Contents

Title Page
Key to symbols used & Bibliography 4
Preface 5
Alekhine
1 Rare Lines 7
2 4...g6 27
3 4...Bg4 43
4 4...dxe5 63
Scandinavian
5 2...Nf6 96
6 2...Qxd5 119
7 5...Bg4 144
8 3...Qd6 163
9 5...g6 181
10 5...c6 207
Pirc/Modern
11 Early Deviations 244
12 Pirc with ...c6 270
13 Modern with ...c6 306
14 Pirc with ...a6 320
15 Modern with ...a6 339
16 Main Line with 6...Bb7 359
Miscellaneous
17 1...Nc6 394
18 1...b6 and Others 409

Variation Index 426

2
 Grandmaster Repertoire

1.e4 vs
Minor Defences
By
Parimarjan Negi

Quality Chess
http://www.qualitychess.co.uk

First edition 2020 by Quality Chess UK Ltd

Copyright © 2020 Parimarjan Negi

All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system or transmitted in any form or by any means, electronic,
electrostatic, magnetic tape, photocopying, recording or otherwise,
without prior permission of the publisher.

Paperback ISBN 978-1-78483-077-9

Hardcover ISBN 978-1-78483-078-6

All sales or enquiries should be directed to Quality Chess UK Ltd,

Suite 247, Central Chambers, 11 Bothwell Street,
Glasgow G2 6LY, United Kingdom
Phone +44 141 204 2073
e-mail: info@qualitychess.co.uk
website: www.qualitychess.co.uk

Distributed in North and South America by National Book Network

Distributed in Rest of the World by Quality Chess UK Ltd through
Sunrise Handicrafts, ul. Szarugi 59, 21-002 Marysin, Poland

3
 Typeset by Jacob Aagaard
Proofreading by John Shaw
Edited by Andrew Greet
Missing lines check by Renier Castellanos Rodriguez
Cover design by adamsondesign.com
Cover photo by capture365.com

4
 Key to symbols used
² White is slightly better
³ Black is slightly better
± White is better
µ Black is better
+– White has a decisive advantage
–+ Black has a decisive advantage
= equality
© with compensation
„ with counterplay
ƒ with an initiative
÷ unclear
? a weak move
?? a blunder
! a good move
!! an excellent move
!? a move worth considering
?! a move of doubtful value
™ only move
# mate

Bibliography
Bauer: Play 1...b6, Everyman Chess 2005
Bauer: Play the Scandinavian, Quality Chess 2010
Hillarp Persson: The Modern Tiger, Quality Chess 2014
Houska: Starting Out: The Scandinavian, Everyman Chess 2009
Kornev: A Practical Black Repertoire with Nf6, g6, d6 1, Chess Stars 2016
Kotronias: The Safest Scandinavian, Chess Stars 2016
Marin: Grandmaster Repertoire – The Pirc Defence, Quality Chess 2019
Moskalenko: The Perfect Pirc-Modern, New In Chess 2013
Negi: Grandmaster Repertoire – 1.e4 1-4, Quality Chess 2014-2016
Palliser, McNab & Vigus: Dangerous Weapons: The Pirc and Modern, Everyman Chess 2009
Schuyler: The Dark Knight System, Everyman Chess 2013
Shaw: Playing 1.e4 – Caro-Kann, 1...e5 & Minor Lines, Quality Chess 2016
Smerdon: Smerdon’s Scandinavian, Everyman Chess 2015
Taylor: Alekhine Alert!, Everyman Chess 2010

5
Vigus: Chess Developments: The Pirc, Everyman Chess 2012
Wisnewski: Play 1...Nc6!, Everyman Chess 2007

Electronic Resources/Periodicals

ChessBase: Mega/Correspondence Databases

The Week In Chess

6
 Preface

This book took a long time to finish. Partly, it was due to my ever busier and changing
professional life. But even more, it was because of the very nature of the systems covered in this
book: the Alekhine, Scandinavian, Pirc/Modern and other miscellaneous replies to 1.e4.

As a young kid, I was always scared of these openings: the Pirc and Modern were endlessly
confusing in move orders; the Scandinavian seemed to me to lead to depressingly dull positions;
and I never prepared any serious lines against the Alekhine, so I was always a bit afraid of that
too. As the years went by and my playing strength increased, fewer and fewer of my opponents
played these lines, so I could essentially get away with never preparing anything deep against
them. Still, my childhood uneasiness against these defences never quite left me.

When I finally started analysing these systems in earnest for this book, I realized that my
uneasiness was not unfounded, as there were just so many interesting possibilities for Black
hidden away in all these lines. Nowadays more than ever, with online chess events and faster
time controls becoming more prevalent, it is vital to have a well-constructed repertoire against
these lines.

In this book, I have endeavoured to address the very features of these defences that made me
uneasy against them. If there is a simple path to an advantage, I will generally favour that over a
more ambitious but intensely theoretical continuation – it does not feel worth the effort to
memorize unnecessarily long lines against inferior continuations which we are never likely to
face anyway. At the same time, there are plenty of cases when White simply must take a
principled approach in order to prove anything – this is a Grandmaster Repertoire book after all.
Still, I have strived to be as practical as possible. When applicable, I have pointed out how
certain variations are related to one another, and have shown how you can follow similar plans
against different options that Black may try. On a related note, I went to a lot of trouble to take
into account the plethora of possible transpositions, especially in and between the Pirc and
Modern, and have given my best efforts to make the proposed repertoire ‘move-order proof.’

The Alekhine, Scandinavian and Pirc/Modern Defences are generally regarded as less
theoretically intense than more popular openings such as 1...e5 and main-line Sicilians. But
paradoxically, my relative lack of practical or theoretical experience in these lines meant that I
was pushed to my analytical limits like never before. I would like to thank my many coaches
who have helped me see different ways of analysing and understanding such challenges. In
particular, I owe my style of chess analysis the most to Vladimir Chuchelov. In this series, I can

7
only hope to match the kind of rigour he demands from his analysis.

Finally, I would like to thank my editor Andrew Greet, who has been a constant throughout this
journey, and dealt effectively with both my writing and my procrastination.

Parimarjan Negi
Cambridge, Massachusetts, July 2020

8
A) 4...Bf5!? 9
B) 4...c6 10
C) 4...Nb6!? 5.a4 a5 6.Nc3 11
C1) 6...Bf5 13
C2) 6...g6 7.exd6 14
C21) 7...exd6 14
C22) 7...cxd6 15
D) 4...Nc6!? 5.c4 Nb6 6.e6 fxe6 7.Nc3 g6 8.h4 Bg7 9.Be3 e5 10.d5 Nd4 11.Nxd4 exd4
12.Bxd4 Bxd4 13.Qxd4 e5 14.Qe3 17
D1) 14...Qe7 22
D2) 14...Bf5 24
D3) 14...0-0 25

9
1.e4 Nf6
Alekhine’s Defence has never really caught on at the highest levels, probably because it
presents White with a space advantage and a lot of freedom. Still, it is not a bad opening, and
many a White player has come unstuck after advancing the central pawns too enthusiastically
and eventually becoming overextended.

2.e5 Nd5
Occasionally Black retreats the knight to its original square:
2...Ng8
You are unlikely to ever face this in a serious game. If you do encounter it, then simple
development will work fine, as shown after:
3.d4 d6 4.Nf3 dxe5
4...Bg4?! 5.h3 Bh5 (5...Bxf3 gives up the bishop early and with no other pieces developed,
which makes no sense for Black) 6.e6!± gives White a great version of a thematic pawn
sacrifice.
4...c6 5.Be2 Bf5 (5...dxe5 6.Nxe5 Nd7 7.Nf3 is similar to variation C2 of Chapter 4, and
the best Black can hope for from here is to transpose to it) 6.0-0 dxe5 7.Nxe5 Nd7 8.Nf3
Ngf6 9.c4 e6 We have transposed to variation C3 of Chapter 4; see page 82 for the
continuation.
5.Nxe5

10
5...Nd7
This is Black’s only real attempt to benefit from having his knight on g8 instead of d5, since
the Nxf7 sacrifice gets nowhere here.
6.Bd3!?
This seems to be the simplest route to an advantage.
6.Nf3 Ngf6 7.Be2 g6 8.0-0 Bg7 resembles variation C2 of Chapter 4, but Black has not had
to commit to ...c6 yet, which may benefit him slightly.
6...Nxe5
6...Ngf6 7.0-0 g6 makes no attempt to dislodge the strong knight on e5, so White’s position
is pleasant.
7.dxe5 Qd5
7...c6 8.0-0± is too slow for Black.

8.0-0!N

11
 A simple novelty.
Giving up the g2-pawn with 8.Qe2 gave White some compensation in Prandstetter –
Vesely, Prague 1977, but it is much easier to develop an initiative after sacrificing on e5, as
White will be left with open central files and diagonals, and pieces which are perfectly
poised to exploit them.
8...Qxe5 9.Re1 Qf6 10.Nc3 e6 11.Nb5 Bd6 12.Bd2 Ne7 13.Nxd6† cxd6 14.Bb4 d5 15.Bc3 Qg5
16.Re3±
Black fails to complete development and get his king to safety.

3.d4 d6

4.Nf3
Simple development works well.

One of the challenges of preparing against the Alekhine is the sheer number of options available
to Black. In this chapter we will deal with four of the relatively minor ones: A) 4...Bf5!?, B)
4...c6, C) 4...Nb6!? and D) 4...Nc6!?.

The three main lines of 4...g6, 4...Bg4 and 4...dxe5 will be analysed in Chapters 2, 3 and 4
respectively.

A) 4...Bf5!?

With this slightly tricky sideline, Black develops a piece while hoping for an improved version
of the ...dxe5 structure.

5.Bd3

12
 This is the simplest route to an edge.

5.Be2 allows 5...dxe5! 6.Nxe5 Nd7 when, compared with variation C3 of Chapter 4 on page 81,
Black has not had to spend a tempo on ...c6.

5...Bxd3 6.Qxd3
There is no clear ‘rule’ as to who is more likely to benefit from the exchange of light-squared
bishops in this structure. However, in the present position there are clear weaknesses in Black’s
position and White will develop much faster, so we have every reason to be happy.

6...e6
This seems like Black’s best move order, as I do not believe the greedy Qb5† to be
worthwhile.

6...Nc6 has been the more popular choice but it gives White the additional option of 7.e6!?, as
well as the simple 7.0-0 e6, reaching the main pathway.

7.0-0 Nc6 8.c4 Nb6 9.exd6 cxd6

9...Bxd6 leads to a type of position which is almost always worse for Black, since there isn’t
any clear plan for him: 10.Nc3 Be7 11.Bf4 0-0 (11...g5?! 12.Bg3 g4 13.Ne5 Nxd4 14.c5 Bxc5
15.Rad1ƒ was also excellent for White in Topalov – Kamsky, Wijk aan Zee 2006)

12.Rad1 Qd7 13.Ne4 Nb4 14.Qb3 Nc6 15.Nc5 Qc8 16.Qc3± Dreev – Vlassov, Internet 2001.

10.Nc3 Be7

13
11.d5!N
In several games White has played less adventurously, in which case he should be able to
maintain a slight edge, but the text move seems like the best way to exploit White’s lead in
development.

11...Nb4
11...exd5 12.cxd5 Ne5 13.Nxe5 dxe5 14.Rd1 Bd6 15.Ne4 0-0 16.Ng5± is also excellent for
White.

12.Qe4 e5 13.c5
White’s initiative flows naturally, for instance:

13...dxc5 14.a3 Na6 15.Rd1 0-0

15...Bd6 should be met by: 16.Qg4! (rather than 16.Nxe5 0-0÷) 16...0-0 17.Bh6±

14
16.d6 Bf6
16...Bxd6 is met by 17.Nb5 of course.

17.Qxb7 Qc8 18.Qxc8 Rfxc8 19.Ne4±

White has regained his pawn while keeping an obvious advantage.

B) 4...c6

This is a fairly common move although usually it transposes to a later chapter.

5.Be2 g6
This is the only serious option with independent significance. Black is trying to avoid ...dxe5,
and may be looking to provoke exd6. At first, this may look like a good thing for Black, since the

15
former option gives White an active knight on e5, while the latter enables Black to develop his
queen to d6. However, as we will see in Chapter 4, the knight on e5 can become a target,
whereas now the queen on d6 can become a target, enabling White to gain time.
5...Bg4 leads to variation A of Chapter 3 on page 44.
5...dxe5 6.Nxe5 and 5...Bf5 6.0-0 dxe5 7.Nxe5 transpose to variations C and C3 of Chapter 4,
on pages 77 and 81 respectively.

6.c4 Nc7

7.exd6! Qxd6
7...exd6 gives Black a solid structure but his pieces are still undeveloped, and 8.Bg5 provokes
a further weakness in his position: 8...f6 (8...Be7 9.Bh6±) 9.Bh4 Bg7 10.Nc3 0-0 11.Qd2 a6
12.0-0² Glatt – Forgacs, Hungary 1989.

8.Nc3 Bg7 9.0-0 0-0

9...Bg4?! is met by the typical reaction of 10.Qb3±.

10.h3!
10.Ne4 Qd8 11.h3 Nd7 12.Bf4 Nf6 is less convincing.

The text move is simple and strong. The queen on d6 isn’t doing Black any favours, so there is
no point in chasing her away just yet. If she voluntarily retreats to d8, then the position will
resemble variation C2 of Chapter 4, but Black will have wasted time moving the queen to d6 and
back, rather than White losing time with Nxe5 followed by Nf3.

10...Ne6
10...Nd7?! invites 11.Ne4 when Black’s queen gets into trouble.

16
10...Qd8 11.Re1 Nd7 12.Bf4± is excellent for White.
The text move is the best attempt to make Black’s position work, but the knight mostly just gets
in the way on e6. It would be justified if Black could exert meaningful pressure on d4 – but he
can’t.

11.Be3 Nf4
11...Rd8 12.Qb3±

12.Re1 Nxe2† 13.Qxe2

Black will continue to struggle to coordinate his pieces. The light-squared bishop is the hardest
piece for Black to develop, so his bishop pair is not particularly useful. White’s plan may be as
simple as Qd2 and Bh6, followed by jumping around with the knights.

C) 4...Nb6!?

17
 Prior to working on this book I had never paid attention to this move, but I found it surprisingly
interesting. My impression of the existing games is that White has not been able to demonstrate a
clear path to an advantage, while Black remains flexible and has avoided being hit by an early
c2-c4, forcing White into slightly more unusual set-ups.

5.a4
Provoking ...a5 seems to make sense.

5.c4 Bg4! forces White to release the tension in the centre because of ...dxe5 threats.

5.Be2!? could be worth exploring, but after 5...g6 (5...Nc6 6.0-0 g6 7.exd6 cxd6 8.d5! Ne5
9.Nxe5 dxe5 10.c4 was good for White in Carlsen – Nakamura, Internet 2016) 6.0-0 Bg7 Black
has a solid set-up and it is up to White to prove what plans he might have.

5...a5
As usual, 5...Nc6?! invites 6.e6! since ...Bxe6 would lose a piece, and after 6...fxe6 7.Ng5
White has a strong initiative for a mere pawn.

5...c6?! was once played by Larsen in 1977 but it is not a good move. I suggest 6.a5 Nd5 7.c4
Nc7 8.Nc3 g6 and now 9.exd6N 9...Qxd6 10.Be2 Bg7 11.0-0 0-0 12.h3± gives White a clear
advantage.

6.Nc3
I also investigated 6.exd6 exd6!? (after 6...cxd6 White may be able to find a better square for
his knight than c3, thus improving on some of the 6.Nc3 lines where we follow up with exd6)
7.Bd3 Be7 when White’s position is of course slightly preferable but Black is solid, and such an
edge can easily disappear if White doesn’t find something convincing.

18
Another possibility is 6.Bb5† c6 7.Bd3, intending to meet 7...g6 with 8.exd6, avoiding the
...cxd6 lines as featured in variation C22 below. However, Black can do better with 7...dxe5!
followed by ...g6 and so on, with a solid position.

We will consider C1) 6...Bf5 and C2) 6...g6.

6...Bg4?! doesn’t make any sense here in view of: 7.h3 Bh5 (7...Bxf3 8.Qxf3±) 8.g4 Bg6 9.e6
This thematic positional pawn sac should always be on your radar when you play against the
Alekhine. 9...fxe6 10.Ng5 Qd7 11.Bb5 c6 12.Bd3 Black’s position feels like it might collapse at
any moment. An illustrative line is:

12...Bxd3 13.Qxd3 Na6 This was Kaufeld – Schlawin, Ruhrgebiet 2004, and now 14.Nxh7N
14...0-0-0 15.Ng5 Nb4 16.Qe2± would have regained the pawn while keeping White’s
positional advantages.

6...dxe5 7.Nxe5 N8d7 gives White a few options but the simplest is:

19
8.Qf3! Nxe5 (8...Nf6 is well met by 9.d5!N 9...g6 [9...Nbxd5? 10.Bb5†! c6 11.Nxc6! bxc6
12.Bxc6† Bd7 13.Nxd5 Bxc6 14.Nxf6†+–] 10.Bb5†±) 9.dxe5 How should Black even continue
developing? 9...c6 (9...g6 10.Be3±)

This position was reached in Radovanovic – Drazic, Backa Palanka 2002, and now 10.Be3!?
Nƒ would have set up a nasty threat of Rd1, so Black is still unable to complete the development
of his kingside.

C1) 6...Bf5

20
7.d5!N
This strong new move spells trouble for Black’s development.

7.Bd3 has been tried by a few strong players but it doesn’t seem so easy for White. After
7...Bxd3 8.Qxd3 e6 9.exd6 Bxd6 the structure reminds me of French ...dxe4 positions, but Black
has got rid of the bad light-squared bishop. The computer evaluates the position in White’s
favour, but such an edge can disappear too easily.

7...e6
Other moves are worse, for instance:

7...dxe5 8.Nxe5 g6?! (8...N8d7 9.Bb5±) 9.Bb5† N8d7 10.g4+–

7...g6 8.e6! fxe6 9.Nd4 exd5 10.Nxf5 gxf5 11.Nxd5 Nxd5 12.Qxd5±

8.dxe6 Bxe6
8...fxe6 seems somewhat ugly. 9.Bg5!? (9.Bd3 dxe5 10.Nxe5 Qf6 11.Nf3² also seems nice for
White) 9...Be7 10.Bxe7 Qxe7 11.exd6 cxd6 12.Qd2 White has a clearly better pawn structure.

9.Bg5 Qd7
9...Be7?! 10.Bxe7 Qxe7 11.exd6 cxd6 12.Qd4±

21
10.exd6 Bxd6
10...Qxd6?! 11.Qxd6 Bxd6 12.Nb5± makes life easier for White.

11.Nb5!
Black cannot save the bishop pair.

11...Nd5
11...Be7? is refuted by: 12.Qxd7† Kxd7 13.0-0-0†+–

11...0-0 12.Nxd6² gives White an easy edge.

12.Nxd6† Qxd6

22
13.c4 Nb4 14.Qxd6 cxd6 15.0-0-0²
White clearly has the more pleasant side of the endgame, due to his bishop pair and the weak
d6-pawn.

C2) 6...g6

This is the most natural and popular continuation.

7.exd6
7.h4!? feels tempting, but there is no need for such aggression when most of our pieces are
undeveloped.

Below we will analyse C21) 7...exd6 and C22) 7...cxd6.

C21) 7...exd6 8.Bg5

This is an annoying move for Black, who will have to weaken himself with ...f6.

White can also consider delaying Bg5 with:

8.Be2!?N 8...Bg7
8...Be7? 9.Bh6 is clearly unpleasant.
8...d5 9.Bg5! f6 10.Bf4 is similar to the main line below, but Black has already committed
to ...d5, which isn’t so great. After 10...Bd6 White keeps a pleasant position, after either
exchanging on d6 or moving his bishop to h6. In either case, Black has some dark-square
weaknesses on the kingside.
9.Bg5 f6
9...Bf6?! is well met by 10.Bh6±.

23
10.Be3 d5
The position is similar to the main line, although we will see below that putting the bishop on
f4 is even nicer in this structure.

8...f6
8...Be7?! allows the typical 9.Bh6±.

9.Bf4!?N
9.Be3 d5 has occurred in a couple of games, in which White struggled to show anything. I
believe White’s position should be slightly better though, with the strongest plan being to castle
on the queenside and then try to expand on the kingside.
The point of the text move is that the bishop will be more actively placed after ...d5.

9...d5
An important alternative is:
9...Bg7 10.Bb5†!
If White played more slowly then, after both sides castle on the kingside, Black may be able
to obtain counterplay with a timely ...g5, turning the placement of the bishop on f4 into a
negative.
10...c6
10...Bd7 is safer, but after 11.Qe2† Qe7 12.Qxe7† Kxe7 13.0-0-0² the endgame seems
more pleasant for White.

24
11.Qe2†! Kf7
11...Qe7? has a simple tactical flaw: 12.Bxd6! Qxe2† 13.Bxe2±
12.Ne4! cxb5 13.Nxd6† Kg8 14.Qxb5
White has two pawns plus a useful initiative for the piece. Black’s only way to wriggle his
king to safety is:
14...h6
14...Nc6? 15.Qb3†+–
15.0-0 Kh7 16.Nf7 Bd7 17.Qd3²
White wins back an exchange, resulting in a somewhat advantageous position with a rook and
two pawns versus two minor pieces.

10.Bd3 Bd6
Black has no real choice, as 10...Bg7?! 11.Nb5 Na6 12.0-0 0-0 13.c3± leaves his pieces
uncomfortably stuck.

11.Bxd6 Qxd6 12.0-0 0-0 13.Qd2²

White has a typical slight plus, and the exchange of dark-squared bishops could make Black’s
weaknesses on the kingside more vulnerable in the long term.

C22) 7...cxd6

25
8.d5!N
Most games in this line have continued with:
8.Bb5† Bd7
White’s main problem here is to try and find a good plan. Black’s position is solid, and it is
hard for White to cause problems by simply manoeuvring his pieces around.
9.0-0
White has tried going for a quick h4-h5 in some games but I do not find it all that
convincing, despite White’s good practical results so far.
9...Bg7 10.Re1 0-0 11.Bg5
11.Bxd7 Qxd7 12.b3 Na6 may also be slightly better for White, but Black’s position
remains solid and playable.
11...Re8
It is no easy task for White to demonstrate an advantage. Black’s position is solid as usual, and
there are many ways for him to continue developing, including ...Bc6 and ...Na6-c7.

The point of the text move is revealed after:

8...Bg7 9.Be3 0-0 10.Bd4

White intends to exchange the dark-squared bishops and start an attack. The general character
of the play is reminiscent of the Sicilian Dragon. As we will see, White has good chances
although a certain amount of precision is needed.

10...N8d7
With White’s king still a couple of moves away from castling, Black may be tempted to open
the centre with 10...e5. However, any danger to the king proves to be illusory after 11.dxe6 Bxe6
12.Bxg7 Kxg7 13.Qd4† followed by 0-0-0, when Black has only succeeded in weakening his d-
pawn.

26
11.Bxg7 Kxg7 12.Qd4† Nf6

13.Nd2!
13.0-0-0 was the first move I looked at, but 13...Bg4! 14.h4 Rc8 gives Black a lot of play.

The text move makes it harder for Black to find an active role for his bishop, and the plan of
Nde4 is surprisingly hard to deal with.

13...Bd7
13...e5?! 14.dxe6 Bxe6 15.0-0-0± is excellent for White.

13...Kg8 avoids the pin, but 14.0-0-0 Bd7 15.h4 h5 16.Nc4!? Nxc4 17.Bxc4± leaves Black with
virtually no counterplay on the queenside, while White is dominating on the kingside.

14.Nde4
14.h4 Rc8 15.h5 is strongly countered by: 15...Rxc3! 16.bxc3 Nbxd5©

14.Nc4 Nxc4 15.Bxc4 Qc7 16.0-0-0 Rac8 might offer White an edge, but it is vanishingly small.

14...Rc8

27
15.h4!
The pin is surprisingly uncomfortable for Black.
15.Nxf6 releases the tension too soon – White should wait for ...Kg8 before making this
exchange.

15.0-0-0? Nbxd5! 16.Nxd5 e5 17.Qe3 Nxe4 18.Qxe4 Bf5 is a nasty tactic which we must be
aware of.

15...h5
15...Qc7?! is too slow in view of 16.h5!±.

I also analysed 15...Kg8 16.Nxf6† exf6 17.0-0-0 Qc7 (17...Re8 18.h5±; 17...h5 18.Ne4±)
18.Rd2 Qc5 when Black forces the queens off, but his problems persist after:

28
19.Qxc5 dxc5 (19...Rxc5? 20.Ne4! shows one of the problems with the ...Kg8 retreat: all these
Ne4 threats are annoying, even in the endgame) 20.Ne4! f5 (20...Kg7 21.h5±) 21.Nf6† Kg7
22.Nxd7 Nxd7 23.Bb5²

16.f3!
Again White must avoid 16.0-0-0? due to the 16...Nbxd5! trick.

The text move is an ideal solution though; Black remains stuck in the pin and there is not much
he can do about it.

16...Kg8
16...Qc7 17.g4! hxg4 18.h5 gives White a winning attack.

17.Nxf6† exf6 18.0-0-0 Qc7 19.Bd3 f5 20.g4!±

Black faces a powerful attack, and if he swaps queens he will have a hard time in the endgame
due to his poor pawn structure.

D) 4...Nc6!?

For a long time I dismissed this as simply a dubious move, as Black will soon be forced to
adopt the ugly ...fxe6 structure. Practical results have been great for White and the engine
suggests multiple ways to get an advantage, and I even won a game in this variation without
knowing any theory. And yet, when I tried to find the ‘simplest’ of these approaches to
recommend for this book, I was surprised to discover that matters are not easy at all. In fact, if I
were still playing, I would consider trying this with Black – it is quite forcing, and White has to
walk a narrow path to obtain an advantage. Even when White plays accurately, I am not entirely
sure of the assessment of some lines, and I may be underestimating some of Black’s resources.

29
Some of the lines where White has achieved good results actually lead to rather messy play, and
White may easily find himself caught off guard.

5.c4 Nb6 6.e6

Black was threatening ...dxe5, and White is practically forced to play this thematic pawn sac if
he wants an advantage.

6.Nc3 seemed tempting to me at first, hoping to avoid the mess of the main lines, but after
6...Bg4!? (6...dxe5 7.d5 Nb4 8.a3 Na6 9.Nxe5 e6 10.Qf3 Qf6 11.Bf4 may be better for White if
you trust the engines, although I am not completely convinced...) 7.e6 (neither 7.exd6 Bxf3 nor
7.Be3 dxe5 are anything special for White) 7...fxe6 I failed to find any good way to exploit the
fact that we will be able to win the light-squared bishop. It seems natural that somehow it should
favour White (as it indeed does after 4...Bg4 5.Be2 Nc6?!, as shown in Chapter 3), but here
Black is quite fast with his own development, and has counterattacking ideas involving ...e5 or
...d5.

I briefly considered 6.Be3 as an attempt to improve on the line above, but Black does fine after:
6...dxe5! (White would be happy after 6...Bg4?! 7.e6!N 7...fxe6 8.Nbd2! intending to recapture
on f3 with the knight, with Ng5 to follow, when Black would really miss the light-squared
bishop) 7.d5 Nb4 8.a3 Na6 9.Nxe5 e6 Black is perfectly solid. In the analogous position after
6.Nc3, White was only able to strive for an advantage with the help of some concrete ideas; here
it seems he is not fast enough, having already spent a tempo on Be3.

6...fxe6 7.Nc3
One of the confusing aspects of this variation is that White has two other options, with all three
lines having the potential to transpose to each other, but each having its own pros and cons. I will
briefly summarize my reasons for avoiding the alternatives, as the drawbacks are not obvious if
you only look at the current theory:

7.h4 g6 (7...e5 8.d5 Nd4 9.Nxd4 exd4 10.Qxd4 e5 11.Qe4 gives White a reasonably nice
version of this structure, for instance: 11...Qf6 12.Bd3 Be7 13.Nc3 Rf8 14.Qe2 Bf5 15.Be3²)
8.h5!? (8.Be3 is the 7.Be3 line below; 8.Nc3 seems best, transposing to our main line) 8...Bg7
9.hxg6 hxg6 10.Rxh8† Bxh8

30
 At first it looks tempting to play this way and target the g6-pawn, but Black’s king will be safe
enough after ...Kd7 and ...Qf8, and White should not forget that he is behind in development.

The other main idea is:

7.Be3 g6 8.h4 Bg7 9.h5 e5 10.d5 e4!
10...Nd4? enables White to profit from having delayed Nc3: after 11.Nxd4 exd4 12.Bxd4
Black has a bad position, as he does not have time to exchange everything on d4 and block
with ...e5.

11.Ng5 Ne5 12.Nxe4

I considered three other ideas:
a) 12.hxg6 hxg6 13.Rxh8† Bxh8 14.Nc3 Bf5 15.Be2 Qd7 16.Qb3 Nd3† 17.Bxd3 exd3ƒ
b) 12.Nc3 should be met by 12...h6! (but not 12...Bf5 13.h6! Bf6 14.Ngxe4 0-0 15.Nxf6†
exf6 16.b3±) 13.Ngxe4 g5 14.Be2 0-0 15.b3 e6 when Black is doing fine. He has clear and

31
 aesthetically nice ways to continue developing, such as ...Bf5-h7, ...Qd7 and ...Rae8.
Meanwhile, White needs to figure out if he has some way to make use of his temporary lead
in development and the slight weaknesses of Black’s kingside. I couldn’t find much.
c) 12.h6!? Bf6 13.Nxe4 0-0 14.Nxf6† exf6 seemed tempting to me. Perhaps with precise
play White can claim an edge, but White has to catch up with development and must also be
ready for ...f5-f4. Also, the knight on e5 is well placed to shield Black’s dark-square
weaknesses for the foreseeable future.
12...0-0 13.hxg6
Now an important resource is:

13...Bf5!N
The position is getting messy. White may still be objectively better, but suddenly Black is
extremely well developed and can claim some initiative! Even if I was sure White could claim
some objective advantage, it is not the kind of position I would want to reach over the board.

7...g6
This is Black’s best and most popular continuation.

7...e5 makes White’s task easier. After 8.d5 Nd4 9.Nxd4 exd4 10.Qxd4 e5 White usually takes
en passant, which is enough to claim an edge, but I prefer:

32
11.Qe4!? just stopping ...Bf5 for the moment. 11...g6 (11...Qf6N 12.Bd3 g6 13.Qe2± is
excellent for White) Now in Lagarde – Jouneau, Sautron 2009, the accurate 12.Qc2!N 12...Bf5
13.Bd3± would have given White a great version of a King’s Indian, due to his lead in
development, control over the e4-square and attacking chances with h4-h5, not to mention the
misplaced knight on b6.

8.h4 Bg7
8...e5 9.d5 Nd4 10.Nxd4 exd4 11.Qxd4 Rg8 12.Bg5± was excellent for White in Kerigan –
Damia, Wachtebeke 2017.

9.Be3
We should avoid the hasty 9.h5?! e5 10.d5 Nd4 11.hxg6 hxg6 12.Rxh8† Bxh8 13.Bd3 c6!
14.Bxg6† Kd7 when Black is doing well.

33
9...e5
9...0-0?! is too committal: 10.h5 e5 11.d5 Nd4

12.hxg6! (I checked 12.Ng5N but found it less clear after 12...Bf5) 12...Nxf3† (12...hxg6
13.Nxd4 exd4 14.Bxd4 Bxd4 15.Qxd4 e5 16.Qe3±) 13.gxf3 hxg6 14.Bd3 Bf5 15.Bxf5 Rxf5
16.Qd3± Fuatai – Dive, Auckland 2011. Black’s bishop is dead and White has an easy attacking
plan involving 0-0-0 and Rdg1, perhaps with Ne4-g5-e6 thrown in at some point.

10.d5 Nd4 11.Nxd4 exd4 12.Bxd4 Bxd4

This is by far the most common line.

I found an engine game in which 12...e5 13.Be3 h5 was played, with the two 3000+ players
going on to make an eventual draw. From a human perspective, it makes little sense for Black to
go for this pawn structure while avoiding the exchange of dark-squared bishops.

34
 The bishop on g7 is clearly worse than the one on e3, and Black will have to look for a way to
either exchange it somehow, or to sacrifice a pawn with ...e4 at an opportune moment, which the
engine eventually did. For our purposes, it is enough to note that 14.Bd3 Bf5 15.Ne4± gives
White firm control over the e4-square, with a pretty one-sided position.

It is also worth considering the rare alternative:

12...0-0 13.Bxg7
White can also consider: 13.h5 Bxd4 14.Qxd4 e5 15.dxe6 (15.Qe3 transposes to the 15.h5
line in the notes to variation D3 on page 25, which I prefer to avoid) 15...Qf6 16.Qxf6 Rxf6
17.e7 Re6† 18.Kd2 Rxe7 19.hxg6 hxg6 This has occurred in a few games; it is obvious that
White has the easier side of the endgame, and the only question is how he can maximize his
winning chances. 20.Rh4!?N looks like a good try, and after something like 20...Bf5 21.Be2
Rh7 22.Rxh7 Kxh7 23.b3 Kg7 24.f4²/± Black will have to suffer for a draw.
13...Kxg7 14.h5 e5
14...g5?! transposes to a high-level game: 15.h6† Kg8 16.Qd2 e5 Now in Ponomariov –
Ivanchuk, Moscow 2002, 17.c5!N would have been the most accurate way to create
additional weaknesses in Black’s position. For instance, 17...dxc5 (17...Nd7 18.cxd6 cxd6
19.Rh5 Kh8 20.Qxg5!± leaves Black with insufficient compensation) 18.Rh5 Bf5 19.Rxg5†
Bg6 20.Qe3±
15.hxg6 hxg6
This occurred in Gutsche – Wigbout, corr. 2003. White’s play can be improved with:

16.Qd2N 16...Rh8 17.Rxh8 Qxh8 18.c5! dxc5 19.Qe3

19.d6!? is also worth considering.
19...Bf5 20.Qxc5²
White intends to castle queenside and ideally play f2-f3 and g2-g4, with chances to create
problems all over the board.

35
13.Qxd4 e5

14.Qe3!?
The tempting 14.dxe6 can be met by 14...0-0 followed by ...Qf6. The engine is unsurprisingly
optimistic about White’s chances at this stage, but somehow the lines always seem to be
converting into drawish endgames, despite a slight edge for White. Technically-inclined players
may be happy with such a scenario, but it is definitely White who has to figure out a plan, as
Black’s next few moves are quite easy.

The position after the text move is hard to assess. I would like to say White is better just because
he controls the e4-square, but this outpost loses much of its significance when Black does not
have a bad dark-squared bishop locked behind the e5-pawn. Moreover, if Black could somehow
exchange his light-squared bishop for the white knight, the pawn structure would simply be great
for him! Black’s knight will find its way around; for instance, it’s not unreasonable to imagine a
long manoeuvre to d4 via c8-e7-f5. Also, the other black pieces can make use of the dark
squares: for instance, his queen and/or rook(s) may go to f4 at some point. White also has a slight
space advantage, but this alone does not count for much when two minor pieces have already
been exchanged, as Black’s remaining pieces have enough breathing room.

We have determined that White cannot hope for much, if any, advantage based on long-term
positional factors. Instead, we need to make use of dynamic factors, such as Black’s lagging
development and slightly vulnerable kingside. For instance, if Black castles short then White
should be doing well after 0-0-0 with the threat of an attack, as Black will be slow to get
counterplay on the queenside. Black can trade queens with ...Qf6-f4, but the endgame should be
more pleasant for White. Black may also consider long castling – indeed, if he could magically
arrange ...Qe7, ...Bd7 and ...0-0-0 right away, he would be doing fine, as it is hard for White to
generate a queenside attack, and castling on the kingside would put White’s king at some risk as
well. However, as we will see, White can make the ...0-0-0 plan harder to achieve by inserting a

36
timely a2-a4.
Black has three main tries: D1) 14...Qe7, D2) 14...Bf5 and D3) 14...0-0.

D1) 14...Qe7

15.Be2!
The problem with 15.h5 g5 is that, whether White plays 16.h6 or allows 16...h6, Black’s
bishop will be able to go to f5 with the benefit of a suitable retreat square after g2-g4.

15...Bd7
The trouble with 15...Bf5 is that White can pick the right moment for g2-g4, when the bishop
will have to go back to d7. 16.a4! (the immediate 16.g4 Bd7 17.Ne4 0-0-0 18.a4 Kb8 19.a5 Nc8
seems fine for Black) 16...a5 (16...Nd7 17.g4+– traps the bishop of course)

37
17.b3 Waiting for Black to commit himself somehow. Note that Black can play ...h5 any time,
but it does not really solve the problem of the bishop, as White can play f2-f3 to give himself the
option of g2-g4, while Black’s king will be less safe on the kingside. 17...0-0 (17...0-0-0? is even
worse after 18.g4 Bd7 19.b4+–) 18.0-0-0±

16.a4!
Forcing a commitment from Black on the queenside, making it hard for him to castle on that
flank.
16.h5 0-0-0! gives Black exactly what he wants. After 17.a4 Kb8 18.a5 Nc8 his position may
look passive, but he is solid and ready to start regrouping, for instance with ...Ne7-f5, or ...gxh5
followed by ...Be8.

16...a5 17.h5
I would like to point out that 17.Nb5? would be positionally the worst idea possible, and
17...Bxb5 18.axb5 0-0³ is excellent for Black. We should avoid trading our knight for the enemy
bishop at all costs!

17.b3!? is also possible, just waiting for Black to commit himself in some way.

17...g5 18.h6
Black’s position is tricky, as his king is not yet able to castle safely on either side.

18...Rf8
18...0-0-0? invites 19.b4!±.

19.0-0
19.Ne4 Bf5! 20.Nxg5 0-0-0 21.b4 Nd7!÷ is rather double-edged.

38
19...Rf6
19...Rf4 20.b3 Rh4 21.Ne4 0-0-0 22.g3 Rxh6 23.b4‚ looks dangerous for Black.

20.Ne4 Rxh6 21.Rac1!?

It is not essential to sacrifice a second pawn, particularly since c4-c5 is not even an immediate
threat, Nevertheless, I don’t think we should worry too much about the a-pawn here. Sooner or
later, White should be able to open the position with c4-c5 and expose Black’s weaknesses.

21.b3 Bf5 gives Black chances to get coordinated with ...Nd7.

21...Bxa4
21...Qf7 22.g3© maintains White’s initiative.

I also considered 21...Nxa4 22.c5 Nxc5 23.Nxc5 dxc5 24.Rxc5 Qd6 25.Rfc1 Rc8 26.Rxa5 when
Black will eventually lose his remaining extra pawn, and the king remains stuck in the centre.

22.b3 Bd7 23.Rfd1!ƒ

All the necessary preparations have been made to advance the c-pawn.

23...a4
I also considered: 23...Nc8 24.c5 dxc5 25.Nxc5 Nd6

39
26.Ne6! Bxe6 27.dxe6 Rxe6 28.Bg4 Rh6 29.Rc5ƒ

24.c5 dxc5

25.b4! c6
25...cxb4? 26.Rxc7+– and 25...c4? 26.d6+– are even worse for Black.

26.d6 Qg7 27.Nxc5±

White has excellent chances to prevail.

D2) 14...Bf5

40
 With this move, Black prepares to develop his queen to f6, since White obviously will not want
to invite an unfavourable exchange with Ne4, allowing ...Bxe4. Instead we can prepare a timely
g2-g4 to kick the bishop away.

15.Be2 Qf6!?
Black plays provocatively, with the idea that g2-g4 leaves some holes on White’s kingside.
Meanwhile, if White plays the same way as in the 14...Qe7 line, the queen will be much more
active on f6.

15...Qe7 transposes to the 15...Bf5 line in the notes to the previous variation.

15...h5 16.f3 Qf6 17.Nb5! is unpleasant for Black, as the attack on c7 forces some kind of
commitment regarding his king. 17...0-0 (17...Qe7 is well met by 18.g4!) 18.Nc3 Having
achieved the goal of provoking short castling, White simply retreats the knight and will look to
prepare 0-0-0, g2-g4 and so on.

16.g4
16.Nb5 0-0! would not be so bad for Black because he hasn’t committed to ...h5, so White’s
attack with g2-g4 will not carry as much force.
I also considered 16.a4 a5 17.b3 h5 18.f3, but found that 18...0-0-0! 19.0-0-0 Kb8= is solid
enough for Black.

16...Bd7 17.Ne4 Qe7 18.Qg5!

After 18.h5 0-0-0 19.a4 Kb8 20.a5 Nc8 I don’t really trust the computer assessment too much.

41
 White may well be slightly better but he needs to make use of it fast, because after a
regrouping plan such as ...gxh5 and ...Be8-g6 Black would be doing quite well. The engines also
seem to assign too much importance to space, and are obviously oblivious to factors such as how
reliable Black’s position would feel in a practical game – and here Black’s position seems to be
functioning just fine.

After the text move, Black will sooner or later be forced to take on g5, after which he will
struggle to find good squares for his minor pieces.

18...c6
18...Qxg5 19.hxg5 Ke7 20.Kd2² is obviously pleasant for White.

19.Rd1
The rook may appear passive here but there is not much Black can do about it, and White will
soon improve his other pieces.

19...cxd5 20.cxd5 Qxg5

20...Rc8 would be risky due to 21.f4!?.

21.hxg5²
White has lasting pressure in the endgame.

D3) 14...0-0

42
 This appears provocative but it may be objectively the best move.

15.0-0-0!?N
I like this flexible approach, which hints at White’s aggressive intentions on the kingside,
without being too committal.

The aggressive 15.h5 g5 16.h6 was seen in Kovar – Leutwyler, Pardubice 2009. White certainly
has some chances on the kingside, but I am not so sure about the evaluation after 16...Kh8!?N
17.0-0-0 Bf5÷ when Black has managed to develop his bishop to its best diagonal, and White
can no longer drive it away with g2-g4.

15...Qf6
15...Bd7 16.Bd3± looks excellent for White.

16.f3 Qf4

43
17.Qxf4 Rxf4
17...exf4?! allows 18.Rd4! Bd7 19.Bd3 Rae8 20.Ne2± when the f4-pawn falls.

18.b3²
White is certainly a bit better, although it is nothing too dramatic, so Black has reasonable
chances to hold an eventual draw.

