of running gait and foot strike, and exercises that focus
on hip mobility and core stability strengthening. Changes to running gait
posture and technique may also help. Recent research suggests that subtle
increases in step rate can substantially reduce the loading to the hip and
knee joints during running and may prove beneficial in the prevention and
treatment of common running-related injuries.13
Hip injury prevention exercises should focus on increasing hip extension
mobility and hip lateral and rotational strength. Along with the exercise
presented here, sidesteps (see figure 35.13) and scorpions (see figure 35.15)
are good for preventing hip injuries.

Clams With an Exercise Band Position 1 (figure 35.25a): Lie on your side
and place an exercise band above both knees. Bend your legs so that your
heels, buttocks, and shoulders align. Position 2 (figure 35.25b): Engage your
abdominal muscles to maintain proper alignment. Keeping your feet together,
slowly lift the upper leg from the lower leg.

„„Figure 35.25 Clams with an exercise band: (a) position 1; (b) position 2.
Triathlon Injuries and Preventive Measures | 489

Knee Injuries
Knee injuries are the most common injury sustained during running. The
most common knee injury occurring in runners is patellofemoral syndrome,
which appears to be related to excessive coronal (side-to-side) and transverse
(rotational) plane motion during the running gait cycle.
Excessive movement or lack of movement at the joints above and below
the knee can negatively affect the function of the knee or patellofemoral joint
and create increased stress. The knee must be allowed to function primarily
in the sagittal (straightforward) plane with the least amount of load and time
on the ground as possible. The knee joint undergoes a five-fold increase in
quadriceps activity to support body weight on the ground. This increase is
associated with a five-fold increase in the ground reaction force moment about
the knee.14 The muscles surrounding the knee joint must be strong enough
to assist with absorption of this force.
The entire chain of movement above and below the knee joint must be
examined to determine the cause of increased stress and injury. Additional
research suggests that knee pain can be affected by alignment of the lower
extremity (the amount of ankle motion, the presence of knee varus, or bowing,
and the functional position of the forefoot).15 Specific knee injuries that can
occur in running are discussed in the following paragraphs.
Anterior knee pain (patellofemoral syndrome, fat pad impingement, patel-
lar tendinitis or tendinosis, patellar bursitis, and chondromalcia patella) is
pain in the front of the knee that is aggravated by actively extending the knee
under load in an open (foot off the ground as in a kicking motion) or closed
(foot on the ground as in a squat) position. During running, pain typically
occurs at the impact and loading phases of the gait cycle.
Lateral knee pain or iliotibial band friction syndrome (ITBFS), as in cycling,
is pain located at the outside aspect of the knee. During running, pain typi-
cally occurs at the end of the swing phase or at impact.
Medial knee pain or pes anserine tendinitis or tendinosis or bursitis, as
mentioned earlier, is pain located at the medial or inside aspect of the knee
below the kneecap. During running, if lower leg rotation is excessive in rela-
tion to the upper leg, this condition may occur.
Meniscus (cartilage) tear or injury can be located on the medial or lateral
aspect of the knee and is typically along the joint line. Acute tearing occurs
from a traumatic twisting or hyperextension event and is usually accompa-
nied by swelling, tenderness, and locking or catching in the joint. A degen-
erative meniscal tear is a wear-and-tear condition that occurs over time and
is typically related to either an old injury or abnormal running mechanics
over many years.
Hamstring strain (tear in the belly of the hamstring muscles) exhibits pain
in the back of the thigh. The athlete usually recalls the exact moment when
the injury occurred. Typically, it is accompanied by swelling, bruising, pain,
and difficulty walking. Evidence suggests that during the terminal swing of
490 } Koch

the running gait, the hamstrings demonstrate the most dramatic increase in
biomechanical load when speed is progressed toward maximal sprinting.16
Therefore, this injury typically occurs during sprint or interval workouts.
Prevention of knee injuries in the runner requires a review of the training
program to assess body readiness for intense workouts. A recent research
study concluded that an 8-week rehabilitation program focusing on strength-
ening and improving neuromuscular control of the hip and core musculature
produces positive patient outcomes, improves hip and core muscle strength,
and reduces excessive movement at the knee, which is associated with devel-
oping patellofemoral syndrome.17
Knee injury prevention exercises include the monster walk (see figure 35.12),
sidesteps (see figure 35.13), and foam rolling for the ITB (see figure 35.16), as
well as the exercise presented here.

Windmill Deadlift Position 1 (figure 35.26a): Stand straight and balance on


one leg. Hold a dumbbell in the hand on the side of the lifted foot. Position 2
(figure 35.26b): Slowly lean forward while maintaining a straight line from the
foot through the hip and shoulder. Reach to the outside of the opposite foot.
Keep leaning until hamstring flexibility limits the movement and slowly return
to position 1. Position 3 (figure 35.26c): Hold the dumbbell in the hand on the
side of the grounded foot. Slowly lean forward while your swinging leg turns
away from your body.

a b c

„„Figure 35.26 Windmill deadlift: (a) position 1; (b) position 2; (c) position 3.
Triathlon Injuries and Preventive Measures | 491

Foot and Ankle Injuries


Foot and ankle injuries in running are common, and potential causes and
prevention techniques are heavily debated. The barefoot running craze has
had more effect on injuries to the foot and ankle than to any other joint.
Although some studies found that running barefoot with a forefoot gait
results in less axial forces, they did not examine the impact and strain on
surrounding muscle and tendon.18
Experts generally accept that running with a forefoot gait technique is
more efficient and creates less impact, although running barefoot has its own
issues of safety and practicality. No research studies are currently available
on barefoot running injuries. A study in Medicine and Science in Sports and
Exercise demonstrated that using a more forefoot style of running results in
a 33 percent decrease in stride length, a 45 percent decrease in vertical dis-
placement of center of mass, and a 35 percent decrease in vertical impacts
compared with rear-foot strike patterns.19 The difference between a trained
(forefoot running) shoe runner and a barefoot runner on incidence of injury
requires more research. Proper footwear selection and running technique
instruction may be the best option for prevention of injury.
Results from a study in Clinical Biomechanics revealed that high-arched
runners reported a greater incidence of ankle injuries, bony injuries, and
lateral lower-leg injuries. Low-arched or flatfooted runners exhibited more
knee injuries, soft-tissue injuries, and medial lower-leg injuries.20 Common
injuries of the foot and ankle during running are discussed in the following
paragraphs.
Achilles tendinitis, tendinosis, tenosynovitis, and bursitis all describe pain
located at the Achilles tendon that increases with increased intensity. The
tendon is typically tender to the touch, swollen, or thickened. This condition
may occur because of increases in training intensity or volume or changes
in footwear.
Posterior tibialis tendinosis, tendinitis, or medial tibial stress syndrome
(shin splints) describes pain located on the inside of the shin. The posterior
tibialis muscle and tendon, located on the inside portion of the tibia (lower leg),
are under tremendous load as they work eccentrically to control the foot and
ankle during pronation in the loading phase. Typically caused by repetitive
overload, overpronation during the load phase, and muscle imbalances, this
condition may also be initiated by changes in footwear or training.
Peroneal tendinosis or tendinitis describes pain located on the outside of
the shin. The peroneal muscles and tendons are primary stabilizers of the
foot and ankle during the loading phase. When injured, they are tender to
the touch. This condition occurs more frequently in a cavus or supinated foot.
Plantar fasciitis or fasciosis is an extremely common inflammatory or
degenerative condition that occurs because of repeated microtrauma to the
plantar fascia as it inserts into the calcaneus (heel). When injured, significant
heel pain occurs. The pain is worse with the first few steps in the morning,
492 } Koch

but the tissue typically warms up and feels a bit better throughout the day.
During running, pain usually occurs in the first few minutes but improves
as the plantar fascia warms up. A tear or stress fracture will cause pain that
does not improve with activity and may exhibit localized swelling or bruising.
Morton’s neuroma, or hot spots, occurs in the forefoot (between the second
and third toes or the third and fourth toes) and is characterized by inflam-
mation of the nerve or nerve sheath located between the bones of the foot.
Typically, the area is tender to the touch and may feel like a marble is in the
shoe. Pain, numbness, and tingling will increase as running volume and
training increase. Walking barefoot is usually avoided.
Metatarsalgia describes pain located at the forefoot that commonly is
directly under the first or second metatarsal heads (ball of the foot). Callous
formation typically occurs at the location of pain, and the athlete overpronates
and has increased pain with weight-bearing activity. Metatarsalgia usually
occurs with increases in training intensity or volume or with changes in
footwear.
Calcaneal cuboid syndrome is a condition that presents as pain on the
outside or outside bottom of the foot, and it typically occurs in runners with
a history of ankle sprain. This condition refers to a subluxation or abnormal
movement of the cuboid bone toward the sole of the foot. It typically presents
as localized tenderness, a slight indentation, and pain on the outside of the foot
with walking, and it seems to occur more frequently in the overpronated foot.
Stress fracture of a metatarsal bone in the foot is the most common fracture
that occurs in runners. Usually presenting as localized tenderness and pain
with weight-bearing activity, a stress fracture often occurs following a sudden
increase in training or a change in footwear. Runners with a supinated or
rigid foot are most at risk for this injury. Caution should be used when using
orthotics made of hard materials (plastic, carbon fiber) for this population.
Additional attention should be paid to the fifth metatarsal because pain there
should be differentiated from an acute fracture versus a stress fracture. Tibial
(lower-leg) stress fractures may also occur in runners, especially those who
have poor lower-extremity mechanics during the stance phase of running.21
Compartment syndrome or chronic exertional compartment syndrome is
not fully understood, although it is theorized that increases in intramuscular
pressure during exercise impair blood flow and subsequently cause pain.
This condition typically occurs in the front, or anterior, portion of the lower
leg and is more common in runners as a chronic syndrome. It is thought to
be caused by lower-extremity malalignment, muscle imbalances, training
errors, improper footwear, or poor running gait technique.
Evaluating shoe selection while actually running is critical to prevention
of injury. Research has shown that motion control shoes caused the greatest
percentage of runners to be injured regardless of foot type.22 Furthermore,
running shoes with more wear resulted in less trunk lean and longer stance
time during running, emphasizing the need to replace running shoes fre-
quently.23
Triathlon Injuries and Preventive Measures | 493

Running technique training, footwear selection assistance by a knowledge-


able professional, and possible integration of custom orthotics can provide
significant injury prevention. In addition, strengthening the foot and ankle
will provide stability when the foot hits the ground under the tremendous
load of the runner’s body.
Arch crunches (see figure 35.21) should be used in conjunction with the
exercise presented here to prevent ankle and foot injuries.

Eccentric Ankle Inversion With Exercise Band Begin with the foot relaxed
and the band placing tension to the inside of the foot (figure 35.27). Scoop the
bottom of the foot as far to the inside as possible, moving only your ankle. Do
not let your knee move with your foot. Slowly return to the starting position
and repeat Perform the action 30 to 40 times on each foot.

„„Figure 35.27 Eccentric ankle inversion with exercise band.


494 } Koch

Heel Raise Stand on the balls of your feet on the edge of a box or step. Lift
your heels up while keeping the knees straight (figure 35.28). Maintain pres-
sure on the big toes. Do not allow the heels to move outward. After the heels
are in the uppermost position, slowly lower the heels to the starting position.
Repeat 30 to 40 times. To progress, try doing heel raises one leg at a time.

„„Figure 35.28 Heel raise.


Triathlon Injuries and Preventive Measures | 495

Training Through Injury


Although training through injuries is normally unwise, in some circum-
stances training can proceed under the guidance of a sports medicine physi-
cian or physical therapist. Table 35.2 provides general guidelines, although
their applicability will vary with the severity and location of the injury. The
various tissues in the body undergo different physiological methods of heal-
ing; some may require full rest, whereas others may even require exercise to
stimulate the body’s natural healing process.

Table 35.2 Training Through Injury


Type of injury Yes No What can I do? (may depend on location of injury)
Bone X Swim and bike

Muscle X Swim, bike, and possibly run

Tendon X Swim and bike

Ligament X Bike

Cartilage X Swim and bike

Nerve X See physician

Fascia X Swim, bike, and run

Swelling X Swim and possibly bike

Redness or fever X See physician

Conclusion
Being successful in an endurance sport such as triathlon, in which triathletes
can compete at any age, requires persistence and knowledge of injury preven-
tion. Understanding how each triathlete’s unique body responds to training,
what signs warn of potential injury, and what preventative measures can be
taken can mean the difference between ringing the cowbell on the side of
the road and qualifying for world championships. The information provided
in this chapter is not intended to diagnosis a current injury but to provide a
resource to healthy triathletes so that they can sustain a lifelong passion for
competition and health.
This page intentionally left blank.
CHAPTER
36
Triathlon Injury
Recovery Techniques
—Nathan Koch, PT, ATC

F ull recovery from injury is critical to a executing a successful training pro-


gram and ultimately achieving the triathlete’s goals. Recovery from injury
is typically multifactorial and may require a team of experts to determine
the path to full recovery. Finding a sports medicine clinician with experience
in helping triathletes recover from injury is the first and most critical step in
resolving injury. Determining the cause of the injury, and not just treating
symptoms, may take more commitment from the athlete and the treating
clinician, but doing so will set the course of treatment in the right direction.
The following are some, but not all, injury recovery techniques used by
sports medicine clinicians. In the sports medicine world and the medi-
cal world in general, the phrase practicing medicine is used for a reason. A
technique or treatment that works well for one triathlete may not work for
the next with the same condition. “More research is needed” is a common
statement in the medical literature; hence, the athlete’s treatment is based on
the current research available, the clinician’s experience, and trial and error.
Therefore, these techniques or procedures are provided as potential solu-
tions to overcoming injury for the triathlete. The following are grouped into
three categories: DIY with a bit of guidance, conservative clinician-assisted
treatments, and invasive clinician-assisted treatments. They may be used
in combination or individually, depending on the clinician’s preference, the
specific injury and, most important, the triathlete’s preference.

DIY (Do It Yourself) With a Bit of Guidance


Several injury recovery methods can be initiated by the triathlete with a little
instruction. Although an appointment with a sports medicine professional is
not necessary to begin these treatments, occasionally a consultation is helpful
to address technique or form.

497
498 } Koch

Core Strengthening
Core strengthening can be described as control of the trunk during com-
plicated movements such as swimming, cycling, and running involving a
complex interaction between the nervous, muscular, and skeletal systems.
Although many exercises are available to athletes, establishing a core strength-
ening program that focuses on the triathlete’s specific weak muscles and
movements is critical to injury recovery and injury prevention.
In other words, programs designed to increase core strength in the triathlete
should have specific goals. The scapular, back, abdominal, and hip muscles
should all be considered when implementing a core training program.
Although researching every available core exercise to determine which
one best activates the intended muscle is virtually impossible, one study at
least got the ball rolling. A study in the Journal of Orthopedic and Sports Physical
Therapy tested the ability of eight Swiss ball exercises (roll-out, pike, knee-
up, skier, hip extension right, hip extension left, decline push-up, and sitting
march right) and two traditional abdominal exercises (crunch and bent-knee
sit-up) on activating core (lumbopelvic hip complex) musculature.1
Results revealed that the roll-out and pike (figure 36.1) were the most effec-
tive exercises in activating upper and lower rectus abdominis, external and
internal obliques, and latissimus dorsi muscles, while minimizing lumbar
paraspinals (back extensors) and rectus femoris (quadriceps) activity. The pike
with physioball exercise in chapter 35 is a good example of the pike exercise
that activates these muscles.

„„Figure 36.1 Pike with physioball. See chapter 35.


Triathlon Injury Recovery Techniques | 499

Eccentric Strengthening
Eccentric strengthening is the gold standard in the research literature for
treating overuse tendon injuries (tendinosis or tendinopathy). For instance,
eccentric exercises have the most evidence of effectiveness in treatment of
Achilles tendinopathy compared with extracorporeal shockwave therapy,
local corticosteroid treatments, prolotherapy (an injection of an irritant
solution such as dextrose or sugar water that is thought to increase blood
supply and repair damaged connective tissue), and topical nitroglycerine
application.2
Eccentric contractions occur when the muscle is contracting as it is length-
ening. These contractions are commonly described as negatives. They are
more forceful contractions than concentric (muscle shortening) contractions,
and they strengthen the connective tissue to a greater degree. This type of
strengthening is critical to fixing a tendon injury long term and preventing
future injury. Achilles tendinosis, patellar tendinosis, and hamstring tears
are three common overuse injuries that are frequently cited in the literature
as responding favorably to eccentric strengthening.

Ice
Ice or cold plunge has been used for decades to control swelling and inflam-
mation from acute injuries. It may not be as beneficial for chronic injuries
except for temporary pain relief. Ice is the simplest form of injury recovery,
and it should be used early and often in an acute injury such as ligament
sprain, muscle sprain, or joint irritation. Temperature should be between
33 and 59 degrees Fahrenheit (between 0.5 and 15 degrees Celsius) for 10
to 20 minutes.
Although ice should be applied immediately in the case of acute injury,
ice baths used as a preventative measure may make sense to avoid aches
immediately after hard efforts. Because ice constricts the blood vessels that
involve normal protein synthesis in the muscle, recovery could possibly be
slowed by using ice sooner than 2 or 3 hours after hard efforts. To address
this theory, more research is needed on how ice or vasoconstriction of blood
vessels affects muscle recovery and how soon after activity to apply ice.

Spinal Stabilization
Spinal stabilization refers to a group of exercises focusing on segmental
spinal stabilization. Exercises focused on deep abdominal and spinal muscles
(transversus abdominis and the multifidus) have been shown to be superior
to exercises focused on superficial muscles (rectus abdominis, obliques, and
erector spinae) in reducing pain and decreasing disability in people with
low-back pain.3
500 } Koch

People with chronic low-back pain typically exhibit atrophy or weakening


of spinal and hip flexors muscles. This weakness typically correlates with the
painful or injured side and would indicate a strong necessity for exercises
that isolate this musculature. Typically, these exercises are prescribed by
physical therapists and may include exercises such as back or hip exten-
sion or supine bridging exercises such as the bridge with long arc quad and
single-leg bridge exercises discussed in chapter 35 (figures 36.2 and 36.3).

„„Figure 36.2 Bridge with long arc quad. See chapter 35.

„„Figure 36.3 Single-leg bridge. See chapter 35.


Triathlon Injury Recovery Techniques | 501

Stretching
Stretching is a heavily debated topic that requires continued research and,
more important, good clinical judgment. Research and opinions vary dra-
matically on what stretches to do, how long to hold them, when to perform
them, and whether stretching can be detrimental to athletic performance.
When used for specific injuries, stretching or range of motion exercises
that are prescribed specifically to target damaged or restricted tissue are
crucial to recovery. Current thinking in the sports medicine world suggests
the use of dynamic or movement stretching that mimics the athlete’s move-
ment in sport before activity and static or sustained position hold stretching
after training.
Evidence for stretching hamstring and hip flexors (specifically the rectus
femoris) in runners was presented in a study in Gait and Posture in Febru-
ary 2010.4 The peak length of the right (trailing limb) hip flexor was nearly
identical between walking and running, whereas the maximum length of
the rectus femoris, a hip flexor and knee extensor, increased during running.
The maximum length of the left (leading limb) biceps femoris (hamstring)
was also unchanged between walking and running. Further, the timing of
peak hip flexor length and peak opposite-leg hamstring length occurred
essentially simultaneously during running, at a time during gait when the
hamstrings are most vulnerable to stretch injury.
Therefore, hip flexor stretching in combination with hamstring stretch-
ing may be beneficial in the treatment and prevention of running-related
hamstring injury. Although an entire book could be written on the topic of
stretching, an excellent clinical research review titled “To Stretch or Not to
Stretch: The Role of Stretching in Injury Prevention and Performance” in
the Scandinavian Journal of Medicine and Science in Sports by M.P. McHugh
and C.H. Cosgrave provides a fantastic summary of the available research.5

Water Running
Water running, or reduced-body-weight running, is a common prescription
for runners who sustain a stress fracture or for athletes as they recover from
surgery. Water running is used to reduce vertical forces through the bone
and soft tissue. A recent study comparing running in water at chest level
versus hip level revealed vertical forces corresponding to 0.80 and 0.98 times
the subject’s body weight at the chest and hip level, respectively. Anteropos-
terior (forward to backward) forces corresponded to 0.26 and 0.31 times the
subject’s body weight at the chest and hip level, respectively. As the water
level decreased, the subjects ran faster.
Therefore, the immersion level and speed must be considered because they
can affect the force components, mainly in the anteroposterior direction.6 The
advent of antigravity treadmills has allowed greater control of these condi-
tions, although they are expensive to use and not widely available.
502 } Koch

Conservative Clinician-Assisted Treatments


Some injury recovery methods require the assistance of a licensed sports
medicine clinician. In most cases, an appointment for an evaluation is neces-
sary before initiating treatment.

Compression Machines
Compression machines have been used for years in the medical profession.
Significant research has shown that these machines are effective in treating
wound healing, venous insufficiency, and lymphedema. More recently, com-
pression machines have entered the sports medicine world in the treatment
of postoperative and postfracture swelling in addition to muscle recovery.
Although results are quantifiable in the treatment of joint or limb swelling
after an acute injury or postsurgery, more research is needed to determine
the effectiveness of their use for muscle recovery and injury prevention. The
most effective devices appear to be those that mimic normal human physiol-
ogy and provide segmented and milking compression.

Dry Needling
Dry needling is a medical treatment modality to produce inoculations in the
injured tissue leading to increased blood and nutrient delivery. Besides caus-
ing localized response in the affected tissue, dry needling is also believed to
create nervous system stimulation leading to endorphin release and reduced
transmission of pain signals.
Dry needling requires specialized training and is performed by physicians,
physical therapists, and chiropractors. It is typically performed every 48 to
72 hours to allow continuation of the process.
Minimal research is available about using dry needling for treatment in
overuse injuries, although it appears promising for overuse tendon and muscle
injuries. Little to no pain is involved with this technique, but some injuries or
conditions such as injuries to the chest or abdomen may be contraindicated
secondary to use of needles.

Electrotherapy
Electrotherapy agents and electrical stimulation are used in a variety of
forms (high-volt galvanic, Russian, TENS, microcurrent, ultrasound, laser
therapy, and others). Electrotherapy is typically used in physical therapy
clinics, athletic training rooms, and chiropractic offices to help alleviate pain
by blocking pain signals, reduce swelling, or activate an atrophied or weak
muscle. These modalities are typically used within the first 1 to 2 weeks
after an acute injury such as an ankle sprain or postsurgery to reduce pain
and swelling.
Triathlon Injury Recovery Techniques | 503

High-voltage galvanic electrical stimulation has been shown to heal


wounds and reduce swelling. Russian electrical stimulation is useful in the
rehabilitation of an atrophied muscle such as the quadriceps after knee sur-
gery. Electrical stimulation in any form does not fix or cure injury and is only
an adjunct to other treatments. Low-level laser therapy is the application of
light by a low-powered laser to an injury. Its clinical effectiveness is debated,
and agreement is limited about the appropriate settings or parameters of the
machines.

Instrument-Assisted Tissue Mobilization


Instrument-assisted tissue mobilization, such as augmented soft-tissue
mobilization (ASTYM) and Graston, is used by certified clinicians to address
fibrotic adhesions that develop in a tendon because of prolonged stress and
trauma over time.
Basic medical science describes an injured tendon (tendinopathy) as a
chronic or degenerative condition rather than inflammation. This degenerative
condition creates abnormal orientation of tendon fibers and scar-tissue-like
properties. An analogy used to describe what is occurring during this type
of treatment is that the injured tissue is like spaghetti in a bowl, whereas
normal tissue should look like spaghetti in a box.
ASTYM may be used only by physical therapists, whereas Graston may
be used by chiropractors or physical therapists. ASTYM uses plastic tools,
graded pressure, and a treatment philosophy that focuses on stimulating the
body’s natural healing process and includes specifically prescribed exercises.
Graston uses metal tools and graded pressure with or without exercises.
These instrument-assisted techniques are most beneficial for tendinosis,
tendinitis, tenosynovitis, muscle strains, carpal tunnel, and postop scarring.
They require as little as 1 treatment to 12 treatments to resolve the issue.7
They may also accelerate early healing following an acute ligament injury.8
An added bonus of these two treatments is that the athlete is instructed to
continue with training and exercise, albeit with possible limitations depend-
ing on severity of injury.

Spinal Manipulations and Mobilizations


Spinal mobilizations and manipulations are performed by physical thera-
pists, chiropractors, and osteopathic physicians. These manual techniques
(clinicians using their hands to treat) are used to increase joint mobility and
motion. Typically, they are used to treat spinal injuries and pain, although
they may be used effectively in other joints.
Thoracic (midback) thrust manipulations have been shown to provide
greater improvements in pain, cervical (neck) range of motion, and disability
in the short and long term when compared with passive treatment involving
electrical stimulation (TENS) and deep heat (infrared).9
504 } Koch

With a lesser degree of effectiveness, the athlete can perform self-mobi-


lization of a joint to increase joint mobility and range of motion. Examples
would include thoracic spine (midback) stretches over a half foam roller for
shoulder and neck injuries or prone press-ups for low-back and disc injuries.
Thoracic extension over half foam (figure 36.4) and press-ups (figure 36.5) are
described in chapter 35.

„„Figure 36.4 Thoracic extension over half foam. See chapter 35.

„„Figure 36.5 Press-ups. See chapter 35.

Massage
Massage is extremely variable in technique and form throughout the world
of massage therapy. Although little research supports the use of massage
as a long-term recovery technique, it may serve as a good adjunct to other
treatment techniques because it can improve range of motion in the limbs
and spine.
Triathlon Injury Recovery Techniques | 505

Anecdotal improvements are frequently reported by athletes who use


massage as a body maintenance or prevention technique, whether using
device-assisted (i.e., foam roller) techniques or when provided by a skilled
therapist. Massage appears most helpful in the recovery of muscle strains
and fascia injuries such as ITB syndrome.
Current research evidence has not affirmed the use of massage to reduce
pain associated with delayed onset muscle soreness (DOMS), to enhance
sports performance or recovery significantly, or to play a significant role in
the rehab of sports injuries.10 Further, using massage immediately after a
hard effort or race may have detrimental effects on removal of lactate from
the muscle and hence muscle recovery.11

Nerve-Gliding Techniques
Nerve-gliding techniques are used by physical therapists to floss the sheath
that surrounds an injured peripheral nerve, such as the median nerve in carpal
tunnel syndrome or the ulnar nerve in handlebar or cyclist palsy. In other
words, these stretching techniques are prescribed by physical therapists to
assist in recovery of nerve injuries that occur away from and are unrelated
to the spine and are typically the result of excessive compression forces.

NSAIDS
Non-steroidal anti-inflammatory drugs (NSAIDS), such as Ibuprofen and
Aleve, are effective at controlling injury pain and inflammation. NSAIDS
appear to provide pain relief in the short term, but their effectiveness in the
long term has not been demonstrated. For most triathletes, a bottle of NSAIDS
is as important as swim goggles in the gear bag.
A recent study in the British Journal of Sports Medicine involving triathletes
competing in the 2008 Brazil Ironman found a high prevalence of NSAID
consumption, limited awareness of their effects and side effects, and a high
rate of nonprescribed use.12 Because these drugs are available over the coun-
ter, have a long history, and provide the ultimate ease of use (pill), they are
extremely common and often abused. Although effective for tendon, muscle,
and joint injuries, they should be used with caution. Long-term use of NSAIDS
for pain management can negatively affect the cardiovascular system, gas-
trointestinal tract, liver, and kidneys.

Orthotics
Significant debate continues regarding the effectiveness of custom and over-
the-counter orthotics in reducing and correcting overuse injury. Because of
the extreme variability of opinion among medical practitioners about process,
materials, and examination, it is not hard to understand the persistence of
this debate.
506 } Koch

Orthotics are commonly prescribed by sports medicine clinicians such


as physical therapists, chiropractors, and podiatrists. Certainly, a degree of
science and art is involved when designing custom orthotics. Significant
evidence is available proving the effectiveness of orthotics in pain reduction
and reduction of functional impairment in runners.
A recent study in Medicine and Science in Sports and Exercise found that
custom foot orthotics used in running shoes alter the muscle activation of
the peroneal muscles (higher preactivation before foot strike) in runners with
overuse injuries.13 Although this study has some limitations, it suggests that
orthotics provide enhanced dynamic control of ankle stability.
A review of the available research reveals many positive and unremarkable
or negative findings with regard to over-the-counter or prefabricated and
custom-made foot orthotics. Thus, the individualistic nature of each athlete
must be recognized, and the need should be assessed based on a dynamic
running or cycling examination.
Orthotics should be considered when dealing with patellofemoral injuries
and foot and ankle conditions. Note that orthotics for cycling and running
should be examined individually because the mechanics of how the foot and
ankle load are dramatically different in these two sports.

Tape
Clinicians and athletes use various types of tape to reduce pain and improve
performance. There is no evidence to prove that tape application has a direct
effect on athletic performance. There is likely no increase in power, muscle
firing, or proprioception (limb awareness in space) when tape is applied to
healthy subjects. Tape has some capability to increase muscle firing patterns
and decrease load on an injured knee or shoulder joint.
Note that the tape should be applied by a licensed sports medicine clinician,
not by the triathlete himself. Taping alone does not address the underlying
culprit of the injury, although it may help speed the healing process. Taping
techniques can be beneficial for the following conditions: calcaneal cuboid
syndrome, patellofemoral syndrome, plantar fasciitis, shoulder impingement,
and elbow tendinitis or tendinosis.

Topical Pharmaceuticals
Topical pharmaceuticals such as nitroglycerine, ketoprofen, and dexametha-
sone can be delivered in an absorbable patch, in a cream, or in a physical
therapy clinic through an electrotherapy method such as ultrasound or ion-
tophoresis. Nitroglycerine (NTG) is a vasodilator commonly used in heart
conditions that has been used in the form of a nitrous oxide patch, contro-
versially, in chronic tendon injuries that require increased blood flow to heal.
This treatment is most commonly used in the Achilles tendon because its
blood supply is poor.
Triathlon Injury Recovery Techniques | 507

One meta-analysis review article concluded that in chronic tendinopathy,


strong evidence supported the use of NTG for relieving pain during activi-
ties of daily living and increasing tendon strength.14 Ketoprofen and dexa-
methasone are anti-inflammatory medications that can be delivered directly
to the injured tendon, ligament, or joint by a form of electrical stimulation
(iontophoresis) or ultrasound (phonophoresis) without some of the side
effects of ingesting medication. But the evidence is questionable regarding
its effectiveness, and theoretical concerns remain about its repeated use and
long-term effects on soft tissue.

Traction
Traction or decompression is used to create disc decompression in conditions
such as herniated disc, degenerative disc, and sciatica. It can be used in the
treatment of cervical (neck) or lumbar (low-back) injuries. Typically prescribed
and performed by physical therapists or chiropractors, spinal decompression
is achieved by a traction or decompression machine.
Traction is a conservative treatment that should be attempted before
considering spinal injections or surgery. Traction appears most effective for
spinal conditions that refer pain into the extremities and have signs of nerve
root compression such as numbness and tingling.

Invasive Clinician-Assisted Treatments


The following injury recovery methods are more aggressive in nature or inva-
sive. These treatments may be more controversial in the medical literature,
have more significant side effects, and should typically be considered after
conservative care has failed.

Corticosteroid Injections
Corticosteroid injections (sometimes referred to as cortisone or steroid injec-
tions) are the most commonly used injections in the treatment of sports
injury and arthritis. Despite the effectiveness of corticosteroid injections in
the short term, noncorticosteroid injections offer fewer immediate and long-
term side effects.
A review of the medical literature involving complications associated with
the use of corticosteroids in the Clinical Journal of Sports Medicine revealed
that the existing medical literature does not provide precise estimates for
complication rates following the therapeutic use of injected or systemic cor-
ticosteroids in the treatment of athletic injuries. Tendon and fascial ruptures
are often reported complications of injected corticosteroids, whereas tibial
stress fractures and multifocal osteonecrosis (bone death) were described
with systemic corticosteroids (i.e., drugs taken by mouth).15
508 } Koch

Hyaluronic Acid Injections


Hyaluronic acid (HA) injections, commonly referred to as viscosupplemen-
tation, are injections indicated for the treatment of pain in osteoarthritis of
the knee in patients who have failed to respond adequately to conservative
care. These joint-lubricating injections must be prescribed and performed by
a trained physician. Although significant improvement appears to occur in
the treatment of joint arthritis, more research is needed to determine effec-
tiveness in other joints and injuries.

Platelet-Rich Plasma Therapy


Platelet-rich plasma therapy, or PRP therapy, is a relatively new and con-
troversial injection therapy that uses the athlete’s own blood to assist with
tissue healing. The first available study about using PRP therapy for Achil-
les tendonitis did not exhibit favorable results.16 More research is needed on
how effective it is, what conditions are appropriate, what specific quantity
of cells should be used, and how many and over what timeframe injections
should be given.
PRP therapy can be prescribed and performed by sports medicine physi-
cians. The high cost of treatment and the amount of time off from training
(ranges from 1 week to months depending on physician protocol) are typi-
cally the primary concerns from athletes. Platelet-rich plasma has also been
used for chronic conditions such as plantar fasciitis, ankle sprains, lateral
epicondylitis, and hamstring insertion strains.

Prolotherapy
Prolotherapy is an injection therapy that typically uses a sugar-based com-
pound that is intended to irritate the region and promote tissue healing. It is
not effective for complete tendon tears and is minimally effective for degen-
erative joint disease. More research is needed to identify the conditions that
respond best to these injections. They are used primarily by naturopathic
physicians, who have reported success for tendon and ligament injuries.
Eccentric loading exercises combined with prolotherapy may provide more
rapid improvements in symptoms than eccentrics alone.17

Stem Cell Therapy


Stem cell therapy (autologus bone-marrow-derived mesenchymal stem cells) is
a controversial topic that may have potential either to fade away or to become
the latest and greatest in sports injury recovery. It is currently being used
experimentally for injuries to cartilage, disc, bone, tendon, and ligament.
Little research is currently available about the effectiveness in the short
and long term. Much like PRP therapy, this treatment is cost prohibitive
Triathlon Injury Recovery Techniques | 509

and requires athletes to refrain from training for a significant time. A recent
study reported that reinjury rates in horses following tendon injury are high
with conventional treatment but that the use of stem cells has resulted in a
significant reduction in tendon reinjury in National Hunt racehorses.18

Surgery
In most cases, surgery is a last resort, because it never restores the injured
area to normal. An athlete dedicated to training will need equal dedication to
conservative treatment measures before resorting to surgery. Unfortunately,
some injuries will probably fail to respond to conservative measures.
For instance, surgery is indicated if a tendon is completely torn. Most par-
tial tears can be rehabbed and must be given 3 months of rehab at the very
least before surgery is considered. If a tendon is completely torn, surgery is
the only available option to restore function to the tendon and its muscle.
Common examples would be rotator cuff tear, patellar tendon tear, and
Achilles tendon tear.
Additional injuries that should have surgery as a viable option would be
disc herniation that causes muscle weakness, cartilage (articular or hyaline)
damage in a joint, some grade 3 ligament tears, displaced or unstable frac-
tures, acute compartment syndrome, and athletic pubalgia, or sports hernia.
In any of the aforementioned cases the triathlete needs to seek the consult
of an experienced orthopedic surgeon who has done a significant number
of the indicated surgery and understands the triathlete’s goals. Setting
realistic goals for return to sport with the surgeon is critical to a satisfactory
outcome following surgery.

Physical Exam
The final and most critical piece of the injury recovery process is what should
happen before using a treatment method with a triathlete: the physical exam.
A physical examination to determine cause is the most critical injury
recovery technique. The exam should include a thorough musculoskeletal
and biomechanical assessment to look for muscular imbalances, asymmetric
motion or strength, abnormal postural conditions such as a leg-length dis-
crepancy or scoliosis, and a review of injury and training history.
Evaluating and treating the entire chain of movement is usually neces-
sary. The assessment would ideally include video analysis of movement to
understand how the triathlete’s individual body functions during a specific
activity. The sports medicine professional treating the injured triathlete must
understand the biomechanics of swimming, cycling, and running to fix the
injury effectively and prevent reoccurrence.
For example, in the case of a recent study on patellofemoral pain syndrome
in runners, gait retraining using real-time feedback to improve hip mechanics
510 } Koch

was used as an evaluation tool and treatment.19 Results revealed a significant


decrease in knee pain and improved function. The results suggest that treat-
ment strategies for patellofemoral pain syndrome should focus on addressing
the underlying mechanics associated with this injury.
This example of assessing movement to address the underlying cause or
dysfunction related to an injury should be equally applied to swimming,
cycling, and running. Complicated movement analysis of a specific sport
requires a highly trained professional with years of experience.

Treatment Suggestions
After a biomechanical assessment, a combination of injury recovery strategies
works most effectively and gets the athlete back to training faster. The follow-
ing treatment suggestions can provide some guidance when addressing an
injury specific to the type of tissue injured. These suggestions are common
treatment approaches by sports medicine clinicians who assist triathletes.
They are not intended to supersede the diagnosis and treatment plan of a
health care professional.

Tendon injuries (Achilles tendinosis)


1. Augmented soft-tissue mobilization—tool assisted (ASTYM)
2. Dry needling
3. Eccentric strengthening
If no improvement occurs over the course of a couple of months with con-
servative measures, rule out a tear in the tendon and consider prolotherapy,
nitric oxide patches, or PRP therapy.
Muscle injuries (hamstring strain or tear)
1. Augmented soft-tissue mobilization—tool assisted (ASTYM) or massage
2. Isolated strengthening (injured muscle) and core strengthening
3. Modalities (electrotherapy, infrared, and cold laser technology)
Ligament injuries (ankle sprain)
1. Augmented soft-tissue mobilization—tool assisted (ASTYM)
2. Compression machine because this type of injury typically results in
swelling
3. Taping techniques
If no improvement occurs after a couple of months with conservative
measures, rule out a complete tear and consider prolotherapy or PRP therapy.
Triathlon Injury Recovery Techniques | 511

Fascial injuries (plantar fasciitis or fasciosis)


1. Augmented soft-tissue mobilization—tool assisted (ASTYM) or massage
2. Isolated and core strengthening
3. Static and dynamic stretching
Bone injuries (stress fracture)*
1. Water running or antigravity treadmill running
2. Core strengthening exercises to address muscle imbalances
3. Assessment of nutrition, running gait, and footwear
Cartilage injuries (meniscus tear in the knee)
1. Core and specific joint-isolated strengthening exercises
2. Injection therapies (prolotherapy, PRP, cortisone, stem cell), mostly
experimental
3. Surgery
Dysfunctional joint movement (patellofemoral syndrome)
1. Core strengthening with particular attention to stabilizing the muscu-
lature that controls the injured joint
2. Taping techniques
3. Massage and stretching to address soft-tissue restrictions (i.e., ITB)
Spinal disc injuries (herniated disc in the lumbar spine or sciatica)
1. Spinal mobilization techniques
2. Spinal stabilization and core strengthening exercises
3. Spinal decompression or traction
If no improvement occurs after a couple of months with conservative
measures, further assessment may be required to determine the benefit of
injections (prolotherapy, PRP, or cortisone).

Conclusion
When injury strikes a triathlete, she typically hits the panic button faster
than the speed of light. But the triathlete must take a deep breath, step
back and review the options, search for the cause, and seek the counsel of
experienced sports medicine professionals. This chapter provides the injured
triathlete with greater knowledge of the treatment tools available. Ultimately,

*Requires diagnostic imaging


512 } Koch

overcoming injury requires a specific plan, much like a race plan, that must be
adhered to if the outcome is to be successful. This process becomes extremely
difficult when the defined recovery plan takes longer than initially expected
or interferes with race goals. But triathletes are known for their persistence,
and never is this trait more important than when they face injury.
Pa r t

X
Nutrition
for Triathletes
This page intentionally left blank.
CHaPTEr
37
Energy Needs, Sources,
and Utilization
—Bob Seebohar, MSc, RD

E fficiency is a commonly used term in the sport of triathlon. From a physical


standpoint, triathletes seek to improve their efficiency in moving their
bodies in an effort to use less oxygen. Participating in triathlon training leads
to a high degree of energy expenditure, so triathletes must replenish their
energy needs efficiently to fuel their bodies properly.
From a nutrition perspective, being efficient means being able to use the
proper nutrients stored in the body at the right times throughout training
based on the duration and intensity. Efficiency is about maximizing the use
of some nutrients while preserving others to provide sustained energy levels
to support performance.

Metabolic Efficiency
The oxidation of fat by the mitochondria is the main source of energy when
the intensity of exercise is low, typically defined in scientific research as rang-
ing from 35 to 65 percent of maximum intensity.1, 2, 3 Depending on gender
and size, a triathlete has about 1,300 to 2,000 calories stored as carbohydrate,
which is distributed in the liver, muscles, and blood.
Under normal conditions, the body has enough glycogen stores to fuel 2
to 3 hours of continuous training at moderate intensity.4 But for durations
longer than this or for higher intensity training sessions, the body must either
receive energy from supplemental sources or be efficient at using its internal
stores. Note that as the intensity of exercise increases, the working muscles
require more blood flow, thus diverting blood flow from the digestive tract.
This effect is often referred to as the blood shunting response.
When a triathlete consumes a large amount of calories under these condi-
tions, the body must decide whether to redistribute blood back to the digestive
tract and take away some of the oxygen-rich blood from the muscles used to

515
516 } Seebohar

exercise or to keep the blood in the muscles and risk not being able to digest
the calories consumed. By improving the body’s efficiency at using its internal
stores of fat and preserving its limited stores of carbohydrate, triathletes can
reduce the amount of calories ingested per hour, supply enough energy to the
body for locomotion, and decrease the risk of gastrointestinal (GI) distress.
The term used to describe this ability is metabolic efficiency.5
To improve the body’s efficiency in using its stored nutrients, a mixture
of macronutrients is required at each meal or snack to provide proper blood
sugar control. Lean protein and fiber, found in carbohydrate-rich foods such
as fruits, vegetables, and whole grains, are the two main components in pro-
viding blood sugar stabilization. Stable blood sugar will lead to enhanced
fat oxidation throughout the day, which will reduce the amount of calories
that need to be consumed during exercise after the body adapts to a higher
fat oxidation rate.

Macronutrients
An understanding of macronutrients is needed to have proper foundational
knowledge before implementing a nutrition periodization or weight manage-
ment plan. Macronutrients consist of carbohydrate, protein, fat, water, vita-
mins, and minerals. Although they all have different functions in the body,
they interact with one another to support the triathlete’s training program.

Carbohydrate
Carbohydrate is the ultimate source of energy for the body and brain. Carbo-
hydrate is vital to triathletes because of its ability to maintain energy levels
and high cognitive functioning. Besides providing energy, carbohydrate plays
a protein-sparing role. When glycogen (stored carbohydrate) levels become
low, the body begins to make glucose from protein and fat, but that process is
inefficient and uses energy to make only a small amount of additional energy.

Simple and Complex Carbohydrates


The two main classifications of carbohydrates, simple and complex, have dif-
ferent chemical structures. Simple carbohydrates, often referred to as simple
sugars, are made up of short chains of sugars and are classified as mono-, di-,
and oligosaccharides. Complex carbohydrates are made up of longer chains,
also called polysaccharides, and include starch and fiber.
Monosaccharides include glucose, fructose, and galactose. These are the
simplest form of carbohydrate. They are made up of one simple sugar mol-
ecule and are the easiest to digest.
Disaccharides include sucrose (glucose, fructose), lactose (glucose, galac-
tose), and maltose (glucose, glucose). These are made up of two monosac-
charides and are not as easy to digest.
Energy Needs, Sources, and Utilization | 517

Oligosaccharides contain three to nine monosaccharides and include malto-


dextrins, corn syrup, and high-fructose corn syrup. Artificially sweetened
foods often contain oligosaccharides. The digestion rate of oligosaccharides
is longer than that of the two previously mentioned.
Polysaccharides are often referred to as complex carbohydrates. They
include starch and fiber. They are made up of long, complex chains of sugars
and require more complex mechanisms for digestion than simple carbohy-
drates do, which extends their digestion time.

Molecular Weight Carbohydrates


Another fairly new topic in the performance arena pertaining to carbohydrate
is high molecular weight carbohydrates. The reason that this type of carbohy-
drate is important is related to blood sugar regulation and movement. Blood
glucose concentrations are influenced by the movement of glucose from the
stomach to the intestine and into the blood.
This process is important because the osmolality of a solution has been
shown to influence the gastric emptying rate from the stomach. Gastric emp-
tying times can be of extreme importance to triathletes during training and
competition. Carbohydrate sources with high osmolality may delay glucose
transportation by slowing gastric emptying. A steady supply of glucose
without much delay of delivery to the blood and the working muscles and
brain is ideal to maintain physical and cognitive functioning.
High molecular weight (HMW) carbohydrates, polyglucosides or glucose
polymers, have low osmolality, which make their entry and exit time out of
the stomach quick and efficient.

Glycemic Index
Another important, and somewhat controversial, area in carbohydrate knowl-
edge is glycemic index. Each sugar has a different glycemic index value, which
means that it will be digested faster or slower. The glycemic index (GI) of foods
can be affected by many variables, such as preparation method, presence of
protein and fat, and fiber content. It is rare to find a food that has only one sugar
in it that will not be affected by other factors that alter its glycemic index value.
Even more important is understanding the various parts that make up
the GI. The glycemic response (GR) of a food is a measure of the ability of
that food to raise blood sugar. The two main components in determining GR
are GI and glycemic load (GL). Many triathletes have used GI at some point
in their training to maintain steady energy levels throughout the day or to
increase glycogen storage after a quality workout, but unfortunately they do
not often use the GL in conjunction with it; therefore, they are plagued with
misinformation and are using only half of the equation.
GI is a standard measure of how quickly 50 grams of carbohydrate of a par-
ticular food are converted to sugar and thus affect blood sugar over a 2-hour
518 } Seebohar

period. In the past, simple carbohydrates were classified as having a high GI,
whereas complex carbohydrates were classified as providing sustained energy
with a gentler rise in blood sugar. Nonrefined, or wholesome, carbohydrates
are typically preferred because they provide important nutrients whereas
refined carbohydrates do not. But many triathletes correlate wholesome with
complex and simple with refined, which is sometimes not accurate.
For example, fruit is considered a simple carbohydrate, but the carbohy-
drate in fruit is unrefined and fruit is dense in nutrients and full of vitamins,
minerals, and fiber. Each food produces its own blood sugar profile, and the
correlation is not strong with whether it contains simple or complex carbo-
hydrates. Some complex carbohydrates can be digested, absorbed, and used
as quickly as simple sugars can, meaning that they have similar glycemic
responses.
Many factors aside from being classified as a simple or complex sugar can
affect the GI of foods. Some of these are the following:
• Biochemical structure of the carbohydrate
• Absorption process
• Size of meal
• Degree of processing
• Contents and timing of previous meal
• Fat, fiber, and protein content
• Ripeness

Glycemic Response
The GI of a food reveals how fast a carbohydrate will increase blood sugar
levels, but it does not provide the information regarding how much of that
carbohydrate is in a serving of that food. Hence, the importance of GL enters
the equation. Both the GI and GL are needed to determine the glycemic
response of a food.
Glycemic load is the numerical value of the GI divided by 100 and multi-
plied by the available carbohydrate content of the food in grams. GL takes
the GI into account, but it is based on how much carbohydrate is in the food
or drink. The GL is numerically lower than the glycemic index.
Here is an example. Watermelon has a GI of 72. A recommended serving
of half a cup is 4 ounces, or 120 grams, of watermelon. This serving has 6
grams of carbohydrate. To calculate the GL, divide the GI by 100 and multiply
by the carbohydrate content in grams:
(72/100) × 6 = 4.32, or 4 when rounded
In this example, a high-GI food becomes a low-GL food. Thus, based on
the serving size and quantity eaten, watermelon will have a better GR and
therefore result in a lower rise in blood sugar than its GI indicates. Keep in
Energy Needs, Sources, and Utilization | 519

mind that as the serving size of a food increases, the amount of carbohydrate
also increases, which will increase the GR of the food.
In many cases, GL is not based on a typical amount of food eaten, so GL
does not provide realistic information unless the food is weighed before
consumption. The important take-home lesson about GL is that it provides
an understanding of the relationship between a specific amount of food and
its biochemical response.

Protein
Protein provides the muscles with the amino acids needed to resynthe-
size and rebuild new muscle cells. In the body, protein plays key roles in
chemical reactions, hormone structure, antibodies, fluid and electrolyte
balance, and structural components of the body such as muscles, tendons,
and ligaments.
A protein is composed of amino acids. Each amino acid consists of an
amine group (the nitrogen-containing part), a carbon atom with a side chain,
and an acid. Each amino acid has a specific side chain that gives it its identity
and chemical nature. Amino acids can also be used to build new proteins.
If the body has a surplus of amino acids and energy, the carbon backbone
of the amino acid is converted to fat and the nitrogen is excreted in the urine.
The protein from food travels to the stomach where stomach acid separates
the protein strands into shorter strands and amino acids. These then travel
to the small intestine where they are broken down further and absorbed
into the blood for delivery to the cells in need of amino acids.
The two types of proteins are classified as essential and nonessential. By
definition, essential proteins provide all the amino acids that the body cannot
make and thus must be obtained through food. These proteins are found in
abundance in any animal product such as meat, fish, and dairy products;
soy products; and some grains such as quinoa. The essential amino acids
are histidine, isoleucine (branched chain amino acid), leucine (branched
chain amino acid), valine (branched chain amino acid), lysine, methionine,
phenyalanine, threonine, and tryptophan.
Nonessential proteins provide the amino acids that the body can make.
These proteins are found in foods such as legumes, seeds and nuts, grains,
and vegetables. The nonessential amino acids are alanine, arginine, aspara-
gine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, proline, serine,
and tyrosine.

Fat
The body has abundant stores of fat, in excess of 80,000 calories for some tri-
athletes.6 These almost unlimited stores can be used as energy. Fat is essential
for body processes such as body insulation, internal organ protection, nerve
transmission, and metabolizing fat-soluble vitamins.
520 } Seebohar

The two classifications of fats are unsaturated and saturated. Unsaturated


fats have three additional subcategories: monounsaturated, polyunsaturated,
and trans fats. Chemically, saturated fats have most of their fatty acids satu-
rated with hydrogen. Unsaturated fats have one or more fatty acids that are
unsaturated.
The unsaturated fats, specifically the polyunsaturates omega-6s and omega-
3s, have gained much popularity among triathletes. These unsaturated fats
cannot be made by the body because the cells cannot convert one fat to another.
Thus, triathletes must receive these fats through food or by supplementation.
Omega-6 fats do not normally need to be supplemented in the diet because
triathletes receive so many through processed and packaged foods. Omega-3
fats should be increased in the diet. Good food examples include walnuts,
salmon, flax, and chia seeds, to name a few. Because food sources are limited,
especially for omega-3 fats, taking supplements is usually the first choice for
triathletes who want to increase dietary intake of these beneficial fats.
Essential fatty acids (EFA) have many important functions in the body,
including the following:
• Regulation of blood pressure and blood lipids
• Formation of blood clots
• Assistance in the immune response
• Improvement of the inflammation profile
A common misconception among triathletes is that polyunsaturated fats
are created equal. In fact, the opposite is true, and understanding the bio-
chemical differences is important. The most significant difference is in the
metabolic process of these two fats. In the biochemical metabolic pathway
of the omega-3 and omega-6 fats, there is one step in which both compete
for one shared enzyme. This process can be problematic because omega-6
fats predominate in the eating program of most triathletes. These fats are
preferentially metabolized instead of omega-3 fats.
The result is a metabolic dominance of omega-6 fats, which can have a
negative effect on health and performance, specifically the inflammatory
response. To alter this situation in a favorable way, triathletes should reduce
their omega-6 fat intake and increase the amount of omega-3 fat that they
receive through food and supplementation. This change will lead to a better
metabolic conversion that favors improved health by providing more of the
beneficial compounds EPA and DHA (eicosapentaenoic and docosahexaenoic
acids) to the body.

Water
Of the essential nutrients required for life, water is by far the most important.
Water makes up 60 to 75 percent of total body weight.7 Water improves fluid
balance, acts in the blood as a transport mechanism, eliminates metabolic
Energy Needs, Sources, and Utilization | 521

waste products, dissipates heat, helps to digest food, and lubricates joints.
Water is an essential nutrient that is crucial to survival as well as athletic
performance.
An interesting side note on the topic of water is thirst. Thirst is defined as
a conscious awareness of the desire for water and other fluids, and it usu-
ally controls water intake. The physiological drive to drink is controlled by a
decrease in blood volume, an increase in blood osmolarity (the total concentra-
tion of particles in solute), and a decrease in the flow of saliva.8 These markers
ultimately lead to sending a signal to the brain, which will increase thirst.