Conclusion

We have looked at a number of sidelines after 1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3. Starting with
4...Bf5!?, the simple 5.Bd3 gives White a pleasant advantage after the bishops are exchanged.

4...c6 is a reasonable move but it generally just transposes to later variations; if Black tries to
follow an independent path, he will find himself in a cramped position with little counterplay.

4...Nb6!? is a more interesting move than I first imagined. 5.a4 a5 6.Nc3 is our choice, when I
found some interesting new ideas, especially after 6...g6 7.exd6 cxd6, when 8.d5!N gives White
excellent prospects, providing White handles the position correctly.

Finally, 4...Nc6!? is one of those variations with a risky reputation which is hard to refute. After
5.c4 Nb6 6.e6 fxe6, I suggest following the well-trodden path of 7.Nc3 g6 8.h4 Bg7 9.Be3 e5
10.d5 Nd4 11.Nxd4 exd4 12.Bxd4 Bxd4 13.Qxd4 e5, before trying the relatively unexplored
option of 14.Qe3, which keeps the position complex and presents Black with some difficult
decisions regarding what to do with his king and how to arrange his pieces. I have pointed out
various improvements and subtleties which should enable you to create a number of problems for
your opponents in this variation.

44
45
A) 6...Nc6 28
B) 6...Bg7 7.Ng5 e6 8.Qf3 30
B1) 8...0-0 9.Qh3 h6 10.Nf3 dxe5 11.dxe5 30
B11) 11...c5 31
B12) 11...Nc6 32
B2) 8...Qe7 9.Ne4 32
B21) 9...h6 33
B22) 9...dxe5 10.Bg5 Qb4† 11.c3 Qa5 12.dxe5!N N8d7 13.Nbd2! Nxe5 14.Qg3
f5 15.0-0-0! 0-0 16.f4 35
B221) 16...Ned7 38
B222) 16...fxe4! 39

46
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 g6
Developing the bishop to g7 is a logical way to put pressure on White’s centre, but it is also a
rather provocative method of development which gives White some aggressive options.

5.Bc4 Nb6
5...dxe5 6.Nxe5 transposes to variation B of Chapter 4 on page 69.

5...Be6? makes no sense in view of 6.Ng5, and after something like 6...dxe5 7.Nxe6 fxe6 8.dxe5
Bg7 9.f4± it is obvious that Black’s opening has been a failure.

5...c6 is a fairly popular move but 6.0-0 is similar to the 4...dxe5 5.Nxe5 g6 line which will be
examined later, and we can angle for a transposition to it. For instance: 6...Bg7 7.Re1 0-0
(7...dxe5 8.Nxe5 is an exact conversion to variation B2 of Chapter 4 on page 72) 8.c3

Other moves are possible, but we will play in the same way as we do in the 4...dxe5 line for the

47
sake of simplicity. I see nothing better for Black than transposing to it with 8...dxe5 9.Nxe5; see
variation B21 of Chapter 4 on page 72. In the event of 8...Bg4?! 9.h3 Bxf3 10.Qxf3 dxe5
11.dxe5± White has an excellent game.

6.Bb3
We will analyse the sideline A) 6...Nc6, followed by the main line of B) 6...Bg7. Other moves
obviously exist, but they basically just lead to worse positions without putting up much of a
challenge.

6...d5 gives up the tension too easily, leaving Black without any effective way of challenging
White’s centre anymore. 7.0-0 I don’t feel the need to include h2-h3. 7...Bg7 8.c3

8...0-0 (8...Bg4 9.Nbd2 0-0 10.h3 Bxf3 11.Nxf3±; 8...Bf5 9.Nh4±) 9.Re1² Without an obvious
square for the light-squared bishop, Black will struggle to finish development. Meanwhile, White
has a few plans, such as manoeuvring the b1-knight to g3, or trying to exchange the dark-squared
bishops.

A) 6...Nc6 7.a4!

The idea is to provoke ...a5, so that Black will not be able to meet a subsequent d4-d5 with
...Na5.

7...a5
7...dxe5?! is riskier after: 8.a5 Nd7 9.Ng5 (9.Bxf7† Kxf7 10.Ng5† Kg8 11.Ne6 Qe8 12.d5
also looks good for White) 9...e6 10.d5 Nd4 11.dxe6 fxe6 This was Kudrin – Binham, Graz
1981, and now the simple 12.Nxe6N 12...Nxe6 13.Bxe6± would have been excellent for White.

48
8.exd6 exd6
8...cxd6 9.d5 Ne5 10.Nxe5 dxe5 was obviously better for White in Klovans – Neckar,
Schwarzach 1998. A natural continuation is:

11.0-0N 11...Bg7 12.Be3 Nd7 (12...Bf5 13.c4 Nd7 14.Nc3 h5 15.Bc2±) 13.Nc3 0-0 14.d6 Nf6
15.Nb5±

9.0-0 d5 10.Re1† Be6

11.Nc3!N
11.Ng5 Qd7 12.Nc3 Bg7 13.Nb5 0-0 14.Bf4 Rac8 15.Nxe6 fxe6 16.Bg3² was decent for
White in Oparin – Pesotsky, Pardubice 2012, but the text move is stronger.

11...Bg7 12.Nb5 0-0 13.Bf4 Rc8 14.c3±

White has a clear superiority in piece activity and coordination. There is no need to memorize

49
anything further, but I have included a few illustrative lines showing how White may convert his
initiative into something tangible.

14...Ne7 15.Ng5 Bf5 16.h3 Qd7

16...h6 17.Nf3 Qd7?! (17...Be6 18.Bc2±) 18.g4! Be4 19.Ne5 Qd8 20.Bg3 g5 21.Nd3+–

17.Qe2 Rfe8

18.g4! h6 19.Nxf7 Kxf7 20.Qf3!

This is the most precise, although 20.Bxc7!? Rxc7 21.Nxc7 Qxc7 22.gxf5 Qc6 23.fxg6†
Qxg6† 24.Kh2+– is also good enough.

20...Rf8 21.gxf5 Nxf5

Or 21...Kg8 22.Re6 Nxf5 23.Nxc7 Rxc7 24.Rxb6 and White wins.

22.Bxc7! Rxc7 23.Nxc7 Qxc7 24.Bxd5† Nxd5 25.Qxd5† Kf6 26.Re6† Kg5 27.Qg2†+–
Black can resign.

B) 6...Bg7

50
7.Ng5
Quieter continuations such as 7.0-0 or 7.Qe2 are quite reasonable, but I favour the most
aggressive choice.

White can also insert 7.a4 a5 before proceeding with 8.Ng5 and so on. It is not obvious to me
how White can benefit from the insertion of those moves on the queenside, but interested readers
may wish to investigate this version and compare it to the variations that follow.

7...e6
7...d5 is played extremely rarely nowadays, for good reason. Black gives up the tension too
easily and allows White easy play with 8.f4. Neither the ...c5 nor the ...f6 break will cause much
of an issue after that, and White is free to pursue any plan he chooses.

8.Qf3
8.f4 is also possible but I prefer the sharper and more ambitious approach. You can already
notice that White has essentially given up on trying to maintain his centre, and is absolutely
focused on tactical threats, beginning with mate in one of course, but followed up by moves like
Ne4 and Bg5. To play so aggressively, White needs to know precisely what he is doing, but I
believe it is worth the effort to create difficult problems for the opponent.
Black has two main options: B1) 8...0-0 and B2) 8...Qe7.

8...Qd7 has occasionally been tried by strong players but it invites trouble on the dark squares:
9.Ne4 dxe5 10.Nf6† Bxf6 11.Qxf6 Rg8 12.dxe5

51
12...Qd4 13.0-0 N8d7 14.Qf3! Nxe5 15.Qg3 White had more than enough compensation for the
pawn in Kaehmann – Reinemer, Germany 2010.

B1) 8...0-0

There seems to be a clear consensus that this move should turn out well for White, but we
should still analyse it of course.
9.Qh3
White continues to play energetically, provoking a weakening of Black’s kingside before
dropping the knight back.

9...h6 10.Nf3 dxe5 11.dxe5

Now B11) 11...c5 is worth considering, but B12) 11...Nc6 is the most natural and popular

52
continuation.

B11) 11...c5

12.Bxh6!?
The piece sacrifice is too tempting to ignore.

With that being said, the calm 12.c3 is also fine. Black is forced to play the ugly move 12...h5,
after which 13.0-0 Nc6 14.Re1 Qc7 15.Qg3² has occurred in a few games. White has a pleasant
edge, with moves like Bg5 and Nbd2 on the agenda.

12...c4 13.Be3!
The immediate 13.Ng5? is rebuffed by 13...Bxe5, but the text move prepares Ng5 with more
force.

13...cxb3N
Black should take the money and hope for the best.

The panicky 13...f6? met with severe punishment in a recent game: 14.Bxb6 Qxb6 15.Bxc4
Qxb2 16.Bxe6† Bxe6 17.Qxe6† Kh7

53
18.0-0 Qxa1 19.Qb3! Nc6 20.Nc3 Qxf1† 21.Kxf1 Despite having two rooks for the queen,
Black’s weak king makes all the difference, and the game was soon over: 21...Nxe5 22.Nxe5
fxe5 23.Ne4 b6 24.Ng5† Kh6 25.Ne6 Rf5 26.g4 1–0 V. Moiseenko – Peng Li Min, Heraklion
2019.

14.Ng5 Re8 15.Nc3

White is ready to bring reinforcements into the attack.

15...N8d7 16.Qh7† Kf8 17.Nce4 Nxe5 18.Bc5† Re7 19.axb3

19...Nd5 20.0-0-0±
20.0-0 is also promising; either way, White’s attack and general initiative offers more than
enough compensation for the sacrificed piece, and Black faces a tough struggle to survive.

54
 B12) 11...Nc6

This way Black develops more quickly and targets the e5-pawn, but he still has a fundamental
problem of being weak on the dark squares.

12.Bxh6 Nxe5 13.Ng5 Nd3†!?

This tactical shot was tried in a GM game but White’s accurate play refuted it, as we will see.

13...Qf6 14.Bxg7 Qxg7 15.Nc3± was not much of a solution for Black in Bellia – Braga,
Chianciano 1988.

13...Qd4!?N is worth considering, but 14.Bxg7 Kxg7 15.Nc3 Rh8 16.Qg3 turns out well for
White, for instance: 16...Nec4 17.0-0 Qh4 (17...Qe5 18.Bxc4 Nxc4 19.Nxf7! Kxf7 20.b3±)
18.a4! With ideas of a4-a5, Ra4 and so on.

14.cxd3 Bxb2 15.Nd2 Bxa1 16.0-0 Bd4

In the event of 16...Qd4 17.Bxf8 Kxf8 18.Qf3! Black collapses on the light squares.

17.Nde4
White has a huge initiative for the sacrificed material.

17...Re8

55
18.Nxf7!
The computer rates 18.Nh7!N as even stronger, but Nisipeanu’s move also works fine.

18...Qd7?
18...Kxf7 was better, but White should still be winning after 19.Qf3† Bf6 (or 19...Kg8 20.Qg4
Kf7 21.Re1+–) 20.Re1+– followed by g2-g4 or Ng5†.

19.Be3 Kxf7 20.Qh7† Bg7 21.Ng5† Kf6 22.Ne4† Kf7 23.Ng5† Kf6 24.Nf3+–
Black soon had to give up in Nisipeanu – Reinderman, Rijeka 2010.

B2) 8...Qe7

This is the most popular and challenging move.

9.Ne4
White continues to attack on the dark squares. Black may respond with B21) 9...h6 or B22)
9...dxe5.

B21) 9...h6

This somewhat unusual move recently began to attract some attention, as it became clear that
landing a knight on d6 (as occurs in a few moves) is not such an amazing achievement for White.
Still, I believe White can claim some advantage with accurate play.

10.exd6

56
 10.Bf4?! Nc6! 11.exd6 Nxd4 12.dxe7 Nxf3† 13.gxf3 Bxb2³ turned out nicely for Black in
Pourramezanali – Firouzja, Bushehr 2019.

I also investigated a tempting new idea:

10.Be3!?N
White sacrifices the e5-pawn for some initiative, but Black is okay after some precise
moves.
10...Nc6!
10...dxe5 11.dxe5 N8d7 12.Nbc3 Bxe5 13.g4!© gives White a strong initiative after the
further 0-0-0, Rhe1 and so on.
11.Nbc3 dxe5 12.dxe5 Bxe5
White’s initiative seems to fizzle away, for instance:

13.Bc5 Qh4
Black intends ...Qf4, so we can’t just continue with 0-0-0.
14.g3 Qh5 15.Nf6†
After 15.g4 Qh4 16.0-0-0 Bd7 17.Be3 Qe7 White can repeat moves with 18.Bc5, and it’s
not clear if he has anything better.
15...Bxf6 16.Qxf6 Rh7 17.Ba3 Qf5
White has enough compensation for the sacrificed pawn, but he cannot claim an advantage.

10...cxd6

57
11.Bf4 d5
11...e5? 12.dxe5 dxe5 13.Be3 leaves Black unable to complete development. 13...f5? occurred
in Erdal – Kirilmaz, Antalya 2019, when White overlooked a sledgehammer blow:

14.Qg3!N+– Hitting g6 while renewing the threat of Bc5.

12.Nd6† Kf8
White seems to have achieved a minor victory in the opening battle, forcing Black’s king to
take a sideways step, followed by trading off one of the enemy bishops. On the other hand,
Black’s king remains safe and it could be argued that Black’s light-squared bishop isn’t such a
prize catch. I still like White’s position, but we should have some clear ideas about what to do
next.

13.Nxc8 Nxc8 14.c3 Nc6

58
 14...Bf6 was played in one game and it should convert to the same kind of position, but the text
move is more flexible.

15.Nd2!N
It is important to delay castling, so that Black cannot simply storm the kingside with his pawns.
15.0-0 is met surprisingly effectively by 15...g5! 16.Be3 f5! when White is not well placed to
exploit the weakened kingside squares, for instance: 17.Nd2 Nd6 18.Rae1 Bf6 19.Qd1 Na5÷
Roesch – Pavlov, corr. 2018 (Black’s last move avoids 19...f4? 20.Bxf4! gxf4 21.Bxd5 when
White gets too many pawns).

15...Bf6
Planning ...Kg7 and ...Bg5.

15...g5 obviously makes less sense when White has yet to castle, and after 16.Be3 White can
play 0-0-0 and h2-h4, with good prospects on the kingside.
After 15...Qf6 16.Qe3 Black cannot force a queen exchange, so this doesn’t change much.

15...Nd6 can be safely met by 16.0-0, and if 16...g5 17.Bxd6 Qxd6 18.Qh5² White can prepare
f2-f4.

16.Qe2!
Preparing to meet ...Bg5 with Bg3, so Black will not be able to trade bishops.

16.0-0 Bg5 is fine for Black, as he has nice squares for his knights and White’s bishop is nothing
special.

59
16...Kg7
As mentioned above, 16...Bg5 17.Bg3 achieves nothing for Black.

16...e5 17.dxe5 Bxe5 18.Bxe5 Qxe5 19.Nf3² gives White pleasant play against the IQP.

17.0-0
White’s bishop pair may become rather potent. For instance, after transferring the b3-bishop to
c2, White can think about creating some attacking ideas on the kingside. Sometimes he may
consider trading the dark-squared bishop for the knight, as the opposite-coloured bishops could
help with attacking the kingside on the light squares.

17...Bg5
I also considered 17...Nd6 18.h3 Na5 19.Bc2 Nac4 20.Nf3ƒ when White avoids exchanging
knights, since the black knights aren’t doing much, and White’s knight may be useful on the
kingside.

18.Bg3 Nd6
I checked a few other ideas:

18...Bh4 is well met by 19.Rae1! intending 19...Bxg3 20.fxg3! Qd7 21.Nf3± with a strong
initiative.

18...h5 19.Nf3! h4 20.Be5† Nxe5 (20...Bf6 21.Bf4²) 21.dxe5 and the opposite-coloured bishops
should be beneficial for the attacking side, especially after the bishop comes to d3.

19.Bc2!?

60
 There is nothing wrong with 19.h3 Nf5 20.Bh2 followed by Bc2 and Nf3, but I want to show
that White does not necessarily need the bishop pair to claim some edge.

19...Nf5 20.Rae1!?
20.Nf3 Nxg3 21.fxg3ƒ is similar.

20...Nxg3 21.fxg3²
Black will have to play well to neutralize the pressure on the kingside.

B22) 9...dxe5

Eating the central pawn is the most challenging approach, as well as the sharpest and riskiest.

10.Bg5

61
 10.dxe5 is another challenging continuation, where White avoids sacrificing anything and aims
for a long-term edge. However, after investigating it I was forced to conclude that 10...h6! gives
Black fair chances.

10...Qb4† 11.c3
11.Nbd2 looks tempting but I found that it leads to nothing.

11...Qa5
I have known of this variation for many years and it always struck me as deeply suspicious for
Black, but I struggled to prove anything against it. The difficulty is that White has many options
that look tempting, but soon simplify to equality after a few accurate defensive moves by Black.

More recently, when working on this book, I found a new and dangerous approach to the
position, with the help of some ridiculously fast engines in the cloud.

12.dxe5!N
12.Nf6† Kf8 leads to insane complications but no advantage.

12.Bf6 has been played in a lot of games but after 12...Bxf6 13.Qxf6 (13.Nxf6† achieves
surprisingly little, because White does not have enough pieces participating in the attack) 13...0-
0 14.Qxe5 Qxe5 15.dxe5 the resulting endgame is equal – or perhaps just marginally better for
White, if you squint at it from a certain angle...

12...N8d7
This is the only move which puts pressure on White to prove something. If Black did not have
this option, then 12.dxe5 is automatically justified, and it would have already become the main
line for sure. Here are two examples of less challenging approaches from Black:

62
12...0-0 13.Be7 Re8 14.Bf6 N8d7 15.Nbd2 prepares long castling, with obvious potential for a
kingside initiative later. The e5-pawn is lost only temporarily after: 15...Nxe5 16.Qg3 Nc6
17.Bxg7 Kxg7 18.0-0-0 White will win back the c7-pawn whenever he wants, and he has
prospects both on the kingside, and also for maintaining pressure in the centre with ideas like
Nd6 etc. Meanwhile, Black will struggle with completing his development.
12...Qxe5 13.Bf4 Qa5 14.Bxc7
Like in the note above, White wins back the pawn and keeps an edge in piece activity,
leaving Black with a challenge to finish development.
14...0-0 15.Nbd2
15.Nf6† Bxf6 16.Qxf6 Nd5 may be slightly favourable for White, but it is better to avoid
having to exchange queens or give up the bishop just yet.

15...Bd7
15...Qf5 gets nowhere after 16.Qe2, when 16...Bd7 17.a4 Nd5?? 18.g4+– is a simple line
showing why Black’s queen was not so well placed on f5.
I also checked 15...N8d7 16.0-0 when Black struggles to continue development.
16.Bd6 Re8 17.Nf6† Bxf6 18.Qxf6²

13.Nbd2!
Simply abandoning the central pawn and preparing 0-0-0 with a great deal of activity.

13...Nxe5
13...h6 14.Bf6 0-0 15.0-0-0 Nxe5 16.Qg3 looks promising for White, for instance: 16...Ned7
17.Bxg7 Kxg7 18.Nf3 White will easily restore material equality by picking up the c7-pawn
some time, while maintaining strong hopes for an initiative.

Black can eat the pawn a different way with:

63
13...0-0 14.0-0 Qxe5
After 14...Nxe5 15.Nf6† Bxf6 16.Qxf6 the dark-squared bishop promises White a strong
initiative unless Black can exchange queens fast. He can do so with 16...Ng4 17.Qf4 Qf5,
but after 18.f3 Qxf4 19.Bxf4 Nf6 20.Bxc7 the bishop pair should ensure White of a
pleasant edge.

15.Bf4 Qf5 16.g4

16.Bxc7 Qxf3 17.Nxf3 may also offer an edge for White but, without the bishop-pair
advantage, I am less certain about it.
16...Qb5 17.Bxc7
Black continues to struggle with his development, because moving the d7-knight will give
White annoying options like Nf6†.

14.Qg3 f5
14...0-0 is well met by: 15.Be7 Re8 16.Bf6! (16.Nf6† Bxf6 17.Bxf6 Ned7 18.Bd4 offers

64
excellent compensation for White, but the text move seems even stronger) 16...Nbd7 17.Bxg7
Kxg7 18.0-0-0 White is ready to advance his h- and/or f-pawns, and his initiative is just starting;
meanwhile, Black has no easy plan for finishing development.

15.0-0-0!
The key idea.

15...0-0
15...fxe4? leads to some pretty mating nets after: 16.Nc4! Nexc4 (16...Nbxc4 17.Rd8† Kf7
18.Qf4†+–) 17.Rd8† Kf7 18.Qxc7†+–

16.f4!
16.Be7 fxe4 17.Bxf8 Bxf8 is obviously okay for Black.

We have reached a critical position, where B221) 16...Ned7 and B222) 16...fxe4! are the two
main options.

White has an easier time after:

16...Nc6?!
The knight is worse here than on d7.
17.Nf2 Nd5 18.Kb1
White is ready to launch his kingside pawns and it is not clear what Black should do next.
Here are a few brief illustrative lines:
18...Qb6
18...h6 19.Bh4±
18...b5 19.h4 h5 (19...b4? 20.h5 bxc3 21.Nc4+–) 20.Nd3©
19.Nc4 Qb5 20.h4 h6

65
21.Bxh6! Bxh6 22.Nd3!
Simply defending f4 and leaving Black without a good defence on the kingside.
22...Bg7
22...Bd7 23.h5!+–
23.h5!+–
The attack will decide.

B221) 16...Ned7

17.Nf2
17.h4!? fxe4 18.Nxe4 seems like fun but we have solid compensation anyway, so why bother?

17...Nf6

66
 17...Nc5 18.h4 Nxb3† 19.axb3 leaves Black unable to make progress on the queenside, and
h4-h5 remains a huge threat. A nice continuation could be:

19...h5 20.Be7 Rf7 21.Qxg6! Rxe7 22.Nc4+– Threatening mate along the 8th rank.

In the event of 17...Nd5 18.h4 N7f6 White should have decent compensation after any normal
move, but there is a precise way to achieve a larger advantage:

19.h5! Nxh5 20.Rxh5 gxh5 21.Bh6 Rf7 22.Nd3! Having sacrificed the exchange to weaken
Black’s kingside, White can simply walk his knight to a great central outpost. 22...h4 (22...c6
23.Ne5 Rc7 24.c4 Qc5 25.Kb1 Ne3 26.Ne4!+–) 23.Qg5 c6 24.Ne5 Qc7 25.Nxf7 Qxf7
26.Bxd5 cxd5 27.Bxg7 Qxg7 28.Qd8† Qf8 29.Qxh4± Followed by Rh1 and Nf3, with an
enduring initiative that counts for a lot more than Black’s extra pawn.

67
18.Bxf6!
Avoiding 18.h4 Nh5! when the knight would be impossible to dislodge.

18...Rxf6 19.h4 Bd7 20.Nd3!

20.h5 Ba4 21.Nd3 gxh5 22.Rxh5 Rg6 is unclear.

20...Ba4
20...Be8 21.Rhe1 Nd5 22.Nc4± is great for White.

21.Ne5 Be8!
After 21...Bxb3 22.axb3 it is not clear what Black should do about the powerful threat of h4-
h5.

22.Ndf3 Na4 23.Bxa4 Qxa4 24.Kb1 Qe4† 25.Ka1©

The game goes on, but it is obvious that White has more than enough of an initiative to
compensate for the missing pawn.

B222) 16...fxe4!

68
 Despite being objectively Black’s best move, I do not regard it as too theoretically challenging,
in the sense that White is the only one who can realistically play for an advantage from here. In
previous variations we saw White getting a useful initiative on the kingside, whereas here Black
keeps his extra pawns as bait, while trying to finish development and/or exchange queens.

17.fxe5 Qxe5 18.Qh4 a5!

This seems best.

18...h6 19.Bxh6 Qh5 20.Qxh5 gxh5 21.Bxg7 Kxg7 22.Nxe4 leaves Black with some weak
pawns; and with Nc5 coming up, he will find it hard to develop his pieces without giving up
some material.

18...e3 19.Nf3 is also unpleasant for Black, for instance:

69
19...Qb5 (19...Qf5 20.Bc2! Qa5 [20...Qc5 21.Be7±] 21.Be7! Re8 22.Ng5+–) 20.Bxe3 a5
21.Rhf1 Qh5 22.Qxh5 gxh5 23.Bc5 Re8 24.a4 Black is struggling with development, and his
extra pawn doesn’t seem worth much as he’ll probably lose one of his weak pawns at least.

19.a4
19.Bc2 a4 20.a3 Qb5 21.Be7 e3! would be annoying.

A more serious alternative is:

19.Nxe4!? a4 20.Bc2 a3 21.Kb1!
21.b3 Nd5! is okay for Black.
21...axb2 22.Be7

22...Rf4!
The computer initially prefers 22...Re8, but 23.Nf6† Bxf6 24.Bxf6 Qh5 25.Qxh5 gxh5
26.Bb3 sees White maintain a safe edge.

70
 The text move runs into some fancy tactics, but Black eventually survives. He must walk a
tightrope though, and White does not risk losing. Best play continues:
23.Ng5! Rxh4 24.Rd8† Bf8 25.Rxf8† Kg7 26.Rhf1

26...Qf5!
The only way to survive!
27.Rf7†! Qxf7 28.Rxf7† Kg8 29.Bf6
White could take a perpetual, but we will focus on the winning attempt.
29...Nd5 30.Rg7† Kf8 31.Be5

31...Ne3 32.Nxh7† Rxh7 33.Rxh7 Nxc2 34.Kxc2 Ra5 35.Bxc7 Rb5 36.Kb1 Ke8
The resulting endgame seems drawish, although even here White can still press for a while
without risking losing.

19...Bd7!

71
 19...e3 20.Nf3 Qf5 21.Bxe3 Nd5 22.Bg5 requires deeper analysis to reach a definite
conclusion, but I am happy to stop here and say that White’s initiative offers at least enough
compensation for the pawn.

20.Rhe1
20.Nc4 Nxc4 21.Rxd7 b5! offers Black enough counterplay.

20...h6 21.Bxh6
I also considered 21.Be3 Nxa4 22.Nxe4 Bc6 23.Rd4 Nb6 24.Bxh6 Bxh6† 25.Qxh6 Qf4†
26.Qxf4 Rxf4 when Black seems to survive.

21...Qh5 22.Qxh5 gxh5 23.Bxg7 Kxg7

White’s position seems slightly more pleasant, but Black has caught up on development and
the a4-pawn is annoyingly weak, which shows how important it was for Black to play ...a5 at the

72
right moment.

24.Nxe4
24.Rxe4 seems okay for Black after 24...Rf2 or 24...Bc6.

24...Bxa4 25.Bxa4 Nxa4 26.b3 Nb6 27.Ng5 Rf5 28.Nxe6† Kf7

29.Nd8† Kf8 30.Nxb7 a4

Black has decent drawing chances, but White can still keep trying to make something of his
extra pawn.

Conclusion

4...g6 should be met by the active 5.Bc4 Nb6 6.Bb3, taking aim at f7. If Black goes for the
unusual 6...Nc6, then 7.a4! is a useful finesse, when 7...a5 8.exd6 gives White an excellent
position no matter how Black recaptures.

6...Bg7 is more natural, at which point I favour the aggressive 7.Ng5 e6 8.Qf3, forcing Black to
decide how to defend f7. 8...0-0 is possible, but 9.Qh3 h6 10.Nf3 dxe5 11.dxe5 gives White an
easy initiative on the kingside. 8...Qe7 is more challenging, when 9.Ne4 leads to an important
branching point. 9...h6 is a sensible move, leading to a position where I had to find some quite
sophisticated ideas to make the most of White’s bishop pair. With that being said, if you study
this section carefully you have good chances to put Black under long-term pressure. We then
looked at the more critical 9...dxe5 10.Bg5 Qb4† 11.c3 Qa5, when 12.dxe5!N is a brand new
idea, avoiding any immediate jumps to the f6-square and instead sacrificing a pawn for fast
development and attacking chances. Engine analysis demonstrates that White’s initiative is real,
and Black will do well to come out of it unscathed, especially if he is unprepared for it.

73
74
A) 5...c6 44
B) 5...e6 6.0-0 Be7 7.c4 Nb6 8.Nc3 0-0 9.h3 Bh5 10.Be3! 46
B1) 10...Nc6 48
B2) 10...N8d7 49
B3) 10...d5 11.c5 Bxf3 12.gxf3 Nc8 51
B31) 13.h4!?N 53
B32) 13.f4! 55
B321) 13...Bh4 55
B322) 13...Nc6 57

75
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 Bg4
Historically, this is the most popular move by a wide margin, although it is not so fashionable
these days.

5.Be2
Black’s two main options are A) 5...c6 and B) 5...e6.

5...dxe5? 6.Nxe5 Bxe2 7.Qxe2 e6 (7...c6 8.Qf3! Nf6 9.Qb3+– has given White a winning
position in several games) 8.Qb5† Nd7 9.Qxb7± Black had no compensation for the lost pawn
in Koster – McFay, East Lansing 1982.

5...g6? is also bad: 6.Ng5! Bxe2 (6...Bf5 7.e6 Bxe6 8.Nxe6 fxe6 9.Bg4± was also tough for
Black in Damaso – Pinheiro, Lisbon 1995) 7.Qxe2 e6 (7...dxe5? 8.Qf3!+–)

8.c4 Nb6 9.Qf3 Qd7 10.Ne4 Be7 11.exd6 cxd6 12.Nf6† Bxf6 13.Qxf6 Rg8 14.Qh4± Black
suffered from an unsafe king and weak dark squares in Mrdja – Joguet, Le Touquet 2001.

5...Nc6?!
This is an inferior version of the 4...Nc6 line examined in Chapter 1, since Black’s light-
squared bishop will soon be exchanged, which will leave him vulnerable on the light
squares.
6.e6
This thematic sacrifice is obvious and strong.
6.c4 Nb6 7.e6 fxe6 8.Nbd2!? is an interesting alternative, for instance: 8...g6 9.h3 Bxf3
10.Nxf3©
6...fxe6 7.0-0 Qd7 8.Re1 Nf6 9.h3

76
9...Bf5
9...Bxf3 10.Bxf3© leaves Black without the light-squared bishop, so the e6-pawn will
inevitably fall sooner or later, leaving Black’s position full of weaknesses.
10.c4 Nb4 11.Nh4 Nc2 12.Nxf5 Nxe1 13.Ne3 Nxg2 14.Nxg2±
Komodo 8 – Rybka 4.1, engine game 2014.

A) 5...c6

6.c4

6.0-0 Bxf3! is a thematic move which may seem surprising to the uninitiated. After 7.Bxf3
dxe5 8.dxe5 e6 there have been many games, and there is no easy way for White to prove
anything.

77
6.Ng5 is another popular move, when 6...Bf5 is best, leading to a complicated game.

6...Nc7
Not exactly fun opening play by Black, but at least the knight controls the e6-square, taking the
sting out of any Ng5 ideas.

6...Nb6 7.Ng5! Bxe2 8.Qxe2 dxe5 9.dxe5 e6 10.0-0 gives White a great version of a typical
pawn structure. A good example continued:

10...N8d7 11.Bf4 Qc7 Attacking the e5-pawn is somewhat challenging, but the tactics work out
well for White. If Black merely develops and castles, White will centralize his pieces and land a
knight on d6, with a considerable advantage. 12.Nc3 h6 13.Nge4 Nxe5 14.c5 Nbd7 15.Nd6†
Bxd6 16.cxd6 Qxd6 17.Rad1 Qc7 18.Rxd7 Qxd7 19.Bxe5± Chandler – Schroll, Germany 1998.

7.exd6!
Black’s last move was directed against the Ng5 plan, so White should convert to a different
structure in which the knight is placed awkwardly on c7.

7...Qxd6
7...exd6 8.0-0 Be7 9.Qb3! also gives White a pleasant advantage. Here is an illustrative line:
9...Qc8 10.Be3 0-0 11.Nc3 Nd7 12.Rfd1

78
12...Nf6 (12...Ne6 13.d5 Nec5 14.Qc2±) Now in Kislik – Freisler, Rakovnik 2013, 13.d5!N
would have been a good way to prevent ...d5, while again emphasizing how bad the c7-knight is.

8.0-0
Even the immediate 8.Qb3!? could be considered.

8...g6 9.Nc3 Bg7 10.Qb3 0-0

Black could defend the b-pawn somehow, but doing so will either weaken the light squares or
misplace a piece in some way, so White is clearly better anyway.

11.Ne4!N
White opted for 11.c5 Qd7 12.Qxb7 and went on to beat his much-lower-rated opponent in
Bologan – Cherepov, Tallinn 2016. However, 12...Nba6!N would have given Black plenty of
activity for the pawn, with chances to regain d4 and possibly c5 as well. The text move maintains

79
control for White.

11...Qd7
11...Qd8? allows 12.Qxb7.

12.Rd1±
Black has no real counterplay, while White enjoys excellent central control while keeping the
possibility of Qxb7 in reserve.

B) 5...e6

This used to be the undisputed main line of the Alekhine before the lines covered in the next
chapter became popular. Compared with many openings, there is not too much deep theory in
this variation, but it can lead to strategically complex positions. Since many 1.e4 players tend not
to study such variations in much detail, I suppose it could be an interesting way to play with
Black, as sidelines don’t work well for White, and in the main lines White needs to have a good
understanding of the plans. Richard Rapport is one of a few strong players who have successfully
incorporated this approach in their repertoires in recent years.

6.0-0 Be7 7.c4 Nb6

White can try to mess around with his move order in various ways: for instance by delaying
Nc3, avoiding Be3, playing h2-h3 early or avoiding it altogether, delaying castling and so on.
However, I was not too impressed by any of those ideas, and I think there are good reasons to
stick to the classical main line, at least up until a later point where I found an interesting new
approach.

8.Nc3 0-0

80
9.h3
I found no compelling reason to avoid this move.

9.b3 would render the ...d5 plan ineffective, but after 9...Nc6! 10.Be3 (10.exd6 cxd6 11.d5 is less
effective here than in the analogous variation B1 below) 10...dxe5 Black is doing fine.

9...Bh5
9...Bf5?! 10.exd6 (10.Bf4 is also possible) 10...cxd6 11.Bf4 leaves Black with a poor choice,
for instance: 11...Nc6 (11...N8d7 12.a4! is also excellent for White)

12.d5! exd5 13.cxd5 Nb4 14.a3 Na6 15.Nd4 Bg6 16.Bxa6 bxa6 17.Nc6 Qd7 18.Qg4!±

81
10.Be3!
10.b3 Nc6! 11.exd6 cxd6 reaches a thematic structure where Black’s knight is better placed on
c6 than on d7, as occurs in variation B2 below. And since White has played the slow b2-b3
instead of Be3, the energetic d4-d5 plan does not come with the necessary bite.

10.exd6 cxd6 releases the tension without offering White a convincing follow-up, and if Black
gets time for ...Nc6 and ...d5 he will be doing fine.

After the text move we will consider the slightly unusual B1) 10...Nc6 and B2) 10...N8d7,
followed by the main line of B3) 10...d5.
A sideline worth mentioning is:
10...a5
It is not altogether clear to me what Black is up to, but the general idea seems to be to play a
useful waiting move.
11.exd6
11.b3 d5! 12.c5 N6d7 is the kind of position Black hopes to reach, where White will have
to work harder to secure the queenside.
11...cxd6
Compared with variation B2 below, Black has not committed his b8-knight to d7, so his
knights are not so awkwardly placed. On the other hand, playing ...a5 does not help him a
great deal, and the b5-square has been weakened.

82
12.Qb3!
After 12.b3 d5! Black’s knights are relatively well placed, compared to variation B2 below.
12...N8d7 13.Qb5!?
Several games have reached this position and it would be possible to analyse more deeply, but
it seems clear to me that the active queen on b5 combined with the constant threats of d4-d5 will
be extremely annoying for Black.

A similar type of idea is:

10...a6
Unlike the line above, this time Black avoids leaving a hole on b5. However, the drawback
is that Black fails to prevent a timely b2-b4, so White will be able to dominate the
queenside space.
11.b3!
11.exd6 cxd6 would justify Black’s play, as he has neither misplaced the b8-knight nor
weakened the b5-square.
11...d5
I don’t see another useful move for Black.
12.c5 N6d7 13.b4 Nc6
White has a serious space advantage, although Black has at least made sense of the ...a6
move, as his knight is stable on c6. I think White’s most logical continuation is:

83
14.a3N
By defending b4 with a pawn, White keeps the queen and rook free for other duties.
14...f6 15.exf6 Bxf6 16.Qd2²
White has the more pleasant game.

B1) 10...Nc6

This is risky due to:

11.exd6 cxd6 12.d5!

White’s pieces are ideally placed to support this active approach.

12...Bxf3

84
 12...exd5 13.Nxd5 Bxf3 14.Bxb6! axb6 15.Bxf3± is great for White.

13.Bxf3 Ne5 14.dxe6 fxe6 15.Bg4

Black’s pawn structure is rather shaky and he doesn’t have any special counterplay to
compensate for it.

15...Nxg4
15...Rf6 16.b3± leaves White in full control.

Perhaps 15...Nexc4 is best, but then 16.Bxe6† Kh8 17.Bxc4 Nxc4 18.Bd4 Bf6 19.Nd5² gave
White an obvious and lasting advantage due to his active pieces and better structure in Smeets –
Tjiam, Netherlands 2012.

16.Qxg4 Rf6 17.b3±

White can improve his position with Qe4, Rfd1 and Rac1, in one order or another. In the
following game, Black’s position soon went downhill.

17...a6?! 18.Rad1 Qc7

85
19.Qd4! Nc8 20.Bg5 Rf8 21.Bxe7 Qxe7 22.Rfe1+–
Perez Garcia – Almeida Quintana, Santa Clara 2000.

B2) 10...N8d7

This move avoids the concrete problems seen in the previous variation, but White gets a
comfortable game after:

11.exd6 cxd6 12.b3

Black’s position is rather passive, and it is hard to find good squares for his knights. At the
same time, White has many options but it is not altogether obvious what his plan should be, so it
would be easy to go wrong if you are not familiar with some ideas. One logical idea is to hunt
down Black’s light-squared bishop if given a chance; for instance, after ...Nf6 White can

86
consider g2-g4 and Nh4xg6, followed by deploying the light-squared bishop on g2, securing the
kingside and putting pressure on Black’s queenside.

12...a6!?
Again we see this move, just awaiting further development without doing anything committal.
This isn’t the most common continuation, but it’s the only one that really makes sense to me.
Black avoids ...Nf6 which would only invite the g2-g4 plan, and can perhaps in the future hope
to plan ...d5, or just continue playing useful waiting moves.

Less logical is:

12...a5
I don’t see the point in weakening the b5-square, and getting in ...a5-a4 will not help Black
to untangle his knights. White doesn’t need to do anything special and can simply improve
his pieces.
13.Rc1
13.d5 e5 14.Nb5 f5 15.Rc1 was somewhat better for White in Caruana – Indjic, Las Vegas
2015, but I see no reason to clarify the central structure and give Black a natural plan of
playing on the kingside.
After the text move, a logical continuation would be:
13...a4N 14.Re1

14...Re8
14...Nf6 allows the standard idea of 15.g4 Bg6 16.Nh4² when White eliminates the light-
squared bishop before transferring his own bishop to the optimal g2-square.
15.Nb5²
Black is running out of useful moves and White can consider b3-b4 next.

12...Nf6!?

87
 This move carries a positional trap of sorts, but White gets a fine position with the accurate
preparatory move:
13.Rc1!
The trick is revealed after 13.g4 Bg6 14.Nh4 Ne4! 15.Nxg6 Nxc3 when Black is fine, so
White starts by defending the c3-knight.
13...Nbd7
13...Bg6 14.Nh4² was pleasant for White in Praemassing – Leistenschneider, Germany
1988, as now he gets to eliminate the bishop without having to weaken his kingside in the
slightest.
13...h6 defends against the plan of g2-g4 and Nh4, but 14.Bd3! created the unpleasant
threat of g2-g4 in Weigel – M. Fischer, Bavaria 2001.
The text move looks like a logical attempt to improve the knight, but a natural and strong
reply is:

14.g4! Bg6 15.Nh4²/±

White’s position was functioning perfectly in Musakaev – Makhnyov, Tyumen 2014.

88
13.a4!
This is the most purposeful continuation, forcing Black to worry about the annoying possibility
of a4-a5.

It is easily possible for White to slowly lose the thread and fail to achieve much, as occurred after
13.Qd2 Bg6 14.Rfd1 Nf6 15.Rac1 Re8 16.Qe1 h6 17.Nd2 Nbd7 18.Bf3 Rb8 when the opening
seems to have gone well for Black, Eljanov – Bischoff, Germany 2017.

13...Rc8!?N
13...Nf6?! 14.Rc1± was excellent for White in Keller – Kirchmann, Frankfurt 2019. White
threatens g2-g4 and Nh4 next, and it isn’t clear what Black can do against it.

13...Re8 is a reasonable move but 14.a5 Nc8 15.b4² is obviously pleasant for White.

14.Rc1!
14.a5 Nd5! shows the point behind Black’s last move.

14...Re8 15.Re1
White can also go for a plan such as 15.Qd2 followed by Qa2, but I like the patient approach
of improving the rook first; what useful move does Black have?

89
15...Bf8
This looks natural although we will see that White has an interesting way of utilizing the g5-
square.

15...h6 is well met by the typical reply of 16.Bd3!, threatening g2-g4.

15...Nf8 is far from an ideal square for the knight, although this regrouping may be the best that
Black can do. White keeps a pleasant edge with 16.Qd2 Nbd7 17.Qa2² intending b3-b4.

16.d5 e5 17.a5 Na8 18.Bd3 Bg6 19.Bxg6 hxg6

20.Ng5!
This move is surprisingly annoying for Black. A possible continuation is:

90
20...Qxa5 21.Qg4 f5
21...Nf6 would of course be met by 22.Qh4! followed by Nce4, with a crushing attack.

22.Qg3ƒ
White has superb play for a mere pawn, with Ne6 one of many ideas in the air.

B3) 10...d5

As mentioned earlier, this is the clear main line.

11.c5
This space-gaining move is normal.

I am not keen on 11.b3 dxc4 12.bxc4 Nc6, as White’s pawn centre is hard to maintain.

11.cxd5!?
This somewhat unusual move deserves attention.
11...Nxd5
11...exd5 was played by Rapport, but I like White’s chances in the resulting pawn structure:
12.Ne1 Bxe2 13.Qxe2 c6 14.f4! (a high-level game continued 14.Nd3 Na6 15.Rae1 Nc7
and with the knight coming to e6, Black was okay in Navara – Rapport, Prague 2016)
14...f5 15.g4!? White had good prospects on the kingside in Perez Garcia – Peralta, Malaga
2001.
12.Nxd5
12.Qb3 could be explored further, but something about giving Black the option of ...Nxe3
does not seem right to me.
12...Qxd5

91
 12...exd5 would again give White a nice structure.
13.Rc1
Although I personally prefer the main line, White has chances for an advantage here too, so
interested readers may wish to investigate this in more detail.

11...Bxf3
Black almost always inserts this exchange, since White must either weaken his structure or
allow the knight to get to c4.

11...Nc8? has occasionally been tried by strong players, including Alekhine specialist Alexander
Baburin in one game, but it is a serious mistake due to the surprisingly blunt response 12.g4!
Bg6, and now a big improvement is:

13.h4!N 13...h5 (13...Bxh4? 14.g5+–) 14.gxh5 Bxh5 15.Ne1 White has excellent attacking
chances.
11...N6d7
This has also been played by some grandmasters, but it seems to me that Black is simply
accepting a worse position in an effort to avoid mainstream theory. Black will struggle to
find good squares for his knights and, without proper development, it is hard to believe that
breaks such as ...f6 will add much to his position.
12.b4 c6
12...f6 13.exf6 Bxf6 14.Qd2² was pleasant for White in Wittmann – Wheeler, Lucerne (ol)
1982.
13.a3 Bxf3 14.Bxf3 f6
A simple improvement is:

92
15.exf6N
15.Bg4 complicated things for no reason in Areshchenko – Indijic, Warsaw 2016.
15...Bxf6 16.Re1
White has a comfortable position and it is still hard for Black to complete development.

12.gxf3 Nc8
12...Nc4? 13.Bxc4 dxc4 14.Qa4 wins a pawn for no compensation whatsoever.

12...N6d7 seems less logical as the knight just gets in the way. A good example continued 13.f4
f5 14.b4 b6 15.Rc1 c6 and now in Soderberg – Nizynski, corr. 1999, a simple and strong
continuation would have been:

16.Kh2!?N Intending Bd3, Ne2 and Rg1, maintaining a big space advantage while gradually
building a kingside attack.

93
 I remember first seeing this position in the Encyclopaedia of Chess Openings, a long time ago.
I was really hesitant about spoiling my pawn structure back then; and even now, I feel a little
uncomfortable about having played gxf3. At the very least, I would like to have a concrete idea
of how to put pressure on the Black position. In many games I see White play f3-f4, and then
proceed to shuffle around with his pieces, while Black regroups with ...Bh4, ...Ne7, ...g6 and
...Nf5. Even though it is obvious that White’s structure is not really bad, and Black is still
passive, there doesn’t seem to be an easy way to break through on the kingside. Thus, I never
liked this line for White too much, and usually planned to deviate somewhere along the way, for
instance with 10.exd6. However, such ideas do not offer White much of an edge, and those who
buy this book expect and deserve more. Therefore I decided to revisit this old line in the hope of
injecting new life into it.
White’s main advantage is in space, which enables him to manoeuvre his pieces comfortably,
whereas Black struggles to find good squares for his pieces – with the exception of the
regrouping plan mentioned above, with ...Bh4 followed by transferring the c8-knight to f5. I
cannot overstate the importance of this plan for Black: not only does it put a stop to White’s
kingside attack, but it is also difficult to suggest any other good squares for those two pieces.
This realization gave me the idea of sacrificing the h-pawn by placing it on h4! Being a pawn
down should not matter too much for White, as he can utilize the additional open file for
attacking purposes. And of course, with the h-file open, Black’s bishop will not be safe on h4, so
he will have to rethink his entire set-up. Black can also ignore the pawn, but in that case his
bishop will still not go to h4.

My first try was B31) 13.h4!?N and it is worth sharing my analysis of it, as it contains some
interesting points. We will then move on to my main recommendation of B32) 13.f4!, with h3-h4
to follow.

I also checked 13.Bd3N, but after 13...Nc6 White should play 14.f4, transposing to variation

94
B322 on page 57. I briefly considered 14.h4, but quickly realized that 14...b6! would be
annoying.

B31) 13.h4!?N

This did not quite work the way I would have hoped, as Black can force the play and survive
with some concrete ideas. Still, the move is playable and would certainly have some surprise
value.

13...b6!
13...f5 would be perfectly nice for White because now the black knight can’t go to f5, and
14.f4 Nd7 15.Kg2² gives us a pleasant game.

In the event of 13...Nd7 14.Kg2 b6 15.b4 a5 16.a3 axb4 17.axb4 Rxa1 18.Qxa1 bxc5 19.bxc5
the opening of the queenside has only helped White.

13...Bxh4 14.Kg2! gives White excellent play for a pawn, for instance:

14...f6 (14...f5 15.f4ƒ; 14...b6 15.Rh1ƒ) 15.Rh1 fxe5 (15...Bg5 16.Bd3‚) 16.dxe5 Black has
nothing better than the ugly 16...g5 here, and although we could analyse more deeply, it is
obvious that White has good compensation and I believe it should be possible to prove an
advantage if we look deep enough. Alas, the text move is stronger.

95
14.b4 a5 15.a3
We must try to maintain the queenside structure. If we were forced to exchange on b6 then
Black’s pieces would come to life and there would not be much point behind White’s set-up.

15...axb4 16.axb4 Rxa1 17.Qxa1 bxc5 18.bxc5 Bxh4

Having lured White’s queen far away from the danger zone, Black grabs the pawn and is ready
to gain some breathing room with ...f6. White’s position is sound but he is not quite in time to
cause Black trouble:

19.Kg2
19.Bd3 f6 20.Qb1 (20.Ne2 fxe5 21.dxe5 Ne7÷) 20...fxe5 21.Bxh7† Kh8 seems messy but
White’s initiative is short lived, for instance:

96
22.Qg6 Qe8 23.dxe5 Qxg6† 24.Bxg6 Nc6=

19...f6 20.Rh1
20.f4 fxe5 21.fxe5 is conveniently met by 21...Ne7! and the knight will be perfect on f5.

20...fxe5 21.dxe5

21...d4!
This forces White to give up on his attack and bring the rook back to the centre.

22.Rd1 Nc6 23.Nb5

White regains the pawn but Black should have little trouble equalizing. For instance:

97
23...N8e7 24.Nxd4 Nxd4 25.Rxd4 Nd5=

B32) 13.f4!

This is the more accurate move order. Advancing the f-pawn is a generally useful move,
gaining space and threatening f4-f5 in some lines. This way, we disguise our intentions and
improve our position before unleashing the h3-h4 idea at the ideal moment.

Since h3-h4 is an integral part of our plan, we should be sure to consider B321) 13...Bh4, before
concentrating on the main line of B322) 13...Nc6.

13...f5 has been played from time to time but I don’t feel the need to analyse it in detail. In
general, if we can provoke this move we should be happy, as Black can no longer bring his
knight to the ideal f5-square. 14.b4² is one good option, when a long battle lies ahead, but White
has a massive space advantage and should be able to press for a long time.

B321) 13...Bh4

This is an important option as it rules out h3-h4. However, there is something a little strange
about moving the bishop again while the rest of Black’s pieces remain on the back rank, so it is
not surprising that we can create concrete problems before Black can reach his dream set-up.

14.Bd3!
14.f5 looks tempting but 14...Ne7! gives Black decent play.

14...g6 15.Qg4 Kh8

15...f5 takes the immediate pressure off the kingside, but robs the knight of its ideal square

98
while leaving Black in a passive position. 16.Qe2 Nc6 17.a3 Kh8 18.b4 a6 19.Kh1 N8e7 was
seen in Hendriks – Bosch, Enschede 1996. White could have solidified his advantage with:

20.Nb1!N± Rerouting the knight to f3, leaving Black in a cramped position without any active
play.

The text move seems best, but Black still has some difficulties achieving his ideal set-up,
because the knight can never go to e7, as it would need to do to get to f5. White could play any
sensible move such as 16.b4 with slightly better chances, but I really like the direct and untested
attacking approach:

16.f5!?N 16...exf5 17.Qf4

As far as I am aware, this type of ‘slow’ pawn sacrifice has not really been seen before in these
positions. The idea is that since Black’s pieces are all jammed up, he can’t really make use of it.
This means that White has time to put a rook on the g-file and bring the rest of his pieces into

99
position to exploit the open files and targets generated.

17...c6
The fact that I found nothing better than this slow move is already an encouraging sign.

Another option is:

17...Be7 18.Kh2
18.h4!? is also tempting.
18...Qd7
18...Rg8 19.Rg1 Nd7 20.Bxf5! gxf5 21.Rxg8† Qxg8 22.Rg1 Qf8 23.Qxf5+– is a nice line
showing the value of the space advantage in bringing additional reserves into the attack.
18...Nd7 19.Rg1ƒ also looks promising for White.
19.Rg1 Nc6
19...b6 20.b4© does not change much.
20.Bb5!?©
Black remains tied up and under pressure. (20.h4!? is also tempting.)

18.Kh2 Na6!?
18...Nd7 allows the blunt 19.Bxf5! gxf5 20.Rg1 when once again White’s space advantage
makes all the difference:

20...Be7?! (20...Ne7 is the only chance although 21.Qxh4± obviously vindicates White’s play)
21.Qxf5 Rg8 22.Rxg8† Qxg8 23.Qxd7+–
If White allows ...Nc7-e6, then Black would suddenly have something nice, so it’s a good thing
that the knight can be eliminated.

19.Bxa6 bxa6 20.Ne2

White can patiently build his attack. He has only sacrificed a single pawn, and Black will take

100
forever to get all his pieces into the game.

20...Be7 21.h4!?
21.Rg1 Rg8 22.Ng3± is an excellent alternative.

The text move is not essential to prove an advantage, but since the h3-h4 sacrifice is central to
my whole concept in this variation, I could not resist giving it as the main line for this variation.

21...Bxh4 22.Qh6 Rg8 23.Rg1 Be7 24.Rg3 Qf8

And now une petite combinaison...

25.Qxh7†! Kxh7 26.Rh3† Qh6 27.Bxh6 Rh8

Black has survived the attack but there is a final finesse.

101
28.Ng3! Kg8 29.Nxf5±
Material is now level. The knight is untouchable and it will land on d6 next, giving White a
dominating endgame.

B322) 13...Nc6

Developing the knight is more flexible, and it feels like the most ‘correct’ continuation. With
the bishop still on e7, White must be on the lookout for tricks involving ...Bxc5 followed by
...d4.

14.Bd3!
We continue to delay h3-h4, instead improving the bishop and threatening f4-f5.

As usual, I would have liked to play 14.h4, but 14...Bxc5! 15.dxc5 d4 enables Black to simplify
things in the centre and on the queenside and get enough activity, which defeats the purpose of
my scheme.

14.b4 is the most common continuation, but it allows Black to achieve the relatively solid set-up
with 14...Bh4! followed by one knight or the other coming to f5, which terrified me as a child!
Even though the computer typically prefers White in these positions, Black has a robust set-up
and I don’t think you should hope for much as White there, unless you can come up with a
concrete way to cause problems.

14...g6
14...Bh4 would of course be met by 15.f5! exf5 16.Bxf5± when, since we have got rid of the
ugly doubled f-pawns, there are no drawbacks in White’s position anymore. Meanwhile, for
Black, the same troubles remain: a lack of space for the minor pieces and a potentially vulnerable

102
kingside.

14...f5 is of course possible, but my assessment is the same as in the note on 13...f5: since Black
can no longer carry out his one attractive plan of ...Bh4 and ...Ne7-f5, he is destined to suffer for
a long time.

A more interesting move is:

14...b6!?
Counterattacking against the centre is a principled approach, but we can ignore it with:

15.f5!?N
15.cxb6 Nxb6 16.f5 f6! is not so clear.
15.Qa4!? has been played and it should offer White some advantage as well, but it feels
more thematic to concentrate on the kingside.
15...bxc5
After 15...exf5 White can flick in 16.cxb6 Nxb6 before 17.Bxf5± with a clear plus.
16.f6! Bxf6
16...gxf6? is easily refuted by 17.Bxh7† Kxh7 18.Qh5† Kg7 19.Kh2 and murder on the g-
file.
17.exf6 Qxf6

103
18.Ne4!?
This flashy move is not the only decent option, but it works well enough.
18...dxe4 19.Bxe4 N8e7 20.Qh5 h6 21.Qxc5±

15.h4!N
Finally, I was able to get my sacrifice in! I believe this is the optimal time for it.

15...Bxh4
Ignoring the pawn doesn’t make much sense for Black. The whole point of my concept is to
prevent the solid blockade with ...Bh4 with a knight coming to f5, and that will be accomplished
just fine if Black doesn’t take the pawn!

15...h5?! seems like a huge weakness. 16.Kg2 is one good answer, and sooner or later something
bad will happen to Black’s kingside.

104
15...f5 16.Kg2 is, as usual, simply better for White if Black doesn’t take on h4, and if he eats the
pawn in the near future, we will convert to one of the lines after 15...Bxh4.

15...Bxc5 16.dxc5 d4 does not have the desired effect here, as after 17.Ne4 dxe3 18.fxe3 the
knight will become monstrously strong on f6 or g5, and White’s attack continues unabated.

16.Kg2 Be7
The bishop is awkwardly placed on the open file, and will have to move at some point.
Nevertheless, I examined a few other options:

16...b6 17.Qa4! and Black’s knight has no good squares.

16...f5
This can be met in more or less the same way as we see in the main line below.
17.a3
White should not just ignore Black’s ideas on the queenside, as it is important to maintain
the pawn structure there. For instance: 17.Rh1 a5 18.Qf3 Rf7 19.Rag1 Be7 20.Kf1 b6
21.Rh2 Rg7÷
17...a5 18.Ne2 b6
18...Be7 gives White a choice: 19.Rh1 transposes exactly to the bolded main line, and
19.Bd2!? also makes sense, supporting b2-b4, similar to the continuation below.

19.Bd2! bxc5 20.dxc5 Be7 21.b4±

White is still a pawn down, but he has a dominant position and Black has no counterplay.

17.Rh1 f5
Black has to take some space on the kingside to avoid being demolished.

105
In an ideal world he would love to get a knight to f5, but a move like 17...Re8?! (naively hoping
for ...Bf8 followed by the desired knight manoeuvre) would allow a quick f4-f5.

18.a3!
With no immediate breakthrough available on the kingside, White should strengthen his
position on the opposite flank.

18...a5 19.Ne2
19.Qa4!? is also possible, trying to play b2-b4 as quickly as possible. White may well have
better chances here as well, but I prefer the text move as it seems more flexible.

19...b6
If Black does not play this, White will just continue with Bd2 anyway – it’s not as if Black has
any clear plan for counterplay.

106
20.Bd2!
The key point is to prepare a quick b2-b4 in order to consolidate the pawn on c5, which will
keep Black’s pieces at bay.

20.cxb6 Nxb6 gives Black enough activity.

20...Qd7
20...Nb8 21.Rc1 Qd7 22.Qc2 Na7 23.b4± and White dominates.

20...bxc5
This may seem like the most logical move, but it only presents the d4-square to the white
knight.
21.dxc5
It remains incredibly hard for Black to complete development, and White maintains
pressure on both sides of the board. Despite the pawn structure on the queenside being
temporarily spoilt, Black’s pieces still struggle to find any good squares, so it is not even
clear if the simplifications on the queenside were to his benefit.
21...Rf7
21...Bxc5? 22.Rc1+– wins material.
21...d4 22.Bc4 does not help Black either.
22.b4 Qf8
22...Bf8 23.b5 N6e7 24.Nd4+– is crushing.

107
23.Qa4!
23.b5 Nd8 24.c6 Bxa3 is not so clear.
23...N8a7 24.b5

This works well, although it isn’t even necessary as White had complete control of the
position anyway.
24...Nd8 25.b6 cxb6 26.cxb6 Nc8 27.Qb5±

21.b4 Nd8!
This difficult move is Black’s best chance.

21...axb4 22.axb4 Rxa1 23.Qxa1 Nd8 gives White complete control after:

24.Qa6! The knight on c8 is essentially trapped, and Black cannot play ...c6 or ...b5.

108
22.Qc2!
White cannot break through immediately on either flank, but he has maintained a major space
advantage while avoiding giving Black any counterplay. Next we can either improve our position
on the kingside, for instance with Ng1-f3, or take some action on the queenside by opening a
file. White’s biggest trump card is his space advantage, so it is important to maintain the
possibility of attacking on either flank.

22.Ng1 Kg7 23.Nf3 axb4! 24.axb4 Rxa1 25.Qxa1 b5 leaves the position too blocked to make
progress.

22.bxa5!? is a decent try, but 22...bxc5! 23.dxc5 Bxc5 24.Bb4 Be7 25.Rc1 is not clear. White
certainly has full compensation for the pawn, but Black is not completely stuck either, and after
something like 25...c5 26.Bxc5 Rxa5 27.Bb4 Ra8 28.Nd4 Qb7÷ Black should have enough
counterplay.

22...c6
22...a4 is met by 23.b5! when Black will not succeed in blocking the queenside, so White’s
pressure on both sides will continue.

22...axb4 23.axb4 Rxa1 24.Rxa1 b5 25.Ra8 sees White dominating on the queenside. A nice
illustrative line continues: 25...c6 26.Qa2 Nf7 27.Qa6 g5

109
28.Kf1 h5 29.Nc3 gxf4 30.Nxb5!+–

23.bxa5!
I find this move the most appealing option.

That being said, 23.a4!? is a decent alternative which also avoids any blocking of the queenside.
This should also give White an edge, with similar ideas as the main line.

23...bxc5
23...bxa5 leaves White with a pleasant position, with pressure on both flanks. Because the b-
file is open, Black’s king can never hope to escape to the queenside, and on the kingside it will
always be somewhat uncomfortable. Besides, Black’s awkward minor pieces ensure that his
rooks will have trouble connecting, so he may not be able to fight for the b-file either. White has
many possible plans, from doubling on the b-file or on the h-file, to bringing the knight to a4, or
to f3, all of which should offer White good chances.

24.dxc5 Nb7 25.Bb4 Nxa5

110
26.Nd4©
Black’s pieces remain cooped up, while White is still playing on both flanks.

Conclusion

4...Bg4 pins our knight, so 5.Be2 is an obvious reply. If Black opts for 5...c6 then 6.c4 is a good
reply, intending 6...Nb6 7.Ng5! or 6...Nc7 7.exd6!, with Black’s knight being misplaced in
either scenario.

5...e6 is the main line, when 6.0-0 Be7 7.c4 Nb6 8.Nc3 0-0 9.h3 Bh5 10.Be3! is my
recommendation. After dealing with a few sidelines, we focused on the old main line of 10...d5
11.c5 Bxf3 12.gxf3 Nc8, when White must look for a suitable way to exploit his space
advantage. I came up with a brand new concept of sacrificing a pawn on h4 to prevent Black’s
traditional blockading plan, but it is important to time it correctly to get the most out of White’s
position. 13.f4! is most accurate, and after the most popular and logical 13...Nc6 we continue
14.Bd3! (threatening f4-f5) 14...g6 and now 15.h4!N 15...Bxh4 16.Kg2 is the new idea which
poses Black difficult and novel challenges. His extra pawn will not be of any benefit for a long
time, and White can use his massive space advantage to switch between both flanks, causing
problems for Black all over the board.

111
A) 5...Nd7? 6.Nxf7! Kxf7 7.Qh5† Ke6 8.c4 N5f6 9.d5† Kd6 10.Qf7 Ne5 11.Bf4 c5 12.Nc3
a6 13.b4! 64
A1) 13...Qb6 66
A2) 13...b6 68
B) 5...g6 6.Bc4 69
B1) 6...Be6 70
B2) 6...c6 7.0-0 Bg7 8.Re1 72
B21) 8...0-0 72
B22) 8...Nd7!? 75
C) 5...c6 6.Be2 77
C1) 6...g6 77
C2) 6...Nd7 79
C3) 6...Bf5 7.0-0 Nd7 8.Nf3 e6 9.c4 N5f6 10.Nc3 81
C31) 10...Bd6 83
C32) 10...Ne4!? 11.Re1!? 89
C321) 11...Be7N 91
C322) 11...Bd6 93

112
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 dxe5
This is perhaps the most solid of all the Alekhine variants, and it has justifiably taken over as
the main line in recent times. In general, the structure can often resemble some type of ...dxe4
French, or even a Scandinavian – especially the lines with 3...Qd6 – but arguably, Black has a
less flexible version here. One crucial idea for White is to minimize piece exchanges. This
follows from basic principles: Black has less space, so exchanges will tend to make his life
easier. For instance, if Black can trade the active knight on e5 for one of his own knights, he
should get a fairly comfortable position – but fortunately for us, this is easier said than done.
Black can often succeed in exchanging his light-squared bishop for a knight, but this obviously
gives White the long-term advantage of the bishop pair, which will become all the more
important if the position opens up.

5.Nxe5
We will examine the ill-fated attempt to trade knights with A) 5...Nd7?, followed by B) 5...g6
and finally the main line of C) 5...c6.

A rare sideline is:

5...Bf5
This slightly awkward move places the bishop on an unstable square, but we need to exploit
it quickly.
6.Qf3!
If 6.Bc4 e6 Black will be able to develop his pieces however he wants to.
6...e6
6...Be6? 7.c4 Nb4 8.Qxb7 N8a6 gives White more than one winning continuation, 9.c5!+–
being the most accurate, threatening both Bb5† and Bxa6.
7.Bd3!
7.Nxf7 looks tempting but 7...Kxf7 8.g4 Qf6 9.gxf5 Qxf5 10.Qxf5† exf5 seems like an
okay endgame for Black.
7...Qf6

113
8.Nc3!N
An obvious improvement, adding to the central pressure.
8.Bxf5 Qxf5 9.Qxf5 exf5 could potentially be better for White due to his superior pawn
structure, but the text move is more ambitious.
8...Bxd3
8...Nxc3? 9.Qxb7 wins easily.
8...c6? is no good due to 9.Nxd5 cxd5 10.Bb5†.
8...Bb4 9.0-0 Bxc3 10.bxc3 is of course good for White.
9.Nxd5
9.Qxd3!? Nxc3 10.bxc3² also gives White a promising initiative.
9...exd5 10.Qxf6 gxf6 11.Nxd3²
White’s superior pawn structure gives him an enduring advantage.

A) 5...Nd7?

114
 This notoriously provocative move has been played a lot over the years, although not so much
recently. For many decades, the debate continued as to whether or not White could prove
anything with the famous knight sacrifice. But more recently, computer analysis has settled the
matter once and for all.

6.Nxf7!
Back in the days when the outcome of this sacrifice was not completely certain, I used to take
the view that it was not worth memorizing a bunch of complicated variations in an effort to
prove something against a move which hardly anyone played as Black, so I decided if I ever
faced it then I would simply reply with 6.Nf3, aiming for a similar type of position as in the
more common main lines. However, it would certainly count as a small victory for Black to
chase the knight away without spending a tempo on ...c6.

Nowadays the situation has changed: the knight sac should be more or less winning outright if
followed up correctly, so there is no question that this is the way to play.

6...Kxf7 7.Qh5† Ke6 8.c4 N5f6 9.d5† Kd6 10.Qf7 Ne5

Other moves make White’s task easier, for instance:

10...Nc5? 11.b4 and Black already has no choice but to return the extra piece.

10...Nb8 leads to a forcing sequence: 11.c5†! Kd7 12.Bb5† c6 13.dxc6† bxc6 14.0-0 Qa5
15.Rd1† Kc7 and now the most efficient winning line is:

115
16.a4!! As pointed out by John Shaw in the relevant volume of his Playing 1.e4 series. (16.Bf4†
Kb7 17.Be2 is also good enough.) White’s last move threats to trap the queen with Bd2, and
after 16...cxb5 17.Bf4† Kb7 18.axb5 Qxa1 19.c6† Kb6 20.Qb3!+– Black’s king will not survive
for much longer.

11.Bf4 c5
Black almost invariably plays this to prevent c4-c5† (which could either occur on the next
move or after a preliminary Nc3).

In the ensuing position, White’s queen is trapped but she is safe enough for the time being, as
Black’s knight is caught in a whopper of a pin. Of course, this means that White’s bishop daren’t
stray away from the h2-b8 diagonal, so we should always be on the lookout for the threat of ...g6
and ...Bh6 – and if White’s bishop retreats to g3, then ...Rf8 really would trap the queen.
Fortunately, White can create damaging threats to the enemy king before Black gets a chance to
carry out this or any other effective plan.

12.Nc3
Developing while threatening a check on b5, which Black needs to prevent.

12...a6 13.b4!
Widening the scope of the attack against the black king.

The two main defensive tries are A1) 13...Qb6 and A2) 13...b6.

13...cxb4 fails to offer much resistance: after 14.c5† Kxc5 15.Na4† Kd6 16.Nb2! b5 17.Nd3
White wins the piece back, while his initiative remains as strong as before.

116
 A1) 13...Qb6

14.Rc1!
It is important to know about this move, as it would not be trivial to find it over the board. The
main idea is to prevent ...cxb4, which could now be refuted by c4-c5† or the even stronger Nb5†.
If Black’s queen takes on b4, White simply completes his development, having pre-emptively
defended the knight. And if Black ignores the pawn on b4, White can include bxc5† at some
point, when the rook will have to support a future advance of the c4-pawn.

14...g6
Alternatives are no better, for instance:

14...g5 15.Bg3 h5 (after 15...Bf5? 16.Bd3! Black had to resign in Manca – Sulskis, Ortisei 2012)
16.bxc5† Qxc5 17.Qg6! Qd4

117
18.Qc2!! It looks weird to retreat the queen, but this redirection is exactly what is needed to
boost White’s attack. 18...h4 19.c5†! Kd7 20.Rd1+– Sakai – Rebaudo, email 2001.

14...Qxb4 15.Bd3 leaves Black unable to do much about the simple plan of castling followed by
annihilating Black’s king in the centre. 15...g5 16.Bg3 h5 seems as good a try as any, but after
17.0-0 h4 18.Bxe5† Kxe5 19.f4† gxf4 we can put an end to Black’s resistance with:

20.Ne4!+– Black soon resigned in Sakai – Zednik, email 2001. (20.Rce1† is also winning but it
is less convincing, and Black subsequently escaped with a draw in Predke – Danilenko, Batumi
2010.)

The text move was played successfully in an old game. Preparing ...Bh6 makes sense, especially
since Black will be targeting not only the bishop, but also the rook on c1. Nevertheless, Black’s
scheme can be refuted in more than one way.

118
15.Be2
We start by simply continuing development.

15...Qc7
15...Bh6?! loses more simply after: 16.Bxe5† Kxe5 17.Qxe7† Be6

18.g3! There is no escaping f2-f4 with a deadly check. 18...Rhe8 19.f4† Bxf4 20.gxf4† Kxf4
21.Qxf6†+–

16.bxc5†!?N
This seems like the simplest winning continuation.

16.Na4? should be avoided due to 16...Bh6! and after the further mistake 17.bxc5†? Kd7
18.Qe6† Ke8 19.Qxe5 Bxf4–+ Black escaped with his extra piece still intact in Prie – Nataf,
Andorra 1997.

119
16.Rd1 Bh6 17.Ne4†! Nxe4 18.Bxe5† Kxe5 19.d6!+– is Shaw’s main line, and 16.0-0 Bh6
17.Ne4†! Nxe4 18.Bxe5† Kxe5 19.d6! Qxd6 20.f4†+– also does the job. However, both of
these lines rely on some fancy tactics, whereas the text move is simpler, without being any less
effective as judged by the engine’s evaluation.

16...Qxc5 17.Bxe5†!
Taking the opportunity to lure the king further up the board, rather than allow 17.0-0 Qd4
when Black has a glimmer of a chance of escaping – although White should be comfortably
winning here as well.

17...Kxe5 18.0-0+–
Intending some combination of Rfe1, Na4, Bf3 and c4-c5(†).

A2) 13...b6

This is Black’s best defensive try. It can be refuted, but some fancy tactics are needed in some
lines.

14.Bd3!
It is important to have the option of Ne4†.

14.b5 a5 is not so clear.

14...g6
14...cxb4 comes unstuck after: 15.Ne4† Nxe4 16.Bxe4 g5 17.Bxe5† Kxe5

120
18.d6! Be6 19.f4†! gxf4 20.Qh5†+– The attack was too strong in Morozov – Lagergren, email
2013.

14...g5 15.Bg3 Bh6 16.Ne4† Nxe4 17.Bxe4 left Black in a horrible position in Dobre – Dombi,
corr. 2014. One way or another, White will avoid exchanging queens and bring his remaining
pieces into the attack. A logical continuation would be:

17...Qf8N 18.Qh5 Bg7 19.bxc5† bxc5 (19...Kxc5 20.0-0 Bd7 21.Qxg5+–) 20.0-0 Kd7 21.Rad1
The attack is too strong.

15.bxc5† bxc5
We have reached a critical position where White needs to produce some excellent moves to
bring home the full point.

121
16.Rb1!!
16.0-0 Bh6 did not give White anything special in Kuvshinov – Manzenkov, email 2003.
The text move may not look like it should warrant two exclams – but the point is that in order to
play it, White needs to have prepared a spectacular follow-up, otherwise he would simply be
losing. The key line continues:

16...Bh6 17.Rb7! Bd7

17...Bxf4 18.Qxf6†! exf6 19.Ne4# is the main point.

17...Bxb7 18.Qe6† Kc7 19.Bxe5† wins trivially.

18.Bg3 Rb8 19.Rxb8 Qxb8 20.0-0 Qf8

122
21.Re1! Nfg4 22.Qf3!
The last difficult move.

22...Qxf3 23.gxf3 Rf8 24.Ne4† Kc7 25.fxg4 Bf4 26.Be2+–

I have been following a line given by Shaw. My only slight criticism is that his final evaluation
of “±” is a bit too modest; White is a pawn up and c5 is weak, so I’m sure the endgame is
winning with accurate play.

B) 5...g6

This is an interesting variation. It is tempting to dismiss Black’s position as passive and

“simply worse”, but the plans aren’t completely trivial for White, and Black’s position remains
extremely solid. This line has been surprisingly popular in engine and correspondence games; the
main explanation seems to be that even though the engines generally assess the position as
slightly better for White, they seem to underestimate how solid Black is – and often, the engine’s
overenthusiastic suggestions lead to weakening moves such as c2-c4, and/or exchanging pieces –
and as long as Black can maintain his solidity, such occurrences should only make his life easier.
Instead, White has to adopt a more patient strategy, which essentially involves restraining pawn
breaks such as ...c5 and ...e5, while refusing to create any weaknesses (as would occur after c2-
c4, for instance) and avoiding minor-piece exchanges, unless Black wants to trade his light-
squared bishop for a knight. While there are some concrete lines that are important, I will
generally put more of an emphasis on ideas and plans. Black has many different options and
move orders, but the ideas and general evaluation should stay consistent throughout.

Some of the main ideas to keep in mind are:

1) Avoid advancing a pawn to c4. I think it is possible to do so in many situations, and the
computer often recommends it – but as a rule of thumb, I would avoid it unless it offers a

123
tangible benefit. Because Black’s position is so solid, it doesn’t help White to be too aggressive.
Instead, we should look to restrict Black’s play and slowly improve our position. Once the pawn
arrives on c4, Black has some targets too, and I don’t like that commitment.

2) Initially we will develop the bishop to c4, putting pressure on the d5-knight in order to prevent
an early ...c5. However, once White has developed and castled, the ...c5 plan is less bothersome,
so usually I prefer to bring the bishop back to f1 (after Re1). Obviously Bb3 is a reasonable
option too, but it gives Black a number of options to harass the light-squared bishop – including
not only ...a5-a4, but also ...Be6 followed by ...Nc7, offering an exchange. If we allow the light-
squared bishops to be exchanged, it clearly contradicts the strategy mentioned above, involving
the avoidance of unnecessary exchanges.

3) Often, Black might just wait around for White to do something. After developing with Bc4, 0-
0, c2-c3, Re1, Bf1, Nbd2 and deploying the knight to somewhere like b3, c4 or e4, what is White
supposed to do next? One useful idea seems to be Bg5. The bishop is quite annoying for Black
on that square, as ...h6 weakens Black’s kingside, and otherwise White has ideas like Qd2 and
Bh6, or Bh4-g3 (to prevent ...e5 for instance), or even a pawn advance with h4-h5, depending on
the situation. Let me clarify that even though exchanges tend to make Black’s life easier, an
exception can be made for exchanging the dark-squared bishops, as it creates clear weaknesses
on Black’s kingside.

6.Bc4
As mentioned above, this is a crucial move to prevent any early ...c5 ideas from Black. By the
way, Taylor in Alekhine Alert considers this move, but only in connection with Bb3 set-ups,
rather than the more subtle Bf1 plan which I have in mind.

6.Be2 Bg7 7.0-0 0-0 can be compared to variation C1 on page 77. The big difference here is that
Black will arrange a quick ...c5, as shown after 8.c4 Nb6 9.Nc3 Be6 10.b3 c5 when he is
completely fine.

We will consider B1) 6...Be6 followed by the main line of B2) 6...c6.

6...Bg7? is met by 7.Nxf7! and although the game goes on, it is obvious that Black should not
allow this.

B1) 6...Be6

124
 Black can try to avoid playing ...c6 entirely, but it doesn’t look like it helps him much.

7.0-0 Bg7 8.Re1 0-0 9.Nd2 Nd7 10.Nef3 N7b6

It makes sense for Black to get rid of the annoying bishop on c4.

10...N7f6 is sometimes played but this set-up doesn’t seem ideal for Black. A good continuation
is:

11.c3N (11.Rxe6!? has been played by Grischuk and also seems promising, so you could
consider this as well. However, the text move is more in line with the overall strategy we are
adopting for this line, so I want to show that it works well here too.) 11...c6 12.h3 Nc7 13.Bf1²
As long as White avoids unnecessary exchanges, Black will continue to suffocate.

125
11.Bf1
It is always an option to retreat the bishop to b3 in these lines, but I distinctly prefer having it
on f1 where it is free from harassment by ...a5-a4 or ...c6 followed by ...Nc7.

11...Bf5 12.c3 c6 13.h3

Notice the awkwardness of the bishop on f5: Black cannot play ...e6, as g2-g4 will trap it.

13...Re8
Such positions are always solid for Black, and it isn’t obvious what kind of strategy White
should use to improve his position. We are following an instructive correspondence game in
which White failed to pose Black any problems, but I found a useful improvement...

14.a4 a5 15.Nb3 Nd7

126
16.Bg5!N
The game continued with 16.Bd2 Nf8 17.Nc5 Nd7! 18.Nxd7 Bxd7= when Black went on to
draw rather easily. As I have already emphasized, under no circumstances should White
exchange knights so easily! So let’s think about how to create a few more problems for Black.

In general, throwing the bishop to g5 seems to me to be a great idea in this type of position. If
Black takes no immediate action, White can continue improving his position with Qd2 and Rad1.
The queen may then drop back to c1, and Bh6 can be played at any moment, as exchanging the
dark-squared bishops seems useful for White. Even h2-h4 can be considered.
How might Black counter these plans? Chasing the bishop with ...h6 would entail a significant
weakening of Black’s kingside, and the bishop will be fine dropping back to h4, and if necessary
g3. Meanwhile, it is useful to prepare against ...e5. Obviously this move will not be possible with
the queen on d8 – and even after it moves away, exchanges on the e5-square will put the
emphasis on the open d-file, and the bishop will control the crucial d8-square. Moreover, there
may be chances to manoeuvre a knight via e4 to f6.

16...Qc7 17.Qd2 e5
17...Nf8 18.Bh6² also favours White.

18.dxe5 Nxe5 19.Nfd4 Bd7 20.Nc5²

White keeps a slight but enduring initiative.

B2) 6...c6

127
 This is the main line, solidifying the knight on d5 and keeping a few options open for the c8-
bishop.

7.0-0 Bg7 8.Re1

8.c3 is less accurate due to 8...Nd7 9.Nf3 N7b6 when the bishop can’t return to the desirable
f1-square.

We will analyse two main options: B21) 8...0-0 is far more popular but the rare B22) 8...Nd7!? is
arguably more challenging.

B21) 8...0-0

9.c3

128
 This is the move which best fits with our plans.

9.Nd2?! Nb6! attacks the d4-pawn while preparing to trade off the c4-bishop, so White can
forget about an opening advantage.

9.Bb3 is the most popular continuation but 9...Be6! makes it hard for White to avoid the bishop
exchange after ...Nd7 and ...Nc7. Remember that the more pieces get exchanged, the easier the
position becomes for Black – with the exception of the dark-squared bishops. But it seems to me
that if the light-squared bishops are exchanged, White’s edge becomes harder to hold on to,
especially in a practical game. One way to avoid the exchange would be to play c2-c4, but this
has its own drawbacks: it is hardly the prettiest move, and it weakens both the d4-pawn and the
long diagonal, much to the delight of the g7-bishop.

9...Be6
The main alternative is:
9...Nd7 10.Nf3 N7b6 11.Bf1
As I have mentioned before, in principle I prefer to put the bishop on f1 rather than b3.
11...Bg4
Black has no other good square for the bishop, so he gets ready to exchange it.

12.Nbd2!
It is best to recapture with a knight on f3.
The hasty 12.h3?! Bxf3 13.Qxf3 allows quick counterplay with 13...Nd7 14.Nd2 e5÷ when
White is not so well placed to react.
12...Qc7
12...Re8N is met by 13.h3 Bxf3 14.Nxf3 Nd7 15.Bg5 Qc7 16.Bh4² when White remains
well placed to deal with any ...e5 ideas.
13.h3 Bxf3 14.Nxf3 e6 15.g3!

129
 The ...e5 break remains a distant dream for Black, while White enjoys the bishop pair plus
absolute control over the centre.

15...c5!?N
The passive 15...Rfe8 allows us to break one of our ‘rules’ with 16.c4! Ne7 17.Bf4± when
Black is in no position to exploit the slight looseness of the d-pawn.
15...Rad8 16.Bg5 Bf6 17.h4! Bxg5 18.hxg5 c5 19.dxc5 Qxc5 20.Qc2± was great for White
in Arnason – Vaganian, Moscow 1990. (20.Qe2!N is even more accurate, but in any case
White’s advantage lies in his superior structure and activity.)
The text move looks like a logical attempt to find counterplay but it comes unstuck after:
16.c4! Ne7 17.Bf4 Qc8 18.Bd6±
Winning material.

10.Nd2!
This is more accurate than 10.Bf1 Nd7 11.Nf3 Bg4, when ...e5 comes soon.

10...Nd7 11.Nef3 Nc7

Simply trading off the bishop with 11...Bg4 12.h3 Bxf3 13.Nxf3 e6 is less than ambitious, to
put it mildly. With White, you just need to find a decent plan, as Black will probably just be
waiting. 14.Bg5! The bishop is always nicely placed here in these lines. 14...Qc7 15.Qd2 Rfe8
Now a promising new idea is:

130
16.h4!?N I think Black’s position is uncomfortable. If he blocks with 16...h5 17.Rad1 then his
kingside has been weakened in the long term. If he waits and refrains from breaking in the
centre, then White just continues improving his pieces; and even if he does break with ...e5 or
...c5, White’s bishop pair should ensure him of an edge.

12.Bf1

12...c5!
This is clearly the critical continuation. Without it, Black’s position remains passive and his
pieces will be awkwardly placed. Even now, the bishop on e6 is uncomfortable, and opening the
centre does not give quite enough room to reshuffle his pieces in the optimal way. Still, White
needs to play precisely to prove his advantage.

13.Nb3 cxd4

131
14.Nbxd4!N
This seems most accurate.

A previous game continued: 14.Nfxd4 Bd5 15.Bg5 (15.c4!?N is worth considering but 15...e5!
16.Nf5 gxf5 17.cxd5 e4 leads to messy positions resembling the Sveshnikov Sicilian, where I
believe Black should be okay) 15...f6 16.Be3 (16.Bh4 e5 17.Nb5 Bc6 seems okay for Black.)
White went on to win in E. Hansen – Yam, Calgary 2013, but he would have had no advantage
after:

16...Bf7!N With ...e5 and ...f5 threatened, Black is suddenly doing well.

14...Bd5 15.Bf4!
Initially I considered 15.Nc2!? with the idea of Nb4, hoping for some advantage with the
bishop pair, but 15...e5! (15...e6 16.Nb4 and 15...Nc5 16.Nb4 Be6 17.Nd4 seem promising for

132
White) 16.Nb4 Be6 17.Ng5 Bf5! is okay for Black, as 18.g4 Be6 leaves some holes in White’s
kingside.

Neither 15.c4 Bxf3 16.Nxf3 Ne6 nor 15.Bg5 f6 16.Bh4 Bf7 were to my liking either.

However, I then returned to my original idea and realized White does not have to hurry to chase
the bishop. Instead, we can continue developing while keeping this and other options open. The
text move keeps the possibility of Nc2 alive, but also threatens Nb5, while preventing ...e5 and
preparing the simple and unassuming plan of Qc1 and Bh6.

15...a6
15...Ne6? 16.Nxe6 Bxe6 17.Ng5± is obviously good for White.

15...Re8 is well met by 16.Nb5! and White will at least get the advantage of the bishop pair.

15...Nc5 gives White a few options but 16.Qc1² followed by Bh6 seems simplest. Following the
disappearance of the dark-squared bishops, White will gain more targets in the open position;
meanwhile, Black still needs to figure out the best squares for his cluster of minor pieces in the
centre.

16.Nc2!
Reverting to my original plan, in a slightly improved scenario.

16...Ne6
16...Bc6?! 17.Nb4± is excellent for White.

17.Qxd5 Nxf4 18.Qe4 e5 19.Rad1²

Despite the opposite-coloured bishops, White’s position seems much easier to play. For

133
instance:

19...Qc7 20.g3 Ne6 21.Nb4²

The knight is heading for the ideal d5-square.

B22) 8...Nd7!?

Even though this move is far less popular, we will analyse it as the main line, because it leads
to some concrete variations where White cannot just idyllically follow the strategies outlined
earlier.