Vitamins and Minerals


Vitamins are metabolic catalysts that regulate biochemical reactions within
the body. They are catalysts needed for metabolic processes to occur, but they
do not provide energy directly. Vitamins are extremely important for sustain-
ing optimal health and can play a significant role in athletic performance.
Minerals are elements that combine to form structures of the body and
regulate body processes. Minerals are found in abundance in most foods, but
they do not serve as a source of energy. The minerals magnesium, sodium,
calcium, potassium, zinc, and iron are the most popular among triathletes
because of their effects on hydration and cramping, oxygen delivery, and
immune system health. Minerals are discussed in chapter 40.

Weight Management
and Body Composition
Many triathletes want to manipulate their body weight or body composition
at some point in their sport participation for a variety of reasons including
aesthetic, performance, or health. The strategies used are of key importance,
and although these plans must be individualized for each triathlete, some
general recommendations are useful.
Although the energy balance equation (calories in versus calories out) can
be a valuable education tool, many triathletes fail to realize that manipulat-
ing body weight is never as easy as counting calories because macronutrients
have different metabolic fates in the body. Rather than counting calories and
attempting to configure daily deficits and surpluses of calories, an easier
approach is to learn about the mind and body connection associated with
food. The quantity of macronutrients that triathletes should eat is discussed
in chapter 38, and the strategy is applicable to quantifying the volume of food
to eat on a daily basis for specific training cycles.
The most important step when it comes to controlling and manipulating
body weight is to relearn the body’s cues for hunger and satiety. Letting
hunger and satiety guide feeding patterns is difficult at first, because of the
psychological and emotional connection to food and the disconnect with
522 } Seebohar

cognitive decision making about food choices that triathletes lose with
age. This change will lead to long-term success because it allows the body’s
physiological systems to align with its psychological systems. The point is that
people need to let hunger and satiety be their primary guides to developing
a healthy relationship with food.
Instinctual eating is a concept that describes feeding the body when it is
biologically hungry. Biological hunger can be identified when physical hunger
pangs exist in the stomach or when cognitive functioning and the ability to
concentrate and focus decrease. Letting hunger and satiety be the primary
guides to feeding will allow a healthy relationship with food to develop. The
cornerstone of using this concept and changing body weight or body com-
position is controlling blood sugar. When blood sugar is high, the hormone
insulin is released. When that happens, the body’s ability to use fat as fuel
is significantly diminished.9 Preventing high ebbs and flows of blood sugar
will help control insulin release, which will in turn help regulate body weight
and body composition.

Controlling Blood Sugar


Controlling blood sugar through the manipulation of diet is much simpler
than triathletes expect. As stated previously, eating sources of lean protein
and fiber together will produce a blood sugar stabilization effect and improve
satiety while improving the body’s ability to use fat and preserve carbohy-
drate. Eating a diet high in carbohydrate or not including good sources of
protein with carbohydrate will promote the oxidation of carbohydrate and
thus cause consistent ebbs and flows of blood sugar. This fluctuation will not
contribute to satiety or optimal fat oxidation opportunities.
Additionally, focusing on a metabolically efficient eating plan will improve
the body’s thermogenic response. This concept describes the metabolic heat
loss associated with food intake. Protein has the highest heat loss, approxi-
mately 20 to 30 percent, followed by carbohydrate at approximately 5 to 10
percent and fat at approximately 3 to 5 percent.10 Many triathletes shy away
from eating lean protein because they are consuming a high-carbohydrate
diet, but eating a bit more lean protein will lead to higher metabolic heat loss,
which will aid in achieving goals pertaining to weight loss or body fat loss.
In normal nondisease states, the body’s blood sugar ebbs and flows about
every 3 to 4 hours. Thus, by following the eating plan described earlier, tri-
athletes should be feeding their bodies along these intervals throughout the
day, depending on the timing of training sessions. Triathletes who tend to eat
every 1 to 2 hours usually do so because blood sugar is not stable. Eating a
higher carbohydrate load will lead to a higher degree of carbohydrate oxida-
tion, which means that the body requires more supplemental carbohydrate
consumed through food at more frequent intervals.
Energy Needs, Sources, and Utilization | 523

Weight Loss Strategies


To complement the hunger and satiety cues and responses offered here, the
following chapter provides quantitative implementation strategies associated
with weight loss in the discussion of the concept of nutrition periodization.
These weight loss strategies usually include using the energy balance equation
to track calories in versus calories out. Many triathletes do this with online
calculators or software programs. Although keeping a food and training
log for energy consumption and expenditure reasons is educational and
may assist the triathlete in making a behavior change, long-term success is
usually not achievable.
Understanding nutrition facts labels, ingredients, and serving sizes of food
is valuable, but when triathletes use numbers without first understanding
their physiological hunger response, they tend to yo-yo with weight and
body composition because of lack of connection between the body and mind.
The nutrition periodization quantitative prescription for weight loss is an
important piece of weight management, but it should not be used before a
clear association is understood with hunger and satiety.

Conclusion
Having a good understanding of the various macronutrients and their
importance in the triathlete’s body is paramount to sustaining optimal per-
formance. The information in this chapter can help develop the ability to use
nutrients in a methodical manner to fuel and recover from training sessions
and improve health.
This page intentionally left blank.
CHAPTER
38
Nutrition Periodization
—Bob Seebohar, MSc, RD

T he future of sports nutrition embraces the relationship of aligning a tri-


athlete’s daily nutrition plan with her physical training plan, a concept
termed nutrition periodization.1 This process will promote the physiological
adaptations with the training response as volume and intensity change from
day to day, week to week, and month to month.
Instead of focusing on proper nutrition in the days leading up to a com-
petition, nutrition periodization should be used to support the training load
shifts that occur throughout the year. Each time the physical training program
cycle changes, the nutrition program should change to support the changing
energy needs. The goal of nutrition periodization is to provide the body the
nutrients needed to improve health, strength, speed, power, and endurance
while helping to maintain a healthy immune system, body weight, and body
composition.
The easier a nutrition plan that a triathlete can have, the more successful
he will be at implementing it on a year-round basis. Further, he will not be
subject to the negative effects of dietary fads. The goals of nutrition periodiza-
tion are simple and straightforward:
• To enhance health
• To improve performance
• To manipulate body weight and composition
Triathletes who are not nutritionally prepared before training sessions
will likely not receive the same positive physiological training adaptations as
triathletes who are prepared and place nutrition on the top of their priority
list. Adequately fueling the body before, during, and after training sessions
allows the body to accomplish the physiological objectives of the workout in
addition to promoting a faster recovery.
Daily macronutrient ranges have been published in scientific research
pertaining to endurance athletes. These numbers provide triathletes, coaches,
and health professionals a starting place for how to build a triathlete’s daily
nutrition plan. The ranges are large but will be properly separated into more
manageable ranges based on training cycles.

525
526 } Seebohar

The ranges seen in research, from 3 to 19 grams of carbohydrate per kilogram


of body weight, from 1.2 to 2.5 grams of protein per kilogram of body weight,
and from 0.8 to 3.0 grams of fat per kilogram of body weight,2, 3, 4, 5, 6 are large
because of the many differences among triathletes, body weight and body
composition changes, and competition cycles. Because of these differences, it
cannot be stated which macronutrient range a triathlete should follow based
purely on gender or even the race distance for which the triathlete trains.
The training cycle and the physical demands along with body weight and
body composition goals should be considered when devising a nutrition plan.
What follows is a description of how to use the general concept of nutrition
periodization and the numbers within each of the training cycles that tri-
athletes will follow throughout the year, from the overall macrocycle down
to the daily microcycle.

Macrocycle Nutrition Guidelines


A handful of nutrition guidelines apply year round on the macrocycle level
of planning. Specifically, managing oxidative stress, inflammation, and iron
deficiency should be main focal points throughout a triathlete’s annual train-
ing plan.

Oxidative Stress
Oxidative stress, sometimes referred to by researchers as reactive oxygen
species, is more commonly known to triathletes as free radicals. Free radicals
are atoms with unpaired electrons that can be formed when oxygen interacts
with certain molecules. These highly reactive free radicals can impart damage
to the cell membrane and DNA.
When this happens, cellular functioning can decrease, which can affect
many body processes. The body has internal enzyme systems that help to
scavenge these free radicals, but the body can be overloaded at certain times,
such as when performing strenuous exercise, being at altitude, being in a
polluted environment, or when experiencing stress.
Antioxidants act as scavengers and interact with free radicals to stop the
destructive chain reaction of cellular damage. Triathletes should choose foods
year round that are rich in beta-carotene, vitamin E, vitamin C, selenium, and
zinc because they can help support immune function and quench free radi-
cal production. Acquiring these antioxidants through food is the preferred
method rather than relying solely on supplements.

Inflammation
Aerobic capacity declines when the body is in an inflammatory state. This
condition can be detrimental to athletic performance because systemic inflam-
mation can have a negative effect on both health and performance.
Nutrition Periodization | 527

Inflammation includes many aspects, but in terms of athletics, the main


detriment to performance is linked to the endothelial lining of the artery
and the inflammation that incurs there. When this lining becomes inflamed,
vasoconstriction is greater and less blood is delivered to working muscles.
Less blood flow equates to less oxygen and nutrient delivery and less waste
removal from the muscles.
Triathletes encounter bacteria and viruses on a regular basis, but the body
is typically well equipped to attack and destroy these invaders through the
inflammation response. A certain amount of inflammation in the body is
beneficial, but excess inflammation begins to cause problems.
The two main types of inflammation are classic and silent. Classic inflam-
mation is the most common because it is the type that is indicated by a bruise.
Triathletes have undoubtedly experienced this type of inflammation because of
falls, scrapes, and muscle contusions. In situations like this, the body is typically
prepared to handle the recovery process. Additionally, triathletes have used
the concept of rest, ice, compression, and elevation to promote faster recovery.
Silent inflammation, which has no visual symptoms such as those found
in classic inflammation, can have significant negative health and performance
effects. This type of inflammation is often involved in disease states such as
heart disease, Alzheimer’s, rheumatoid arthritis, and cancer.
One of the methods of determining the amount of silent inflammation in
the body is through blood testing that looks at the level of C-reactive protein
(CRP) in the body. CRP is a marker of total-body inflammation. This type of
inflammation is more difficult to manage, which is why improving the overall
daily nutrition plan to include more anti-inflammatory omega-3 fats is highly
beneficial.
The differences between the omega-6 and omega-3 fats were explained in
chapter 37, but to appreciate the importance of these fats and their role in inflam-
mation, more of the science should be understood. The metabolism and effects
of fatty acids are complex, and understanding what happens when certain types
of fats become more or less initiated in their biochemical pathways is important.
Omega-6 fats, found in corn, safflower, sunflower, and soybean oils, are
separated into two main constituents through their metabolic process: arachi-
donic acid (AA) and gamma-linolenic acid (GLA). Having an overabundance
of AA leads to a higher degree of inflammation. AA is usually high because
triathletes typically consume many of the oils listed.
GLA is an omega-6 fat, but it has anti-inflammatory properties because it
is not converted to the proinflammatory AA compound. Instead, GLA is con-
verted to dihomogamma-linolenic acid (DGLA), which competes with AA in
the metabolic pathway. If DGLA is present more than AA, DGLA will negate
the proinflammatory effects of AA.
The underlying message is that triathletes should reduce their consump-
tion of oils and foods rich in omega-6 and increase their intake of omega-3
food sources, such as salmon, walnuts, flax, and chia seeds. The metabolic
528 } Seebohar

fate of omega-3 fats concludes with the compounds EPA and DHA but only
if omega-6 fats are reduced so that the shared enzyme can allow omega-3 to
convert to EPA and DHA rather than allow omega-6 fats to convert to AA.

Iron
The last macrocycle nutrition component is iron. Iron stores and training load
are inversely related, so as the volume and intensity increase in a triathlete’s
training program, body iron stores may decrease. Menstruating females are
more susceptible to low iron stores, but males can also be vulnerable.
About two-thirds of the iron found in the body is contained in hemoglobin,
and the rest is distributed in the liver, spleen, and bone marrow, with small
amounts in myoglobin. Some triathletes are susceptible to high iron losses,
and some are not. Several signs and symptoms may classify a triathlete as
at risk for iron deficiency or anemia:
• Unrecognized bleeding from the intestinal tract
• Hematuria (presence of blood in the urine)
• Heavy sweating (iron is found in small amounts in sweat)
• Foot strike hemolysis (destruction of red blood cells because of the impact
of the foot strike on the ground)
• Females who regularly menstruate
• Inflammation (which can lead to an increase in hepcidin, a protein that
reduces gastrointestinal iron absorption)
• Restriction of calories

Role of Iron
Iron has many roles in the body, such as being part of neural and immune
function, thyroid hormone metabolism, and mitochondrial oxidative enzymes
and being involved in the formation of new red blood cells. One of the most
important roles that iron plays is as a component of the protein hemoglobin,
which carries oxygen from the lungs to the body’s cells.
Iron stores can be hard for triathletes to maintain because the typical iron
absorption from food is rather low. Some foods may be high in iron, but their
absorption rates are low. In addition, iron inhibitors found in the diet can
decrease iron absorption even more. These inhibitors include calcium, zinc,
phytates, and fiber found in whole grains and nuts; tannins found in coffee
and tea; bran; and soy products.
Additionally, triathletes with Celiac disease or Crohn’s disease may have
lower rates of iron absorption because of the effects on the gastrointestinal
system. But on a positive note, iron promoters found in the diet promote a
higher rate of absorption. These foods include meat, fish, poultry, broccoli,
Brussels sprouts, tomatoes, potatoes, green and red peppers, and other foods
rich in vitamin C.
Nutrition Periodization | 529

The two types of iron in foods are heme and nonheme. Heme iron comes from
animal sources such as beef, chicken, shrimp, oysters, sardines, and fish and
has a high absorption rate. Nonheme iron comes from vegetable sources such
as enriched cereals, blackstrap molasses, pumpkin seeds, beans, lentils, and
tofu and has a much lower absorption rate compared with heme iron sources.

Consequences of Low Iron Stores


Triathletes often wonder what it feels like to have low iron stores. Without
a laboratory blood work analysis, the easiest way to know whether iron
deficiency is present is fatigue that worsens with exertion during exercise.
Fatigue is common in sport, of course, and it can result from many causes
such as other nutritional imbalances, illness, or stress. If a triathlete experi-
ences normal fatigue throughout the day that does not worsen with exercise,
the cause is likely not iron deficiency by itself, although a blood test should
be done to rule out this possibility.
Other signs and symptoms are associated with iron deficiency, but note
that an iron stores blood work test is needed make an accurate diagnosis of
iron deficiency or anemia. The other signs and symptoms are the following:
• Decreased performance
• Sleepiness and fatigue (outside of normal)
• Poor concentration
• Moodiness or irritability
• Always feeling cold
• Decreased immune function
• Eating nonnutritive substances such as dirt, clay, and ice (often referred
to as pica)

Three Stages of Iron Deficiency


Three stages are associated with iron deficiency, which can be determined
through laboratory blood work analysis and proper interpretation by a sports
physician or sports dietitian:
1. Stage 1, iron depletion, is the mildest form of iron deficiency. Normally,
serum ferritin levels will be decreased. If the condition is detected early
enough, this stage can be easily managed through food and does not
normally require supplementation.
2. Stage 2, iron deficiency without anemia, is characterized by a low serum
ferritin, a decrease in percent transferrin saturation, and an increase in
total iron binding capacity. This stage can also be treated with food if
caught early enough, although supplementation may be needed.
3. Stage 3, anemia, is the most serious stage. It includes all the previously
mentioned blood clinical markers in addition to a decrease in hemoglobin.
530 } Seebohar

Triathletes should be aware of the four types of anemia can be diagnosed


with laboratory blood work analysis:
1. Macrocytic: Red blood cells are larger than normal. The most common
type is called megaloblastic, or pernicious, anemia. It is caused by a
deficiency of vitamin B12 or folic acid.
2. Microcytic: Red blood cells are smaller than normal. The main cause is
iron deficiency anemia.
3. Normocytic: Red blood cells are normal in size but low in quantity and
have a low hemoglobin level. Typical causes are chronic illness, medica-
tions, pregnancy, and hemolysis.
4. Sports: This is not a clinically recognized type but is common among
triathletes in the initial phases of training. Because of higher plasma
volume (training response), iron in the body appears to be diluted and
hemoglobin levels are lower. A triathlete should not worry or supple-
ment with iron if this occurs because the body will adapt. Performance
has not been shown to be compromised.
High training loads may cause a decrease in iron stores; therefore, a triath-
lete’s nutrition should be periodized to include more iron-rich foods at certain
times of the annual training year. In a perfect world, supplementation would
not be necessary, but some triathletes may need supplementation in addition
to a sound, iron-rich nutrition program. Those with clinically diagnosed iron
deficiency anemia must consume foods high in iron as well as take an iron
supplement. If iron stores are too low, increasing those stores through food
alone is extremely difficult, especially if a triathlete is in a high training load
period or in race season.

Iron Supplements
A triathlete who is clinically diagnosed with iron deficiency or anemia may
be asked to take an iron supplement. Supplemental iron may cause gastroin-
testinal (GI) distress such as constipation, which can negatively affect train-
ing sessions. Luckily, the replenishment of iron stores typically takes about
6 to 8 weeks depending on the stage, and in cases of the first stage of iron
deficiency, supplementation may not be necessary.
Triathletes should emphasize eating iron-rich foods, pay special atten-
tion to scheduled changes in training load, and be sure to have frequent
blood tests to monitor the status of iron stores throughout the year in rela-
tion to training, competitions, and possible supplementation. Iron supple-
ments should be used wisely and only after iron stores blood testing has
been completed and evaluated by a qualified sports physician or sports
dietitian.
After the aforementioned macrocycle nutrition guidelines are understood,
mesocycle and microcycle nutrition plans and goals can be introduced.
Nutrition Periodization | 531

Base Period Nutrition Plan


As stated previously, nutrition should be aligned with the physiological goals
and objectives for each training cycle. The common foundational goals among
triathletes during the base season are improving endurance, strength, and
flexibility. Additional ancillary goals may be included, such as improving
technique and economy, but in general training intensity is low and volume
is low to moderate.
During this cycle, changing body weight or composition may be the primary
goal. Triathletes who want to pursue that goal should do so only during the
first three-quarters of the cycle so that energy demands do not exceed energy
consumption as training load increases. Refer to the section about weight
management in table 38.1 for more information.
Table 38.1 provides ranges of the macronutrients that can be used to help
build a triathlete’s eating plan.
Carbohydrate intake should range from 3 to 7 grams per kilogram of body
weight during base training depending on body weight or composition goals.
On the average, 3 or 4 grams per kilogram can be used for weight loss, and 5 to 7
grams per kilogram can be used for triathletes who do not have a weight loss goal.1
Protein intake should range from 1.2 to 2.0 grams per kilogram.1 The higher
end of the protein range may be needed for triathletes engaging in a heavier
strength training program or for those seeking weight loss because it will
have a more positive effect on satiety.
Fat intake should range from 0.8 to 1.2 grams per kilogram. For triathletes
who have weight or body composition goals, daily fat intake can range from
0.8 to 1.0 grams per kilogram, and those in a weight maintenance phase can
consume 0.9 to 1.2 grams per kilogram.1
Research does not support a definitive daily hydration recommendation,
but triathletes can use a few simple assessment tools to monitor daily hydra-
tion status. The first is urine color. The normal color of urine throughout the
day should ideally be pale yellow. Many factors can alter this color, such as
the first void of the morning and certain dietary supplements, so triathletes
should assess urine after the first void in the morning.
The second tool is frequency. Urinating every 2 to 3 hours throughout the
day will provide a rough indication of hydration status for triathletes in non-
disease states. Urinating too often or not often enough can indicate improper
hydration.

Table 38.1 Nutrition Consumption for Body Weight Goals


Body weight goal Carbohydrate Protein Fat
Weight loss 3 to 4 g/kg 1.6 to 2.0 g/kg 0.8 to 1.0 g/kg
(1.4 to 1.8 g/lb) (0.7 to 0.9 g/lb) (0.4 to 0.5 g/lb)
Weight management 5 to 7 g/kg 1.2 to 1.6 g/kg 0.9 to 1.2 g/kg
(2.3 to 3.2 g/lb) (0.5 to 0.7 g/lb) (0.4 to 0.5 g/lb)
532 } Seebohar

Specific Preparation Period Nutrition Plan


This training cycle progresses from mostly aerobic training to triathlon-
specific training goals such as improving strength, speed, power, force,
and economy. Training sessions will likely be focused on specificity and
the intensity is higher, so energy expenditure is greater. Therefore, nutri-
ent needs will be different. Triathletes need to enter training sessions well
fueled and hydrated. Table 38.2 lists nutrition consumption for various
training loads.
Daily carbohydrate intake will likely need to be increased during this
cycle, and it should range from 5 to 12 grams per kilogram per day.1 The
range is large to account for the volume of training. Less carbohydrate may
be needed on days that have a lower training load, and the opposite also
holds true.
Protein intake should range from 1.4 to 2.0 grams per kilogram.1 Normally,
triathletes tend to follow a lower range for protein during a competition
build or race cycle because carbohydrate consumption is higher.
Fat intake should range from 1.0 to 1.5 grams per kilogram.1 Most tri-
athletes will find themselves needing only the lower end of this range, but
endurance and ultraendurance triathletes may need a larger amount of
daily fat to maintain energy stores.
A triathlete may be going through multiple competitions during this
training cycle, and that schedule offers multiple tapering opportunities.
Although the distance of the upcoming race and tapering schedule are
factors, training volume is generally reduced and intensity is usually
maintained or lessened during a taper. Therefore, carbohydrate intake can
be decreased to 5 to 7 grams per kilogram, protein can be maintained at
1.4 to 1.6 grams per kilogram, and fat can be held at around 1.0 grams per
kilogram.1
As with the base training cycle, triathletes should monitor their daily
hydration by the color and frequency of urination. Because of the higher
intensity and duration of training sessions, however, triathletes may have
a higher degree of fluid loss, so they should monitor their daily hydration
status closely.

Table 38.2 Nutrition Consumption for Training Loads


Training load Carbohydrate Protein Fat
Low 5 to 8 g/kg 1.6 to 2.0 g/kg 1.0 to 1.2 g/kg
(2.3 to 3.6 g/lb) (0.7 to 0.9 g/lb) (0.45 to 0.5 g/lb)
High 7 to 12 g/kg 1.4 to 1.6 g/kg 1.2 to 1.5 g/kg
(3.2 to 5.5 g/lb) (0.6 to 0.7 g/lb) (0.5 to 0.7 g/lb)
Nutrition Periodization | 533

Off-Season Cycle Nutrition Plan


With less structured training or complete rest comes significant changes in
energy expenditure. The off-season typically includes a reduction in volume,
intensity, and frequency of training for triathletes. Because of the quick
transition from competition season to the off-season, triathletes often have
difficulty making the necessary nutrition changes that are needed to coincide
with reduced energy expenditure. With that challenge comes potential gains
in weight and body fat.
But by following macronutrient ranges that support lower energy expen-
diture, triathletes can successfully manage the transition from competition
season to off-season without unnecessary weight gain.
During the off-season (table 38.3), daily carbohydrate intake should range
from 3 to 4 grams per kilogram if the triathlete is exercising without specific
objectives and training less than 2 hours per day.1
Protein intake should range from 1.5 to 2.0 grams per kilogram.1 The higher
range can be used by triathletes to improve satiety, stabilize hunger response
throughout the day, and contribute to positive net protein synthesis if strength
training will be part of the program.
Fat intake should range from 1.0 to 1.2 grams per kilogram.1 Fat is still
important to include in the daily nutrition program, and healthier sources
should be the focal points.
During this time of the year, triathletes should attempt to establish their
fluid balance instinct again without adhering to specific hydration guidelines.
Water alone does not have to be the focus because some portion of fluid
requirements can be achieved by eating foods such as fruits and vegetables
that contain a lot of water.

Table 38.3 Nutrition Consumption for Training Cycles


Base training cycle
Training cycle Carbohydrate Protein Fat
Weight loss 3 to 4 g/kg 1.6 to 2.0 g/kg 0.8 to 1.0 g/kg
(1.4 to 1.8 g/lb) (0.7 to 0.9 g/lb) (0.36 to 0.45 g/lb)
Weight management 5 to 7 g/kg 1.2 to 1.6 g/kg 0.9 to 1.2 g/kg
(2.3 to 3.2 g/lb) (0.5 to 0.7 g/lb) (0.4 to 0.5 g/lb)
Specific preparation
Training cycle Carbohydrate Protein Fat
Low training load 5 to 8 g/kg 1.6 to 2.0 g/kg 1.0 to 1.2 g/kg
(2.3 to 3.6 g/lb) (0.7 to 0.9 g/lb) (0.45 to 0.5 g/lb)
High training load 7 to 12 g/kg 1.4 to 1.6 g/kg 1.2 to 1.5 g/kg
(2.3 to 5.5 g/lb) (0.2 to 0.7 g/lb) (0.5 to 0.7 g/lb)
Off-season 3 to 4 g/kg 1.5 to 2.0 g/kg 1.0 to 1.2 g/kg
(1.4 to 1.8 g/lb) (0.7 to 0.9 g/lb) (0.45 to 0.5 g/lb)
534 } Seebohar

Conclusion
Using the quantitative ranges associated with nutrition periodization will help
triathletes better manage their caloric intake relative to the ever-changing
energy expenditure that results from the cycle shifts in the training program
that happen throughout the year. Energy needs must be reassessed when the
training cycle changes to keep triathletes properly fueled without under- or
overeating.
Guiding a nutrition program and associating energy needs with fluctuating
energy expenditures based on training load changes throughout the year is
the primary goal of nutrition periodization. This approach will lead to better
management of weight, body composition, and energy needs associated with
training.
CHAPTER
39
Nutrient Timing
for Triathlon Training
and Racing
—Bob Seebohar, MS, RD

S pecific nutrient timing protocols are of great benefit when used before,
during, and after certain training sessions. A specific nutrient timing
schedule for a triathlete must consider that a triathlete’s energy needs may be
higher or lower depending on the type, duration, and intensity of workouts
and the overall physical goals for the specific training cycle. The following
guidelines will be separated based on the nutrient to allow a customized
approach to training for triathlons of different distances.

Before Training
Nutrition needs before training depend largely on whether the upcoming
training session will be short or long.

Shorter Sessions
Carbohydrate is necessary before most training sessions, but if the training
cycle does not have high energy demands as defined by long duration (greater
than 2 to 3 hours) or high intensity, then a carbohydrate feeding may not be
necessary until after the workout. This circumstance is typical during the
off-season and early preparatory (base) training cycles and for athletes using
the metabolic efficiency concept described in chapter 37.
Teaching the body to use more of its internal fat stores as energy and thus
preserving the limited carbohydrate stores means that fewer calories are
needed for any training session. For sessions that are less than 2 hours in
duration and are of low to moderate intensity, focusing on hydration may be
the only nutrient timing recommendation needed.

535
536 } Seebohar

The form of carbohydrate that an athlete chooses depends largely on the


mode (swim, bike, or run) of the training session and the athlete’s individual
digestive system response. Liquid sources of carbohydrate are much easier
and quicker to digest compared with semisolid sources and, especially, solid
sources.
For example, Kevin Collington, elite short-course triathlete, chooses high-
carbohydrate foods such as oatmeal with honey to eat before training sessions.
He states that these foods provide him the quick energy that he needs before
training and do not weigh him down. They also help him prepare for his
next workout, often 1 to 2 hours later.
In contrast, Matt Chrabot, elite short-course triathlete, typically eats white
toast and eggs before some training sessions and races. (Kevin and Matt are
two of several triathletes whom I have interviewed about nutritional aspects
of training and competing. All quotes and information from triathletes in
this chapter came from my personal interviews with them.)

Longer Sessions
For longer training sessions (more than 3 hours), the normal pretraining
carbohydrate recommendation is to consume 200 to 300 grams, 1 to 4 hours
beforehand.1 As the start of the workout approaches, less carbohydrate should
be consumed to allow adequate digestion. For sessions lasting 2 to 3 hours,
triathletes need to assess their hunger cues and energy needs. Often, no
calories are needed during a 2- to 3-hour moderate-intensity aerobic train-
ing session, but if any higher intensity will occur within the session, energy
expenditure and thus energy needs will increase.
Protein needs depend on many factors, including the triathlete’s training
session goal, mode of training, intensity, and digestive challenges, but the
normal recommendation is to consume 5 to 20 grams of protein before train-
ing.2 This nutrient can be taken in the form of a liquid such as a smoothie
with milk or protein powder or a semisolid such as yogurt or cottage cheese.
Alternatively, protein-enhanced sports drinks can be used.
The guidelines for consuming fat before a workout are not concrete, and
many triathletes try to minimize their intake of fat before training because
it has a slower rate of digestion. If it is consumed, fat should be kept to a
minimum so that it does not significantly disrupt the digestive response.
Surprisingly, the nutrient often forgotten in a triathlete’s pretraining nutri-
tion plan is fluid. Triathletes typically wake up dehydrated, and this condition
can impair their performance. Thus, triathletes need to begin the day with
an adequate hydration plan.
At least 4 hours before exercise, triathletes should consume approximately 5
to 7 milliliters of fluid per kilogram of body weight. For example, a 160-pound
(73 kg) triathlete would consume 364 to 509 milliliters of fluid. Additional fluid
should be consumed slowly, approximately 3 to 5 milliliters per kilogram of
body weight, 2 hours before exercise if the triathlete is not urinating or if the
Nutrient Timing for Triathlon Training and Racing | 537

urine is dark in color.3 For the sample triathlete described here, this quantity
would be 218 to 364 milliliters of fluid.
Consuming sodium-rich foods during this time can help stimulate thirst
and retain fluids. If sodium is consumed in a beverage, the recommended
amount is 460 to 1,150 milligrams per liter.3 But triathletes can obtain sodium
from food sources and add additional sodium to foods or beverages if needed.
Sarah Haskins, 2008 Olympian, structures what she eats before training
based on the mode and the intensity of the session. For a lower intensity
workout, she chooses a meal with more protein (like eggs) before the work-
out. For a longer or higher intensity workout, she chooses something that is
easier to digest. As she explains,
Before a longer brick session (3-hour bike and 45-minute run), I will have
8 ounces of Greek yogurt, strawberries, and two slices of toast with natural
peanut butter. During the workout, I will have an entire First Endurance
Liquid Shot Gel Flask, two bottles of First Endurance EFS sport drink,
and one bottle of water. After the workout I will have one bottle of water
and one bottle of First Endurance Ultragen recovery drink.

During Training
As stated previously, nutrition must be periodized based on the triathlete’s cur-
rent training cycle. Nutrition recommendations during training sessions will
differ based on fluctuating energy expenditures and needs throughout the year.
During most off-season training, nutritional needs during exercise will be low
to nonexistent. Even during the early phases of the preparatory (base) training
cycle, nutritional needs during training will be low. Not until the duration of
training sessions exceeds 2 to 3 hours will triathletes need to consume calories
during training.
A classic example of this concept is Kevin Collington. He consumes noth-
ing for workouts that are 90 minutes or shorter. For workouts longer than 90
minutes he consumes 200 to 300 calories per hour from sports drinks, energy
gels, and energy bars.
The following information can be used to assist a triathlete in building a nutri-
ent timing protocol for training sessions or races, but it should be customized to
the time of year and the triathlete’s individual goals, as discussed in chapter 38.

Carbohydrate Consumption
Scientific research has long supported the hourly consumption of 30 to 60
grams of carbohydrate. Recent research, however, has supported increasing
the maximum grams of carbohydrate if multiple sources of carbohydrate are
consumed4 because of multiple transport mechanisms of carbohydrate across
the small intestine. These guidelines suggest the consumption of up to 90 to
105 grams of carbohydrate per hour of exercise.4
538 } Seebohar

Note that the body has enough stored carbohydrate to fuel about 2 to 3 hours
of moderately intense exercise; thus, as discussed previously, some training
sessions may either not need supplemental carbohydrate or need only a small
amount. Additionally, if a triathlete has developed metabolic efficiency and
is better at using fat as an energy source during training, less carbohydrate
may be needed during certain workouts.5 Matt Chrabot consumes only an
energy gel during competitions because they last less than 2 hours.
Note that many triathletes who consume too much carbohydrate during
training experience GI distress. Thus, triathletes should seek to develop
metabolic efficiency during the off-season or early preparatory training
cycle to reduce their hourly calorie needs from carbohydrate. This, in turn,
will decrease the risk of GI distress in triathletes who have experienced it
in the past or have sensitive stomachs. Matt Chrabot explains his individual
concerns:
I experience GI distress during run workouts; thus, the amount of fiber
I eat throughout the day if I’m running in the afternoon plays an impor-
tant role. What’s considered healthy by most people’s standards isn’t in a
runner’s best interest. Healthy in many cases translates to high in fiber.

Protein Consumption
The research in support of consuming protein during training is inconclusive,
but it does suggest that consuming protein during training may decrease
posttraining muscle soreness and damage.6 For the sport of triathlon, how-
ever, offering protein as an energy source during training is generally not
necessary. If protein is to be used during training, the mainstream consen-
sus is to use branched chain amino acids rather than whole protein sources.
Many branched chain amino acids can be found in popular sports drinks
and sports nutrition products.

Fat Consumption
Similarly, triathletes are not likely to need to consume any fat during most
workouts or competitions. One exception is with ultraendurance triathletes
who may have competitions spanning multiple days and may require more
calories overall. In general, most triathletes do not need to include fat in
their during-training or competition nutrition plan. They should focus on
carbohydrate and possibly protein.

Sodium Consumption
The quantity of sodium consumed during training is one of the most con-
troversial topics among researchers, athletes, and coaches. Because of the
positive role that sodium plays in helping maintain fluid balance, triathletes
need to consume sodium, but the quantities and the timing are debated in
Nutrient Timing for Triathlon Training and Racing | 539

the scientific community. Some research supports the consumption of 460 to


1,150 milligrams per liter3 of fluid consumed, whereas the American College
of Sports Medicine recommends consuming 0.5 to 0.7 grams of sodium per
liter of fluid per hour. Other researchers recommend a higher range of 1.7 to
2.9 grams of sodium per liter of fluid per hour.7
For triathletes who have high sweat rates or have difficulty staying hydrated,
more sodium per hour may be required. Experimentation is encouraged to
identify their sodium needs. Many sport drinks cannot provide the amount
of sodium that most triathletes need, so sodium supplementation is an option.
Sodium supplements can be found in many forms including capsules,
tablets, powders, liquids, and strips. The triathlete should choose a product
that provides the amount of sodium required by his body and find a delivery
system that is simple to implement and easy on his digestive system.

Fluid Consumption
Dehydration is a key concern for triathletes. The most serious consequence
of exercise-induced dehydration is hyperthermia, which places added stress
on the cardiovascular system. The increase in the body’s core temperature
through exercise combined with warm environmental conditions can quickly
predispose a triathlete to becoming dehydrated.
As fluids are lost from the body, an increase occurs in the concentration
and osmolality of dissolved substances and particles in the body’s fluids,
including sodium. These increases in osmolality and sodium concentration
reduce blood flow to the skin and thus decrease the rate of sweating. The
cooling mechanism for the body becomes less efficient, which can accelerate
the process of contracting a heat illness.
Another negative consequence of dehydration-induced hyperthermia is a
large decline in cardiac output. Less blood is delivered to the muscles and
the extremities, so less oxygen is delivered to the working muscles, both of
which will negatively affect training.
The primary benefit of sufficient fluid replacement during training is
maintaining cardiac output and allowing blood flow to the skin to increase
to high levels, which will promote heat dissipation from the skin, thereby
preventing excessive storage of body heat.
Dehydration of as little as 1 percent loss of body weight through sweat can
have a negative effect on training and racing. For a 160-pound (73 kg) triath-
lete, this loss equates to only 1.6 pounds (.7 kg). Most triathletes can lose this
amount of fluid in a short time, especially if training in a warm and humid
environment. Some research supports the recommendation of drinking 3
to 8 ounces (90 to 240 ml) of fluid every 15 to 20 minutes during training.3, 8
This approach, known as drinking ahead of thirst, is being challenged
in the scientific community. Some suggest that drinking according to thirst
instead of ahead of thirst may be more beneficial. Although no scientific
studies to date have determined which approach is best, triathletes need to
540 } Seebohar

adapt their fluid replacement strategies to their training and environment


and experiment with their approach throughout the year.

After Training
Triathletes need to consume nutrients within 30 to 60 minutes following qual-
ity training because during this period insulin sensitivity is at its highest and
muscles are more apt to accept nutrients. Refueling and rehydrating in the first
30 to 60 minutes following a training session or competition is particularly
important for glycogen-depleting workouts, typically those longer than 2 or
3 hours or shorter (about 90 minutes) with higher intensity intervals. Both
scenarios produce stress to the internal glycogen stores and can contribute
to dehydration at a faster rate, particularly in warmer environments.
Recommended carbohydrate intake posttraining is based on body weight.
The current guidelines for replenishing carbohydrate are to consume 1.0 to 1.2
grams of carbohydrate per kilogram of body weight.9 Protein intake is based
on the absolute quantity consumed, 10 to 25 grams,10 and the triathlete’s body
weight is not a factor.
Most research agrees that higher glycemic index sugars and essential
amino acids are preferred sources of carbohydrate and protein, respectively.
After challenging training sessions or races, triathletes often do not feel
hungry, but they still need to provide their bodies with nourishment. Thus,
a good strategy is to consume a liquid source of calories that includes both
carbohydrate and protein.
Fat is typically not included in the first feeding following training because
it can impede carbohydrate absorption. But it can certainly be included in the
second feeding, usually about 2 hours after the initial feeding. When fat is
consumed, healthy sources, including mono- and polyunsaturated fats, are
recommended. Sodium is also a necessary component of the postworkout
or race nutrition plan. Current guidelines suggest consuming 2 to 5 grams
of sodium per liter of fluid when triathletes have lost a significant amount of
fluid (more than 2 liters) during the workout.10
Triathletes often do not consume enough fluids after training or competi-
tion to replenish what they lost during their efforts. Research suggests that for
proper rehydration, 150 percent of fluid losses should be consumed following
training sessions.11 The recommendation of drinking 24 ounces of fluid for
every pound (1.5 liters for every kilogram) of body weight lost during train-
ing will accomplish this goal.11
For example, if a triathlete loses 4 pounds (1.8 kg) in a training session or
competition, she should consume 96 ounces (2.7 L) of fluid for optimal rehy-
dration. Of course, this is quite a bit of fluid, so instead of having to make
up for a significant amount of dehydration after training or competition, the
triathlete can hydrate before and during the effort to prevent a large fluid
loss that may be unrealistic to handle afterward.
Nutrient Timing for Triathlon Training and Racing | 541

After the last workout of the day, Kevin Collington likes to have something
like a salad with vegetables and a good protein source.

Conclusion
Implementing a nutrient timing system to use before, during, and after train-
ing sessions is an evolving task that will change as energy needs, training
cycles, and environmental factors change. But as triathletes progress closer
to their competition days, they should have a well-formed race-week and
race-day nutrition plan already developed from their experimentation with
nutrient timing strategies during each of their training cycles.
Race-week nutrition should be implemented by simply following what
has worked during higher training loads but accounting for reduced energy
expenditure within the taper. Race-day nutrition will no longer be a mys-
tery because the athlete has been developing nutrient timing strategies for
higher intensity and longer training sessions during the specific preparation
training cycle. Using the knowledge presented in this chapter along with
individual experimentation, the triathlete will already have developed a
race-day nutrition plan during training and will have tested it many times
before the competition.
This page intentionally left blank.
CHAPTER
40
Supplements
for Triathletes
—Bob Seebohar, MSc, RD

I t is no surprise that triathletes use nutrition supplements throughout the


year. Some do so with good reason, but others have with no reason at all
besides having viewed them in a magazine or having heard testimonials from
professional triathletes. Supplements encompass a wide range of nutrition
products and extend beyond the popular vitamins and minerals. Supplement
categories are discussed after a brief description of the supplement industry
is given.

Supplement Regulation
The Dietary Supplements Health and Education Act (DSHEA) of 1994 states
that nutritional supplements that do not claim to diagnose, prevent, or cure
disease are not subject to regulation by the Food and Drug Administration
(FDA). Unfortunately, many supplement manufacturers have concluded that
DSHEA guidelines mean that they have no need to prove claimed benefits,
to show safety with acute or chronic administration, to commit to accepted
quality assurance practices, or to follow the stringent labeling regulations
required for food products.
As a result, dietary ingredients used in dietary supplements are no longer
subject to the premarket safety evaluations required of other new food ingre-
dients or for new uses of old food ingredients.
The intent of DSHEA was to meet the concerns of manufacturers and
consumers that safe and well-labeled products would be available to people
who wanted to use them.
The purposes of DSHEA included the following:
• Define dietary supplements and dietary ingredients
• Establish a new framework for assuring safety
• Outline guidelines for literature displayed where supplements are sold

543
544 } Seebohar

• Provide for use of claims and nutritional support statements


• Require ingredient and nutrition labeling
• Grant the FDA the authority to establish good manufacturing practice
(GMP) regulations
The nutritional supplement industry is a multibillion-dollar-a-year growth
industry. The industry is moving toward compiling more scientific data and
validating products and claims for formulation and marketing purposes.
For that reason, companies will continue to outsource their production to
contract manufacturers who already have good quality control procedures
and laboratories in place.

Oversight of Supplement Manufacturing


Some supplements do not contain the stated amounts of ingredients or, even
worse, have additional substances that are not listed on the label. These
inaccuracies can be of concern to triathletes who have health issues because
a certain herbal product or manufactured supplement could interfere with
medications. Likewise, elite athletes who undergo regular drug testing could
face sanctions if they unknowingly consume a prohibited substance.
Contamination of a supplement can occur at many levels in the supply and
manufacturing process. Luckily, several industry-driven good manufactur-
ing practices (GMP) programs are in place to assist athletes and coaches. One
industry program is overseen by the National Nutritional Foods Association
(NNFA), which has certified dozens of companies and has a strategic alliance
with NSF International, a third-party institution that provides supplement
contamination testing.
Membership in NNFA requires compliance with its GMPs, and its web-
site lists companies that are members. NSF International has an athlete-
certification program and can test specific products for interested athletes.
Other resources include consumerlab.com, which has independently tested
dietary supplements and has an Athletic Banned Substances Screened
Products Program; the FDA, which has also presented an outline of a GMP
program on its website at www.cfsan.fda.com; and Informed Choice (www.
informed-choice.org), another third-party institution that provides supple-
ment contamination testing.

Evaluating the Efficacy of a Supplement


The following criteria can be used to help evaluate the efficacy of using a
supplement:
• The label carries the USP (United States Pharmacopoeia) mark. The USP
mark means that the supplement passes tests for dissolution (how well it
Supplements for Triathletes | 545

dissolves), disintegration, potency, and purity. The manufacturer should


also be able to demonstrate that the product passes tests for content
potency, purity, and uniformity.
• A nationally known food and drug manufacturer made the supplement.
Reputable manufacturers follow strict quality control procedures.
• Research supports product claims. Reputable companies should provide
research from peer-reviewed scientific journals to support claims.
• Claims are accurate and appropriate. If statements are unclear or the
label makes preposterous claims, the company probably did not follow
good quality control procedures.
Additionally, a few simple criteria can be employed when advising tri-
athletes about supplements. Figure 40.1 is a simple, 12-step checklist that
can be used to educate triathletes. When a supplement is being considered,
a triathlete should progress through this checklist before purchase. If any of
the statements on the checklist is true, there is good reason to be skeptical of
taking the supplement. The triathlete should investigate the supplement in
question in greater depth.

• The product promises quick improvement in health or physical performance.


• The product contains some secret ingredient or formula.
• The product is advertised mainly with anecdotes, case histories, or testimonials.
• The product features popular personalities or star athletes in its advertisements.
• The product takes a simple truth about a nutrient and exaggerates that truth in
terms of health or physical performance.
• The product questions the integrity of the scientific or medical establishment.
• The product is advertised in a health or sports magazine whose publishers also
sell nutritional aids.
• The product is sold by the person who recommends it.
• The product uses the results of a single study or outdated and poorly con-
trolled research to support its claims.
• The product is expensive, especially when compared with the cost of obtaining
the equivalent nutrients from ordinary foods.
• The product is a recent discovery not available from any other source.
• The product makes claims that are too good to be true and promises the
impossible.

„„Figure 40.1 Checklist for evaluating a supplement.


546 } Seebohar

Primary Types of Supplements for Triathletes


Many nutrition supplements on the market claim to improve performance
in some way, shape, or form. To make it easier to navigate the supplement
world, supplements can be classified into three categories:
1. Dietary: micronutrients such as calcium, iron, zinc, and multivitamins.
These supplements are used to counter certain vitamin and mineral defi-
ciencies. Triathletes typically use these supplements because laboratory
testing has identified vitamin and mineral deficiencies or because they
want to have an insurance policy in place to mask inadequate food choices.
2. Sports: energy bars, gels, sports drinks, and electrolyte supplements.
Their primary use is before, during, and after training.
3. Ergogenic: performance-enhancing substances such as amino acids, cre-
atine, caffeine, sodium bicarbonate, energy drinks with extra ingredients
added for mental focus or energy production, adaptogens, androsten-
dione, beta-hydroxy-beta-methylbutyrate (HMB), carnitine, conjugated
linoleic acid (CLA), glucosamine, chondroitin, and others. Triathletes
typically do not take these substances in great amounts, but they may
take them throughout the day and possibly before exercise, depending
on the ergogenic aid.

Popular Supplements Used by Triathletes


Discussing all supplements used by triathletes is not feasible, but the following
are among the more popular in terms of efficacy and use in endurance sport.
Table 40.1 shows some of the common benefits and drawbacks of supplements.

Iron
Iron is a supplement that should be taken only under the supervision of a
medical or nutrition professional in response to a laboratory-tested proven
deficiency. Iron is a component of hemoglobin, the protein found in red blood
cells that carries oxygen to cells, and myoglobin, the protein found in heart
and skeletal muscle that carries oxygen to those muscles. The daily recom-
mended intake of iron does not account for what is lost because of a high
training load; iron stores and training load are inversely related. Although
iron stores may be compromised in the body when training load is high,
taking any supplement is not recommended until a sports physician or sports
dietitian conducts an evaluation.
Some athletes, particularly menstruating females, vegetarians, and those
following calorie-restricted diets, are more prone to iron deficiency anemia,
but any athlete can be prone to iron deficiency. Maintaining iron balance and
getting enough iron in the body can be difficult because iron is lost through
sweat, urine, the GI tract, and foot-strike hemolysis. If an athlete has low
Supplements for Triathletes | 547

Table 40.1 Benefits and Drawbacks of Supplements


Supplement Benefits Drawbacks
Iron Improves oxygen-carrying capacity Iron overload; supplementing iron
possibly fatal for athletes who have
the disease hemochromatosis
Calcium Maintains bone health, muscle function None
Multivitamins Correct deficiencies Possible overload of fat-soluble
vitamins (A, D, E, K)
Energy gels Provide immediate energy Possible GI distress because of high
carbohydrate concentration
Electrolytes Improve fluid balance Increases thirst in some athletes
Adaptogens Increase resistance to stress Inconclusive but strengthening
evidence to validate its use
Beta-alanine Serves as buffering tool in delaying No scientific validation yet of its worth
fatigue in endurance sport
Caffeine Stimulates central nervous system, which Anxiety, irritability, possible GI
can improve mental alertness, increase distress at excessive doses
usage of fat as fuel, and spare glycogen
MCTs Converts to energy faster; spares Anecdotal reports of GI distress
glycogen
Creatine Improves anaerobic energy system No data regarding its safety for use in
training young athletes

iron stores, the body’s ability to make red blood cells is limited, which leads
to a decrease in the oxygen-carrying capacity of the blood. Because athletes
depend on oxygen delivery to working muscles, even a slight imbalance of
iron stores can have significant effects on performance. All triathletes should
have a full blood draw, including iron stores, taken once per year, preferably
in the base training cycle. This information will provide a baseline regarding
iron stores and will indicate whether a supplement should be taken.

Calcium
Calcium is the most abundant mineral in the body. The majority of calcium is
present in the bones and teeth. The remaining stores are used for muscular
contractions and metabolic functions such as activating enzymes that break
down muscle glycogen for energy production.
The regulation of calcium in the body requires two hormones, parathy-
roid and calcitonin, and vitamin D. When blood calcium levels decrease,
parathyroid hormone is released and increases the activation of vitamin D
in the kidneys, which act to decrease the amount of calcium lost in the urine.
In addition, vitamin D increases the absorption of calcium from the small
intestine and resorption from the bones.
The body is efficient in regulating calcium. Thus, excessive or poor dietary
intake of calcium rarely results in acute changes in calcium levels in the
548 } Seebohar

body. A dietary imbalance coupled with an athlete’s demanding training


schedule can result in calcium alterations over time that can create a chronic
deficiency state over years. There is really no deficiency test for calcium, nor
do symptoms become evident. But triathletes can have their bone mineral
density measured, which can show whether adequate dietary calcium is
being consumed.
But some athletes may not be meeting their bodies’ calcium needs. Spe-
cifically, athletes who restrict their daily caloric intake for the purpose of
aesthetics, health, or performance or those who did not consume adequate
amounts of calcium during childhood could be susceptible to poor calcium
deposition. This situation could lead to bone deformities and the possibility
of having a lower peak bone mass at an early age, thus increasing the risk of
osteoporosis. Eating lower amounts of calcium as an adult athlete can lead
to excessive calcium resorption from the bone, which increases the risk of
osteomalacia, a softening of the bones.
Intakes of calcium may be higher for athletes than for nonathletes because
calcium is lost in sweat. Current daily recommendations for calcium intake
are to consume between 1,300 to 1,500 milligrams of calcium per day. The
tolerable upper intake level (UL) per day is 2,500 milligrams. Toxicity symp-
toms over the UL include constipation, kidney stones, cardiac arrhythmia,
and malabsorption of iron, magnesium, and zinc.

Multivitamins
The use of a daily multivitamin has always been controversial. Whether to
take one is not an easy question to answer. Experts generally believe that ath-
letes who engage in strenuous or long-duration training may have increased
needs for some vitamins and minerals. What is not known exactly is which
vitamins and minerals are specifically needed in excess by some athletes, so
no specific recommendations can be made. Generally, a good practice for some
athletes may be to take a multivitamin that meets their expenditure needs
and is associated with their gender and possible micronutrient deficiencies
as long as its use is aligned with the athlete’s physical periodization cycles.

Energy Gels
Energy gels are a mix between a sports drink and an energy bar in a semisolid
structure. These products are extremely popular among triathletes because
they are a source of concentrated calories that is easy to carry and consume.
The primary purpose of energy gels is to provide carbohydrate that can be
used immediately during training or competition to fuel working muscles.
The type and quantity of carbohydrate contained in the energy gel should
be evaluated to determine whether it will meet the athlete’s energy needs.
Additionally, energy gels have roughly a 55 to 65 percent carbohydrate
concentration (compared with 4 to 8 percent in most sports drinks), which
Supplements for Triathletes | 549

makes gels more challenging for an athlete to digest during training and
competition. Athletes with sensitive stomachs may have difficulty absorbing
these high-concentration gels, and higher intensity training will sometimes
lead to a reduction in digestion of calories. Adequate water must be consumed
with an energy gel, and consuming a gel simultaneously with a sports drink
is not recommended.

Electrolyte Supplements
Electrolyte supplements are typically seen as tablets, capsules, powders,
fluids, and strips. Also available are full-spectrum electrolyte supplements
that offer additional minerals such as calcium, magnesium, and potassium.
The basis for including more minerals in electrolyte supplements is to make
up for what is lost in sweat. Many compounds are found in sweat including
ammonia, copper, creatinine, iodine, iron, lactic acid, manganese, phosphorus,
urea, uric acid, sodium, chloride, calcium, potassium, and magnesium. The
last five are the micronutrients found in higher quantity in sweat.
For athletes who have a high sweat rate and sodium concentration and
train in a hot and humid environment, additional electrolyte supplements
may be needed. No sports drink on the market can supply the necessary
quantities of electrolytes for these athletes. A typical sodium loss per hour
for athletes is 1 gram,1 but anecdotal information from coaches and athletes
is that sodium losses can be as high as 5 grams per hour.

Adaptogens
Adaptogens, consisting mostly of herbal compounds, are a fairly new inclu-
sion to the supplement market even though they have existed for thousands
of years. An adaptogen is a substance that increases resistance to physical
and emotional stress, stress-related imbalances, and environmental pol-
lution.2 Some also contain polysaccharides. These have been reported to
stimulate the immune system, which works in a synergistic way to increase
the body’s ability to prevent illness. Some common examples of adaptogens
include ginseng, ashwagandha, schisandra, rhodiola, cordyceps, reishi, and
maitake.
One of the major actions of adaptogens relative to athletes is their ability
to increase resistance to the catabolic effects of high-intensity training and
stress. As the intensity of training increases, the body has to use its natural
defense mechanisms to help repair and replace the damage that occurs. Using
adaptogens could allow this defense mechanism to work more efficiently.
The benefits of taking adaptogens are more likely to occur in the more
intense training cycles. Most of the current research
. has targeted physio-
logical markers important to athletes such as VO2peak, pulmonary ventila-
tion, time to exhaustion, and others, such as creatine kinase and C-reactive
protein activity. Some recent research has shown a link to prevention of
550 } Seebohar

overtraining syndrome, providing protective effects from inflammation


and infection, increasing fat utilization, and slowing glycogen utilization.

Beta-Alanine
Beta-alanine, a nonessential amino acid, is part of the amino acid carnosine
and is needed to replenish the body’s stores of carnosine to fuel high-inten-
sity exercise. Carnosine is a potent buffer of hydrogen ions, and increased
amounts of this amino acid have direct correlation with improved anaero-
bic and aerobic performance. Muscle carnosine synthesis is limited by the
availability of beta-alanine. Thus, supplementing with beta-alanine would
cause an increase in carnosine stores, which could then provide a buffering
mechanism during training.
Consistent, high-intensity exercise increases the amount of muscle car-
nosine, but research has shown that taking a daily beta-alanine supplement
increases it more than training and eating carnosine-rich foods3 such as beef,
pork, poultry, and fish does. Beta-alanine, through its effect on carnosine, can
also synthesize lactic acid to be reused as fuel and can generate nitric oxide
synthase, which makes the vasodilator nitric oxide.3
Although no definitive agreement has been reached among researchers
who have studied endurance athletes, it is generally accepted that supple-
mental beta-alanine can increase carnosine levels, which can have a positive
effect in delaying neuromuscular fatigue, using lactic acid as fuel, improving
intra- and extracellular buffering capabilities, augmenting total work com-
pleted, and increasing time to exhaustion.3

Caffeine
Caffeine is one of the most well-researched ergogenic aids. Caffeine acts
on the central nervous system as a stimulant by increasing the release of
adrenaline, which can improve mental alertness, using body fat as fuel, and
sparing glycogen at low- to moderate-intensity exercise.4 In higher intensity
exercise, caffeine has been shown to decrease the rating of perceived exer-
tion, increase carbohydrate oxidation when consumed with a carbohydrate
source, and allow athletes to reach a higher levels of lactate.
Caffeine is often mistakenly classified as a diuretic. Recent research,
however, suggests that over a 24-hour period, caffeine does not exhibit a
dehydrating effect on the body.5 Moderate amounts of caffeine can be con-
sumed throughout the day as long as other beverages are consumed to aid
in positive net fluid balance.
The protocol that can be used to seek improvement in performance with
caffeine includes the following:
• Consume 3 to 9 milligrams of caffeine per kilogram of body weight 60
to 75 minutes before competition.6
Supplements for Triathletes | 551

• If the competition is prolonged, consuming 1 to 3 milligrams of caf-


feine every 90 to 120 minutes may be justified.7 More is not better in this
scenario because an excessive dose taken too frequently can cause GI
distress.
Caffeine has been proved to improve performance when used correctly.
But using caffeine as an ergogenic aid should be done with caution because
of the higher levels needed to stimulate the central nervous system. Too
much caffeine can cause athletes to become anxious, delirious, and irritable.
Athletes must not abuse it in sport.

Medium Chain Triglycerides (MCTs)


Most athletes do not view fat as a supplement or even as an ergogenic aid.
But MCTs are different. Some athletes have used these fats to enhance per-
formance during longer duration training and competition.
MCTs are a type of saturated fat with shorter lengths of carbon molecules
that allow them to have characteristics different from other fats. MCTs are
digested and metabolized more quickly by the liver and can be converted to
energy faster. MCTs act similarly to carbohydrate, which has led to the belief
that they could be used as a glycogen-sparing fuel source.
Unfortunately, research has shown mixed results. Evidence that supports
the use of MCTs as an ergogenic aid to improve performance is not convinc-
ing, although anecdotal reports continue to come from athletes who have
had great success using it.

Creatine
Creatine, a nitrogenous organic compound, is made from the amino acids
arginine, glycine, and methionine. Most creatine in the body is found in the
skeletal muscle, although a small amount is synthesized in the liver, kidneys,
and pancreas. The theory behind the use of creatine among athletes is that
supplementing with it can improve athletic performance in events that last
between 90 seconds and 4 minutes. The proposed mechanism underlying
the benefit of creatine is that it decreases the stress on anaerobic glycolysis
and reduces lactate levels. This, in turn, would delay fatigue. Although the
benefits for anaerobic athletes are obvious, creatine could benefit endurance
athletes for specific training sessions that are targeted at improving anaerobic
endurance or for heavier strength training cycles.
As mentioned previously, creatine exhibits its positive effect on the anaero-
bic energy pathway, specifically the phosphocreatine system (sometimes
referred to as the creatine phosphate system). This energy system is the first
to be initiated in any training session, and it can become depleted quickly,
usually in less than 15 seconds. The benefit of supplementing with creatine is
that it can improve the ability to generate more ATP (adenosine triphosphate),
552 } Seebohar

which can improve energy supply during shorter, more intense interval
training sessions. The result could be more forceful muscle contractions and
possibly higher velocities of movement.
Many methods of supplementing with creatine are used. The traditional
method has been to use loading phase consisting of consuming 20 grams of
creatine per day for 5 to 7 days and then to implement a maintenance dose of
3 to 5 grams per day. Another method is to consume 3 to 5 grams per day on a
consistent basis without a loading phase. Additionally, research has provided
shorter protocol testing, consisting of up to only 5 days of creatine loading,
and has discovered improvements in anaerobic markers of performance.9
Regardless of the method chosen, creatine supplementation should be
customized based on each triathlete’s needs. Cycling the use of creatine is
important because some evidence suggests that using creatine for an extended
period may reduce its positive effects. Normal cycling patterns include using
creatine for 3 to 8 weeks and then discontinuing its use for 3 to 4 weeks.
Supplementing with creatine has not come without controversy. In the
past, concerns have surfaced about using creatine in warmer climates and
its possible effects on decreasing performance in warmer environments, usu-
ally because of dehydration concerns. But research has concluded that using
creatine does not decrease an athlete’s ability to dissipate heat, nor does it
negatively affect the fluid balance state as long as recommended doses are
taken.8
Creatine supplementation has been studied in normal, healthy adults but
not in youth under 18 years of age. No conclusive data suggests that using
creatine is unsafe for adults, but no information is available regarding the use
of creatine in young athletes or over extended periods with this age group.
Additionally, because the physiological benefit of creatine lies in regenerat-
ing ATP to improve energy, athletes should not use creatine without making
sure that they are consuming adequate calories on a daily basis. The use of
creatine should be justified and used carefully by each triathlete.

Conclusion
Because many triathletes use supplements in their training and competi-
tion, it is important to know about the three categories of supplements
and to understand why triathletes use them. Often times, supplements are
not necessary if nutrition periodization and nutrient timing protocols are
used consistently. In some cases, a triathlete may reach for a supplement
depending on its specific category, the way in which it may improve health
or performance, and the specific deficiency that needs to be addressed. In
most cases, triathletes can obtain a large portion of their vitamin and mineral
needs through whole foods, but the use of sport supplements and ergogenic
aids, such as energy bars, energy gels, sports drinks, and caffeine, will be
efficacious before, during, and after some training sessions and competitions.
Pa r t

XI
Psychology
of Multisport
This page intentionally left blank.
CHaPtEr
41
Mental Toughness
for Triathlon
—JoAnn Dahlkoetter, PhD

T riathletes face significant psychological barriers in their training and


competitions as they push themselves to their physical limits. Whether
it’s the fear of a recurring injury, going out too fast in the swim, losing focus
during the last section of the bike leg, or hitting the wall at mile 20 of the
marathon, athletes all have mental or physical obstacles to overcome. With
proper mental training, triathletes can learn to work through these blocks
and enhance their motivation and self-confidence. Triathletes can find that as
beliefs about their limits change, the limits themselves begin to shift. Beliefs
give rise to goal realization.
In some cases, research has revealed that mental practices are almost as
effective as true physical practice, and that doing both is more effective than
either alone.1 Thus, more triathletes are seeking out mental training tools and
sport psychology experts to assist them with the mental side of their perfor-
mance. Both males and females are seeking this help, but some interesting
gender differences are appearing when examining who is searching for help
in the mental training area. One study by Dolan, Houston, and Martin found
that female triathletes were more likely than male triathletes to train with
others, use mental preparation strategies, and report feeling anxious before
competitions.2
In triathlon training and racing, the mind and body are so well connected
that achieving good consistent race outcomes is difficult when the proper
mind-set is not present. The right internal state must be created first. After a
triathlete feels right inside, a quality triathlon performance can occur more
naturally and with less effort. The appropriate internal state can bridge the
gap between what the triathlete thinks she can accomplish and what she
actually achieves. The right mind-set can make the difference between just
having the capacity and realizing true potential.

555
556 } Dahlkoetter

One top triathlete recently noted that during his best event,
It was like I was in an invisible envelope where the only thing that existed
was this race. I was essentially unaware of myself . . . like my body was
being directed from an unconscious part of myself, rather than by my
thoughts. I did not have to think about each move . . . I was on automatic
pilot.
This is exactly the mind-set that will help triathletes achieve their best, but
that mind-set can be elusive to achieve. This chapter explores the keys to
achieving that mind-set, beginning with performance motivation.

Achieving Peak Performance


Triathletes should ask themselves a few questions. Have they ever had both
their mind and body totally in sync with their triathlon training or racing.
Have they ever had the sense that, for a single moment, they were in complete
control of what they were doing? Have they ever felt such an intense pleasure
and motivation in an action that they could continue doing it all day just for
the experience, regardless of the outcome?
Research, interviews, and mental performance coaching with a wide
variety of athletes over 30 years have established the foundation for a better
understanding of the peak motivation state. Through this experience, it’s been
discovered that several emotional and behavioral patterns and mind-sets are
clearly linked to achieving this state. This chapter describes how triathletes
can achieve this ideal state of high-level motivation more often, and it identi-
fies the circumstances that allow it to occur.
What do triathletes report about their motivation and peak performance or
zone experience? This is a state of mind at which a person becomes completely
immersed in the task at hand, to the exclusion of most other outside influ-
ences. Triathletes are completely focused on the present moment; the rest of
the world seems to fade into the background. During the peak performance
they feel more self-confident and more fully integrated. Their mind, body,
and spirit are tuned in to the moment. These are the times when they feel
most energetic and alive.
Psychologist Abraham Maslow first studied people with peak experiences,
called self-actualizations, in the early 1940s. He researched human behavior at its
most personally fulfilling levels. He found that during peak experience, “The
human powers come together in a particularly efficient and intensely enjoy-
able way . . . in which the person is more open to experience, more expressive
or spontaneous, fully functioning, more truly oneself, more fully human.”3
Motivated triathletes also report a feeling of power during optimal per-
formance that transcends their usual level of strength and energy. This
outpouring of power is often apparent to other triathletes or spectators. In
personal interviews, spectators reported watching the eyes of Mark Allen in
Mental Toughness for Triathlon | 557

the marathon section of the Hawaii Ironman Triathlon, stating, “He looks so
driven, so consumed by the momentum of the race itself, that nothing can
stop him.” The act of swim, bike, and run itself takes over and the person
feels completely synchronized within it.
Maintaining this type of internal balance is key for staying motivated, for
performing well, and for producing an enjoyable experience. Performing well
is a natural outcome of having the right kind of internal thoughts and feel-
ings. When an athlete feels good, training improves. Level of performance
is a direct reflection of motivation, mind-set, and the way that the triathlete
thinks and feels. Thus, to deliver an optimal performance, triathletes need to
build and sustain the right type of internal environment regardless of what
is going on around them.
Key characteristics of motivation and peak performance include having a
feeling of being totally immersed in an activity, of being completely in tune
with the task at hand. Triathletes have an expectation that they have the abil-
ity to meet the challenge ahead. They are focused on the present moment
so intensely that they can often foresee actions before they occur. This focus
gives triathletes a sense of unusual power that appears to come from a new
source within themselves. From this, they have a sense of total joy and elation,
a harmonious experience of mind, body, and spirit. They have a perception
that the past and future are fading away and the present action is the only
thing that matters.
Anyone can experience high levels of motivation and peak performance in
a wide range of triathlon settings. We discuss specific techniques for develop-
ing motivation through goal setting in the next section.

Goal Setting and Commitment


Goal setting is essential for making steady progress in any triathlon training
program. Without goals we tend to flounder, like a ship without a rudder.
We may go a long way, but we might well end up running in circles without
a sense of direction to our training.
The question of how goal setting and perfectionism affects performance is
highly debated. One research study investigated how perfectionism affected
race performance and what role competitive triathletes’ goals played in this
relationship (study 1: N = 112; study 2: N = 321).4
The research analysis indicated that perfectionistic personal standards,
high performance-approach goals, low performance-avoidance goals, and
high personal goals predicted race performance beyond the triathletes’
normal performance level.
The implications from this study are that triathletes need to set high per-
sonal goals that are specific but attainable based on their training records
(e.g., run at a 6:30 per mile [4:00 per km] pace for the next half-Ironman). The
more specific the goals are and the more that the triathlete can incorporate
558 } Dahlkoetter

mental rehearsal and visualization into the goals, the more likely he is to
achieve them.
The way that triathletes approach goals can have a powerful effect on the
results. How can triathletes make goal setting more effective? Several factors
can improve success and results, including the types of goals set and the way
that they are measured and evaluated. Here are some principles to keep in
mind when establishing a program.
• Focus on aspects within the triathlete’s control: Triathletes should direct
energy toward aspects of training and competition that are potentially within
their control, such as triathlon training schedule, plan, diet, warm-up, form,
mental focus, and effort level. They should avoid focusing on elements that
are beyond their control such as placing, winning, the weather, and their
competitors.
• Create measurable goals: Triathletes should write goals in a logbook so
that they see them every day. Goals should be specific rather than general,
and they should provide a framework for evaluating progress, such as a time.
If necessary, goals can be adjusted. Triathletes can give themselves a reward
for progress made or readjust the goal if it is unrealistic.
• Set both long-term and short-term goals: Successful triathletes devise
short-term goals that bring them closer to their final goals. Many make the
mistake of skipping short-term goals.
• Make the goal public: Research has shown that those who make their
goals public in some way perform significantly better than those who keep
their goals to themselves.

Mental Discipline:
The Champion’s Mind-Set
How do triathletes develop the mental discipline to keep the training
schedule on course with all of life’s setbacks and distractions? To excel as
triathletes, they need to be eager for success and results, and they must focus
on becoming the best triathlete they can be. It should begin with a vision,
but then the triathlete must be disciplined to reach the goal.
We often read about triathletes who overcome physical barriers. Out of
these challenges triathletes develop an intense, burning desire to succeed.
They need to demonstrate to themselves that they can complete their goals
even with their physical limits. Through these kinds of examples triathletes
and coaches can begin to understand that mental discipline and desire are
sometimes more important than genetic capacity or a healthy body.
To be a disciplined triathlete, the drive must come from within, whether
a novice, a serious triathlete, or an elite competitor. Few people realize that
building and maintaining a high level of self-motivation is a learned skill
that anyone can acquire. Discipline and motivation is energy, and that
Mental Toughness for Triathlon | 559

sense of self-directedness is one of the most powerful sources of energy


available to an athlete. From internal discipline triathletes gain the willing-
ness to persevere with training, to endure discomfort and stress, and to
make priorities with time and energy as they move closer to fulfilling their
goal.
In a 2004 study Grand’Maison reported that although two-thirds of the
Ironman respondents stated that their knowledge of sport psychology was
limited or inexistent, 97 percent said that they believed strongly or very
strongly that discipline and mental skills were key to success.5 In addition,
the survey revealed two deep-rooted misconceptions about the use of sport
psychology: that sports psychology is mainly for those with deep emotional
issues and that it’s only for top-level elite triathletes. Ironman triathletes’
issues clearly reflect what many performers want and what sport psychology
consultants should be providing—practical and effective guidelines that
work in the real world of performance.

Key Traits of the


Highly Disciplined Triathlete
What are the key characteristics of well-disciplined triathletes? Through
extensive work with numerous triathletes over several years, a constellation
of traits that defines the champion’s mentality has developed. High-level
triathletes do not possess superhuman powers or extraordinary traits lim-
ited to a select few. Anyone who wants to excel in triathlon can develop the
characteristics that make a champion.
• Internal discipline and self-direction: Champion triathletes decide
from the outset that they are training and competing for themselves, not for
the awards, not for the prize money, not for their coaches. Direction and drive
need to come from within. The objectives must be chosen because that’s
precisely what they want to be doing. Triathletes should ask themselves,
“What keeps me swimming, biking, and running? Who am I doing it for?”
• Commitment to excellence: Does the triathlete set a high standard for
herself? Elite triathletes know that to excel at their sport, they must decide
to make it a priority in their life, to be the best at what they do. They set
challenging yet realistic standards that are specific, and they are honest in
evaluating their abilities and the amount of time and energy that they can
put into their program.
• Determination, consistency, organization: Winning triathletes know
how to self-energize and work hard on a daily basis. Because they are pas-
sionate about what they do, they find it easier to maintain consistency in
training and competition. Regardless of personal problems, fatigue, or dif-
ficult circumstances, they can generate the excitement and energy needed
to do their best.
560 } Dahlkoetter

• Concentration and focus: Disciplined triathletes have the ability to


maintain focus for long periods. They can tune in what’s critical to their per-
formance and tune out what’s not. They can easily let go of distractions and
take control of their attention, even under pressure. They put their attention
on the aspects of the competition that are within their control and recognize
that they can make that choice.
• Capacity to deal with obstacles: Top triathletes know how to deal with
difficult situations. Adversity builds character and becomes an opportunity
for learning, opening the way for personal growth and renewal. When elite
triathletes know that the odds are against them, they embrace the opportunity
to explore the outer limits of their potential. Rather than avoiding pressure
they feel challenged by it. They are calm and relaxed under fire, realizing
that nervousness is normal and that some nervousness can contribute to
performance. Breathing deeply and doing a mental rehearsal of exactly how
the race should go can also help triathletes remain calm and relaxed.
• Enthusiasm and desire, love for the sport: Triathletes who win have a
drive, a fire inside that fuels their passion to achieve a key goal, regardless of
their level of talent or ability. They begin with a vision, and as they see that
vision with more clarity, it becomes more likely to turn into reality. Wherever
attention goes, energy flows.

Mental Imagery
Images are the mental representations of experience. Although verbal lan-
guage is the most common means for communicating with the external world,
imagery is a powerful means for internal communication. The visualization
process can be defined as the conscious creation of mental or sensory images
for the purpose of enhancing training and life. It is the deliberate attempt
to select positive mental images to affect how the body responds to a given
situation.
By using imagery or visualization, athletes can create, in vivid detail, a
replay of a best performance in the past. Alternatively, they can mentally
rehearse an upcoming event and see themselves doing it right. Imagery
guides much of an athlete’s experience because it is a more efficient, complete
language than self-talk. For example, trying to describe to someone how to
execute the perfect freestyle swim stroke, in detail, using words is difficult. A
person could write an entire book. The same stroke shown through a video
replay of a top triathlete conveys the same message in a few seconds.
Most athletes daydream and reexperience situations in their minds in a
haphazard way. The fact that people can remember previous experiences
in detailed fashion is why visualization works so well for athletes. Most
good triathletes have discovered this technique on their own and may use
it occasionally to improve learning and performance. For maximum results,
however, they need to control their imagery and practice it on a regular basis
Mental Toughness for Triathlon | 561

rather than just let thoughts pass in and out. Through imagery, triathletes
can re-create their best performances in detail and then use that energy to
help them through any situation that they may encounter.

Principles for Building Discipline


and Maximizing Performance
With the constellation of traits that disciplined triathletes possess, how do
triathletes begin to build them into their lives? How do they turn these quali-
ties into useful behaviors that will make a difference in the way that they
train and race?
The following suggestions have helped many athletes excel in their sport.
• Create a disciplined mind-set: Direct focus to what is possible, to what
can happen, on the path toward success. Rather than complaining about the
weather or criticizing the competition, mentally trained triathletes attend
to only those things that they can control. They have control over thoughts,
emotions, training form, and perception of each situation. Triathletes have
a choice in what they believe about themselves. Positive energy makes peak
performances possible.
• Practice being focused yet relaxed: Develop the ability to maintain
concentration for longer periods. Tune in to what’s critical to performance
and tune out what’s not. Triathletes can easily let go of distractions and take
control of their attention. As they focus more on the task at hand (e.g., train-
ing form, how they’re feeling), negative thoughts will have fewer ways to
enter the mind. Athletes at any level can experience the same single-minded
focus and drive seen in elite athletes. Concentration is a learned skill that
can improve with practice. The key is to recognize that good concentration
is a constant vigil. The mind is like a gypsy; a certain amount of mind drift-
ing will always occur. But if triathletes regularly work on their attention
span, they can begin to take control of the most distracting situations. When
athletes want to concentrate on something, they must find a way to become
more attracted to it, such as thinking of a song that has an upbeat cadence
and synchronizing their movements to that rhythm.
• Picture goals every day: Enter a relaxed state through deep abdominal
breathing. Then, as vividly as possible, create an image in the mind of what
you want to achieve in the sport. Triathletes can produce a replay of one of
their best performances in the past and then use all those positive feelings
of self-confidence, energy, and strength in the mental rehearsal of an upcom-
ing event. See yourself doing it right and then use the imagery during the
event itself.
• Build a balanced lifestyle: Create a broad-based lifestyle with a variety
of interests. Strive for a balance among work and fun, social time, personal
quiet time, and time to be creative. Develop patterns of healthy behavior.
562 } Dahlkoetter

Eat regularly, get a consistent amount of sleep each night, reduce workload
at times if possible, and allow time to relax and reflect between activities.
Develop a social support network of close friends and family, some who are
sports oriented and some who have other interests. Learn to communicate
openly. Resolve personal conflicts as they occur, so that they don’t build to a
crisis on the night before an important race.
• Enjoy the process and take the pressure off: Triathletes should make
a deliberate effort each day to create enjoyment in triathlons, to renew their
enthusiasm and excitement for training, and to avoid trying to force physical
improvement. Especially when in the recovery phase after a race or when
healing from an injury, lighten up on the rigid training schedule and exercise
according to feelings each day. Remove the strict deadlines and race dates
that have been cast in stone. Let the next breakthrough occur naturally, at its
own pace, when the internal conditions are right. Trust that there’s a time to
push hard in races and a time to hold back during recovery.
• Use setbacks as learning opportunities: Draw out the constructive les-
sons from every workout and race and then move on. Look for advantages
in every situation, even if conditions are less than ideal.

Conclusion
Mental toughness is the key to performing well on a consistent basis for tri-
athletes at all levels. The mind and body are so well connected that achieving
a good outcome becomes difficult when mental toughness is not present. The
right internal state must be created first. When a triathlete feels right inside,
a quality performance can occur naturally. The appropriate internal state can
bridge the gap between what a triathlete thinks he can accomplish and what
he actually achieves. The right mind-set can make the difference between just
having the ability and realizing true potential. Triathletes should incorporate
the mental toughness profile into mental preparation to train more health-
fully and perform at their best.
CHAPTER
42
Psychology of
Triathlon Training
—JoAnn Dahlkoetter, PhD

F or most people who are passionate about their sport, triathlon training
provides a sense of self-assurance, a feeling of control over life, and the
chance to test personal limits and reach for new challenges. Yet when goals
become unrealistically high, internal pressure builds, making workouts more
like an obligation than a pleasure. When this happens, triathletes may cross
the line into obsession. They may become desensitized to their body’s feedback
and unknowingly overtrain, creating serious risk for injury, stress, or burnout.
A study by Main et al. reported that exposure to stressors (either training
or psychological) had significant effects on several negative health factors
including mood disturbance and burnout.1
The disciplines of swimming, biking, and running are effective ways
of handling so many needs that for some triathletes, training becomes an
all-consuming obsession that preempts all other interests in life. Having a
focused commitment to physical improvement is not in itself harmful, but
when triathletes become so attached to their training that it becomes the sole
means for coping with life, its positive effects can quickly fade away.

Obsession and Staleness


Triathletes frequently make comments such as, “I’m working harder than
ever, putting in more time training, and there’s absolutely no progress. Is
it something in my head?” Or they say, “I’m recovered from my injury, but
I can’t seem to get my form back.” Others might offer, “I’m at 100 percent
physically, but the motivation just isn’t there.”
These statements may sound familiar. Plateaus and staleness are universal
experiences that all triathletes and performers have gone through, regardless
of ability level or type of sport. Experiencing peaks and valleys is normal, but
there are ways to regain passion and ensure that high points are experienced
more often than low ones.

563
564 } Dahlkoetter

Research examining obsession and the overtraining syndrome in elite


athletes indicates an incidence rate of about 20 to 30 percent. A relatively
higher occurrence is seen in individual-sport athletes, females, and those
competing at the highest representative levels.2

Symptoms of Obsession
Any triathlete who has challenged himself for a long time has probably experi-
enced a period of frustration. How long this period lasts depends largely on
his ability to recognize the symptoms and seek help early.
The most commonly reported symptoms are similar to those observed in
overtrained athletes: increased perception of effort during exercise, frequent
upper respiratory tract infections, muscle soreness, sleep disturbances, loss
of appetite, mood disturbances, shortness of temper, decreased interest in
training and competition, decreased self-confidence, and inability to concen-
trate. The appearance of these many symptoms underlines the importance
of taking a holistic approach when trying to treat or prevent overtraining
in the triathlete; both training and nontraining stressors must be con-
sidered.

Obsession Profile
The obsessed triathlete who displays a high risk for injuries, staleness, or burn-
out often shows classic type A personality characteristics: a high-achieving
perfectionist who is independent, has a strong need to be in control, finds it
difficult to relax, and is unexpressive emotionally. If the triathlete’s training
pattern fits the following description, she may have moved from passion to
obsession and be unknowingly setting herself up for disappointment.
• Obsession and obligation: Triathletes believe that they must complete
the daily distance in their program regardless of weather, personal state of
health, or enjoyment level. Training becomes a duty.
• Lack of balance: They tend to avoid developing a broad-based lifestyle
that includes social support, a variety of interests, and a firm sense of self.
They cling to training as their primary coping mechanism and use exercise
to deal with most of life’s stresses.
• Never-ending goals: Triathletes may use their athletic goals as a way
to prove themselves, but they are rarely satisfied even when they reach their
goals. They are constantly looking toward the next target, but they can’t relax
and enjoy the journey.
• No pain, no gain: Triathletes continue to train through pain that worsens
during a workout, a sure road to injury or loss of motivation. Triathletes often
get into trouble by testing pain thresholds on a daily basis. Many triathletes
are afraid to rest because they have trouble changing gears. They can’t relax
and trust that their bodies will remember what to do.
Psychology of Triathlon Training | 565

• Lack of variety: Doing the same repetitive activity week after week
with little or no variation can bring about a sense of boredom or declining
interest. Even if physical injury doesn’t affect enthusiasm, mental strain can
eventually take its toll.
• Continual pressure and seriousness: With the demands of a rigid
training schedule and the pressures of competition, triathletes can lose the
element of fun in workouts. Building more enjoyable periods into a training
routine is critical for preventing burnout.
• Negative thinking: Triathletes may frequently condemn themselves for
failures, setbacks, and mistakes rather than realizing that these things are
inevitable and may offer good opportunities for learning.
• Avoidance: They can become overanxious about meeting a goal, which
may lead to their subconsciously seeking a way out. A sick or injured triath-
lete doesn’t have to face the pressures of the real competition.
Obsession and burnout are barriers that prevent triathletes from fully
enjoying their training and reaching their best performances. Overtrain-
ing without having regular breaks can often lead to injury and suboptimal
performances.

Tips for Preventing Staleness


The following strategies may help build variety into a triathlete’s life and
create renewed interest and passion in training. A triathlete who notices any
of the symptoms of staleness should address them immediately, because
they will not go away on their own.
• Change the routine by cross-training: Modify the normal training
schedule, even if only for a few days. Try active rest by doing a different
sport such as hiking, in-line skating, or cross-country skiing for a few days.
Breaks and variation in routine can give triathletes the emotional energy
that they need to make them hungry again for training and racing.
• Build in more diversity: Go to a new, scenic place to train at least once
a week. Take a running or biking vacation to explore new trails in national
parks, tropical beaches, or high mountain peaks.
• Listen to the body: Coaches and triathletes should become partners
with the triathlete’s body, giving it regular attention and dialoguing with it.
Triathletes need to become acutely aware of what the body and mind need
each day for proper recovery. Both body and mine need to know when it’s
time to push hard and when there is a need to back off.
The tips offered in chapter 41 on mental toughness apply here as well:
Create a disciplined mind-set, practice being focused and relaxed, picture
goals every day, build a balanced lifestyle, reduce pressure, and use setbacks
as learning opportunities.
566 } Dahlkoetter

For triathletes, the overall goal is to stay passionate but not become
obsessed. Coaches and triathletes should create a healthy, sustainable training
program and lifestyle to prevent slumps, injuries, and burnout. Triathletes
need to discover the point of pushing just hard enough without going over
the edge in training. In each area of life they need to know how much is
enough. They should try to put these mental training techniques to work
and learn to live more fully, train more healthfully, and perform best when
it counts the most.

Mind–Body Awareness
Sport psychologists working with triathletes at every level often hear their
patients say, “Whenever an important race comes up something goes wrong,
either mentally or physically. I can’t seem to pull my mind and body together
when it really counts. I seem to be continually sabotaging myself.”
For example, running requires so many bones, muscles, and connective
tissue to work well together that the triathlete can easily misjudge the body’s
limits and incur an injury.
Coaches often teach triathletes not to pay too much attention to the body’s
subtle messages of aches and pains. Coaches might say, “Don’t be a wimp.
Just get through the workout. We have a race this Sunday.” So triathletes
often ignore the subtle symptoms and plod on, making sure to complete the
designated weekly distance, doing a small amount of tissue damage each
day, until the body screams for attention. After an injury sets in, the healing
and recovery process then takes much longer because they did not have the
mind–body awareness skills to pay attention each day.

Tuning in Early to the Body’s Messages


In fact, the body has critical information for triathletes and coaches. They
need only to listen carefully, sense its subtle messages, and follow its direc-
tion regularly. Triathletes first need to check in with their bodies before
they blindly move ahead with a training plan. To train consistently and stay
healthy, triathletes need to treat and care for their bodies as they would a
finely tuned machine.
When working to improve mind–body awareness, a technique involving
listening and dialoguing allows triathletes to tune in to the body and mind
with a sense of compassion and curiosity. Many athletes have successfully
used this dialoguing technique. The process does not take extra time, only
greater awareness during the time when athletes are training. Triathletes can
learn a method of sensing how they feel, physically and mentally. They can
then have a dialogue with their feelings and sensations, and do most of the
listening. Triathletes tend to be action oriented, driven, and goal directed.
They are accustomed to telling their bodies what to do, most of the time not
so tactfully.
Psychology of Triathlon Training | 567

A triathlete may say to himself, “Pick up the pace, you slug. Your shoul-
ders are too stiff. Why aren’t you catching that guy in front of you?” What
do triathletes do when an uncomfortable feeling develops in the body (e.g.,
calf cramping or knee aching)? The typical reaction is to try to override it.
Perhaps yell at it a bit. “Why does this stupid pain have to come on right
now, when I’ve got 13 miles to go in the run?” Or they might beat them-
selves up mentally, “If only I had trained harder, my knee wouldn’t freeze
up like this.”

Mind–Body Awareness Strategies


The key to doing well in triathlon and preventing injuries is to become quiet
and listen to the body’s sensations on a daily basis. The key to mind–body
awareness for preventing injury is for triathletes to let the body give this
crucial feedback, especially while warming up for regular workouts. Then
they can take in and respect its wisdom before deciding on a course of action.
When triathletes let the body speak, they are allowing themselves to be
open to the depth and richness of their whole self—body, mind, and spirit.
When they pay attention to a sensation such as tension or pain, it is more
likely to be resolved so that they can go on with their training in a clear,
focused way. Coaches and triathletes may also gain a better sense of what
they need to do, if anything, to help the body function more efficiently (e.g.,
alter posture, training form, breathing, and so on). Often they don’t need to
change anything at all. The body can heal and correct itself. For instance, if
a minor calf strain occurs, awareness and backing off on the training plan
can often take care of the problem.
During a workout, triathletes may wonder about many issues: “Am I doing
the right thing for my body today? Should I be going this far or this fast when
I’m just returning from an injury? Is this workout helping to strengthen that
area, or is it further aggravating the tendons that are still weak and vulnerable?
Am I ready to take my training to the next level, or should I be conservative
and stay with the same workouts for another week?” Regular listening and
dialoguing with the body, as described in the following section, can fully
answer all these questions and give triathletes a strong sense that they are
doing the right thing.

Checking in With the Mind and Body


The body is wise in many ways beyond what the average person acknowl-
edges. If it is allowed to do so, the body can show the path to optimal health
and fitness. The body carries knowledge not only about how the triathlete
is training but also about how she is treating herself, what she values and
believes, how she has been hurt emotionally, and how she is living her life.
Triathletes’ physical selves know the best path to follow to move up to the
next level in performance.
568 } Dahlkoetter

Many triathletes recognize that they can confer with their bodies to assess
whether they are eating the proper food and getting enough sleep. They
understand that their bodies know what it feels like to be in good health and
what it feels like when they are on the edge of a cold or flu or injury. They
know the positive sensations of having energy to spare and being focused,
and when they’ve become too tired or let their bodies get torn down.
Indeed, coaches and triathletes understand how to identify the extreme
states, but do they recognize the subtle stages in between? By training ardu-
ously and regularly challenging the muscles and tissues to work harder, tri-
athletes are continually tearing down the body. And they need to remember
that they are also challenging the mind. It, too, needs recovery. This repair
and healing process is going on constantly, for both the body and mind. The
question is, What can triathletes do to be more aware and to facilitate the
healing process? With some practice and exercises, triathletes can improve
their mind–body awareness.

Mind–Body Awareness Exercise


Triathletes should think about an area of the body that they would like to
work on and begin a dialogue with. They should ask themselves, “What
bodily sensation am I most aware of? What part needs my attention?” After
they identify a certain area, they should ask, “What exactly does it feel
like?” Triathletes should find the right words to describe it, making sure
that they have the best words to articulate how it feels (e.g., tight shoulders,
heavy feet, mind racing, and so on). They can write down these thoughts
in a journal later.
Now they can begin asking that part of the body what it needs: “What
needs to happen next for it to get better?” They can learn, for instance,
whether it needs more rehab care or more rest. Does it need some special
treatment?
Next, they could ask the body to show how it would feel to be completely
healed. How would it feel to have the problem cleared up, resolved satis-
factorily? The body knows how it would feel to have the situation resolved,
even though a triathlete’s analytical mind may have no idea how that could
happen. That kind of suggestion helps create an opening for new ways of
being and perhaps healthier ways of treating the body.
The body’s solutions are infinitely better, more creative, and healthier than
anything the logical mind can dream up. Triathletes may feel stuck, but the
body isn’t. After they begin to listen, understand, and dialogue with their
bodies, their bodies will show their wisdom. The body will show better
ways for staying healthy and moving forward with training. Triathletes
can then train more consistently and remain healthy throughout the year,
gaining a new level of sensitivity to the body that can take them to new
performance highs.
Psychology of Triathlon Training | 569

Relaxation
One of the most overlooked elements in triathlon training programs also
happens to be the one most crucial for good performance—the ability to relax
the mind and body. Championship performances look smooth and effort-
less because the top athletes know how to relax. Elite triathletes consistently
report that the key to smoother, faster training is to focus on being relaxed
rather than trying to create more power.
Triathletes who try to muscle through a workout can create a lack of syn-
chronicity in the muscle groups. For instance, for an athlete to run efficiently,
the hamstrings must be relaxed as the quads are contracting. This aspect can
be critical in the final stages of a race when fatigue sets in and the pressure
mounts. Remaining loose and calm can make all the difference, enhancing
the ability to perform when it really counts.
One study examined the relationship between basic psychological skills
usage (i.e., goal setting, imagery, self-talk, and relaxation) and the intensity and
dimensions of competitive anxiety.3 Interviews were conducted on a sample of
elite athletes from a variety of team and individual sports. Findings revealed
that the participants maintained the intensity of their anxiety response before
competition but could then use goal setting, imagery, or self-talk to reinterpret
their anxiety-related symptoms and enhance performance. Higher levels of
self-confidence and an optimistic outlook toward forthcoming competitions
were also expressed. The athletes reported that they perceived control over
their anxiety responses when they used relaxation techniques.
Triathletes are continually moving between tension and relaxation, both
mentally and physically. They need a certain amount of tension to perform
well, but if their muscles become too tight, they lose fluidity and sense of
control. Each triathlete needs to find that right balance to move and stay
within the comfort zone.

What Happens When Triathletes Relax


Relaxation can be done both while being active and while being still. Relax-
ation does for the mind what stretching does for the body. It is both a mental
and physical experience. Relaxation is state of physical and mental stillness
characterized by the absence of tension and anxiety. It means letting go and
sometimes doing absolutely nothing with the mind and muscles. Although
relaxation is one of the more natural and satisfying states that human beings
can attain, the feeling of calmness must be experienced to be fully understood.
Two positive effects occur when people discover how to relax. First, a
physiological response occurs. Those who are most relaxed go into a slow,
deep abdominal breathing pattern and experience a decrease in heart rate.
The electromyograph (EMG) shows diminished muscle tension, and the hand
thermometer shows warmth, indicating more blood flow to the hands and feet.
570 } Dahlkoetter

Second, a psychological response ensues. The electroencephalograph (EEG)


indicates that relaxed triathletes who are still while visualizing go into an
alpha state, creating more brain waves in this creative and healthful state
of consciousness. This state enhances their ability to concentrate and move
away from anxiety and negative thoughts, a state of mind more conducive
to performing well. Relaxation is also an enabling condition. Those who are
physically and mentally relaxed are empowered to accomplish feats not pos-
sible at other levels of consciousness.

Benefits of Relaxation
Learning relaxation has many positive rewards that extend far beyond the
arena of sport. People can begin to see gains in every part of life. Triathletes
need to clear their minds and calm themselves more completely for a number
of reasons.
• Prepare for a triathlon: Calm the mind and conserve energy in the body
to perform well in an important event. Triathletes need to be fully rested for
maximal exertion.
• Decrease tension levels: Find the optimal stress level and reduce unnec-
essary tension and anxiety.
• Create fluidity of muscles: Triathletes can train more efficiently because
their antagonistic (nonworking) muscles are loose. Their coordination and
strength are greatly enhanced because they are using only the muscles nec-
essary to perform the task at hand.
• Lower the risk of injury: Injury is often a result of muscle tightness. The
healthiest muscles are those that are loose and relaxed.
• Decrease fatigue: Triathletes often exhaust themselves with the day’s
activities. Those who are overanxious are unconsciously contracting their
muscles throughout the day, which slows the blood flow process needed for
rapid recovery. Periodic relaxation exercises can help triathletes feel recharged
and energetic.
• Enhance quality of sleep: Triathletes need to be able to reduce their
internalized pressure and fall into a deep and restful sleep on the night before
a competition or after traveling to a new environment.
• Accelerate the natural healing process: During any given winter,
chances are high that triathletes will have a flu virus. Yet their own healing
system (immune system) will deal with it effectively without their awareness,
as long as they are allowing enough relaxation and recovery time.
• Repair of bone and soft tissue: Running and other weight-bearing
activities are high-maintenance activities. The body is constantly repairing
itself. Some have found that broken bones and connective tissue strains can
be healed in a shorter time when they use mental relaxation and imagery
techniques.
Psychology of Triathlon Training | 571

• Recharge the mind: Triathletes who learn the art of relaxation can
quickly discharge any pressure and restore the feelings of excitement and
joy about training.
• Open and expand consciousness: When the mind is fully relaxed, nega-
tive thinking and self-criticism are greatly reduced. The right brain becomes
more open to new ideas, and triathletes can begin to work with visualiza-
tion techniques in a powerful way. They can create vivid mental images of
exactly what they want to happen and increase the chances of turning those
images into reality.

Self-Assessment for Tension and Stress


The key to successful performance in any sport is to recognize the early
warning signs of rising stress levels. Triathletes then need to take immedi-
ate action to bring stress levels down before the body screams at them for
attention. Each person typically has a characteristic pattern in the response
to stress. What system in the body overacts to pressure? Do triathletes notice
any of the following reactions? Triathletes should rate themselves on each
of the reactions listed on a scale of 1 to 10 (10 being the highest or worst) by
putting a number next to each symptom:
• Racing heart
• Muscle tension in jaw, neck, shoulders
• Rapid, shallow breathing
• Stomach upset or vomiting
• Desire to urinate
• Irritability
• Forgetting details
• Inability to focus or make decisions
• Resorting to old habits
• Feeling fatigued
• Catching colds frequently
Triathletes should know their individual responses to stressful situations
such as a competition and begin doing relaxation exercises before the symp-
toms get out of hand.

Muscle Relaxation Exercise


When Usain Bolt broke the world record in the relay in the 100 meters at the
world championships, his face looked peaceful and serene as he was coming
in to the finish line. Triathletes often report, “Relaxed running seems so easy
that it feels as if you’re hardly working at all.”
572 } Dahlkoetter

With regular practice of relaxation techniques, triathletes can begin to


look and feel like world-class competitors. Here is a useful exercise to reach
deeper levels of relaxation and higher levels of performance.
When doing various tasks, triathletes should try monitoring the muscle
tension levels that accompany common activities throughout the day. They
should notice how they hold their wrists and shoulders while typing at the
computer, how tightly they are gripping the steering wheel while driving,
what muscles they are using to hold the phone (is their neck crimped?), how
tight their neck and shoulders are while they are doing the last interval in
a workout. Triathletes frequently use far more muscle power than needed
to accomplish a task effectively. When they muscle through a workout, they
block the body’s natural timing, flow, and rhythm. The movement becomes
jerky and uncoordinated, causing fatigue to develop more quickly.
The goal in this exercise is to match the effort with the task. Top triathletes
are skilled in detecting subtle differences in tension levels and fine-tuning
their responses, adjusting muscle tension to whatever job they are doing.
They should relax the muscles not needed for the task.

Conclusion
The long-range challenge for triathletes in using the psychology of training
is to create a healthy, balanced lifestyle and prevent staleness, injuries, and
burnout. Triathletes need to find that point where they are pushing just hard
enough without going over the edge in each part of their training. After
practicing these techniques both at home and during training, triathletes
will find that triathlons begin to feel easier. They can take their minds off
the feeling of tension and fatigue and move straight into focusing on the
task at hand. They’ll look forward to pressure situations as an opportunity
to fine-tune relaxation and have their bodies and minds perform well when
it really counts.
CHAPTER
43
Mental Skills for Peak
Triathlon Performance
—JoAnn Dahlkoetter, PhD

C onfidence, self-worth, and optimism are all key elements in determining


how triathletes view a triathlon challenge and transform it into a power-
ful performance. Developing a realistic, healthy self-image is the foundation
for success in sport and in life. When triathletes know themselves well and
are confident in their abilities, they’re well on their way to realizing their
goals. Self-confidence gives them the ability to create and sustain an optimal
performance state regardless of the external conditions.
For example, many triathletes gain a confidence boost when their fast
training times indicate that they can do well in an upcoming race. A study
by Knechtle, Wirth, and Rosemann showed that faster training runs and
setting a fast personal best time in a marathon and in an Olympic-distance
triathlon were associated with a subsequent fast Ironman race time.1
Top triathletes have a strong, solid belief in themselves and their ability
to perform well. Their confidence is so deep that it is almost indestructible,
unaffected by outside influences. Extremely self-confident triathletes are
resilient to setbacks, can deal with pressure, and are not easily intimidated.
For example, six-time Hawaii Ironman Champion Mark Allen had a presence
about him, an extraordinary self-assurance that at times could appear intimi-
dating to others. His belief in himself was reflected in his performances.
Confidence, like determination, discipline, and strength, are fortunately
all part of a constellation of mental skills that are learned and not inherited.
Therefore, with regular practice they can become part of a triathlete’s skill set.

Self-Confidence
Self-esteem and positive self-image are essential for performing well in any
sport. One of the main factors differentiating humans from other animals is
the awareness of self. Humans have the ability to form an identity and then

573
574 } Dahlkoetter

attach a value to it. They define themselves by certain standards and then
decide whether they like themselves.
The struggle that often occurs comes from the human capacity for self-
judgment. If a triathlete rejects or criticizes himself for doing poorly in a race,
he’ll find himself avoiding anything that might bring on the pain of further
self-rejection. He may then take fewer risks in training and racing, as well as
in social and professional situations. In short, people may limit their ability
to realize their full potential.
The good news is that people can learn to stop making these judgments
and direct their energies toward building a more constructive self-image.
Triathletes can change how they feel about themselves. They can learn to
recognize their positive qualities and acquire an attitude of acceptance toward
themselves and others. After those perceptions change, they will see improve-
ments in training and in every part of their life with a gradually expanding
sense of inner freedom.

Transforming Weaknesses Into Strengths


If triathletes are content with a mediocre performance, they will never chal-
lenge themselves to reach their full potential in a competition. If they really
want to improve their training and build confidence, they need to go out of
their way to address and correct their weaknesses.
Many triathletes fail in their workouts because they are unable to over-
come a singular personal deficiency. They adopt a negative attitude toward
weaknesses. They justify a shortcoming by saying that they were born with
this particular problem or have since acquired it and that they can do noth-
ing about it. Alternatively, they defend a weakness, such as their swimming
ability, and in the process start to structure their lives around it, saying, “I’m
not a good swimmer, I’ll never get faster.” They make the weakness the center
of their thinking instead of facing the issue head on and conquering it by
joining a masters swim program or taking a swim class or lessons.
So to build confidence, triathletes must first understand their specific prob-
lem areas and then go to work. By making the weakest link in their chain
strong, they can’t help but improve. The best triathletes in the world are those
who actively work to overcome their weaknesses. Practicing strengths is fun
because we excel in those areas and can easily get positive feedback. But the
way for triathletes to improve is to test themselves, to learn where they are
making mistakes, to face them, and to turn them around.
For instance, in a race in which a competitor surges ahead of them, dedi-
cated triathletes try to stay with the competitor and use that triathlete’s energy
to pull them along to a personal best. If they cannot meet the challenge, they’ll
know that it’s time to go back to the pool or the track and work on their ability
to increase the pace and recover quickly within the workout.
When a race is on the line, there is no substitute for both physical and mental
skill. If triathletes have addressed their weaknesses, they won’t be vulnerable.
Mental Skills for Peak Triathlon Performance | 575

They’ll be fully prepared, confident, and ready to encounter anything that


the competition might give them.

Building Confidence
Triathletes should try using some of the following techniques and choose the
strategies that seem to fit them best, including modeling positive qualities,
engaging in healthy self-talk, building self-acceptance, and making positive
self-statements.

Modeling Positive Qualities


Let’s say that a triathlete wants to train more efficiently, with a sense of con-
fidence, lightness, and power. She should select another triathlete who pos-
sesses those qualities and visualize that person’s style during her workout.
She imagines being that triathlete, floating effortlessly with endless amounts
of energy and self-assuredness. For example, a triathlete who runs with tight
shoulders or a posture either too far forward or too straight up and down
can model a training partner who has a better form. A triathlete who slows
down at the end of races can mimic her partner, who always finishes strong,
in both practices and events.

Engaging in Healthy Self-Talk


Triathletes can work to diminish the intensity of negative self-attacks while
nourishing more healthy self-talk. They may never entirely turn off the inner
voice that says, “You really screwed up that workout; you’re stupid,” but they
can tone down its volume and significance. An effective way to go about
quieting this negative voice is to imagine turning down the volume on a
radio or portable media player. Triathletes can do the same thing. With each
breath out, they can turn down the volume on the negative voice. And when
someone gives them a compliment, rather than dismissing it they should take
it in and let it enhance their self-esteem.

Building Self-Acceptance
Triathletes frequently use their watches, their scores, or their performances to
define their self-worth. They may say, “I need to run a sub-40 minute 10K to
feel OK about myself.” A disappointing race performance does not indicate
that they are poor triathletes, nor is it a commentary on their real potential.
Triathletes need to begin with a firm foundation, a secure sense of self. They
must consciously lighten up on harsh self-judgments and come to accept
themselves as valuable people regardless of the outcome.
Boosting confidence is critical to good performance, and many other
attributes follow in its wake. Self-confident triathletes are also optimistic,
motivated, focused, and unafraid to take risks. They move toward challenges
with inner strength and courage and find personal rewards in each endeavor
576 } Dahlkoetter

regardless of the outcome. As their self-image becomes more positive, the


degree of excellence will correspondingly rise in all areas of their lives.

Making Positive Self-Statements


Triathletes should write out their list of personal strengths and build positive
affirmations around each one. They should create simple positive phrases to
say to themselves silently or aloud to reinforce their positive qualities. Here
are some examples:
• I am a strong, efficient triathlete; I am improving rapidly.
• I believe in myself; I radiate confidence.
• I strive to be positive and enthusiastic, no matter what happens.
• I feel a sense of power, confidence, and inner strength when I compete.
• I thoroughly enjoy myself as I train and compete.
• I am consistently working to address my weaknesses.
• The results will take care of themselves; I simply perform.
• My body and mind are growing stronger and healthier every day.
• Mistakes are simply feedback; I am open to learning.
• I focus on doing the very best that I can at every moment.
• I am willing to do whatever it takes to meet my goal.

Focus and Determination


Success in triathlon training requires total concentration and determination,
for both novices and those at the highest levels of performance. If the mind
begins to wander, a triathlete can easily become distracted and lose the edge.
The decline affects not only performance but also the quality of the exper-
ience. The person cannot take pleasure in an activity when not fully present.
For instance, a 2011 study by Memmert indicated that skilled athletes who
had high attention scores performed better than skilled players who had
low attention scores, in accordance with specific creative-thinking abilities.2
Complete and undivided attention is difficult to come by. Triathletes can
easily become sidetracked by myriad external factors or the mind’s thought
processes. Focusing is a challenge when numerous tasks compete for atten-
tion (e.g., preparation for their event, work responsibilities, family demands,
children’s needs, and so forth).
Triathletes may have input from several sources at once—a coach, a train-
ing group, or a sports medicine doctor. The challenge is to integrate all this
information, separate the important issues from the nonessential ones, and
make important decisions about training. Knowing how to focus and being
able to maintain concentration throughout the day is critical for triathletes
at any level.
Mental Skills for Peak Triathlon Performance | 577

Present-Centered Focus
Successful concentration depends on a present-centered focus in which tri-
athletes are totally connected to the current task. A present-centered focus is
one in which all attention is directed to what is occurring now. So, for instance,
in a race, a triathlete may be focusing on what the competitor in front of him
is doing, or on how his body is feeling, or what decision he is going to make
by having that information.
Concentration is the learned skill of fully attending to the task at hand and
excluding irrelevant external cues and internal distractions. Internal factors
for triathletes might include self-doubt, fears, expectations, or fatigue. External
distractions may involve heavy traffic on the travel route to a competition,
a rival competitor showing up unexpectedly, or equipment problems. Tri-
athletes need to be able to concentrate in spite of those disruptions. The true
test comes when the requirements of a particular situation to stay focused
extend beyond their current abilities.
Unfortunately, deep concentration is not as simple as it appears on the
surface. The brain is not programmed to stay in the present moment. The
capability to assimilate large quantities of information can make it difficult to
sustain attention on any one subject for a long period. Thoughts about the past
or future can cry out for attention whenever people try to focus on one subject.
The critical mind often focuses on the mistakes and the what-ifs—what
might have occurred and what should have occurred. Triathletes think about
the training that they should have done, or the critical move that was made
late in the competition, or the great shape that they were in last year. Focus-
ing on what could have happened will not change the past, but the present
can suffer because of trying to change it.
Now look at how top triathletes tune in to the current moment during a
peak performance. Throughout these periods, triathletes focus on the precise
details of the event that allow them to respond optimally. They often report
intense focus on one activity—cycling up a hill and choosing the right gears
or finishing the race at a certain pace. Top triathletes are flexible enough to
assess and make changes in their mental focus easily.
Triathletes need to focus on the present to be successful. Key aspects and
tools needed to do so include process goals, refocusing, and changing focus.