9.Nf3 N7b6 10.Bf1 Bg4 11.c3

11.Nbd2N is generally desirable, preparing to recapture on f3 with a knight, but here it allows
11...Bxd4. True, White can win back the pawn with 12.Ne4 Bg7 13.c4 Nc7 14.Nd6† Kf8

134
15.Nxb7, but 15...Qxd1 16.Rxd1 Ne6= seems fine for Black.

11...Bxf3!?N
This is the critical continuation, forcing White to recapture with the queen. Anything else
would allow Nbd2 followed by h2-h3, with normal play.

11...0-0 transposes to the 9...Nd7 10.Nf3 N7b6 11.Bf1 Bg4 line in the notes to the variation B21
above; this was the actual move order of the Arnason – Vaganian game quoted there.

12.Qxf3 0-0 13.Bg5!

White needs to play accurately to find an advantage. As usual in these lines, the bishop can be
pretty annoying to deal with on g5.

13.Nd2 Re8 is less convincing, for instance: 14.Qd1 (14.Ne4 is met by 14...Qc7 when I would
like to avoid playing c2-c4, but without this move it will be hard to prevent ...e5; I also checked
14.g3 Qc7 and once again Black will easily prepare ...e5) 14...e5 15.dxe5 Rxe5 16.Rxe5 Bxe5
17.Nf3 Bg7 White may potentially be able to claim an edge with the two bishops, but it is not
much.

13...h6
Kicking the bishop weakens Black’s kingside, but leaving it on g5 would have had a downside
as well.

For instance: 13...Re8 14.Nd2 Qc7 15.Rad1 e5

135
16.c4! Nb4 17.dxe5 Bxe5 18.Ne4 With a strong initiative for White.

14.Bd2!
14.Bh4 f5! leads to insane complications, for instance: 15.h3 f4 (15...g5 16.Bg3 f4 17.Bh2 e5
18.dxe5 Nd7÷) 16.Bd3 g5 17.Bxg5 hxg5 18.Nd2÷

The text move avoids any such trouble with the bishop, while keeping an eye on the h6-pawn,
making it harder for Black to break with ...e5 in the near future. It may look strange to block the
knight from coming to d2, but it will be fine going to a3.

14...Re8 15.Na3 Qc7

15...Nd7 16.Nc4² is also pleasant for White.

136
16.Rad1 e6
16...e5? 17.c4 wins material, for instance: 17...Nf6 18.dxe5 Rxe5 19.Bf4+–

17.Nc2²
Having prevented the ...e5 break, White has free rein to try various plans, such as h4-h5, or
expanding on the queenside, or continuing to manoeuvre the pieces to better squares.

C) 5...c6

This is the main line, stabilizing the knight before deciding how to develop.

6.Be2
Several other moves have been tried but the simple developer is best. Black has three main

137
tries: C1) 6...g6, C2) 6...Nd7 and C3) 6...Bf5.

C1) 6...g6

This is similar to variation B, except that White has already committed his bishop to e2 while
Black has played ...c6. I believe these differences favour White slightly: even though his bishop
is less active here, Black has lost out on the option of an early ...c5, which would have been a
good option after 5...g6 6.Be2. In this version, White’s slow manoeuvring plan does not work
quite as well, but we should not even be trying to play in that way. Instead, we can go for a more
aggressive and ambitious set-up with c2-c4 and f2-f4, maintaining the strong knight on e5 rather
than retreating it. This will save time and put Black under immediate pressure. The ...c5 plan
remains his best chance for counterplay, but he is fighting an uphill battle due to the loss of time
caused by the early ...c6.

7.0-0 Bg7

8.c4!
8.Re1 0-0 resembles the earlier variation B21, except White’s bishop is on e2 instead of c4.
Even though the bishop could be headed for f1 anyway, there is a subtle but important
difference, as the bishop on c4 put some pressure on the d5-knight, while keeping the e-file clear
for the rook, whereas the bishop on e2 invites Black to prepare ...e5. Here is a sample line
showing how those ideas might play out: 9.Bf1 Bf5! 10.h3 (10.c3 Nd7 11.Nf3 e5! 12.dxe5
Re8=) 10...Nd7 11.Nf3 Now in Korneev – Baburin, Chelyabinsk 2017, Black could have
obtained interesting counterplay with:

138
11...e5!N 12.dxe5 Re8 13.g4 Otherwise Black simply regains the pawn with a good game.
13...Be6 14.c4 N5b6 15.Nbd2 Qc7 16.Qe2 h5! 17.gxh5 Bf5 Obviously it would take much
deeper analysis to reach a definitive verdict, but it seems to me that Black has good
compensation, and we are certainly a long way from the kind of stable, long-term edge we were
aiming for.

Fortunately, all this is no problem. Even if White could force a transposition to the earlier line, I
would still recommend the more ambitious approach with c2-c4 and f2-f4 when this specific
move order occurs.

8...Nc7 9.f4!
Reinforcing the knight before Black gets a chance for ...c5.

9...0-0 10.Be3 Nd7 11.Nc3

139
11...Nxe5
11...c5N 12.dxc5 will soon convert to the main line below.

Another example continued 11...Nb6 12.Qe1 f6 13.Nf3 Be6 14.b3 Bf7 15.Rd1± and White was
in full control in Matsuura – Lima, Rio de Janeiro 2004.

12.fxe5
12.dxe5 may also offer White an edge, but it would be minor due to the symmetrical pawn
structure. The text move is more ambitious, and just a stronger move overall.

12...c5 13.dxc5
13.d5!? is a good alternative which has been played several times, with perfect results so far. I
am happy enough with the text move, but it’s a matter of taste.

13...Bxe5
13...Qxd1 14.Raxd1 Bxe5 15.Bf3 leaves Black struggling with development, so despite the
doubled or even tripled pawns on the c-file (in the event that Black exchanges on c3), White’s
advantage is clear.

140
14.Qe1
14.Bf3N² is a good alternative.

14...Bd4 15.Bf3 e5 16.Rd1±

Svidler – Baburin, Bunratty 2009.

C2) 6...Nd7

In general, this move doesn’t seem like the best bet for Black, because he will not be able to
develop his light-squared bishop without it being traded for an enemy knight. When White has
the bishop pair with this pawn structure, it generally ensures him of a pleasant game.

7.Nf3 g6

141
 Black has a solid set-up resembling some of the earlier lines in the chapter, with the bishop
coming to g7 and the knight ready to drop back to c7 if chased. It sounds like it should be
playable enough but Black hasn’t done too well in practice, one reason being that it is hard for
him to arrange an effective central challenge with ...c5 or ...e5.

Another option is:

7...N7f6 8.0-0
Black may develop his bishop to either f5 or g4, but in either case he will lose the bishop
pair and White will enjoy a long-term edge. If Black refrains from developing his bishop,
then why did the knight move to f6?
8...Bf5
8...Bg4 9.h3 Bxf3 10.Bxf3 e6 11.c4 Nb6 12.b3 Be7 13.Bb2 0-0 14.Nd2 was also pleasant
for White in Karjakin – Carlsen, Nice 2008.
9.Nh4 Bg6 10.c4 Nb6 11.Nc3 e6 12.g3 Be7 13.Be3 Qc7 14.Nxg6 hxg6

15.Qb3 Rd8 16.Rfd1 0-0 17.Rac1

White has a pleasant advantage, as is typical for such positions. Black does not have much
counterplay so White has no need to hurry, and can patiently improve his pieces while
contemplating moves like a2-a4 and/or h2-h4 when the time is right. In the game, Carlsen
tried to open up the centre, but White was well placed to meet that idea:
17...e5 18.d5 Nbd7 19.Bf3²
Anand – Carlsen, Nice 2008.

8.0-0 Bg7 9.c4 Nc7 10.Nc3 0-0

142
11.Bf4!
White’s play is mostly quite obvious in this line, whereas Black struggles to find good squares
for his pieces and create counterplay.

11...c5
I checked three other possibilities:

11...b6 12.Qd2 Bb7 13.Bh6± is excellent for White.

11...Nf6 12.h3 Bf5 13.Be5!? is a logical move, as the exchange of dark-squared bishops should
generally favour White, as it clearly weakens the black kingside. 13...Ne4 14.Bxg7 Kxg7 15.Re1

15...Nf6 (15...Nxc3 16.bxc3 strengthens White’s centre and the doubled pawns are not weak in
the slightest) 16.Qd2 Qd6 17.Rad1 Rad8 18.Qe3± White had a typical, pleasant position in

143
Eljanov – Kasimdzhanov, Isle of Man 2017.

11...Ne6 12.Be3
The knight on e6 looks cute but it doesn’t help with the rest of Black’s development, and
there is nowhere for it to go.

12...Nf6 13.h3 Qc7 14.Qd2 Rd8 15.Rad1 b6 16.Rfe1 Bb7 17.Qc1

White should not be in any particular hurry to do anything. Black has completed
development but it is hard to suggest a decent plan for him, as ...c5 seems impossible in the
near future.
17...Nh5 18.Bf1

18...Rac8
18...Rd7 19.d5 cxd5 occurred in Hammer – Lie, Kragero 2018, when 20.cxd5!N would have
preserved a big advantage for White.

144
19.b4 Qb8 20.c5±
Anand – Nakamura, Zurich 2016.

12.d5 e5 13.Bg5 f6 14.Be3

This type of structure would offer Black great counterplay, if only he had time to manoeuvre
the c7-knight to d6 (or even d4) and mobilize his kingside pawns. As things stand, he cannot get
coordinated and White’s initiative flows easily.

14...Rf7 15.Ne4 Qf8 16.b4! b6

16...cxb4 17.c5±

17.bxc5 Nxc5 18.Nxc5 bxc5 19.Nd2

White was clearly on top in Navara – Varga, Bilbao 2014.

145
 C3) 6...Bf5

This has been Black’s most popular choice and it is reasonable to think of it as the main line of
the Alekhine today. With the bishop actively developed on f5, Black prepares to challenge the
active knight with ...Nd7, avoiding the coordination problems of the previous variation where the
bishop was stuck on c8.

7.0-0
I favour simple development. Despite the more favourable placement of Black’s bishop
compared to previous variations, the fact remains that Black suffers from a slight lack of space,
so we should be able to make his life harder as long as we restrict his play.

It is worth briefly checking the more aggressive attempt:

7.c4!?
This leads to interesting complications and I feel there are a couple of ideas worth
mentioning; but in general, I prefer to aim for a solid positional edge against this type of set-
up.
7...Nb4 8.0-0! Nd7
8...Nc2 9.Bg4! leads to some messy, forced lines, slightly favouring White after: 9...Bxg4
10.Qxc2 Qxd4 11.Re1 Nd7 (11...Be6 12.Nd2©) 12.Re4 Qd1† 13.Qxd1 Bxd1 14.Nxd7 Bc2
15.Nc5 b6 16.Na3 Bxe4 17.Nxe4²

146
9.g4!?
9.Ng4 Nc2 10.Ne3 is simply met by 10...Nxe3 11.Bxe3 e6= when the fancy manoeuvring
did not help White at all, as we already know that piece exchanges generally help Black.
9...Be4!
9...Bxb1 10.Rxb1 Nxe5 11.dxe5² and 9...Bc2 10.Qd2 Nxe5 11.Qxb4 Qxd4 12.Be3ƒ both
look promising for White.
After the text move, White has a host of interesting sacrificial ideas which could be
explored further. As mentioned earlier, I feel we should be able to find a strategic edge
against this variation, so it doesn’t seem necessary to be involved in these messy tactics.
Nevertheless, for readers who wish to investigate this direction in more detail, a few
possibilities are:

10.Nxf7!?
There is also 10.Nf3!? Nc2 11.Nc3 Bg6 12.Rb1 Nb4 13.Bf4 Bxb1 14.Qxb1 when White

147
 can aim for long-term compensation, especially since the knight on b4 is somewhat
misplaced. At the same time, after 14...g6 White still needs to find a way to develop his
initiative.
10...Kxf7 11.Nc3 Bg6 12.f4
White has obvious attacking chances in return for the sacrificed piece.

7...Nd7 8.Nf3
As usual, we avoid unnecessary simplifications which would make Black’s life easier. Black
will continue to struggle with the placement of his minor pieces and strive to exchange some of
them. We will try to avoid that – with the exception of clearly favourable trades such as a knight
for an enemy bishop.

8...e6 9.c4 N5f6

9...N5b6?! leaves the knight passive, and 10.Nc3 Be7 11.Bf4! 0-0 12.h3 (12.Qd2 h6 13.Rad1
Nf6 14.Rfe1 is also strong) 12...Nf6 13.g4 Bg6 14.Ne5± was great for White in Castellanos
Rodriguez – T. Lund, Kragero 2019.

9...Nb4?!
White should not be afraid of this move at all, and can choose between two promising
replies, according to personal taste:
10.a3
This is simple and effective.
10.Nc3!? is the maximalist try. After 10...Nc2 11.Rb1 Nb4 12.Bg5! f6 (12...Be7 13.Bxe7
Qxe7 14.Rc1±) 13.Bf4 Bxb1 14.Qxb1 we should not even speak of compensation – White
is simply a lot better.
10...Nc2 11.Ra2

148
11...Nxd4
11...c5? 12.dxc5 Nxc5 13.b4 Nb3 14.Rxc2 just wins.
12.Qxd4 Bxb1 13.Bg5 f6 14.Rxb1 fxg5 15.Rd1!
Black’s position is a wreck.

10.Nc3
10.Bf4 may seem like an attractive idea to prevent ...Bd6, but 10...Qb6!? is annoying.

10.Nh4 would be desirable but after 10...Bxb1! 11.Rxb1 Bd6 the knight on h4 is misplaced and
it will have to return to f3 almost immediately, wasting more time. Meanwhile, Black can castle
and play a quick ...e5 to free his position. In the ensuing open structure, White has some vague
chances to claim a minor edge with his bishop pair, but Black’s coordination is good and White
has failed to show much in the games.

After the text move, Nh4 is a genuine positional threat. Black has two main options: C31)
10...Bd6 and C32) 10...Ne4!?.

10...h6 creates an escape square for the bishop but 11.d5! shines the spotlight on Black’s lack of
development. A logical continuation is: 11...cxd5N (11...Bc5 12.Nd4 Bxd4 13.Qxd4 cxd5
14.cxd5 Qb6 occurred in S. Kasparov – Schnabel, Berlin 2017, when White has several ways to
keep the upper hand but 15.Be3!N seems best) 12.cxd5 Bc5 13.Nd4 0-0 14.dxe6 Bxe6 15.Nxe6
fxe6

149
 Black has a certain amount of activity to make up for his worse pawn structure and the bishop
pair. Nevertheless, 16.Na4 Qe7 17.Nxc5 Nxc5 18.Bc4² clearly favours White.

C31) 10...Bd6

11.Nh4
If White can safely take the light-squared bishop, then I will generally declare a long-term
edge.

11...Bg6
White does not have to hurry to exchange on g6 – although even if he were to do so, the open
h-file should not lead to anything too serious, unless White does something drastically wrong.

150
12.c5!?
This is not a simple decision. To understand how I arrived at it, consider the two other main
options:

Perhaps the most obvious try is:

12.g3 0-0
12...Qa5!? is an interesting attempt to play on the kingside, but White keeps an edge after:
13.Nxg6!N (the bishop should be removed, as 13.Be3 Bh5! 14.g4 g5!„ is quite appealing
for Black) 13...hxg6 14.c5! Cutting off the queen. (14.Bf3 would be met by 14...Qf5!
15.Bg2 g5 and Black suddenly has the initiative) 14...Be7 15.Bf4 0-0 With his queen so far
away, Black can hardly exploit the open h-file, so he may as well castle. 16.Rb1 Qd8 17.b4
Nd5 18.Nxd5 exd5 19.b5²
13.Nxg6
13.c5 Be7! is the reason why I favour 12.c5. The bishop on e7 seems comfortably placed,
and Bf4-g3 is not possible now that g2-g3 has been played, so we can’t really exploit the
fact that Black has ceded control of the h2-b8 diagonal.
13...hxg6 14.Bf3

This position can be evaluated somewhere between equal and slightly better for White,
depending on your understanding of the game. Compared with the earlier note on 10.Nh4,
White’s pieces are nicely coordinated, and the two bishops must count for something. On
the other hand, it is not easy to suggest a clear plan for White, whereas Black has a solid
position and an obvious plan of preparing either ...e5 or ...c5. In a recent high-level game,
Black soon equalized after:
14...Re8 15.Bg2 Qc7 16.Be3 e5 17.c5 Bf8 18.b4 exd4 19.Bxd4 Rad8=
Ragger – Movsesian, Austria 2020.

When it became clear that the above line did not yield anything special, I turned to an interesting

151
alternative:
12.f4!?
This is similar to the main line in the sense that White is playing rather concretely, rather
than just coasting along with the two bishops. However, a good reply is:
12...Ne4!
Before White gets too far up the kingside, Black takes the opportunity to relieve his position
with another exchange.
White would love to see 12...0-0?! 13.Nxg6 hxg6 14.g4! followed by g4-g5, h4-h5 and so
on. Black is not really fast enough to open the centre with any great effect, so he is in grave
danger of being steamrolled on the kingside.
12...Bf5 blocks the kingside but has an awkward feeling about it, and White’s bishop pair
combined with the mobile queenside pawn majority should ensure White of an edge.
13.Nxg6 Nxc3 14.bxc3 hxg6
It seems to me that Black keeps a solid position, and I would rather not have committed my
f-pawn in this way. Play could continue:

15.g3
15.c5? Nxc5! was a cold shower for White in Sutovsky – Miroshnichenko, Kragujevac
2009.
15...Nf6
15...0-0N 16.Be3 c5 is another idea. In either case, the position remains tense, with a lot of
subtleties to explore.
16.Bf3 0-0 17.a4 Qa5 18.Qc2 Rac8 19.Be3 b6 20.Bf2 Rfd8÷
Slyusar – Vicoli, corr. 2019. It is certainly possible to search for an edge in this line, but
personally I don’t feel too happy about White’s pawn structure, particularly with the dark-
squared bishop stuck behind the pawns.

152
12...Bc7
This has been the usual choice in correspondence games and I consider it the most challenging
move.

12...Be7 13.Bf4!
In the similar situation in the 12.g3 line in the previous note, White’s bishop would not have
had much of a future on this diagonal – but with the pawn on g2, it’s a different story.
13...0-0 14.Nxg6 hxg6 15.Re1 Nd5
Black will have to play this move sooner or later. Whenever it happens, White will trade
knights and apply pressure on the queenside.
15...b6 16.b4 a5 17.a3² is pleasant for White, and 17...Nd5?! 18.Nxd5 exd5 19.b5!± only
makes matters worse for Black.
16.Nxd5 exd5
In Efimenko – Kopylov, Germany 2017, White missed the strongest continuation:

153
17.Qb3!N
Posing awkward problems on the queenside.
17...Bf6
17...Qc8 18.Ba6! is unpleasant.
If 17...b6 then White maintains the pressure with 18.Ba6 or 18.Qa4.
18.Rad1 b5
18...b6 gives White a few ideas; I like 19.Bg4!?², when 19...bxc5? is inadvisable due to
20.Qb7.

19.Bg4 Re8 20.Rxe8† Qxe8 21.Qc3 Nf8 22.Re1 Qd8 23.Bd6²

With lasting pressure.

13.Bg5 0-0

154
14.Nxg6
It is hard to avoid this for too long – but once Black has castled, there is no reason to delay it
any longer.

14...hxg6 15.b4
Not only preparing a future queenside advance, but also defending the c5-pawn so that ...e5 can
be met by d4-d5.

15...Re8
15...Qb8 16.g3 gets nowhere for Black, and if 16...Nd5 17.Nxd5 exd5 18.b5± White is in full
control.

The text move is more logical. The rook supports a future ...e5 break, while also making room
for ...Nf8-h7 to untangle from the annoying pin. Black has done decently in a handful of
correspondence games from this position. The general pattern of events has been that White
failed to pose concrete problems or apply meaningful pressure with the bishop pair, and Black
eventually equalized.

I believe that practically, White has a more pleasant position, but it is important to have a clear
sense of what we are doing, rather than just manoeuvring pieces around and hoping for the best.
Black’s main ideas are the ...e5 pawn break and the knight manoeuvres ...Nf8-h7 and ...Nd5,
aiming to simplify things. We need to devise a strategy directed against those ideas, and to this
end I suggest:

155
16.Qb3!?N
16.Rb1 a6 17.g3 Nf8 18.a3 N8h7 19.Be3 Nd5 20.Nxd5 Qxd5 21.Bf3 Qf5 22.a4 Rad8 23.Qe2
e5= occurred in Groot – De Ridder, corr. 2014; a good example of how Black can equalize when
he is allowed to carry out the aforementioned ideas.

I like the queen on b3, where it may support b4-b5, while also keeping the d5-square in sight.
Still, we could also put the queen on c2 and proceed in a similar manner as shown below. The
real novelty here is not so much the individual move marked with an “N”, but rather the plan I
have devised in order to improve White’s pieces and anticipate Black’s ideas. Our next move
should be Rad1 so that when ...Nf8-h7 arrives, the bishop can drop neatly back to c1. White can
then look to advance on either or both flanks. On the kingside we will aim for g2-g3, Bf3, Kg2
and h2-h4, at which point attacking ideas with h4-h5 will start to become relevant and
dangerous. On the queenside, we will always have the option of b4-b5, although there is no hurry
for it, and I tend to favour the idea of strengthening White’s position across the board before
launching an attack anywhere. Black’s position remains pretty solid of course, but I like White’s
chances with the plan I have described. Let’s see how these ideas may play out.

16...Nf8 17.Rad1 N8h7 18.Bc1 a6

Black usually includes this move at some point, to avoid the possibility of b4-b5 coming at any
moment.

19.a4
The inclusion of these two pawn moves does not change much from our perspective.

19...Nd5 20.g3 Nhf6 21.Bf3 Nxc3 22.Qxc3 Nd5 23.Qb3

Black has achieved something worthwhile in exchanging a pair of knights and getting his other

156
knight to the big central outpost – and yet he is still some distance away from full equality. The
...e5 break is hard to pull off because of the pressure on d5, so Black must either sit and shuffle
his pieces around or try to do something purposeful – which comes at a cost, as we will see.

23...Ne7!?
Given enough time, Black would love to regroup with ...Nf5, ...Qd7, ...Bd8-f6, ...Rad8,
perhaps reshuffling the heavy pieces on the d-file and then finally picking off the d4-pawn, or at
least reducing White to a horribly passive defence. However, in the time it takes him to carry out
those manoeuvres, White can launch a strong attack.

23...Qd7 24.Kg2² does not really change anything.

23...b6!? 24.Kg2 Rb8 is another idea Black might try. However, White keeps control after
25.Bd2 Qd7 26.Qd3 b5 27.a5 Bd8 28.h4² when he still has good chances to develop threats on
the kingside. Moreover, the changes in the queenside structure carry some strategic risks for
Black: if the position opens up, especially in conjunction with some major-piece exchanges, the
a6- and c6-pawns will be obvious targets for White’s light-squared bishop – and the pawns on a5
and c5 are only a few steps away from becoming queens.

24.Bb2
24.Kg2 Nf5 25.h4!? Nxd4 26.Qc4 e5 27.Bb2 Qf6 28.Bxd4 exd4 29.Rxd4² gives White a bit
of pressure with b4-b5 coming up, but there is no need to allow such simplifications to occur.

24...Qd7 25.Kg2 Nf5?!

Objectively speaking, Black should accept that he is worse and settle for something like
25...Nd5 26.h4². In that case, White remains in control, with two bishops and possibilities for
pawn breaks on either flank.

157
The text move is more critical, as Black’s regrouping threatens to put serious pressure on the d4-
pawn, while the knight on f5 and bishop on f6 seem like they could be useful for defending the
kingside as well. However, there is a flaw in Black’s scheme, which White can exploit in
instructive fashion.

26.h4 Bd8
26...Rad8? is too slow, and 27.Rh1+– leaves Black without a good answer to h4-h5.

27.Rh1 Bf6 28.h5 g5

28...gxh5? allows a faster win by means of:

29.d5!! (29.Rxh5± is not a bad reserve option) 29...exd5 (29...Bxb2 30.dxc6+–) 30.Bxf6 gxf6
31.Rxh5 Ng7 32.Rh4+– The kingside threats will decide.

158
29.h6! Nxh6! 30.d5!
This dynamic breakthrough is the reason why Black would have been safer leaving his knight
on d5.

30...exd5 31.Bxf6 Qf5 32.Ba1 g4

White is temporarily a piece up but his bishop is trapped. The forcing line continues:

33.Bxd5 Re2!
A good try, but it is not quite enough.

34.Rhf1 cxd5 35.Qxd5 Rae8 36.Qxb7 R8e4 37.Kg1 Qf3

Black is still in the game: he threatens ...Rxf2, and White only has one route to victory.

159
38.Rd4!!
38.Rd8† Kh7 39.Rd4 Nf5 40.Rxe4 Nxg3! gives White no more than a draw.

38...Nf5
38...Rxf2? fails to 39.Qa8†!+– and the e4-rook falls with check.

39.Rd8†!
39.Rxe4 Rxe4 enables Black to draw.

39...Kh7 40.Qxf7+–
Obviously nobody needs to remember all this – but it’s good to see how the dynamic attacking
possibilities of h5-h6 and d4-d5 can be put to use when the opportunity arises.

C32) 10...Ne4!?

This is a modern continuation which has found favour among several strong players. Black
looks to exchange a knight for a knight, rather than bishop for knight as in the previous variation
where Nh4 was permitted.

11.Re1!?
Surprisingly, this natural move has yet to be tested in a practical game, although it has occurred
in correspondence play. We don’t want to exchange pieces too easily, which rules out certain
options such as taking on e4 immediately. So instead we make a useful developing move, which
happens to create a subtle threat of Nxe4 followed by Bf1, highlighting the fact that Black’s king
is still in the centre. If Black exchanges on c3, we should consider it a success as we get the open
b-file, while the doubled pawns are not problematic at all.

160
As mentioned above, 11.Nxe4 is too cooperative, and 11...Bxe4 12.Re1 Nf6!?N seems fine for
Black.

We can also rule out 11.Bf4 because 11...Bd6! leads to simplifications which help Black.

11.Na4!?N takes things to the opposite extreme, avoiding simplifications at the expense of some
time. Perhaps it is worth considering, but it feels a bit weird to play such a move, and I don’t
want to recommend violating opening principles without a compelling reason if I can avoid it.

11.Qb3!?
Out of the many alternatives to have been tried, this seems like the best attempt to cause
problems. However, Black seems to be okay with precise play, as shown after:
11...Nxc3
11...Qc7 allows 12.Nh4.
12.bxc3 Qc7
12...b6 13.Bf4 Be7 14.h3 (or 14.Rad1 0-0 15.h3 h6 16.Bg3) 14...0-0 15.a4 seems pleasant
for White.
13.Nh4

13...Be4!
This situation resembles a popular variation of the Slav Defence, where Black looks to
provoke a slight softening of White’s kingside before allowing his bishop to be exchanged.
13...Bg6 is met by 14.g3 when a tough battle lies ahead, but White is getting the kind of
position he wants with two bishops.
14.Re1
14.f3 Bg6 gives Black what he wants, and ...Bd6 will be annoying.
14.g3 can be considered but I think Black is okay; 14...Bd6, 14...Be7 and 14...b6!? all
deserve consideration.

161
14...Be7 15.Bf1 Nf6 16.f3 Bg6
Having succeeded in provoking f2-f3, Black was doing fine in Wei Yi – Zhou Jianchao, China
2016.

Returning to our main line, Black’s two most logical options are C321) 11...Be7N and C322)
11...Bd6. Other moves are easier for us to handle, for instance:

11...Ndf6?! 12.Qb3!
Black runs into tactical problems, as shown after:

12...Qc7
12...Rb8 13.Bd3 Nd2 (13...Nc5 14.dxc5 Bxd3 15.Rd1+–; 13...Nxc3 14.Bxf5+–) 14.Bxd2
Bxd3 15.d5 and Black is in trouble.
12...Qb6 13.Bd3 Qxb3 (13...Nd6 14.Bxf5 Nxf5 15.d5 leads to the same thing, as Black will
hardly benefit from keeping the queens on) 14.axb3 Nd6 15.Bxf5 Nxf5 16.d5±
13.Bd3 Nc5
13...Nxc3 14.Bxf5+– is obviously no good for Black.
The text move seems like the best try, but White has a great position after:
14.dxc5 Bxd3

162
15.Bf4! Qc8
Taking the bishop would be hopeless, as the loss of the b7- and c6-pawns combined with
Black’s vulnerable king would lead to a quick collapse.
16.Rad1 Bf5 17.Bd6±
11...Nxc3 12.bxc3
Black has to be careful here too.

12...Be7
12...Bd6 13.c5! shows one of the benefits of the bxc3 structure: we can play this advance
without permanently weakening the d4-pawn or giving away the d5-square. 13...Bc7
14.Bg5 Qb8 15.Nh4! Bg6 16.g3±
12...Qc7 13.Nh4² is pleasant for White.
13.Bf4
13.d5!? 0-0 14.Nd4 also looks tempting.

163
13...Qa5!?
13...0-0 14.Qb3 b6 is safer, although White’s position remains slightly more pleasant.

14.d5! cxd5 15.cxd5 0-0 16.Nd4 Nc5 17.Be5

With a promising initiative for White.

C321) 11...Be7N

12.Bf4!?
This option is much more appealing here than on the previous turn, because now ...Bd6 would
waste time – and more importantly, run into immediate trouble because of a quick d4-d5.

12.Nxe4 Bxe4 13.Bf1 is a reasonable alternative.

164
12...0-0
As alluded to above, 12...Bd6? 13.Bxd6 Nxd6 14.d5!± is no good for Black.

12...Nxc3 13.bxc3 seems promising for White. If 13...0-0 14.Qb3 b6 15.a4² White intends a4-a5
with pressure on the queenside. Black can prevent this with 13...Qa5!? 14.Qb3 b6, but after
15.a4² his queen seems rather awkward.

13.Qb3 Qb6
This is the only way of defending b7 which isn’t completely passive.

13...Nxc3 14.bxc3 transposes to the note above, where 14...b6 15.a4² was mentioned. It is worth
adding that 14...Qb6? would give Black a worse version of our main line, due to 15.c5! Qxb3
16.axb3± when White’s pawn structure is amazing.

14.Bd3!
Surprisingly, Black suddenly starts to experience trouble along the e-file.

14...Qxb3
14...Bd6 15.Bxe4 Bxf4 16.Bxf5 exf5 17.Qc2² gives White a pleasant game with the superior
pawn structure.

15.axb3 Ng3
15...Nxc3? 16.Bxf5 wins a piece for White.

15...Nd6?! 16.Bxd6! Bxd6 17.Bxf5 exf5 18.c5± gives White a clear structural advantage.

15...Bd6 16.Bxe4 Bxf4 17.Bxf5 exf5² and the game goes on, but again White’s better structure
leaves his advantage in no doubt.

165
16.hxg3
16.Bxf5 Nxf5 17.d5 also offers White good chances for an edge.

16...Bxd3 17.d5 Nc5

I also considered 17...cxd5 18.cxd5 when Black remains under some pressure after: 18...Bg6
(18...Bf5?! 19.Nd4± is even worse) 19.dxe6 fxe6 20.Rxe6 Bf6 21.b4²

18.dxc6
18.b4 Nb3 19.dxc6 Nxa1! (19...bxc6 20.Ra3 Bxc4 21.Ne5 Bxb4 22.Nxc4²) 20.cxb7 Nc2
21.bxa8=Q Rxa8 22.Rd1 Nxb4 should be holdable for Black.

18...bxc6
White has a pleasant choice between two different types of slightly advantageous endgames.

166
19.b4
19.Ne5 Bf6 may lead to an opposite-coloured-bishop endgame after: 20.Be3 (20.Nxc6 Nxb3
21.Ra3 Rfc8 22.Nxa7 Bxc3 23.bxc3 Bxc4 should not be too difficult for Black to hold, despite
White’s extra pawn) 20...Bxe5 21.Bxc5 Bxc3 22.bxc3

22...Rfb8 23.Re3 Bc2 24.b4² It is not clear how much better White is, but he certainly has a
pleasant endgame with no risk.

19...Nb3 20.Ra3 Bxc4 21.Ne5 Bxb4 22.Nxc4 Bxa3 23.bxa3

Black is materially fine, but he will have to suffer to equalize because White’s pieces
coordinate so well.

C322) 11...Bd6

167
 This has been the choice of correspondence players. Black develops the bishop actively, not
fearing the tactical possibilities along the e-file.

12.Nxe4 Bxe4 13.Bf1 Nf6

13...Bxf3 14.Qxf3² reaches a typical bishop-pair situation that White can be happy with.
13...Bf5 invites 14.d5 Nf6 15.dxe6 Bxe6 (15...fxe6 16.h3²) 16.Ng5² with an edge for White.

13...Bg6 is also well met by 14.d5, due to the following point: 14...cxd5 15.cxd5 e5

16.Bf4! f6 (16...0-0 17.Nxe5² is a safe pawn grab) 17.Nd4ƒ

14.Bg5

168
14...h6!?N
This move has not been tested, but it forces the play and is therefore critical.

14...Bg6?! 15.d5!± gives White an even better version of a familiar theme.

14...Bf5 was played in Bolz – Nepustil, corr. 2017, and now 15.d5!?N would be the natural and
thematic attempt to put Black under pressure. Play may continue:

15...0-0 (15...cxd5?! 16.cxd5 0-0 17.dxe6 Bxe6 18.Bxf6±) 16.dxc6 bxc6 17.g3² White has a
nicer structure and good chances to press.

15.Bxf6!
15.Bh4 is well met by 15...Bxf3 16.Qxf3 g5! 17.Bg3 Bxg3 18.hxg3 Kf8= when, with only a
single bishop, White has no claim to an advantage; I would actually take Black’s side if given the

169
choice.

15...Bxf3 16.Bxd8 Bxd1

17.Ba5! Bh5 18.d5

Even in the endgame, this thematic break works well. Black’s best defensive attempt is to give
up a pawn.

18...cxd5
18...0-0 allows a neat tactic: 19.c5! Bxc5 20.dxc6 b6 (20...bxc6? 21.Re5+– exploits the
alignment of Black’s bishops) 21.Bc3 Rac8 22.b4 Be7 23.b5± Material is equal but the pawn
structure most definitely isn’t.

19.cxd5 0-0
19...e5?! is simply met by 20.Bc3 f6 21.f4± when White wins a pawn in more favourable
circumstances.

170
20.dxe6 fxe6 21.Rxe6 Bc5 22.Be1²
Black’s activity gives him a certain amount of compensation; nevertheless, White has realistic
chances to convert his extra pawn.

Conclusion

This chapter has dealt with the most popular branch of the Alekhine these days, namely 4...dxe5
5.Nxe5. The notorious 5...Nd7? variation has long been regarded as suspicious and there have
always been good reasons for Black not to play it, but only in recent years has it been
conclusively refuted, thanks to the engines of course.

We then moved on to 5...g6, which is solid and generally not a bad line at all. I recommend
meeting it with 6.Bc4, before castling and later rerouting the bishop to f1, with good chances of a
long-term squeeze.

5...c6 is the big main line, when 6.Be2 is our choice. Then 6...g6 seems like an inferior cousin of
5...g6, as Black’s slow set-up with ...c6 invites us to seize a lot of space with c2-c4 and f2-f4,
gaining excellent chances. 6...Nd7 is also not too impressive, as White simply avoids
unnecessary exchanges with 7.Nf3 and leaves Black with a cramped position.

6...Bf5 is the best and most popular continuation. Then 7.0-0 Nd7 8.Nf3 e6 9.c4 N5f6 10.Nc3
reaches an important position, where both 10...Bd6 and 10...Ne4!? present their own challenges.
In both cases, I had to come up with some rather sophisticated ideas to counter Black’s set-up,
which has held up quite well even in correspondence games. Although Black’s system is
certainly playable, I believe I have shown some nice ways to create problems for our opponents
in these lines.

171
172
A) 3...Nxd5 4.c4 97
A1) 4...Nf6?! 98
A2) 4...Nb6 5.Nf3 99
A21) 5...Bg4 100
A22) 5...g6 6.Nc3 Bg7 7.c5! Nd5 8.Bc4 101
A221) 8...Nxc3 102
A222) 8...c6 104
B) 3...Bg4 4.Bb5†!? 105
B1) 4...c6!? 106
B2) 4...Nbd7 5.Be2 Bxe2 6.Qxe2 Nxd5 7.c4! N5f6 8.Nf3 e6 9.Nc3!? 109
B21) 9...Bd6 111
B22) 9...Bb4!? 113
B221) 10.Bd2!? 114
B222) 10.0-0 115

173
1.e4 d5 2.exd5 Nf6
In this branch of the Scandinavian, Black delays the recapture on d5, often with the intention of
sacrificing a pawn. Even if White refrains from an immediate c2-c4 in favour of developing with
3.d4 or 3.Nf3, Black often continues with 3...Bg4, aiming to provoke White into either making
an ugly concession with f2-f3 or allowing simplifications that may help Black to equalize by
delaying the capture on d5. In recent years, this system has been championed by Australian GM
David Smerdon, who wrote an excellent book building this into an easily accessible repertoire
for Black. Rather shamelessly, I also used Smerdon’s Scandinavian to identify potentially
promising lines to use as White!

I once suffered a rather painful defeat as a kid when playing with White against one of these
gambit lines, and I have been wary of such positions ever since. I was adamant on avoiding 3.d4
Bg4 4.f3 followed by clinging to the extra pawn in this repertoire, as everything I have seen
about this line makes me uncomfortable about White’s side. Even though extensive analysis
could help assuage those doubts, I do not believe there are many players who would wish to
analyse a sub-variation of the Scandinavian in such depth as a main line of the Najdorf.
Moreover, accepting the gambit gives Black exactly the kind of position he wants: even if his
compensation is objectively suspicious, he still gets a certain initiative with a lot of unexplored
territory and room for surprising ideas, as Smerdon shows in his book. Fortunately, among the
various lines where White declines the gambit, there was a particular move order covered in
Smerdon’s Scandinavian which seemed promising to me, and on closer inspection I found a nice
refinement which the Australian GM did not consider. In this way, I found a route to a slight
advantage, while nicely avoiding the complications which would have offered Black the active
counterplay he desires.

3.d4
My initial thought was to recommend 3.Nf3, which blunts Black’s 3...Bg4 plan – although the
move is still playable. However, it then dawned on me that 3...Qxd5!? would present a big
problem to our repertoire, as it converts to the 2...Qxd5 lines where Black has avoided my

174
preferred set-up delaying Nf3. This was a disappointing revelation for me, but then I returned to
the text move and eventually found a way of playing which ticks all the necessary boxes.

We will analyse A) 3...Nxd5 and B) 3...Bg4.

3...Qxd5 4.Nc3 will convert to a variation covered in the next few chapters, depending on where
Black moves the queen.

A) 3...Nxd5

This move has fallen a bit out of fashion these days: it doesn’t have the flashiness of the gambit
approach from variation B below, and it allows White to build a nice pawn centre with c2-c4. At
the same time, we will need to show some precision if we are to cause Black problems.

4.c4
We will consider A1) 4...Nf6?! followed by the more challenging A2) 4...Nb6.

A1) 4...Nf6?!

The knight simply seems to be worse here than on b6, where it would exert some pressure on the
queenside. For instance, Black would have ideas such as ...Bg4xf3, when recapturing with the
bishop could leave the c4-pawn hanging. Also, the f6-knight will block the bishop on g7, should
Black develop it there. Essentially, Black is just accepting a worse position with limited
prospects for counterplay.

5.Nf3
5.Nc3?! needlessly allows 5...e5!.

175
5...Bg4
5...g6 is possible although, as mentioned above, the f6-knight just gets in the way of the
bishop. 6.Nc3 Bg7 7.Be2 0-0 8.0-0 c6 9.Bf4 Black is obviously a long way from equalizing. A
good example continued:

9...Bf5 10.h3 Ne4 11.Qb3 Qb6 12.Qa3 Nxc3 13.bxc3 c5 14.g4 Bc8 15.Rab1 Qe6 16.Rfe1 Na6
17.Bf1 Qc6 18.Ne5± Ponkratov – Rakhmanov, Kaliningrad 2015.

White also gets a pleasant game after:

5...e6 6.Nc3 Be7 7.Be2 0-0 8.0-0
Black has a passive position. A good example continued:

8...b6
8...c5 is well met by 9.dxc5! Bxc5 (matters are similar after 9...Qxd1N 10.Rxd1 Bxc5 11.a3

176
 or even 11.Ne5!?) 10.a3N²/± intending b2-b4, Be3 (or Bb2) and so on. Despite the
simplifications, Black is struggling to complete development.
9.Ne5 Bb7 10.Bf3 Qc8 11.Bf4 Rd8
11...Nbd7 12.Nc6±
11...Nfd7!?N may be the best try, although 12.Bxb7 Qxb7 13.Qg4 Nf6 (13...Nxe5 14.Bxe5
g6 15.d5±) 14.Qh3² still favours White.
12.Re1 c6 13.Bg5±
Movsesian – Piorun, Warsaw 2010.

6.Be2 c6
Trying to do without this move does not really help Black. For instance, after 6...e6 7.0-0 Be7
White can start to target b7 right away: 8.Qb3 Qc8 9.h3 Bh5 10.Nc3 0-0

11.g4 Bg6 12.Ne5 Nbd7 13.Nxg6 hxg6 14.Bf3 c6 15.Bf4 White had an excellent position in
Degraeve – Kovarcik, Saint Affrique 1999.

177
7.0-0
This is a good example of why the knight is not well placed on f6; if it was on b6, White would
have to worry about ...Bxf3 due to the loose c4-pawn.

7...e6 8.Nc3 Be7 9.h3 Bh5

10.Re1!
10.Be3 has been much more common but after 10...Nbd7 the d4-d5 advance is not so effective,
and it is still too early to go hunting the light-squared bishop since Black has not yet castled. Of
course White should still be a bit better, but the text move simply seems stronger.

10...0-0
10...Nbd7 is well met by 11.d5!, for instance: 11...cxd5 (11...Nc5 12.b4 Nce4 13.Nxe4 Nxe4
14.dxc6±) 12.cxd5 exd5 13.Nd4 Bg6 14.Ba6!±

11.g4
Now that Black has castled and there are no worries about the h-file, this works perfectly.