Process Goals
Besides tuning into the present moment, a triathlete needs to focus on the
right objective to do well in training and competition. What are they focus-
ing on? Attending to process goals is far more effective than thinking about
the outcome. A process focus involves concentration on the specific task
being done rather than the result. So, for instance, a triathlete may focus on
technique, breathing, pace, or mental attitude. A process goal keeps attention
directed toward what needs to be done right now.
578 } Dahlkoetter

Refocusing
If triathletes lose their focus, their aim should be to regain it as quickly as
possible. When distracting thoughts interrupt their flow, they want to retune
the brain into the correct psychological channel. Awareness of inattention
is the first step. The sooner that triathletes notice the lapse in attention, the
quicker they can turn it around. The second step is selecting another point at
which to direct the mind. They can return their attention to the same focus
that they had before they were interrupted, or they can go to something else
of importance at that moment.
When doing a triathlon, for example, the right focus depends on what is
happening at that moment (e.g., finishing the swim or making a quick transi-
tion to the bike). So concentration becomes dynamic and changes according
to particular challenges of the event. Knowing what to focus on is a skill
that can be developed with time and experience. As triathletes encounter
a variety of scenarios, they can store their responses in memory, improve
their ability to take action, and learn to return their focus to a relevant object
more quickly.

Changing Focus
Triathletes also need to be flexible in their range of focus, able to switch
gears whenever necessary. For instance, during the start of a triathlon they
need to have a wide focus and attend to what is happening all around. They
want to be aware of others so that they don’t get run over by the people
behind them. By the middle of the race, they are more attuned to proper
pacing, their breathing, and their form. In the final stages of the race, they
switch to a narrower focus, pouring all their energies into putting one
foot in front of the other and getting themselves to the finish line. In other
words, they are focusing on the present, what is in front of them at that
moment.

Three Focusing Exercises


The following exercises will help triathletes hone their ability to maintain
focus and determination.

Breathing Exercise #1
Practice this exercise before beginning training each day. Find a comfortable,
quiet place where you will not be interrupted and note the time on a watch.
Close your eyes and narrow your focus to one point—breathing. Notice
the inhale and exhale. Continue this exercise for as long as you can sustain
your focus. When your mind begins to wander, open your eyes and note the
time on the watch. How much time passed since you started the task? Ten
seconds? One minute?
Mental Skills for Peak Triathlon Performance | 579

Breathing Exercise #2
Begin by doing deep abdominal breathing. Feel your belly rise and fall with
each breath. Upon the inhale, imagine that your breath is carrying particles
of concentration and determination into your body. With the exhale, notice
that all the distractions and stress are drifting away. The incoming breath is
like a sedative that supplies the peace needed to focus on the present.

Mental Rehearsal Exercise


Successful triathletes prepare for their events in advance. They have practiced
every scenario in their minds a hundred times before it occurs. The more
they rehearse their plan ahead of time in training, the more automatic it will
become and the less thought will be required during the actual performance.
They will be able to react appropriately to what is happening each moment.
After they have a game plan in place, they can keep their mind on what is
relevant and follow the plan when it really counts.
Triathletes at any level can experience the same single-minded focus and
drive that elite athletes use. Concentration is a learned skill that can improve
with practice. The key is to recognize that good concentration is a constant
vigil.
But if triathletes regularly work on their attention span, they can begin to
take control of the most distracting situations. As they sharpen their focus,
they can experience the effortless stretching of the mind and body because
they are totally immersed in the present. They will gain a keen sense of the
big picture and be able to anticipate the correct moves to handle any training
or racing situation that comes their way.

Visualization
A powerful resource for channeling performance energies more efficiently
is visualization, also known as mental imagery. In the Olympics a wide
range of elite triathletes have reported using visualization in their training
to improve performance. Practicing visualization can make all the difference
in how triathletes of any ability level experience their workouts and races.
For example, in a 2004 study by Ranganathan et al., exercise psychologists
compared “people who went to the gym with people who carried out virtual
workouts in their heads.” They found a 30 percent muscle increase in the
group who went to the gym. But the group of participants who conducted
mental exercises of the weight training increased muscle strength as well,
achieving nearly half as much gain (13.5 percent) as those who really worked
out. Those in the virtual workout group maintained their average for 3 months
following the mental training.3
The use of mental imagery for enhanced performance is not new. The
practice of martial arts in Asia, meditation and yoga in ancient India, and
580 } Dahlkoetter

hypnotherapy are other illustrations of how the mind’s capacity to picture


situations can be a critical part of athletic performance. Whereas mental train-
ing may have been viewed skeptically in the past, imagery and other similar
techniques have now become an integral part of most sport competitions.
Serious triathletes who want to engage in complete preparation train both
the body and mind for top-level performance.
To test the effect of imagery in the training of skilled movements, in 2007
Fontani et al. designed an experiment in which athletes learned a new motor
action and trained themselves for a month either by overt action or by mental
imagery of the action. The results show that mental imagery can influence
muscular abilities such as strength and power.4
Just as athletes work out their bodies, they also need to exercise their brains.
Visualization and mental training is like taking the brain to the gym. Coaches
and triathletes want their bodies to have many different gears and speeds. They
also need the mind to be flexible, to be able to shift into different gears depend-
ing on the task. Mental imagery is a powerful tool for achieving this purpose.
Images are the mental representations of our experience. Although verbal
language is the most common means for communicating with the external
world, images are a powerful means for internal communication. By using
imagery or visualization, triathletes can create, in vivid detail, a replay of one
of their best past performances or mentally rehearse an upcoming event and
learn to see themselves doing it right.
Imagery guides much of a triathlete’s experience because it is a more effi-
cient, complete language than self-talk. Using words to describe to someone
how to execute the perfect running style in detail could fill an entire book.
Showing the same running form through a video replay of a top triathlete
doing hill training can convey the same message in a few seconds.

Incorporating Visualization Into Training


Most people daydream and reexperience situations in their minds in a
haphazard way. The fact that they can remember previous experiences in
detailed fashion is why visualization works so well for triathletes. Most
good triathletes have discovered this technique on their own and may use
it occasionally to improve learning and performance. For maximum results,
they need to control their imagery and practice it regularly rather than just
let thoughts pass in and out. Through their visualization they can re-create
the past in detail and transform it to fit any situation that they encounter.

Types of Imagery
There are three primary types of imagery:
1. Internal visual imagery: With this type of imagery, triathletes mentally
rehearse what they see with their own eyes as they execute an event. It’s as
Mental Skills for Peak Triathlon Performance | 581

if they have a video camera on their head that records what they are seeing
as they are moving.
2. External visual imagery: Triathletes become outside observers, as
though they are watching a movie of themselves performing. This method is
useful for analyzing form or distancing themselves from pain during a race.
3. Kinesthetic imagery: Triathletes don’t see anything but experience the
event through their sense of touch. For instance, they feel the wind on their
face or notice their arms propelling themselves forward as they run.
Visualization is not wishful thinking or daydreaming about being a great
triathlete. It is a learned skill that requires effort, concentration, discipline,
and regular practice to gain the maximum benefits. Following are the key
principles for triathletes to use in creating effective mental imagery.

Guidelines for Effective Visualization


Use visualization almost any time—at home or before, during, or after train-
ing and racing. In the learning phases imagery is easier to do in a quiet,
nondistracting environment. Visualization is most effective when the mind
is calm and the body is relaxed.
Begin the session with a few minutes of deep abdominal breathing. Put
one hand on your belly about 2 inches (5 cm) below the belly button and feel
your hand rise and fall with each breath. With each inhale, imagine filling
up a balloon inside your abdomen; during the exhale, the balloon collapses
back down.
Create in your mind an image as vivid as possible of what you want to
achieve in the sport. Just let distracting thoughts and feelings float away and
refocus on the image.
Bring in the senses to see, hear, and feel what it’s like to have a great train-
ing run or race. Sight, hearing, and touch are the most powerful senses for
incorporating day-to-day visualization into training. Bring the scene into the
present tense to be focused totally on the task at hand.
Don’t replay the mistakes. You should see what you want to see, not what
you don’t want to see. Strive to remove the memory of errors. If you see
yourself doing something incorrectly, edit the film in your mind and replay
it exactly as you wish it to happen, imagining that your performance is equal
to or better than your previous best.
Mentally rehearse training at the same rhythm and pace that you want to
achieve in actual execution to establish the appropriate neurological pattern
within your brain.
Use visual models. Before going to sleep at night, try watching a video
of a superior performance (go to YouTube or find recordings from the last
Olympics). Then visualize yourself moving just as fluidly and powerfully as
your favorite triathlete does.
582 } Dahlkoetter

To train the body properly, you must first train the mind. Inspiring images
can create powerful emotions and produce superior performances. So the
best approach is to focus on positive images and memories.
At first, triathletes may not fully believe that they can perform up to the
level of their visualizations. Acting as if it is already happening is OK. With
practice the body will come into line with mental images. If a negative image
comes into the mind, triathletes just need to breathe deeply, let it go with
the exhale, and then bring in a positive image with the next breath. As they
practice and refine their mental training, their images will become clearer
and more convincing.
Remember that visualization can be effective at any level of triathlon
training or competition. Mental and physical training can complement one
another. The more familiar that triathletes become with the intricacies of
triathlon, the more effective their mental practice will be.
Of course, visualization is not a substitute for physical training; triathletes
still have to put in the distances. But visualization can make all the difference
in the quality and enjoyment of training and racing. It will move triathletes
much closer to realizing their true potential.

Conclusion
Gaining mental skills can give triathletes an edge in training and racing. Tri-
athletes are indeed multidimensional, a synthesis of body and mind, capable
of accomplishing extraordinary things. This chapter has explained the many
elements of the successful triathlete, both psychological and physical.
The purpose of mental skills training is to nurture and develop all these
attributes so that each part supports every other part. Triathletes should seek
to experience the broader, deeper elements of the sport. By doing so, they can
begin to comprehend the profound connection between mental training and
the subsequent changes in performance.
After they understand these concepts, triathletes can take the lessons from
triathlon and enhance the way that they think, feel, and behave in everyday
life. They can enjoy and appreciate the value of the triathlete lifestyle and
realize their true potential.
Epilogue
The Application of Science in
Triathlon
—Joe Friel, MSc

I s training to perform at a high level in triathlon only about training longer,


harder, and more often? Having read this book and now perhaps having a
deeper understanding of the science of training for triathlon, you know the
answer: A lot more goes into triathlon performance than just lots of swim-
ming, biking, and running.
In the early days of triathlon, training was simple—work out long, hard,
and often. Science has taken us well beyond this oversimplified approach
to performance. The requirements for success in triathlon are many and
varied, from physiology and genetics to equipment and psychology—and
much more. Therein lies the challenge for you.
The sport of triathlon is complex. Many variables are involved, not the
least of which is the unique characteristics of the athlete. As a coach, athlete,
sport scientist, or student of sport, you must give a great deal of thought to
what is deemed necessary for boosting performance given these individual
considerations. Where should the triathlete invest time, energy, and money?
What approach is best to meet the triathlete’s particular needs? There are
no easy or one-size-fits-all answers; there are only options. And that is what
makes sport, especially one as complex as triathlon, so intriguing.
Science has not come up with any guaranteed training methods that fit
all athletes all the time. If we have learned one thing from research, it is that
what works on average for a group of subjects doesn’t affect all the same way.
In the final analysis, it boils down to the individual. And an effective, well-
planned approach isn’t even a fixed position for any given athlete. Changes
are always taking place in the triathlete’s physiological and psychological
makeup, lifestyle, goals, and more. What works—and what doesn’t—depends
not only on what is known from science but also on what the triathlete has
experienced and how he currently responds to a training dose.
So now the fun begins. Application. I hope that Triathlon Science has given
you a platform on which to begin applying what you have learned from the
best that sport science has to offer.

583
This page intentionally left blank.
References
Chapter 1 Physiology and the Multi- 5. Bray, M.S., Hagberg, J.M., Pérusse, L., Ranki-
sport Athlete nen, T., Roth, S.M., Wolfarth, B., and Bouchard,
C. 2009. The human gene map for performance
1. Tannerstedt, J., Apro, W., and Blomstrand, and health-related fitness phenotypes: The
E. 2009. Maximal lengthening contractions 2006–2007 update. Medicine and Science in Sports
induce different signaling responses in and Exercise, 41, 35–73.
the type I and type II fibers of human skel-
6. Flynn, M.G., Costill, D.L., Kirwan, J.P., Fink,
etal muscle. Journal of Applied Physiology, 106,
W.J., and Dengel, D.R. 1987. Muscle fiber com-
1412–1418.
position and respiratory capacity in triathletes.
2. Wernbom, M., Augustsson, J., and Thomee, International Journal of Sports Medicine, 8(6),
R. 2007. The influence of frequency, intensity, 383–386.
volume and mode of strength training on
7. Simoneau, J.A., and Bouchard, C. 1995.
whole muscle cross-sectional area in humans.
Genetic determinism of fiber type proportion
Sports Medicine, 37, 225–264.
in human skeletal muscle. FASEB Journal, 9,
3. Yan, Z., Okutsu, M., Akhtar, Y.N., and Lira, 1091–1095.
V.A. 2011. Regulation of exercise-induced
8. Ahmetov, I.I., and Rogozkin, V.A. 2009. Genes,
fiber type transformation, mitochondrial bio-
athlete status and training—an overview.
genesis, and angiogenesis in skeletal muscle.
Medicine and Sport Science, 54, 43–71.
Journal of Applied Physiology, 110, 264–274.
9. Ahmetov, I.I., Druzhevskaya, A.M., Lyubaeva,
4. Bentley, D.J., Millet, G.P., Vleck, V.E., and
E.V., Popov, D.V., Vinogradova, O.L., and Wil-
McNaughton, L.R. 2002. Specific aspects of
liams, A.G. 2011. The dependence of preferred
contemporary triathlon: Implications for
competitive racing distance on muscle fibre
physiological analysis and performance. Sports
type composition and ACTN3 genotype in
Medicine, 32, 345–359.
speed skaters. Experimental Physiology, 96,
5. Laursen, P.B. 2011. Long distance triathlon: 1302–1310.
Demands, preparation and performance.
10. Bouchard, C., An, P., Rice, T., Skinner, J.S.,
Journal of Human Sport and Exercise, 6, 247–263.
Wilmore, J.H., Gagnon, J., Pérusse, L., Leon,
6. Laursen, P.B., and Rhodes, E.C. 2001. Factors A.S., and. Rao, D.C. 1999. Familial aggrega-
affecting performance in an ultraendurance tion of VO2max response to exercise train-
triathlon. Sports Medicine, 31, 195–209. ing: Results from the HERITAGE Family
Study. Journal of Applied Physiology, 87, 1003–
Chapter 2 Genetics and Inheritance in 1008.
Triathlon Performance 11. Timmons, J.A., Knudsen, S., Rankinen, T.,
1. Lettre, G. 2009. Genetic regulation of adult stat- Koch, L.G., Sarzynski, M., Jensen, T., Keller, P.,
ure. Current Opinion in Pediatrics, 21, 515–522. Scheele, C., Vollaard, N.B., Nielsen, S., Aker-
ström, T., et al. 2010. Using molecular classi-
2. Yang, J., Benyamin, B., McEvoy, B.P., Gordon,
fication to predict gains in maximal aerobic
S., Henders, A.K., Nyholt, D.R., Madden, P.A.,
capacity following endurance exercise training
et al. 2010. Common SNPs explain a large pro-
in humans. Journal of Applied Physiology, 108,
portion of the heritability for human height.
1487–96.
Nature Genetics, 42, 565–569.
12. Bouchard, C., Sarzynski, M.A., Rice, T.K.,
3. Knechtle, B., Wirth, A., Baumann, B., Knechtle,
Kraus, W.E., Church, T.S., Sung, Y.J., Rao, D.C.,
P., Rosemann, T., and Oliver, S. 2010. Differ-
and Rankinen, T. 2011. Genomic predictors of
ential correlations between anthropometry,
the maximal O2 uptake response to standard-
training volume, and performance in male and
ized exercise training programs. Journal of
female Ironman triathletes. Journal of Strength
Applied Physiology, 110, 1160–1170.
& Conditioning Research, 24(10), 2785–2793.
13. Collins, M., Xenophontos, S.L., Cariolou, M.A.,
4. Peterson, R.E., Maes, H.H., Holmans, P., Sand-
Mokone, G.G., Hudson, D.E., Anastassiades,
ers, A.R., Levinson, D.F., Shi, J., Kendler, K.S.,
L.C., and Noakes, T.D. 2004. The ACE gene
Gejman, P.V., and Webb, B.T. 2011. Genetic risk
and endurance performance during the South
sum score comprised of common polygenic
African Ironman triathlons. Medicine and Sci-
variation is associated with body mass index.
ence in Sports and Exercise, 36, 1314–1320.
Human Genetics, 129, 221–230.

585
586 } References

14. Saunders, C.J., Xenophontos, S.L., Cariolou, 25. Collins, M., and Raleigh, S.M. 2009. Genetic risk
M.A., Anastassiades, L.C., Noakes, T.D., and factors for musculoskeletal soft tissue injuries.
Collins, M. 2006a. The bradykinin beta 2 Medicine and Sport Science, 54, 136–149.
receptor (BDKRB2) and endothelial nitric oxide 26. September, A.V., Mokone, G.G., Schwellnus,
synthase 3 (NOS3) genes and endurance per- M.P., and Collins, M. 2006. Genetic risk factors
formance during Ironman triathlons. Human for Achilles tendon injuries. International Sport
Molecular Genetics, 15, 979–987. Medicine Journal, 7, 201–215.
15. Posthumus, M., Schwellnus, M.P., and Collins, 27. Collins, M., and Posthumus, M. 2011. Type V
M. 2011. The COL5A1 gene: A novel marker of collagen genotype and exercise-related pheno-
endurance running performance. Medicine and type relationships: A novel hypothesis. Exercise
Science in Sports and Exercise, 43, 584–589. and Sport Sciences Reviews, 39, 191–198.
16. O’Connell, K., Posthumus, M., and Collins, 28. Noakes, T.D., Sharwood, K., Speedy, D., Hew,
M. 2011. COL6A1 Gene and Ironman triathlon T., Reid, S., Dugas, J., Almond, C., Wharam, P.,
performance. International Journal of Sports and Weschler, L. 2005. Three independent bio-
Medicine, 32(11), 896–910. logical mechanisms cause exercise-associated
17. Woods, D. 2009. Angiotensin-converting hyponatremia: Evidence from 2,135 weighed
enzyme, renin-angiotensin system and human competitive athletic performances Proceedings
performance. Medicine and Sport Science, 54, of the National Academy of Sciences of the United
72–87. States of America, 102, 18550–18555.
18. Yang, N., Garton, F., and North, K. 2009. Alpha- 29. Saunders, C.J., de Milander, L., Hew-Butler, T.,
actinin-3 and performance. Medicine and Sport Xenophontos, S.L., Cariolou, M.A., Anastassia-
Science, 54, 88–101. des, L.C., Noakes, T.D., and Collins, M. 2006b.
19. Saunders, C.J., September, A.V., Xenophontos, Dipsogenic genes associated with weight
S.L., Cariolou, M.A., Anastassiades, L.C., changes during Ironman triathlons. Human
Noakes, T.D., and Collins, M. 2007. No associa- Molecular Genetics, 15, 2980–2987.
tion of the ACTN3 gene R577X polymorphism 30. de Milander, L., Kun, M.A., September, A.V.,
with endurance performance in Ironman tri- Schwellnus, M.P., Noakes, T.D., and Collins,
athlons. Annals of Human Genetics, 71, 777–781. M. 2012. AVPR2 gene and weight changes
20. Ahmetov, I.I., Druzhevskaya, A.M., Astraten- during triathlons. International Journal of Sports
kova I.V., Popov, D.V., Vinogradova, O.L., Medicine, 33(1), 67–75.
and Rogozkin, V.A. 2010. The ACTN3 R577X
polymorphism in Russian endurance athletes. Chapter 3 Gender and Age
British Journal of Sports Medicine, 44, 649–652. Considerations in Triathlon
21. Ahmetov, I.I., Williams, A.G., Popov, D.V., 1. Lepers, R., and Cattagni, T. 2012. Do older ath-
Lyubaeva, E.V., Hakimullina, A.M., Fedotovs- letes reach limits in their performance during
kaya, O.N., Mozhayskaya, I.A., Vinogradova, marathon running? Age (Dordr), 34(3), 773–781.
O.L., Astratenkova, I.V., Montgomery, H.E., and 2. Whipp, B.J., and Ward S.A. 1992. Will women
Rogozkin, V.A. 2009. The combined impact of soon outrun men? Nature, 335(6355), 25.
metabolic gene polymorphisms on elite endur-
ance athlete status and related phenotypes. 3. Sparkling, P.B., O’Donnell E.M., and Snow T.K.
Human Genetics, 126(6), 751–761. 1998. The gender difference in distance run-
ning performance has plateaued: An analysis
22. Maciejewska, A., Sawczuk, M., Ciszczyk, P., of world rankings from 1980 to 1996. Medicine
Mozhayskaya, I.A., and Ahmetov, I.I. 2012. The and Science in Sports and Exercise, 30, 1725–1729.
PPARGC1A gene Gly482Ser polymorphism in
Polish and Russian athletes. Journal of Sports 4. Knechtle, B., Knechtle, P., and Rosemann, T.
Sciences, 30, 101–113. 2011. Upper body skinfold thickness is related
to race performance in male Ironman tri-
23. Tucker, R., and Collins, M. 2012. Athletic per- athletes. International Journal of Sports Medi-
formance and risk of injury: Can gene explain cine, 32(1), 20–27.
all? Dialogues in Cardiovascular Medicine, 17(1),
31–39. 5. Lepers, R. 2008. An analysis of Hawaii Iron-
man performances in elite triathletes from
24. Dallam, G.M., Jonas, S., and Miller, T.K. 2005. 1981 to 2007. Medicine and Science in Sports and
Medical conditions in triathlon competition: Exercise, 40, 1828–1834.
Recommendations for the triathlon organis-
ers, competitors and coaches. Sports Medicine, 6. Tanaka, H., and Seals, D.R. 1997. Age and
35, 143–161. gender interactions in physiological func-
tional capacity: Insight from swimming per-
References | 587

formance. Journal of Applied Physiology, 82(3), elderly subjects. American Journal of Physiology,
846–851. 74, E785–790.
7. Schumacher, Y.O., Mueller, P., and Keul, J. 2001. 20. Lepers, R., Rüst, C.A., Stapley, P., and Knechtle,
Development of peak performance in track B. 2012 Relative improvements in endurance
cycling. Journal of Sports Medicine and Physical performance with age: Evidence from 25 years
Fitness, 41, 139–146. of Hawaii Ironman racing. Age (Dordr), doi:
8. Swain, D.P. 1994. The influence of body mass 10.1007/s11357-012-9392-z
in endurance bicycling. Medicine and Science in
Sports and Exercise, 26, 58–63. Chapter 4 Swimming Biomechanics
9. Knechtle, B., Knechtle, P., and Lepers, R. for Triathlon
2010. Participation and performance trends 1. Toussaint, H.M. 1990. Differences in propelling
in ultra-triathlons from 1985 to 2009. Scandi- efficiency between competitive and triathlon
navian Journal of Medicine & Science in Sports, swimmers. Medicine and Science in Sports and
doi:10.1111/j.1600-0838.2010.01160.x Exercise, 22(3), 409–415.
10. Lepers, R., Knechtle, B., Knechtle, P., and 2. Toussaint, H.M., Beelen, A., Rodenburg, A.,
Rosemann, T. 2011. Analysis of ultra-triathlon Sargeant, A.J., deGroot, G., Hollander, A.P., et
performances. Open Access Journal of Sports al. 1988. Propelling efficiency of front-crawl
Medicine, 2, 131–136. swimming. Journal of Applied Physiology, 65(6),
11. Ransdell, L.B., Vener, J., and Huberty, J. 2009. 2506–2512.
Master athletes: An analysis of running, swim- 3. Zamparo, P. 2006. Effects of age and gender
ming and cycling performance by age and on the propelling efficiency of the arm stroke.
gender. Journal of Exercise Science & Fitness, 7, European Journal of Applied Physiology, May 97(1),
S61–S73. 52–58.
12. Lepers, R., and Maffiuletti, N. 2011. Age and 4. Bixler, B. 2005. Resistance and propulsion.
gender interactions in ultra-endurance per- In J.M. Stager and D.A. Tanner (Eds.), Hand-
formance: Insight from triathlon. Medicine and book of sports medicine and science: Swimming
Science in Sports and Exercise, 43(1), 134–139. (2nd ed., pp. 59–101). Malden, MA: Blackwell
13. Reaburn, P., and Dascombe, B. 2008. Endurance Science.
performance in masters athletes. European 5. Counsilman, J.E. 1969. The role of sculling
Review of Aging and Physical Activity, 5, 31–42. movements in the arm pull. Swimming World,
14. Bernard, T., Sultana, F., Lepers, R., Hauss- December, 6, 7, 43.
wirth, C., and Brisswalter, J. 2010. Age related 6. Counsilman, J.E. 1971. The application of
decline in Olympic triathlon performance: Bernoulli’s principle to human propulsion in
Effect of locomotion mode. Experimental Aging water. In Biomechanics in swimming, waterpolo,
Research, 36(1), 64–78. and diving (pp. 59–72). Brussels, Belgium: Uni-
15. Lepers R., Sultana F., Bernard T., Hausswirth versite Libre de Bruxelles.
C., and Brisswalter, J. 2010. Age-related changes 7. Miller, D.I. 1975. Biomechanics of swimming.
in triathlon performances. International Journal Exercise and Sport Sciences Reviews, 3, 219–48.
of Sports Medicine, 31(4), 251–256. 8. Rushall, B.S., Sprigings, E.J., Holt, L.E., and
16. Lepers, R., and Stapley, P.J. 2011. Age-related Cappaert, J.M. 1994. A re-evaluation of forces
changes in conventional road versus off-road in swimming. Journal of Swimming Research,
triathlon performance. European Journal of Fall 10, 6–30.
Applied Physiology, 111(8), 1687–1694. 9. Schleihauf, R.E. 1979. A hydrodynamic analysis
17. Balmer, J., Bird, S., and Davison, R. 2008a. of swimming propulsion. International sympo-
Indoor 16.1-km time-trial performance in sium on biomechanics in swimming (pp. 70–109).
cyclists aged 25–63 years. Journal of Sports Sci- Baltimore: University Press.
ences, 26, 57–62. 10. Schleihauf, R.E., Higgins, J., Hinricks, R.,
18. Mittendorfer, B., and Klein, S. 2001 Effect of Luedtke, D., Maglischo, C., Maglischo, E., et al.
aging on glucose and lipid metabolism during 1984. Biomechanics of swimming propulsion.
endurance exercise. International Journal of American Swimming Coaches Association world
Sport Nutrition and Exercise Metabolism, 11, clinic yearbook, 19–24.
S86–91. 11. Toussaint, H.M., and Truijens, M. 2005. Bio-
19. Sial, S., Coggan, A.R., Hickner, R.C., and Klein, mechanical aspects of peak performance in
S. 1998 Training-induced alterations in fat and human swimming. Animal Biology, March 55(1),
carbohydrate metabolism during exercise in 17–40.
588 } References

12. Toussaint, H.M., Van den Berg, C., and Beek, 9. Ryan, M., and Gregor, R. 1992. EMG profiles
W.J. 2002. “Pumped-up propulsion” during of lower extremity muscles during cycling
front crawl swimming. Medicine and Science in at constant workload and cadence. Journal of
Sports and Exercise, February 34(2), 314–319. Electromyography and Kinesiology, 2(2), 69–80.
13. Maglischo, E.W. 2003. Swimming fastest. Cham- 10. Chapman, A., Vincenzino, B., Blanch, P.,
paign, IL: Human Kinetics. Dowlan, S., and Hodges, P. 2008. Does cycling
14. Costill, D.L., Kovaliski, J., Porter, D., Fielding, effect motor coordination of the leg during
R., and King, D. 1985. Energy expenditure running in elite triathletes? Journal of Science
during front crawl swimming: Predicting and Medicine in Sport, 11, 371–380.
success in middle distance events. International 11. Chapman, A., Vincenzino, B., Hodges, S.,
Journal of Sports Medicine, 6, 266–270. Dowlan, A., Alexander, M., and Milner, T.
15. Craig, A.B.J., Skehan, P.L., Pawelczyk, J.A., and 2009. Cycling impairs neuromuscular control
Boomer, W.L. 1985. Velocity, stroke rate, and during running in triathletes: Implications for
distance per stroke during elite swimming performance, injury and intervention. Journal
competition. Medicine and Science in Sports and of Science and Medicine in Sport, 12S, S1–S83,
Exercise, 17(6), 625–634. 142.
16. Toussaint, H.M., Bruinink, L., Coster, R., De 12. Bonacci, J., Green, D., Saunders, P., Blanch, P.,
Looze, M., Van Rossem, B., Van Veenen, R., et Franettovich, M., Chapman, A., and Vicenzino,
al. 1989. Effect of a triathlon wet suit on drag B. 2010. Changes in running kinematics after
during swimming. Medicine and Science in cycling are related to alterations in running
Sports and Exercise, 21(3), 325–328. economy in triathletes. Journal of Science in
Medicine in Sport, 13, 460–464.
Chapter 5 Cycling Biomechanics for 13. Sutter, K., Sayers, M., and Askew, C. 2009.
Triathlon The effects of cycling intensity on running
economy in triathletes. Journal of Science in
1. Sharp, A. 1896. Bicycles and tricycles. Longmans Medicine in Sport, 12S, S1–S83, 143.
Green (reprinted MIT Press, Cambridge, 1977).
14. Bonacci, J., Green, D., Saunders, P., Franet-
2. Broker J. 2003. Cycling biomechanics: Road and tovich, M., Blanch, P., and Vicenzino, B. 2011.
mountain. In E. Burke (Ed.), High tech cycling II Plyometric training as an intervention to
(pp. 119–146). Champaign, IL: Human Kinetics. correct altered neuromuscular control during
3. Cavanaugh, P., and Sanderson, D. 1986. The bio- running alter cycling in triathletes: A pre-
mechanics of cycling: Studies of the pedaling liminary randomized controlled trial. Physical
mechanics of elite pursuit riders. In E. Burke Therapy in Sport, 12, 15–21.
(Ed.), Science of cycling (pp. 91–122). Champaign,
IL: Human Kinetics. Chapter 6 Running Biomechanics for
4. LaFortune, M., and Cavanaugh, P. 1983. Effec- Triathlon
tiveness and efficiency during bicycle riding. 1. Anderson, T. 1996. Biomechanics and running
In Matsui and Kobayashi (Ed.), Biomechanics economy. Sports Medicine, 22(2), 76–89.
VIIB; International Series on Sport Science 4B.
Champaign, IL: Human Kinetics 2. Fletcher, G., Bartlett, R., Romanov, N., and
Fotouhi, A. 2009. Pose Method technique
5. Kautz, S., and Hull, M. 1993. A theoretical basis improves running performance without
for interpreting the force applied to the pedal economy changes. International Journal of
in cycling. Journal of Biomechanics 14, 843–855 Sports Science and Coaching, 3(3), 365–380. doi:
6. Broker J. 2003. Cycling power: Road and moun- 10.1260/174795408786238506
tain. In E. Burke (Ed.), High tech cycling II (pp. 3. Hausswirth, C., and Brisswalter, J. 2008. Strate-
147–174). Champaign, IL: Human Kinetics. gies for improving performance in long dura-
7. Chapman, A., Vicenzino, B., Blanch, P., and tion events: Olympic distance triathlon. Sports
Hodges, P. 2007. Leg muscle recruitment Medicine, 38(11), 881–891. doi: 1 [pii]
during cycling is less developed in triathletes 4. Kyröläinen, H., Belli, A., and Komi, P.V. 2001.
than cyclists despite matched cycling train- Biomechanical factors affecting running
ing loads. Experimental Brain Research, 181 (3), economy. Medicine and Science in Sports and
503–518. Exercise, 33(8), 1330–1337.
8. Munro, C., and Miller, D. 1987. Ground reaction 5. Saunders, P.U., Pyne, D.B., Telford, R.D., and
forces in running: A re-examination. Journal of Hawley, J.A. 2004. Factors affecting running
Biomechanics, 20 (2), 147–155.
References | 589

economy in trained distance runners. Sports 17. Støren, Ø., Helgerud, J., and Hoff, J. 2011. Run-
Medicine, 34(7), 465–485. doi: 3475 [pii] ning stride peak forces inversely determine
6. Williams, K.R. 2007. Biomechanical factors running economy in elite runners. Journal
contributing to marathon race success. Sports of Strength and Conditioning Research, 25(1),
Medicine, 37(4–5), 420–423. doi: 37NaN38 [pii] 117–123. doi: 10.1519/JSC.0b013e3181b62c8a
7. Lake, M.J., and Cavanagh, P.R. 1996. Six weeks 18. Bentley, D.J., Cox, G.R., Green, D., and Laursen,
of training does not change running mechan- P.B. 2008. Maximising performance in triath-
ics or improve running economy. Medicine and lon: Applied physiological and nutritional
Science in Sports and Exercise, 28(7), 860–869. aspects of elite and non-elite competitions.
Journal of Science and Medicine in Sport, 11(4),
8. Hausswirth, C., and Lehénaff, D. 2001. Physio- 407–416. doi: S1440-2440(07)00155-7 [pii]
logical demands of running during long 10.1016/j.jsams.2007.07.010
distance runs and triathlons. Sports Medicine,
31(9), 679–689. 19. Squadrone, R., and Gallozzi, C. 2009. Biome-
chanical and physiological comparison of bare-
9. Arendse, R.E., Noakes, T.D., Azevedo, L.B., foot and two shod conditions in experienced
Romanov, N., Schwellnus, M.P., and Fletcher, barefoot runners. Journal of Sports Medicine and
G. 2004. Reduced eccentric loading of the knee Physical Fitness, 49(1), 6–13.
with the pose running method. Medicine and
Science in Sports and Exercise, 36(2), 272–277. 20. Hanson, N.J., Berg, K., Deka, P., Meendering,
doi:10.1249/01.MSS.0000113684.61351.B0 J.R., and Ryan, C. 2011. Oxygen cost of running
barefoot vs. running Shod. International Journal
10. Dallam, G.M., Wilber, R.L., Jadelis, K., Fletcher, of Sports Medicine, 32(6), 401–406. doi: 10.1055/s-
G., and Romanov, N. 2005. Effect of a global 0030-1265203
alteration of running technique on kinemat-
ics and economy. Journal of Sports Science, 21. Robbins, S.E., and Hanna, A.M. 1987. Running-
23(7), 757–764. doi: W6T5U3W0436V0U52 related injury prevention through barefoot
[pii]10.1080/02640410400022003 adaptations. Medicine and Science in Sports and
Exercise, 19(2), 148–156.
11. Romanov, N., and Fletcher, G. 2007. Runners
do not push off the ground but fall forwards 22. Giuliani, J., Masini, B., Alitz, C., and Owens,
via a gravitational torque. Sports Biomechanics, B.D. 2011. Barefoot-simulating footwear
6(3), 434–452. doi:10.1080/14763140701491625 associated with metatarsal stress injury in
2 runners. Orthopedics, 34(7), e320–323. doi:
12. Divert, C., Mornieux, G., Baur, H., Mayer, F., 10.3928/01477447-20110526-25
and Belli, A. 2005. Mechanical comparison
of barefoot and shod running. International 23. Jenkins, D.W., and Cauthon, D.J. 2011. Bare-
Journal of Sports Medicine, 26(7), 593–598. doi: foot running claims and controversies: A
10.1055/s-2004-821327 review of the literature. Journal of the American
Podiatry Medical Association, 101(3), 231–246. doi:
13. Lieberman, D.E., Venkadesan, M., Werbel, 101/3/231 [pii]
W.A., Daoud, A.I., D’Andrea, S., Davis, I.S.,
Pitsiladis, Y. 2010. Foot strike patterns and 24. Egermann, M., Brocai, D., Lill, C.A., and Schmitt,
collision forces in habitually barefoot versus H. 2003. Analysis of injuries in long-distance
shod runners. Nature, 463(7280), 531–535. doi: triathletes. International Journal of Sports Medi-
nature08723 [pii]10.1038/nature08723 cine, 24(4), 271–276. doi: 10.1055/s-2003-39498
14. Williams, K.R. 1985. The relationship between 25. Neely, F.G. 1998. Biomechanical risk factors
mechanical and physiological energy esti- for exercise-related lower limb injuries. Sports
mates. Medicine and Science in Sports and Exer- Medicine, 26(6), 395–413.
cise, 17(3), 317–325. 26. Siegele, J., Horstmann, T., Bunc, V., Shifta, P.,
15. Arampatzis, A., Knicker, A., Metzler, V., Verle, S., and Niess, A. 2010. [Relation between
and Bruggeman n, G. 2000. Mechanical pelvis malposition and functional knee
power in running: A comparison of differ- pain by long distance running]. Sportverletz
ent approaches. Journal of Biomechanics, 33(4), Sportschaden, 24(3), 144–149. doi: 10.1055/s-0029-
457–463. doi: 10.1016/S0021-9290(99)00187 1245638
16. Unnithan, V.B., Dowling, J.J., Frost, G., and 27. Wen, D.Y. 2007. Risk factors for overuse injuries
Bar-Or, O. 1996. Role of cocontraction in the in runners. Current Sports Medicine Reports, 6(5),
O2 cost of walking in children with cerebral 307–313.
palsy. Medicine and Science in Sports and Exercise, 28. Hreljac, A. 2004. Impact and overuse injuries
28(12), 1498–1504. in runners. Medicine and Science in Sports
590 } References

and Exercise, 36(5), 845–849. doi: 00005768- running performance. Journal of Sports Science,
200405000-00017 [pii] 18(10), 825–833. doi: 10.1080/026404100419883
29. Kerrigan, D.C., Franz, J.R., Keenan, G.S., 40. Chapman, A.R., Vicenzino, B., Blanch, P., and
Dicharry, J., Della Croce, U., and Wilder, R.P. Hodges, P.W. 2007. Leg muscle recruitment
2009. The effect of running shoes on lower during cycling is less developed in triathletes
extremity joint torques. PM & R, 1(12), 1058– than cyclists despite matched cycling train-
1063. doi: S1934-1482(09)01367-7 [pii]10.1016/j. ing loads. Experimental Brain Research, 181(3),
pmrj.2009.09.011 503–518. doi: 10.1007/s00221-007-0949-5
30. Bentley, D.J., Millet, G.P., Vleck, V.E., and 41. Toussaint, H.M. 1990. Differences in propelling
McNaughton, L.R. 2002. Specific aspects of efficiency between competitive and triathlon
contemporary triathlon: Implications for swimmers. Medicine and Science in Sports and
physiological analysis and performance. Sports Exercise, 22(3), 409–415.
Medicine, 32(6), 345–359. doi: 320601 [pii] 42. Chapman, A.R., Vicenzino, B., Blanch, P., and
31. Millet, G.P., and Vleck, V.E. 2000. Physiological Hodges, P.W. 2008. Is running less skilled in
and biomechanical adaptations to the cycle to triathletes than runners matched for run-
run transition in Olympic triathlon: Review ning training history? Medicine and Science in
and practical recommendations for training. Sports and Exercise, 40(3), 557–565. doi: 10.1249/
British Journal of Sports Medicine, 34(5), 384–390. MSS.0b013e31815e727a
32. Bernard, T., Vercruyssen, F., Grego, F., Hauss-
wirth, C., Lepers, R., Vallier, J.M., and Briss- Chapter 7 In the Water
walter, J. 2003. Effect of cycling cadence on 1. Dallam, G.M., Jonas, S., and Miller, T.K. 2005.
subsequent 3 km running performance in Medical considerations in triathlon competi-
well trained triathletes. British Journal of Sports tion: Recommendations for triathlon organis-
Medicine, 37(2), 154–158; discussion 159. ers, competitors and coaches. Sports Medicine,
33. Guezennec, C.Y., Vallier, J.M., Bigard, A.X., 35(2), 143–161.
and Durey, A. 1996. Increase in energy cost 2. Foster, S. 2007. Open water swimming: Some basic
of running at the end of a triathlon. European principles. Retrieved from www.fluidmove-
Journal of Applied Physiology and Occupational ments.com/ss/documents/30.pdf
Physiology, 73(5), 440–445.
3. Glover, D. 2009. When “stuff” happens during
34. Hausswirth, C., Bigard, A.X., and Guezennec, the swim. Triathlon Life, 12(3), 42–43.
C.Y. 1997. Relationships between running
4. International Triathlon Union— competition
mechanics and energy cost of running at the
rules. 2011 Retrieved from www.triathlon.
end of a triathlon and a marathon. International
org/images/uploads/itusport_competition-
Journal of Sports Medicine, 18(5), 330–339. doi:
rules_20110222.pdf
10.1055/s-2007-972642
5. British Triathlon Federation. 2011. Retrieved
35. Cala, A., Veiga, S., García, A., and Navarro, E.
from www.britishtriathlon.org/events/
2009. Previous cycling does not affect running
page.php?article=383&category=/events/
efficiency during a triathlon world cup com-
&folder=rules/
petition. Journal of Sports Medicine and Physical
Fitness, 49(2), 152–158. 6. Clarys, J.P. 1979. Human morphology and
hydrodynamics. In Third International Sympo-
36. Hue, O., Valluet, A., Blonc, S., and Hertogh,
sium of Biomechanics in Swimming (pp. 3–41).
C. 2002. Effects of multicycle-run training on
Edmonton, Canada: University of Alberta, and
triathlete performance. Research Quarterly for
Baltimore: University Park Press.
Exercise and Sport, 73(3), 289–295.
7. USA Triathlon. 2011. Retrieved from www.
37. Lucía, A., Hoyos, J., and Chicharro, J.L. 2001.
usatriathlon.org/resources/for-race-directors/
Preferred pedalling cadence in professional
sanctioning/getting-started
cycling. Medicine and Science in Sports and Exer-
cise, 33(8), 1361–1366. 8. Triathlon Canada—Triathlon medical manual.
2000. Retrieved from www.triathlon.mb.ca/
38. Gottschall, J.S., and Palmer, B.M. 2002. The
admin/images/Seasons/sanction_triathlon-
acute effects of prior cycling cadence on run-
canada_medicalmanual.pdf
ning performance and kinematics. Medicine
and Science in Sports and Exercise, 34(9), 1518– 9. International Triathlon Union—uniform rules—
1522. doi: 10.1249/01.MSS.0000027712.03976.B6 age group. 2011. Retrieved from www.triathlon.
org/images/uploads/itusport_uniform-rules-
39. Garside, I., and Doran, D.A. 2000. Effects of
age-group_2010-11-29_1.pdf
bicycle frame ergonomics on triathlon 10-km
References | 591

10. International Triathlon Union—uniform rules— 22. Tomikawa, M., Shimoyama, Y., and Nomura,
juniors/ U23/elite athletes. 2011. Retrieved T. 2008. Factors related to the advantageous
from www.triathlon.org/images/uploads/ effects of wearing a wetsuit during swimming
itusport_uniform-rules-elite-junior-u23_09- at different submaximal velocity in triathletes.
11-2010-clean-vf.pdf Journal of Science and Medicine in Sport, 11(4),
11. Triathlon Australia policy documents. 2011. 417–423.
Ret rieved f rom w w w.t riat h lon.org.au/
Assets/Triathlon+Australia+Digital+Assets/ Chapter 8 On the Bike
Resources/Technical/TA+Policy+Document_ 1. Ricard, D., Hills-Meyer, P., Miller, M., and
June+2011.pdf Michael, T. 2006. The effects of bicycle frame
12. International Triathlon Union—approved national geometry on muscle activation and power
federation AG uniforms 2011. 2011 Retrieved during a Wingate anaerobic test. Journal of
from www.triathlon.org/images/uploads/ Sports Science and Medicine, 5, 25–32.
approved_catalogue_ag_2011.pdf 2. Browning, R. 1991. Lower extremity kinetics
13. International Triathlon Union—approved national during cycling in elite triathletes in aerodynamic
federation uniforms 2011. 2011. Retrieved cycling. Unpublished master’s thesis, Univer-
from www.triathlon.org/images/uploads/ sity of California at Los Angeles.
approved_catalogue_ag_2011.pdf 3. Garside, I., and Doran, D. 2000. Effects of
14. Triathlon Australia—approved unifrom design. bicycle frame ergonomics on triathlon10-km
2011. Retrieved from www.triathlon.org.au/ running performance. Journal of Sport Sciences,
Assets/Triathlon+Australia+Digital+Assets/ 18(10), 825–833.
Resources/High+Performance/Australian 4. International Triathlon Union. 2011. ITU com-
+Elite+Tri-Suit+-+revision+18+November petition rules. Author: North Vancouver, British
+2010.pdf Columbia, Canada.
15. FINA. 2011. List of FINA approved swim suits. 5. Kyle, C. 2003. Selecting cycling equipment. In
Retrieved from www.fina.org/H2O/index. E. Burke (Ed.), High tech cycling II (pp. 1–48).
php?option=com_content&view=article&id= Champaign, IL: Human Kinetics.
917&Itemid=461
6. USA Triathlon competitive rules, article V—cycling
16. Triathlon Australia competition rules. 2011. conduct. 2011. Retrieved from www.usatriath-
Ret rieved f rom w w w.t riat h lon.org.au/ lon.org/about-multisport/rulebook.aspx
Assets/Triathlon+Australia+Digital+Assets/
7. Hamley, E., and Thomas, V. 1967. Physiological
Resources/Technical/TA+Race+Competition+
and postural factors in the calibration of a bicy-
Rules.pdf
cle ergometer. Journal of Physiology, 191, 55–57.
17. USA Triathlon list of approved swimsuits. 2011.
8. LeMond, G., and Gordis, K. 1987. Greg LeMond’s
Retrieved from www.usatriathlon.org/
complete book of bicycling (pp. 118–145). New
resources/about-events/rules/approved-
York: Perigree Books.
skinsuits-speedsuits
9. Pruitt, A., and Matheny, F. 2001. Andy Pruitt’s
18. Parsons, L., and Day, S. 1986. Do wet suits
medical guide for cyclists. Chapel Hill, NC: RBR.
affect swimming speed? British Journal of Sports
Medicine, 20(3), 129. 10. Norgren, K. 2011. Validation of an on-road motion
capture system for cycling research and bicycle fit-
19. Toussaint, H.M., DeLooze, M., Van Rossem,
ting. Unpublished master’s thesis, University
B., Leijdekkers, M., and Dignum, H. 1990.
of Colorado at Colorado Springs.
Differences in propelling efficiency between
competitive and triathlon swimmers. Medicine 11. Hagberg, J., and McCole, S. 1996. Energy expendi-
and Science in Sports and Exercise, 22(3), 409–415. ture in cycling. In E. Burke (Ed.), High tech cycling
(pp. 167–184). Champaign, IL: Human Kinetics.
20. Chatard, J., and Millet, G. 1996. Effects of
wetsuit use in swimming events. Practical 12. Broker, J. 2003. Cycling power: Road and moun-
recommendations. Sports Medicine (Auckland, tain. In E. Burke (Ed.), High tech cycling II (pp.
NZ), 22(2), 70. 147–174). Champaign, IL: Human Kinetics.
21. Chatard, J.C., Senegas, X., Selles, M., Dreanot,
P., and Geyssant, A. 1995. Wet suit effect: A
Chapter 9 For the Run
comparison between comeptitive swimmers 1. Boyer, K.A., and Nigg, B.M. 2004. Muscle activ-
and triathletes. Medicine and Science in Sports ity in the leg is tuned in response to impact
and Exercise, 27(4), 580–586. force characteristics. Journal of Biomechanics,
592 } References

37(10), 1583–1588. doi: S0021929004000314 [pii] sion. Medicine and Science in Sports and Exercise,
10.1016/j.jbiomech.2004.01.002 41(5), 1138–1145. doi: 00005768-200905000-00021
2. Nigg, B.M., and Wakeling, J.M. 2001. Impact [pii]10.1249/MSS.0b013e318192ce58
forces and muscle tuning: A new paradigm. 13. Damavandi, M., Dixon, P.C., and Pearsall, D.J.
Exercise and Sport Sciences Reviews, 29(1), 37–41. 2011. Ground reaction force adaptations during
3. Dixon, S.J., Collop, A.C., and Batt, M.E. 2000. cross-slope walking and running. Human
Surface effects on ground reaction forces and Movement Science. doi: S0167-9457(11)00098-4
lower extremity kinematics in running. Medi- [pii]10.1016/j.humov.2011.06.004
cine and Science in Sports and Exercise, 32(11), 14. Fields, K.B., Sykes, J.C., Walker, K.M., and Jack-
1919–1926. son, J.C. 2010. Prevention of running injuries.
4. Kerdok, A.E., Biewener, A.A., McMahon, T.A., Current Sports Medicine Reports, 9(3), 176–182.
Weyand, P.G., and Herr, H.M. 2002. Energet- doi: 00149619-201005000-00014 [pii] 10.1249/
ics and mechanics of human running on JSR.0b013e3181de7ec5
surfaces of different stiffnesses. Journal of 15. Richards, C.E., Magin, P.J., and Callister, R.
Applied Physiology, 92(2), 469–478. doi: 10.1152/ 2009. Is your prescription of distance running
japplphysiol.01164.2000 shoes evidence-based? British Journal of Sports
5. Brughelli, M., and Cronin, J. 2008. Influence Medicine, 43(3), 159–162. doi: bjsm.2008.046680
of running velocity on vertical, leg and joint [pii] 10.1136/bjsm.2008.046680
stiffness: Modelling and recommendations for 16. Robbins, S.E., and Hanna, A.M. 1987. Running-
future research. Sports Medicine, 38(8), 647–657. related injury prevention through barefoot
doi: 3883 [pii] adaptations. Medicine and Science in Sports and
6. Bosco, C., Saggini, R., and Viru, A. 1997. The Exercise, 19(2), 148–156.
influence of different floor stiffness on mechan- 17. Warburton, M. 2001. Barefoot Running. Sport-
ical efficiency of leg extensor muscle. Ergonom- science, 5(3), online.
ics, 40(6), 670–679. doi: 10.1080/001401397187964 18. Divert, C., Mornieux, G., Freychat, P., Baly, L.,
7. McMahon, T.A., and Greene, P.R. 1979. The Mayer, F., and Belli, A. 2008. Barefoot-shod
influence of track compliance on running. running differences: Shoe or mass effect?
Journal of Biomechanics, 12(12), 893–904. doi: International Journal of Sports Medicine, 29(6),
0021-9290(79)90057-5 [pii] 512–518. doi: 10.1055/s-2007-989233
8. Zamparo, P., Perini, R., Orizio, C., Sacher, M., 19. Stefanyshyn, D.J., and Nigg, B.M. 2000. Energy
and Ferretti, G. 1992. The energy cost of walk- aspects associated with sport shoes. Sportver-
ing or running on sand. European Journal of letz Sportschaden, 14(3), 82–89. doi: 10.1055/s-
Applied Physiology and Occupational Physiology, 2000-7867
65(2), 183–187. 20. Roy, J.P., and Stefanyshyn, D.J. 2006. Shoe
9. Janakiraman, K., Shenoy, S., and Sandhu, midsole longitudinal bending stiffness and
J.S. 2011. Intravascular haemolysis during running economy, joint energy, and EMG.
prolonged running on asphalt and natural Medicine and Science in Sports and Exercise, 38(3),
grass in long and middle distance runners. 562–569. doi: 00005768-200603000-00023 [pii]
Journal of Sports Sciences, 29(12), 1287–1292. doi: 10.1249/01.mss.0000193562.22001.e8
10.1080/02640414.2011.591416 21. Nigg, B.M., Bahlsen, H.A., Luethi, S.M., and
10. Miller, B.J., Pate, R.R., and Burgess, W. 1988. Stokes, S. 1987. The influence of running velo-
Foot impact force and intravascular hemolysis city and midsole hardness on external impact
during distance running. International Journal forces in heel-toe running. Journal of Biomechan-
of Sports Medicine, 9(1), 56–60. doi: 10.1055/s- ics, 20(10), 951–959.
2007-1024979 22. Wakeling, J.M., Von Tscharner, V., Nigg, B.M.,
11. Paradisis, G.P., Bissas, A., and Cooke, C.B. 2009. and Stergiou, P. 2001. Muscle activity in the leg
Combined uphill and downhill sprint running is tuned in response to ground reaction forces.
training is more efficacious than horizontal. Journal of Applied Physiology, 91(3), 1307–1317.
International Journal of Sports Physiology and 23. McKenzie, D.C., Clement, D.B., and Taunton,
Performance, 4(2), 229–243. J.E. 1985. Running shoes, orthotics, and inju-
12. Peeling, P., Dawson, B., Goodman, C., Land- ries. Sports Medicine, 2(5), 334–347.
ers, G., Wiegerinck, E.T., Swinkels, D.W., and 24. Nigg, B.M. 2001. The role of impact forces
Trinder, D. 2009. Training surface and intensity: and foot pronation: A new paradigm. Clinical
Inflammation, hemolysis, and hepcidin expres- Journal of Sports Medicine, 11(1), 2–9.
References | 593

25. Van Meerhaeghe, T. 2006. [When and why increasing altitude in endurance athletes. Euro-
functional orthotics?]. Revue médicale de Brux- pean Journal of Applied Physiology, 96(4), 404–412.
elles, 27(4), S327–329. doi: 10.1007/s00421-005-0081-9
26. Landorf, K.B., and Keenan, A.M. 2000. Efficacy 37. Rusko, H.K., Tikkanen, H.O., and Peltonen, J.E.
of foot orthoses. What does the literature tell 2004. Altitude and endurance training. Journal
us? Journal of the American Podiatric Medical of Sports Sciences, 22(10), 928–944; discussion
Association, 90(3), 149–158. 945. doi: 10.1080/02640410400005933
27. Landorf, K.B. 2011. Foot orthoses can reduce 38. Beidleman, B.A., Muza, S.R., Fulco, C.S.,
lower limb overuse injury rate. Journal of Physio- Cymerman, A., Sawka, M.N., Lewis, S.F.,
therapy, 57(3), 193. doi: S1836-9553(11)70041-8 and Skrinar, G.S. 2008. Seven intermittent
[pii] 10.1016/S1836-9553(11)70041-8 exposures to altitude improves exercise per-
28. Nigg, B.M., Nurse, M.A., and Stefanyshyn, D.J. formance at 4300 m. Medicine and Science in
1999. Shoe inserts and orthotics for sport and Sports and Exercise, 40(1), 141–148. doi: 10.1249/
physical activities. Medicine and Science in Sports mss.0b013e31815a519b
and Exercise, 31(7 Suppl.), S421–428. 39. Saunders, P.U., Pyne, D.B., and Gore, C.J. 2009.
29. Burkett, L.N., Kohrt, W.M., and Buchbinder, Endurance training at altitude. High Altitude
R. 1985. Effects of shoes and foot orthotics Medicine and Biology, 10(2), 135–148. doi: 10.1089/
. ham.2008.1092
on VO2max and selected frontal plane knee
kinematics. Medicine and Science in Sports and 40. Stray-Gundersen, J., and Levine, B.D. 2008.
Exercise, 17(1), 158–163. Live high, train low at natural altitude.
30. Ely, M.R., Cheuvront, S.N., Roberts, W.O., Scandinavian Journal of Medicine and Science in
and Montain, S.J. 2007. Impact of weather on Sports, 18(Suppl. 1), 21–28. doi: SMS829 [pii]
marathon-running performance. Medicine and 10.1111/j.1600-0838.2008.00829.x
Science in Sports and Exercise, 39(3), 487–493. 41. Chapman, R., and Levine, B.D. 2007. Altitude
doi: 00005768-200703000-00012 [pii] 10.1249/ training for the marathon. Sports Medicine,
mss.0b013e31802d3aba 37(4–5), 392–395. doi: 37NaN31 [pii]
31. Montain, S.J., Ely, M.R., and Cheuvront, S.N. 42. Wilber, R.L. 2007. Application of altitude/
2007. Marathon performance in thermally hypoxic training by elite athletes. Medicine and
stressing conditions. Sports Medicine, 37(4–5), Science in Sports and Exercise, 39(9), 1610–1624.
320–323. doi: 37NaN12 [pii] doi: 00005768-200709000-00024 [pii] 10.1249/
32. Gosling, C.M., Gabbe, B.J., McGivern, J., mss.0b013e3180de49e6
and Forbes, A.B. 2008. The incidence of 43. Millet, G.P., Roels, B., Schmitt, L., Woorons,
heat casualties in sprint triathlon: The tale X., and Richalet, J.P. 2010. Combining hypoxic
of two Melbourne race events. Journal of methods for peak performance. Sports Medi-
Science and Medicine in Sport, 11(1), 52–57. cine, 40(1), 1–25. doi: 1 [pii] 10.2165/11317920-
doi: S1440-2440(07)00196-X [pii] 10.1016/j. 000000000-00000
jsams.2007.08.010 44. Kargarfard, M., Poursafa, P., Rezanejad, S., and
33. Dallam, G.M., Jonas, S., and Miller, T.K. 2005. Mousavinasab, F. 2011. Effects of exercise in pol-
Medical considerations in triathlon competi- luted air on the aerobic power, serum lactate level
tion: Recommendations for triathlon organis- and cell blood count of active individuals. Inter-
ers, competitors and coaches. Sports Medicine, national Journal of Preventive Medicine, 2(3), 145–150.
35(2), 143–161. doi: 3524 [pii] 45. Pierson, W.E., Covert, D.S., Koenig, J.Q., Name-
34. Buskirk, E.R., Iampietro, P.F., and Bass, D.E. kata, T., and Kim, Y.S. 1986. Implications of
2000. Work performance after dehydration: air pollution effects on athletic performance.
Effects of physical conditioning and heat accli- Medicine and Science in Sports and Exercise, 18(3),
matization. 1958. Wilderness and Environmental 322–327.
Medicine, 11(3), 204–208. 46. Sacha, J.J., and Quinn, J.M. 2011. The envi-
35. Garrett, A.T., Rehrer, N.J., and Patterson, M.J. ronment, the airway, and the athlete. Annals
2011. Induction and decay of short-term heat of Allergy, Asthma, and Immunology, 106(2),
acclimation in moderately and highly trained 81–87; quiz 88. doi: S1081-1206(10)00601-0 [pii]
athletes. Sports Medicine, 41(9), 757–771. doi: 5 10.1016/j.anai.2010.06.004
[pii] 10.2165/11587320-000000000-00000 47. Poussel, M., and Chenuel, B. 2010. [Exercise-
36. Wehrlin, J.P.,. and Hallén, J. 2006. Linear induced bronchoconstriction in non-asthmatic
decrease in VO2max and performance with athletes]. Revue des maladies respiratoires,
594 } References

27(8), 898–906. doi: S0761-8425(10)00370-0 and performance time in trained cyclists.


[pii]10.1016/j.rmr.2010.08.004 European Journal of Applied Physiology and Occu-
48. Campbell, M.E., Li, Q., Gingrich, S.E., Macfar- pational Physiology, 65(1), 79–83.
lane, R.G., and Cheng, S. 2005. Should people be 8. Padilla, S., Mujika, I., Angulo, F., and Goiriena,
physically active outdoors on smog alert days? J.J. 2000. Scientific approach to the 1-h cycling
Canadian Journal of Public Health, 96(1), 24–28. world record: A case study. Journal of Applied
49. Daigle, C.C., Chalupa, D.C., Gibb, F.R., Morrow, Physiology, Oct. 89(4), 1522–1527. Erratum in
P.E., Oberdörster, G., Utell, M.J., and Frampton, Journal of Applied Physiology, Dec. 89(6).
M.W. 2003. Ultrafine particle deposition in 9. Van Schuylenbergh, R., Eynde, B.V., and Hespel,
humans during rest and exercise. Inhalation P. 2004. Prediction of sprint triathlon perfor-
Toxicology, 15(6), 539–552. doi: 00F0G36CKYU- mance from laboratory tests. European Journal
W1B7B [pii] 10.1080/08958370304468 of Applied Physiology, Jan. 91(1), 94–99.
10. Laursen, P.B., and Rhodes, E.C. 2001. Factors
Chapter 10 Triathlon Training affecting performance in an ultraendurance
Technologies triathlon. Sports Medicine, 31(3), 195–209.
1. Friel, J. 2009. The triathlete’s training bible (3rd 11. Laursen, P.B., Knez, W.L., Shing, C.M., Langill,
ed., pp. 54–55). Boulder, CO: VeloPress. R.H., Rhodes, E.C., and Jenkins, D.G. 2005.
2. Friel, J. 2009. The triathlete’s training bible (3rd Relationship between laboratory-measured
ed., p. 49). Boulder, CO: VeloPress. variables and heart rate during an ultra-
endurance triathlon. Journal of Sports Sciences,
3. Coggan, A., and Allen, H. 2009. Training and Oct 23(10), 1111–1120.
racing with a power meter (2nd ed., p. 48). Boul-
der, CO: VeloPress. 12. Hue, O., Le Gallais, D., Chollet, D., and Préfaut,
C. 2000. Ventilatory threshold and maximal
4. McGregor, S., and Fitzgerald, M. 2010. The run- oxygen uptake in present triathletes. Canadian
ner’s edge. Champaign, IL: Human Kinetics. Journal of Applied Physiology, Apr. 25(2), 102–113.

Chapter 11 Aerobic Capacity 13. Bunc, V., Heller, J., Horcic, J., and Novotny,
J. 1996. Physiological profile of best Czech
1. Hill, A.V. 1925. Muscular activity (pp. 1–115). male and female young triathletes. Journal of
London: Bailliere Tindall and Cox. Sports Medicine and Physical Fitness, Dec. 36(4),
2. Taylor, H.L., Buskirk, E., and Henschel, A. 265–270.
1955. Maximal oxygen uptake as an objective 14. Galy, O., Manetta, J., Coste, O., Maimoun,
measure of cardio-respiratory performance. L., Chamari, K., and Hue, O. 2003. Maximal
Journal of Applied Physiology, 8, 73–80. oxygen uptake and power of lower limbs
3. Day, J.R., Rossiter, H.B., Coats, during a competitive season in triathletes.
. E.M., et al. 2003.
The maximally attainable VO2 during exercise Scandinavian Journal of Medicine and Science in
in humans: The peak vs. maximum issue. Sports, June 13(3), 185–193.
Journal of Applied Physiology, 95, 1901–1907. 15. Lucía, A., Hoyos, J., Pérez, M., Santalla, A.,
4. Lucia, A., Rabadan, and Chicharro,
. J.L. 2002. Inverse relationship
. M., Hoyos, J., et al. 2006.
Frequency of the VO2max plateau phenomenon between VO2max and economy/efficiency in
in world-class cyclists. International Journal of world-class cyclists. Medicine and Science in
Sports Medicine, 27, 984–992. Sports and Exercise, Dec. 34(12), 2079–2084.
5. Doherty, M., Nobbs, L., and Noakes, T.D. 2003. 16. Coyle, E.F. 2005. Improved muscular effi-
Low frequency of the ‘‘plateau phenomenon’’ ciency displayed as Tour de France champion
during maximal exercise in elite British ath- matures. Journal of Applied Physiology, June
letes. European Journal of Applied Physiology, 89, 98(6), 2191–2196. Epub 2005 Mar 17.
619–623. 17. Bouchard, C., Sarzynski, M.A., Rice, T.K., et
6. Noakes, T.D., St. Clair Gibson, A., and Lambert, al. 2011. Genomic predictors of the maximal
E.V. 2005. From catastrophe to complexity: O2 uptake response to standardized exercise
A novel model of integrative central neural training programs. Journal of Applied Physiology,
regulation of effort and fatigue during exercise 110, 1160–1170.
in humans: Summary and conclusions. British 18. Bouchard, C., Leon, A.S., Rao, D.C., et al. 1995.
Journal of Sports Medicine, Feb. 39(2), 120–124. The HERITAGE Family Study. Aims, design,
7. Hawley, J.A., and Noakes, T.D. 1992. Peak and measurement protocol. Medicine and Sci-
power output predicts maximal oxygen uptake ence in Sports and Exercise, 27, 721–729.
References | 595

19. Morss, G.M., Jordan, A.N., Skinner, J.S., et al. 7. Millet, G.P., Millet, G.Y., Hofmann, M.D., and
2004. Dose Response to Exercise in Women Candau, R.B. 2000. Alterations in running econ-
aged 45–75 yr (DREW): Design and rationale. omy and mechanics after maximal cycling in
Medicine and Science in Sports and Exercise, 36, triathletes: Influence of performance level. Inter-
336–344. national Journal of Sports Medicine, 21, 127–132.
20. Kraus, W.E., Torgan, C.E., Duscha, B.D., et al. 8. Kjendlie, P.L., Ingjer, F., Stallman, R.K., and
2001. Studies of a targeted risk reduction inter- Stray-Gundersen, J. 2004. Factors affect-
vention through defined exercise (STRRIDE). ing swimming economy in children and
Medicine and Science in Sports and Exercise, 33, adults. European Journal of Applied Physiology,
1774–1784. Oct., 93(1–2), 65–74. Epub 2004 Jul 8.
21. Helgerud, J., Høydal, K., Wang, E., Karlsen, T., 9. Hopker, J., Passfield, L., Coleman, D., Jobson,
Berg, P., Bjerkaas, M., Simonsen, T., Helgesen, S., Edwards, L., and Carter, H. 2009. The effects
C., Hjorth, N., Bach, R., and Hoff, J. 2007.
. Aero- of training on gross efficiency in cycling: A
bic high-intensity intervals improve VO2max review. International Journal of Sports Medicine,
more than moderate training. Medicine and Sci- Dec., 30(12), 845–850.
ence in Sports and Exercise, Apr. 39(4), 665–671. 10. Sacchetti, M., Lenti, M., Di Palumbo, A.S., and
22. Gibala, M.J., Little, J.P., van Essen, M., Wilkin, De Vito, G. 2010 Different effect of cadence on
G.P., Burgomaster, K.A., Safdar, A., Raha, S., cycling efficiency between young and older
and Tarnopolsky, M.A. 2006. Short-term sprint cyclists. Medicine and Science in Sports and Exer-
interval versus traditional endurance training: cise, 42, 2128–2133.
Similar initial adaptations in human skeletal 11. Louis, J., Hausswirth, C., Easthope, C., Briss-
muscle and exercise performance. Journal of walter, J. 2011. Strength training improves
Physiology, Sep. 15, 575(Pt. 3), 901–911. cycling efficiency in master endurance athletes.
European Journal of Applied Physiology, June 3.
Chapter 12 Economy [Epub ahead of print]
1. Foster, C., and Lucia, A. 2007. Running econ- 12. Hopker, J.G., Coleman, D.A., and Wiles, J.D.
omy: The forgotten factor in elite performance. 2007. Differences in efficiency between trained
Sports Medicine, 37(4–5), 316–319. and recreational cyclists. Applied Physiology,
2. Capelli, C., Pendergast, D.R., and Termin, B. Nutrition and Metabolism, 32(6), 1036–1042.
1998. Energetics of swimming at maximal 13. Hopker, J.G., Coleman, D.C., and Passfield, L.
speeds in humans. European Journal of Applied 2009. Changes in cycling efficiency during a
Physiology and Occupational Physiology, Oct., competitive season. Medicine and Science in
78(5), 385–393. Sports and Exercise, 41(4), 912–919.
3. Lucía, A., Hoyos, J., Pérez, M., Santalla, A., 14. Hopker, J., Coleman, D., Passfield, L., and Wiles
and Chicharro,
. J.L. 2002. Inverse relationship J. 2010. The effect of training volume and inten-
between VO2max and economy/efficiency in sity on competitive cyclists’ efficiency. Applied
world-class cyclists. Medicine and Science in Physiology, Nutrition and Metabolism, Feb., 35(1),
Sports and Exercise, Dec., 34(12), 2079–2084. 17–22.
4. Noakes, T.D., and Tucker,
. R. 2004. Inverse 15. Lu´cıa, A., Pardo, J., Durantez, A., Hoyos, J., and
relationship between VO2max and economy Chicharro, J.L. 1998. Physiological differences
in world-class cyclists. Medicine and Science between professional and elite road cyclists. Inter-
in Sports and Exercise, June, 36(6), 1083–1084; national Journal of Sports Medicine, 19(5), 342–348.
author reply 1085–1086. 16. Jeukendrup, A.E., Craig, N.P., and Hawley, J.A.
5. Hausswirth, C., Bigard, A.X., and Guezennec, 2000. The bioenergetics of World Class Cycling.
C.Y. 1997. Relationships between running Journal of Science and Medicine in Sport, Dec., 3(4),
mechanics and energy cost of running at the 414–433.
end of a triathlon and a marathon. International 17. Bonacci, J., Chapman, A., Blanch, P., and Vicen-
Journal of Sports Medicine, 18, 330–339. zino, B. 2009. Neuromuscular adaptations to
6. Hue, O., Gallais, D.L., Chollet, D., Boussana, training, injury and passive interventions:
A., and Prefaut, C. 1997. The influence of Implications for running economy. Sports
prior cycling on biomechanical and cardio- Medicine, 39(11), 903–921.
respiratory response profiles during running 18. Paavolainen, L., Häkkinen, K., Hämäläinen,
in triathletes. European Journal of Applied Physi- I., Nummela, A., Rusko, H. 1999. Explosive-
ology and Occupational Physiology, 77, 98–105. strength training improves 5-km running time
596 } References

by improving running economy and muscle 2. Solberg, G., Robstad, B., Skjønsberg, O.H., and
power. Journal of Applied Physiology, May, 86(5), Borchsenius, F. 2005. Respiratory gas exchange
1527–1533. indices for estimating the anaerobic threshold.
19. Nummela, A.T., Paavolainen, L.M., Sharwood, Journal of Sports Science and Medicine, 4, 29–36.
K.A., Lambert, M.I., Noakes, T.D., and Rusko, 3. Anderson, G.S., and Rhodes, E.C. 1989. A
H.K. 2006. Neuromuscular factors determin- review of blood lactate and ventilatory meth-
ing 5 km running performance and running ods of detecting transition thresholds. Sports
economy in well-trained athletes. European Medicine, 8, 3–55
Journal of Applied Physiology, May, 97(1), 1–8. 4. Quinn, T.J., and Coons, B.A. 2011. The Talk
20. Gleim, G.W., Stachenfeld, N.S., and Nicholas, Test and its relationship with the ventilatory
J.A. 1990. The influence of flexibility on the and lactate thresholds. Journal of Sports Science,
economy of walking and jogging. Journal of Aug., 29(11), 1175–1182.
Orthopaedic Research, Nov., 8(6), 814–823. 5. Gaskill, S.E., Ruby, B.C., Walker, A.J., Sanchez,
21. Craib, M.W., Mitchell, V.A., Fields, K.B., Cooper, O.A., Serfass, R.C., and Leon, A.S. 2001. Valid-
T.R., Hopewell, R., and Morgan, D.W. 1996. The ity and reliability of combining three methods
association between flexibility and running to determine ventilatory threshold. Medicine
economy in sub-elite male distance runners. and Science in Sports and Exercise, Nov., 33(11),
Medicine and Science in Sports and Exercise, June, 1841–1848.
28(6), 737–743. 6. Brooks, G.A. 2007. Lactate: Link between glyco-
22. Godges, J.J., MacRae, P.G., and Engelke, K.A. lytic and oxidative metabolism. Sports Medicine,
1993. Effects of exercise on hip range of motion, 37(4–5), 341–343.
trunk muscle performance, and gait economy. 7. Brooks, G.A. 2009. Cell-cell and intracellular
Physical Therapy, 73, 468–477. lactate shuttles. Journal of Physiology, Dec. 1,
23. Nelson, A.G., Kokkonen, J., Eldredge, C. et 587(Pt. 23), 5591–5600.
al. 2001. Chronic stretching and running 8. Bernard, T., Hausswirth, C., Le Meur, Y., Bignet,
economy. Scandinavian Journal of Medicine and F., Dorel, S., and Brisswalter, J. 2009. Distribu-
Science in Sports, 11, 260–265. tion of power output during the cycling stage
24. Godges, J.J., MacRae, H., Longdon, C. et al. 1989. of a Triathlon World Cup. Medicine and Science
The effects of two stretching procedures on in Sports and Exercise, June, 41(6), 1296–1302.
hip range of motion and gait economy. Journal 9. Laursen, P.B., Rhodes, E.C., Langill, R.H.,
of Orthopaedic and Sports Physical Therapy, 10, McKenzie, D.C., and Taunton, J.E. 2002. Rela-
350–357. tionship of exercise test variables to cycling
25. Chapman, A., Vicenzino, B., Blanch, P. et al. performance in an Ironman triathlon. European
2007. Leg muscle recruitment during cycling Journal of Applied Physiology, Aug., 87(4–5),
is less developed in triathletes than cyclists 433–440.
despite matched cycling training loads. Experi- 10. Hopker, J., Coleman, D., Passfield, L., and Wiles,
mental Brain Research, 181, 503–518. J. 2010. The effect of training volume and inten-
26. Bonacci, J., Green, D., Saunders, P.U., Blanch, sity on competitive cyclists’ efficiency. Applied
P., Franettovich, M., Chapman, A.R., et al. 2010. Physiology, Nutrition and Metabolism, Feb., 35(1),
Change in running kinematics after cycling 17–22.
are related to alterations in running economy 11. Gibala, M.J., Little, J.P., van Essen, M., Wilkin,
in triathletes. Journal of Science and Medicine in G.P., Burgomaster, K.A., Safdar, A., Raha, S.,
Sport, 13, 460–464. and Tarnopolsky, M.A. 2006. Short-term sprint
27. Millet, G.P., and Vleck, V.E. 2000. Physiological interval versus traditional endurance training:
and biomechanical adaptations to the cycle to Similar initial adaptations in human skeletal
run transition in Olympic triathlon: Review muscle and exercise performance. Journal of
and practical recommendations for training. Physiology, Sep. 15, 575(Pt. 3), 901–911.
British Journal of Sports Medicine, 34, 384–390. 12. Van Schuylenbergh, R., Eynde, B.V., and Hespel,
P. 2004. Prediction of sprint triathlon perfor-
Chapter 13 Anaerobic Threshold mance from laboratory tests. European Journal
1. Wasserman, K., and McIlroy, M.B. 1964. Detect- of Applied Physiology, Jan., 91(1), 94–99. Epub
ing the threshold of anaerobic metabolism in 2003 Sep 4.
cardiac patients during exercise. American 13. Schabort, E.J., Killian, S.C., St. Clair Gibson, A.,
Journal of Physiology, 14, 844–852. Hawley, J.A., and Noakes, T.D. 2000. Prediction
References | 597

of triathlon race time from laboratory testing of elite distance runners. Part II: Muscle fiber
in national triathletes. Medicine and Science in composition and enzyme activities. Annals of
Sports and Exercise, Apr., 32(4), 844–849. the New York Academy of Sciences, 301, 323–327.
14. Hue, O. 2003. Prediction of drafted-triathlon 10. Costill, D.L., Daniels, J., Evans, W., Fink, W.J.,
race time from submaximal laboratory testing Krahenbuhl, G., and Saltin, B. 1976. Skeletal
in elite triathletes. Canadian Journal of Applied muscle enzymes and fiber composition in male
Physiology, Aug., 28(4), 547–560. and female track athletes. Journal of Applied
15. Swart, J., and Jennings, C.L. 2004. Use of blood Physiology, 40, 149–154.
lactate concentration as a marker of training 11. Parcell, A.C., Sawyer, R.D., and Craig Poole, R.
status: Review article. South African Journal of 2003. Single muscle fiber myosin heavy chain
Sports Medicine, 16, 1–5. distribution in elite female track athletes.
16. Pyne, D.B., Lee, H., and Swanwick, K.M. 2001. Medicine and Science in Sports and Exercise, Mar.,
Monitoring the lactate threshold in world- 35(3), 434–438.
ranked swimmers. Medicine and Science in 12. Saltin, B., Henriksson, J., Nygaard, E., and
Sports and Exercise, Feb., 33(2), 291–297. Andersen, P. 1977. Fiber types and metabolic
potentials of skeletal muscles in sedentary man
Chapter 14 Muscle Types and and endurance runners. Annals of the New York
Triathlon Performance Academy of Sciences, 301, 3–44.

1. Andersen, J.L., Schjerling, P., and Saltin, B. 13. Gollnick, P.D., and Matoba, H. 1984. The
2000. Muscle, genes and athletic performance. muscle fiber composition of skeletal muscle
Scientific American, 283, 48–55. as a predictor of athletic success. An overview.
American Journal of Sports Medicine, May–June,
2. Andersen, J.L., and Aagard, P. 2010. Effects 12(3), 212–217.
of strength training on muscle fiber types
and size; consequences for athletes training 14. Holloszy, J.O., and Coyle, E.F. 1984. Adaptations
for high-intensity sport. Scandinavian Journal of skeletal muscle to endurance exercise and
of Medicine and Science in Sports, 20(Suppl. 2), their metabolic consequences. Journal of Applied
32–38. Physiology, Apr., 56(4), 831–838.

3. Zierath, J.R., and Hawley, J.A. 2004. Skeletal 15. Luden, N., Hayes, E., Minchev, K., Louis, E.,
muscle fiber type: Influence on contractile and Raue, U., Conley, T., and Trappe, S. 2011. Skel-
metabolic properties. PLoS Biology, Oct. 2(10), etal muscle plasticity with marathon training
e348. in novice runners. Scandinavian Journal of
Medicine and Science in Sports, Apr 8, 662–670.
4. Harridge, S.D. 1996. The muscle contractile
system and its adaptation to training. In P. Mar- 16. Aagaard, P., Andersen, J.L., Bennekou, M., Lars-
connet, B. Saltin, P.V. Komi, and J. Poortmans son, B., Olesen, J.L., Crameri, R., Magnusson,
(Eds.), Human muscular function during dynamic S.P., and Kjaer, M. 2011. Effects of resistance
exercise (pp. 82–94). Basel, Switzerland: Karger. training on endurance capacity and muscle
fiber composition in young top-level cyclists.
5. Bottinelli, R. 2001. Functional heterogeneity of Scandinavian Journal of Medicine and Science in
mammalian single muscle fibres: Do myosin Sports, Mar. 1. e298–307.
isoforms tell the whole story? Pflugers Archiv,
443, 6–17. 17. Wang, Y.X., Zhang, C.L., Yu, R.T., Cho, H.K., and
Nelson, M.C., et al. 2004. Regulation of muscle
6. Lionikas, A., Li, M., and Larsson, L. 2006. fiber type and running endurance by PPAR\
Human skeletal muscle myosin function at gd\. PLoS Biology, 2, e294.
physiological and non-physiological tempera-
tures. Acta Physiologica, 186, 151–158. 18. Gollnick, P.D., Armstrong, R.B., Saltin, B.,
Saubert, C.W. 4th, Sembrowich, W.L., and Shep-
7. Simoneau, J.A., and Bouchard, C. 1989. Human herd, R.E. 1973. Effect of training on enzyme
variation in skeletal muscle fiber-type propor- activity and fiber composition of human skel-
tion and enzyme activities. American Journal of etal muscle. Journal of Applied Physiology, Jan.,
Physiology, Oct. 257(4 Pt. 1), E567–572. 34(1), 107–111.
8. Tesch, P.A., and Karlsson, J. 1985. Muscle 19. Simoneau, J.A., Lortie, G., Boulay, M.R.,
fiber types and size in trained and untrained Marcotte, M., Thibault, M.C., and Bouchard,
muscles of elite athletes. Journal of Applied C. 1985. Human skeletal muscle fiber type
Physiology, Dec., 59(6), 1716–1720. alteration with high-intensity intermittent
9. Fink, W.J., Costill, D.L., and Pollock, M.L. 1977. training. European Journal of Applied Physiology
Submaximal and maximal working capacity and Occupational Physiology, 54(3), 250–253.
598 } References

20. Helgerud, J., Høydal, K., Wang, E., Karlsen, T., 7. Tucker, R. 2009. The anticipatory regulation
Berg, P., Bjerkaas, M., Simonsen, T., Helgesen, of performance: The physiological basis for
C., Hjorth, N., Bach, R., and Hoff, J. 2007.
. Aero- pacing strategies and the development of a
bic high-intensity intervals improve VO2max perception-based model for exercise perfor-
more than moderate training. Medicine and mance. British Journal of Sports Medicine, June,
Science in Sports and Exercise, Apr., 39(4), 665– 43(6), 392–400. Epub 2009 Feb 17.
671. 8. Rowell, L.B., Marx, H.J., Bruce, R.A., et al. 1966.
21. Burgomaster, K.A., Howarth, K.R., Phillips, Reductions in cardiac output, central blood
S.M., Rakobowchuk, M., Macdonald, M.J., volume, and stroke volume with thermal stress
McGee, S.L., and Gibala, M.J. 2008. Similar in normal men during exercise. Journal of Clini-
metabolic adaptations during exercise after cal Investigation , 45, 1801–1816.
low volume sprint interval and traditional 9. Nadel, E.R., Cafarelli, E., Roberts, M.F., and
endurance training in humans. Journal of Physi- Wenger, C.B. 1979. Circulatory regulation
ology, Jan., 1;586(1), 151–60. Epub 2007 Nov 8. during exercise in different ambient tempera-
22. Gibala, M.J., Little, J.P., van Essen, M., Wilkin, tures. Journal of Applied Physiology, Mar., 46(3),
G.P., Burgomaster, K.A., Safdar, A., Raha, S., 430–437.
and Tarnopolsky, M.A. 2006. Short-term sprint 10. Savard, G.K., Nielsen, B., Laszczynska, J.,
interval versus traditional endurance training: Larsen, B.E., and Saltin, B. 1988. Muscle blood
Similar initial adaptations in human skeletal flow is not reduced in humans during moder-
muscle and exercise performance. Journal of ate exercise and heat stress. Journal of Applied
Physiology, Sep. 15, 575(Pt. 3), 901–911. Physiology, Feb., 64(2), 649–657.

Chapter 15 Fatigue Resistance and 11. Cheung, S.S., and Sleivert, G.G. 2004. Multiple
triggers for hyperthermic fatigue and exhaus-
Recovery tion. Exercise and Sport Sciences Review, 32,
1. Nybo, L., and Nielsen, B. 2001. Hyperthermia 100–106.
and central fatigue during prolonged exercise 12. Tatterson, A.J., Hahn, A.G., Martin, D.T., and
in humans. Journal of Applied Physiology, Sep., Febbraio, M.A. 2000. Effects of heat stress on
91(3), 1055–1060. physiological responses and exercise perfor-
2. Nybo, L., and Nielsen, B. 2001. Perceived exer- mance in elite cyclists. Journal of Science and
tion is associated with an altered brain activity Medicine in Sport, June, 3(2), 186–193.
during exercise with progressive hyperther- 13. Kay, D., Marino, F.E., Cannon, J., et al. 2001.
mia. Journal of Applied Physiology, Nov., 91(5), Evidence for neuromuscular fatigue during
2017–2023. high-intensity cycling in warm, humid condi-
3. Drust, B., Rasmussen, P., Mohr, M., Nielsen, tions. European Journal of Applied Physiology, 84,
B., and Nybo, L. 2005. Elevations in core and 115–121.
muscle temperature impairs repeated sprint 14. Marino, F.E., Lambert, M.I., and Noakes, T.D.
performance. Acta Physiologica Scandinavica, 2004. Superior performance of African runners
Feb., 183(2), 181–190. in warm humid but not in cool environmental
4. Tucker, R., and Noakes, T.D. 2009. The physio- conditions. Journal of Applied Physiology, 96,
logical regulation of pacing strategy during 124–130.
exercise: A critical review. British Journal of 15. Tucker, R., Rauch, L., Harley, Y.X., and Noakes,
Sports Medicine, June, 43(6), e1. Epub 2009 Feb T.D. 2004. Impaired exercise performance in
17. the heat is associated with an anticipatory
5. Lambert, E.V., St. Clair Gibson, A., and Noakes, reduction in skeletal muscle recruitment.
T.D. 2005. Complex systems model of fatigue: Pflugers Archiv, July, 448(4), 422–430.
Integrative homoeostatic control of periph- 16. Rauch, H.G., St. Clair Gibson, A., Lambert,
eral physiological systems during exercise in E.V., and Noakes, T.D. 2005. A signalling role
humans. British Journal of Sports Medicine, Jan., for muscle glycogen in the regulation of pace
39(1), 52–62. during prolonged exercise. British Journal of
6. Noakes, T.D., St. Clair Gibson, A., and Lambert, Sports Medicine, Jan., 39(1), 34–38.
E.V. 2005. From catastrophe to complexity: 17. Noakes, T.D. 2000. Physiological models to
A novel model of integrative central neural understand exercise fatigue and the adapta-
regulation of effort and fatigue during exercise tions that predict or enhance athletic perfor-
in humans: Summary and conclusions. British mance. Scandinavian Journal of Medicine and
Journal of Sports Medicine, Feb., 39(2), 120–124. Science in Sports, June, 10(3), 123–145.
References | 599

18. Noakes, T.D. 2004. Linear relationship between 6. Bruyn-Prevost, P. 1980. The effects of various
the perception of effort and the duration of warming up intensities and durations upon
constant load exercise that remains. Journal of some physiological variables during an exer-
Applied Physiology, Apr., 96(4), 1571–1572; author cise corresponding to the WC170. European
reply 1572–1573. Journal of Applied Physiology and Occupational
19. Amann, M., Eldridge, M.W., Lovering, A.T., Physiology, 43(2), 93–100.
Stickland, M.K., Pegelow, D.F., and Dempsey, 7. Proske, U., Morgan, D., and Gregory, J. 1993.
J.A. 2006. Arterial oxygenation influences Thixotropy in skeletal muscle and in muscle
central motor output and exercise perfor- spindles: A review. Progress in Neurobiology,
mance via effects on peripheral locomotor 41(6), 705–721.
muscle fatigue in humans. Journal of Physiol- 8. Orlick, T., and Partington, J. 1987. The sport
ogy, Sep. 15, 575(Pt. 3), 937–952. Epub 2006 psychology consultant: Analysis of critical
Jun 22. components as viewed by Canadian Olympic
20. Amann, M., Proctor, L.T., Sebranek, J.J., Pegelow, athletes. Sport Psychologist, 2, 105–130.
D.F., and Dempsey, J.A. 2009. Opioid-mediated 9. Law, R.Y., and Herbert, R.D. 2007. Warm-up
muscle afferents inhibit central motor drive reduces delayed-onset muscle soreness but
and limit peripheral muscle fatigue develop- cool-down does not: A randomised controlled
ment in humans. Journal of Physiology, Jan. 15, trial. Australian Journal of Physiotherapy, 53, 91–95.
587(Pt. 1), 271–283. Epub 2008 Nov 17.
10. Murphy, J.R., Di Santo, M.C., Alkanani, T.,
21. Amann, M., and Dempsey, J.A. 2008. The con- and Behm, D.G. 2010. Aerobic activity before
cept of peripheral locomotor muscle fatigue as and following short-duration static stretching
a regulated variable. Journal of Physiology, Feb. improves range of motion and performance
14. [Epub ahead of print] vs. a traditional warm-up. Applied Physiology,
22. Nielsen, B., and Davies, C.T. 1976. Tempera- Nutrition, and Metabolism, 35(5), 679–690.
ture regulation during exercise in water and 11. Stewart, I.B., and Sleivert, G.G. 1998. The effect
air. Acta Physiologica Scandinavica, Dec., 98(4), of warm-up intensity on range of motion and
500–508. anaerobic performance. Journal of Orthopaedic
23. St. Clair Gibson, A., Baden, D.A., Lambert, M.I., and Sports Physical Therapy, 27(2), 154–161.
Lambert, E.V., Harley, Y.X., Hampson, D., Rus- 12. Gregson, W.A., Drust, B., Batterham, A., and
sell, V.A., and Noakes, T.D. 2003. The conscious Cable, N.T. 2002. The effects of pre-warming on
perception of the sensation of fatigue. Sports the metabolic and thermoregulatory responses
Medicine, 33(3), 167–176. to prolonged submaximal exercise in moder-
ate ambient temperatures. European Journal of
Chapter 16 Warm-Up and Cool-Down Applied Physiology, 86(6), 526–533.
1. Wright, V., and Johns, R.J. 1961. Quantitative 13. Gonzalez-Alonso, J., Teller, C., Andersen, S.L.,
and qualitative analysis of joint stiffness in Jensen, F.B., Hyldig, T., and Nielsesn, B. 1999.
normal subjects and in patients with connec- Influence of body temperature on the devel-
tive tissue diseases. Annals of the Rheumatic opment of fatigue during prolonged exercise
Diseases, 20, 36–46. in the heat. Journal of Applied Physiology, 85(3),
2. Buchthal, F., Kaiser, E., and Knappeis, G.G. 1032–1039.
1944. Elasticity, viscosity and plasticity in the 14. Bishop, D., Bonetti, D., and Dawson, B. 2001.
cross striated muscle fibre. Acta Physiologica The effect of three different warm-up inten-
Scandinavica, 8(1), 16–37. sities on kayak ergometer performance.
3. Brechue, W.F., Ameredes, B.T., Barclay, J.K., Medicine and Science in Sports and Exercise, 33(6),
and Stainsby, W.N. 1995. Blood flow and . pres- 1026–1032.
sure relationships which determine VO2max. 15. Saltin, B., Gagge, A.P., and Stolwijk, J.A. 1968.
Medicine and Science in Sports and Exercise, 27(1), Muscle temperature during submaximal exer-
37–42 cise in man. Journal of Applied Physiology, 25(6),
4. Barcroft, J., and King, W.O. 1909. The effect of 679–688.
temperature on the dissociation curve of blood. 16. Ozyener, F., Rossiter, H.R., Ward, S.A., and
Journal of Physiology, 39(5), 374–384. Whipp, B.J. 2001. Influence of exercise intensity
5. Andzel, W.D. 1982. One mile run performance on the on- and off-transient kinetics of
as a function of prior exercise. Journal of Sports pulmonary oxygen uptake in humans. Journal
Medicine and Physical Fitness, 22(1), 80–84. of Physiology, 533, 891–902.
600 } References

17. Tomaras, E.K., and MacIntosh, B.R. 2011. Less and Maresh, C. 2009. Factors affecting flutter
is more: Standard warm-up causes fatigue and kicking speed in women who are competitive
less warm-up permits greater cycling power and recreational swimmers. Journal of Strength
output. Journal of Applied Physiology, 111(1), and Conditioning Research, 23(7), 2130–2136.
228–235. 3. Hahn, T., Foldspang, A., Vestergaard, E., and
18. Burnley, M., Doust, J.H., and Jones, A.M. 2006. Ingemann-Hansen, T. 1999. Active knee joint
Time required for . the restoration of normal flexibility and sports activity. Scandinavian
heavy exercise VO2 kinetics following prior Journal of Medicine and Science in Sports, 9(2),
heavy exercise. Journal of Applied Physiology, 74–80.
101(5), 1320–1327. 4. Jones, A.M. 2002. Running economy is nega-
19. Kenny, G.P., Reardon, F.D., Zaleski, W., Rear- tively related to sit-and-reach test performance
don, M.L., Haman, F., and Ducharme, M.B. in international-standard distance runners.
2003. Muscle temperature transients before, International Journal of Sports Medicine, 23(1),
during, and after exercise measured using an 40–43.
intramuscular multisensor probe. Journal of 5. Ibid., 40.
Applied Physiology, 94, 2350–2357.
6. Treheam, T.L., and Buresh, R.J. 2009. Sit-and-
20. Ahmaidi, S., Granier, P., Taoutaou, Z., Mer- reach flexibility and running economy of men
cier, J., Dubouchaud, H., and Prefaut, C. 1996. and women collegiate distance runners. Journal
Effects of active recovery on plasma lactate and of Strength and Conditioning Research, 23(1),
anaerobic power following repeated intensive 158–62.
exercise. Medicine and Science in Sports and Exer-
cise, 28(4), 450–456. 7. Craib, M.W., Mitchell, V.A., Fields, K.B., Cooper,
T.R., Hopewell, R., and Morgan, D.W. 1996. The
21. Wilmore, J.H., and Costill, D.L. 2008. Physiol- association between flexibility and running
ogy of sport and exercise (4th ed., pp. 116–117). economy in sub-elite male distance runners.
Champaign, IL: Human Kinetics. Medicine and Science in Sports and Exercise, 28(6),
22. Wilmore, J.H., and Costill, D.L. 2004. Physiology 737–743.
of sport and exercise (3rd ed.). Champaign, IL: 8. Ibid., 737.
Human Kinetics, p. 100.
9. Nelson, A.G., Kokkonen, J., Eldredge, C., Corn-
23. De Pauw, K., De Geus, B., Roelands, B., Lauw- well, A., and Glickman-Weiss, E. 2001. Chronic
ens, F., Verschueren, J., Heyman, E., and Meeu- stretching and running economy. Scandinavian
sen, R.R. 2011. Effect of five different recovery Journal of Medicine and Science in Sports, 11(5),
methods on repeated cycle performance. 260–265.
Medicine and Science in Sports and Exercise, 43(5),
890–897. 10. Mojock, C.D., Kim, J.S., Eccles, D.W., and
Panton, L.B. 2011. The effects of static stretch-
24. Foss, M.L., Ketevian, S.J., and Fox, E.L. 1998. ing on running economy and endurance per-
Fox’s Physiological Basis for Exercise and Sport. formance in female distance runners during
New York: WCB/McGraw-Hill. treadmill running. Journal of Strength and
25. Vaile, J., Halson, S., Gill, N., and Dawson, B. Conditioning Research, 25(8), 2170–2176.
2008. Effect of hydrotherapy on recovery from 11. Allison, S., Bailey, D., and Folland, J. 2008.
fatigue. International Journal of Sports Medicine, Prolonged static stretching does not influence
29(7), 539–544. running economy despite changes in neuro-
26. Ingram, J., Dawson, B., Goodman, C., Wallman, muscular function. Journal of Sports Sciences,
K., and Beilby, J. 2009. Effect of water immer- 26(14), 1489–1495.
sion methods on post-exercise recovery from 12. Hayes, P.R., and Walker, A. 2007. Pre-exercise
simulated team sport exercise. Journal of Science stretching does not impact upon running
and Medicine in Sport, 12(3), 417–421. economy. Journal of Strength and Conditioning
Research, 21(4), 1227–1232.
Chapter 17 Flexibility and Core Strength
13. Prichard, B. 1984. Lower extremity injuries in run-
1. Araujo, C.G. 1999. Body flexibility profile and ners induced by upper body torque (UBT). Lecture
clustering among male and female elite ath- presented at Biomechanics and Kinesiology in
letes. Medicine and Science in Sports and Exercise, Sports U.S. Olympic Sports Medicine Confer-
31(5), S115. ence, Colorado Springs, CO.
2. McCullough, A., Kraemer, W., Volek, J., 14. Egermann, M., Brocai, D., Lill, C.A., and
Solomon-Hill, G.F., Hatfield, D., Vingren, J., Schmitt, H. 2003. Analysis of injuries in long-
References | 601

distance triathletes. International Journal of with and without patellofemoral pain. Journal
Sports Medicine, 24(4), 271–276. of Orthopaedic and Sports Physical Therapy, 33(11),
15. Byrnes, W.C., McCullagh, P., Dickinson, A., and 671–676.
Noble, J. 1992. Incidence and severity of injury 27. Leetun, D.T., Ireland, M.L., Willson, J.D., Ballan-
following aerobic training programs empha- tyne, B.T., and Davis, I.M. 2004. Core stability
sising running, racewalking, or step aerobics. measures as risk factors for lower extremity
Medicine and Science in Sports and Exercise, 25(5), injury in athletes. Medicine and Science in Sports
S81. and Exercise, 36(6), 926–934.
16. Korkia, P.K., Tunstall-Pedoe, D.S., and Maffulli, 28. Fredericson, M., Cookingham, C., Chaud-
N. 1994. An epidemiological investigation of hari, A.M., Brian, C., Oestreichner, N., and
training and injury patterns in British tri- Sahrmann, S.A. 2000. Hip abductor weakness
athletes. British Journal of Sports Medicine, 28(3), in distance runners with iliotibial band syn-
191–196. drome. Clinical Journal of Sport Medicine, 10(3),
17. Krivickas, L., and Feinberg, J.H. 1996. Lower 169–175.
extremity injuries in college athletes: Rela- 29. Sherry, M.A., and Best, T.M. 2004. A com-
tion between ligamentous laxity and lower parison of 2 rehabilitation programs in the
extremity muscle tightness. Archives of Physical treatment of acute hamstring strains. Journal
Medicine and Rehabilitation, 77(11), 1139–1143. of Orthopaedic and Sports Physical Therapy, 34(3),
18. Wilber, C.A., Holland, G.J., Madison, R.E., and 116–125.
Loy, S.F. 1995. An epidemiological analysis of 30. Morton, D.P., and Callister, R. 2010. Influence
overuse injuries among recreational cyclists. of posture and body type on the experience
International Journal of Sports Medicine, 16(3), of exercise-related transient abdominal pain.
201–206. Journal of Science and Medicine in Sport, 13(5),
19. American College of Sports Medicine position 483–488.
stand. The recommended quantity and quality 31. Sato, K., and Mokha, M. 2009. Does core
of exercise for developing and maintaining strength training influence running kinetics,
cardiorespiratory and muscular fitness, and lower-extremity stability, and 5000-M per-
flexibility in healthy adults. 1998. Medicine and formance in runners? Journal of Strength and
Science in Sports and Exercise, 30(6), 975. Conditioning Research, 23(1), 133–140.
20. Pereles, D., Roth, A., and Thompson, D. 2011, 32. Schibek, J.S., Guskiewicz, K.M., Prentice, W.E.,
February. The impact of a pre-run stretch on the Mays, S., and Davis, J.M. 2001. The effect of core
risk of injury in runners. American Academy of stabilization training on functional performance in
Orthopaedic Surgeons 2011 Annual Meeting swimming. Master's thesis, University of North
Paper Presentations, San Diego, California. Carolina, Chapel Hill.
21. Thacker, S.B., Gilchrist, J., Stroup, D.F., and 33. Stanton, R., Rearburn, P.R., and Humphries, B.
Kimsey, C.D. 2004. The impact of stretching 2004. The effect of short-term Swiss ball train-
on sports injury risk: A systematic review of ing on core stability and running economy.
the literature. Medicine and Science in Sports and Journal of Strength and Conditioning Research,
Exercise, 36(3), 371–378. 18, 522–528.
22. Alter, M.J. 2004. Science of flexibility. Cham- 34. Okada, T., Huxel, K.C., and Nesser, T.W. 2011.
paign, IL: Human Kinetics. Relationship between core stability, func-
23. Khaund, R., and Flynn, S.H. 2005. Iliotibial band tional movement, and performance. Journal
syndrome: A common source of knee pain. of Strength and Conditioning Research, 25(1),
American Family Physician, 71(8), 1545–1550. 252–261.
24. Maffulli, N., Pankaj, S., and Luscombe, K.L. Chapter 18 Strength Training
2004. Achilles tendinopathy: Aetiology and
management. Journal of the Royal Society of 1. Telephone interview with Tudor O. Bompa,
Medicine, 97(10), 472–476. PhD [Interview by D.W. Warden]. (2007,
25. Hodges, P.W., and Richardson, C.A. 1997. Con- November 28). Retrieved from www.tri-talk.
traction of the abdominal muscles associated com
with movement of the lower limb. Physical 2. Bompa, T.O. 1999. Periodization: Theory and meth-
Therapy, 77(2), 132–142. odology of training (4th ed., p. 316). Champaign,
26. Ireland, M.L., Wilson, J.D., Ballantyne, B.T., IL: Human Kinetics.
and Davis, I.M. 2003. Hip strength in females 3. Ibid., p. 318.
602 } References

4. Weyland, P.G., Sternlight, D.B., Bellizzi, M.J., 14. Ploutz, L.L., Tesch, P.A., Biro, R.L., and Dudley,
and Wright, S. 2000. Faster top running speeds G.A. 1994. Effect of resistance training on
are achieved with greater ground forces not muscle use during exercise. Journal of Applied
more rapid leg movements. Journal of Applied Physiology, 76(4), 1675–1681.
Physiology, 89(5), 1991–1999. 15. Goreham, C., Green, H.J., Ball-Burnett, M., and
5. Hickson, R.C., Dvorak, B.A., Gorostiaga, E.M., Ranney, D. 1999. High-resistance training and
Kurowski, T.T., and Foster, C. 1988. Potential muscle metabolism during prolonged exercise.
for strength and endurance training to amplify American Journal of Physiology, Endocrinology
endurance performance. Journal of Applied and Metabolism, 276(3), E489–E496.
Physiology, 65(5), 2285–2290. 16. Johnston, R.E., Quinn, T.J., Kertzer, R., and
6. Marcinik, E.J., Potts, J., Schlabach, G., Will, Vroman, N.B. 1997. Strength training in female
S., Dawson, P., and Hurley, B.F. 1991. Effects distance runners: Impact on running economy.
of strength training on lactate threshold and Journal of Strength and Conditioning Research,
endurance performance. Medicine and Science 11(4), 224–229.
in Sports and Exercise, 23(6), 739–743.
17. Aagaard, P., Andersen, J.L., Bennekou, M.,
7. Paavolainen, L., Hakkinen, K., Hamalainen, I., Larsson, B., Olesen, J.L., Crameri, R., Kjaer, M.
Nummela, A., and Rusko, H. 1999. Explosive- 2011. Effects of resistance training on endur-
strength training improves 5-km running time ance capacity and muscle fiber composition
by improving running economy and muscle in young top-level cyclists. Scandinavian
power. Journal of Applied Physiology, 89(5), Journal of Medicine and Science in Sports. doi:
1527–1533. 10.1111/j.1600-0838.2010.01283.x
8. Mikkola, J., Vesterinen, V., Taipale, R., Capost- 18. Louis, J., Hausswirth, C., Easthope, C., and
agno, B., Hakkinen, K., and Nummela, A. 2011. Brisswalter, J. 2012. Strength training improves
Effect of resistance training regimens on tread- cycling efficiency in master endurance athletes.
mill running and neuromuscular performance European Journal of Applied Physiology, 112(2),
in recreational endurance runners. Journal of 631–640.
Sports Science, 29(13), 1359–1371.
19. Bishop, D., Jenkins, D., Mackinnon, L.T., McE-
9. Storen, O., Helgerud, J., Stoa, E.M., and Hoff, niery, M., and Carey, M.F. 1999. The effects of
J. 2008. Maximal strength training improves strength training on endurance performance
running economy in distance runners. Medi- and muscle characteristics. Medicine and Science
cine and Science in Sports and Exercise, 40(6), in Sports and Exercise, 31(6), 886–891.
1087–1092.
20. Ferrauti, A., Bergermann, M., and Fernandez-
10. Millet, G.P., Jaquen, B., Borrant, F., and Candau, Fernandez, J. 2010. Effects of a concurrent
R. 2002. Effects of concurrent endurance and strength and endurance training on running
strength
. training on running economy and performance and running economy in recre-
VO2 kinetics. Medicine and Science in Sports and ational marathon runners. Journal of Strength
Exercise, 34(8), 1351–1359. and Conditioning Research, 24(10), 2770–2778.
11. Yamamoto, L.M., Lopez, R.M., Klau, J.F., Casa, 21. Aagaard, P., and Andersen, J.L. 2010. Effects
D.J., Kraemer, W.J., and Maresh, C.M. 2008. The of strength training on endurance capacity
effects of resistance training on endurance in top-level endurance athletes. Scandinavian
distance running performance among highly Journal of Medicine and Science in Sports. doi:
trained runners: A systematic review. Journal 10.1111/j.1600-0838.2010.01197.x
of Strength and Conditioning Research, 22(6), 2036.
22. Tanaka, H., Costill, D.L., Thomas, R., Fink,
12. Yamamoto, L.M., Klau, J.F., Casa, D.J., Arm- W.J., and Widrick, J.J. 1993. Dry-land resistance
strong, L.E., and Maresh, C.M. 2010. The
training for competitive swimming. Medicine
effects of resistance training of road cycling
and Science in Sports and Exercise, 25(8), 952–
performance among highly trained cyclists:
959.
A systematic review. Journal of Strength and
Conditioning Resistance, 24(2), 560. 23. Trappe, S.W., and Pearson, D.R. 1994. Effects
of weight assisted dry-land strength training
13. Esteve-Lanao, J., Rhea, M.R., Fleck, S.J., and
on swimming performance. Journal of Strength
Lucia, A. 2008. Running-specific, periodized
and Conditioning Research, 8(4), 209–275.
strength training attenuates loss of stride
length during intense endurance running. 24. Swimming, Athens and Aquaman: Michael
Journal of Strength and Conditioning Research, Phelps interview [Interview by P. Mauro].
22(4), 1176–1163. (2008, August 13). IGN Sportsw.
References | 603