11...Bg6 12.Ne5 Nbd7 13.Nxg6 hxg6 14.Bf4

White had an obvious advantage in Westerberg – Ajrula, Skopje 2014.

A2) 4...Nb6

178
 When investigating this variation, I was surprised to find that if White simply develops his
pieces and castles, Black will be doing fine. Therefore White needs to go for some concrete
variations involving a quick c4-c5 in order to put pressure on Black.

5.Nf3
It is best to start with the king’s knight, as 5.Nc3 e5! would be quite alright for Black.

We will analyse A21) 5...Bg4 followed by the main line of A22) 5...g6.

5...Nc6 6.d5 Nb8 seems a little too provocative, and after 7.Nc3 e6 8.Bd3² White should be able
to count on a pleasant game.

5...Bf5 has occurred in a reasonable number of games but I don’t think it requires any special
analysis. After 6.Nc3 e6 7.Be2 Be7 8.0-0 0-0 9.Be3² White enjoys a clear space advantage and
Black’s pieces are not doing much to challenge White’s pawn centre.

A21) 5...Bg4

179
 With this move, Black at least makes a meaningful attempt to put pressure on the centre.
However, White can solve that problem while also seeking to exploit the weakened b7-pawn:

6.c5!
Instead after 6.Be2 e6 7.0-0 Be7 8.Nc3 0-0 9.Be3 Nc6 Black seems to be doing quite okay.
There is definite pressure on White’s central pawns, and Black may consider ...f5 next.

6...N6d7
Black also has a hard time after the alternative:
6...Nd5 7.Qb3!
There is no easy way to defend the b7-pawn.
7...Nc6
7...b6?? loses a piece to 8.Ne5.
7...Bxf3 8.Qxb7 is a safe pawn grab, and after 8...Nd7 9.gxf3 e6 10.Nc3± the doubled
pawns are not too big a problem.
7...Bc8 avoids an immediate calamity but it is too passive, and White can finish his
development however he wants.
8.Qxb7 Ndb4

180
9.Bb5 Bd7 10.Bxc6 Nxc6 11.0-0±
White keeps an extra pawn.

7.Bc4
7.Qb3 Nc6! is not so clear. Instead White should develop as usual, exploiting the fact that
Black’s knight is passively placed.

7...e6 8.Nc3
8.h3 allows 8...Bxf3! 9.Qxf3 Nc6 10.Be3 Qf6 when the ensuing endgame doesn’t feel too bad
for Black.

8...Be7
8...Nc6 9.Be3 should soon convert to the main line below. Note that 9...Bxf3?! 10.Qxf3 Qf6
does not work here because 11.Qg3!± is excellent for White. A practical example continued:

181
11...Nxd4 12.0-0-0 Bxc5 13.Qxc7 0-0 (13...Bb6 14.Qxb7 Rb8 15.Qe4±) 14.Qxd7+– Black had
nothing like enough compensation for the piece in Green – Radnai, Budapest 2009.

9.h3 Bh5 10.Be3 Nc6

The more popular 10...0-0 allows White to launch a virulent attack with: 11.g4 Bg6 12.h4! h5
13.Ng5 (13.Bd3!?N could also be considered)

13...Nf6 14.gxh5 Bxh5 15.Be2± This occurred in Suetin – Gipslis, Minsk 1993, and several
other games. Obviously the game goes on and the outcome is far from certain, but White’s
kingside attack makes him the clear favourite.

11.Qe2
Preparing to castle on the queenside. After that White will easily develop an initiative on the
kingside, while Black’s pieces remain rather passive.

182
11...0-0
11...e5 was of no help to Black after 12.g4 Bg6 13.0-0-0 exd4 14.Nxd4 0-0 15.Ne6 fxe6
16.Bxe6†± in Santo-Roman – Goldenberg, France 1992.

12.g4 Bg6 13.0-0-0 Na5 14.h4! Nxc4 15.Qxc4 h5 16.d5ƒ

Black faced some difficult problems in Starke – Williams, corr. 2000.

A22) 5...g6

This is the main line and the most logical choice. The bishop will exert pressure on the centre
from g7, and in the meantime Black prepares to castle and avoids a premature ...Bg4, which we
have already seen exposes both the bishop and the b7-pawn.

183
6.Nc3 Bg7 7.c5!
As mentioned earlier, simple development does not lead anywhere special, for instance:

7.h3 0-0 8.Be3 c5 9.d5 Na6 10.Qd2 e6 and Black seems fine.

7.Be3 0-0 8.Be2 Nc6 9.0-0 Bg4 and again, Black seems to be doing quite well.

7...Nd5
The alternative is too passive:
7...N6d7 8.Bc4 0-0 9.0-0 b6 10.Re1!
Rather than worry about the c5-pawn, White develops another piece and prepares Bg5 with
an unpleasant attack on e7 – a typical plan in these lines.
10...bxc5
10...Bb7 11.Bg5 Nf6 12.Ne5± is excellent for White.
Accepting the pawn sac must be critical, but White gets more than enough play for it.
11.Bg5 Bf6 12.dxc5N 12...Nxc5

13.Bh6
White has many tempting alternatives, but one good option is enough.
13...Re8
13...Bg7 14.Bxg7 Kxg7 (14...Qxd1 15.Raxd1 Kxg7 16.Rxe7±) 15.Nd5+– gives White an
overwhelming initiative.
14.Ne5 Bxe5 15.Qf3! Be6 16.Bxe6 Nxe6 17.Qxa8±

184
8.Bc4
One of the big advantages of the c4-c5 move is that White’s bishop can come to this excellent
square.

Black can either trade off his knight with A221) 8...Nxc3 or support it with A222) 8...c6.

A221) 8...Nxc3 9.bxc3 0-0 10.0-0 Nc6

11.Bg5!
I regard this as the most precise move order.

It has been more popular to start with 11.Re1, which often transposes to one of the lines below.
However, Black can take the game in a different direction with: 11...Bg4!? 12.Bg5 (12.h3? runs

185
into 12...Bxf3 13.Qxf3 Nxd4! 14.cxd4 Qxd4³) 12...h6 13.Bh4 g5 14.Bg3 e5 Although White
may still be able to prove an advantage here, the position is quite complicated, so we may as well
avoid it with my recommended move.

11...Na5
11...Bg4 is met by 12.h3± when, unlike the note above, there is no ...Nxd4 trick to worry
about.

11...b6 12.Re1 h6 (12...bxc5 13.d5±) 13.Bh4 puts Black under serious pressure, especially since
13...g5? allows White to win with a crushing attack after: 14.Bxg5! hxg5 15.Nxg5 Bf5 16.Qh5
e6 17.Rxe6!+– Ladva – Vicas, Narva 2015.

12.Bd3 b6
12...Be6 13.Re1± maintains strong pressure. 13...Re8 This position was reached via a slightly
different move order in Madar – Vrbova, Slovakia 2010. An obvious improvement is:

14.Rxe6!N This thematic sacrifice is strong, although White would remain better after any
sensible move. 14...fxe6 15.Qe2 Qd7 16.h4 White is absolutely dominating and, with h4-h5
coming next, Black is in danger of being crushed on the kingside.

186
13.Re1 Re8!?N
13...f6 14.Bf4 Bb7 (14...bxc5 15.Qa4 c6 16.Rad1 gives White a huge initiative) 15.Be4 Bxe4
16.Rxe4 Rf7 17.Qa4 c6 18.Re6² saw White exploit the weakness created by Black’s 13th move
in Simonyan – Nugumanov, Yerevan 2017.

The text move is a natural attempt to improve Black’s play, but White has more than one
promising continuation. For instance:

14.Rb1
Another option is 14.h3 bxc5 15.Qa4 c6 16.Rab1© with promising play for a pawn.

14...bxc5 15.Rb5 Nb7 16.Qb3 cxd4 17.Rbe5!

I could not resist including this pretty move – although of course the immediate 17.Rxb7 is
better for White as well.

17...Bxe5 18.Nxe5 Be6 19.Qxb7²

A222) 8...c6

187
 This is the more solid choice, trying to block the c4-bishop.

9.0-0 0-0
Sometimes Black tries 9...Be6, but 10.Qb3 b5 11.cxb6 axb6 12.Ng5 leaves him in an
uncomfortable position where he is unable to avoid Nxe6.

10.Re1 Bg4
Several other moves have been tried, but White gets an easy advantage against all of them.

10...b6 11.Bg5 Re8 12.Qb3± is excellent for White.

10...h6 prevents the typical Bg5 move but is too slow. After 11.Qb3 e6 12.Ne4± the best thing
one can say about Black’s position is that it is solid and will be hard to break down. Still, Black
is passive and White has all the space in the world, along with excellent squares for his knights,
such as e5 and d6.

10...Be6 has occurred numerous times but for some reason no one has gone for the thematic
sacrifice:

188
11.Rxe6!N 11...fxe6 12.Ng5 Rf6 13.Qg4±

10...Nxc3 11.bxc3 Nd7

11...b6 12.Bg5 Bf6 13.Qd2!N‚ gives White a dangerous initiative, especially since
13...bxc5? 14.Bxf6 exf6 15.Qh6 is unplayable for Black.
11...b5 12.Bb3 does not help Black at all. For example, 12...Bg4 13.Bg5 Bf6 14.Bxf6 exf6
15.h3 Bf5 16.a4 a5 17.Qe2 bxa4 18.Rxa4 Qc7 19.Qe7 Nd7 20.g4+– and White was already
winning in Greet – Stefansson, Beijing (blitz) 2008.
12.Bg5 Nf6
This position has been reached in several games. White is better after practically any
sensible move, but I especially like the new idea:

13.a4!?N
A useful move, taking some extra space on the queenside. A sample continuation is:

189
13...Bf5
13...b6 invites 14.Ne5 with a strong initiative for White.
14.Qb3
14.a5!? also makes sense.
14...Qc7 15.Ne5 e6 16.Bf4 Qc8 17.h3±

11.Bg5 Nf6
11...Bf6 12.Bh6 Bg7 (12...Re8 13.h3 Bxf3 14.Qxf3 Nxc3 15.bxc3±) 13.Bxg7 Kxg7 has
occurred in a few games but no one went for the straightforward win of a pawn with White:

14.Bxd5!N 14...cxd5 15.Qb3 Bxf3 16.Qxb7 Nd7 17.gxf3 e6 18.b3 The spoilt structure on the
kingside may look slightly worrying, but Black has no good way to make use of it. For instance:
18...Qg5† (18...Nf6? 19.Rxe6!+– is an important detail) 19.Kf1 Nf6 20.Qc7± The queen covers
some important kingside squares, while also threatening a possible queen trade on e5.
Meanwhile, White’s queenside pawns ensure that virtually any type of endgame will be winning

190
for him, even if he were to lose one of the kingside pawns.

12.h3
There are other playable moves, but the text is simple and good.

12...Bxf3 13.Qxf3 Qxd4 14.Rxe7 Nbd7 15.Bb3 Qxc5 16.Bxf6 Nxf6 17.Rxb7²
We have been following Mallahi – Doostkam, Iran 2000. White has an obvious advantage due
to his more active pieces, especially the rook on b7 and the bishop, which combine to exert nasty
pressure on f7. Black faces a difficult defence, with no counterplay in sight.

B) 3...Bg4

This is known as the Portuguese Gambit, whereby Black attempts to provoke f2-f3 while
offering a pawn sacrifice.

4.Bb5†!?
I find this the most attractive option, the main idea being to return the extra pawn and obtain a
lasting positional edge.

Black may insist on a gambit approach with B1) 4...c6!?, but the main line is B2) 4...Nbd7.

4...Bd7
This can be safely met by:
5.Be2 Nxd5 6.Nf3 Bf5 7.0-0
White enjoys a safe positional edge, similar to the 3...Nxd5 lines analysed earlier where
Black avoids the most challenging set-up with ...g6.
7...e6 8.c4 Nb4

191
 This position occurred in Skoulakis – Galopoulos, Kallithea 2008, when White meekly
defended the c2-square by placing the knight on a3. We actually dealt with a similar
position in the 9...Nb4 line in the note to variation C3 of the previous chapter on pages 82-
83. Here Black is missing the moves ...c6 and ...Nd7, so it is hardly surprising that White
can obtain a huge advantage with:

9.Nc3!N
It is worth noting that 9.a3? does not work here because of 9...Nc2 10.Ra2 Nxd4 when
taking on d4 with the queen would allow a queen exchange.
9...Nc2 10.Rb1 Nb4 11.Bg5! Be7
After 11...f6 12.Be3± White has finished his development, and taking the exchange would
leave Black too weak on the light squares.
12.Bxe7 Qxe7 13.Rc1±

B1) 4...c6!?

192
 This interesting side variation is described as “the Elbow” in Smerdon’s book. This is certainly
in the spirit of Black’s previous move, as he insists on sacrificing a pawn in pursuit of the
initiative. White has not done anything particularly ugly (such as f2-f3 and g2-g4) in order to
stay a pawn ahead – but even so, we still need to be quite accurate to prove an advantage.

5.dxc6 Nxc6
Inserting 5...Qa5† 6.Nc3 before 6...Nxc6 doesn’t help Black: 7.Nge2 0-0-0 8.Bxc6 bxc6

9.h3! (9.0-0 e5 10.Qd3 exd4 11.Nxd4 Qb6 is getting slightly messy) 9...Bh5 (9...Bf5 10.0-0 c5
11.d5±) Now in Cernousek – Mroziak, Czech Republic 2019, White could have obtained a clear
advantage with 10.0-0!N e5 11.Qd3± because the possibility of Qf5† removes any danger of the
pin along the d-file.

6.Nf3 Qb6

193
 Creating the option of ...0-0-0 next.

7.Bxc6†
In the event of 7.Nc3 0-0-0 8.Bxc6 (8.Be3 e5!÷) 8...Qxc6 9.0-0 Bxf3 10.Qxf3 Qxf3 11.gxf3
Rxd4 12.Be3 Rb4 White does not seem to have much.

7...Qxc6 8.Be3!
I believe this is the most accurate move, trying to play Nbd2 as soon as possible. In general, I
would prefer not to allow ...Bxf3 spoiling my structure, so I am attempting to prevent that as
soon as I can. If Black goes for an immediate ...Bxf3 then we get an improved version of the
endgame, having saved a tempo on h2-h3. In general, Black of course has some compensation
due to his bishop pair, and in the long run it may enable him to convert to some opposite-
coloured-bishop endgame a pawn down, but that is hardly a great achievement, especially when
Black was aiming for exciting gambit play.

The text move was only mentioned briefly by Smerdon, as he considered it to be similar enough
to 8.0-0 which he analysed in more detail, but there are some differences that favour White, as I
will point out. I would also like to stress that it is crucial to play with purpose in these positions,
as just shuffling the pieces around would likely allow Black to utilize his bishops to get good
compensation. Thus, I have analysed it in more concrete detail than I might otherwise have done
for this type of sideline.

8...Rc8
8...0-0-0 9.Nbd2± completely blocks all of Black’s activity, Savicevic – Petrusic, Zrenjanin
2018.

8...Bxf3?! 9.Qxf3 Qxf3 10.gxf3 is excellent for White; suffice to say that Smerdon calls 8.h3
White’s best move, leading to the same endgame except that here we have the extra developing

194
move Be3. 10...Rc8 (10...Nd5 11.c4±) Now in Marie – Mevel, Avoine 2006, White should have
continued:

11.c3N 11...e6 12.Nd2± With every chance to convert the extra pawn.

8...e5!?N is interesting, but 9.h3 Bxf3 10.Qxf3 Qxf3 11.gxf3² takes all the fun out of Black’s
position.

8...e6N
This allows White to exploit his delay in castling to set up an ideal position in the centre.
9.Nbd2 Bd6
9...Nd5 10.c4±
9...b5 prevents c2-c4 but costs a valuable tempo and presents us with a weakness to target
later. 10.0-0 Bd6 11.c3 0-0 12.h3 Bh5 13.Bg5 Nd5 14.Bh4 Nf4 15.Bg3±

10.c4!

195
 It is useful to prevent any ...Nd5 ideas.
10...0-0 11.0-0²
White can easily improve his position with Rc1, Re1, Qb3 and so on, while Black must
struggle to prove his compensation, as the bishop pair alone is not enough.

9.c3 e6
9...Bxf3?! 10.Qxf3 Qxf3 11.gxf3 converts to the same endgame as the 8...Bxf3?! line
mentioned above.

10.Nbd2 Bd6 11.0-0 0-0

11...Nd5 12.c4± is great for White.

12.h3 Bh5 13.Re1 h6

13...Rfd8?! 14.Bg5± forces Black to worry about an annoying pin in addition to the missing
pawn.

In the event of 13...Nd5 14.Ne4 Nxe3 15.Rxe3² Black’s bishop pair does not offer much
compensation.

196
14.Qb3!
After 14.c4 Rfd8 it is not clear to me what White should do next.

I considered 14.g4 Bg6 15.Ne5 Bxe5 16.dxe5 Nd5 when White should be somewhat better, but I
am uncomfortable with the idea of exposing the king and the potential for Black’s light-squared
bishop to become a nuisance. The text move prepares Ne5 without the need for the weakening
g2-g4.

14...Rfd8
14...Qc7 and 14...Nd5 can both be met by 15.Ne5².

14...Nd7 covers the e5-square but it isn’t clear what Black is doing next, so White faces a
pleasant choice: 15.a4 is a sensible continuation, gaining space on the queenside, while 15.Nh4!?
is another idea, hunting the light-squared bishop.

15.Ne5 Bxe5 16.dxe5 Nd7 17.Bd4

197
17...Nc5
17...Bg6 18.a4² maintains some edge for White.

18.Bxc5 Qxc5
18...Rxd2?! 19.Bxa7± sees White bag a second pawn.

19.Ne4 Qxe5 20.Qxb7²

Black remains a pawn down and must fight for a draw.

B2) 4...Nbd7

This is Black’s best and most popular reply, but we can meet it rather effectively with a
surprising retreat.

198
5.Be2
Losing a tempo in this way is not the most ambitious choice for White, but there is a clear logic
to it. The d7-knight is somewhat misplaced, as it prevents Black recapturing on d5 with the
queen and putting pressure on the centre with ...Nc6 and ...0-0-0. Of course Black continues to
have a solid position, and many players are still willing to play this way. Still, as Smerdon
admits, this is not the type of position Black is yearning for when he offers the Portuguese
Gambit. I consider this a minor psychological victory for White, who can play for a typical space
advantage with the thematic d4-c4 pawn structure.

5...Bxe2 6.Qxe2 Nxd5 7.c4!

Smerdon focuses on 7.Nf3 but the text move is an important finesse in the more order, for
reasons explained below.

7...N5f6
7...N5b6?! misplaces the knight, and after something like 8.Nf3 g6 9.Nc3 Bg7 10.Bg5 h6
11.Bh4 Nf6 12.0-0 0-0 13.Rad1± Black has an unpleasant, passive position.

Smerdon mentions this position ever so briefly, commenting that it will transpose to one of the
lines beginning with 7.Nf3. In the event that Black follows up with ...g6, this will be true.
However, if Black goes for an ...e6 set-up, as most players do, then there is a significant
difference, as Black has not had the chance to meet c2-c4 with ...Bb4†. This in turn means that
White will not have to castle before playing c2-c4, and as we will see, delaying castling increases
our possibilities.

8.Nf3 e6
Black’s slightly unusual and risky alternative is:

199
8...g6 9.0-0 Bg7 10.Nc3
The main idea behind Black’s 8th move is to sacrifice a pawn with 10.Re1 0-0! 11.Qxe7 c5,
when he has interesting compensation. I think White is better here too if he can show some
accuracy, but I don’t see any need to go in for this.
10...0-0 11.Bf4
The bishop is ideally placed here, especially as it prevents Black’s natural set-up with ...c6
and ...Qc7.
11.Be3 c6 12.Ne5 Qc7 13.f4 is another plan. Black remains reasonably solid though, and I
am not sure if f2-f4 is so impressive. Nevertheless, White remains better, as Black lacks a
convincing plan for counterplay.

11...c5!?N
This is more active than 11...c6 12.Rfe1 Re8 as played in Pichot – Benedetti, Buenos Aires
2013. After 13.h3N² Black has no counterplay, and White can start expanding on the
queenside with b2-b4 whenever he feels like it.
12.d5 Re8
Black’s hopes are restricted to the ...e6 break – but even after playing it, his position will
not be easy, for instance:
13.Rad1 e6 14.dxe6 Rxe6 15.Qc2²

200
9.Nc3!?
9.0-0 Bd6 has been far more common. I checked a few lines here and although White has some
potential to fight for the advantage, delaying castling gives us some valuable extra options.

We will analyse two main possibilities. B21) 9...Bd6 would be the normal move if White had
already castled, while B22) 9...Bb4!? is an alternative which is specifically directed against our
chosen move order.

9...Be7 is rather passive and the simple 10.0-0!? gives White a pleasant version of the normal
lines. A good example continued: 10...0-0 11.Bf4 c6 12.a3 Re8 13.Rfe1 Nf8

14.h3 Ng6 15.Bg3 Bd6 16.Ne5 Qe7 17.h4± Zorko – A. Muzychuk, Rogaska Slatina 2009.

B21) 9...Bd6

201
10.g4!N
Black will not enjoy being rammed by the g-pawn.

I will include some brief analysis of an interesting but ultimately less strong alternative:
10.Bg5!?
This was my initial idea, keeping the options of castling on either side. However, I think
Black is okay after:
10...h6!
White is happier after the two alternatives I checked:
a) 10...c6 11.d5! 0-0 (11...h6 12.Bxf6 Qxf6 13.dxc6 bxc6 14.0-0²) 12.dxc6 bxc6 13.0-0²
b) 10...0-0 11.Ne5! Be7 (11...Qc8? 12.Nxd7 Nxd7 13.c5+–; 11...Qe8?! 12.0-0± and
Black’s queen is awkwardly placed; 11...c6 12.Nxd7 Qxd7 13.Bxf6 gxf6 14.0-0-0² leaves a
nice structure for the knight against the bishop) 12.h4!? (12.0-0-0 Nd5!) 12...c5 (12...h6
13.Rh3ƒ) 13.0-0-0ƒ This can be analysed further, but White has clearly put the possibility of
0-0-0 to good use.
11.Bh4
11.Bxf6 Qxf6 12.c5 Be7 13.Qb5 Rb8 14.0-0 c6 15.Qa5 a6=
11...0-0

202
 I was unable to find anything truly convincing here. For instance:
12.Ne5!?N
12.g4!? seems interesting but leads to crazy complications, and it is hard to say if White’s
initiative is strong enough.
12...Be7! 13.0-0-0 Nxe5 14.dxe5
14.Bxf6 Ng6=
14...Nd7 15.Bg3
15.Bxe7 Qxe7÷
15...Qe8 16.f4 Nb6 17.Bf2 Qc6 18.b3 Rfd8÷

10...c6
It seems wise for Black to avoid committing his king for now.

10...Nxg4 gives White some edge after: 11.c5! (11.Rg1 Nxh2 is messy) 11...Be7 12.Rg1 Nh6

203
13.Rxg7 (13.Bxh6 gxh6 14.0-0-0 c6 15.Ne5 Qc7 16.f4© is also interesting) 13...Nf5 14.Rg1 c6
15.Bf4²

10...0-0
This would be risky to say the least; although White needs to play accurately to prove
something. My analysis continues:
11.g5 Nh5
I also considered 11...Ne8 12.c5 Be7 13.Be3 c6 14.0-0-0 b6 15.h4 Nc7 16.Kb1 Nd5
17.Nxd5 exd5 18.Qc2 when White keeps the upper hand.
12.c5!
Avoiding ...Nf4 ideas, as well as the possibility of counterplay with ...c5.
12.Be3 can be met by 12...Nf4! when ...Ng2† ideas are annoying. White can still try for an
advantage with 13.Qf1, but the queen is not ideally placed and it seems best to stop the
knight from coming to f4 in the first place.
12...Bf4
12...Be7 13.Be3 gives White a comfortable position; he can play 0-0-0 next, and then start
playing on the kingside. The knight on h5 only blocks White’s plans temporarily; there will
always be ways around it, like Ne4-g3. Meanwhile, Black is unlikely to create much on the
queenside.

204
13.Be3!
Black’s minor pieces on the kingside are somewhat stuck, while White is free to castle
before choosing a plan of attack. For instance:
13...c6 14.0-0-0 b6
If Black plays ...f6 any time, White will switch to attacking along the g-file.
15.Ne4 Qc7 16.Ng3±
The kingside blockade did not last for long.

11.Bd2!?
By delaying g4-g5 in this way, we avoid the possibility of the knight going via h5-f4 to g2 with
check.

Compare the alternative: 11.g5 Nh5 12.Be3 (12.Bd2 Nf4 13.Qe4 f5! is becoming rather messy)
12...Nf4 13.Qf1 White should cover the g2-square, but 13...Qe7! 14.0-0-0 0-0-0 sees Black’s

205
king escape to safety. After 15.Ne2 Nxe2† 16.Qxe2 White may still be slightly better, but we
could have aimed for more.

11...Nxg4
This is the only challenging continuation.

11...Qc7?! 12.g5 Nh5 13.0-0-0 0-0-0 (13...Nf4 14.Qe4±) 14.Kb1± gives White a much easier
life. There are many ways to continue, but one simple plan could be Ne4 with c4-c5 to follow.
The position resembles certain variations of the Classical Caro-Kann, where Black would
typically have his knight on f6, with the white pawn back on g2. The present version is superior
for White, as we have free rein with ideas like Ne4 whenever we feel like it.

12.Rg1 Nh6

13.Rxg7
13.c5 Be7 14.Bxh6 gxh6 15.0-0-0 Qc7÷ is messy.

Taking back the pawn immediately seems simplest. Black will be able to complete development
and castle long, but White can aim for a long-term edge.

13...Nf5 14.Rg1 Qe7 15.0-0-0 0-0-0 16.Ne4

In general, White is not in a hurry. There are ideas like d4-d5 and c4-c5, but he can also
continue just playing slowly, and it seems more challenging for Black to find good squares for
his pieces.

16...Bc7
16...Bb4? 17.Bf4!± is unpleasant for Black.

206
16...Rhg8 17.Nxd6† Qxd6 18.Rge1² also keeps an edge for White.

17.Kb1 Rhg8 18.Rge1!?

There is no need to simplify things, and Black is unable to make much use of the g-file.

18...Kb8 19.a3²
White has various ideas such as Bb4 or Ng3, as well as just improving his position slowly with
Bc3. It seems hard for Black to arrange ...c5 in an effective way – and without that, he has little
counterplay.

B22) 9...Bb4!?

This is an extra option for Black, presented by White’s move order involving delayed castling.
Here Black is clearly ready to exchange his bishop for the c3-knight, which could lead to
interesting pawn structures more typically associated with other openings such as the Nimzo-
Indian. The bxc3 structure will be discussed in more detail under variation B222 below.

First of all, White must decide whether or not to allow his pawns to be doubled in the first place.
There are arguments for either approach, so both B221) 10.Bd2!? and B222) 10.0-0 deserve to be
analysed.

B221) 10.Bd2!?

This modest-looking move certainly deserves attention.

10...0-0

207
 10...Bxc3 is unnecessary, and White easily claims an edge after something like 11.Bxc3 0-0
12.0-0 c6 13.Rad1 Qc7 14.Ne5 Rfd8 15.f4².

11.0-0N
11.a3 has been played in one game but I would rather not chase the bishop yet, as it may retreat
voluntarily, as shown in our main line below.

I also considered 11.0-0-0, but quickly dismissed it due to 11...Bxc3 12.Bxc3 b5! when Black is
doing quite well.

11...c6
11...c5 12.a3 Bxc3 13.Bxc3 cxd4 14.Nxd4² gives White an easy edge.

12.Ne5!
12.a3 Bd6 13.Rad1 Qc7 14.Rfe1 a6 15.b4 Bf4= seems like nothing special.

12...Bd6!?
Other moves seem less challenging, for instance:

12...Qb6 13.a3 Be7 14.Be3 Qc7 15.f4 Rfe8 (15...a6 16.g4±) 16.g4± with easy kingside play for
White.

12...Nxe5 13.dxe5 Nd7 14.a3 Bxc3 (14...Be7 may be the lesser evil although 15.Rad1 Qc7 16.f4
Rad8 17.Be3 a6 18.Ne4² gives White lasting pressure) 15.Bxc3 Qc7

208
16.Rad1 Rfd8 17.Rd6 Nc5 18.Rfd1 b6 19.h4 Nb7 20.R1d3! White is dominating, and 20...Nxd6
21.exd6 Qd7 allows 22.Bxg7!+– with destruction on the kingside.

13.Rad1
13.f4 c5! gives Black annoying counterplay.

13...Qc7
Now 13...c5 can be conveniently met by 14.Bf4 when White maintains some initiative.

13...a6!? is not a bad move. Play may continue 14.f4 (14.c5 Bxe5 15.dxe5 Nd5÷) 14...c5 15.Be3
cxd4 16.Bxd4 Qc7 when White should be better, but I’m not sure by how much.

14.f4 a6 15.c5!? Be7 16.Rf3

An interesting battle lies ahead. White has a backward d-pawn and a potentially bad bishop,
but none of that will matter if he crashes through on the kingside. Here are a few illustrative
lines.

209
16...Rfd8!
Black’s top priority should be to deal with the kingside threats, and this is the best way.

16...b6 17.b4 b5 allows 18.f5! (18.a4 is fine but the text move is more direct and stronger)
18...exf5 19.Rxf5 a5 20.a3 Rfe8 21.Qf3± with strong pressure on the kingside.

16...Rfe8 allows White to press ahead with: 17.g4! Nf8 18.g5 Nd5 19.Ne4 b6

20.Rdf1ƒ White has a lot of firepower near the kingside. 20.Kh1!? is also interesting, preparing
against ...f6/...f5 and intending Rdf1 with similar attacking ideas: f4-f5 could come at any
moment, and White could also aim for piece play with Rh3 and perhaps Qh5 to provoke a
weakening pawn move.

17.Na4!?²

210
 Black’s last move prepared to meet 17.g4 with 17...Nxe5! 18.fxe5 Nxg4 19.Bf4 Rxd4 20.Rxd4
Bxc5 21.Rfd3 Rd8 22.Qd2 Rxd4 23.Rxd4 Qd8 24.Ne2 Qd5© when Black holds his own in the
complications.

The text move prepares Nc4, when Black’s ...b6 plan is completely blocked, and he has no other
obvious counterplay in sight. Meanwhile, White can continue manoeuvring and playing on both
flanks.

B222) 10.0-0

With this move we give Black the option of exchanging on c3 to leave us with doubled c-
pawns. In that case, the pawn on c3 strengthens White’s pawn centre and he can then put a rook
on the open b-file. Still, Black’s position is extremely solid, and after something like ...c5 and
...Qc8-c6 it is not clear how White should direct his pieces. These positions are somewhat hard
to analyse because there are hardly any games and the play is not particularly forcing. I think
White’s position should be slightly preferable, but he needs to know what he is doing and play
with a clear purpose. If he just shuffles his pieces aimlessly, then Black’s solidity and better
structure will give him much easier play once he gets fully developed. I think White’s main plan
should be to develop his bishop on a decent square and then move the f1-rook to b1, leaving the
a1-rook in place to support a4-a5. If White can provoke ...b6, he gets a clear target for a4-a5. If
Black leaves the pawn on b7, then his queen or some other piece will constantly have to remain
on defensive duty. Aside from all that, we also want to ensure that both the black knights remain
passive and deprived of good outposts. Finally, White must of course take care to maintain the
good health of his central pawns.

10...0-0
10...Bxc3 11.bxc3 0-0 12.Bf4 converts to our main line below.

211
11.Bf4!?N
This seems like the most attractive choice to me.

11.Bd2N transposes to variation B221 above.

11.Bg5 was played in Tuncer – Gucin, Antalya 2013, when 11...h6N 12.Bh4 c5 13.Ne4 cxd4
14.Rad1 Be7 15.Rxd4 Nxe4 16.Bxe7 Qxe7 17.Qxe4 Nc5= would have been fine for Black. The
text move is directed against ...c5, as shown in the next note.

11...Bxc3
Black has nothing better than making this exchange. For instance:

11...c6?! 12.Rac1± is too passive for Black.

11...c5?! invites 12.Bd6 Re8 13.Nb5±.

11...Bd6 12.Ne5 is obviously pleasant for White.

11...Re8 12.h3
This is always a useful move in such a position.
12...Bxc3
It is hard to find another useful move for Black – and if he delays this exchange for much
longer, White can consider safeguarding his pawn structure with Rac1.
12...c5?! runs into 13.Nb5 cxd4 14.Nc7±.
13.bxc3 c5
Now White follows the same plan as in the main line below.

14.Rfb1 Qc8

212
 14...b6 15.a4² gives White a long-term target on the queenside.
15.a4 Qc6 16.a5 Rac8 17.Nd2²
With a slightly more pleasant game for White.

12.bxc3

12...c5 13.Rfb1 Qc8

13...b6² can be played here or at any moment, but it will always give White a clear target for
a4-a5.

14.a4 h6
14...Re8 15.h3² transposes to the 11...Re8 line in the notes above.

14...Qc6 gives White an interesting way of transforming the position: 15.Ne5!? (there is nothing
wrong with 15.Nd2 Rfd8 16.Bg5² when White maintains an edge in the same manner as the
main line below) 15...Nxe5 16.dxe5 Nd7 17.Rd1² White’s control over the d-file and the outpost
on d6 are more relevant than the queenside pawn structure.

14...Rd8 15.Bg5!? is an extremely annoying move for Black because he can hardly contemplate
breaking the pin with ...h6 and ...g5, so he will probably have to move the d8-rook again, which
will not be fun.

213
 Another advantage of the text move is that White will want to play Nd2 at some point to
protect c4, and if he had played it too quickly then Black would have been able to respond with
...Nf8, regrouping his pieces while putting pressure on the d4-pawn. Now that the bishop is
poised to take on f6, the ...Nf8 manoeuvre is obviously less appealing. A possible continuation is
15...h6 16.Bh4 Qc7 17.a5 Rac8 18.Nd2² when White keeps the upper hand.

15.Nd2 Qc6 16.Bg3!

Aiming for something similar to the above note on 14...Rd8.

16...Rfd8 17.Bh4²
Once again we have established the annoying pin. Black remains pretty solid and a long battle
lies ahead, but White’s position remains slightly more comfortable.

214
 Conclusion

1.e4 d5 2.exd5 Nf6 is a tricky move order, not only because of the various gambits associated
with it, but also because Black might still revert to a ...Qxd5 set-up if given the chance. 3.d4 is a
good option for us, when the queen recapture will transpose to a subsequent chapter, so we
focused on two other options.

3...Nxd5 is a reasonable move but White should get a comfortable game, provided he plays with
a certain amount of energy. This becomes apparent after 4.c4 Nb6 5.Nf3 g6 6.Nc3 Bg7, when
7.c5! Nd5 8.Bc4 gives White a pleasant game in all variations, especially thanks to the natural
plan of Bg5 and Re1, developing quickly while creating an awkward attack on the e7-pawn.

3...Bg4 is the other main option, keeping some gambit ideas alive. I suggest meeting it with
4.Bb5†!?, when Black must make an important decision. 4...c6!? is an interesting gambit, but
White has good chances to consolidate after taking the pawn and playing d2-d4, then bringing
the bishop to e3 before castling, which takes a lot of the fun out of the position for Black and
leaves him with a long struggle a pawn down.

Finally we looked at the main line of 4...Nbd7, when 5.Be2 is a nice idea, emphasizing that the
knight on d7 robs Black of the attractive plan of ...Qxd5 followed by long castling. So 5...Bxe2
6.Qxe2 Nxd5 is forced, and then 7.c4! N5f6 8.Nf3 e6 9.Nc3 is an accurate move order, keeping
all options open for the white king. If 9...Bd6 then 10.g4!N is an attractive idea which, if
followed up correctly, poses plenty of problems to Black. 9...Bb4!? seems like a safer choice,
although both 10.Bd2!? and 10.0-0 offer White good chances of an opening edge, with the
choice coming down to personal preferences regarding the prospect of doubled c-pawns.

215
A) 3...Qd8 120
B) 3...Qa5 4.d4 123
B1) 4...c6 124
B2) 4...Nf6 5.Bd2 c6 6.Bc4 126
B21) 6...Bg4?! 127
B22) 6...Qb6?! 128
B23) 6...Qc7 131
B24) 6...Bf5 7.Nd5 Qd8 8.Nxf6† gxf6 133
B241) 9.Qe2!? 134
B242) 9.Qf3! 137
B2421) 9...Bxc2N 137
B2422) 9...e6 140

216
1.e4 d5 2.exd5 Qxd5
This is the traditional Scandinavian move.

3.Nc3
After this standard tempo-gainer Black has several options. In this chapter we will deal with A)
3...Qd8 and B) 3...Qa5, with coverage of the latter move spanning the next chapter as well.

3...Qd6 has taken over as the main line in recent years, and it will receive detailed coverage in
Chapters 8-10.

3...Qe5†?! and 3...Qe6†?! have been played in hundreds of games, but in both cases 4.Be2
leaves Black’s queen misplaced, and I don’t see much point in analysing them as serious moves.

A) 3...Qd8

217
 This looks rather unassuming but it is very much a playable variation. In the 3...Qa5 variations
it is not uncommon for the queen to get driven back to d8, so moving it there immediately is not
as strange as it may appear at first.

4.d4 Nf6 5.Nf3 Bg4

This is by far the most popular continuation. With the queen on d8, Black’s set-up is less active
and flexible than it would be after 3...Qa5 or 3...Qd6, so he cannot afford to play too
provocatively.

For instance, 5...Bf5 6.Ne5!± and 5...c6 6.Bc4 Bf5 7.Ne5! e6 8.g4± both give White improved
versions of more mainstream Scandinavian variants.

5...g6!? deserves to be mentioned, as it has been played by Nakamura and other high-level GMs;
usually in blitz and rapid events, but sometimes in classical games as well. White has many
reasonable options but 6.Bf4 Bg7 7.Qd2 seems like the simplest approach for us. We will
encounter similar positions in Chapter 9, when Black’s queen goes to d6 and White answers the
...g6 set-up with Bg5 and Qd2. As we will see throughout that chapter, White quite often drops
the bishop back to f4 at some point, when sometimes the queen retreats to d8, transposing to the
present move order – but there are also times when the queen does well to choose another square
such as b6. Here Black has lost that option as his queen has committed to d8 so early, so White
should be happy enough.

6.h3 Bxf3
6...Bh5?! 7.g4 Bg6 8.Ne5± gives White easy play.

7.Qxf3 c6
Despite having given away the light-squared bishop so easily, Black’s position remains solid,
and we must take care to avoid certain concrete lines in which Black can force favourable
simplifications.

8.Be3
Black’s main idea can be seen perfectly after: 8.Bf4 e6 9.0-0-0 Bb4! 10.Ne4 (10.Kb1 is met by
10...Bxc3 11.Qxc3 Nd5 12.Qg3 Nxf4 13.Qxf4 Nd7= when there is no reason why White should
be better with so many minor pieces having been exchanged) 10...Nxe4 11.Qxe4 Qd5 Leading
to an endgame where Black seems to be doing just fine, despite the engine’s initial enthusiasm
for White.

218
8...e6 9.Bd3!
It is best to avoid castling too quickly, as in that case Black’s typical ...Bb4 move would be
annoying, whereas after the text move it does him no good at all. At the same time, our last move
prepares 0-0-0 next, since the bishop on d3 enables us to meet ...Bb4 with Ne4, followed by
recapturing with the bishop.

9.0-0-0 Bb4 10.Ne4 Nxe4 11.Qxe4 Qd5 leads to the typical queen exchange which Black is
aiming for.

9.Bc4 can be met by 9...Nbd7 10.0-0-0 Nb6, and if 11.Bd3 Bb4 (11...Nbd5 12.Ne2! keeps
things interesting for White) 12.Ne4 Nxe4 13.Bxe4 f5! 14.Bd3 Qd5= Black is once again fine.

9...Qa5!?
This move is extremely rare but it seems to me like the only serious way to disrupt our plans by
making 0-0-0 harder to play. If White was allowed to settle his king comfortably on the
queenside, he would happily launch a kingside attack with easy play.
9...Bb4 10.a3 Bxc3† 11.bxc3± gives White easy play with the bishop pair, and the king will go
to the kingside of course.

9...Be7 10.0-0-0! works perfectly for White, for instance: 10...Nd5 (10...Bb4?! 11.Ne4 Nxe4
12.Bxe4 f5 13.Bd3 Qd5 can simply be refuted by 14.Qe2! Qxa2 15.c3+–)

219
11.Ne4 0-0 12.c4 Nxe3 13.fxe3 Nd7 14.g4 The opposite-coloured bishops ensure that White’s
attacking chances are great. 14...a5 15.h4 a4 16.g5 a3 17.b3 Qa5 18.Rhg1ƒ Quesada Perez –
Slipak, Buenos Aires 2015.

9...Nbd7
This has been by far the most common continuation, but the drawback is that it blocks the
queen’s route to d5. This proves significant after:
10.0-0-0! Bb4

11.Ne2!
With ...Qd5 unavailable, White seems to be doing great, as a few games show:
11...Nd5
11...Qa5 12.Kb1 0-0-0 occurred in Wyss – Berchtold, Switzerland 2016, when the direct
13.g4N± would have been strong.

220
12.Qg3 g6 13.Bh6 Bf8
Another game continued 13...b5 14.h4 Qc7 15.Qf3 N7b6 16.Nf4 Bd6 17.Nxd5 Nxd5
18.Kb1 Bf4 19.Bxf4 Qxf4 20.Qe2 Qf6 21.h5 0-0-0 22.c4± and Black’s exposed king
remained the defining feature in Vandenbussche – Fridman, Germany 2014.
14.Bxf8 Rxf8 15.c4 N5f6 16.Nc3 Qa5

17.d5! cxd5 18.cxd5 Nxd5 19.Nxd5 Qxd5 20.Kb1 Qa5 21.Rhe1©

Black’s exposed king gave him a lot of problems in Heberla – Djukic, Plovdiv 2013.

10.0-0
Ideally I would prefer to avoid committing the king to the kingside, but it seems like the most
appropriate reaction to Black’s last move. The queen would have been well placed after long
castling – but with the king on g1, Black’s last move is hardly the most useful. Moreover, it
gives us the opportunity to play a2-a3 and b2-b4 with gain of tempo. If the queen was still on d8,
Black could have easily parried this plan with ...a5.