25. Michaelis, V. 2008, August 3. Built to swim, Journal of Strength and Conditioning Research,
Phelps found a focus and refuge in water. USA 20(4), 947–954.
Today.
26. Tanaka, H., and Swensen, T. 1998. Impact of Chapter 19 General and Specific
resistance training on endurance performance. Training
A new form of cross-training. Sports Medicine, 1. Duffield, R., Dawson, B., and Goodman, C.
25(3), 191. 2005. Energy system contribution to 400-metre
27. Ibid, 191. and 800-metre track running. Journal of Sports
28. Friel, J. 2009. The triathlete’s training bible (3rd Sciences, 23(3), 299–307.
ed., pp. 242–247). Boulder, CO: VeloPress. 2. Wilmore, J.H., Costill, D.L., and Kenney, W.L.
29. Chtara, M., Chamari, K., Chaouachi, M., Cha- 2008. Physiology of sport and exercise (p. 241).
ouachi, A., Koubaa, D., Feki, Y., and Amri, Champaign, IL: Human Kinetics.
M. 2005. Effects of intra-session concurrent 3. Seiler, K.S., and Kjerland, G.O. 2006. Quanti-
endurance and strength training sequence fying training intensity distribution in elite
on aerobic performance and capacity. British endurance athletes: Is there evidence for an
Journal of Sports Medicine, 39(8), 555–560. “optimal” distribution? Scandinavian Journal
30. Ronnestad, B.R., Hansen, E.A., and Raastad, T. of Medicine and Science in Sports, 16(1), 49–56.
2010. In-season strength maintenance training 4. Esteve-Lanao, J., Juan, A.S., Earnest, C.P., Foster,
increases well-trained cyclists' performance. C., and Lucia, A. 2005. How do endurance run-
European Journal of Applied Physiology, 110(6), ners actually train? Relationship with competi-
1269–1282. tion performance. Medicine and Science in Sports
31. Kraemer, W.J., and Fry, A.C. 1991. Strength and Exercise, 37(3), 496–504.
testing: Development and evaluation of 5. Esteve-Lanao, J., Foster, C., Seiler, S., and Lucia,
methodology. In P.J. Maud and C. Foster A. 2007. Impact of training intensity distribu-
(Eds.), Physiological assessment of human fitness tion on performance in endurance athletes.
(2nd ed., pp. 129–130). Champaign IL: Human Journal of Strength and Conditioning Research,
Kinetics. 21(3), 943.
32. Brzycki, M. 1993. Strength testing: Predicting 6. Coggan, A., Raguso, C., Gastaldelli, A., Sidos-
a one-rep max from reps-to-fatigue. JOPERD, sis, L., and Yeckel, C. 2000. Fat metabolism
64, 88–90. during high-intensity exercise in endurance-
33. Gourgoulis, V., Aggeloussis, N., Vezos, N., trained and untrained men. Metabolism, 49(1),
Kasimatis, P., Antoniou, P., and Mavromatis, G. 122–128.
2007. Estimation of hand forces and propelling 7. Toussaint, H.M., and Beek, P.J. 1992. Biome-
efficiency during front crawl swimming with chanics of competitive front crawl swimming.
hand paddles. Journal of Biomechanics, 41(1), Sports Medicine, 13(1), 8–24.
208–215. 8. Lucia, A., Hoyos, J., and Chicharro, J.L. 2001.
34. Toussaint, H.M., and Vervoom, K. 1990. Effects Preferred pedalling cadence in professional
of specific high resistance training in the water cycling. Medicine and Science in Sports and Exer-
on competitive swimmers. International Journal cise, 33(8), 1361–1366.
of Sports Medicine, 11(3), 228–233. 9. Leirdal, S., and Ettema, G. 2011. The relation-
35. Girold, S., Calmels, P., Maurin, D., Milhau, ship between cadence, pedalling technique
N., and Chatard, J.C. 2006. Assisted and and gross efficiency in cycling. European Journal
resisted spring training in swimming. Journal of Applied Physiology, 111(12), 2885–2893.
of Strength and Conditioning Research, 20(3), 10. Chavarren, J., and Calbet, J.L. 1999. Cycling
547–554. efficiency and pedalling frequency in road
36. Turner, A.M., Owings, M., and Schwane, J.A. cyclists. European Journal of Applied Physiology,
2003. Improvement in running economy after 6 80(6), 555–563.
weeks of plyometric training. Journal of Strength 11. Gottschall, J.S., and Palmer, B.M. 2002. The acute
and Conditioning Research, 17(1), 60–67. effects of prior cycling cadence on running
37. Saunders, P.U., Telford, R.D., Pyne, D.B., performance and kinematics. Medicine and
Peltola, E.M., Cunningham, R.B., Gore, C.J., Science in Sports and Exercise, 34(9), 1518–1522.
and Hawley, J.A. 2006. Short-term plyometric 12. Dantas, J.L., Smirmaul, B.P., Altimari, L.R.,
training improves running economy in highly Okano, A.H., Fontes, E.B., Camata, T.V., and
trained middle and long distance runners. Moraes, A.C. 2009. The efficiency of pedaling
604 } References

and the muscular recruitment are improved aged men. Journal of Strength and Conditioning
with increase of the cadence in cyclists and Research, 25(4), 1104–1112.
non-cyclists. Electromyography and Clinical 8. Bartlett, J.D., Close, G.L., MacLaren, D.P.,
Neurophysiology, 49(6–7), 311–319. Gregson, W., Drust, B., and Morton, J.P. 2011.
13. Nimmerichter, A., Eston, R., and Williams, C. High-intensity interval running is perceived
2012. Effects of low and high cadence interval to be more enjoyable than moderate-intensity
training on power output in flat and uphill continuous exercise: Implications for exer-
cycling time-trials. European Journal of Applied cise adherence. Journal of Sports Science, 29(6),
Physiology, 112(1), 69–78. 547–553.
14. Paton, C.D., Hopkins, W.G., and Cook, C. 2009. 9. Ekkekakis, P., Hall, E.E., and Petruzzello, S.J.
Effects of low- vs. high-cadence interval train- 2008. The relationship between exercise inten-
ing on cycling performance. Journal of Strength sity and affective responses demystified: To
and Conditioning Research, 23(6), 1758–1763. crack the 40-year-old nut, replace the 40-year-
old nutcracker! Annals of Behavioral Medicine,
Chapter 20 Interval Training 35(2), 136–149.
1. Margaria, R., Oliva, R.D., Di Prampero, P.E., 10. Nekoliczak, A., and Petruzzello, S. 2011. Effect of
and Cerretelli, P. 1969. Energy utilization in differing intensities of exercise on affect and enjoy-
intermittent exercise of supramaximal inten- ment. Unpublished master’s thesis. IDEALS
sity. European Journal of Applied Physiology, 26(6), College of Applied Health Sciences.
752–756. 11. Williams, D., Dunsiger, S., Ciccolo, J., Lewis, B.,
2. Billat, V., Pinoteau, J., and Petit, B. 1996. Calibra- Albrecht, A., and Marcus, B. 2008. Acute affec-
tion de la duree des repetitions d'une seance tive response to a moderate-intensity exercise
d'interval
. training a la vitesse associee a stimulus predicts physical activity participa-
VO2max en reference au temps limite continu. tion 6 and 12 months later. Psychology of Sport
Science and Motricite, 28, 13–20. and Exercise, 9(3), 231–245.
3. Billat, V.L., Slawinski, J., Bocquet, V., Demarle, 12. Tomlin, D.L., and Wenger, H.A. 2001. The rela-
A., Lafitte, L., Chassaing, P., and Koralsztein, tionship between aerobic fitness and recovery
J. 2000. Intermittent runs at the velocity asso- from high intensity intermittent exercise.
ciated with maximal oxygen uptake enables Sports Medicine, 31(1), 1–11.
subjects to remain at maximal oxygen uptake 13. Wilkinson, J.G., Martin, D.T., Adams, A.A., and
for a longer time than intense but submaximal Liebman, M. 2002. Iron status in cyclists during
runs. European Journal of Applied Physiology, high-intensity interval training and recovery.
81(3), 188–196. International Journal of Sports Medicine, 23(8),
4. Esfarjani, F., and Laursen, P.B. 2007. Manipu- 544–548.
lating
. high-intensity interval training: Effects 14. Seller, S. 2010. What is best practice for training
on VO2max, the lactate threshold and 3000 m intensity and duration distribution in endur-
running performance in moderately trained ance athletes? International Journal of Sports
males. Journal of Science and Medicine in Sport, Physiology and Performance, 5(3), 276–291.
10(1), 27–35. 15. Laursen, P.B. 2010. Training for intense exercise
5. Laursen, P.B., Shing, C.M., Peake, J.M., performance: High-intensity or high-volume
Coombes, J.S., and Jenkins, D.G. 2002. Inter- training? Scandinavian Journal of Medicine and
val training program optimization in highly Science in Sports, 20, 1–10.
trained endurance cyclists. Medicine and Science 16. Lindsay, F.H., Hawley, J.A., Myburgh, K.H.,
in Sports and Exercise, 34(11), 1801–1807. Schomer, H.H., Noakes, T.D., and Dennis, S.C.
6. Sperlich, B., Zinner, C., Heilemann, I., Kjendlie, 1996. Improved athletic performance in highly
P., Holmberg, H., and Mester, J. 2010. . High- trained cyclists after interval training. Medi-
intensity interval training improves VO2peak, cine and Science in Sports and Exercise, 28(11),
maximal lactate accumulation, time trial and 1427–1434.
competition performance in 9-11-year-old 17. Westgarth-Taylor, C., Hawley, J.A., Rickard,
swimmers. European Journal of Applied Physiol- S., Myburgh, K.H., Noakes, T.D., and Dennis,
ogy, 110(5), 1029–1036. S.C. 1997. Metabolic and performance adapta-
7. Ziemann, E., Grzywacz, T., Luszczyk, M., Las- tions to interval training in endurance-trained
kowski, R., Olek, R.A., and Gibson, A.L. 2011. cyclists. European Journal of Applied Physiol-
Aerobic and anaerobic changes with high- ogy and Occupational Physiology, 75(4), 298–
intensity interval training in active college- 304.
References | 605

18. Stepto, N.K., Hawley, J.A., Dennis, S.C., and ment in lactate threshold. International Journal
Hopkins, W.G. 1999. Effects of different inter- of Sports Medicine, 31(08), 567–571.
val-training programs on cycling time-trial 29. Fox, E.L., Bartels, R.L., Billings, C.E., O'Brien, R.,
performance. Medicine and Science in Sports and Bason, R., and Mathews, D.K. 1975. Frequency
Exercise, 31(5), 736–741. and duration of interval training programs and
19. Helgerud, J., Hoydal, K., Wang, E., Karlsen, T., changes in aerobic power. Journal of Applied
Berg, P., Bjerkaas, M., and Hoff, J. 2007.
. Aerobic Physiology, 38(3), 481–484.
high-intensity intervals improve VO2max more 30. Paton, C.D., and Hopkins, W.G. 2005. Combin-
than moderate training. Medicine and Science in ing explosive and high-resistance training
Sports and Exercise, 39(4), 665–671. improves performance in competitive cyclists.
20. Smith, T.P., Coombes, J.S., and Geraghty, D.P. Journal of Strength and Conditioning Research,
2003. Optimising high-intensity treadmill 19(4), 826–830.
training using the running speed at maximal
O2 uptake and the time for which this can be Chapter 21 Duration, Frequency, and
maintained. European Journal of Applied Physiol- Intensity
ogy, 89(3), 337–343.
1. Powers, S., and Howley, E.T. 2008. Exercise
21. Weston, A.R., Myburgh, K.H., Lindsay, F.H., physiology: Theory and application to fitness and
Dennis, S.C., Noakes, T.D., and Hawley, J.A. performance level. New York: McGraw-Hill.
1996. Skeletal muscle buffering capacity and
endurance performance after high-intensity 2. Zaryski, C., and Smith, D.J. 2005. Training prin-
interval training by well-trained cyclists. Euro- ciples and issues for ultra-endurance athletes.
pean Journal of Applied Physiology, 75(1), 7–13. Current Sports Medicine Reports, 4(3), 165–170.
22. Franch, J., Madsen, K., Djurhuus, M., Mogens, 3. Main, L.C., Landers, G.J., Grove, J.R., Dawson,
S., and Pedersen, P.K. 1998. Improved run- B., and Goodman, C. 2010. Training patterns
ning economy following intensified training and negative health outcomes in triathlon:
correlates with reduced ventilatory demands. Longitudinal observations across a full com-
Medicine and Science in Sports and Exercise, 30(8), petitive season. Journal of Sports Medicine and
1250–1256. Physical Fitness, 50(4), 475–485.
23. Paton, C.C., and Hopkins, W.G. 2004. Effects 4. Gosling, C.M., Forbes, A.B., McGivern, J., and
of high-intensity training on performance and Gabbe, B.J. 2010. A profile of injuries in athletes
physiology of endurance athletes. Sportscience, seeking treatment during a triathlon race
8, 25–40. series. American Journal of Sports Medicine, 38(5),
1007–1014.
24. Laursen, P.B., Blanchard, M.A., and Jenkins,
D.G. 2002. Acute high-intensity interval train- 5. Vleck, V.E., Bentley, D.J., Millet, G.P., and
ing improves Tvent and peak power output Cochrane, T. 2010. Triathlon event distance spe-
in highly trained males. Canadian Journal of cialization: Training and injury effects. Journal
Applied Physiology, 27(4), 336–348. of Strength and Conditioning Research, 24(1),
30–36.
25. Belcastro, A.N., and Bonen, A. 1975. Lactic acid
removal rates during controlled and uncon- 6. Gosling, C.M., Gabbe, B.J., and Forbes, A.B.
trolled recovery exercise. Journal of Applied 2008. Triathlon related musculoskeletal
Physiology, 39(6), 932–936. injuries: The status of injury prevention
knowledge. Journal of Science and Medicine in
26. Declan, A.J., Brennan, K.M., and Lauzon, C.D. Sport, 11(4), 396–406.
2003. Effects of active versus passive recovery
on power output during repeated bouts of 7. Hickson, R.C., Hagberg, J.M., Ehsani, A.A., and
short term, high intensity exercise. Journal of Holloszy, J.O. 1981. Time course of the adaptive
Sports Science and Medicine, 2, 47–51. responses of aerobic power and heart rate to
training. Medicine and Science in Sports and
27. Felix, S.D., Manos, T.M., Jarvis, A.T., Jensen, Exercise, 13(1), 17–20.
B.E., and Headley, S.A. 1997. Swimming perfor-
mance following different recovery protocols 8. Hickson, R.C., Overland, S.M., and Dougherty,
in female collegiate swimmers. Journal of Swim- K.A. 1984. Reduced training frequency effects
ming Research, 12, 1–6. on aerobic power and muscle adaptations
in rats. Journal of Applied Physiology, 57(6),
28. Dalleck, L., Bushman, T.T., Crain, R.D., Gajda, 1834–1841.
M.M., Koger, E.M., and Derksen, L.A. 2010.
Dose-response relationship between interval 9. Knechtle, B., Knechtle, P., Rust, C.A., and Rose-
training frequency and magnitude of improve- mann, T. 2011. A comparison of anthropometric
606 } References

and training characteristics of Ironman tri- 20. Tabata, I., Nishimura, K., Kouzaki, M., Hirai,
athletes and Triple Iron ultra-triathletes. Jour- Y., Ogita, F., Miyachi, M., et al. 1996. Effects
nal of Sports Sciences, 29(13), 1373–1380. of moderate-intensity endurance and high-
10. Neal, C.M., Hunter, A.M., and Galloway, S.D. intensity intermittent
. training on anaerobic
2011. A 6-month analysis of training-intensity capacity and VO2max. Medicine and Science in
distribution and physiological adaptation in Sports and Exercise, 28(10), 1327–1330.
Ironman triathletes. Journal of Sports Sciences, 21. Green, H., Tupling, R., Roy, B., O’Toole, D.,
29(14), 1515–1523. Burnett, M., and Grant, S. 2000. Adaptations in
11. Iaia, F.M., and Bangsbo, J. 2010. Speed endur- skeletal muscle exercise metabolism to a sus-
ance training is a powerful stimulus for tained session of heavy intermittent exercise.
physiological adaptations and performance American Journal of Physiology, Endocrinology,
improvements of athletes. Scandinavian Journal and Metabolism, 278(1), E118–E126.
of Medicine and Science in Sports, 20(suppl. 2), 22. Green, H.J., and Fraser, I.G. 1988. Differential
11–23. effects of exercise intensity on serum uric acid
12. Iaia, F.M., Hellsten, Y., Nielsen, J.J., Fernstrom, concentration. Medicine and Science in Sports
M., Sahlin, K., and Bangsbo, J. 2009. Four weeks and Exercise, 20(1), 55–59.
of speed endurance training reduces energy 23. Laursen, P.B., Blanchard, M.A., and Jenkins,
expenditure during exercise and maintains D.G. 2002. Acute high-intensity interval train-
muscle oxidative capacity despite a reduction ing improves Tvent and peak power output
in training volume. Journal of Applied Physiol- in highly trained males. Canadian Journal of
ogy, 106(1), 73–80. Applied Physiology, 27(4), 336–348.
13. Laursen, P.B., and Jenkins, D.G. 2002. The 24. Lindsay, F.H., Hawley, J.A., Myburgh, K.H.,
scientific basis for high-intensity interval train- Schomer, H.H., Noakes, T.D., and Dennis, S.C.
ing: Optimising training programmes and 1996. Improved athletic performance in highly
maximising performance in highly trained trained cyclists after interval training. Medi-
endurance athletes. Sports Medicine, 32(1), 53– cine and Science in Sports and Exercise, 28(11),
73. 1427–1434.
14. Laursen, P.B., Shing, C.M., Peake, J.M., 25. Stepto, N.K., Hawley, J.A., Dennis, S.C., and
Coombes, J.S., and Jenkins, D.G. 2002. Inter- Hopkins, W.G. 1999. Effects of different inter-
val training program optimization in highly val-training programs on cycling time-trial
trained endurance cyclists. Medicine and Science performance. Medicine and Science in Sports
in Sports and Exercise, 34(11), 1801–1807. and Exercise, 31(5), 736–741.
15. Esteve-Lanao, J., Foster, C., Seiler, S., and Lucia, 26. Weston, A.R., Myburgh, K.H., Lindsay, F.H.,
A. 2007. Impact of training intensity distribu- Dennis, S.C., Noakes, T.D., and Hawley, J.A.
tion on performance in endurance athletes. 1997. Skeletal muscle buffering capacity and
Journal of Strength and Conditioning Research, endurance performance after high-intensity
21(3), 943–949. interval training by well-trained cyclists.
16. Seiler, S. 2010. What is best practice for training European Journal of Applied Physiology and
intensity and duration distribution in endur- Occupational Physiology, 75(1), 7–13.
ance athletes? International Journal of Sports 27. Westgarth-Taylor, C., Hawley, J.A., Rickard,
Physiology and Performance, 5(3), 276–291. S., Myburgh, K.H., Noakes, T.D., and Dennis,
17. Laursen, P.B., Shing, C.M., Peake, J.M., S.C. 1997. Metabolic and performance adapta-
Coombes, J.S., and Jenkins, D.G. 2005. Influence tions to interval training in endurance-trained
of high-intensity interval training on adapta- cyclists. European Journal of Applied Physiology
tions in well-trained cyclists. Journal of Strength and Occupational Physiology, 75(4), 298–304.
and Conditioning Research, 19(3), 527–533. 28. Iaia, F.M., Perez-Gomez, J., Thomassen, M.,
18. Londeree, B.R. 1997. Effect of training on lac- Nordsborg, N.B., Hellsten, Y., and Bangsbo,
tate/ventilatory thresholds: A meta-analysis. J. 2011. Relationship between performance
Medicine and Science in Sports and Exercise, 29(6), at different exercise intensities and skeletal
837–843. muscle characteristics. Journal of Applied Physi-
ology, 110(6), 1555–1563.
19. Tabata, I., Irisawa, K., Kouzaki, M., Nishimura,
K., Ogita, F., and Miyachi, M. 1997. Metabolic 29. Iaia, F.M., Thomassen, M., Kolding, H., Gun-
profile of high intensity intermittent exercises. narsson, T., Wendell, J., Rostgaard, T., et al.
Medicine and Science in Sports and Exercise, 29(3), 2008. Reduced volume but increased train-
390–395. ing intensity elevates muscle Na+-K+ pump
References | 607

alpha1-subunit and NHE1 expression as well 39. Bosquet, L., Leger, L., and Legros, P. 2002.
as short-term work capacity in humans. Ameri- Methods to determine aerobic endurance.
can Journal of Physiology, Regulatory, Integrative, Sports Medicine, 32(11), 675–700.
and Comparative Physiology, 294(3), R966– 40. Lydiard, A. 1990. Osaka lecture. Retrieved from
974. www.lydiardfoundation.org/pdfs/al_lecture.
30. Balsom, P.D., Gaitanos, G.C., Ekblom, B., and pdf
Sjodin, B. 1994. Reduced oxygen availability 41. McGregor, S.J., and Fitzgerald, M. 2010. The
during high intensity intermittent exercise runner's edge. Champaign, IL: Human Kinetics.
impairs performance. Acta Physiologica Scan-
dinavica, 152(3), 279–285. 42. Allen, H., and Coggan, A. 2010. Training and
racing with a power meter. Boulder, CO: Velo-
31. Gaitanos, G.C., Williams, C., Boobis, L.H., and Press.
Brooks, S. 1993. Human muscle metabolism
during intermittent maximal exercise. Journal Chapter 22 Periodization
of Applied Physiology, 75(2), 712–719.
1. Bompa, T.O. 2009. Periodization: Theory and
32. Parolin, M.L., Spriet, L.L., Hultman, E.,
methodology of training (5th ed.). Champaign,
Hollidge-Horvat, M.G., Jones, N.L., and Hei-
IL: Human Kinetics.
genhauser, G.J. 2000. Regulation of glycogen
phosphorylase and PDH during exercise 2. Selye, H. 1978. The stress of life. New York:
in human skeletal muscle during hypoxia. McGraw-Hill.
American Journal of Physiology, Endocrinology, 3. Bompa, T.O. 2005. Periodization training for sports
and Metabolism, 278(3), E522–E534. (2nd ed.). Champaign, IL: Human Kinetics.
33. Trump, M.E., Heigenhauser, G.J., Putman, C.T., 4. Issurin, V.B. 2010. New horizons for the meth-
and Spriet, L.L. 1996. Importance of muscle odology and physiology of training periodiza-
phosphocreatine during intermittent maximal tion. Sports Medicine, 40(3), 189–206.
cycling. Journal of Applied Physiology, 80(5), 5. Rowbottom, D.G., Keast, D., Garcia-Webb, P.,
1574–1580. and Morton, A.R. 1997. Training adaptation
34. Parolin, M.L., Chesley, A., Matsos, M.P., Spriet, and biological changes among well-trained
L.L., Jones, N.L., and Heigenhauser, G.J. 1999. male triathletes. Medicine and Science in Sports
Regulation of skeletal muscle glycogen phos- and Exercise, 29(9), 1233–1239.
phorylase and PDH during maximal intermit- 6. Rowbottom, D.G., Keast, D., and Morton, A.R.
tent exercise. American Journal of Physiology, 1998. Monitoring and preventing of overreach-
277(5 Pt. 1), E890–E900. ing and overtraining in endurance athletes.
35. Harmer, A.R., McKenna, M.J., Sutton, J.R., In R.B. Kreider (ed.) Overtraining in sport (pp.
Snow, R.J., Ruell, P.A., Booth, J., et al. 2000. 47–66). Champaign, IL: Human Kinetics.
Skeletal muscle metabolic and ionic adapta- 7. Friel, J. 2009. The triathlete's training bible (3rd
tions during intense exercise following sprint ed.). Boulder, CO: VeloPress.
training in humans. Journal of Applied Physiol-
ogy, 89(5), 1793–1803. 8. Matveyev, L.P. 1965. Periodization of sports train-
ing. Moscow: Fizkultura i Sport.
36. McKenna, M.J., Heigenhauser, G.J., McKelvie,
R.S., Obminski, G., MacDougall, J.D., and 9. Smith, D.J. 2003. A framework for understand-
Jones, N.L. 1997. Enhanced pulmonary and ing the training process leading to elite perfor-
active skeletal muscle gas exchange during mance. Sports Medicine, 33(15), 1103–1126.
intense exercise after sprint training in men. 10. Stone, M.H., O'Bryant, H., and Garhammer,
Journal of Physiology, 501(Pt. 3), 703–716. J. 1981. A hypothetical model for strength
37. Fiskerstrand, A., and Seiler, K.S. 2004. Train- training. Journal of Sports Medicine and Physical
ing and performance characteristics among Fitness, 21(4), 342–351.
Norwegian international rowers 1970–2001. 11. Turner, A. 2011. The science and practice of
Scandinavian Journal of Medicine and Science in periodization: A brief review. Strength and
Sports, 14(5), 303–310. Conditioning Journal, 33(1), 34–46.
38. Billat, V., Bernard, O., Pinoteau, J., Petit, B., 12. Baechle, T.R., and Earle, R.W. (Eds.). 2008.
and. Koralsztein, J.P. 1994. Time to exhaustion Essentials of strength training and conditioning.
at VO2max and lactate steady state velocity Champaign, IL: Human Kinetics.
in sub elite long-distance runners. Archives 13. Zaryski, C., and Smith, D.J. 2005. Training prin-
Internationales de Physiologie, de Biochimie et de ciples and issues for ultra-endurance athletes.
Biophysique, 102(3), 215–219. Current Sports Medicine Reports, 4(3), 165–170.
608 } References

14. Mujika, I. 2009. Tapering and peaking for optimal 29. Buford, T.W., Rossi, S.J., Smith, D.B., and
performance. Champaign, IL: Human Kinetics. Warren, A.J. 2007. A comparison of periodiza-
15. Friel, J. 2009. The cyclist’s training bible. Boulder, tion models during nine weeks with equated
CO: VeloPress. volume and intensity for strength. Journal
of Strength and Conditioning Research, 21(4),
16. Fry, R.W., Morton, A.R., and Keast, D. 1991. 1245–1250.
Overtraining in athletes. An update. Sports
Medicine, 12(1), 32–65. 30. Hartmann, H., Bob, A., Wirth, K., and Schmidt-
bleicher, D. 2009. Effects of different periodiza-
17. Brown, L.E. 2001. Nonlinear versus linear tion models on rate of force development and
periodization models. Strength and conditioning power ability of the upper extremity. Journal
journal, 23(1), 42–44. of Strength and Conditioning Research, 23(7),
18. Issurin, V.B. 2008. Block periodization versus 1921–1932.
traditional training theory: A review. Journal of 31. Miranda, F., Simao, R., Rhea, M., Bunker, D.,
Sports Medicine and Physical Fitness, 48(1), 65–75. Prestes, J., Leite, R.D., et al. 2011. Effects of
19. Kiely, J. 2010. New horizons for the methodol- linear vs. daily undulatory periodized resis-
ogy and physiology of training periodization: tance training on maximal and submaximal
Block periodization: New horizon or a false strength gains. Journal of Strength and Condition-
dawn? Sports Medicine, 40(9), 803–805 [author ing Research, 25(7), 1824–1830.
reply]. 32. Prestes, J., Frollini, A.B., de Lima, C., Donatto,
20. McGregor, S.J., Weese, R.K., and Ratz, I.K. 2009. F.F., Foschini, D., de Cassia Marqueti, R., et al.
Performance modeling in an Olympic 1500-m 2009. Comparison between linear and daily
finalist: A practical approach. Journal of Strength undulating periodized resistance training to
and Conditioning Research, 23(9), 2515–2523. increase strength. Journal of Strength and Con-
21. Lydiard, A. 1990. Osaka lecture. Retrieved from ditioning Research, 23(9), 2437–2442.
www.lydiardfoundation.org/pdfs/al_lecture. 33. Rhea, M.R., Ball, S.D., Phillips, W.T., and Bur-
pdf kett, L.N. 2002. A comparison of linear and
22. Lydiard, A., and Gilmour, G. 2000. Running daily undulating periodized programs with
with Lydiard (2nd ed.). Oxford, UK: Meyer and equated volume and intensity for strength.
Meyer Sport. Journal of Strength and Conditioning Research,
16(2), 250–255.
23. Stewart, A.M., and Hopkins, W.G. 2000. Sea-
sonal training and performance of competitive 34. Rhea, M.R., Phillips, W.T., Burkett, L.N.,
swimmers. Journal of Sports Sciences, 18(11), Stone, W.J., Ball, S.D., Alvar, B.A., et al. 2003.
873–884. A comparison of linear and daily undulating
periodized programs with equated volume
24. Mujika, I., Chatard, J.C., Busso, T., Geyssant, A.,
and intensity for local muscular endurance.
Barale, F., and Lacoste, L. 1995. Effects of train-
Journal of Strength and Conditioning Research,
ing on performance in competitive swimming.
17(1), 82–87.
Canadian Journal of Applied Physiology, 20(4),
395–406. 35. Vanni, A.C., Meyer, F., da Veiga, A.D., and
Zanardo, V.P. 2010. Comparison of the effects of
25. Pyne, D.B., Lee, H., and Swanwick, K.M. 2001.
two resistance training regimens on muscular
Monitoring the lactate threshold in world-
and bone responses in premenopausal women.
ranked swimmers. Medicine and Science in
Osteoporosis International, 21(9), 1537–1544.
Sports and Exercise, 33(2), 291–297.
36. Banister, E.W., Carter, J.B., and Zarkadas, P.C.
26. Issurin, V.B. 2009. Generalized training effects
1999. Training theory and taper: Validation in
induced by athletic preparation. A review.
triathlon athletes. European Journal of Applied
Journal of Sports Medicine and Physical Fitness,
Physiology and Occupational Physiology, 79(2),
49(4), 333–345.
182–191.
27. Bondarchuk, A.P. 1988. Constructing a training
37. Busso, T., Candau, R., and Lacour, J.R. 1994.
system. Track Technique, 102, 3254–3269.
Fatigue and fitness modelled from the effects
28. Apel, J.M., Lacey, R.M., and Kell, R.T. 2010. A of training on performance. European Journal of
comparison of traditional and weekly undu- Applied Physiology and Occupational Physiology,
lating periodized strength training programs 69(1), 50–54.
with total volume and intensity equated. Jour-
38. LeMond, G., and Gordis, K. 1987. Greg LeMond's
nal of Strength and Conditioning Research, 25(3),
Complete Book of Bicycling. New York: Perigree.
694–703.
References | 609

39. Calvert, T.W., Banister, E.W., Savage, M.V., and weeks of training leading to the Sydney 2000
Bach, T. 1976. A systems model of the effects of Olympic Games. International Journal of Sports
training on physical performance. IEEE Transac- Medicine, 23(8), 582–587.
tions on Systems, Man and Cybernetics, 6, 94–102. 9. Jeukendrup, A.E., Hesselink, M.K., Snyder,
40. Allen, H., and Coggan, A. 2010. Training and A.C., Kuipers, H., and Keizer, H.A. 1992.
racing with a power meter. Boulder, CO: Velo- Physiological changes in male competitive
Press. cyclists after two weeks of intensified training.
41. Coggan, A.R. 2006. The Performance Manager. International Journal of Sports Medicine, 13(7),
Paper presented at the USA Cycling Coaching 534–541.
Summit. Retrieved from http://home.train- 10. Martin, D.T., Scifres, J.C., Zimmerman, S.D.,
ingpeaks.com/articles/cycling/the-science- and Wilkinson, J.G. 1994. Effects of interval
of-the-performance-manager.aspx training and a taper on cycling performance
42. McGregor, S.J., and Fitzgerald, M. 2010. The and isokinetic leg strength. International Journal
runner’s edge. Champaign, IL: Human Kinetics. of Sports Medicine, 15(8), 485–491.
43. Mureika, J.R., Covington, D., and Mercier, D. 11. Steinacker, J.M., Lormes, W., Kellmann, M.,
2000. The 1999 Mercier scoring tables: A how-to Liu, Y., Reissnecker, S., Opitz-Gress, A., et al.
guide. Athletics: Canada’s National Track and 2000. Training of junior rowers before world
Field / Running Magazine (April/May). championships: Effects on performance, mood
state and selected hormonal and metabolic
44. Fitz-Clarke, J.R., Morton, R.H., and Banister, responses. Journal of Sports Medicine and Physi-
E.W. 1991. Optimizing athletic performance by cal Fitness, 40(4), 327–335.
influence curves. Journal of Applied Physiology,
71(3), 1151–1158. 12. Mujika, I., and Padilla, S. 2000. Detraining:
Loss of training-induced physiological and
Chapter 23 Tapering and Peaking for performance adaptations. Part II: Long term
insufficient training stimulus. Sports Medicine,
Races 30(3), 145–154.
1. Bosquet, L., Montpetit, J., Arvisais, D., and 13. Mujika, I., and Padilla, S. 2000. Detraining:
Mujika, I. 2007. Effects of tapering on perfor- Loss of training-induced physiological and
mance: A meta-analysis. Medicine and Science performance adaptations. Part I: Short term
in Sports and Exercise, 39(8), 1358–1365. insufficient training stimulus. Sports Medicine,
2. Thomas, L., Mujika, I., and Busso, T. 2009. 30(2), 79–87.
Computer simulations assessing the potential 14. Mujika, I., and Padilla, S. 2001. Muscular
performance benefit of a final increase in train- characteristics of detraining in humans.
ing during pre-event taper. Journal of Strength Medicine and Science in Sports and Exercise, 33(8),
and Conditioning Research, 23(6), 1729–1736. 1297–1303.
3. Mujika, I. 2009. Tapering and peaking for optimal 15. Hickson, R.C., Foster, C., Pollock, M.L., Galassi,
performance. Champaign, IL: Human Kinetics. T.M., and Rich, S. 1985. Reduced training inten-
4. Houmard, J.A. 1991. Impact of reduced training sities and loss of aerobic power, endurance, and
on performance in endurance athletes. Sports cardiac growth. Journal of Applied Physiology,
Medicine, 12(6), 380–393. 58(2), 492–499.
5. Mujika, I., and Padilla, S. 2003. Scientific 16. Mujika, I. 2010. Intense training: The key to
bases for precompetition tapering strategies. optimal performance before and during the
Medicine and Science in Sports and Exercise, 35(7), taper. Scandinavian Journal of Medicine and Sci-
1182–1187. ence in Sports, 20(suppl. 2), 24–31.
6. Mujika, I., Padilla, S., Pyne, D., and Busso, T. 17. Costill, D.L., King, D.S., Thomas, R., and Har-
2004. Physiological changes associated with greaves, M. 1985. Effects of reduced training
the pre-event taper in athletes. Sports Medicine, on muscular power in swimmers. Physician
34(13), 891–927. and Sportsmedicine, 13, 94–101.
7. Pyne, D.B., Mujika, I., and Reilly, T. 2009. Peak- 18. Shepley, B., MacDougall, J.D., Cipriano, N.,
ing for optimal performance: Research limita- Sutton, J.R., Tarnopolsky, M.A., and Coates, G.
tions and future directions. Journal of Sports 1992. Physiological effects of tapering in highly
Science, 27(3), 195–202. trained athletes. Journal of Applied Physiology,
8. Mujika, I., Padilla, S., and Pyne, D. 2002. Swim- 72(2), 706–711.
ming performance changes during the final 3 19. Houmard, J.A., Scott, B.K., Justice, C.L., and
610 } References

Chenier, T.C. 1994. The effects of taper on 2007. Practical tests for monitoring perfor-
performance in distance runners. Medicine and mance, fatigue and recovery in triathletes.
Science in Sports and Exercise, 26(5), 624–631. Journal of Science and Medicine in Sport, 10(6),
20. Houmard, J.A., Costill, D.L., Mitchell, J.B., Park, 372–381.
S.H., Fink, W.J., and Burns, J.M. 1990. Testoster- 31. Coutts, A.J., Wallace, L.K., and Slattery, K.M.
one, cortisol, and creatine kinase levels in male 2007. Monitoring changes in performance,
distance runners during reduced training. physiology, biochemistry, and psychology
International Journal of Sports Medicine, 11(1), during overreaching and recovery in tri-
41–45. athletes. International Journal of Sports Medicine,
21. Houmard, J.A., Costill, D.L., Mitchell, J.B., Park, 28(2), 125–134.
S.H., Hickner, R.C., and Roemmich, J.N. 1990. 32. Banister, E.W., and Calvert, T.W. 1980. Planning
Reduced training maintains performance in for future performance: Implications for long
distance runners. International Journal of Sports term training. Canadian Journal of Applied Sport
Medicine, 11(1), 46–52. Science, 5(3), 170–176.
22. Houmard, J.A., Kirwan, J.P., Flynn, M.G., and 33. Busso, T. 2003. Variable dose-response rela-
Mitchell, J.B. 1989. Effects of reduced train- tionship between exercise training and per-
ing on submaximal and maximal running formance. Medicine and Science in Sports and
responses. International Journal of Sports Medi- Exercise, 35(7), 1188–1195.
cine, 10(1), 30–33. 34. Busso, T., Candau, R., and Lacour, J.R. 1994.
23. McConell, G.K., Costill, D.L., Widrick, J.J., Fatigue and fitness modelled from the effects
Hickey, M.S., Tanaka, H., and Gastin, P.B. 1993. of training on performance. European Journal of
Reduced training volume and intensity main- Applied Physiology and Occupational Physiology,
tain aerobic capacity but not performance in 69(1), 50–54.
distance runners. International Journal of Sports 35. Busso, T., and Thomas, L. 2006. Using math-
Medicine, 14(1), 33–37. ematical modeling in training planning.
24. Houmard, J.A., and Johns, R.A. 1994. Effects of International Journal of Sports Physiology and
taper on swim performance. Practical implica- Performance, 1(4), 400–405.
tions. Sports Medicine, 17(4), 224–232. 36. Coggan, A.R. 2006. The Performance Manager.
25. Neufer, P.D., Costill, D.L., Fielding, R.A., Flynn, Paper presented at the USA Cycling Coaching
M.G., and Kirwan, J.P. 1987. Effect of reduced Summit. Retrieved from http://home.train-
training on muscular strength and endurance ingpeaks.com/articles/cycling/the-science-
in competitive swimmers. Medicine and Science of-the-performance-manager.aspx www.
in Sports and Exercise, 19(5), 486–490. cyclingpeakssoftware.com/pmc_summit.pdf
26. Banister, E.W., Carter, J.B., and Zarkadas, P.C. 37. Hellard, P., Avalos, M., Lacoste, L., Barale, F.,
1999. Training theory and taper: Validation in Chatard, J.C., and Millet, G.P. 2006. Assessing
triathlon athletes. European Journal of Applied the limitations of the Banister model in moni-
Physiology and Occupational Physiology, 79(2), toring training. Journal of Sports Science, 24(5),
182–191. 509–520.
27. Zarkadas, P.C., Carter, J.B., and Banister, E.W. 38. Hellard, P., Avalos, M., Millet, G., Lacoste, L.,
1995. Modelling the effect of taper on per- Barale, F., and Chatard, J.C. 2005. Modeling the
formance, maximal oxygen uptake, and the residual effects and threshold saturation of
anaerobic threshold in endurance triathletes. training: A case study of Olympic swimmers.
Advances in Experimental Medicine and Biology, Journal of Strength and Conditioning Research,
393, 179–186. 19(1), 67–75.
28. Margaritis, I., Palazzetti, S., Rousseau, A.S., 39. McGregor, S.J., Weese, R.K., and Ratz, I.K.
Richard, M.J., and Favier, A. 2003. Anti- 2009. Performance modeling in an Olympic
oxidant supplementation and tapering exer- 1500-m finalist: A practical approach. Journal
cise improve exercise-induced antioxidant of Strength and Conditioning Research, 23(9),
response. Journal of the American College of 2515–2523.
Nutrition, 22(2), 147–156. 40. Calvert, T.W., Banister, E.W., Savage, M.V., and
29. Vollaard, N.B., Cooper, C.E., and Shearman, Bach, T. 1976. A systems model of the effects
J.P. 2006. Exercise-induced oxidative stress in of training on physical performance. IEEE
overload training and tapering. Medicine and Transactions on Systems, Man and Cybernetics, 6,
Science in Sports and Exercise, 38(7), 1335–1341. 94–102.
30. Coutts, A.J., Slattery, K.M., and Wallace, L.K. 41. Allen, H., and Coggan, A. 2010. Training and
References | 611

racing with a power meter. Boulder, CO: VeloPress. choneuroimmunology. Sports Medicine, 32(3),
42. Fitz-Clarke, J.R., Morton, R.H., and Banister, 185–209.
E.W. 1991. Optimizing athletic performance by 11. Uusitalo, A.L., Vanninen, E., Valkonen-Korho-
influence curves. Journal of Applied Physiology, nen, M., and Kuikka, J.T. 2006. Brain serotonin
71(3), 1151–1158. reuptake did not change during one year in
overtrained athletes. International Journal of
Chapter 24 Physiology of Overtraining Sports Medicine, 27(9), 702–708. Epub 2006 Feb 1.
1. Purvis, D., Gonsalves, S., and Deuster, P.A. 12. Budgett, R., Hiscock, N., Arida, R.M., and
2010. Physiological and psychological fatigue Castell, L.M. 2010. The effects of the 5-HT2C
in extreme conditions: Overtraining and elite agonist m-chlorophenylpiperazine on elite ath-
athletes. PM & R, 2, 442–450. letes with unexplained underperformance
syndrome (overtraining). British Journal of
2. Budgett, R., Newsholme, E., Lehmann, M.,
Sports Medicine, 44(4), 280–283.
Sharp, C., Jones, D., Peto, T., Collins, C.,
Nerurkar, R., and White, P. 2000. Redefining 13. Anderson, O. 2002. How do you know when
the over training syndrome as the unexplained you are at risk of overtraining? It's a simple
underperformance syndrome. British Journal of matter of well you feel, sleep and eat. Peak
Sports Medicine, 34, 67–68. Performance, 163, 1–4.
3. Ackel-D’Elia, C., Vancini, R.L., Castelo, A.,
Nouailhetas, V.L., and Silva, A.C. 2010. Absence
Chapter 25 Swim Base Building
of the predisposing factors and signs and 1. Friel, J. 2010. Your best triathlon. Boulder, CO:
symptoms usually associated with overreach- VeloPress.
ing and overtraining in physical fitness centers. 2. Knechtle, B., Baumann B., Knechtle, P., and
Clinics (Sao Paulo), 65(11), 1161–1166. Rosemann, T. 2010. Speed during training and
4. Koutedakis, Y., Frischknecht, R., Vrbová, G., anthropometric measures in relation to race
Sharp, N.C., and Budgett, R. 1995. Maximal performance by male and female open-water
voluntary quadriceps strength patterns in ultra-endurance swimmers. Perceptual and
Olympic overtrained athletes. Medicine and Motor Skills, 111(2), 463–474.
Science in Sports and Exercise, 27(4), 566–572. 3. Naemi, R., and Sanders, R.M. 2008. A "hydro-
5. Agostini, F., and Biolo, G. 2010. Effect of physi- kinematic" method of measuring the glide
cal activity on glutamine metabolism. Current efficiency of a human swimmer. Journal of
Opinion in Clinical Nutrition and Metabolic Care, Biomechanical Engineering, 130(6), 061016.
13(1), 58–64. Review. 4. Chatard, J.C., Chollet, D., and Millet, G. 1998.
6. Robson, P. 2003. Elucidating the unexplained Performance and drag during drafting swim-
underperformance syndrome in endurance ming in highly trained triathletes. Medicine and
athletes: The interleukin-6 hypothesis. Sports Science in Sports and Exercise, 30(8), 1276–1280.
Medicine, 33(10), 771–781. 5. Toussaint, H.M., and Beek, P.J. 1992. Biome-
7. Urhausen, A., Gabriel, H., and Kindermann, W. chanics of competitive front crawl swimming.
1995. Blood hormones as markers of training Sports Medicine, 13(1), 8–24.
stress and overtraining. Sports Medicine, 20, 6. Maglischo, E. 2003. Swimming fastest. Cham-
251–276. paign, IL: Human Kinetics.
8. Shephard, R.J., and Shek, P.N. 1998. Acute 7. Sterkel, J. 2001. Long and short range planning.
and chronic over-exertion: Do depressed In D. Hannula and N. Thornton (Eds.), The
immune responses provide useful markers? swim coaching bible (pp. 107–108). Champaign,
International Journal of Sports Medicine, 19(3), IL: Human Kinetics.
159–171. Review.
8. Byrn, G., and Friel, J. 2009. Going long (2nd ed.).
9. Budgett, R., Newsholme, E., Lehmann, M., Boulder, CO: VeloPress.
Sharp, C., Jones, D., Peto, T., Collins, D.,
Nerurkar, R., and White, P. 2000. Redefining 9. Naemi, R., Easson, W.J., and Sanders, R.H. 2010.
the overtraining syndrome as the unex- Hydrodynamic glide efficiency in swimming.
plained underperformance syndrome. British Journal of Science and Medicine in Sport, 13(4),
Journal of Sports Medicine, 34(1), 67–68. 444–451. Epub 2009 Jun 18.
10. Armstrong, L.E., and VanHeest, J.L. 2002. The 10. Robertson, E., Pyne, D., Hopkins, W., and Anson,
unknown mechanism of the overtraining J. 2009. Analysis of lap times in international
syndrome: Clues from depression and psy- swimming competitions. Journal of Sports
Science, 27(4), 387–395.
612 } References