221
10.0-0-0 would be ideal, but the point of Black’s last move was of course to meet this with
10...Bb4! 11.Bd2 Bxc3 12.Bxc3 Qxa2. This is far from the end of the story, as 13.d5! cxd5
14.Qg3 Nbd7 15.Qxg7 Rg8 16.Qh6 Rxg2 17.Qh4 (but not 17.Bxf6? Rxf2–+) gives White some
initiative, but the position remains complicated and Black has certainly succeeded in putting the
queen to good use.

I also considered 10.Bd2!? Qb6 11.0-0-0 Qxd4 when White definitely has compensation for the
pawn, but it is not clear if we have anything more than that. A possible continuation is 12.Ne4
Nbd7 13.Bc3 Qb6 14.Bxf6 Nxf6 15.Nxf6† gxf6 16.Qxf6 Rg8÷ when any result is possible.

10...Nbd7
10...Bd6? 11.Nb5!± is a nice trick.

11.a3!?
It may not seem that crucial, but placing the pawns on a3 and b4 definitely helps White. Any
...c5 plans will be prevented for the foreseeable future, and eventually White can think about b4-
b5 ideas.

11.Ne2 Bd6 12.c4 0-0 13.Ng3 c5= seems fine for Black.

11.Qg3!?N 11...g6 12.Rab1 is also possible, with similar ideas to the main line below.

11...Be7
11...Bd6 was played in Sarbay – D. Petrov, Orsha 2018. A logical continuation would have
been: 12.b4N 12...Qc7 (12...Qh5 13.Qxh5 Nxh5 14.Ne4 Bc7 15.c4² is pleasant for White)
13.Rfe1 0-0

222
14.Ne4 Nxe4 15.Qxe4 g6 (15...Nf6?! 16.Qh4 Be7 17.Bg5± puts Black in trouble, as 17...h6 is
met by 18.Bxh6!) 16.c4²

12.Ne2
There is no need to commit to 12.b4 Qc7 just yet.

12...0-0 13.c4 Rfe8

13...e5 can be met by 14.Ng3 g6 15.Rfe1² when opening the centre does not solve Black’s
problems, because the bishop pair remains strong.

14.Nc3 Rad8 15.b4N

I have already mentioned the merits of this advance, and now seems like the ideal time for it.

15.Rfd1 Qc7 16.Rac1 a5 17.g4 was also quite pleasant for White in Xiong – Iotov, Richardson
2015, although Black went on to hold a draw without much drama.

15...Qc7 16.Rfc1!? a5
16...g6 17.Rab1² and 16...e5 17.d5² favour White as well.

17.Rab1²
Black remains pretty solid, but there is no doubt that White’s bishop pair and space advantage
give him the upper hand.

B) 3...Qa5

Even though 3...Qd6 has become trendier in recent years, the text move remains the overall

223
leader in total games played, by a wide margin.

4.d4
Black has tried virtually every legal move from this position, but I think B1) 4...c6 is the only
serious alternative to the usual B2) 4...Nf6.

It is obviously risky for Black to try:

4...e5?! 5.Nf3 exd4
5...e4 6.Ne5 Nd7 7.Nc4 leaves the e4-pawn vulnerable, and after something like 7...Qf5
8.Ne3 Qa5 9.Bd2 White is already winning.
5...Bg4 6.Bc4! gives White an overwhelming initiative, for instance: 6...Nf6 (6...Nc6 loses
to 7.Bxf7† Kxf7 8.Ng5†+–) 7.dxe5 Bxf3 8.Qxf3 Qxe5† 9.Be3 Nbd7 10.0-0-0+–
6.Qxd4
You don’t really have to remember any lines from here, as it is obvious that White is doing
well – but still, it’s fun to see how we may capitalize on our lead in development.
6...Nc6
6...Be6 7.Bb5† Nd7 8.Bg5 c6 9.Bc4 Ngf6 10.Bxe6 fxe6 11.0-0-0± Perini – Cosentino, corr.
2012.
6...Nf6 7.Bb5† c6 8.Bc4 Be7 9.0-0 0-0 10.Re1 Qd8 (10...Bd8 11.Bf4 Bb6 12.Qd2±)
11.Qh4 Bf5 12.Bg5± and White keeps an obvious advantage in activity.
7.Bb5 Nge7
In Ljubicic – Popovic, Zenica 2007, the most accurate continuation would have been:

224
8.Qf4!N 8...Bd7 9.Rb1! Qb4
9...a6? 10.b4+– catches the queen.
10.Qxc7±
With a safe extra pawn.

B1) 4...c6

Black often plays this move at some point, and it would be easy to transpose to a later variation
by developing the knight to f6 within the next few moves. To give this move order any real
significance, I will consider a set-up where Black avoids ...Nf6 with a view to developing the
knight to g7 instead.

5.Bd2 Bf5 6.Bc4 e6 7.Qe2 Bb4

7...Nf6?! 8.Nd5 gives White an improved version of the more popular lines examined later in

225
the chapter. 8...Qd8 9.Nxf6†

9...Qxf6 (9...gxf6 10.0-0-0 looks great for White, who will go on the attack with g2-g4, f2-f4 and
so on. Usually, White is unable to able to pull off Qe2 and 0-0-0 in such a position because the
c2-pawn is hanging, as you will see in the later variation B241.) 10.0-0-0 White has excellent
prospects; one nice example continued 10...Bg6 11.Nf3 Nd7 12.Bg5 Qf5 13.d5! with a decisive
initiative in Hamdouchi – Boe Olsen, Plovdiv 2010.

8.g4! Bg6
Black grabbed a pawn with 8...Bxc2 in Koch – H. Luther, Nuremberg 2004, when 9.a3!N
would have been strong, for instance: 9...Bg6 (9...Bxc3 10.Bxc3 Qa4 11.d5±) 10.f4 Bauer covers
this variation in a note and marks the text move as interesting/unclear, but I think Black has
serious difficulties, for instance: 10...Ne7 (10...Be7 11.Nf3±) 11.Nf3 Nd7 12.0-0 Bxc3 13.Bxc3
Qd8

226
14.f5 exf5 15.gxf5 Bh5 16.Rae1±

9.f4 Ne7 10.0-0-0 Nd7 11.a3 Bxc3

11...Bxa3 is a reasonable try but White has more than one route to a clear advantage: 12.Na2±
is a simple option, while 12.h4!? h5 13.f5 exf5 14.Nd5 Qd8 15.Nf4± is flashier.

12.Bxc3 Qc7 13.Nh3

13.f5!? also looks dangerous for Black but we don’t need to hurry.

13...Nb6 14.d5 exd5

So far, White had done everything right in Hector – Nakamura, Malmo/Copenhagen 2005. My
improvement is:

227
15.Ba2!N
This seems simplest. Avoiding Bxg7 for now makes the threat of f4-f5 even stronger.

15...f6 16.f5 Bf7 17.Rde1

Black remains a pawn up for now, but that is the end of the good news for him, as his king is
stuck in the centre and g4-g5 is threatened. The game goes on of course, but White should be
able to make his initiative count with precise play.

B2) 4...Nf6

This can be a confusing variation to study, due to the multitude of move orders available to
both sides. In most games, White seems to play Nf3, Bc4, and Bd2 in some order, while Black is
likely to play some combination of ...Bf5 (or ...Bg4), ...c6 and ...e6. I recommend:

5.Bd2
This is not the most popular choice, but I believe there are some distinct advantages to be
gained from delaying Nf3, as will be pointed out later.

5...c6
This is the most popular continuation, although it gives us some opportunities to exploit the
advantages of leaving the knight on g1.

5...Bg4 is the most challenging continuation and we will analyse it separately in the next chapter.
Other moves are of little concern to us, for instance:

5...Bf5 6.Bc4 c6 leads to variation B24 below.

228
5...Nc6? runs into 6.Bb5! when the threat of Nd5 forces the awkward 6...Qb4.

7.d5 a6 8.Ba4 b5 occurred in Mauko – Theissl Pokorna, Szombathely 1993, when the simple
9.Nxb5N 9...Qe4† 10.Qe2± would have led to material gains for White.

5...Qb6?!
This doesn’t make much sense since White was not yet threatening anything, and the b2-
pawn cannot be touched anyway.
6.Nf3 Bg4
6...Qxb2?? loses in more than one way, 7.Rb1 Qa3 8.Nb5 being the simplest.
6...c6 7.Bc4 and 6...Bf5 7.Bc4 e6 8.Qe2 c6 both transpose to variation B22 below.
7.h3

7...Bxf3
7...Bh5 8.g4 Bg6 9.Bc4 also gives White a massive lead in development. 9...e6 10.Qe2 c6

229
 11.0-0-0 Nbd7 12.h4 h5 This position was reached in J. Howell – A. Ledger, Norwich
1994, and now the most convincing winning line would have been: 13.Bxe6!N 13...fxe6
14.Qxe6† Kd8 15.Ne5 Be8 16.g5 Nd5 (16...Ng4 17.Rhe1+–) 17.Nxd5 cxd5 18.Qxd5+–
8.Qxf3 e6
8...Nc6 9.d5! Nd4 (9...Ne5 10.Qg3±) 10.Qd3 e5 11.Na4! Qd6 12.c3 b5 13.cxd4 bxa4
14.Qb5† Nd7 15.Qxa4 was clearly better for White in Debray – Munoz Pantoja, Barcelona
2017.
9.0-0-0 c6 10.g4 Nd5
Another game continued 10...h6 11.h4 Nbd7 12.Bh3 Qxd4 13.g5 hxg5 14.hxg5 Nd5 and
now 15.Bxe6!N+– would have been crushing. (15.Be3? Ne5! 16.Qg2 Nxe3µ led to an
unfortunate reversal in Milliet – Gunina, Astana 2013.)

11.Ne4
White enjoys the bishop pair plus a big advantage in development. A good example
continued:
11...Nd7 12.g5 Qxd4 13.c4 Ne5 14.Qe2 Ba3 15.Ba5+–
With a decisive material gain in Firat – Ilic, Skopje 2016.

6.Bc4
We continue to leave the knight on g1 for the time being, instead developing the bishop and
introducing the possibility of Nd5.

We will analyse four main possibilities: B21) 6...Bg4?!, B22) 6...Qb6?!, B23) 6...Qc7 and B24)
6...Bf5.

6...Qd8 7.Nf3 gives White a big lead in development and an easy game, especially since
7...Bg4?? runs into 8.Ne5 (or 8.Bxf7†).

230
 B21) 6...Bg4?!

If Black wanted to put the bishop here, it would have been better to do so on the previous turn, as
discussed in the next chapter. Here it is too provocative and White swiftly builds a powerful
initiative.

7.f3 Bf5
7...Bh5 8.g4 Bg6 transposes to the main line below.

7...Bd7 may be the lesser evil although it is rather passive. A good example continued: 8.Qe2 e6
9.Ne4 Qd8 10.0-0-0 b5 11.Bd3 Na6 12.f4 (12.c3 is a good alternative but there is no particular
reason to fear Black’s next move in the game)

12...Nb4 13.Bxb4 Bxb4 14.c3 Be7 15.Nxf6† Bxf6 16.Nf3 Qa5 17.Kb1 b4 18.c4 White
remained in control in Fedorov – Mamedyarov, Batumi 2002. Black’s position is cramped and
his light-squared bishop is a sorry sight.

231
8.g4 Bg6
8...Bd7 is hardly an improvement on the previous note. We will follow another example in
which White soon built up a dominating position: 9.Qe2 Qc7 10.0-0-0 a5 11.a4 Na6 12.Nh3
Nb4 13.Nf4 e6

14.g5 Nfd5 15.Bxd5 cxd5 16.Nfxd5 Black was a pawn down in addition to suffering from an
unsafe king in Floegel – David, Germany 1996.

9.f4 e6 10.Qe2 Be7 11.d5!

This thematic move smashes open the centre, giving White a huge initiative at no material cost.

11...cxd5 12.Nxd5 Qd8 13.Nxf6† gxf6 14.0-0-0 Qc7

This position has arisen in a couple of games, and both times White missed the crushing
continuation:

232
15.f5!N exf5 16.g5! fxg5
16...0-0 17.h4! leads to a devastating attack.

17.Bxg5 Nc6 18.h4 f4 19.Nf3+–

The game goes on, but Black will not be able to recover from the shattering of his kingside if
White plays accurately.

B22) 6...Qb6?!

This has been used by some strong players but I find it rather suspicious, as it is hard to
imagine that Black can get away with grabbing either of the b2- or d4-pawns while so far behind
in development.

233
7.Nf3 Bf5
7...Bg4?? is obviously no good due to: 8.Bxf7†! Kxf7 9.Ne5†+–

Of course we should also check the greedy approach:

7...Qxb2?! 8.Rb1 Qa3 9.0-0 e6

10.Ne5 Be7 11.Re1 Qd6!?N

11...0-0 is strongly met by 12.Ne4! with two main possibilities:
a) 12...Nxe4 13.Rxe4 soon led to a rout after: 13...Qd6 14.Qh5 b5 15.Rb3 f6 16.Rg4! fxe5
17.Rxg7† Kxg7 18.Rg3† Kh8 19.Bd3 1–0 Degraeve – Tomczak, Cappelle-la-Grande 2006.
b) 12...Nbd7 occurred in Nouro – Raaste, Helsinki 1996, when 13.Rb3!N 13...Qa4 14.Nc3!
Qa5 15.Nd5 Qd8 16.Nxe7† Qxe7 17.Rh3+– would have been crushing.
The text move seems like the best chance to improve Black’s play, but he is still in trouble
after:

234
12.Bf4!
12.Ne4 Qxd4! and I was unable to find anything conclusive.
12.Nxf7!? Kxf7 13.Ne4 Nxe4 14.Rxe4 is certainly dangerous, but after 14...b5 Black
seems to be able to survive; I will omit my further analysis as the recommended move is
both stronger and easier to understand.
12...Qd8 13.Ne4 Nxe4 14.Rxe4 0-0

15.Rb3!
A nice way of bringing the final piece into the attack.
15...b5
15...Nd7 16.Nxd7 Qxd7 17.Qh5+– leaves Black facing too much firepower on the
kingside.
16.Qh5!
The attack decides.

235
8.Qe2 e6 9.0-0-0 Nbd7
9...Bb4 10.Ne5 Nbd7 11.Nxd7 Nxd7 (11...Kxd7 is a better try although after 12.Kb1!?N²
Black cannot be entirely comfortable with his king in such a position) 12.g4! Bg6 Now in
Rabadan Velasco – Menor Villamor, Galapagar 2005, White missed the strongest continuation:

13.h4!N 13...h5 14.Bxe6±

10.Ne5!?
Many other moves have been tried but I like this active choice.

10...Qxd4
Taking the pawn must be critical, although White gets more than enough compensation for it.

10...Be7 is safer, when 11.Nxd7 Kxd7 (11...Nxd7? 12.g4 Bg6 13.f4 is horrible for Black)
reaches a funny-looking position.

236
 Something similar occurred in the note to Black’s 9th move above, where it was noted that
taking with the king on d7 was the lesser evil for him. You should always keep this possibility in
mind in such situations, as it would be easy to ‘forget’ that putting the king on d7 was a serious
possibility. But with the d5-square securely defended, White does not have any immediate way
of smashing through in the centre. So how should we continue? Many moves are possible but
somehow I like the patient and flexible 12.f3!?N², simply taking control of some light squares
while keeping various options open. White can consider expanding on the kingside with g2-g4,
followed by advancing either the f- or h-pawn. Alternatively, he can just focus on the centre with
Rhe1. His knight may go to a4 or perhaps e4 at any moment, and then on to c5. Meanwhile Black
has to keep an eye on those ideas while coming up with a suitable way to safeguard his king, as it
can hardly stay on d7 indefinitely.

11.Nxd7 Nxd7
11...Qxd7? 12.Bf4 Qc8 13.Nb5! would be crushing.

237
12.Be3 Qh4
This was reached in Panchenko – Kosanski, Prague 1982. White has more than one promising
continuation. Most logical is:

13.Rd4!N 13...Qe7 14.Rhd1

All White’s pieces are in play while Black’s king remains stuck in the centre, so White
certainly has more than a pawn’s worth of compensation.

B23) 6...Qc7

Compared to the previous variation, this seems like an attempt to play more solidly, rather than
paying any attention to the loose pawns on b2 and d4. White still has every chance of developing
a meaningful initiative though.

238
7.Nf3 Bg4
I considered a couple of other options:
7...Bf5 has been played in quite a few games but White gets an easy initiative with the typical
plan: 8.Ne5 e6

9.g4! Bg6 10.Qe2 (10.h4 Nbd7 11.Qe2 Nxe5 12.dxe5 Nd5 is not so clear) 10...Nbd7 11.f4!?N
With good prospects on the kingside and in the centre.

7...e6!?
This may look rather passive but it actually makes a good deal of sense, as developing the
bishop actively on f5 only makes it into a target for White’s kingside pawns, as we saw in
the note above. So instead, Black keeps a compact position and intends to deploy the bishop
on b7, where it will not be so exposed.
8.Qe2 Nbd7

239
9.0-0!
Short castling makes the most sense here, for a few reasons. Firstly, there is no real need to
advance our kingside pawns without the bishop on f5. And secondly, the more common 9.0-
0-0?! invites 9...b5!N when the piece sac on e6 does not seem to get anywhere, so White
must settle for 10.Bd3 b4 11.Ne4 Nxe4 12.Bxe4 a5 (or 12...Bb7) when the outcome is still
not clear, but I don’t see any point in giving Black a clear target on the queenside.
9...Be7 10.Rfe1 b5!?N
It seems logical to gain both time against the bishop and space on the queenside.
10...0-0? is inadvisable due to 11.Bxe6!.
10...Nb6 11.Bd3 Nbd5 was played in Rutkowski – Escher, Frankfurt 2004, when
12.Ne4N² would have maintained a pleasant edge for White.
11.Bd3 Bb7 12.Ne4 0-0 13.b4!?
It is worth taking the opportunity to block up the queenside and bury the bishop on b7.
13...Nxe4 14.Bxe4 Nf6 15.Bd3 Nd5 16.c3
White enjoys a lasting advantage.

240
8.h3 Bxf3 9.Qxf3 e6 10.0-0-0 Nbd7 11.g4
This scenario is pretty typical for these positions: having developed all the pieces and sheltered
his king on the queenside, White can advance his kingside pawns with no real worries, especially
since any opening of the position is likely to favour his bishop pair.

11...Bb4!?
This is arguably Black’s best chance for counterplay. Black’s remaining bishop is not doing
much, so he intends to swap it off in order to establish his knights on the light squares and play
on the queenside.

11...Nb6 has been played more frequently but White just seems to have a clearly more
comfortable position after 12.Bd3!?. (I prefer this to the more common 12.Bb3, when 12...a5
13.a3 a4 14.Ba2 Nbd5 gives Black some ideas on the queenside.)

12.Kb1 Bxc3 13.Bxc3 Nd5 14.Bd2 b5 15.Bd3 N7b6

Black proceeds with his plan, but White can more or less ignore it with:

241
16.h4!
We are following a game of Yu Yangyi, in which he improves on his own play!

16.g5 Nc4 17.Bxc4 bxc4 18.Qe2 c3 19.Bxc3 Nxc3† 20.bxc3 Qa5 gave Black a lot more
counterplay than was necessary in Yu Yangyi – Megaranto, Subic Bay 2009.

16...Nc4 17.Bc1 0-0-0

Black was unable to conjure up any serious threats on the queenside, so he takes the
opportunity to shelter his king before the kingside opens up. Still, even without the black king as
a target, White is doing great.

18.g5 Rd7 19.h5 c5 20.Be4±

White was in full control in Yu Yangyi – Yu Ruiyuan, Qinhuangdao 2011 – and with the
position starting to open up, the superiority of the bishops over the knights becomes more

242
evident.

B24) 6...Bf5

This is Black’s most common continuation, which we will meet with a thematic knight-
exchanging manoeuvre:

7.Nd5 Qd8 8.Nxf6† gxf6

Black usually recaptures this way, for good reason.

8...exf6?!
This simply gives us a pleasant version of an ...exf6 Caro-Kann.
9.c3! Be7
9...Bd6 has been the most popular choice but 10.Qe2† Qe7 11.0-0-0² leads to an endgame
where White’s superior pawn structure clearly makes him the favourite.
9...Nd7 hardly seems the most accurate move order in view of: 10.Qe2†!N 10...Be7
(10...Qe7 11.0-0-0 is similar to the note above) 11.g4 Bg6 12.f4 f5 13.g5 Qc7 14.0-0-0ƒ

243
10.Ne2!N
10.Qb3 has been played a few times, and objectively White should be somewhat better after
gobbling the b7- and c6-pawns. However, I see no reason to play materialistically and risk
giving Black a meaningful initiative when we can instead gain a clear advantage by
positional means.
10...0-0 11.Ng3 Bg6 12.h4 h6 13.Qg4 Kh8 14.h5 Bh7 15.Nf5 Bxf5 16.Qxf5±
Black will have to trade queens to avoid a disaster on the kingside, so White will be left with a
dominating position, with two bishops and the better pawn structure.

This position has arisen in a few hundred games, which sounds like a decent number – but if
we added the moves Nf3 and ...e6, the game count would multiply many times over. The version
with those extra moves included has been tested and analysed extensively, and even though
White’s position seems slightly more pleasant in general, I always struggled to grasp exactly
what should be trying to achieve in the main lines of that variant. This led me to explore the

244
sacrificial idea of 10.Qe2!? (in the position with Nf3 and ...e6 included), preparing 0-0-0 without
spending a tempo on Bb3 or c2-c3. Black replies by capturing the c2-pawn, and I was able to
capitalize on White’s lead in development to score a number of victories. Nevertheless, the fact
remains that Black should be doing fine with accurate play.

When it became clear that my formerly favourite continuation could be neutralized, the natural
next step was to explore White’s options in the related position without Nf3 and ...e6 included. I
started by analysing B241) 9.Qe2!?, a rare move which leads to some interesting possibilities. I
found that Black should be okay after accurate play, but I included my analysis anyway to
provide some background information. Some readers may also wish to use it as a secondary
surprise weapon. I then investigated B242) 9.Qf3! which I present as my main recommendation.
Once again, we do not mind giving up the c2-pawn for the sake of quick development and
activity, but the queen has certain advantages on f3 compared to e2, as we will see.

B241) 9.Qe2!? Bxc2

The less challenging 9...e6?! leads to a position which was discussed briefly on page 124 – see
the 7...Nf6?! line in the notes to variation B1.

9...Qxd4 10.0-0-0 Qg4 (10...Qe4 11.Be3 Qxg2 12.Nf3 Bg4 13.Qd2 is dangerous for Black)
11.f3 Qg6 12.Bc3© could be analysed in more detail, but White has at least enough play for the
pawn so I will concentrate on the more challenging continuation.

10.Nh3!?
This is an interesting way to exploit the fact that we have not yet committed the knight to f3. I
only found one game from this position, so almost all of the following analysis is original.

245
10...Nd7!N
This is Black’s most useful move, catching up on development and threatening ...Nb6.
In the game Black got into trouble after: 10...b5 11.Rc1! Bf5 12.Bd3 Qd5 13.Nf4 Bxd3 14.Qxd3
Qd6 15.0-0 Bh6 16.Qh3 Bg5 17.Qc8† Qd8 18.Qb7 Nd7 19.Qxc6± Gharamian – Rocha, Lisbon
2017.

10...Bf5?! also looks risky due to: 11.g4! Bg6 (11...Bc8 12.Nf4 leaves Black horribly
undeveloped, and that alone gives White more than enough compensation) 12.f4 f5

13.Ng5! Qd7 (this is the only try, as 13...h6 14.gxf5+– and 13...fxg4 14.f5+– are terrible for
Black) 14.0-0-0 fxg4 15.f5! Qxf5

16.Bxf7†! Bxf7 17.Rhf1 Qd5 18.Nxf7 Rg8 19.Rf5!+–

10...Qxd4!?N
This is riskier than our main line, but still a playable move which leads to interesting

246
 complications.
11.Be3 Qe4 12.Rc1 Qxg2
12...Ba4? leaves Black horribly uncoordinated, and 13.0-0 Rg8 14.Nf4 Bh6 15.Qh5! gives
White a deadly initiative.
12...b5 13.Rxc2 bxc4 14.Nf4 Bh6 15.Rxc4 Qb1† 16.Rc1 Qe4 (16...Qxa2?! 17.0-0±) 17.0-0
Bxf4 18.Rc4 Bxh2† 19.Kxh2© gives White a dangerous lead in development while it is not
clear what Black should do with his king.

13.Qxc2! Nd7!
13...Qxh3? 14.Qb3! leaves Black defenceless.
13...Qxh1†?! 14.Ke2 Qg2 15.Qb3! also turns out great for White, for instance: 15...Qg4†
(15...e6 16.Qxb7+–) 16.Ke1 e6 (16...Qxh3 17.Bxf7† Kd8 18.Be6+–) 17.Qxb7 Bd6
18.Rd1! (18.Qxa8 0-0ƒ) 18...0-0 19.Rxd6±
14.Qf5!
Keeping the initiative alive.
14.Qb3 Ne5 15.Bxf7† Nxf7 16.Qxb7 Rd8 17.Qxc6† Qxc6 18.Rxc6 Bh6 results in an equal
endgame.
14...e6
14...Qxh1†? 15.Ke2 Qg2 16.Rg1 Qxg1 17.Nxg1+–
15.Bxe6 fxe6 16.Qxe6† Be7 17.Rg1 Qd5 18.Qxd5 cxd5 19.Rc7
Black is not far away from equalizing, but he is under a certain amount of pressure and still
needs to figure out where to put his king. Meanwhile, White is only one pawn down, and he will
easily regain it while keeping some chances to press in the endgame.

11.Rc1
The careless 11.0-0?! Nb6 would lead to the exchange of our light-squared bishop.

11...Bf5!

247
 11...Bg6 12.Nf4 Nb6 13.d5! would make things trickier for Black.

12.Nf4
12.g4 Bg6 13.f4 looks tempting, but 13...f5 14.Ng5 e6 15.d5 cxd5 16.Bxd5 Bd6÷ seems
acceptable for Black.

12...Nb6 13.d5!?
13.Bxf7†!? looks tempting, but after 13...Kxf7 14.Qh5† Bg6 15.Nxg6 hxg6 16.Qxh8 Qxd4©
the tables have turned: the position is still dynamically balanced, but Black now has the initiative
in return for a small amount of material, which is not really the outcome we are looking for.

I also checked 13.Bb3 Qxd4! 14.0-0 e6 when White seems to have a dangerous lead in
development for the two sacrificed pawns, but it is hard to prove anything concretely.

13...Nxd5
This seems like the most obvious move so I will present it as the main line.

13...Nxc4 14.Rxc4 Bh6! seems like a safe route to equality, for instance:

248
15.dxc6 bxc6 16.0-0 Bxf4 17.Rxf4 Bd3 18.Qf3 Bxf1 19.Qxc6† Kf8 20.Rd4 h5 21.Rxd8† Rxd8=

14.Bxd5 cxd5 15.0-0

15...Bh6
15...e6? is too slow, and 16.Nh5! a6 (16...Bg6 17.Qb5†+–) 17.Bc3 Be7 18.Ng7†+– duly
punishes Black.

16.Rfd1 Rg8÷
White certainly has decent compensation for the two-pawn deficit, but it doesn’t seem like
enough to be playing for an advantage, especially considering the simplifications that are likely
to occur if White tries to win back material.

249
 B242) 9.Qf3!

Having seen why the lines with the queen on e2 did not quite work the way I would have
wanted, we now come to my preferred queen move. In general, we can already get a sense that
the queen is more active on f3, especially when it comes to supporting a future d4-d5 break. On
the other hand, if Black declines to capture on c2, the queen is actually slightly worse on f3 than
e2, as it gets in the way of the thematic pawn storm with g2-g4 and f2-f4. Still, being able to go
0-0-0 without having spent a tempo safeguarding the c2-pawn is an achievement in itself;
moreover, despite blocking the f-pawn, the queen on f3 is actually surprisingly well placed to
support an alternative plan involving g2-g4 and h2-h4 combined in conjunction with the knight
coming to f4.
Black now should choose between two serious options: B2421) 9...Bxc2N and B2422) 9...e6.

9...Qxd4?? is out of the question here because 10.Qxf5 Qxc4 11.Qc8 is mate!

B2421) 9...Bxc2N 10.Rc1

This move gives Black no choice but to retreat the bishop to g6, whereas 10.Ne2 gives Black
additional options such as 10...e6 when the bishop could go to f5. This position is reached in the
note to Black’s 10th move in variation B2422 below, where we see that White is doing
excellently, so Black’s extra option is of questionable value. Therefore 10.Ne2 seems like a valid
move order, but after 10...Nd7 we should bring the rook to c1 anyway, and it feels more natural
to just play it immediately and force the play.

10...Bg6 11.Ne2 Nd7

After 11...e5? 12.0-0 Black is obviously not helped by the prospect of the centre opening up.

250
12.h4 h5
12...h6? 13.Nf4 Nb6 14.Nxg6 fxg6 15.Be6± leaves Black too weak on the light squares.

12...Nb6 leads to a fascinating forced line: 13.h5!? (13.Bb3 is also possible of course, when
White maintains good compensation) 13...Nxc4 14.Rxc4 Qd5 (14...Bb1? 15.Rc1 Bxa2 16.b3
Qb6 17.Rh3+–)

15.hxg6! Qxc4 16.Rxh7 Bg7 17.Rxh8† Bxh8 18.Qh3 Bg7 19.Qh7 Bf8 20.g7 Bxg7 21.Qxg7²
White’s bishop and knight are slightly more valuable than Black’s rook and pawn.

13.0-0
13.Nf4 Bh6! 14.0-0 Bxf4÷ is far from clear, but eliminating the dangerous knight at least
helps Black to keep his light squares defended.

251
13...Nb6
This seems like Black’s most logical move. I checked several other possibilities:

13...Bg7? is strongly met by 14.Nf4± of course.

13...e5 looks suspicious: 14.d5ƒ is one good reply, with a lasting initiative.

13...e6 is simply met by 14.Nf4 Bf5 15.Nxh5, when the open h-file doesn’t give Black much
counterplay, because White will easily solidify his kingside with g2-g3, and Black has little
chance of shifting his heavy pieces into attacking positions when his own king remains
vulnerable. Play may continue 15...Qc7 (15...Bg6?! 16.Nf4 Bf5 17.Rfe1±) 16.b4! when it is
obvious that Black’s king will find no peace on the queenside.

Finally, 13...Qb6 might not suffer from a direct refutation but Black’s situation remains perilous,
and White has several tempting possibilities. We should not be afraid of losing the pawns on b2
and/or d4, as taking them would cost Black more time while opening additional lines for White’s
better-developed pieces. 14.Bc3 is one good option – the idea being not to defend the
aforementioned pawns, but rather to add force to a subsequent d4-d5. 14.Rc3 is another
promising idea, mobilizing the rook along the third rank. Finally, the simple Rfe1 is a good
move, bringing the last piece into play and keeping various options open.

14.Nf4!?
14.Bb3© should offer White decent long-term compensation, but I decided to focus on the
more forcing continuation.

14...Nxc4
14...Bg7? 15.Nxg6 fxg6 16.Qg3± makes White’s task much easier.

252
14...Bh6 15.Rfe1 Bxf4 16.Qxf4 Nxc4 17.Rxc4 transposes to our main line below.

14...Qxd4!? is not the worst move but White keeps the upper hand after: 15.Be3! (15.Nxg6 fxg6
16.Be6 Qxd2 17.Rfd1 Qxb2 18.Rxc6 leads to a strange mess) 15...Be4 16.Bxd4 Bxf3 17.gxf3
Nxc4 18.Rxc4 e5 19.Be3 exf4

20.Re4†! Kd7 21.Rd1† Kc7 22.Bxf4† Kb6 23.Rd7± White should be able to pick up both of the
remaining f-pawns, with good winning chances in the ensuing endgame.

15.Rxc4

15...Bh6
Black’s best bet is to eliminate the troublesome knight.

15...Bg7 fails to solve Black’s problems after: 16.d5! 0-0 17.Bc3 cxd5 18.Nxd5 Kh8 19.Bb4±

253
16.Re1 Bxf4
Black should not delay this move.

16...Qd7? is refuted by: 17.Nxg6 fxg6 18.Bxh6 Rxh6 19.Qxf6+–

16...e6?! 17.Bb4± is also no fun for Black.

17.Qxf4
The disappearance of the nasty knight takes some of the pressure off Black’s position.
However, the opposite-coloured bishops help to fuel White’s initiative during the middlegame.

17...0-0
17...Qd7? runs into the unpleasant 18.Bb4±.

17...Qd5? 18.Qxf6 0-0-0 19.Rxe7 Qd6 (even worse is 19...Qxc4? 20.Qe5 Rd7 21.Qxh8†+–)
20.Qxd6 Rxd6 21.Rb4± reaches a miserable endgame for Black: he is already one pawn down
and is unable to avoid a further loss on the queenside.

18.Rc3!
White’s queen will be perfectly placed on h6, and the rook will go to g3. Meanwhile, Black has
trouble getting his rooks into the game at all. A few possible lines are:

18...Re8
18...Qd6?! 19.Qh6 leaves Black defenceless.

19.Qh6! Qd7 20.Rg3 Qf5 21.Bb4!? e5 22.Ree3

254
22...exd4
22...e4 allows an attractive breakthrough after: 23.Bc3 Rad8 24.d5! cxd5 (24...Rxd5
25.Rgf3+–)

25.Rg5!+– The defence collapses.

22...Rad8 is a better defensive try although Black remains in serious trouble after: 23.Bc3 exd4
24.Bxd4 Rxe3

255
25.fxe3! Rd6 26.Rf3 Qb1† 27.Rf1 Qxa2 28.Bc5± Black must give up the exchange, and his
chances of surviving the ensuing endgame are not good at all.

23.Ref3 Qb1† 24.Kh2 Re6

Black seems to be holding his position together, but White has a forced win.

25.Bc5!
25.Rxg6† fxg6 26.Rg3 looks tempting but 26...g5! keeps the game going.

25...Rd8 26.Rxg6† fxg6 27.Rg3

Having lured the rook to d8, White wins easily, because 27...g5 can be met by 28.Rxg5† when
White will soon pick up both rooks.

256
 B2422) 9...e6

This has been Black’s usual reply on the rare occasions that this variation has arisen in
practice.

10.Ne2
10.0-0-0?! Qxd4 leaves White with inadequate compensation, as he has to waste time dealing
with the threats of ...Qxc4 and ...Bg4.

10.c3 has been played a few times but to play such a move defeats the whole purpose of our
move order up to this point, which is to prepare 0-0-0 without loss of time, while being ready to
sacrifice the c2-pawn.

10...Nd7
Accepting the pawn sacrifice seems like a worse idea here than on the previous move:
10...Bxc2
The problem for Black is that, having committed to ...e6 rather early, the e6-point becomes
a target for sacrifices.
11.0-0
11.Rc1 Bf5 12.Ng3 Bg6 13.Nh5 Nd7 14.0-0 also looks decent but the text move is more
flexible.
11...Be7
11...Nd7 12.Nf4 Nb6 13.Nxe6! fxe6 14.Bxe6 gives White a vicious attack.
12.Rac1

257
12...Bf5
12...Bg6 13.Rfe1 also puts Black under a lot of pressure.
13.Bh6 Bf8
13...Nd7 14.Ng3 Bg6 15.h4 catches the bishop.
14.Qe3±
Intending Ng3, and after Bg6 then h4-h5.

11.0-0-0
We have carried out the first part of our plan, namely castling without having had to spend an
additional tempo defending the c2-pawn. It is true that the queen on f3 gets in the way of the
attractive plan of g2-g4 and f2-f4-f5, but advancing the g- and h-pawns in conjunction with Nf4
remains as an effective alternative.

11...Qb6!?N

258
 Preparing to shelter the king on the queenside seems best.

11...Nb6 is met by 12.Bb3 when the focus will soon come back to the kingside, as 12...a5 13.a3²
doesn’t change much.

11...Qc7 12.g4 Bg6 is hardly an improvement on the main line from Black’s perspective. This
can be emphasized by:

13.Ng3! Exploiting the fact that the queen does not attack the d4-pawn from c7. (There is
nothing wrong with 13.h4 h5 14.g5 0-0-0 [14...f5? 15.Nf4+–] 15.gxf6² as per the main line
below, so all in all White has a pleasant choice.) 13...0-0-0 14.h4 h6 15.Rhe1 Kb8 16.h5 Bh7
17.Ne4 Bxe4 18.Qxe4± The h6-pawn will be a huge weakness in the long run – and even if
Black avoids losing it outright, the possibility of forcing a distant passed pawn with f4/g5 makes
future endgames dangerous for Black.

We should also consider the only move to have been tested in practice:
11...Rg8?!
This seems like a dubious move in view of:

259
12.h3!
Preventing any ...Bg4 ideas while renewing the threat of g2-g2 followed by h2-h4, when the
position of the rook on g8 takes away the possibility of ...h5.
12...Nb6
12...Qb6 invites White to go full steam ahead on the kingside: 13.g4 Bg6 14.h4±
13.Bb3 a5 14.a3 a4 15.Ba2 Nd5
15...Bd6 16.g4 Bg6 17.h4+– leaves Black without a good move.
16.c4 Ne7
White should simply press ahead with the familiar plan:

17.g4N
17.d5? cxd5 18.cxd5 Nxd5 19.Bxd5 Qxd5 20.Qxd5 exd5 let Black off the hook in Vouldis
– Kotronias, Athens 2003.
17...Bg6 18.h4 h6 19.Nf4±

260
12.g4 Bg6 13.h4

13...h5

13...h6? has some positional drawbacks at the best of times, but here matters are even worse
for Black due to 14.Nf4 Bh7 15.g5!+– when his position collapses.

14.g5
This simple move seems best.

14.Nf4 hxg4 15.Qe2!? leads to interesting complications after: 15...Bf5 16.Nxe6 Bxe6 17.Bxe6
fxe6 18.Qxe6†

18...Be7! (18...Kd8? loses to 19.Rhe1 Kc8 20.Qxg4 Bd6 21.Re6 Qc7 22.Rxf6 Rd8 23.Rf7+–)

261
19.Rhe1 Rh7 The position is messy but Black seems to be okay.

14...0-0-0 15.gxf6
White has captured a pawn and it is not clear how Black should proceed.

15...Bf5
15...Bh6 16.Bg5² and 15...c5 16.Bc3² also favour White.

16.Bg5²
The onus is on Black to regain the pawn or demonstrate compensation somehow.

Conclusion

This was the first of several chapters dealing with the 2...Qxd5 Scandinavian. After 3.Nc3 we
started by analysing 3...Qd8, which Black usually follows with an early ...Bg4 followed by
exchanging on f3. This may seem timid, but if White castles on the queenside too quickly, he
allows an annoying simplifying plan involving an early ...Qd5, hitting a2 and thus forcing a
queen exchange. My recommended move order is designed to avoid this, and I like White’s
chances.

We then moved on to 3...Qa5 4.d4, when our repertoire is based on the concept of meeting
4...Nf6 with 5.Bd2, keeping the knight on g1 for the time being. After 5...c6 6.Bc4 we
considered several options, the most popular and thematic being 6...Bf5 7.Nd5 Qd8 8.Nxf6†
gxf6. Here it took some effort and creativity to achieve my objective of preparing 0-0-0 without
spending time defending the c2-pawn, while also utilizing the fact that the knight has not yet
moved to f3. 9.Qe2!? has been analysed but 9.Qf3! is strongest. Then 9...Bxc2N offers White a
lasting initiative for the pawn, while 9...e6 can be met by 10.Ne2 followed by 0-0-0 and pushing

262
the kingside pawns, with fine prospects for White.

263
A) 6...Bf5 145
B) 6...Bd7 7.f4!?N 147
B1) 7...e6 148
B2) 7...a6!? 8.Nf3 Qb6 9.Bc4 Nc6 10.Na4! Qa7 11.Nc5 152
B21) 11...Bg4 153
B22) 11...Bf5!? 154
B3) 7...Bg4!? 8.Nf3 e6 156
B31) 9.Bb5†!? 157
B32) 9.h3 158

264
1.e4 d5 2.exd5 Qxd5 3.Nc3 Qa5 4.d4 Nf6 5.Bd2 Bg4

This has not been the most popular choice overall, but it seems to be the weapon of choice for
most of the leading exponents of Black’s set-up (this is true after 5.Bc4 as well as 5.Bd2). Black
plays more flexibly than in the classical variations with ...c6 and ...Bf5, which we considered in
the previous chapter. Even though theory isn’t exactly settled there, White at least has a clear
plan of Bc4 and Nd5 over the next few moves. With the text move, on the other hand, Black
provokes a slight weakening with f2-f3 and then retreats the bishop, giving White many more
options – which translates to more ways to go wrong. This is especially true because many
players don’t seem to prepare this line as thoroughly for White.

6.f3
6.Be2 is a safe alternative. I briefly considered 6...Bxe2 7.Ncxe2!? but found 7...Qa6! to be a
good reply, preventing c2-c4 for the time being. Black will continue with ...e6, ...Nd7, ...Bd6 and
later ...e5 or ...c5, with a decent game.

A) 6...Bf5 has been the most popular choice, but B) 6...Bd7 is the more challenging move.

6...Bh5 7.g4 Bg6 transposes to variation A.

A) 6...Bf5

265
 This move is recommended by both Houska and Bauer in their respective books on the
Scandinavian, but I think it pushes the provocation slightly too far, as g2-g4 is useful for
attacking purposes while any weakening of the kingside is not really felt by White.

7.g4 Bg6
7...Bd7 8.g5! has led to rather unpleasant results for Black, for instance: 8...Nd5 (8...Nh5 9.f4
g6 10.Qf3 [this is not considered by Houska, whose line continues with 10.Bg2] 10...c6 11.Bc4
Bg7 12.Nge2 Na6 13.0-0-0 Qc7 14.Rde1 Rd8 15.Ng3 Nxg3 16.hxg3+– Pinchon – Ibanez, corr.
2014) 9.Ne4 Qa4 10.c4 Qxd1† 11.Rxd1 Nb6 12.Bf4

12...Na6 13.b3 e6 14.Ne2 0-0-0 15.h4 Bc6 16.Bh3 Kb8 17.Kf2 Be7 18.h5 Rhf8 19.Rhg1 Black
suffered from a serious lack of space in Stevic – B. Kovacevic, Zagreb 2013.