11. Zaidi, H., Taiar, R., Fohanno, S., and Polidori, 23. Gatta, G., Ditroilo, M., Sisti, D., Cortesi, M.,
G. 2008. Analysis of the effect of swimmer's Benelli, P., and Bonifazi, M. 2008. The assess-
head position on swimming performance ment of path linearity in swimming: A pilot
using computational fluid dynamics. Journal study. International Journal of Sports Medicine,
of Biomechanics, 41(6), 1350–1358. 29(12), 959–964. Epub 2008 Jun 11.
12. Tanaka, H., Costill, D.L., Thomas, R., Fink, 24. Wa k ayo sh i, K., Yo sh ida, T., I k ut a, Y.,
W.J., and Widrick, J.J. 1993. Dry-land resistance Mutoh, Y., and Miyashita, M. 1993. Adapta-
training for competitive swimming. Medicine tions to six months of aerobic swim train-
and Science in Sports and Exercise, 25(8), 952–959. ing. Changes in velocity, stroke rate, stroke
13. Trappe, S.W., and Pearson, D.R. 1994. Effects length and blood lactate. International Journal
of weight assisted dry-land strength training of Sports Medicine, 14(7), 368–372.
on swimming performance. Journal of Strength 25. Goldsmith, W. 2011, August 1. Mental skills
and Conditioning Research, 8(4), 209–275. training in swimming: A new approach. Mes-
14. Tanaka, H., and Swensen, T. 1998. Impact of sage posted to www.swimcoachingbrain.com
resistance training on endurance performance.
A new form of cross-training. Sports Medicine, Chapter 26 Bike Base Building
25(3), 191. 1. Seiler, S., and Tonnessen, E. 2009. Intervals,
15. Koop, S., and Martin, G.L. 1983. Evaluation threshold, and long slow distance: The role of
of a coaching strategy to reduce swimming intensity and duration in endurance training.
stroke errors with beginning age-group swim- Sportscience, 13, 32–53.
mers. Journal of Applied Behavior Analysis, 16(4), 2. St. Pierre, A., and Henderson, N. 2009. Labora-
447–460. tory performance evaluations, time trial per-
16. Moreno, F.J., Saavedra, J.M., Sabido, R., Luis, V., formance, and training intensity distribution
and Reina, R. 2006. Visual search strategies of in elite masters cyclists. Medicine and Science in
experienced and non-experienced swimming Sports and Exercise, 41(6), 461–462.
coaches. Perceptual and Motor Skills, 103(3), 3. National Strength and Conditioning Asso-
861–872. ciation. 2012. Developing endurance, edited by
17. Issurin, V. 2008. The workout: General con- Reuter, B. Champaign, IL. Human Kinetics.
cepts and structure guidelines. In M. Yessis 4. Allen, H., and Coggan, A. 2010. Training and
(Ed.), Block periodization (pp. 62–63). Michigan: racing with a power meter (2nd ed.). Boulder,
Ultimate Athlete Concepts. CO: VeloPress.
18. Arnett, M.G. 2001. The effect of a morning 5. Wilmore, J., Stanforth, P., Gagnon, J., Rice, T.,
and afternoon practice schedule on morning Mandel, S., Leon, A., Rao, D.C., Skinner, J., and
and afternoon swim performance. Journal Bouchard, C. 2001. Cardiac output and stroke
of Strength and Conditioning Research, 15(1), volume changes with endurance training: The
127–131. HERITAGE Family Study. Medicine and Science
19. Peeling, P., and Landers, G. 2009. Swimming in Sports and Exercise, 33(1), 99–106.
intensity during triathlon: A review of cur-
rent research and strategies to enhance race Chapter 27 Run Base Building
performance. Journal of Sports Sciences, 27(10), 1. Midgley, A.W., McNaughton, L.R., and Jones,
1079–1085. A.M. 2007. Training to enhance the physio-
20. Chatard, J.C., and Wilson, B. 2003. Drafting logical determinants of long-distance running
distance in swimming. Medicine and Science in performance: Can valid recommendations be
Sports and Exercise, 35(7), 1176–1181. given to runners and coaches based on current
21. Chollet, D., Hue, O., Auclair, F., Millet, G., and scientific knowledge? Sports Medicine, 37(10),
Chatard, J.C. 2000. The effects of drafting on 857–880. doi: 37103 [pii]
stroking variations during swimming in elite 2. Gaskill, S.E., Serfass, R.C., Bacharach, D.W.,
male triathletes. European Journal of Applied and Kelly, J.M. 1999. Responses to training in
Physiology, 82(5–6), 413–417. cross-country skiers. Medicine and Science in
22. Basset, D.R. Jr., Flohr, J., Duey, W.J., Howley, Sports and Exercise, 31(8), 1211–1217.
E.T., and Pein, R.L. 1991. Metabolic responses 3. Esteve-Lanao, J., Foster, C., Seiler, S., and
to drafting during front crawl swimming. Lucia, A. 2007. Impact of training intensity
Medicine and Science in Sports and Exercise, 23(6), distribution on performance in endurance
744–747. athletes. Journal of Strength and Condition-
References | 613

ing Research, 21(3), 943–949. doi: R-19725 and Exercise, 41(9), 1797–1807. doi: 10.1249/
[pii]10.1519/R-19725.1 MSS.0b013e3181a154f3
4. Esteve-Lanao, J., San Juan, A.F., Earnest, C.P., 13. Miranda, F., Simão, R., Rhea, M., Bunker, D.,
Foster, C., and Lucia, A. 2005. How do endur- Prestes, J., Leite, R.D., and Novaes, J. 2011.
ance runners actually train? Relationship Effects of linear vs. daily undulatory peri-
with competition performance. Medicine and odized resistance training on maximal and
Science in Sports and Exercise, 37(3), 496–504. submaximal strength gains. Journal of Strength
doi: 00005768-200503000-00022 [pii] and Conditioning Research, 25(7), 1824–1830. doi:
5. Seiler, K.S., and Kjerland, G. 2006. Quantify- 10.1519/JSC.0b013e3181e7ff75
ing training intensity distribution in elite 14. Prestes, J., Frollini, A.B., de Lima, C., Donatto,
endurance athletes: Is there evidence for an F.F., Foschini, D., de Cássia Marqueti, R., and
“optimal” distribution? Scandinavian Journal of Fleck, S.J. 2009. Comparison between linear
Medicine and Science in Sports, 16(1), 49–56. doi: and daily undulating periodized resistance
SMS418 [pii]10.1111/j.1600-0838.2004.00418.x training to increase strength. Journal of Strength
6. Seiler, S., Haugen, O., and Kuffel, E. 2007. and Conditioning Research, 23(9), 2437–2442. doi:
Autonomic recovery after exercise in trained 10.1519/JSC.0b013e3181c03548
athletes: Intensity and duration effects. Medi- 15. Rhea, M.R., Ball, S.D., Phillips, W.T., and Bur-
cine and Science in Sports and Exercise, 39(8), kett, L.N. 2002. A comparison of linear and
1366–1373. doi: 00005768-200708000-00020 daily undulating periodized programs with
[pii]10.1249/mss.0b013e318060f17d equated volume and intensity for strength.
7. Fiskerstrand, A., and Seiler, K.S. 2004. Train- Journal of Strength and Conditioning Research,
ing and performance characteristics among 16(2), 250–255.
Norwegian international rowers 1970–2001. 16. Rhea, M.R., Phillips, W.T., Burkett, L.N., Stone,
Scandinavian Journal of Medicine and Science in W.J., Ball, S.D., Alvar, B.A., and Thomas, A.B.
Sports, 14(5), 303–310. doi: SMS370 [pii]10.1046/ 2003. A comparison of linear and daily undu-
j.1600-0838.2003.370.x lating periodized programs with equated
8. Enoksen, E., Shalfawi, S.A., and Tønnessen, E. volume and intensity for local muscular
2011. The effect of high- vs. low-intensity train- endurance. Journal of Strength and Conditioning
ing on aerobic capacity in well-trained male Research, 17(1), 82–87.
middle-distance runners. Journal of Strength 17. Vanni, A.C., Meyer, F., da Veiga, A.D., and
and Conditioning Research, 25(3), 812–818. doi: Zanardo, V.P. 2010. Comparison of the effects of
10.1519/JSC.0b013e3181cc2291 two resistance training regimens on muscular
9. Yamamoto, L.M., Lopez, R.M., Klau, J.F., Casa, and bone responses in premenopausal women.
D.J., Kraemer, W.J., and Maresh, C.M. 2008. The Osteoporosis International, 21(9), 1537–1544. doi:
effects of resistance training on endurance 10.1007/s00198-009-1139-z
distance running performance among highly 18. Vøllestad, N.K., and Blom, P.C. 1985. Effect of
trained runners: A systematic review. Journal varying exercise intensity on glycogen deple-
of Strength and Conditioning Research, 22(6), tion in human muscle fibres. Acta Physiologica
2036–2044. doi: 10.1519/JSC.0b013e318185f2f0 Scandinavica, 125(3), 395–405.
10. Apel, J.M., Lacey, R.M., and Kell, R.T. 2011. A 19. Buist, I., Bredeweg, S.W., van Mechelen, W.,
comparison of traditional and weekly undu- Lemmink, K.A., Pepping, G.J., and Diercks,
lating periodized strength training programs R.L. 2008. No effect of a graded training
with total volume and intensity equated. Jour- program on the number of running-related
nal of Strength and Conditioning Research, 25(3), injuries in novice runners: A randomized
694–703. doi: 10.1519/JSC.0b013e3181c69ef6 controlled trial. American Journal of Sports
11. Buford, T.W., Rossi, S.J., Smith, D.B., and Medicine, 36(1), 33–39. doi: 0363546507307505
Warren, A.J. 2007. A comparison of periodiza- [pii]10.1177/0363546507307505
tion models during nine weeks with equated 20. Jobson, S.A., Passfield, L., Atkinson, G., Barton,
volume and intensity for strength. Journal G., and Scarf, P. 2009. The analysis and utiliza-
of Strength and Conditioning Research, 21(4), tion of cycling training data. Sports Medicine,
1245–1250. doi: R-20446 [pii]10.1519/R-20446.1 39(10), 833–844. doi: 3 [pii] 10.2165/11317840-
12. Kok, L.Y., Hamer, P.W., and Bishop, D.J. 2009. 000000000-00000
Enhancing muscular qualities in untrained 21. Busso, T. 2003. Variable dose-response
women: Linear versus undulating peri- relationship between exercise training and
odization. Medicine and Science in Sports performance. Medicine and Science in Sports
614 } References

and Exercise, 35(7), 1188–1195. doi: 10.1249/01. Chapter 30 Half-Ironman


MSS.0000074465.13621.37
1. Britt, R. 2010. Easiest half-Ironman 70.3? Hard-
22. Esteve-Lanao, J., Rhea, M.R., Fleck, S.J., and est? RunTri’s ranking of best half-Ironman races.
Lucia, A. 2008. Running-specific, periodized Retrieved from www.runtri.com/2010/08/
strength training attenuates loss of stride toughesthardest-vs-easiestfastest-half.html
length during intense endurance running. Jour-
nal of Strength and Conditioning Research, 22(4), 2. Seiler, S., and Espen, T. 2009. Intervals, thresh-
1176–1183. doi: 10.1519/JSC.0b013e31816a861f olds, and long slow distance: The role of
intensity and duration in endurance training.
23. Klein, R.M., Potteiger, J.A., and Zebas, C.J. 1997. Sportscience, 13, 32–53. Retrieved from www.
Metabolic and biomechanical variables of two sportsci.org/2009/ss.htm
incline conditions during distance running.
Medicine and Science in Sports and Exercise, 3. Bernhardt, G. 2007. Training plans for multi-
29(12), 1625–1630. sport athletes (2nd ed.). Boulder, CO: VeloPress.
4. Burtscher, M., et al. 2006. Effects of short-term
Chapter 28 Sprint acclimatization to altitude (3200 m) on aerobic
and anaerobic exercise performance. Inter-
1. Bernhardt, G. 2004. Triathlon training basics.
national Journal of Sports Medicine, 27, 629–635.
Boulder, CO: VeloPress.
2. Bernhardt, G. 2007. Training plans for multi- Chapter 31 Ironman
sport athletes (2nd ed.). Boulder, CO: VeloPress.
1. Britt, R. 2011. How much time does it take to
3. Burtscher, M., et al. 2006. Effects of short-
finish an Ironman triathlon? Average Iron-
term acclimatization to altitude (3200 m) on
man finish times. Retrieved from www.runtri.
aerobic and anaerobic exercise performance.
com/2011/06/how-long-does-it-take-to-finish-
International Journal of Sports Medicine, 27,
ironman.html
629–635.
2. Bernhardt, G. 2007. Training plans for multi-
Chapter 29 Olympic sport triathletes (2nd ed.). Boulder, CO: Velo-
Press.
1. Bernhardt, G. 2007. Training plans for multi-
sport triathletes (2nd ed). Boulder, CO: Velo- 3. Seiler, S., and Espen, T. 2009. Intervals, thresh-
Press. olds, and long slow distance: The role of
intensity and duration in endurance training.
2. Bernhardt, G. 2004. Triathlon training basics. Sportscience, 13, 32–53. Retrieved from www.
Boulder, CO: VeloPress. sportsci.org/2009/ss.htm
3. Bernhardt, G., and Hansen, N. 2011. Swim
workouts for triathletes (2nd ed). Boulder, CO: Chapter 32 Duathlon
VeloPress.
1. International Triathlon Union website, www.
4. International Triathlon Union, listing of fed- triathlon.org
erations. Retrieved from www.triathlon.org/
federations/ 2. USA Triathlon website, www.usatriathlon.org
5. Le Meur, Y., et al. 2011. Relationships between 3. USA Triathlon. 2011. National Championship
triathlon performance and pacing strategy Results. Retrieved from http://ultramaxtri.
during the run in an international competition. com/t im i ng/2011/2011_USAT_Du_Nat/
International Journal of Sports and Physiology USATDuN_results.html
Performance, 6(2), 183–194. Retrieved from 4. Bernhardt, G. 2007. Training plans for multi-
www.ncbi.nlm.nih.gov/pubmed/21725104 sport athletes (2nd ed.). Boulder, CO: VeloPress.
6. Hausswirth, C., et al. 2009. Pacing strategy 5. Seiler, S., and Espen, T. 2009. Intervals, thresh-
during the initial phase of the run in triathlon: olds, and long slow distance: The role of
Influence on overall performance. European intensity and duration in endurance training.
Journal of Applied Physiology, 108(6), 1115– Sportscience, 13, 32–53. Retrieved from www.
1123. Retrieved from www.ncbi.nlm.nih.gov/ sportsci.org/2009/ss.htm
pubmed/20024576 6. Burtscher, M. et al. 2006. Effects of short-
7. Burtscher, M., et al. 2006. Effects of short- term acclimatization to altitude (3200 m) on
term acclimatization to altitude (3200 m) on aerobic and anaerobic exercise performance,
aerobic and anaerobic exercise performance. International Journal of Sports Medicine, 27,
International Journal of Sports Medicine, 27, 629–635.
629–635.
References | 615

Chapter 33 Combination Workout plete triathlon book. New York: Grand Central.
Training 13. Ingraham, S.J. 2003. The role of flexibility in
injury prevention: Have we stretched the
1. Bonacci, J., et al. 2011. Neuromuscular control
truth? MN Med, 86(5), 58–61.
and running economy is preserved in elite
international triathletes after cycling. Sports 14. www.grastontechnique.com
Biomechanics, 10(1), 59–71. Retrieved from 15. www.activerelease.com
www.ncbi.nlm.nih.gov/pubmed/21560752 16. Leahy, P., and Michael, D.C. 2011. Founder,
2. Bonacci, J., et al. 2010. Altered movement active release technique. Personal communica-
patterns but not muscle recruitment in mod- tion.
erately trained triathletes during running 17. Hall, H. 2009. The Graston technique: Inducing
after cycling. Journal of Sports Science, 28(13), microtrauma with instruments. Science Based
1477–87. Retrieved from www.ncbi.nlm.nih. Medicine, Dec. 29, www.sciencebasedmedi-
gov/pubmed/20945251 cine.org/index.php/the-graston-technique-
3. Chapman, A.R., et al. 2008. Does cycling effect inducing-microtrauma-with-instruments
motor coordination of the leg during running 18. Holland, T. 2011. The 12-week triathlete (2nd
in elite triathletes? Journal of Science and ed.). Beverly, MA: Fair Winds Press.
Medicine in Sport, 11(4), 371–80. Epub 2007
Apr 26. Retrieved from www.ncbi.nlm.nih. Chapter 35 Triathlon Injuries and
gov/pubmed/17466592
Preventive Measures
Chapter 34 Triathlete Body 1. Burns, J., Keenan, A.M., and Redmond, A.C.
Maintenance and Medical Care 2003. Factors associated with triathlon-related
overuse injuries. Journal of Orthopaedic and
1. Friel, J. 2009. The triathlete’s training bible (3rd Sports Physical Therapy, 33(4), 177–184.
ed.). Boulder, CO: VeloPress.
2. Vleck, V.E., Bentley, D.J., Millet, G.P., and
2. Pruitt, A. 2006. Andy Pruitt’s complete medical Cochrane, T. 2010. Triathlon event distance spe-
guide for cyclists. Boulder, CO: VeloPress. cialization: Training and injury effects. Journal
3. Scott, D. 1986. Dave Scott’s triathlon training. of Strength and Conditioning Research, 24(1),
New York: Simon and Schuster. 30–36.
4. Noakes, T. 2001. Lore of running (4th ed.). 3. Heinlein, S.A., and Cosgarea, A.J. 2010. Bio-
Champaign, IL: Human Kinetics. mechanical considerations in the competitive
5. Dale, P. 2011. Definition of muscular strength swimmer’s shoulder. Sports Health: A Multi-
and endurance. www.livestrong.com/arti- disciplinary Approach, 2(6), 519–525.
cle/106346 4. Clarsen, B., Krosshaug, T., and Bahr, R. 2010.
6. Burns, J. 2003. Factors associated with triath- Overuse injuries in professional road cyclists.
lon-related overuse injuries. Journal of Ortho- American Journal of Sports Medicine, 38(12),
pedic Sports Physical Therapy, 33(4), 177–184. 2494–2501.
7. Korkia, P. 1994. An epidemiological investiga- 5. Villavicencio, A.T., Hernández, T.D., Bur-
tion of training and injury patterns in British neikiene, S., and Thramann, J. 2007. Neck pain
triathletes. British Journal of Sports Medicine, in multisport athletes. Journal of Neurosur-
28(3), 191–196. gery: Spine 7(4), 408–413.
8. O’Toole, M. 1989. Overuse injuries in ultra 6. Bini, R., Hume, P.A., and Croft, J.L. 2011. Effects
endurance athletes. American Journal of of bicycle saddle height on knee injury risk and
Sports Medicine, 17, 514–518. cycling performance. Sports Medicine, 41(6),
463–476.
9. Meehan, H. 2000. Do triathletes really have
fewer injuries, and which ones get hurt? www. 7. Gregersen, C.S., Hull, M.L., and Hakansson,
sportsinjurybulletin.com. N.A. 2006. How changing the inversion/
eversion foot angle affects the nondriving
10. Jenkins, J. 2010. Flexibility for runners. Clinics
intersegmental knee moments and the rela-
of Sports Medicine, 29, 365–377.
tive activation of the vastii muscles in cycling.
11. McCann, P. 2010. Flexibility in the athlete. Journal of Biomechanical Engineering, 128(3),
University of Virginia Foundation, University 391–398.
of Virginia. Personal communication.
8. Jarboe, N.E., and Quesada, P.M. 2003. The
12. Fitzgerald, M. 2003. Staying healthy. The com- effects of cycling shoe stiffness on forefoot
616 } References

pressure. Foot and Ankle International, 24(10), 19. Arendse, R.E., Noakes, T.D., Azevedo, L.B.,
784–788. Romanov, N., Schwellnus, M.P., and Fletcher,
9. Van Gent, R.N., Siem, D., Van Middelkoop, M., G. 2004. Reduced eccentric loading of the knee
Van Os, A.G., Bierma-Zeinstra, S.M.A., and with the pose running method. Medicine and
Koes, B.W. 2007. Incidence and determinants Science in Sports and Exercise, 36, 272–277.
of lower extremity running injuries in long 20. Williams, D.S., McClay, I.S., and Hamill, J. 2001.
distance runners: A systematic review. British Arch structure and injury patterns in runners.
Journal of Sports Medicine, 41(8), 469–480. Clinical Biomechanics, 16(4), 341–347.
10. Bonacci, J., Chapman, A., Blanch, P., and Vicen- 21. Milner, C.E., Hamill, J.D., and Davis, I.S. 2010.
zino, B. 2009. Neuromuscular adaptations to Distinct hip and rearfoot kinematics in female
training, injury and passive interventions: runners with a history of tibial stress fracture.
Implications for running economy. Sports Journal of Orthopedic and Sports Physical
Medicine, 39(11), 903–921. Therapy, 40(2), 59–66.
11. Franz, J.R., Paylo, K.W., Dicharry, J., Riley, 22. Ryan, M.B., Valiant, G.A., McDonald, K., and
P.O., and Kerrigan, D.C. 2009. Changes in the Taunton, J.E. 2009. The effect of three different
coordination of hip and pelvis kinematics with levels of footwear stability on pain outcomes in
mode of locomotion. Gait and Posture, 29(3), women runners: A randomized control trial.
494–498. British Journal of Sports Medicine, 45, 715–721.
12. Myers, W.C., Kahan, D.M., Joseph, T., Butry- 23. Kong, P.W., Candelaria, N.G., and Smith, D.R.
mowicz, A., Poor, A.E., and Zoga, A.C. 2011. 2009. Running in new and worn shoes: A com-
Current analysis of women athletes with pelvic parison of three types of cushioning footwear.
pain. Medicine and Science in Sports and Exer- British Journal of Sports Medicine, 43, 745–749.
cise, 43(8), 1387–1393.
13. Heiderscheit, B.C., Chumanov, E.S., Michalski, Chapter 36 Triathlon Injury Recovery
M.P., Wille, C.M., and Ryan, M.B. 2011. Effects Techniques
of step rate manipulation on joint mechanics 1. Escamilla, R.F., Lewis, C., Bell, D., Bramblet, G.,
during running. Medicine and Science in Daffron, J., Lambert, S., Pecson, A., Imamura, R.,
Sports and Exercise, 43(2), 296–302. Paulos, L., and Andrews, J.R. 2010. Core muscle
14. Biewener, A.A., Farley, C.T., Roberts, T.J., and activation during Swiss ball and traditional
Temaner, M. 2004. Muscle mechanical advan- abdominal exercises. Journal of Orthopedic
tage of human walking and running: Impli- and Sports Physical Therapy, 40(5), 265–276.
cations for energy cost. Journal of Applied 2. Magnussen, R.A., Dunn, W.R., and Thomson,
Physiology, 97(6), 2266–2274. B.A. 2009. Nonoperative treatment of mid-
15. Lun, V., Meeuwisse, W.H., Stergiou, P., and portion Achilles tendinopathy: A systematic
Stefanyshyn, D. 2004. Relation between run- review. Clinical Journal of Sports Medicine,
ning injury and static lower limb alignment in 19(1), 55–64.
recreational runners. British Journal of Sports 3. Franca, F.R., Burke, T.N., Hanada, E.S., and
Medicine, 38(5), 576–580. Marques, A.P. 2010. Segmental stabilization
16. Chumanov, E.S., Schache, A.G., Heiderscheit, and muscular strengthening in chronic low
B.C., and Thelen, D.G. 2011. Hamstrings are back pain: A comparative study. Clinics, 65(10),
most susceptible to injury during the late 1013–1017.
swing phase of sprinting. British Journal of 4. Riley, P.O., Franz, J., Dicharry, J., and Kerrigan,
Sports Medicine, 45(9), 684–685. C.D. 2010. Changes in hip joint muscle tendon
17. Earl, J.E., and Hoch, A.Z. 2011. A proximal lengths with mode of locomotion. Gait and
strengthening program improves pain, func- Posture, 31(2), 279–283.
tion, and biomechanics in women with patello- 5. McHugh, M.P., and Cosgrave, C.H. 2010. To
femoral pain syndrome. American Journal of stretch or not to stretch: The role of stretching
Sports Medicine, 29(1), 154–163. in injury prevention and performance. Scan-
18. Lieberman, D.E., Venkadesan, M., Werbel, dinavian Journal of Medicine and Science in
W.A., Daoud, A.I., D’Andrea, S., Davis, I.S., Sports, 20, 169–181.
Mang’Eni, R.O., and Pitsiladis, Y. 2010. Foot 6. Haupenthal, A., Ruschel, C., Hubert, M., de
strike patterns and collision forces in habitu- Brito Fontana, H., and Roesler, H. 2010. Loading
ally barefoot versus shod runners. Nature, forces in shallow water running at two levels of
463(7280), 531–535. immersion. Journal of Rehabilitation Medicine,
References | 617

42(7), 664–669. exercises for painful Achilles tendinosis: A


7. Wilson, J.K., Sevier, T.L., Helfst, R.H., Honing, randomized trial. British Journal of Sports
E.W., and Thomann, A.L. 2000. Comparison Medicine, 45, 421–428.18. Avella, C.S., and
of rehabilitation methods in the treatment of Smith, R.K. 2010. Ultrasound evaluation of
patellar tendinitis. Journal of Sports Rehabili- stem cell treated tendon injuries in the horse:
tation, 9(4), 304–314. Repair or regeneration. Lancet, 376(9754),
1751–1767.
8. Loghmani, M.T., and Warden, S.J. 2009. Instru-
ment-assisted cross-fiber massage accelerates 18. Avella, C.S., and Smith, R.K. 2010. Ultrasound
knee ligament healing. Journal of Orthopaedic evaluation of stem cell treated tendon injuries
and Sports Physical Therapy, 39(7), 506–514. in the horse: Repair or regeneration. Lancet,
376(9754), 1751–1767.
9. González-Iglesias, J., Fernández-De-Las-Peñas,
C., Cleland, J.A., and Gutiérrez-Vega, M.D.R. 19. Noehren, B., Scholz, J., and Davis, I. 2011.
2009. Thoracic spine manipulation for the The effect of real-time gait retraining on hip
management of patients with neck pain: A ran- kinematics, pain and function in subjects with
domized clinical trial. Journal of Orthopaedic patellofemoral pain syndrome. British Journal
and Sports Physical Therapy, 39(1), 20–27. of Sports Medicine, 45, 691–696.
10. Brummitt, J. 2008. The role of massage in sports Chapter 37 Energy Needs, Sources,
performance and rehabilitation: Current evi-
dence and future direction. North American
and Utilization
Journal of Sports Physical Therapy, 3(1), 7–21. 1. Arnos, P.M., Sowash, J., and Andres, F.F. 1997.
11. Wiltshire, E.V., Poitras, V., Pak, M., Hong, T., Fat oxidation at varied work intensities using
Rayner, J., and Tschakovsky, M.E. 2010. Mas- different exercise modes. Medicine and Science
sage impairs postexercise muscle blood flow in Sports and Exercise, 29, S199.
and “lactic acid” removal. Medicine and Sci- 2. Jeukendrup, A.E. 2004. Carbohydrate intake
ence in Sports and Exercise, 42(6), 1062–1071. during exercise and performance. Nutrition,
12. Gorski, T., Cadore, L.E., Pinto, S.S., Silva, E.M., 20, 669–677.
Correa, S.C., Beltrami, G.F., and Kruel, M.L.F. 3. Acheton, J., Gleeson, M., and Jeukendrup, A.E.
2011. The use of NSAIDs in triathletes: Preva- 2002. Determination of the exercise intensity
lence, level of awareness and reasons for use. that elicits maximal fat oxidation. Medicine
British Journal of Sports Medicine, 45, 85–90. and Science in Sports and Exercise, 34, 92–97.
13. Baur, H., Hirschmuller, A., Muller, S., and 4. Coyle, E.F. 1995. Substrate utilization during
Mayer, F. 2011. Neuromuscular activity of the exercise in active people. American Journal of
peroneal muscle after foot orthoses therapy in Clinical Nutrition, 61, 968S–979S.
runners. Medicine and Science in Sports and 5. Seebohar, B. 2009. Metabolic efficiency train-
Exercise, 43(8), 1500–1505. ing: Teaching the body to burn more fat. Fuel-
14. Gambito, E.D., Gonzales-Suarez, C.B., Oqui- 4mance.
ñena, T.I., and Agbayani, R.B. 2010. Evidence 6. van Loon, L.J.C. 2004. Use of intramuscular
on the effectiveness of topical nitroglycerin triacylglycerol as a substrate source during
in the treatment of tendinopathies: A system- exercise in humans. Journal of Applied Physi-
atic review and meta-analysis. Archives of ology, 97(4), 1170–1187.
Physical Medicine and Rehabilitation, 91(8),
7. Dunford, M., and Doyle, J.A. 2008. Nutrition
1291–1305.
for sport and exercise. Belmont, CA: Thompson
15. Nichols, A. 2005. Complications associated Higher Education.
with the use of corticosteroids in the treatment
8. Kenney, W.L., and Chiu, P. 2001. Influence of
of athletic injuries. Clinical Journal of Sports age on thirst and fluid intake. Medicine and
Medicine, 15(5), E370. Science in Sports and Exercise, 33(9), 1524–
16. Weir, A., van Schie, H.T.M., Bierma-Zeinstra, 1532.
S.M.A., Verhaar, J.A.N., Weinans, H., and Tol, 9. Horowitz, J.F., Mora-Rodriguez, R., Byerley,
J.L. 2010. Platelet-rich plasma injection for L.O., and Coyle, E.F. 1997. Lipolytic suppression
chronic Achilles tendinopathy. JAMA, 303(2), following carbohydrate ingestion limits fat
144–149. oxidation during exercise. American Journal
17. Yelland, M.J., Sweeting, K.R., Lyftogt, J.A., Ng, of Physiology, 273(4), E768–E775.
S.K., Scuffham, P.A., and Evans, K.A. 2011. 10. Halton, T.L., and Hu, F.B. 2004. The effects of
Prolotherapy injections and eccentric loading high protein diets on thermogenesis, satiety
and weight loss: A critical review. Journal
618 } References

of the American College of Nutrition, 23(5), cise Metabolism, 17(1), 109–123.


373–385. 7. Maughan, R.J. 1991. Fluid and electrolyte loss
and replacement in exercise. Journal of Sports
Chapter 38 Nutrition Periodization Sciences, 9, 117–142.
1. Seebohar, B. 2010. Nutrition periodization for 8. Jeukendrup, A.E., Jentjens, R., and Moseley, L.
athletes (2nd ed.). Boulder, CO: Bull. 2005. Nutritional considerations in triathlon.
2. Burke, L.M., Cox, G.R., Cummings, N.K., and Sports Medicine, 35, 163–181.
Desbrow, B. 2001. Guidelines for daily carbohy- 9. Burke, L.M., Kiens, B., and Ivy, J.L. 2004. Car-
drate intake: Do athletes achieve them? Sports bohydrates and fat for training and recovery.
Medicine, 31(4), 92–97. Journal of Sports Sciences, 22, 15–30.
3. Burke, L.M., Kiens, B., and Ivy, J.L. 2004. Car- 10. Burke, L., and Deakin, V. 2006. Clinical sports
bohydrates and fat for training and recovery. nutrition (3rd ed.). McGraw-Hill.
Journal of Sports Sciences, 22, 15–30.
11. Maughan, R., and Shirreffs, S.M. 2008. Devel-
4. Tarnopolsky, M. 2004. Protein requirements opment of individual hydration strategies for
for endurance athletes. Nutrition, 20, 662–668. athletes. International Journal of Sports Nutri-
5. Rodriguez, N.R., DiMarco, N.M., and Langley, tion and Exercise Metabolism, 18, 457–472.
S. 2009. Nutrition and athletic performance.
Journal of the American Dietetic Association, Chapter 40 Supplements for Triathletes
109(3), 509–527. 1. Burke, L., and Deakin, V. 2006. Clinical sports
6. Campbell, B., Kreider, R.B., Ziegenfuss, T., La nutrition (3rd ed.). McGraw-Hill.
Bounty, P., Roberts, M., Burke, D., Landis, J., 2. Panossian, A., Wikman, G., and Wagner, H.
Lopez, H., and Antonio, J. 2007. International 1999. Plant adaptogens III. Earlier and more
society of sports nutrition position stand: Pro- recent aspects and concepts on their mode of
tein and exercise. Journal of the International action. Phytomedicine 6(4), 287–300.
Society of Sports Nutrition, 4, 8.
3. Artioli, G.G., Gualano, B., Smith, A., Stout, J.,
Chapter 39 Nutrient Timing for and Lancha, A.H. 2010. Role of beta alanine
supplementation on muscle carnosine and
Triathlon Training and Racing exercise performance. Medicine and Science
1. Hargreaves, M., Hawley, J.A., and Jeukendrup, in Sports and Exercise, 42(6), 1162–1173.
A. 2004. Pre-exercise carbohydrate and fat 4. Graham, T.E. 2001. Caffeine and exercise:
ingestion: Effects on metabolism and perfor- Metabolism, endurance and performance.
mance. Journal of Sports Sciences, 22, 31–38. Sports Medicine, 31, 785–807.
2. Tipton, K.D., and Wolfe, R.R. 2004. Protein and 5. Armstrong, L.E., Pumerantz, A.C., Roti, M.W.,
amino acids for athletes. Journal of Sports Sci- Judelson, D.A., Watson, G., Dias, J.C., Sokmen, B.,
ences, 22, 65–79. Casa, D.J., Maresh, C.M., Harris, L., and Kellogg,
3. Sawka, M.N., Burke, L.M., Eichner, E.R., M. 2005. Fluid, electrolyte and renal indices of
Maughan, R.J., Montain, S.J., and Stachenfeld, hydration during 11 days of controlled caffeine
N.S. 2007. Exercise and fluid replacement posi- consumption. International Journal of Sport
tion stand. Medicine and Science in Sports and Nutrition and Exercise Metabolism, 15, 252–265.
Exercise, 39(2), 377–389. 6. Hoffman, J.R., Kang, J., Ratamess, N.A., Jen-
4. Jeukendrup, A.E. 2010. Carbohydrate and nings, P.F., Mangine, G.T., and Faigenbaum,
exercise performance: The role of multiple A.D. 2007. Effect of nutritionally enriched
transportable carbohydrates. Current Opinion coffee consumption on aerobic and anaerobic
in Clinical Nutrition and Metabolic Care, 13(4), exercise performance. Journal of Strength and
452–457. Conditioning Research, 21(2), 456–459.
5. Seebohar, B. 2009. Metabolic efficiency train- 7. Yeo, S.E., Jentjens, R.L., Wallis, G.A., and Jeu-
ing: Teaching the body to burn more fat. Fuel- kendrup, A.E. 2005. Caffeine increases exog-
4mance. enous carbohydrate oxidation during exercise.
6. Luden, N.D., Saunders, M.J., and Todd, M.K. Journal of Applied Physiology, 99, 844–850.
2007. Postexercise carbohydrate-protein- 8. Lopez, R.M., Casa, D.J., McDermott, B.P., Ganio,
antioxidant ingestion decreases plasma M.S., Armstrong, L.E., and Maresh, C.M. 2009.
creatine kinase and muscle soreness. Inter- Does creatine supplementation hinder exercise
national Journal of Sports Nutrition and Exer- heat tolerance or hydration status? A system-
References | 619

atic review with meta-analyses. Journal of Chapter 42 Psychology of Triathlon


Athletic Training, 44(2), 215–223. Training
9. Law, Y.L., Ong, W.S., GillianYap, T.L., Lim, S.C.,
1. Main, L.C., Landers, G.J., Grove, J.R., Dawson,
and Von Chia, E. 2009. Effects of two and five
B., and Goodman, C. 2010. Training patterns
days of creatine loading on muscular strength
and negative health outcomes in triathlon:
and anaerobic power in trained athletes. Jour-
Longitudinal observations across a full com-
nal of Strength and Conditioning Research,
petitive season. Journal of Sports Medicine and
23(3), 906–914.
Physical Fitness, 50(4), 475–485.
Chapter 41 Mental Toughness for 2. Winsley, R., and Matos, N. 2011. Overtraining
Triathlon and elite young athletes. Medicine and Sport
Science, 56, 97–105.
1. Ranganathan, V.K., Siemionow, V., Liu, J.Z., 3. Wadey, R., and Hanton, S. 2008. Basic psycho-
Sahgal, V., and Yue, G.H. 2004. From mental logical skills usage and competitive anxiety
power to muscle power: Gaining strength responses: Perceived underlying mechanisms.
by using the mind. Neuropsychologia, 42(7), Research Quarterly for Exercise and Sport,
944–956. 79(3), 363–373.
2. Dolan, S.H., Houston, M., and Martin, S.B.
2011. Survey results of the training, nutrition, Chapter 43 Mental Skills for Peak
and mental preparation of triathletes: Practi- Triathlon Performance
cal implications of findings. Journal of Sports
Science July 29(10), 1019–1028. 1. Knechtle, B., Wirth, A., and Rosemann, T.
2010. Predictors of race time in male ironman
3. Maslow, A.H. 1943. A theory of human motiva-
triathletes: Physical characteristics, training,
tion. Psychological Review, 50(4), 370–396.
or prerace experience? Perceptual and Motor
4. Stoeber, J., Uphill, M.A., and Hotham, S. 2009. Skills, 111(2), 437–446.
Predicting race performance in triathlon: The
2. Memmert, D. 2011. Creativity, expertise, and
role of perfectionism, achievement goals, and
attention: Exploring their development and
personal goal setting. Journal of Sport and
their relationships. Journal of Sports Sciences,
Exercise Psychology, 31(2), 211–245.
29(1), 93–102.
5. Grand’Maison, K. 2004. What mental skills
3. Ranganathan, V.K., Siemionow, V., Liu, J.Z.,
Ironman triathletes need and want. Journal
Sahgal, V., and Yue, G.H. 2004. From mental
of Excellence, 10, 86–87.
power to muscle power: Gaining strength
by using the mind. Neuropsychologia, 42(7),
944–956.
4. Fontani, G., Migliorini, S., Benocci, R., Fac-
chini, A., Casini, M., and Corradeschi, F. 2007.
Effect of mental imagery on the development
of skilled motor actions. Perceptual and Motor
Skills, 105(3 Pt. 1), 803–826.
This page intentionally left blank.
Index
Note: The italicized f and t following page numbers refer to figures and tables, respectively.

1RM 24, 259-260 affirmations 576


3-to-1 rule 378 age-related declines
10 percent rule 378 flexibility 34
hunger cues 522
training effects on 15
A triathlon performance 49t
AA (arachidonic acid) 527-528 event distance comparison 51-53, 52f
AA phase, in strength training 256, 256t gender differences in 47-48, 47f
acclimation 138 physiological factors 48
acclimatization 138, 139 sport comparisons 49-51, 50f
accumulation mesocycle 315 air pollution 140-141, 141f, 526, 549
acetyl coenzyme A (acetyl CoA) 7-8, 184 Aleve 505
Achilles tendon injuries Alexander, Craig 49t
in cyclists 481 alleles 27-28, 28f
flexibility training effects on 241 Allen, Mark 556-557, 573
injury models 33-35, 34f Alter, Michael J. 240
recovery techniques 499, 508, 509, 510 altitude
in runners 491 in free radical activity 526
acidosis 23, 23f, 182 in race tactics 398, 407, 417, 439
actin 16, 18f, 19-20, 212 training for 12, 138-140
action potentials 19 American College of Sports Medicine
active cool-downs 219-220 flexibility training statement 240
active drag 105 sodium intake guidelines 539
active release technique (ART) 459 amino acids 519
active warm-ups 211 anaerobic energy systems
acute compartment syndrome 509 ATP-CP (immediate) 5-6, 6f, 23, 186, 551-
acute training load (ATL) 319-324, 320f, 335 552
adaptogens 549-550 glycolysis (short-term) 6-7, 7f, 23, 184-185
adenosine diphosphate (ADP) 6, 6f usage ratio in training 270-271
adenosine triphosphate. See ATP anaerobic metabolism 183-185
ADP 6, 6f, 20 anaerobic threshold (AT)
adversity 560 defined 181
aero bars 117, 122-123, 123f gender differences in 40
aerobic capacity. See VO2 max markers of 182-183
aerobic energy system as performance determinant 184, 187-188
physiology of 7-8, 7f, 184 strength training effects on 252
training effects on 22-23 training effects on 14
usage ratio in training 270-271 anaerobic training
aerobic exercise, defining 269-270 base building 273
aerobic metabolism 165-167, 183-185 metabolic adaptations 303
aerobic threshold testing 364, 377 muscular adaptations 23, 23f
aerobic training. See endurance training speed work as 302
aerodynamics, in cycling anatomical adaptation (AA), in strength train-
bicycle frames 116-117 ing 256, 256t
bike fit adjustments 127-128, 127f anatomical movement limitations 95, 95f
cycling power 73 Anderson, Owen 347-348
drafting 125-127, 126f anemia
handlebar placement 122-123, 123f risk factors 132, 528
rider position 80-82, 80f, 81f types 529-530

621
622 } Index

ankle flexion beach starts 103


exercises for 229-230, 229f, 230f beliefs, in peak performance 555
in running 95, 95f, 226, 227f beta-alanine 550
in swimming 65-66, 65f, 224 beta-carotene 526
ankle injuries bicarbonate 23, 23f
preventive exercises 482-483, 482f, 483f, 493- bicycles
494, 493f, 494f dismounting 444
recovery techniques 506, 508, 510 frames 116-117
from running shoes 133 handlebars 117
types 481-482, 491-493 mounting 443
anticipatory regulation of fatigue 202-203, 205 rider position 80-81, 80f, 81, 81f, 114, 118
antioxidants 526 seat tube angle 82, 97-98, 97f, 114-115, 115f
aorta 10 tri-bikes vs. road bikes 115-116
arachidonic acid (AA) 527-528 weight 73, 124
arch crunches 483, 483f wheels 117-118, 125
arm pull, in swimming bike fit
mistakes and corrections 66-67, 66f aerodynamics and 127-128, 127f
S pull pattern 60-61, 60f handlebar placement 122-123, 123f
Armstrong, Lance 128 in injury prevention 468, 470, 473, 478
ART 459 rider position for 118
artificial sweeteners 517 saddle fore–aft position 120, 121f
ASTYM 503 saddle height 119-120
asymmetry, in pedaling mechanics 76-77, 76f bike shop pros 455
AT. See anaerobic threshold bike-to-run transition
athletica pubalgia (sports hernia) 473, 487, 509 bicycle geometry effects on 82, 97-98, 97f,
athletic trainers 455 114-115, 115f
ATL 319-324, 320f, 335 cycling perspective 83-84
ATP running economy and 84, 173-174, 178
in energy production 4-8, 5f, 6f, 7f, 184 running performance factors in 96-98
molecular structure of 4, 4f in sprint events 397-398
in muscle contraction 20 tips for 444
ATP-CP system bilateral breathing drill 359
creatine supplements for 551-552 biological hunger 522
energy production in 5-6, 6f, 186 biomechanics. See cycling biomechanics; running
training effects on 23 biomechanics; swimming biomechanics
augmented soft-tissue mobilization (ASTYM) 503 block periodization 315-317, 316f
avoidance, in obsession profile 565 blood flow
in heat tolerance 201
B warm-up effects on 212
blood pressure 15
back extensions blood shunting response 515
over physioball 470, 470f blood sugar control 516, 517, 522
standing 467, 467f blood volume
Badmann, Natascha 39, 49t heat tolerance and 15
balance, in swimming training effects on 13, 366
as function of kick 62, 62f body composition (BMI)
mistakes and corrections 66-67, 66f of elite thriathletes 29
balanced lifestyle 561-562, 564 genetic effects on 29-30
ballerina drill 283 manipulating 521-523
barefoot running body fat. See also dietary fat; fat metabolism
injury risk 96, 491 in body composition 29
running efficiency and 88, 92-94, 93f effects on performance 29
vs. shod running 133-134, 134f fat stores 519
for skill development 382 gender differences in 40, 44
base building. See also cycling base building; run training effects on 15
base building; swim base building body mass index. See body composition (BMI)
aerobic 269-273, 271t, 272f body roll, in swimming 67, 68f
anaerobic 273 body temperature. See also heat and humidity
nutrition during 530, 531t, 533t, 537 in failure concept of fatigue 200
vs. specific training 274-277, 275t fluid intake and 539
in sprint training 390 as warm-up risk 213-214
Index | 623

body type, in swim performance 354 cardiac output


body weight in dehydration 539
as cycling advantage 73, 124 training effects on 13, 366
manipulating 521-523, 531, 531t, 533t cardiopulmonary system
strength training effects on 252, 253 anatomy of 9-10, 10f
training effects on 15 oxygen transport in 10-11
Bolt, Usain 571 term defined 9
Bompa, Tudor O. 249 training effects on 12-15
bone death 507 carnosine 550
brain function carpal tunnel syndrome 503, 505
heat effects on 201-202 cartilage injuries 509, 511
muscle activation reserve and 203, 205 catch-up drill 280, 280f
breathing exercises Celiac disease 528
for focus 578-579 center of mass, in swimming 57-58, 58f
in swimming 359 central fatigue 342-343
brick workouts central nervous system (CNS), in swimming 356-357
defined 441 chair lower-back stretch 236, 236f
for duathletes 432-434 chest press 263, 263f
frequency and volume of 446-447 chia seeds 520, 527
in Olympic-distance training 402 Child’s Pose 236, 236f
as running performance factor 97 chiropractic treatments 502-505, 507
term etymology 446 chondromalcia patella 489
bridge Chrabot, Matt 536, 538
with long arc quad 471, 471f, 500, 500f chronic exertional compartment syndrome 492
plank 245, 245f chronic training load (CTL) 319-324, 320f, 321f,
single-leg 472, 472f, 500, 500f 322f, 324t, 335, 336f
British Triathlon Federation 104 clams with exercise band 488, 488f
Brownlee, Alistair 28 clock diagrams 75, 75f
buoyancy, in swimming 44, 61-62, 63 CM phase, in strength training 259, 259t
buoy press 281 CNS, in swimming 356-357
burnout, avoiding 366 coaches, technology use by 150, 157
butterfly stroke 361 Coggan. A.R. 319
cold therapy 220-221, 499
Collington, Kevin 536, 537, 541
C combo workouts
cadence defined 441
cycling rate and drills 281-282, 367 for duathletes 432-434, 447-448
running drill 283 for triathletes 445-446
caffeine 550-551 commitment 557-558, 559
calcaneal cuboid syndrome 492, 506 compartment syndromes 492, 509
calcium supplements 547-548 competition maintenance (CM), in strength
calorie counting 521 training 259, 259t
Calvert, T.W. 318 complex carbohydrates 516-517
canted-surface running 133 compliments, accepting 575
carbohydrate compression machines 502
in blood sugar control 522 CompuTrainer system 283
daily intake ranges concentration 560, 561, 576-579
base period 531, 531t, 533t concentric muscle actions 83-84
general guidelines 526 consciousness, expanding 571
off-season 533 consistency, in disciplined athletes 559
specific training period 532, 532t, 533t consumerlab.com 544
glycemic index and response 517-519 contractile filaments 16, 18f, 19-20
training intake ranges contraction phase, of muscle contraction 20, 21f
posttraining 540 contrast therapy 220-221
pretraining 535-537 cool-downs
during training 537-538 active 219-220
types 516-517 passive 220-221
carbohydrate metabolism core muscles 241, 242f
measurement of 165-166, 166t core strength
in metabolic efficiency 515-516 exercises for 244-247, 245f, 245t, 246f, 247f,
carbon monoxide 140 486, 486f
624 } Index

core strength, continued platform pedal attachment 444, 445f


in injury prevention 242-244 preparing for transition 442-443, 443f
in injury recovery 498, 498f cycling training
performance and 244 Olympic-distance 402
corner pec stretch 234, 234f skill drills 281-283
corn oil 527 sprint distance 391
corn syrup 517 for VO2max improvement 168-169
corticosteroid injections 507 cyclist palsy 505
cortisol–testosterone ratio 343
cortisone shots 507
coupling phase, of muscle contraction 19, 19f D
C-reactive protein (CRP) 527 Dave Scott’s Triathlon Training (Scott) 456
creatine kinase 343 decompression 507
creatine phosphate (CP). See ATP-CP system dehydration 539, 550. See also hydration
creatine supplements 551-552 delayed-onset muscle soreness (DOMS). See
Crohn’s disease 528 muscle soreness, preventing
cross bridging, in muscle contraction 20 delta efficiency, vs. gross efficiency 175
cross-training 369, 421, 565 depression 346-347
crotch height method, in bike fitting 119 desire 560
CRP 527 determination 559, 576-579
CTL 319-324, 320f, 321f, 322f, 324t, 335, 336f detraining 299-300, 316-317
cutoff times dexamethasone 507
in Ironman events 419 dietary fat. See also body fat; fat metabolism
water conditions for 106 daily intake ranges
cycling. See also bicycles; bike fit; injury descriptions base period 531, 531t, 533t
acceleration 124-125 general guidelines 526
age-related decline in 49-51, 50f off-season 533
climbing and descending 124, 125f specific training period 532, 532t, 533t
drafting 125-127, 126f MCT supplements 551
effects on running economy 84, 173-174, 178 training intake ranges
flexibility effects on 225 posttraining 540
gender differences in 43f, 44-45, 44f, 45t, 46-47, pretraining 536
46f during training 538
governing bodies 113 types 519-520
heart rate zones 145, 146t-147t dietary supplements, defined 546. See also
multisport issues in 83, 177 supplements
power formulas 71-72 Dietary Supplements Health and Education Act
skill training drills 281-283 (DSHEA) 543-544
strength training effects on 250-252, 253 disaccharides 516
in swim-to-bike transition 443 disc injuries 507, 509, 511
taper in 329, 330 discipline, mental 558-560
cycling base building distractions 577
intensity and duration in 363-366 diversity, in training 565
sample schedule 370 diving starts 103
volume in 366, 370 dizziness
workout guidelines 369-370 in combo workouts 445-446
workout structure 367-368 postexercise 220
cycling biomechanics DNA. See genetics
aerodynamics in 80-82, 80f, 81f DNA markers, defined 28
in bike-to-run transition 83-84 DOMS. See muscle soreness, preventing
pedaling drills 82-83 doorway shoulder stretch 232, 232f
pedaling mechanics 73-79, 74f, 75f, 76f, 77f Douglas bag method 161-162
power and resistance in 71-73 drafting (cycling)
cycling cleats 478, 481 bicycle choice and 115, 116
cycling efficiency effect on bike-to-run transition 97
measurement of 172 energy savings from 125-126, 126f
in Tour de France cyclists 172, 173f race guidelines 127
training effects on 175 drafting (swimming) 361
cycling helmets 444 drag, in swimming
cycling shoes defined 59
in foot and ankle injuries 481-482 in propulsion 59-60
Index | 625

types 63-65 in run base building 377


water temperature effects on 105 in swim base building 354-355, 356
drag-generated propulsion 59-60 energy balance equation 521
DREW study 167 energy gels 548-549
dropped elbow, in swimming 68-70, 69f energy levels, monitoring 347-348
drug use, illegal 12 energy production
dry needling 502 ATP in 4-5, 4f, 5f
DSHEA 543-544 energy systems in 5-8, 5f, 6f, 7f
duathlon exercise intensity and 185-187
distances 431 training effects on 22, 23
finish times 431, 432t energy systems
racing tactics 437-439 ATP-CP (immediate) 5-6, 6f, 23, 186, 551-552
sample training plan 436, 436t-437t dynamics of 8-9, 9f, 270-271
training strategies 432-435, 447-448 glycolysis (short-term) 6-7, 7f, 23, 183-184
duration of exercise overview 5, 5f
in cycling base building 363-366 oxidative phosphorylation (long-term) 7-8,
guidelines 301 7f, 22-23, 184
in interval training 291 usage ratio in training 270-271
dynamic gluteal muscles 480, 480f enjoyment of training 286-287, 562, 565
dynamic knee angle method, in bike fitting 120 enthusiasm 560
environmental factors. See also altitude; heat and
humidity; pollution
E in muscle fiber composition 30, 30t
EAHE 36 in tactical training 276
early vertical forearm (EVF) 68-70, 69f in warm-ups 213-214, 216
eccentric ankle inversion with band 493, 493f epicondylitis 508
eccentric muscle actions epigenetic, defined 27
in cycling vs. running 83-84 EPO 12
for injury recovery 499, 508 equipment. See also bicycles
economy. See also running economy in bike-to-run transition 444
defined 171 in duathlon transition 444, 445f
gender differences in 40 swim gear 109-110, 110t
measurements of 171-172 swimsuits 106-107, 107f
multisport interference 177-178 in swim-to-bike transition 442-443, 443f
as performance determinant 172 in tactical training 276
relation to VO2max 172 wetsuits 61, 63, 108-109, 108t
strength training effects on 251-252 ergogenic aids 12, 546. See also supplements
training effects on 14-15, 173-176, 177f erythrocytes (RBCs) 11, 12, 288
effective force essential amino acids 519
in aerodynamic positioning 81, 81f essential fatty acids 520, 527-528
in pedaling mechanics 73-75, 74f, 77-78 ETAP 243
elbow position ETS 7f, 8
in swim drill 279 EVF 68-70, 69f
in swim technique 68-70, 69f excitation phase, of muscle contraction 18-19, 19f
elbow tendinitis or tendinosis 506 exercise-related transient abdominal pain (ETAP) 243
electrolyte homeostasis 36 exponential taper 331, 331f, 332t. See also taper
electrolyte supplements 549
electron transport system (ETS) 7f, 8
electrotherapy 502-503
F
encephalopathy (EAHE) 36 FAI 473, 487
endurance training. See also training adaptations fainting
adaptations in 271-273, 271t, 273f, 365-366 in combo workouts 445-446
aerobic-anaerobic ratios in 270-271 postexercise 220
in cycling base building 366, 367 family physicians 455
effect on lactate levels 185 fascia injuries 505, 511
effect on VO2max 168 fascial ruptures 507
as general preparation 274, 275t, 390 fast-decay taper 331, 331f, 332t. See also taper
genetic effects on 31-33, 32t fast glycolytic (FG) muscle fibers 20-22. See also
heat tolerance and 15, 205-206 fast-twitch muscle fibers
vs. high-intensity training 186-187, 196-197, fast oxidative glycolytic (FOG) muscle fibers 20-22
287, 374 See also fast-twitch muscle fibers
626 } Index

fast-twitch muscle fibers (Type II) stretching in cool-down 221


age-related decline in 48, 50 in swimming performance 224-225, 224f
metabolic features 21-22 training guidelines 228t
in muscle composition 21, 30 floatation aids 109, 110t
performance and 193-194 fluid balance. See hydration
physiology and characteristics 192-193, 192t flying dismount 444
training effects on 194-195 flying mount 443
fat. See body fat; dietary fat; fat metabolism foam rolling exercises
fat-free mass 29 gastrocsoleus 483, 483f
fatigue ITB 479, 479f
anticipatory regulation 202-203, 205 focus 560, 561, 576-579
as context-specific 199-200 FOG muscle fibers 20-22. See also fast-twitch
failure concept of 200-201 muscle fibers
H+ role in 6 Food and Drug Administration (FDA) 543, 544
from low iron stores 529 foot injuries
as overtraining sign 342-343 preventive exercises 482-483, 482f, 483f, 494,
pacing and regulation 201-204 494f
in performance modeling 318-319, 318f recovery techniques 506
relaxation for 570 types 481-482, 491-493
as sensation vs. event 201, 205 foot strike 90-91
training effects on 205-206 foot strike hemolysis 132-133
as warm-up risk 214 force, in running speed 249
fat metabolism. See also body fat; dietary fat form drag 63
age-related decline in 53 Foster, Carol 172
as fitness indicator 271-273, 272f fractures 509. See also stress fractures
measurement of 165-166, 166t free radicals 526
as metabolic efficiency 515-516 frequency of exercise
fat pad impingement 489 brick workouts 446-447
FDA 543, 544 guidelines 300, 309
Federation Internationale De Natation 102, 107 interval training 293
femoral acetabular impingement (FAI) 473, 487 swim base building 355
femoral stress fractures 487 during taper 329-330
FG muscle fibers 20-22. See also fast-twitch frictional drag 64
muscle fibers fructose 516
fiber 517, 538 FTP. See functional threshold pace; functional
FINA rules and guidelines 102, 107 threshold power
finger spread 281 fuel utilization. See metabolic efficiency
fist drill 281 full-body stretch 233, 233f
fitness functional overreaching 340
fat metabolism and 271-273, 272f functional threshold pace (FTP) 154
and health assessments 451-454, 452f functional threshold power (FTP) 151, 183, 187
in performance modeling 318-319, 318f functional training 382
sprint race tactics based on 396-398
sprint training based on 389-393
tracking with power meter 149 G
warm-up duration and 215-216 galactose 516
flat days (functional overreaching) 340 gamma-linolenic acid (GLA) 527-528
flax 520, 527 GAS 312-313, 312f
flexibility gastric emptying rate 517
age-related decline in 34 gastrocsoleus exercises 482, 482f, 483, 483f
in cycling performance 225 gender differences
defined 458 mental training 555
exercises for triathlon performance
lower body 229-230, 229f, 230f, 234-238, age-gender interaction 47-48, 47f
234f, 235f, 236f, 237f, 238f event distance comparison 46-47, 46f
upper body 231-234, 231f, 232f, 233f, 234f participation rates 41
in injury prevention 239-241, 458 performance density 41-42, 42t
in injury recovery 501 physiological factors 40-41
multisport issues in 227-228 sport comparisons 42-46, 43f, 44f, 45t
in running economy 176 VO2max 40, 164
in running performance 225-227 General Adaptation Syndrome (GAS) 312-313, 312f
Index | 627

general training. See base building; cycling base hands on head drill 283
building; endurance training; run base Haskins, Sarah 537
building; swim base building Hawaii Ironman Triathlon
genetics age-gender interaction in performance 47-48,
DNA variations 27-28, 28f 47f
in electrolyte homeostasis 35-36 gender differences in performance 42-46, 43f,
in endurance performance 31-33, 32t 44f, 45t
in injury risk 33-35, 35t qualification for 423, 424, 428
in physical characteristics 29-31, 30t Hb. See hemoglobin
in VO2max 167 Hct 11, 288
genetic testing 36 health, defined 460. See also medical care
genotype, defined 27 heart, in cardiopulmonary system 9-10, 10f
GI (glycemic index) 517-518 heart rate
GI distress 538 in aerobic threshold testing 365, 377
GL 517-519 as anaerobic threshold marker 183
GLA 527-528 limits in monitoring 144
glucose 516 training effects on 12-13
glucose polymers 517 heart rate monitors
glutamine 343 limits of 144
glute kicks drill 283 as pacing tool 145
glycemic index (GI) 517-518 training zones for 145, 146t-148t
glycemic load (GL) 517-519 heat and humidity
glycemic response 518-519 acclimation to 138, 398
glycogen depletion effects on performance 137-138
in failure concept of fatigue 200 role in fatigue 201-203
in fatigue regulation 203-204 training effects on heat tolerance 15, 205-206
training effects on 206 warm-ups in 213-214, 216
as warm-up risk 214 heat illness and injuries 138
glycogenolysis 183-184 heat therapy 220-221
glycogen stores 515 heel raises 266, 266f, 494, 494f
glycolysis height, genetic effects on 28-29
energy production in 6-7, 7f hematocrit (Hct) 11, 288
rate of and lactate formation 184-185 heme iron 529
training effects on 23 hemoglobin (Hb)
goals 557-558, 564, 577 gender differences in 40
goggles, swimming 109 iron in 528
GPS devices 153-155 in overtraining 288
grapevine step 246, 246f in oxygen transport 11, 12
Graston technique 459-460, 503 herbal compounds 549-550
gross efficiency, vs. delta efficiency 175 Heritage studies 167
Guimard, Cyrille 317 high elbow drill 279
high-fructose corn syrup 517
high-intensity training (HIT). See also interval
H training
H+ 6, 23 for advanced athletes 404
HA injections 508 argument for 302-304
half-Ironman triathlons for cycling efficiency improvement 175, 186
distances 409 vs. endurance training 186-187, 196-197, 287, 374
finish times 409 as general preparation 274, 275t
racing tactics 415-417 for VO2max improvement 168-169, 169f
sample training plan 413, 414t-415t high knees drill 283
training strategies 409-413 high–low training model 139
warm-ups 215, 416 high molecular weight carbohydrates 517
Hall, Harriet 460 Hill, A.V. 161-162
hamstring curls 265, 265f, 479, 479f hill work
hamstring injuries for race-specific training 390, 402
core strength effects on 243 for speed improvement 132-133, 132f, 267
in cyclists 473 hip abductor exercise 247, 247f
recovery techniques 499, 501, 508, 510 hip flexor
in runners 487, 489-490 stretches for 237-238, 237f, 238f, 476, 476f
hamstring stretches 235, 235f tendinitis or tendinosis 473
628 } Index