8.f4 e6

266
 Now White gets a strong initiative with a thematic pawn sacrifice.

9.f5! exf5 10.g5 Nc6!?

Other moves also fail to solve Black’s problems, for instance:
10...Nd5 11.Qe2† Be7 (11...Ne7N is covered under the 10...Ng8 move order below) 12.Qb5†
Qxb5 13.Bxb5† c6 14.Nxd5 cxb5 15.Nc7†± and White won material in Nataf – Sanchez,
Belfort 2012.
10...Ng8 11.Qe2† Ne7N (11...Be7? led to an eventual success for Black in Boidman – Mann,
Germany 2006, but the simple 12.h4N+– would have given White an overwhelming advantage)

12.h4 h5 13.gxh6 Rxh6 (13...gxh6 14.Nb5 Qb6 15.Qe5+–) 14.Bxh6±

10...Nfd7 11.Qe2† Kd8 12.0-0-0ƒ

When I first went through these lines I ended this particular branch here, as it seemed clear
that White should be doing well. However, I later realized it was worth paying a bit more

267
 attention to the following move:
12...f4!?
GM Eric Prie once won a game with this move, and his compatriot Christian Bauer
provided some analysis of it in Play the Scandinavian. Black is trying to get some
counterplay on the light squares but White can take away all his fun with a nice
improvement:

13.Nh3!N 13...Qf5
If 13...f3 14.Qxf3 Qf5 15.Qxf5 Bxf5 16.Bg2 Nc6 17.Bf4+– it is obvious that the queen
trade has been of no help to Black.
14.Bxf4 Bb4
Targeting the knight is Black’s only serious idea, but White has more than one good
answer.
15.Nb5!
This is the most accurate, although 15.Qd3!?± is an acceptable alternative.
15...Na6 16.Qg2! c6
16...Qe4 17.a3 Be7 (17...Ba5 18.Qxe4 Bxe4 19.Rg1+–) 18.Nc3 Qxg2 19.Bxg2 Rb8
20.Rhf1+– gives White an overwhelming advantage in piece activity.

268
17.d5!!
A beautiful finesse, although it is important to stress that White’s success does not depend
on being able to memorize or calculate such details, as the simpler 17.Bd3 Qe6 18.d5 leads
to a near-decisive advantage as well.
17...Qxc2†
17...cxb5 18.Bd3 traps the queen.
18.Qxc2 Bxc2 19.Kxc2 cxb5 20.Kb1+–
Black is unable to get coordinated and it should not take long for White to utilize his active
pieces to make decisive inroads.

11.Qe2†!N
11.gxf6 has been played a few times but it leads to some messy complications.

11...Kd7 12.0-0-0 Re8 13.Qb5 Qxb5 14.Bxb5±

269
 White wins material without allowing any serious counterplay.

B) 6...Bd7

Surprisingly, I can already recommend a brand new idea.

7.f4!?N
The usual continuation is 7.Bc4 Qb6 8.Nge2 e6, which has been tested in many high-level
games. It seems to me that Black has generally been doing fine here, and the onus is on White to
create new problems for the defence. The position is still rich with possibilities, but I could not
find anything truly convincing.

The text move may seem slightly odd because the pawn of f4 appears weakening – but to my
eye, it looks better here than on f3. More importantly, it enables White to continue developing in
the most natural way, Nf3, Bd3, Qe2, Ne5 and so on. If White can play all those moves, then
suddenly the f4-pawn would fit in perfectly, while the bishop on d7 might seem out of place.
Therefore the onus is on Black to try and exploit White’s play, rather than simply developing in a
routine fashion.

I analysed three main options: B1) 7...e6, B2) 7...a6!? and B3) 7...Bg4!?.
7...Nc6?! 8.Nf3 0-0-0 enables White to target the queen with:

270
9.Nb5! Qb6 10.a4 a5 11.c4±

B1) 7...e6

This natural move may be followed by ...c5. Alternatively, Black may look to activate his
bishop on c6, or perhaps exchanging it on b5 after trading off the c3-knight.

8.Nf3 Qb6
Safeguarding the queen is prudent. I checked three other possibilities:

8...c5 9.d5! Qc7 10.Bc4 exd5 11.Nxd5 Nxd5 12.Bxd5 Be7 enables White to seize at least a
slight edge with:

271
13.Ne5 0-0 14.Qh5 Be6 15.Bxe6 fxe6 16.Qg4 Rf5 17.0-0-0² The e-pawn is a clear target.

8...Bc6
This seems a natural way to improve the bishop, although of course it also presents a big
target for a future Ne5.
9.Bc4 Bb4
9...Bxf3? 10.Qxf3 Nc6? 11.Nd5+– is an example of what Black must avoid.

10.Ne5!
10.Qe2? is too slow, and 10...Bxf3 11.Qxf3 Nc6³ gives Black excellent counterplay.
10...0-0
10...Bxg2 11.Rg1 Bd5 12.Bxd5 exd5 13.Rxg7±
11.0-0 Bd5 12.Bd3 Nc6 13.Qe1
White’s forces are ideally placed whereas Black’s pieces, especially the d5-bishop and the

272
 queen on a5, are quite awkward.
A possible continuation is:
13...Qb6 14.Nxd5 Nxd5 15.c3±

A final option is:

8...Bb4 9.Bd3!
9.a3?! Bxc3 10.Bxc3 Qd5 11.Bd3 Bb5= gives Black exactly what he is looking for. With
only the dark-squared bishop remaining, White won’t have much to play for.
9...Bxc3
9...c5 is met by 10.a3 Bxc3 11.Bxc3 Qc7 12.dxc5 Qxc5 13.Qe2± when White’s dark-
squared bishop is great.
9...0-0 10.Qe2 Nc6 11.0-0 Qb6 12.Be3 Ne7 13.Ne5± also leaves White in control.
10.Bxc3 Qd5

11.Qe2!
As long as our light-squared bishop stays alive, we will have an excellent position with two
bishops and Ne5 coming up, while Black is struggling for space. If Black could somehow
exchange his light-squared bishop – even for the f3-knight – then his position would be fine.
11...a6 12.Bb4±
By preparing c2-c4, White continues to thwart Black’s hopes of swapping off his problem
bishop.

273
9.Bc4!
White’s main priority is to prepare against ...c5, which can now be met by d4-d5. The bishop
may not be ideally placed for other purposes, but simply discouraging ...c5 is enough, as if Black
cannot play this move then he is liable to suffocate slowly.

9...Bd6
Again, in the absence of any games or established theory, I considered a number of other
plausible options:

9...c5?!
I have already mentioned that White’s last move was directed against this possibility – but
still, it is worth seeing the ‘proof’ of why it is bad.
10.d5 Be7 11.Qe2 exd5
11...0-0 accepts a big pawn weakness after: 12.dxe6 Bxe6 (or 12...fxe6 13.0-0-0±) 13.Bxe6
Qxe6 14.Qxe6 fxe6 15.Ng5±
12.Nxd5 Nxd5 13.Bxd5
Black has serious development troubles.
13...Nc6
13...0-0? 14.Qxe7 Re8 fails to: 15.Bxf7†+–
14.0-0-0 0-0 15.Bc3!
15.Bxc6 Bf6! is Black’s only trick.
15...Qc7
15...Rfe8? 16.Bxf7† Kxf7 17.Rxd7+–
16.Ne5±
9...Bc6!?
Threatening ...Bxf3 followed by ...Nc6 is critical.
10.a4!

274
 10.Qe2?! is too slow, and 10...Bxf3 11.Qxf3 Nc6³ favours Black.
10.f5!? Bxf3 11.Qxf3 Nc6 12.fxe6 fxe6 13.0-0-0 Nxd4÷ gives White definite
compensation for the pawn, but Black is also not without his chances.
10...a5
10...Bxf3 11.Qxf3 Nc6 12.a5 Qxd4 13.Bb5 Bc5 (13...Qd6 14.a6+–) 14.Bxc6† bxc6
15.Qxc6† Ke7 16.0-0-0 Bd6 17.Rhe1ƒ seems promising for White.

11.f5! Bxf3
11...exf5? 12.Ng5+– catches Black’s king in the centre.
12.Qxf3 Nc6 13.fxe6 fxe6 14.Nb5!
This is the key benefit of including a2-a4 to provoke ...a5.
14...0-0-0
14...Nxd4? 15.Nxd4 Qxd4 16.Qxb7+– is no good for Black.
15.0-0-0 Nxd4 16.Be3 e5 17.c3 Qxb5
17...Nxf3 18.Bxb6 cxb6 19.Be6†! Nd7 20.gxf3±
18.Rxd4 Qe8

275
19.Ba6! c6 20.Rxd8† Kxd8
Black must take with the king to avoid Qxc6†.
21.Bxb7 Kc7 22.Ba6±

Just like on the previous move, Black may consider:

9...Bb4
With ideas of ...Bxc3 followed by ...Bb5 – which of course we will prevent.
10.Qe2 0-0 11.0-0-0

11...Bc6
After developing his bishop to b4 it is clear that Black is not aiming for ...c5. Instead, he
should strive to activate or exchange his light-squared bishop, otherwise he will have no
counterplay.
12.Ne5 Bxc3 13.Bxc3 Bd5 14.Bd3!

276
 Of course!
14.Rhe1 Bxc4 15.Qxc4 Nc6 16.Nxc6 bxc6= gives White nothing.
14...Bxa2
Alternatives such as 14...Nc6 15.a3± leave Black with no counterplay, so the text move is
the only critical try.

15.g4! Nd5 16.g5

White’s attack is overwhelming, while there is no risk on the queenside.
16...Nxc3 17.bxc3 c5
17...g6 18.h4+– leaves Black without a good defence.
18.Qh5 g6 19.Qh6+–
White will soon break through in one way or another.

10.Qe2 0-0 11.0-0-0 c5

This seems like Black’s best chance for counterplay.

277
11...Nc6 12.g4 gives White a promising initiative, for instance: 12...Rad8 13.Kb1 a6 (13...Ne7
14.Ne5²) 14.g5

14...Nh5 15.Ne5 g6 16.Nxd7 Rxd7 17.d5±

12.dxc5!
12.d5 may appear more natural, and the position after 12...exd5 13.Nxd5 Nxd5 14.Bxd5 Nc6
15.Ng5!ƒ would certainly deserve further analysis, but I like the text move even more. Opening
the position should favour White due to his lead in development.

12...Bxc5 13.Rhe1 Bc6

13...Qc7 14.f5! exf5 15.Bg5‚ looks dangerous for Black.

13...Nc6 14.Ne4! (14.Na4 Nd4! is okay for Black) 14...Nxe4 15.Qxe4 gives Black serious

278
problems, for instance: 15...Qc7 (15...Ne7 16.Ng5 Ng6 17.Ba5! Qxa5 18.Rxd7+–) 16.Ng5 g6
17.f5!+–

14.Ne5 Nbd7 15.Nxc6 Qxc6 16.g4 Nb6 17.Bd3²

Black is still in the game, but White’s bishop pair and edge in activity give him the better
chances.

B2) 7...a6!?

This move may look a little strange, and at first glance you might struggle to see the purpose of
it. However, the point becomes clear when you consider the plan of ...Nc6 and ...Qb6, attacking
the d4-pawn. Without ...a6 inserted, this would be refuted by the simple Na4, trapping the queen.
Another feature of Black’s set-up is that, by delaying ...e6, Black keeps the option of ...Bg4 in
the future, adding even more pressure on the d4-pawn.

279
8.Nf3 Qb6 9.Bc4
9.Bd3!? is interesting but Black has a good reply in 9...e6! (9...Nc6? 10.Na4! Qa7 11.c3± is
White’s idea) 10.Qe2 Nc6! when White must give up a pawn for unclear compensation: 11.0-0-0
(11.Qf2 is conveniently met by 11...Nb4) 11...Nxd4 12.Nxd4 Qxd4÷

9...Nc6
9...e6?! 10.Qe2± makes life harder for Black, as ...Nc6 will invite d4-d5.

10.Na4!
I must admit there is a part of me that feels uneasy about committing to Na4-c5 without having
completed development, as it costs time and leaves the centre momentarily unstable. However,
the loss of time is balanced out by the fact that Black has already spent several tempos moving
his queen about, so we can afford a bit of leeway. The knight manoeuvre also prepares c2-c3,
which will make the centre much more stable. And most importantly, White’s position holds up
to concrete analysis – although Black has some interesting resources as well.
10.Qe2 Bg4! puts d4 under increasing pressure.

10.d5 Nd4 11.Nxd4 Qxd4÷ could be analysed further, but Black seems okay.

10...Qa7 11.Nc5
If White was given time for Nxd7 and c2-c3, he would be in complete control. Thus, B21)
11...Bg4 and B2) 11...Bf5!? are the moves that we should take seriously.
11...Nxd4 is less of a problem than you might think, as 12.Nxd7 Nxd7 13.Bc3 Nxf3† 14.Qxf3
gives White excellent compensation. A sample line is:

280
14...e6 15.0-0-0 0-0-0 16.f5!± White’s wonderful bishops and lead in development are clearly
worth more than a pawn.

B21) 11...Bg4

Pinning the knight may seem more natural although, with c2-c3 imminent, it is not clear if
Black will be able to exert meaningful pressure on White’s centre.

12.c3 e6 13.Nd3!
The knight is well placed here, and it can go to e5 when the time is right.

13.Nb3 Bd6 14.h3 Bf5 15.Ne5 Ne7 16.0-0 c5! 17.dxc5 Bxc5† 18.Nxc5 Qxc5† 19.Kh2 Ne4„
would not be so clear.

281
13...b5
It seems logical to take a bit of space on the queenside.

White is doing well after:

13...Bd6 14.Nde5 Bf5
14...Bxe5 15.fxe5 Nd5 16.0-0±
15.Qe2 0-0 16.Be3!
16.0-0-0 b5 17.Bb3 Ne7÷ is less convincing.
After the text move, the queen on a7 is beginning to feel rather misplaced.

16...Nd5
16...Ne7?! invites 17.d5±.
16...Qb8 17.Nxc6 bxc6 18.Ne5² is pleasant for White.
16...Rae8!? is an interesting try but 17.0-0² keeps a pleasant edge for White. (17.d5?! Nxe5!
18.Bxa7?! [18.fxe5 Bc5÷] 18...Nxc4 would justify Black’s 16th move.)
17.g4 Be4 18.0-0 Bxf3 19.Rxf3ƒ
White has excellent chances on the kingside.

282
14.Bb3 Be7
In the event of 14...Na5 15.Bc2 Nc4 16.Nde5 Nxd2 17.Nxg4 Nxg4 18.Qxd2± White’s bishop
should give much better attacking prospects than Black’s, which is often the key to success in
these opposite-coloured-bishop middlegames.

15.Bc2 0-0 16.Qe2

I find White’s regrouping rather aesthetically pleasing. It’s not clear if White is really better,
but his plans for the middlegame seem clearer.

16...Qb6
16...Bf5?! seems like the wrong idea, and 17.0-0 Qb6 18.h3 h6 19.g4 Bh7 20.Qg2± sees White
building a menacing attacking position.

17.0-0 Rad8 18.Rae1

18.Nde5!? can also be considered, although it has the slight drawback of allowing Black to
trade off the c6-knight, which was not doing much.

18...Bf5 19.h3 Ne4 20.Kh2 Nxd2 21.Nxd2

White intends g2-g4 next, with decent attacking prospects.

B22) 11...Bf5!?

283
 This allows us to spoil Black’s pawn structure; nevertheless, it is arguably the more
challenging move.

12.Ne5
12.Nb3!? e6 13.c3÷ is an interesting alternative which could be explored further. By delaying
Ne5, we can be more flexible. Still, it seems right to concentrate on the more ambitious and
concrete approach.

12...e6 13.Nxc6 bxc6 14.Qf3

White’s position seems like it should be more pleasant. However, we have yet to complete
development and fully coordinate our pieces, which gives Black a chance to fight back on the
queenside.

14...Qb6
14...Bxc5 15.Qxc6† Kf8 16.Qxc5† Qxc5 17.dxc5 Bxc2 18.Ba5² gives White the more
pleasant endgame with the two bishops.

14...Nd5 can be met by 15.Bb3² when it is not clear what Black should do next. Note that it
would be a bad idea to ‘win’ material by taking twice on c5, because g2-g4 followed by f4-f5
would trap the bishop.

284
15.b4!?
This is the most challenging continuation, trying to shut off the black queen completely.

15.Na4 Qb7 16.c3 is the alternative, when Black’s pawn structure remains a problem and he
needs to figure out a way to solve it soon. If he fails to carry out a quick ...c5 then White will
most likely be able to claim some advantage. Therefore 16...Be4! 17.Qe2 c5 seems best, and
after 18.0-0 cxd4 19.cxd4 Be7÷ the position remains complicated.

15...a5!
Fighting back on the queenside seems best.

15...Bxc2 16.0-0 Bd6 is well met by:

17.a4! 0-0 18.a5 Qa7 19.g4ƒ I think White’s initiative is worth more than a pawn.

285
16.a4

16...Nd5!
16...axb4? 17.a5!+– exploits the hanging c6-pawn.

16...Bxc5 17.dxc5 Qa7 18.0-0 0-0 19.c3 seems risky for Black, whose queen is shut out of play
for a while. This gives White chances to launch an attack on the kingside, either with a timely
g2-g4 or by transferring the dark-squared bishop to the a1-h8 diagonal.

17.c3
Another interesting line is: 17.g4!? Bxc2 18.Bxd5 cxd5 19.Rc1

19...Bxa4!? 20.Nxa4 Qxd4 21.Nc5 axb4 22.Nb3 Qb2© The position remains messy but I
suspect Black is okay.

286
17...axb4 18.Bxd5 cxd5 19.cxb4 Be7 20.Qe2 0-0 21.0-0 c6

22.g4 Be4 23.a5 Qa7 24.Rfc1

Black remains under slight pressure, and exchanging on c5 does not seem to help him. Perhaps
Black should just sit and do nothing, when it is not exactly clear how White should improve, but
at least the position seems pretty one-sided.

B3) 7...Bg4!?

This is another important move to consider. Does the inclusion of f2-f4 even benefit White?

8.Nf3 e6
8...c6?! is too slow in view of 9.h3 Bxf3 10.Qxf3± when Black has been forced to trade off his

287
bishop, leaving a position where having the pawn on f4 is certainly not a negative for White, and
he has a clear superiority due to his bishop pair and space advantage.

8...Nc6?! is also inadvisable due to: 9.Nb5! Qb6

10.a4! This forces 10...a5, after which 11.h3 Bxf3 12.Qxf3± leaves White in a dominating
position with the knight enjoying the b5-square unchallenged.

Being in uncharted territory, I decided to investigate a few interesting ideas: B31) 9.Bb5†!? is
worth considering, but the calm B32) 9.h3 would be my main recommendation.

9.Bd3?! Nc6! puts immediate pressure on White’s centre.

9.Bc4 Bb4 10.a3 Bxc3 11.Bxc3 Qf5 seems quite okay for Black. Trading his bishops for the
enemy knights is not so bad here, as his knights have some good squares available and White’s
dark-squared bishop is not too active.

B31) 9.Bb5†!?

288
 The idea of this move is simple: we saw in the note above that the immediate Bd3 would have
been well met by ...Nc6, so we are provoking ...c6 in order to prevent Black from doing that. At
the same time, it does lose a move, which is not ideal.

9...c6
9...Nc6?? is unplayable due to 10.Nd5+– of course.

10.Bd3
I also considered 10.Nd5, reminding us of the previous chapter, but Black seems to be okay
after: 10...Qd8 11.Nxf6† Qxf6 (11...gxf6?! turns out badly for Black after 12.Bd3 Bxf3 [or
12...Nd7 13.h3 Bxf3 14.Qxf3 Bd6 15.0-0-0±] 13.Qxf3 Qxd4 14.0-0-0 Nd7 15.Kb1 0-0-0
16.Ba5±) 12.Bd3

12...Nd7! (12...Bxf3 13.Qxf3 Qxd4 14.0-0-0© is more interesting for White) 13.c3 Bd6 14.0-0

289
Bf5 15.Bxf5 Qxf5 16.Ne5 Nf6 17.Qb3 Rb8 18.Rae1 0-0 Black seems to be doing just fine. In
general, with fewer minor pieces, or if White doesn’t have the bishop pair, then this type of
structure should be okay for Black.

10...Nbd7
10...c5? 11.d5± is no good for Black.

10...Be7?! 11.h3 Bxf3 12.Qxf3 0-0 13.0-0-0 Nbd7 14.g4 Nd5 15.Kb1± also turns out
excellently for White.

10...Qb6!? is more interesting but White can make a promising pawn sacrifice: 11.h3 Bxf3
12.Qxf3 Qxd4

13.0-0-0 This could be analysed in more detail, but I would be pretty optimistic about White’s
chances.

10...Bb4 is a natural move to consider, but White can fight for an advantage with: 11.a3 Bxc3
12.Bxc3 Qc7 13.0-0 0-0 (13...Nbd7 14.Qe1! 0-0 15.Nh4 Bh5 16.Qg3ƒ sees White develop
some play on the kingside) 14.Qe1 Bf5 15.Bxf5 exf5

290
16.Qe5! Qd7 (16...Qxe5 17.fxe5 Ne4 18.Nd2 also favours White) 17.Nh4²

11.h3 Bf5 12.Bxf5 Qxf5

It is possible to explore this position further, but after exchanging these bishops I think Black
should be okay. White still has some ideas floating around involving a timely Ne5 which might
cause Black some problems – but I can’t escape the feeling that all this would have been so much
better if we could have kept our light-squared bishop alive.

B32) 9.h3

291
9...Bxf3
9...Bh5? would unsurprisingly put the bishop in danger after: 10.g4! Bg6 11.Qe2 Be7
(11...Bxc2 12.Nb5! Qb6 13.Qc4+–)

12.Ne5 Sooner or later, f4-f5 will be crushing. For instance: 12...0-0 13.Bg2 c6 14.f5+–

10.Qxf3 Nc6
Clearly, Black has to go after the d4-pawn to cause White any problems.

11.Bb5
11.Nb5!? Qb6 12.a4 looks tempting, but Black has enough tactical resources after: 12...Bb4!
(12...a5? 13.c3±) 13.c3 a6! 14.cxb4 axb5 15.Bxb5 Qxb5!? 16.axb5 Rxa1† 17.Kf2 Rxh1 18.bxc6
b6= With the knight heading for the ideal d5-square, I am not sure how realistic White’s winning
chances really are, despite the computer’s optimism. Nevertheless, this could certainly be

292
analysed further and some players may consider this line a decent practical try.

11...Qb6 12.d5!?
Aiming to simplify into a symmetrical pawn structure while maintaining the bishop pair. I
checked a couple of alternatives:

12.f5 0-0-0! (White would be happier after 12...exf5 13.d5; or 12...a6 13.Bxc6† Qxc6 14.Qxc6†
bxc6 15.fxe6) 13.Bxc6 Qxc6 14.Qxc6 bxc6 15.fxe6 fxe6

16.Ne2 (16.0-0-0 Rxd4 17.Rhe1 Kd7 seems okay for Black) 16...c5 17.Be3 cxd4 18.Bxd4 The
position is close to equal and I think Black should be okay; nevertheless, a strong technical
player might try to make something out of White’s superior structure, with two pawn islands
against four.

The other possibility is:

12.Bxc6†!? Qxc6
12...bxc6 13.f5 exf5 14.0-0-0 seems promising for White.
13.Qxc6† bxc6 14.Ke2
This seems like another one of those endgames which should result in a draw, but Black has
to be slightly careful to avoid an unpleasant defence, whereas White does not take much
risk. Here are a few brief possibilities.

293
14...Be7
a) 14...0-0-0 is similar and may transpose within a move or two.
b) 14...Bb4 15.Na4 maintains an edge for White.
c) 14...c5 and now 15.dxc5 Bxc5 16.Be3 seems like the simplest way to maintain a tiny
edge. 15.Be3 is also possible, when Black can choose between simplifying with 15...cxd4
and making things a little more complex with 15...c4!?.
15.Be3 0-0-0 16.Rad1
16.Kf3 Nd5 is similar.
16...Nd5
16...a6 17.Kf3 does not change much.

17.Ne4
17.Nxd5 cxd5 improves Black’s structure, and 17.Kf3 Nxc3 18.bxc3 Rd5 damages
White’s.

294
 17.Rd3!? could be considered though.
17...f5 18.Nf2
Again, it’s close to equal but marginally more comfortable for White.

12...Nxd5
12...exd5 needlessly gives White the additional option of 13.Qe2†!?, but he can simply
transpose to our main line with 13.Nxd5 Nxd5 14.Qxd5 of course.

13.Nxd5 exd5 14.Qxd5 Bc5!

Forcing White to exchange on c6.

14...Rd8 15.Qe5† would be annoying for Black.

14...Be7
This move is playable, but White has more freedom compared to the main line.
15.0-0-0 0-0
15...Rd8 16.Bxc6† Qxc6 looks pleasant for White after either 17.Qe5!? or 17.Qxc6† bxc6
18.Rhe1.

295
16.Bc4!?
White’s only difficulty is choosing between several pleasant continuations:
a) 16.Qc4 a6 (16...Bf6 17.c3²) 17.Bxc6 bxc6 18.Bc3²
b) 16.Bxc6 Bf6 It seems worthwhile to provoke the following pawn move. 17.b3 bxc6
18.Qc4 Rad8 (18...a5 19.a4²; 18...Qb5 19.Qxb5 cxb5 20.Ba5!) 19.Bc3² This leads to quite
a pleasant major-piece endgame for White. Of course Black has drawing chances – but it
isn’t easy, especially since White’s king is ideally placed to exploit the queenside
weaknesses.
16...Rad8
16...Bf6 leads to a similar position after 17.c3 followed by Qb5.
16...a6?! is too slow, as 17.c3 prevents all of Black’s counterplay and White no longer even
has to exchange queens. Play may continue 17...Rad8 18.Qe4 Bf6 19.Be3 Qa5 20.Rxd8
Rxd8 21.Rd1 with unpleasant threats.
17.Qb5
17.Qe4?! is too slow here due to 17...Bf6 18.c3 Rfe8.

296
17...Qxb5
17...Qd4 18.Bb3 leaves Black’s queen awkwardly placed.
18.Bxb5 Nd4 19.Bc4²
White has the more pleasant endgame due to the bishop pair. (19.Bd3!? Ne6 20.Be2 c6 21.f5
seems like a promising alternative.)

15.Bxc6†
This seems best.

15.Qe4† Kf8 16.Bxc6 Qxb2! would be annoying.

15.0-0-0 0-0 turns out okay for Black after 16.Bxc6 (16.Bd3 Nb4! 17.Bxb4 Qxb4=) 16...bxc6!
(avoiding a queen exchange, which would give White an improved version of our main line
because the black king is far away on the kingside) 17.Qf3 Bd4 18.Bc3 c5= when Black has

297
enough activity to compensate for his weak structure.

15...Qxc6
15...bxc6? 16.Qe4† Kf8 17.0-0-0± is hardly a serious alternative.

16.Qxc6†
After 16.Qe5† Qe6 17.Qxe6† fxe6 the isolated e-pawn is less of a weakness than the split
queenside pawns. Black has good chances to liquidate it with ...e5 – and even if he is unable to
do so, the e-pawn is still only a single target, whereas in the main line, all three of the black
queenside pawns will be weak.

16...bxc6

17.0-0-0 0-0-0!²
Keeping the king on the queenside will aid in the defence of the pawns. Still, White retains an
obvious edge because of his superior pawn structure. Having the pawn on f4 might seem slightly
weird but it’s not really a weakness, and it may enable us to seize some extra space on the
kingside with f4-f5 and g2-g4 at some point. Of course Black keeps good drawing chances, but
he will have to work hard for it.

Conclusion

After 3.Nc3 Qa5 4.d4 Nf6 5.Bd2, the provocative 5...Bg4 has taken over as Black’s first choice
among Scandinavian specialists. After 6.f3 we started by checking 6...Bf5, but 7.g4 is a useful
space-gaining move which enhances White’s attacking chances on the kingside, so it seems
unwise for Black to invite it with gain of tempo.

298
6...Bd7 is the more reliable choice, when 7.f4!?N is a brand new idea, after which I analysed
three main responses. 7...e6 should be met by 8.Nf3, followed by developing the bishop to d3 or
c4 according to what Black does. The two main ‘rules’ are as follows: White should be ready to
meet an early ...c5 with d4-d5; and he should also prevent an exchange of the light-squared
bishops. With the help of these guidelines and a few other precise details, White should obtain a
promising position.

We then turned to 7...a6!?, which looks weird but is connected with a direct plan of attacking the
d4-point. Play continues 8.Nf3 Qb6 9.Bc4 Nc6 10.Na4! Qa7 11.Nc5, reaching a weird situation
where both sides have made some exotic manoeuvres early in the opening. I found some
interesting ideas after both 11...Bg4 and 11...Bf5!? which should enable you to pose problems to
your opponents.

Finally, 7...Bg4!? is a natural choice, moving the bishop yet again. After 8.Nf3 e6 we first
considered 9.Bb5†!?, provoking 9...c6 in order to stop Black’s knight from utilizing that square.
Although White’s idea is not without merit, the slight loss of time is a drawback and I think
Black is okay. Therefore 9.h3 is my main recommendation, with the main line leading to a
slightly better pawn structure for White, after he meets ...Nc6 with Bb5 followed by taking on
c6. (The 11.Nb5!? line in the notes is another interesting option which could cause Black
problems.) The main line and notes contained several variations of the same theme: in general,
White’s superior structure gives him a small but lasting advantage with minimal risk.

299
A) 4...c6 164
B) 4...Nf6 5.Nf3 166
B1) 5...Bg4 6.h3 166
B11) 6...Bxf3 166
B12) 6...Bh5 169
B2) 5...a6 6.g3 Bg4 7.h3! 173
B21) 7...Bxf3 175
B22) 7...Bh5 177

1.e4 d5 2.exd5 Qxd5 3.Nc3 Qd6

300
 It wasn’t so long ago that this was considered a fringe variation, which was only really played
regularly by one world-class GM – Sergey Tiviakov. Times change though, and the text move
has since been played regularly by some of the strongest players in the world, including Magnus
Carlsen and Fabiano Caruana. Part of the appeal of this variation is that the lines tend to be less
forcing, so it is harder to prepare long, forced lines against it, thus making it less risky than a lot
of other openings.

4.d4
We will start by considering A) 4...c6, which is sometimes connected with the scheme of
leaving the knight on g8 for a while. We will then move on to the most natural and popular B)
4...Nf6.

4...a6 5.Nf3 Bg4 (Black almost always prefers 5...Nf6, which transposes to variation B2) 6.h3
Bh5 7.g4 Bg6 8.Ne5² gives White a pleasant game.

4...g6 is played from time to time, but after 5.Nf3 Bg7 6.Bg5!? White can just continue playing
in the same way as in the next chapter, and I cannot see anything better for Black than
transposing to it by playing ...Nf6 at some point.

A) 4...c6

5.Nf3 Bg4!?
If Black wants to postpone ...Nf6, this seems like the most challenging option.

5...Bf5
This sideline has been played by Tiviakov a few times, but it is not too difficult to handle.
6.Ne5 Nd7

301
 6...Nf6 leads to variation A of Chapter 10 on page 208.
6...e6? is asking for trouble. 7.g4 Bg6 occurred in Broehl – Forchert, Germany 2013, and
now 8.Bf4N 8...Qd8 9.h4+– would have been the easiest way to get a decisive initiative.
7.Nc4

7...Qf6?!
This was a surprising choice from Tiviakov, who is usually extremely well prepared.
7...Qg6 8.Ne3 Ngf6 9.Nxf5 Qxf5 10.Bd3² gave White an easy edge with the two bishops
in Welling – Sant Anna, Douglas 2014.
7...Qc7 8.Qf3 e6 9.Bf4 Qd8 10.0-0-0 Ngf6 occurred in Szamoskozi – Schmidt, Budapest
2008, when 11.Ne3N Bg6 12.h4² seems logical.
8.d5!?
8.g4 Bg6 9.Be3± (or 9.h4!?) would also have been excellent for White.
8...Ne5?!
This was the dubious continuation of Sukandar – Tiviakov, Jakarta 2015. White chose a
suboptimal continuation, but still went on to defeat his much higher rated opponent. However, at
this point any of 9.Be3N, 9.Bf4N or even 9.Bg5!?N 9...Qxg5 10.Nxe5 would have been great
for White.

302
6.h3!
Sacrificing the d4-pawn is clearly the most principled continuation. If something more cautious
like 6.Be3 or 6.Be2 is played, then Black will simply play 6...Nf6 with a better version of a
standard variation.

6...Bxf3
6...Bh5 7.g4 Bg6 8.Ne5 gives White a clearly improved version of variation B12.

7.Qxf3 Qxd4
7...Nf6 transposes to variation B11, which will be analysed shortly.

White should be doing excellently from here, but in practice he has often failed to make the most
of his chances. He still needs to show some precision, while being aware of the crucial Nb5
resource. The most accurate continuation is:

303
8.Bf4! Nf6
Other moves are no better, for instance:

8...Qb6 9.0-0-0+– gives White an overwhelming advantage in development.

8...Qd8 9.Nb5! Na6 occurred in Ucucu – Acar, Konya 2019, when 10.Bc4!N would have given
White a huge initiative.

8...Nd7 9.Rd1 Qb6 was seen in Krzyzanowski – Pultorak, Zalakaros 2017, when White could
have won with:

10.Rxd7!N 10...Kxd7 11.Bc4!+– With decisive threats.

8...e5 9.Bg5 Be7 10.Rd1 Qb6 led to an eventual victory for Black in Ivanchuk – Karpov, Cap

304
d’Agde 2013, but it was not because of the quality of his opening play.

11.Ne4!N+– is the simplest of many possible ways to get a winning advantage.

9.Nb5!
As mentioned earlier, this is a vital attacking resource. The next few moves are forced.

9...cxb5 10.Bxb5† Nbd7 11.Qxb7 Rd8 12.Bc7 Qe4† 13.Qxe4 Nxe4 14.0-0-0 a6N
14...Nxf2 15.Rhe1 Nxd1 16.Rxd1 g6 17.Bxd8 Kxd8 18.Rxd7† Kc8 19.Rxa7 was close to
winning for White in Fridman – S. Kasparov, Porticcio 2018. The text move is a slight
improvement, but Black is still in deep trouble.

15.Bc6 Nef6 16.Bxd8 Kxd8 17.Rxd7† Nxd7 18.Rd1 Kc7 19.Bxd7±

White has an extra pawn and the more active pieces, so his winning chances are excellent.

305
 B) 4...Nf6 5.Nf3

In the rest of this chapter we will analyse two important options: B1) 5...Bg4 and B2) 5...a6.
5...g6 and 5...c6 will be examined in Chapters 9 and 10 respectively.

B1) 5...Bg4 6.h3

Black has the usual choice between B11) 6...Bxf3 and B12) 6...Bh5.

B11) 6...Bxf3 7.Qxf3

7...c6
7...Nc6?! has been played in a bunch of games but 8.Bf4!N is a powerful reply. The critical
continuation is 8...Nxd4 (8...Qe6† 9.Be3 is also no fun for Black) 9.Qxb7 Nxc2† 10.Ke2 Nd4†

306
11.Kd1 Qxf4 12.Qxa8† Kd7 13.Qxa7 when Black is the exchange down and has no way to
exploit White’s slightly exposed king. For instance:

13...e6 14.Qa4† Kd8 15.Bd3 Qxf2 16.Qa8† Kd7 17.Rf1 Qe3 18.Ba6 White still has to remain
vigilant with his king in the centre, but he should be able to convert his material advantage with
careful play, especially since Black’s king is also less than safe.

8.Bf4 Qd8
8...Qxd4 takes us back to variation A, where we saw that 9.Nb5! is more or less winning for
White.

8...Qe6† 9.Be2 Nbd7 occurred in Mista – Proszynski, Ustron 2008, when the most accurate
continuation would have been:

10.0-0!N Queenside castling would allow some ...Nd5 ideas when the vulnerability of the a2-

307
pawn would at least slow down White’s attack. The text move avoids any such problems, and
after something like 10...Nb6 11.Rfe1 Qc8 12.Be5+– Black will not be able to complete
development without allowing something bad to happen.

9.d5! Nxd5
If 9...cxd5 then simply 10.Bxb8! followed by Bb5† creates major problems for Black’s king.

10.0-0-0 e6 11.Nxd5 cxd5

11...exd5 was played in Dragomirescu – Spulber, Mamaia 2019. White has several strong
continuations but my favourite is: 12.Qg3!N 12...f6

13.Bc4!? The flashy approach is the easiest. (13.c4+– is also fine) 13...Na6 14.Rhe1† Kf7
15.Qg4 Qc8 16.Rxd5 Qxg4 17.Rd7† Kg6 18.hxg4+–

308
12.Bxb8 Qxb8 13.Bb5† Ke7 14.Rhe1 a6
Black is unable to catch up in development. 14...g6 15.Rxd5 Bh6† 16.Kb1 Rd8 is too slow in
view of: 17.Rxd8 Qxd8 18.Rd1+–

15.Qxd5 axb5
15...Qf4† 16.Kb1 axb5 17.Qxb7† wins easily.

16.Qg5† Ke8
16...f6 allows a forced mate after: 17.Rxe6†! Kf7 (17...Kxe6 18.Qd5† Ke7 19.Qd7#) 18.Rd7†
Kg8 19.Qd5 h6 20.Rxf6† Kh7

21.Qe4† Kg8 22.Qe6† Kh7 23.Rxh6#

309
17.Qxb5† Ke7 18.Qg5†
We are following a game, in which White opted to repeat moves before proceeding with the
winning continuation.

18...Ke8 19.Qb5† Ke7 20.Rd7† Kf6

21.Rxf7†! Kxf7 22.Qd7† Be7 23.Qxe6† Kf8 24.Qxe7† Kg8 25.Qe6† Kf8 26.Qf5† Kg8
27.Qd5† Kf8 28.Qf5† Kg8
White has no immediate mate, but winning Black’s queen is the next best thing.

29.Re7 Qe8 30.Qd5† Kf8 31.Rxe8† Rxe8 32.Qxb7

Black soon had to resign in R. Horvath – Fauland, Deutschlandsberg 2018.

B12) 6...Bh5

310
7.g4 Bg6 8.Ne5 Nbd7
It seems advisable to challenge the strong knight.
The passive 8...c6? has been played in a surprisingly high number of games. White has several
strong continuations but the rare 9.f4! seems best. Black is already in severe difficulties, as
shown after: 9...e6N (9...Be4 10.Nxe4 Nxe4 11.Qd3 Qd5 12.Bg2 Qa5† 13.c3 gave White a
huge advantage in Schnegg – Boric, Vienna 2013)

10.Qe2!? (if you don’t feel like sacrificing a pawn then 10.Bg2± is an excellent alternative, when
Black does not really have a good answer to h4-h5) 10...Qxd4 (or 10...Be7 11.h4+– and Black
will be steamrollered on the kingside) 11.Be3 Qd6 12.Bg2+– The game continues, but White’s
massive advantage in activity should decide the game if he continues to play accurately.

9.Qe2!
This move is virtually untested. Evidently the idea of sacrificing a pawn did not occur to many

311
players, but it gives White superb play.

9...Qxd4
Black should accept the challenge, as 9...0-0-0 10.Bg2 e6 (or 10...Qxd4? 11.Nc6! bxc6
12.Qa6† Kb8 13.Be3+–) 11.h4!± gives White a serious advantage without much effort.

After the text move I found a couple of games where White exchanged on g6. However, it is
much better to leave the bishop as a target for the f4-f5 advance. Thus, my improvement is:

10.Nxd7!N 10...Qxd7
10...Nxd7 11.Be3 Qf6 12.0-0-0± gives White a tremendous amount of activity.

11.f4 h6
11...h5 is hardly an improvement, and after 12.f5 Bh7 13.Bd2 0-0-0 14.0-0-0 Black is likely to
be murdered on the queenside before he can get his kingside pieces into the game. For instance:
14...e6 15.Bg2 Nd5 (15...exf5 16.Bf4 gives White a deadly attack)

312
16.Nxd5 exd5 17.Bxd5! Qxd5 18.Bg5!+– Black is busted.

12.f5
White could also start with 12.Bg2, but it feels more natural to drive the bishop back before
anything else.

12...Bh7

13.Bg2
This is the most accurate move order, forcing Black to defend b7 in one way or another.

Instead, 13.Bd2 e6 14.0-0-0 Bd6 15.Bg2 allows 15...0-0! (rather than 15...c6 16.h4ƒ) 16.Bxb7
Rab8 17.fxe6 fxe6 18.Bg2 Ba3!? when the outcome is uncertain.

313
13...0-0-0
This seems like Black’s best try.

13...c6
This feels a little slow, although White still has to play accurately to exploit his dynamic
advantage.
14.Bd2 e6 15.0-0-0 Bd6
15...0-0-0 16.Qf2!? b6 17.Bg5 Qc7 18.Rxd8† Kxd8 19.Qd4† Kc8 20.Bf4 gives White a
decisive attack.

16.h4! 0-0-0
16...Qe7 17.g5 Nd5 18.Nxd5 cxd5 19.g6 fxg6 20.fxe6± is highly unpleasant for Black.
17.g5 Nd5
17...Ne8 18.g6 fxg6 19.fxe6 Qe7 20.Ne4ƒ is also good for White.
18.g6 Bg8
Black must avoid 18...fxg6? 19.fxe6+– when his position collapses.

314
19.fxe6 fxe6
19...Qxe6? is worse: 20.Bh3 f5 21.Qf2!+–
20.Ne4±
White’s initiative is definitely worth more than a pawn.

14.Bd2 e6 15.0-0-0 Nd5

15...c6 16.Qf2 b6 17.Bg5± keeps Black under unpleasant pressure.

15...exf5?!
This more is as risky as it looks, and we can refute it with some attractive tactics:

315
16.Bf4! Bd6
16...Qe6 allows 17.Bxb7†! Kxb7 18.Qb5† Kc8 19.Rxd8† Kxd8 20.Qb8†+– and the king
is caught.
17.Rxd6! cxd6 18.Nb5 Qe6 19.Qd2 d5 20.Qb4 Qb6 21.Re1!
Tightening the noose.

21...Rd7
21...Rhe8 allows White to execute the threat created by his previous move: 22.Nxa7†!
Qxa7 23.Qc3†! Kd7 24.Qc7 mate!
22.Qc3† Kd8 23.Qe5 Ne4 24.Bxe4 dxe4 25.gxf5 Re8 26.Qb8† Ke7 27.Rxe4†+–
Finally, it’s all over.

316
16.Qf2!
16.Nxd5 exd5 17.Be3 c6 18.Bxa7 Bd6÷ sees Black return the extra pawn and stay in the
game.

16...b6
16...Kb8 17.Nxd5 exd5 18.Be3± wins back the pawn in a much more favourable way than the
line above.

16...Qc6 17.Qxa7 Bc5 18.Qa5 Kb8 19.Nxd5 exd5 20.Bf4± is also excellent for White.

17.Qd4!
This does not threaten anything immediately, but the queen eyes a range of targets across the
board, and White is ready to bring his one inactive piece into play. Black is still a pawn up for
now, but he has a hard time coordinating his pieces or doing anything useful.

17...c6
17...Nxc3? 18.Qxc3+– only helps White.

17...Kb8 is more sensible but 18.Rhe1 leaves Black stuck for a good move, for instance: 18...Be7
(18...Rg8 19.Bf4 Bc5 20.Qc4+–; 18...Nxc3 19.Qxc3 Bd6 20.Qf3±) 19.Qxg7 Rhg8 20.Qe5 Bf6

317
21.fxe6 fxe6 22.Qxe6 Nxc3 23.bxc3± No fancy mating combination this time, but an extra pawn
will do nicely.

18.Rhe1 Rg8
18...Nxc3 19.Qxc3 Bd6 20.fxe6 fxe6 21.Rxe6± regains the pawn while keeping a big
initiative.

18...Bd6 19.Qxg7 exf5 20.Nxd5 cxd5 21.Bxd5 fxg4 22.hxg4± is also great for White.

18...Kb7
If Black maintains the tension in this way, White increases the pressure as follows.
19.Bf4 Qc8 20.Nxd5 cxd5 21.a3!?
We can afford to spend a tempo on a small improving move.
21.Kb1 Qc5 gives Black a marginally improved version of the note below.

318
 21.Qa4!? is a tempting alternative though.
21...Rg8
21...Qc5 is the lesser evil although it leads to a clearly better endgame for White after:
22.Qxc5 Bxc5 23.fxe6 fxe6 24.Rxe6 Rhf8 25.Bxd5† Rxd5 26.Rxd5 Bg8 27.Rd7† Kc8
28.Rc7† Kd8 29.Rec6 Rxf4 30.Rc8† Kd7 31.Rxc5 bxc5 32.Rxg8±
22.Qa4 Bd6
22...Bc5 23.c4+– and White will soon break through.

23.Bxd6 Rxd6 24.Rxd5! exd5 25.Re7†+–

19.Qa4! Bd6
19...Kb7 20.Ne2! creates a horrible threat of c2-c4.

20.Ne4!?
20.Ne2!? is a promising alternative.

20...Qc7
Other moves are no better, for instance:

20...Kb7 and 20...Bb8 fail to address the main threat: 21.c4+–

20...exf5 21.Nxd6† Qxd6 22.Qxa7 Qc7 23.Qa8† Kd7 24.Qa4± leaves Black’s king vulnerable.

20...Bh2 is the computer’s initial suggestion, but it soon becomes clear that 21.c4 exf5 22.gxf5
Bxf5 23.cxd5 cxd5 24.Qxd7† Rxd7 25.Nc3± reaches a position where White’s extra piece is
worth more than Black’s three pawns.

319
21.c4 Ne7 22.Nxd6† Rxd6 23.c5! bxc5
After 23...Rd7 24.Bf4 the defence comes unglued. For instance: 24...b5 25.Rxd7 Qxd7 26.Qa5
Rd8 27.fxe6 fxe6 28.Bd6+–

24.Bf4 Rxd1† 25.Rxd1 Qb6 26.fxe6 fxe6 27.Be4!±

Ironically, after the bishop on h7 was Black’s problem piece for such a long time, White makes
sure to exchange it at the first opportunity! In this way, he gets closer to Black’s numerous pawn
weaknesses, and Black has a difficult road ahead.

B2) 5...a6

Just like the main line covered in Chapter 10, Black prevents Nb5, so it’s a logical alternative.
However, this variant has a markedly different character from the 5...g6 and 5...c6 lines covered

320
in the next two chapters. Here Black intends to play actively, perhaps with ...b5 followed by
...Bb7, or more likely with a quick ...Bg4, ...Nc6, and ...0-0-0, followed by ...e5 if permitted.
Obviously such ideas may lead to sharp, forcing play, which won Black’s set-up some fans in the
formative years of the 3...Qd6 Scandinavian. However, such concrete ideas are contrary to the
modern interpretation of the ...Qd6 systems. Nowadays, Black mostly chooses this opening
precisely to avoid a sharp, forcing battle in the opening. Thus its popularity has dipped a lot in
over-the-board play.

6.g3
It is useful to have the option of Bf4 gaining a tempo. Besides, the bishop will be well placed
on g2.

6...Bg4
This is deservedly the main line, although naturally I checked some alternatives as well:
6...Nc6?! is the wrong move order, as 7.Bf4 Qd8 8.Bg2 is pleasant for White. A possible
continuation is: 8...Bg4 9.0-0 e6 10.h3 Bh5?! (10...Bxf3 is the lesser evil but after 11.Bxf3 White
can clearly be satisfied with the outcome of the opening)

11.g4 Bg6 12.Ne5!± White has a large advantage. 12...Nd5?! has been played several times but
it only makes matters worse after 13.Nxd5 exd5 and now 14.c4!N+– is crushing.

6...b5 used to be one of the main lines of the ...Qd6 Scandinavian but its popularity has dwindled
since White’s best response became established. 7.Bg2 Bb7 8.0-0 e6 Black seems to have an
active game but in reality the ...a6/...b5 structure provides White with an easy target to attack, as
shown after: 9.Bf4 Qb6

321
10.a4! Bd6 11.Be3! Black has serious difficulties, and I will not spend any more time on this line
as nobody plays it anymore.
6...g6 aims for a set-up resembling the next chapter, where the early g2-g3 is not conducive to
some of White’s more aggressive ideas against the fianchetto set-up. But of course, the early ...a6
is hardly the most useful way for Black to spend a tempo, so White should have no trouble
securing a slight advantage. 7.Bg2 Bg7 8.0-0 0-0 9.Bf4 Qd8 10.Qd2 Black has tried a few
different moves here but I don’t see much value in detailed analysis, as White should maintain a
lasting edge with mostly natural moves. We will follow one model example: 10...c6 11.Rfe1 Bg4
12.Ne5 Be6 13.Rad1 Nbd7

14.Qc1 (14.Nxd7!?N 14...Qxd7 15.Na4 Qd8 16.Rxe6! fxe6 17.Nc5 also looks extremely
promising) 14...Nb6 15.Bh6 Qc8 16.Bxg7 Kxg7 17.Ne4 Bh3 18.Nc5± Here Black blundered a
pawn with 18...Bg4?? and resigned immediately in Asrian – Kurajica, Istanbul 2003, but his
position was already distinctly unpleasant.

322
A final alternative is:
6...Bf5 7.Bg2 Nc6?!
This attempt to play actively will not end well for Black.
7...c6!? is a more solid approach, whereby Black essentially writes off ...a6 as a lost tempo
in order to make a firm barrier against the bishop on g2. However, 8.0-0 e6 9.Nh4!N
ensures White of some advantage, for instance: 9...Bg4 (9...Bg6 10.Nxg6 hxg6 11.Re1 is
excellent for White, thanks to his two bishops and nice pawn structure) 10.f3 Bh5 11.g4
Bg6 12.f4‚
8.0-0 0-0-0
White has a few good options, the strongest of which has not yet been tested in practice:

9.Ng5!N 9...Bg6 10.d5 Nb4 11.Bf4 Qc5 12.Na4 Qa5 13.c4

Black is in big trouble, for instance:
13...e6
13...Bc2 loses to 14.Qd4 Bxa4 15.Nxf7, and if 15...Rg8 then 16.Qa7 Kd7 17.Bh3† Ke8
18.Nxd8 Kxd8 19.Qb8 is mate.
14.a3 Nc2 15.b3 Nxa1 16.b4+–
White wins material.

323
7.h3!
It is useful to force Black to make an immediate decision about his light-squared bishop. If he
gives it up, then his hopes for any quick activity with ...Nc6 and ...0-0-0 are as good as dead. If
he keeps it alive with ...Bh5, then he runs into some other problems that would not have occurred
if White had not chased the bishop.

7.Bg2 Nc6 leads to a huge amount of theory, and while Black’s position seems on the verge of
falling apart in some of the ensuing lines, I could not find a sure-fire route to an advantage.

We will analyse B21) 7...Bxf3 and B22) 7...Bh5.

B21) 7...Bxf3 8.Qxf3

This just leads to a typical scenario where White enjoys more space plus the bishop pair. Of

324
course White players often manage to fumble away the slight edge in such positions, so you may
find it useful to familiarize yourself with some plans. Out of various possibilities, I recommend a
simple plan involving Be3 and 0-0-0 followed by expanding on the kingside by advancing the g-
and h-pawns. Such ideas are common against the Sicilian, and here White has more control over
the rest of the board. Black can shelter his king on the queenside, but even then White’s space
advantage should give him some targets.

8...c6
8...Nc6?! seems worse, as Black has no meaningful pressure against the centre and 9.Be3 0-0-0
10.Bg2 leaves his knight pinned.

9.Be3
Adventurous players may wish to investigate 9.Bd2!? Qxd4 10.0-0-0 or 9.Bf4!? Qxd4
10.Bd3!?N, but there is no real need for such antics.

9...e6 10.0-0-0 Nd5

White also has a pleasant game after:
10...Nbd7 11.g4
Sticking to the same plan.
11...Be7
11...0-0-0? is impossible due to: 12.Bf4 Qb4 13.Bxa6!+–
This position was reached in Nasuta – Tavoularis, Newcastle 2018. White has various ways
to maintain the upper hand, but I suggest sticking to the plan outlined earlier:

12.h4!?N 12...0-0
After 12...Qc7 13.g5 Nd5 then 14.Nxd5!? is perhaps even easier than the usual Ne4.
14...cxd5 15.h5² White is already poised to create kingside threats with g5-g6.
13.g5 Nd5 14.Ne4

325
 Here we should avoid 14.Nxd5 Qxd5 since, even though the endgame might still be a tad
better for White, it is obvious that we can pose Black more serious problems with a
middlegame attack.
14...Qc7
14...Qb4 15.c3 Qa4 16.Kb1 leaves Black in no position to create meaningful threats on the
queenside.

15.c4!? Nxe3 16.fxe3²

The ...c5 break is firmly under control, because White can simply answer it with d4-d5.
Meanwhile, White’s imposing kingside pawns should help to generate a lot of attacking chances.

11.Ne4 Qc7 12.h4

White should not worry about ...Nxe3, as fxe3 will strengthen his centre while leaving
opposite-coloured bishops, which could add potency to our attack if we can target the light
squares.

326
 For this reason, 12.Bd2 is unnecessary and Black can switch to preparing ...c5. For instance:
12...b5 13.g4 (13.Bd3 Nd7 14.Rhe1 Be7 15.h4 0-0 16.Ng5 ooks like a better bet, but I still prefer
leaving the bishop on e3 in the first place) 13...Nd7 14.g5 Rc8 Already, ...c5 is a strong threat.
Of course, g4-g5 was not the best plan in the last couple of moves, but I would like to be able to
stick to that plan if possible.

12...Nd7 13.g4 Nxe3

13...c5 14.g5² is also not easy for Black.

14.fxe3

14...Be7 15.g5 Rf8

15...0-0 16.Bd3 is obviously dangerous for Black.

16.Qh5!
Black cannot save his h-pawn.

16...h6
16...0-0-0?! 17.Qxh7 Rh8 18.Qxg7 Rdf8 sees Black trying to force a draw by perpetually
attacking the queen, but White can easily avoid this with:

327
19.Nd2 Qg3 20.Bd3 Rxh4 21.Rxh4 Qxh4 22.Ne4± Preserving an extra pawn.

17.gxh6 gxh6 18.Qxh6 0-0-0 19.Qh5²

This time the extra pawn is a bit less secure, but Black still has a hard time winning it back.

B22) 7...Bh5

This is the main line and the most ambitious choice. However, as mentioned earlier, White can
use the insertion of these moves to his advantage.

8.Bg2
8.g4 Bg6 9.Ne5 Nc6! gives Black good play.

328
8...Nc6
8...c6 is not so much in the spirit of Black’s earlier play, although objectively it limits White’s
advantage to a more manageable level. 9.0-0 e6 10.Re1 Be7 11.Bf4 Qd8 Now in Sosovicka –
Orgon, Liptovsky Mikulas 2019, a good way to organize White’s position would have been:

12.Ne2!?N 12...Nbd7 13.c4 (13.g4 Bg6 14.Nh4 is also possible) 13...0-0 14.g4 Bg6 15.Nh4 Re8
16.Nxg6 hxg6 17.a3 Rc8 18.Nc3² With two bishops and a useful space advantage.

9.d5!
9.0-0 converts to a version of the 7.Bg2 lines which I found to be less convincing.

Somehow, it took a long time for players to realize how powerful the text move is, despite the
existence of some correspondence games from 2006/07 showing its effectiveness. For instance,
Houska does not mention it in her 2009 book Starting Out: The Scandinavian. It has still not
been tested in many practical games, possibly because most players have not studied these lines

329
in enough detail to be aware of it. White has scored exceptionally well in the games where it has
been played, and with good reason. It was vital to include h2-h3 and ...Bh5 here, because
otherwise Black could just simplify the position with ...Ne5.

9...Nb4
9...Ne5?! is no good here due to 10.g4! Bg6 11.0-0 Nxf3† 12.Qxf3 followed by Bf4 when it
looks practically impossible for Black to complete development.

9...Bxf3N 10.Bxf3 Qe5† (10...Ne5 11.Bg2±) 11.Kf1 0-0-0 12.Kg2± is also great for White.

10.Bf4 Qc5 11.Be3 Qd6

Going back and forth with the queen is hardly ideal for Black, but he has nothing better.

The alternative is:

11...Qa5 12.0-0 Nbxd5 13.Nxd5 Nxd5?!
13...Qxd5 is the lesser evil, when 14.Qe2 transposes to our main line below.
14.Bd2

14...Qc5N
14...Qa4 occurred in Calistri – Chabanon, Clichy 2006, when 15.g4!N 15...Bg6 16.b3 Qa3
17.c4+– would have given White a decisive advantage.
The text move is a better try, but Black is still in deep trouble after:
15.c4! Qxc4 16.g4 Bg6 17.Rc1 Qxa2 18.Ne5 c6 19.Nxg6 hxg6

330
20.Qc2! Nb6
20...e5 allows 21.Ra1+– trapping the queen.
21.Bxc6†+–
And White wins.

12.Qe2! Nbxd5 13.Nxd5 Qxd5

13...Nxd5N 14.0-0-0 Qe6 15.Qd2 c6 16.Rhe1ƒ gives White tremendous play for a mere pawn.

14.0-0 c6
14...e6 15.g4 Bg6 occurred in Hartl – Stagl, Velden 2010, when White could have obtained a
decisive advantage with:

331
16.Nd2!N 16...Qb5 (16...Ne4 17.Bf4+–) 17.Qxb5† axb5 18.Bxb7 Rb8 19.Bc6† Kd8 20.Ba7+–

I also considered 14...Qb5N 15.Qxb5† axb5 16.Ne5 when the disappearance of queens does
little to temper White’s initiative, as shown after: 16...0-0-0 (16...Rb8 17.a4 bxa4 18.Ba7! Ra8
19.Rxa4+–)

17.a4! b4 (17...bxa4 18.Rxa4 makes it impossible for Black to defend the queenside) 18.a5 Rd6
19.c3! bxc3 20.bxc3+– Black has no good defence against Rfb1 and so on.

332
15.g4!N
15.Rad1 Qa5 16.g4 Bg6 was the move order of the Dothan – Spitz game quoted below, but
Black could have considered 15...Qb5N 16.c4 Bxf3 17.Bxf3 Qe5 when his survival chances are
better than in the game.

15...Bg6 16.Rad1 Qa5

With this, we re-join the game.
17.Bf4

17...Nd5
17...e6 18.Ne5 also gives White an overwhelming initiative.

18.Bg3 e6 19.c4+–

333
 Black’s position was already beyond salvation in Dothan – Spitz, corr. 2007.

Conclusion

3...Qd6 can be considered the modern main line of the Scandinavian, although the lines covered
in this chapter are not particularly in vogue at present. After 4.d4 we first looked at 4...c6 5.Nf3
lines where Black delays ...Nf6. It is hard to see how he can benefit from such ideas, especially
considering that 5...Bg4 6.h3! works out so well for White if Black gobbles the d4-pawn.

4...Nf6 is normal, and after 5.Nf3 we looked at two of the four main branches (the other two
being reserved for the upcoming chapters). Firstly, 5...Bg4 should be met by 6.h3, when White
gets a strong initiative regardless of whether Black exchanges on f3 or retreats his bishop. It’s
amazing how often White is able to sacrifice the d-pawn with powerful effect in these lines.

Finally, 5...a6 6.g3 Bg4 seems like the most reliable of the options examined in this chapter,
although 7.h3! once again presents Black with a difficult choice. 7...Bxf3 is relatively safe, but
White gets an easy game thanks to his bishop pair and the easy plan of advancing his kingside
pawns. 7...Bh5 is the move Black would like to play but, not for the first time in this chapter,
White gets a great game with a timely d4-d5 pawn sacrifice.

334
A) 7...a6!? 8.Bf4 Qd8 183
A1) 9.0-0-0 184
A2) 9.Be2!? 186
A21) 9...0-0 186
A22) 9...b5 188
B) 7...0-0 8.0-0-0 191
B1) 8...Bf5 192
B2) 8...Bg4 9.Be2 193
B21) 9...c6 194
B22) 9...Nbd7!?N 195
B3) 8...a6 9.Bf4! 197
B31) 9...Qd8 10.h4! 197
B311) 10...Bg4 198
B312) 10...h5 199
B32) 9...Qb6! 10.Ne5!? 200
B321) 10...Nc6 201
B322) 10...Nbd7N 202

335
1.e4 d5 2.exd5 Qxd5 3.Nc3 Qd6 4.d4 Nf6 5.Nf3 g6
Recently, this has become the trendiest version of the ...Qd6 Scandinavian. The main appeal of
this move is its flexibility. It doesn’t really commit Black to a specific plan (apart from
developing the bishop to g7 obviously) – he may revert to ...a6, followed by either ...Nc6 to
attack d4, or ...b5 and ...Bb7; alternatively, he may develop with ...Bf5 preparing ...Ne4, or even
...Bg4xf3. The text move also gives White a wider array of options than the other main lines.
This might seem like a drawback, but I actually consider it a positive feature of Black’s set-up.
The multitude of options can easily confuse White players, even during preparation, as there is
no well-established theory to follow. With so many choices, White has a lot of room to go
wrong, whereas Black does not need to know too much theory to get through the opening and
reach a playable position from which he can look to outplay his opponent, regardless of any
engine assessment of +0.2 for White after the opening, or whatever it might be.

So how should we meet Black’s last move? If White continues meekly with something like
6.Be2 followed by 0-0, then Black will finish his development rather fast, and White’s pieces
don’t seem sufficiently well placed to claim any advantage. The most popular continuation has
been 6.Nb5, and in the event of 6...Qb6 White can either play 7.c4 or regroup with 7.Na3
intending Nc4. This set-up has certain merits, as the knight was far from ideally placed on c3,
but I don’t like the aesthetics of it – we are still far from completing development, and spending
time moving the knight around seems to play right into Black’s provocations.

Instead, I favour a simple yet aggressive plan of preparing 0-0-0. Even though I mentioned that
...g6 isn’t very committal in terms of plans, it does seem rather likely that Black will play ...Bg7
and ...0-0 in the near future. This naturally makes me curious about exploring ideas involving
long castling and attacking with h4-h5. With such plans, the knight on c3 seems satisfactorily
placed as well. All that remains is to decide on the best way to develop the dark-squared bishop,
which might go to g5 or e3. The two options are quite similar, as after Qd2 White can consider
Bf4 to kick Black’s queen off to a less comfortable square, and/or Bh6 to trade bishops.
Ultimately, I slightly prefer:

6.Bg5!?
Initially, I felt 6.Be3 might make more sense because it overprotects d4, and ...Ng4 will never
be much of a worry. However, when I looked a bit deeper I realized it offers Black a couple of
important additional options. For instance, after 6...Bg7 7.Qd2 0-0 8.0-0-0 Black can play
8...Nd5!; and there are also ideas of ...Bf5 and ...Ne4 to consider. Either way, Black finds it easy
to exchange a pair of knights, which gives the rest of his pieces more room in the centre and on
the queenside. As we will see, the bishop on g5 makes it harder for Black to carry out such ideas.

336
6...Bg7
An interesting alternative is:
6...Bf5!?
As played in T. Kosintseva – Zhao Xue, Beijing (rapid) 2014. White’s best response is:
7.Nb5!N
An unexpected way to exploit Black’s unusual method of development.
7.Qd2 Ne4! would be just fine for Black.
7...Qb6
7...Qe6†? 8.Ne5 Na6 9.Bc4 Nd5 10.g4+– exploits Black’s poor coordination.
8.Bxf6 exf6
The change in the pawn structure ...exf6 is not ideal for Black, but normally her two bishops
would compensate for it. However, she also faces problems completing development with
her king exposed in the centre.

337
9.Qe2† Be7 10.0-0-0 a6
10...Kf8 11.Nc3 Qe6 12.Qd2 Kg7 13.Nh4² sees White eliminate one of the bishops while
keeping the better structure.
11.Nc3 Qe6 12.Qd2²
Black has yet to castle, while Nh4 is an annoying threat. If White can exchange one of the two
bishops, Black will have no real compensation for her inferior structure.

7.Qd2
We will examine these two main options: A) 7...a6!? and B) 7...0-0.
7...Bf5 should be met by 8.0-0-0 and since 8...Ne4?? isn’t possible because of 9.Nxe4 Bxe4
10.Re1, I see nothing better for Black than 8...0-0, which converts to variation B1 below.

7...h6 seems too committal – White often wants to play Bf4 anyway at some point in these lines,
and now Black will not be able to castle for the foreseeable future. After 8.Bf4 Qd8 9.0-0-0 c6
10.Bc4 Nbd7 11.Rhe1 Nb6 an interesting new idea is:

12.Bxf7†!?N It is hard to resist this move, although it is certainly not essential. (12.Bb3 Nbd5
13.Nxd5 Nxd5 14.Be5 0-0 15.Bxg7 Kxg7 16.Ne5² saw White maintain an edge by simpler
means in Strikovic – Gashimov, Ourense 2009) 12...Kxf7 13.Ne5† Kg8 14.Nxg6 Nbd5
15.Nxd5 Nxd5 16.Nxh8 Nxf4 17.Qxf4 Kxh8 18.Rd3²

A) 7...a6!?

This has not been played so often, but delaying ...0-0 seems like a logical way to counter White’s
plan of attacking on the kingside.

8.Bf4!

338
 It seems best to force a commitment from Black’s queen before deciding what to do next.

8.h4 plays into Black’s hands: 8...b5 9.h5 (9.Bf4 is met by 9...Qb6! when the queen is
excellently placed behind the b-pawn) 9...b4! (9...Nxh5 10.d5 would be more to White’s liking)
10.Na4 Ne4÷

8...Qd8
The alternative is:
8...Qb6
As played in Kovalevskaya – Kurajica, Solin 2005. White should answer with:

9.Ne5!N
9.0-0-0 Bg4! 10.Be2 Nc6 followed by ...0-0-0 would work out perfectly for Black.
9...Nbd7
9...Qxb2?? 10.Rb1 Qa3 11.Nc4 catches the queen.
9...0-0 10.0-0-0 converts to variation B32, and 9...Nc6 10.0-0-0 0-0 is variation B321.
10.0-0-0
We have transposed to variation B322 on page 202.

For a long time I tried to get the principled plan with A1) 9.0-0-0 followed by h2-h4 to work.
This certainly leads to interesting play, but in some sense it is what Black is best prepared to deal
with, because he has not committed to ...0-0 yet. So finally, I opted for the more subtle A2)
9.Be2!? as my main recommendation.

A1) 9.0-0-0

339
9...b5
This is most consistent with the early ...a6, although Black also has the option of transposing to
variation B31 with 9...0-0.

10.h4 Bb7
10...Bg4?! prevents h4-h5 for the time being but it does not combine well with the ...a6/...b5
plan. White simply plays 11.Be2 intending Ne5, with a dominating position.

10...h5 weakens the kingside. 11.Ne5 Bb7 12.Bd3! is a strong reply, intending 12...Nc6
13.Nxg6 fxg6 14.Bxg6† Kf8 15.Rhe1© when Black’s position seems rather unpleasant.

11.h5!?N
This seems like the most interesting try.

340
11.Ne5 Nbd7 led to an eventual victory for White in Sychev – Golubov, Moscow 2018, but
Black’s play can be improved. I considered 12.f3!?N, just preparing for a subsequent kingside
expansion while waiting for Black to commit his king. However, after 12...0-0! Black seems to
be ready to withstand any immediate attack. The position contains much to explore, but I failed
to find anything special for White.

My next attempt to improve on the above game was 11.d5N, intending to meet 11...Nbd7 with
12.h5. However, Black can avoid this with 11...h5! 12.Ne5 0-0 when the position is ripe for
exploration, but again I could not find anything convincing. This led me to investigate the text
move, whereby White launches his attack before Black has a chance to block with ...h5. As we
will see, White can soon follow up with d4-d5, reaching the position I originally had in mind.

11...Nxh5 12.Rxh5!
This is the only way to justify White’s previous move.

12...gxh5
Inserting 12...Bxf3?! 13.gxf3 before 13...gxh5 hardly seems advisable for Black. Giving up the
light-squared bishop is a significant concession, while the doubled f-pawns do not harm White
too much – perhaps the rook will even benefit from the open g-file. My analysis continues:
14.Bd3 c6 (14...Qxd4 15.Be3 Qd7 16.Bh6! Bxh6 17.Qxh6 c6 18.Qxh5±)

15.Ne4! Qxd4 (15...Nd7 16.Re1! 0-0 17.Ng3ƒ) 16.c3 Qd5 17.Qe3 Qe6 18.Bc2!ƒ With an
ongoing initiative for the sacrificed material.

341
13.d5 Nd7 14.Nd4
The position looks dangerous for Black and I was tempted to end my analysis here, but it’s
actually not easy to prove anything after:

14...Rg8! 15.Qe3!?
White has two other interesting tries:

15.Be2 b4 16.Ne4 Bxd5 17.Qe3 Nf6 18.Nc5 Qc8 reaches a complex position with lots of
options, but Black seems to be holding on as far as I could analyse.

15.g3!? b4! 16.Na4 Bxd5 17.Nf5 looks dangerous but Black can survive with:

17...e6! 18.Nxg7† Rxg7 19.Qxb4© Here too, White certainly has decent compensation, but
Black seems to have enough defensive resources.

342
15...Bxd4
15...Nb6 16.Nf5! Bxc3 17.d6! Bf6 18.dxc7 Qxd1† 19.Kxd1 Nd5 20.Qc5 Kd7 21.Bh2ƒ is
more dangerous for Black.

16.Qxd4 Nf6©
It would take a lot more analysis to reach a definitive conclusion. I feel that the dark-squared
bishop should be worth about as much as a rook, but it is hard to prove any concrete advantage.
If these kinds of positions appeal to you then you may still wish to consider playing this way, but
I decided in the end that it made more sense to recommend a more subtle approach, to which we
now turn.

A2) 9.Be2!?

343
 The main advantage of this move is that it avoids committing White’s king to either flank,
enabling us to choose whichever option will be most unpleasant against Black’s reply. Thus, if
Black commits his king to the kingside, the bishop on e2 will help to support h4-h5, or to prepare
g2-g4 if Black blocks with ...h5. On the other hand, if Black delays castling and commences his
queenside play with ...b5, we can switch to a positional scheme of exploiting the weaknesses
created by Black’s queenside advance.

Let’s see how these ideas unfold after A21) 9...0-0 and A22) 9...b5.

9...b6 makes little sense as White can continue with 10.0-0-0 followed by h2-h4, with a clearly
improved version of variation A1 above because ...b6 is a waste of time.

A21) 9...0-0

This position has occurred a few times via transposition. White’s best reply is:

10.0-0-0!N
We will see later in variation B31 that the early ...0-0 does not combine well with retreating the
queen on d8, as Black is too slow to create counterplay on the queenside. The same is true here.

10...c5
The other logical try is:
10...b5
With the queen on d8, this plan gives White no real problems.
11.h4

11...Bb7

344
 Black gains nothing from 11...b4?! 12.Na4± when White’s knight is comfortable and
Black’s queenside structure has lost its fluidity.
11...h5 is met by 12.Ne5 Bb7 13.Rhg1ƒ with g2-g4 to follow.
11...Bg4 12.Ne5 Bxe2 13.Nxe2ƒ also looks great for White.
12.h5ƒ
White’s attack proceeds uninhibited.

11.d5!
By keeping the centre blocked, we give ourselves the necessary time to launch a kingside
offensive.

11.dxc5 Qa5 makes it harder for White to launch something on the kingside, and I don’t really
believe in White’s prospects of keeping an edge by other means.

11...Nbd7
11...b5 is met by 12.Ne5! Bb7 13.Bf3 and because of the possibility of Nc6, Black finds it
hard to develop any further.

12.h4! h5
12...b5 13.h5 b4 14.Na4 Nxh5 15.Bh6 Ndf6 16.Bxg7 Nxg7 17.Ne5ƒ gives White a promising
attack.

12...Ng4 can simply be ignored: 13.h5! Nxf2

14.Bh6! Nxh1 15.Rxh1 Bf6 16.Ne4 With a crushing attack.

345
13.d6!
13.Ng5!? followed by g2-g4 leads to some crazy fun, but the text move is stronger.

13...e6 14.Ne5 Nxe5 15.Bxe5 b5

15...Ne4? is impossible due to 16.Nxe4 Bxe5 17.d7! and White wins; this is why d5-d6 was
useful.

16.Bf3 Rb8 17.g4ƒ

White has excellent chances of success on the kingside.

A22) 9...b5

Again, this position has occurred via transposition. As mentioned earlier, our general plan is to

346
castle on the kingside and attack on the queenside, but it is important to go about it in the most
accurate way possible. I found a remarkable novelty:

10.b4!N
If White plays a premature a2-a4, either immediately or after castling, Black simply replies
with ...b4 when we have to waste time retreating the knight, losing time and coordination.
Therefore it makes sense to fix the pawn on b5.

10.0-0N looks obvious and natural, but after 10...0-0 11.b4 Black has the annoying 11...a5!
12.Nxb5 (12.a3 c6 is okay for Black, whose pieces can target the d5- and c4-squares) 12...c6
with an unclear mess. This would clearly not be an option if the knight could go to c7 with
check, which leads to the text move.

10...0-0
10...c6 11.a4 0-0 transposes to the main line below.

10...Nc6 11.a4! Nxb4 (11...bxa4 12.Rxa4±) 12.axb5 0-0 transposes to the following note on
11...Nc6.

We should also consider:

10...a5?! 11.Nxb5 Nd5
Black gives up a pawn in order to plunge the game into chaos. Objectively this would be a
mistake, but White has to play accurately and energetically to prove it.

347
12.Bh6 Bxh6 13.Qxh6 c6
13...axb4? 14.Qg7 Rf8 15.Ng5 gives White a decisive attack.
14.Qg7!
14.Ng5!? should lead to an advantage with precise play but the variations are just as
complicated, and the text move leads to greater rewards.
14...Rf8 15.Ng5 Nf6
At first glance it may appear that things have gone wrong for White, as Black is avoiding an
immediate disaster on the kingside while the b5-knight and b4-pawn are in trouble.
However, White can exploit his dynamic advantages with:

16.0-0-0!!
Now matter how Black responds, his position breaks down. The chances of getting this
position on the board are pretty low, so I would not recommend trying to memorize any
lines from here, but you may find it interesting and enjoyable to play through some of the

348
 moves just to see how all of Black’s defensive tries come unstuck.
16...h6!?
16...cxb5 17.Bxb5† Bd7 18.Bc4 e6 19.d5 Nh5 20.Qh6 Qf6 21.Qxf8† Kxf8 22.Nxh7† Kg7
23.Nxf6 Kxf6 24.b5† leaves White with a winning endgame.
16...Bf5 17.Rhe1 axb4 18.g4 Rxa2 19.gxf5 b3 and now 20.Bc4!! is a lovely idea, giving up
a piece in order to open the e-file, divert Black’s rook and buy time for White’s attack to
break through, as shown after: 20...Rxc2† 21.Kb1 Rxc4 22.Qxf6 cxb5 23.Nxh7 Rg8
24.Ng5 Rf8 25.fxg6+– Black’s defences are broken.
16...Be6 17.d5!! makes full use of the rook on d1. My main line continues 17...Bxd5
18.Rhe1 Nbd7 19.Nd6†! exd6 20.Bb5† Ne5 21.Bxc6† Bxc6 22.Rxe5† dxe5 23.Rxd8†
Rxd8 24.Qxf6 Ba4 25.Nxh7 and White wins.
The text move is perhaps the trickiest attempt, but White can once again break through with
some flashy tactics.
17.d5!! cxd5
17...hxg5 loses more quickly after: 18.dxc6 Qb6 19.Rhe1 Be6 20.Bd3! Nxc6 21.Rxe6+–

18.Nh7!! Nxh7 19.Rxd5!

Once again, the d-file plays a key role in White’s success.
19...Nd7 20.Qc3 e6 21.Nc7† Ke7 22.Rhd1!
Bringing the last piece into play.
22...exd5 23.Nxd5†

349
23...Kd6
Black has no choice, as 23...Ke8 24.Qe3† wins easily.
24.Bb5!?
My favourite of several winning moves. The finish might be:
24...Ra6 25.Bxa6 Bxa6 26.Ne3† Ke7 27.Nf5† gxf5 28.Qe3† Kf6 29.Qxh6† Ke7 30.Qd6† Ke8
31.Re1†+–
With mate to follow.

11.a4 c6
11...bxa4 12.Nxa4² leaves Black with no real compensation for his queenside weaknesses.

11...Nc6 12.axb5 Nxb4 is an attempt to liquidate the queenside, but White can thwart this plan
with: 13.0-0 Bb7

350
14.Rfb1! (but not 14.bxa6 Bxa6 15.Bxa6 Nxa6 when White’s edge has all but vanished, as Black
is ready to play ...c5 and get rid of his final weakness) Now 14...a5 is forced, and after 15.Be5²
Black’s a-pawn is permanently weak.

12.0-0
Taking the pawn on b5 seems somewhat premature – perhaps White can somehow consolidate
and keep his extra pawn, but I wouldn’t feel comfortable opening up all these lines without even
castling. Besides, Black has a rather passive position, and trying to take away his active plans
seems a much more appropriate way to continue.

12...Bb7 13.a5!
13.Ne5 a5! gives Black some much-needed activity.

351
There were some other ways to prevent ...a5, but White didn’t really have any pressure on the
b5-pawn, so why leave the pawn on a4? Besides, the pawn on a5 prevents Black from activating
his b8-knight with ...Nbd7-b6.

13...Nbd7
White doesn’t have to be in any particular hurry now: as long as the ...c5 and ...e5 breaks are
prevented, it will be hard for Black to obtain any counterplay. White can manoeuvre his pieces
around in various ways, such as Ne5-Nd3, Bf3 and so on. Black’s most obvious idea is ...Nd5,
but White can simply meet it with Ne4 – the knight would then be headed towards the ideal c5-
square, while giving up the dark-squared bishop is no big deal for White.

14.Rad1!
This is a useful prophylactic move directed against any surprising ...c5 ideas – and the rook
seems perfectly natural here anyway.

The point of the text move can be seen when we compare the natural-looking alternative:
14.Rfe1 Nd5 15.Ne4 c5! 16.Nxc5 Nxc5 17.bxc5 Nxf4 18.Qxf4 Rc8 White is a pawn up for
now, but Black suddenly has much more active bishops and his compensation may well be
enough.

14...Nd5
14...e6 15.Ne5²

15.Ne4! Nxf4 16.Qxf4²

White remains in control.

B) 7...0-0

352
8.0-0-0
We continue with the plan outlined at the start of the chapter. Black’s three most important
options are B1) 8...Bf5, B2) 8...Bg4 and B3) 8...a6.

8...Nd5N was a problem in the analogous position with the bishop on e3, but here it can be met
by 9.Ne4 because 9...Qb4? would simply blunder away the e7-pawn.

8...c6 9.Bf4 Qd8 was played in Manuel – Pagente, Manila 2013, and now 10.h4!N looks to me
like an improved version of variation B31 for White. The different is that here Black has played
...c6 instead of ...a6, which does not fit in so well with plans such as ...Bg4 followed by ...Nc6 or
...c5, as well as ...b5 ideas.

8...c5N
This is another natural move to check, but it allows a typical response from White:
9.d5
9.Bf4 Qd8 10.dxc5 Qa5 could be analysed further but, after the centre gets opened, I would
be less optimistic about the h4-h5 plan working as desired – and without it, I am not sure if
White can hope for much of an advantage.
9...a6

353
10.Bf4!
It is useful to kick the queen away before ...b5 is played, so that ...Qb6 isn’t really an
option.
10...Qd8 11.h4 h5
11...b5 12.h5 Nxh5 13.Bh6 gives White a powerful attack.
12.Be2 b5 13.Ne5
Black is barely able to develop, and g2-g4 will come soon.

B1) 8...Bf5

9.Bf4!
This move is well timed, as finding a good square for the queen isn’t easy at all. If it goes to
d8, Black will have a hard time putting much pressure on d4. And if the queen goes to b6, Black
will have to worry about possible knight jumps to c4 or even a4.

354
9.d5 c6 10.Nd4 Bc8! seems okay for Black.

9.h4 is another natural try, but after 9...Rd8! 10.h5 gxh5! (rather than 10...Nxh5 11.Rh4ƒ) Black
is surprisingly safe.

9...Qb6
I also checked: 9...Qd8 10.Bh6 Nbd7!?N (this seems like a logical attempt to improve on
10...Nc6 11.Bxg7 Kxg7 12.d5± as played in Smolkov – Dimov, Sunny Beach 2015)

11.h4 Bxh6 12.Qxh6 Ng4 13.Qd2 h5 14.d5ƒ Intending Nd4, f2-f3 and g2-g4.

10.Ne5 Nc6
10...c5 11.d5 Nbd7 is well met by:

355
12.Nc4! Qd8 13.f3 Intending to advance the g- and h-pawns.
10...Nbd7 11.Na4! exploits the unfortunate placement of the queen: 11...Qe6 12.Bc4 Nd5 13.g4!
Be4 14.Rhe1+– Winning material.

11.Na4!N
Improving on 11.Nc4 Qa6! 12.Ne3 Qa5 when Black was doing just fine in Davidavicius –
Fels, corr. 2013.

11...Qb4 12.Nxc6 Qxd2† 13.Rxd2 bxc6 14.Bc4! Ne4

14...Rac8?! 15.Re1± gives White an ideal set-up.

14...Be4 15.f3 Bd5 16.Ba6! leaves Black’s bishop somewhat stuck.

14...Nd5 should be met by 15.Bxd5! cxd5 16.Nc5² when Black’s light-squared bishop has no
prospects, and we can continue targeting it with f2-f3, g2-g4 and so on. If White can exchange
the dark-squared bishops, then it will be perfect, as the knight-versus-bishop endgame will be
more than comfortable. As long as Black has two bishops, he has decent chances to hold, but his
position will remain mostly passive.

15.Re2 Nd6 16.Bb3 Bxd4 17.Rxe7 Bxf2 18.Rf1

356
18...Bb6
18...Bh4 19.Rxc7 Nb5 20.Rxc6 Nd4 21.Rc4 Ne2† 22.Kb1 Nxf4 23.Rcxf4 Bg5 24.R4f2 sees
White maintain an extra pawn.

19.Nxb6 axb6 20.Rd1 Be6 21.a3 Bxb3 22.cxb3

Black is still struggling to save all the queenside pawns.

B2) 8...Bg4

This move has yielded decent results for Black in the few games in which it has been played.
Trading off the light-squared bishop is a common idea across many Scandinavian lines: White
has to waste some time to eliminate it, in this case, with Be2 and h2-h3; and once our bishop
arrives on f3, it can be restricted by a pawn on c6. At the same time, it is also hard for Black to

357
find much counterplay, since the typical plan of ...c5 makes no sense without the light-squared
bishop. At first, I did not think much of this move, as I assumed Black would not have much
counterplay for the reason just described, and that White would be able to expand on the
kingside at his leisure. However, Black does have some interesting ideas connected with timely
knight manoeuvres like ...Nd7-b6-c4 or ...Nd5.

9.Be2
We will start by analysing B21) 9...c6, which has been played in a few games, before moving
on to the more flexible B22) 9...Nbd7!?N.

B21) 9...c6

10.h4!N
10.Ne5 Bxe2 11.Qxe2 allows 11...Nd5! when the simplifications help Black.

10.h3
This is the most obvious move but Black seems to be okay after:
10...Bxf3 11.Bxf3 Nbd7
Now the ...Nb6-c4 plan becomes surprisingly annoying.
12.Bf4
12.h4 can safely be met by 12...h5, and if 13.Qe2 Rfe8! 14.g4 e5 Black’s central
counterplay comes in time.
I also considered 12.g4 Nb6! 13.Qe2 Rfe8 14.h4 e5 and again Black is fine.
12...Qb4 13.a3
13.g4 Nb6 14.a3 Qa5! 15.h4 Nc4 16.Qd3 Nd5! 17.Nxd5 cxd5 turns out nicely for Black.

358
13...Qa5!N
13...Qb6 gives White an easy advantage by advancing on the kingside. 14.g4!N seems
simplest to me. (14.Bh6 also gave White a more pleasant position in Grigoriants – Kruchev,
Voronezh 2014, but I don’t see any reason to exchange the dark-squared