hip injuries. See pelvis and hip injuries running-specific


HIT. See high-intensity training foot and ankle injuries 491-493
“hitting the wall” 200 knee injuries 489-490
Holland, Tom 460 lumbar spine injuries 484
Hollander, Lew 48 pelvis and hip injuries 487-488
homeostasis failure 200 swimming-specific
homeostasis regulation 201 shoulder injuries 462-463
horsepower, in cycling power 72 spine and neck injuries 464-465
hot spots, on feet 481, 492 injury incidence 239, 457, 461, 462t, 484
hunger and satiety 521-522, 523 injury prevention
hyaluronic acid (HA) injections 508 as base building benefit 366
hydration bike fit as 468
genetic factors in 35-36 core strength in 242-244
monitoring 531, 532, 533 exercises for
posttraining 540 feet and ankles 482-483, 482f, 483f, 493-
pretraining 535, 536-537 494, 493f, 494f
during training 539-540 knees 479-480, 479f, 480f, 490, 490f
hydrogen ions ( H+) 6, 23 low back (lumbar) 466-467, 466f, 467f,
hypernatremia 36 470-472, 470f, 471f, 472f, 485-486,
hyperthermia. See also heat and humidity 485f, 486f
in failure concept of fatigue 200 midback 463-464, 463f, 464f, 469, 469f
fluid intake and 539 neck 465, 465f
as warm-up risk 213-214 pelvis and hips 474-477, 474f, 475f, 476f,
hypertrophy 23 477f, 488, 488f
hyponatremia 36 shoulders 463-464, 463f, 464f
hypothermia 104 flexibility training in 239-241, 458
health assessments for 452-454, 452f
for heat-related injuries 138
I mind-body awareness skills and 566, 567
ibuprofen 505 relaxation and 570
ice treatments 220-221, 499 running biomechanics and 95-96
iliotibial band syndrome (ITBS) strength training for 456-458
in cyclists 473, 478 training progression and 378
prevention of 243 warm-ups as 213
recovery techniques 241, 505 injury recovery techniques
in runners 489 conservative clinician-assisted 502-507
imbalance do-it-yourself 497-501, 570
in life 565 invasive clinician-assisted 507-509
in swimming 66-67, 66f physical exams in 509-510
immediate energy system. See ATP-CP system suggested approaches 510-511
immune system 549, 570 injury risk
impact forces, in running in bike-to-run transition 83
injury risk and 96 in exercise duration 301
running surfaces and 129, 130, 131f genetic effects on 33-35, 35t
impingement syndrome (shoulder) 462 in high-intensity training 169
impulse, in swimming 61 in interval training volume 287
impulse–response (IR) model 318, 318f, 334 risk factors 457
inchworm exercise 485, 485f running biomechanics and 95, 95f
inclines, for strength endurance training 382 running shoes and 133
ineffective force, in pedaling mechanics 73-74, running surfaces and 129
77-78 inland swims 104
inflammation 526-528, 550 instinctual eating 522
Informed Choice 544 instrument-assisted tissue mobilization 503
injury, training with 495, 495t, 564 insulin 522
injury descriptions intensity of exercise. See also high-intensity
cycling-specific training
foot and ankle injuries 481-482 aerobic-anaerobic usage ratios 270-271
knee injuries 477-478 in base building 272-273, 363-366, 377
low-back (lumbar) pain 469-470 cross-reference chart 388t
neck and midback pain 468 energy production and 185-187
pelvis and hip injuries 473 guidelines 309
Index | 629

inconsistency in defining 387-389 kneeling stretches


in interval training 290-291 ankles 229, 229f
low-intensity argument 304-308, 305f, 306f, hip flexors 238, 238f
307f lats 231, 231f
during taper 328 knee-over-pedal-spindle (KOP) method, in bike
interleukin-6 343-344 fitting 120
International Triathlon Union (ITU) knees
race videos 441 injury prevention exercises 479-480, 479f,
rules and guidelines 480f, 490, 490f
cycling equipment and racing 113, 116, injury recovery techniques 511
117, 127 injury types 477-478, 489-490
duathlon distance standards 431 pain in 242-243
water conditions and temperature 102, stretch for 234, 234f
106 Koechli, Paul 317
wetsuits 109 KOPS method, in bike fitting 120
World Championships 402 Krebs cycle 7f, 8, 22, 184
interval training
benefits of 285-287
general interval structure 288, 289-293, 294t L
periodization of 293-294 lactate
physiological effects of 289, 289t in anaerobic metabolism 184
specific interval structure 288, 294, 295t cool-down effects on 219
volume of 287-288 in glycolysis 6
intrinsic motivation 559 in muscle soreness 220
in-water starts 102-103 as performance determinant 188
IR model 318, 318f, 334 as threshold marker 182
Ironman triathlons training effects on 184-185
cutoff times 419 lactate dehydrogenase (LDH) 184
distances 419 lactate shuttle 185
finish times 419 lactate threshold. See anaerobic threshold
racing tactics 425-429 lactate threshold training
sample training plan 425, 426t-427t effect on VO2max 168
training strategies 421-425, 447 in half-Ironman training 412
warm-ups 215 in Ironman training 422
iron stores lactose 516
altitude training and 139 laser therapy 503
deficiencies in 528, 529-530 lat pull-down 263, 263f
food sources 528-529 lat stretches 231, 231f
foot strike hemolysis and 133 LDH 184
function of 528 leg extension method, in bike fitting 119
in overtraining 288 leg-length discrepancies 478, 481
iron supplements 530, 546-547 leg press 262, 262f
Issurin, V.B. 316-317 LeMond, Greg 317
ITBFS. See iliotibial band syndrome LeMond method, in bike fitting 119
ITBS. See iliotibial band syndrome lift-generated propulsion 59-60
ITU. See International Triathlon Union ligament injury treatments 503, 508, 509, 510. See
also specific types
J lightheadedness
in combo workouts 445-446
joint dislocations 458 postexercise 220
limitations
analyzing with power meter 149
K correcting 368, 574-575
ketoprofen 507 in health assessment 453-454
kick, count, stroke drill 281 in Ironman training planning 423-424
kick, in swimming swim base building goals and 353-354
function of 61-63, 62f linear taper 331, 331f. See also taper
mistakes and corrections 65-66, 65f listening to body 565
kick drills 280 long, slow distance training. See endurance
knee drive drill 283 training
knee extensions 265, 265f long rides 369
630 } Index

long-term energy system. See oxidative phos- medial tibial stress syndrome (shin splints) 491
phorylation medical care
long-term goals 558 health and fitness assessment 451-454, 452f
Lore of Running (Noakes) 456 for injury recovery
low-back (lumbar) injuries or pain conservative treatments 502-507
preventive exercises 466-467, 466f, 467f, 470- invasive treatments 507-509
472, 470f, 471f, 472f, 485-486, 485f, 486f physical exam for 509-510
recovery techniques 499-500, 500f, 504, 504f, massage 458-460
507, 511 off-season recovery 451
types 464-465, 469-470, 484 resource team for 454-456
low-back stretches 236-237, 236f, 237f medium chain triglycerides (MCTs) 551
low-intensity training. See also intensity of meniscus tears 489, 511
exercise mental imagery. See visualization
cycling power distribution data 306-308, 307f mental rehearsal exercise 579
pace zone distribution data 306, 306f mental toughness. See mental training
training adaptations in 304 mental training
volume distribution data 305-306, 305f discipline 558-562
LSD training. See endurance training effectiveness of 555
Lucia, Alejandro 172 focus and determination 576-579
lung capacity. See VO2max gender differences in 555
lungs, in cardiopulmonary system 9-10, 10f goals and commitment 557-558
Lydiard, Arthur 304, 375 peak performance mind-set 555-557
lying full-body stretch 233, 233f self-confidence 573-576
visualization 212-213, 560-561, 579-582
Mercier scores 322, 322f
M mesocycles
macrocycles in block periodization 315
defined 313 defined 313
nutrition guidelines 526-530 in nonlinear periodization 317-318
in traditional periodization 314f in run base building 377-379, 379t
macrocytic anemia 530 in swim base building 355-356
macronutrients in traditional periodization 314, 314f
carbohydrate 516-519 metabolic efficiency
daily intake ranges adaptogens effects on 550
in base period 531, 531t described 515-516
general guidelines 525-526 as fitness indicator 271-273, 272f
off-season 533 measurement of 165-166, 166t
specific training period 532, 532t to reduce GI distress 538
fat 519-520 training effects on 206
protein 519 metatarsalgia 481, 492
vitamins and minerals 521 metatarsal injuries 93, 96, 492
water 520-521 microcycles
maltodextrins 517 defined 313-314
maltose 516 in swim base building 356-357
Maslow, Abraham 556 in traditional periodization 314f, 315
massage 458-460, 504-505 microcytic anemia 530
massage therapists 455 midback pain
masters athletes. See also age-related declines in cyclists 468
duathlon finish times 431, 432t exercises for 463-464, 463f, 464f, 469, 469f
gender differences in performance 47-48, 47f mind–body connection
muscle fiber changes in 50 awareness exercise 568
performance examples of 39, 49, 49t awareness skills 566-568
masters groups, for swim training 356 effect on performance 555
maximal exercise performance 14 in preventing staleness 565
maximal oxygen uptake. See VO2max minerals 521, 549
maximum strength (MS), in strength training mitochondria 17f
257, 258t ATP production 8, 11
maximum transition (MT), in strength training training effects on 22, 195-196
256, 257t mixed training 375
Mb 22-23, 528 mobile training apps 157
MCTs 551 modeling 575
Index | 631

molecular weight carbohydrates 517 NNFA 544


momentum, in swimming 59 Noakes, Tim 456
monosaccharides 516 nonessential amino acids 519
monster walk 474, 474f nonfunctional overreaching 340-341
mood states nonheme iron 529
monitoring 347-348 nonlinear (undulating) periodization 317-318,
as overtraining sign 343 375-377
Morton’s neuroma 481, 492 non-steroidal anti-inflammatory drugs (NSAIDs)
motivation 558-560 505
motor units, training effects on 24 normal force, in pedaling mechanics 73-74, 74f
MS phase, in strength training 257, 258t normocytic anemia 530
MT phase, in strength training 256, 257t nose clips 109-110
multifocal osteonecrosis 507 NSAIDS 505
multisport athletes, vs. single-sport NSCA 456
economy in 94-95, 177-178 NSF International 544
flexibility training for 227-228 NTG 506
muscle recruitment in cycling 83 nutrition
study limitations 178 macronutrients 516-521, 525-526
multivitamins 548 metabolic efficiency 515-516
muscles planning with power meter data 150
activation reserves 203, 205 posttraining 540-541
age-related decline in 48 pretraining 535-537
anatomy of 16, 17f race-week and race-day plans 541
in bike-to-run transition 83-84 in tactical training 276
contraction of 16-20, 18f, 19f during training 537-540
core 241, 242f weight management 521-523
fiber composition 30-31, 30t, 50, 193-194 nutrition periodization
fiber types 20-22, 192-193, 192t base period guidelines 531
gender differences in 40 concept and goals 525
heat effects on 201-202 macrocycle guidelines 526-530
oxygen transport to 11, 204 off-season 533
recruitment and multisport training 83 specific training period 532, 532t
strength training effects on 252
training effects on 22-24, 191, 194-197
in VO2max 162 O
warm-up effects on 212, 213f, 217 Ober’s test 241
muscle soreness, preventing 213, 220, 505, 538 OBLA 182. See also anaerobic threshold
muscle strains 503, 505, 510 obligation, in obsession profile 564
muscular endurance 24, 278, 289, 289f obsession symptoms and profile 563-565
muscular power 24 ocean swims 103, 104
muscular strength 24 off-season 451, 533, 537
myoglobin (Mb) 22-23, 528 oligosaccharides 517
myosin 16, 18f, 19-20, 192, 212 Olympic-distance triathlons
distances 399
N finish times 400
racing tactics 404-407
National Nutritional Foods Association (NNFA) 544 sample training plan 404, 405t
National Strength and Conditioning Association training strategies 400-404
(NSCA) 456 warm-ups 215, 217t
naturopathic physicians 508 Olympic games 399
neck injuries or pain omega-3 fats 520, 527-528
injury recovery techniques 504, 504f, 507 omega-6 fats 520, 527-528
preventive exercises 465, 465f, 469, 469f one-repetition maximum lift (1RM)
types 464-465, 468 determining 259-260
neck rotation stretch 465, 465f as muscular strength measure 24
negative thinking 565, 571, 575 onset of blood lactate accumulation (OBLA) 182.
nerve-gliding techniques 505 See also anaerobic threshold (AT)
neuromuscular function 177-178 open-water swimming
neuromuscular training 302, 367 vs. pool swimming 101-102
Newton’s laws of motion 59 skills training 357, 361
nitroglycerine (NTG) 506 water conditions 104-106, 105t
632 } Index

organization skills 559 aerodynamics and 80-82, 80f, 81f


orthopedic surgeons 455 cyclist type comparisons 79
orthotics 136-137, 493, 505-506 pedal force and crank torque 73-78, 74f, 75f,
osteoarthritis 508 76f, 77f
osteopathic physicians 503-504 skill training drills 282-283
OTS. See overtraining peloton 442
overload 299-300 pelvic position, in running 95, 95f
overreaching 340-342 pelvis and hip injuries
overtraining preventive exercises 474-477, 474f, 475f, 476f,
causes 345 477f, 488, 488f
defined 341 types 473, 487-488
diagnosing 344-345 perfectionism 557
incidence rate 564 performance decline, in overtraining 342
obsession in 563-565 performance determinants
vs. overreaching 340-342 anaerobic threshold 184, 187-188
prevention and treatment 347-348, 549-550 economy 172-173
as psychological disorder 346-347 muscle fiber composition 193-194
risk in interval training 288 VO2max 165, 168
self-monitoring questionnaire 347-348 performance-enhancing substances 12, 546. See
signs of 342-344 also supplements
overuse injuries. See injury descriptions; injury performance management charts 156, 157f
incidence; injury prevention; injury performance modeling
recovery; injury risk equations for 318-319, 318f, 335
oxidative phosphorylation of periodized training loads 319-324, 320f,
energy production in 7-8, 7f, 184 321f, 322f, 323f, 324t
training effects on 22 of taper 333-337, 334f, 336f
oxidative stress 526 performance monitoring 347, 423. See also tech-
oxygen-enriched blood 9-11 nology training tools
oxygen-reduced blood 10-11 performance plateaus 358
oxygen supply, in fatigue 200, 204 periodization
oxygen transport base period intensity 273
in cardiopulmonary anatomy 9-10, 10f block 315-317, 316f
physiology of 10-11 cycles in 313-314
training effects on 12-15, 23 of interval training 293-294
original model 311, 312f, 375
performance modeling of 318-324, 318f, 320f,
P 321f, 322f, 323f, 324t
pacing strategies rationale for 312-313
duathlon 437-438 of strength training 255, 267
in fatigue regulation 201-204 target events and approach choice 324
half-Ironman 415-416 traditional 314-315, 314f, 375-377
Ironman 425-428 undulating (nonlinear) 317-318, 375-377
Olympic-distance events 406 Periodization (Bompa) 249
sprint events 396-397 peripheral fatigue 343
pacing tools peroneal tendinosis or tendinitis 491
heart-rate monitors 145 pes anserine tendinitis, tendinosis, or bursitis
power meters 150, 152-153, 153t 478, 489
paddles, swimming with 110t, 359, 360f pharmaceuticals, topical 506-507
pain 455-456 Phelps, Michael 254
partial pressure of oxygen (PO2) 10f, 11 phenotypes, defined 27
passive cool-downs 220-221 physical exams, for injury treatment 509-510
passive warm-ups 211 physical therapists 455
patella, described 477 physical therapy treatments 502-507
patellar tendon injuries 489, 499, 509 physicians 455, 502-504, 508
patellofemoral injuries 506 physio tape 506
patellofemoral syndrome 489, 506, 511 pike with physioball 486, 486f, 498, 498f
PDH 184 Pilates, Joseph and Clara 241
peak performance skills. See mental training piriformis syndrome 487-488
peak power 290, 367 plank (bridge) 245, 245f
pec stretches 233-234, 233f, 234f planning, based on health assessment 451-454,
pedaling mechanics 452f
Index | 633

plantar fasciitis essential vs. nonessential 519


described 491-492 training intake ranges
recovery techniques 136, 506, 508, 511 posttraining 540
running shoes and 133 pretraining 536
platelet-rich plasma (PRP) therapy 508 during training 538
plyometric training PRP therapy 508
for bike-to-run transition 84 Pruitt, Andy 455
for running economy 176, 177f, 267 psychological issues. See mental training
PO2 10f, 11 push-ups 264, 264f
podiatrists 506 push-ups with rotation 247, 247f
pollution 140-141, 141f, 526, 549 pyruvate dehydrogenase (PDH) 184
polyglucosides 517 pyruvic acid 6, 7, 183-184
polymorphisms 27-28, 28f
polysaccharides 517
pontoon starts 361 R
pool swimming, vs. open-water 101-102 race analyses 149
Popov, Alexander 317 race belts 444
pose method of running race-specific training. See also specific race
injury risk and 96 distances
running efficiency and 88, 91-92 in cycling base building 368
for skill development 382 vs. general training 274-277, 277t
positive affirmations 576 interval training 294, 295t
positive attitude 561 nutrition planning 541
posterior tibialis tendinosis or tendinitis 491 by race distance 276, 277t
power running surfaces for 132
cycling 71-72, 77, 77f, 124, 125f strength training 267
swimming 61 terrain in 390, 402
VO2max measurements 163-164 race tactics and strategies 276, 387. See also
PowerCranks 283 specific race distances
power meters range of motion. See flexibility
cycling aerodynamic assessments 127, 127f rating of perceived exertion (RPE)
as pacing tool 152-153, 153t in aerobic threshold testing 377
pedaling efficiency measures 283 cross-reference chart 388t
pros and cons 149-150 in fatigue regulation 203, 204, 206
steps in using 150-152 in pacing 152-153, 389
wattage measures 149 training effects on 205
power profile 151t, 152 RBCs 11, 12, 288
power stroke, in muscle contraction 20 reaction force 59, 90
power training levels 151, 151t reactive oxygen species 526
predicted pace 377 realization mesocycle 315
prerace checklists 213 reciprocal inhibition 89-90
present-centered focus 577-578 recombinant human erythropoietin (rhEPO) 12
press-ups 470, 470f, 504, 504f recovery periods. See rest or recovery periods
pressure drag 63 red blood cells (RBCs) 11, 12, 288
pressure gradient 10f, 11 refueling 540
process goals 577 rehearsal, mental 579
projection angle, in running 90-91, 91f rehydrating 540. See also hydration
prolotherapy 508 relaxation
pronation 136 benefits of 570
propulsion, in swimming effects of 569-570
arm pull 60-61, 60f exercise for 571-572
defined 59 stress self-assessment 571
kick 61-63, 62f RER
types 59-60 in aerobic metabolism measure 165-166, 166t,
protein 172
in blood sugar control 522 as anaerobic threshold marker 182
daily intake ranges vs. RQ 165
base period 531, 531t, 533t resistance, in cycling 72-73, 118
general guidelines 526 resistance training. See strength training
off-season 533 respiratory exchange ratio (RER). See RER
specific training period 532, 532t, 533t respiratory quotient (RQ) 165
634 } Index

rest intervals, in interval training 292 as performance determinant 173


restoration cycle 378 running form and 87-88
rest or recovery periods strength training effects on 251-252, 267
enjoyment during 562 running shoes
off-season 451 for barefoot progression 93-94
in Olympic-distance training 400, 404 in bike-to-run transition 444
training adaptations and 313 injury risk and 492
reversibility 299-300, 316-317 pro and cons 133-136
rhEPO 12 recommendations 136
rigor mortis 20 running shop pros 455
road bikes 115-116. See also bicycles run-to-bike transition training 447-448
roller training, for cycling 82 run training
rotational inertia 125 Olympic-distance 402
rotator cuff strengthening 463, 463f skill drills 283
rotator cuff tears 509 sprint distance 391
RPE. See rating of perceived exertion Russian electrical stimulation 503
RQ 165
run base building
conflicting evidence on 373-374 S
evaluating adaptation in 380 sacral stress fractures 487
vs. higher-intensity training 374 safflower oil 527
mesocycle design concepts 377-379 salmon 520, 527
sample four-week mesocycle 379, 379t sarcomere 16, 17f
skill development in 382 sarcoplasmic reticulum 16, 17f, 19, 20
strength endurance training in 382 saturated fats 520
strength training in 381-382 sciatica 507, 511
training progression in 378, 380-381 Science Based Medicine (Hall) 459-460
runner’s stretch 237, 237f Science of Flexibility, The (Alter) 240
running. See also injury descriptions scorpions (exercise) 477, 477f
10 percent rule 378 Scott, Dave “The Man” 39, 49t, 456
age-related decline in 49-51, 50f scraping drill 282
at altitude 138-140 seated rows 264, 264f
barefoot 88, 92-94, 93f, 96, 133-134, 134f seated trunk rotation 466f
flexibility and performance 225-227 seat tube angle 82, 97-98, 97f, 114-115, 115f
gender differences in 43f, 44f, 45-46, 45t, selenium 526
46-47, 46f self-acceptance 575-576
heart rate zones 145, 147t-148t self-actualizations 556
in heat and humidity 137-138 self-confidence 573-576
injury rates 239, 457 self-criticism 571, 575
orthotics 136-137 self-direction 559
pace zones 154, 154t, 389 self-mobilization 504, 504f
with pollution 140-141, 141f self-talk, healthy 575
pose method 88 Selye, H 312-313, 312f
skill training drills 283 serotonin, in overtraining 346-347
strength training effects on 250-252, 253, 267 setbacks, attitude toward 562
taper in 329, 330 shark fin drill 67
training surfaces 129-133, 130f, 131f Shepherd, Roy 345
running biomechanics shin splints 491
athlete comparisons 94-95 shoes. See cycling shoes; orthotics; running shoes
cycling effects on 96-98 short-term energy system. See glycolysis
in economy and form 87-88 short-term goals 558
in efficiency and performance 88-95, 91f shoulders
in injury risk and prevention 95-96, 95f injury prevention exercises 463-464, 463f,
running surfaces and 129 464f
running economy injury recovery techniques 504, 504f, 506
barefoot running and 134, 134f injury types 462-463
cycling effects on 84, 173-174, 178 stretches for 232, 232f
factors affecting 90-91, 91f side bridge (side plank) 245, 245f
flexibility effects on 225-226 side plank (side bridge) 245, 245f
improving 89-90, 176 side step 246, 246f, 475, 475f
measurement of 171 side swimming drill 281, 281f
Index | 635

simple carbohydrates 516-517 warm-ups 215, 217t, 396


single-leg drills, for cycling 82, 282 squats 261, 261f
sitting hamstring stretch 235, 235f SSTRIDE studies 167
skill training staleness (nonfunctional overreaching) 340-341, 565
cycling 281-283 standing stretches
general vs. specific 278 hamstrings 235, 235f
running 283 lats 231, 231f
swimming 278-281, 279t, 280f, 281f, 354-355, lower back 237, 237f
358-361, 360f starch 517
skin drag 64 static stretching
sleep, improving 570 guidelines 228t
sliding filament theory 20 lower body exercises 229-230, 229f, 230f, 234-
slow-decay taper 331, 331f, 332t. See also taper 238, 234f, 235f, 236f, 237f, 238f
slow oxidative (SO) muscle fiber. See slow-twitch prerun 226, 227f
muscle fibers upper body exercises 231-234, 231f, 232f, 233f,
slow-twitch muscle fibers (Type I) 234f
metabolic features 21-22 stationary bike training 367, 369
in muscle composition 21, 30 stem cell therapy 508-509
performance and 193-194 step taper 331. See also taper
physiology and characteristics 192-193, 192t steroid injections 507
training effects on 194-195 streamline position, in swimming 63, 63f
SM phase, in strength training 258-259, 258t strength endurance training 382
snorkels 110t, 359, 360f strength maintenance (SM), in strength training
social support network 562 258-259, 258t
socks 397, 442, 444 strengths, affirming 576
sodium intake strength training
posttraining 540 arguments against 252-253
pretraining 537 benefits of 175, 250-255
during training 538-539 best approach 255
software, for training analysis 155-156, 156f, exercises 255t, 261-266, 261f, 262f, 263f, 264f,
157f, 283 265f, 266f
SO muscle fibers 20-22. See also slow-twitch importance of 249
muscle fibers for injury prevention 456-458
soybean oil 527 load goals for 259-260
specificity 299-300 muscular adaptations 23-24
specific training nutrition 532, 532t, 533t. See also phases of
race-specific training anatomical adaptation 256, 256t
speed endurance training 303, 304 competition maintenance 259, 259t
speed maintenance 303 maximum strength 257, 258t
speed production 303 maximum transition 256, 257t
speed suits 107 strength maintenance 258-259, 258t
speed work 302 race-specific 267, 421
spinal injuries or pain. See low-back injuries or in run base building 381-382
pain; midback pain sport-specific 267-268
spinal manipulations 503-504 stress
SpinScan software 283 nutrition and 526
Spirig, Nicola 28 in overtraining 345
sport psychology 559. See also mental training self-assessment 571
sports anemia 530. See also anemia supplements for 549
sports hernia 473, 487, 509 stress fractures
sports massage 459 as corticosteroid side effect 507
sports physicians 455. See also injury recovery recovery techniques 501, 511
techniques; medical care types 487, 492
sports supplements, defined 546. See also supple- stretching. See flexibility
ments stretch-shortening cycle 50, 176
spotters 428 stride length
sprint triathlons flexibility and 226, 227, 227f
distances 389 in running economy 90-91, 91f
racing tactics 396-398 strength training and 249, 251-252
sample training plan 393, 394t-395t stride rate
training strategies 389-393 cycling effects on 97
636 } Index

stride rate, continued training effects on 174-175, 174f


in running economy 90-91 swimming strokes, for base building 361
strength training and 249 swimsuits 106-107, 107f, 442
strides 283, 379 swim-to-bike transition
sucrose 516 tips for 442-443, 443f
sunflower oil 527 training for 445-446
sunglasses 444 swim training
superbikes 117. See also bicycles Olympic-distance 400-402
Superman on BOSU 486, 486f skill drills 278-281, 279t, 280f, 281f, 358-361,
super suits 106 360f
supplements sprint distance 391
benefits and drawbacks 547t
contamination testing 544
evaluating 544-545, 545f T
illegal 12 tactical training. See race tactics and strategies;
most popular 546-552 specific race distances
regulation and oversight 543-544 talk test 366, 377
types 546 tangential force, in pedaling mechanics 73-74, 74f
surgery 509 tape, physio 506
surgery recovery techniques taper
conservative clinician-assisted 502-503 approach comparison 332, 332t
do-it-yourself 501 defined 327
swim base building distinguishing criteria 327
fitness goals 353-355 duration of 330
individual training sessions 357 elements of 328-330
mesocycle 355-356 guidelines for 337-338
microcycle 356-357 performance gains from 328
progress monitoring 358 shape of 330, 331f
skill training drills 358-361, 360f triathlon-specific studies of 333
swim caps 109 Taylor, H.L. 162
swim clubs 356 technology training tools
swim golf 279, 359 benefits of 143, 389
swimming. See also injury descriptions effectiveness of 144
age-related decline in 51, 52f heart rate monitors 144-149, 146t-148t
economical strokes 171-172 mobile apps 157
flexibility and performance 224-225, 224f power meters 149-153, 151t, 153t
gender differences in 42-44, 43f, 44f, 45t, speed distance devices 153-155, 154t
46-47, 46f training analysis software 155-156, 156f, 157f
inefficiency 57 using in races 416
race gear 109-110 tendon injuries 506-507, 508, 509, 510. See also
skill training drills 278-281, 279t, 280f, 281f, specific types
358-361, 360f tendon ruptures 507
start types 102-103 tennis balls swim drill 359
strength training effects on 254-255, 267 terminal cisternae of sarcoplasmic reticulum 16,
swimsuits 106-107, 107f 17f, 19, 20
taper in 329, 330 terrain, in training 390, 402
training gear 110, 110t testing
water conditions 101-102, 104-106, 105t aerobic threshold 364, 377
wetsuits 61, 63, 108-109, 108t in cycling base building 364-366
swimming accelerometers 155 genetic 36
swimming biomechanics supplement contamination 544
center of mass position 57-58, 58f in swim base building 358
drag 63-65, 64f, 105 testosterone–cortisol ratio 343
mistakes and corrections 65-70, 65f, 66f, 68f, thermogenic response 522
69f thirst 521, 539
momentum 59 thoracic extension 463, 463f, 504, 504f
propulsion 59-62, 60f, 62f threshold heart rate 145
streamline position 63, 63f threshold pace (FTP) 154
swimming blind drill 361 threshold power (FTP) 151, 183, 187
swimming economy tibial stress fractures 492, 507
measurement of 171-172 topical pharmaceuticals 506-507
Index | 637

torque 12-Week Triathlete, The (Holland) 460


in cycling power 72 Type II muscle fibers. See fast-twitch muscle
in pedaling mechanics 74-77, 75f, 76f fibers
in swimming 58 Type I muscle fibers. See slow-twitch muscle
Tour de France cyclists 172, 173f fibers
traction (medical) 507
training adaptations
in anaerobic training 303 U
body composition 15 UBT 227
cardiopulmonary 12-15 UCI 113
economy improvement 14-15, 173-176, 177f ulnar nerve, in cyclist palsy 505
evaluating in run base building 380 undulating (nonlinear) periodization 317-318,
event duration and 304 375-377
heat tolerance 15, 205-206 uniforms 106-107, 107f, 442
in low-intensity aerobic training 365-366 Union Cycliste Internationale (UCI) 113
maintenance of 300 United States Pharmacopoeia (USP) mark 544-545
metabolic and physiologic 271t unsaturated fats 520
metabolic efficiency 271-273, 272f up-and-over shoulder stretch 232, 232f
muscular 22-24, 191, 194-197 upper-body torque (UBT) 227
nutrition effects on 525 urine color and frequency 531, 532
rest or recovery in 313 USA Triathlon. See USAT rules and guidelines
VO2max improvement 167-169, 286 USAT rules and guidelines
training analysis software 155-156, 156f, 157f cycling equipment and racing 113, 127
training log uses duathlon distance standards 431
goal setting 558 swimsuits 108
injury cause determination 453 USP mark 544-545
overtraining prevention 347
training plans, based on health assessment 451-
454. See also specific race distances V
training residuals 316-317
training stress balance (TSB) 319-324, 320f, 321f, variety, lacking in training 565
323f, 336f VAS 454
training stress score (TSS) 319-320, 335, 336f VCOM 84
training time 150, 252, 253 velocity, in swimming 59
training tools. See technology training tools vena cavae 10
training zones ventilatory threshold (VT) 182-183
cross-reference chart 388t vertical oscillations, in running 90-91, 91f
heart rate 145, 146t-148t vertical running center of mass (VCOM) 84
inconsistency in defining 387-389 viscosupplementation 508
power levels 151, 151t visual analog pain scale (VAS) 454
transitions. See also bike-to-run transition visualization
bike-to-run tips 444 for peak performance 560-561, 579-582
in duathlon 444-445, 445f of warm-up 212-213
in sprint events 397-398 vitamin C 526
swim-to-bike 442-443, 443f vitamin D 547
training for 445-448 vitamin E 526
videos of 441 vitamins 521
transmutation mesocycle 315 V-line drill 280, 280f
treadmill training 140-141, 141f volume of training
Triathlete’s Training Bible, The (Friel) 145, 255 age-related decline in 51
Triathlon Canada 106 brick workouts 446-447
tri-bikes 115. See also bicycles in cycling base building 366, 370
trochanteric bursitis 487 guidelines 301, 309
tropomyosin 16, 18f, 19, 20 interval vs. aerobic 287-288
troponin complex 16, 18f, 19, 20 for Olympic-distance events 400
TSB 319-324, 320f, 321f, 323f, 336f during taper 329
Ts on physioball 464, 464f VO2max
TSS 319-320, 335, 336f age-related decline in 48, 49
T1 transition. See swim-to-bike transition gender differences in 40, 164
T2 transition. See bike-to-run transition genetic effects on 31
T-tubules 16, 17f, 19 historical view 161-162
interval training effects on 285-286, 289, 289f
638 } Index

VO2max, continued water running 501


measurement of 163-164, 163f water temperature
as overvalued 161 pool vs. open-water 102
relation to economy 172 race modifications due to 104-105, 105t
training effects on 13-14, 164, 167-169 for wetsuit use 108, 108t
variation in 164-165, 165t wattage measurements 149
warm-up effects on 212, 217 wave drag 64-65, 64f
VO2max plateau 162 weaknesses. See limitations
VO2max velocity 290 weather, in race tactics. See also heat and humidity
VO2peak 162 duathlon 438
VT 182-183 half-Ironman 416
Ironman 428-429
Olympic-distance events 407
W sprint events 398
wall angels 469, 469f weighted vests 382
walnuts 527 weight gain, from strength training 252, 253
warm-ups weight loss 523, 531, 531t. See also body weight
active vs. passive 211 Wellington, Chrissie 45, 49t
benefits of 211, 212-213, 213f wetsuits
duration of 214-216 performance and 61, 63, 108-109
for half-Ironman 215, 416 removing in transition 442
intensity and specificity 216-217 water temperature for 108, 108t
for Ironman 215 windmill deadlift 490, 490f
for Olympic-distance events 215, 217t Wingate tests 114
risks in 213-214 World Triathlon Corporation (WTC) 409, 419
for sprint events 215, 217t, 396
timing of 217-219, 218f
water, as nutrient 520-521. See also hydration
Z
water quality, in swimming 106 zinc 526
About the Editors

Joe Friel, MSc, has trained endurance athletes since 1980. He served as head
coach of the USA National Triathlon Team at the World Championships in
2000, and athletes he has worked with have appeared in the Olympic Games,
world and national championships. He is co-founder of the USA Triathlon’s
National Coaching Association and served on the USA Triathlon Coaching
Certification Committee. Friel is a Colorado State Masters Triathlon cham-
pion, a Rocky Mountain region and Southwest region duathlon age-group
champion, and has been a perennial USA Triathlon All-American duath-
lete. As a member of several national duathlon teams, Friel was a top-five
contender in world class events and competed in road running and United
States Cycling Federation races. He is the author of The Triathlete’s Training
Bible, Your First Triathlon, Your Best Triathlon, Total Heart Rate Training, and The
Paleo Diet for Athletes, and he is a contributor to Precision Heart Rate Training
and USA Triathlon’s Complete Triathlon Guide.

639
640 } About the Editors

Jim Vance is a triathlon/duath-


lon, running, and cycling coach
at TrainingBible Coaching and
the founder and head coach of
TriJuniors, a USAT High Perfor-
mance Team in San Diego. For
his coaching, he was awarded the
2009 Tri Club of San Diego Coach
of the Year and was appointed US
Elite National Team coach for the
Duathlon World Championships
in 2011 and 2012. He has coached
athletes who have won or quali-
fied for events including the US
Elite National Championship,
Elite ITU World Championship,
Ironman World Championship,
70.3 World Championship, and
XTERRA European Tour Elite. A
former elite triathlete who spent
time at the US Olympic Training Center, Vance placed third in the Florida
Ironman and was an International Triathlon Union Age Group World
Champion, a XTERRA Amateur World Champion, and a Letter winner at
the University of Nebraska in track and field and cross country.
About the
Contributors

Ildus I. Ahmetov is the head of the Laboratory


of Molecular Genetics at Kazan State Medical
University. He also serves as a Senior Research
Fellow of the Sports Genetics Laboratory at the
St. Petersburg Research Institute of Physical
Culture and as a Senior Research Fellow of the
Laboratory of Exercise Physiology at the Insti-
tute for Biomedical Problems of the Russian
Academy of Sciences. He won Young Investi-
gators Awards in 2007 and 2010 for projects
documenting genetic variants in athletes and
is author of Molecular Sports Genetics.

Hunter Allen is a former professional


cyclist, renowned coach, and expert in using
power meters to train endurance athletes.
As a professional racer for 17 years, he
earned more than 40 career victories in com-
petitions around the world. Upon retiring
from racing, Allen became a USA Cycling
elite-level cycling coach and certified nutri-
tion consultant. He has coached more than
400 athletes, including the 2008 USA Cycling
BMX Olympic team, champions of the
European road racing circuit, and champion
mountain bikers. He writes for Road maga-
zine and Cycling Weekly, and he coauthored
Training and Racing with a Power Meter and Cutting-Edge Cycling. Allen is the
founder of Peaks Coaching Group, and is a co-developer of TrainingPeaks
WKO software, a leading program for analyzing data from power meters.

641
642 } About the Contributors

Gale Bernhardt is a triathlon, cycling, and


endurance coach. She served as the 2004
USA Triathlon Olympic coach for both the
men's and women's teams and was selected
by USA Triathlon to serve as the 2003 Pan
American Games men’s and women’s coach.
She has served as a USA Triathlon World
Cup coach for the International Triathlon
Union (ITU) Sport Development squad, and
has worked internationally for the ITU as an
expert World Cup coach. She has worked
with Olympic athletes and winners of the
USA Cycling Pro National Championship
race. Bernhardt is author of five books for triathletes: Training Plans for Mul-
tisport Athletes; Triathlon Training Basics; Bicycling for Women; Training Plans
for Cyclists; and Swim Workouts for Triathletes. She is also a contributor to The
Woman Triathlete.

Jeff Broker, PhD, has been involved in the study


of pedaling mechanics and optimal integration
of rider and bicycle since 1987. He worked at the
United States Olympic Committee (USOC) as a
Senior Sport Biomechanist for nearly 10 years,
through which he worked with the USA Triath-
lon and US National Cycling team to optimize
rider/bicycle integration. He has worked with
nine United States sports federations including
the US Cycling Federation, USA Triathlon, and
USA Track & Field. Broker is author of Bicycle
Accidents: Biomechanics, Engineering, and Legal
Aspects and is a contributor to High-Tech Cycling and International Olympic
Committee’s Olympic Handbook of Sports Medicine: Road Cycling. He currently
serves as an Associate Professor and Chair of the Biology department at the
University of Colorado.
About the Contributors | 643

Professor Malcolm Collins leads a productive


research group investigating the biological
mechanisms of musculoskeletal soft tissue
injuries and endurance performance. His
research interests include genetic elements
that determine the endurance phenotype and
the inter-individual physiological responses
during participation in endurance events. He
has published more than 70 papers in scientific
journals and book chapters, several focused on
the Ironman triathlete. He serves as a professor
in the Research Unit for Exercise Science and
Sports Medicine (ESSM) at the University of
Cape Town.

JoAnn Dahlkoetter, PhD, is founder of


Performing Edge Coaching International,
a global resource and certification training
program for sports psychology coaches. She
is an internationally recognized keynote
speaker and world-class athlete. She is a
past winner of the San Francisco Marathon
(2:43:20), placed 2nd in the Hawaii Ironman
Triathlon, and was rated the No. 1 triathlete
in the U.S. by Triathlete Magazine. The author
of the bestselling book Your Performing Edge,
her work has been published in Runners
World, Fitness Magazine, Time, and Sports Illus-
trated. She is currently a regular contributor
to the Huffington Post and Triathlete Magazine. Dr. Dahlkoetter has appeared
as an expert guest on numerous shows including Oprah and Friends, ABC
sports, and NBC Olympics. She is a licensed clinical psychologist and
medical staff member at Stanford University Medical Center and currently
maintains a fulltime private practice. In her 30 years of clinical practice Dr.
Dahlkoetter has worked with five Olympic Gold Medalists and numerous
Olympic and professional athletes. She continues to train and race in the
San Francisco Bay Area.
644 } About the Contributors

George M. Dallam, PhD, a 30-year triath-


lete participant and coach, is the found-
ing member of the National Coaching
Commission of USA Triathlon and was
USA Triathlon's first national team coach.
Throughout his coaching career, Dallam
has served as a personal coach to several
elite triathletes including Hunter Kemper,
Amanda Stevens, Marcel Vifian, Callahan
Hatfield, Michael Smedley, Ryan Bicker-
staff, Nick Radkewich, Susan Williams,
Laura Reback, Becky Lavelle, and Doug
Friman. Athletes under his direction have
won seven National Elite Championships,
Pan American Games gold and silver medals, World Cup medals, World and
USA Age Group Championships, as well as qualifying for and being among
the top American male finishers in three Olympic Games. In 2005, he was a
finalist for the United States Olympic Committee’s esteemed “Doc” Counsil-
man Award for Science in Coaching. He was named USA Triathlon's Elite
Coach of the Year in 2006. Dallam has been a professor in Exercise Science
and Health Promotion at Colorado State University at Pueblo for 15 years. As
a sport scientist he has authored and coauthored numerous scientific papers
and books relating to triathlon, including Championship Triathlon Training. He
is a contributor to USA Triathlon’s Complete Triathlon Guide.

Matt Fitzgerald is a running and triathlon coach,


sports nutritionist, and USA Triathlon All-American.
He placed second overall in the 2004 Long Beach
Triathlon and was an Ironman finisher. He has
contributed to many triathlon publications includ-
ing Triathlete Magazine (for which he was the former
senior editor), Triathlete, and Inside Triathlon. He has
been featured on TrainingPeaks and Active.com.
His many writing accomplishments include authoring Triathlete Magazine’s
Complete Triathlon Book, Triathlete Magazine’s Essential Week-by-Week Training
Guide, Racing Weight, Iron War, RUN: The Mind-Body Method of Running by
Feel, and Brain Training For Runners and co-authoring Run Faster from the 5K
to the Marathon and The Runner's Edge.
About the Contributors | 645

Neal Henderson is a triathlon coach, the


sports science director at the Boulder Center
for Sports Medicine, and a consultant to USA
Cycling, USA Triathlon, and Specialized Bicy-
cle Components. He has coached numerous
elite triathletes including Cameron Dye, Flora
Duffy, Seth Wealing, and Jamie Whitmore,
and has worked with cyclists Taylor Phin-
ney and Roman Kreuziger. He was named
USA Cycling National Coach of the Year in
2009 and USA Cycling Developmental Coach
of the Year in 2007. He is a USA Triathlon
Level 3 Elite Coach and USA Cycling Level
1 Elite Coach. He was a member of the USA
Triathlon National Coaches Commission from 2003 to 2008. As a triathlon
participant, he was a Top American Finisher at the International Triathlon
Union Winter Triathlon World Championships in 2002 and 2003 and was
1st Place Amateur Overall at XTERRA Keystone in 1999.

Nathan Koch is founder, director,


and owner of Endurance Rehab LLC,
the official training center for Joe
Friel's Ultrafit. He is also an instruc-
tor of Advanced Bike Fit Classes at
the Serotta International Cycling
Institute. He writes regularly about
sports injuries and recovery for Tri-
athlete Magazine and LAVA Magazine.
Koch has provided sports medicine to
NCAA Division I men's and women's
athletics at Saint Louis University and
the University of Nebraska, and he
spent three seasons working for the
St. Louis Rams and a sports medicine
clinic (ProRehab) in St. Louis, where he
worked with professional athletes. He has presented “Core Strengthening
for the Endurance Athlete” to teams of elite medical professionals and has
worked closely with them to provide the most technologically-advanced
sports medicine available.
646 } About the Contributors

Sean Langlais is a consultant for Biome-


chanics Engineering, where he examines
the biomechanics of injuries, including
those occurring due to cycling. He recently
had his “Grip Pressure Distributions and
Associated Variability in Golf: A Two-Club
Comparison” published in the Journal of
Biomechanics. After receiving his MSc in
Exercise Science from the University of
Colorado, Langlais worked with Carmi-
chael Training Systems where he helped
train cyclists and triathletes in endurance.
At CTS, he also served as the in-house bicycle mechanic. He has also worked
with USA cycling as the team mechanic for the Junior National Mountain
Bike skills camp.

Romuald Lepers is an assistant professor at the


Faculty of Sport Sciences in Dijon at Burgundy
University in France where his research interests
include exercise physiology, neuromuscular
fatigue, and the effects of age and gender on
endurance performances. He has published
more than 70 articles, many on triathlon and
endurance in publications including Journal
of Sports Sciences, European Journal of Applied
Physiology, and Scandinavian Journal of Medicine
& Science in Sports. He is a frequent triathlon
participant, having completed 21 Ironman races,
four of which were the Hawaii Ironman.
About the Contributors | 647

Bruce R. Mason is the head of the Aquatics


Testing Training and Research Unit at the Aus-
tralian Institute of Sport where he provides bio-
mechanical servicing to the Australian Institute
of Sport swim team and Australian swim team.
He was the chief biomechanical advisor to the
Australian swim team from 1994 to 2002, for
which he received the Order of Australia Medal
and Australian Sports Medal from the Austra-
lian government. He also twice received awards
for Outstanding Contribution to Swimming in
Australia from the Australia Swim Coaches &
Teachers Association. His research in swim-
ming biomechanics has led him to present at ISBS and BMS international
conferences. Mason has contributed to three books on swimming: The Swim
Coaching Bible, World Book of Swimming, and Triathlon Into the Nineties.

Stephen J. McGregor, PhD, a former triathlete


and elite competitive cyclist, coached endur-
ance athletes for more than 15 years, advising
numerous cyclists, triathletes, and runners at
the national and international level. He has
been appointed as physiological advisor to both
the 2008 U.S. Olympic BMX team and Eastern
Michigan University men's cross country/dis-
tance track team and has presented at numerous
conferences including USA Triathlon Art and
Science of Coaching. He is a USA Cycling Level
I coach and is an instructor for the USA Cycling
Level II Coaching Certification Sports Sciences,
USA Cycling Power Training Certification Course, USA Cycling Level I (Elite)
Coaching Certification Interval Training, and South African Coaching and
Power Training Certification. He is currently the director of both the Applied
Physiology Laboratory and the Human Factors-Dynamical Systems Labora-
tory at Eastern Michigan University. He is co–author of The Runner's Edge.
648 } About the Contributors

David Pease is a senior biomechanist and the


deputy head of discipline in the Aquatic Testing,
Training, and Research Unit at the Australian
Institute of Sport where he conducts research into
enhancing swimming performance and helps the
swim program monitor athletes during training
sessions. With more than 20 years’ experience
in swimming biomechanics, he has served as a
biomechanist for USA Swimming and New Zea-
land Swimming. He has also worked with United
States Swimming’s International Center for Aquatic
Research (ICAR) and the United States Olympic
Committee’s Sport Science division. He was a competitive swimmer from
the age of 6 through university level, when he swam for the University of
Southern California.

John Post, MD, is the medical director at Train-


ingBible Coaching and an orthopedic surgeon at
Martha Jefferson Hospital in Charlottesville, Vir-
ginia. He is a widely-read medical and triathlon
author and lecturer who presents nationally and
runs his own endurance athlete blog. For more
than 20 years, his focus has included surgical and
nonsurgical knee and shoulder care in endurance
athletes. A veteran triathlete, Post has finished the
Hawaii Ironman six times, in addition to participat-
ing in the Manhattan Island Marathon Swim and
the English Channel Relay Swim.
About the Contributors | 649

Gina Sacilotto is a research assistant in the


Aquatics Testing Training and Research Unit
at the Australian Institute of Sport Aquatics
where she assists Bruce R. Mason in research
for athletes and coaches to enhance perfor-
mance. An open-water competitive swimmer
herself, through her research she has explored
topics including active drag in swimming,
anthropometry in swimming, and the develop-
ment of new aquatics biomechanical systems
to enhance swimmer performance.

Bob Seebohar, MS, RD, CSSD, CSCS, is owner


of Fuel4mance, a leading nutrition consulting
firm serving amateur and elite athletes; Kids
that TRI, 501c3 Youth Triathlon Team; Perfor-
mance Webinars, an educational webinar com-
pany; and a co-owner of Elite Multisport coach-
ing. In 2008, Seebohar traveled to the summer
Olympic Games in Beijing as a sport dietitian
for the U.S. Olympic team and as the personal
sport dietitian and exercise physiologist for the
Olympic triathlon team. He has worked closely
with triathletes including Susan Williams (2004
Olympic Bronze medalist), Sarah Haskins, and Jasmine Oeinck (2009 National
Elite Champion). He has also been appointed a sports nutrition consultant
for the USA Triathlon Olympic and Developmental teams. Seebohar is a USA
Triathlon Elite Level III coach and a frequent presenter for USA Triathlon
coaching certification clinics. In 1996, he represented the United States as
a member of the duathlon team at the World Championships. He has com-
peted in numerous endurance events including the Boston Marathon and
six Ironman races. He is author of eight books on endurance and nutrition
including Performance Nutrition: Applying the Science of Nutrient Timing and
Nutrition Periodization for Athletes. He is a contributor to USA Triathlon’s
Complete Triathlon Guide.
650 } About the Contributors

Dr. Ross Tucker is senior lecturer with the


University of Cape Town's Exercise Science and
Sports Medicine department where he has stud-
ied topics including fatigue and the role of the
brain in determining pacing strategy and exer-
cise performance. He serves as a sports scientist
and strategist for the Springbok Sevens rugby
team and has recently worked with Olympic
kayakers Shaun Rubenstein and Mike Arthur in
preparation for the 2012 London Olympics. He
currently serves as the scientific editor of Run-
ner’s World South Africa and as a contributor and
editor to Health24, South Africa's largest fitness- and health-related website.
He is a consultant technical expert and physiologist with Adidas South
Africa and consults Discovery Health, Powerade, and Sports Illustrated. He
has co-authored Runner's World Magazine’s The Runner's Body and participates
regularly in 10k races and half-marathons.

David Warden is co-founder of PowerTri.


com. He is the founder of David Warden
Coaching where he coaches triathletes of
all abilities. An internationally-recognized
triathlon coach and overall winner of 19
triathlon events, he is former vice presi-
dent of the USA Triathlon Rocky Mountain
Regional Council. He has more than 15
years of experience in endurance sports,
having been named 2011 USA Triathlon
Rocky Mountain Region sprint-distance
champion and a three-time USA Triathlon
All-American. He also produces the Tri Talk
Triathlon Podcast, the #1 multi-sport podcast on iTunes, and he has had
articles published in Triathlete and Inside Triathlon magazines.
About the Contributors | 651

Randall L. Wilber is a senior sport physiolo-


gist at the U.S. Olympic Training Center in
Colorado Springs. He has worked with the
U.S. National Team in triathlon since 1993,
has worked with every U.S. Olympic Team
in triathlon since 2000, has been named a
member of the official U.S. Olympic Team
delegation in three summer Olympics and
three winter Olympics, and has provided
support for Team USA at two Pan American
Games. He has worked with many famous
triathletes including Hunter Kemper, Barbara
Lindquist, Sheila Taormina, Nick Radkewich,
Susan Williams, Laura Bennett, and Matt Charbot. A well-published author,
his scientific papers have appeared in numerous journals. He is author of
two books, Altitude Training and Athletic Performance and Exercise-Induced
Asthma: Pathophysiology and Treatment. He has been an invited speaker at
several USA Triathlon coaching clinics as well as International Triathlon
Union sponsored conferences. Wilber has been recognized as a fellow of the
American College of Sports Medicine (ACSM) and has served on the ACSM
Olympic and Paralympic Sports Medicine Issues Committee since 2005. He
was named chair of that committee in 2009.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy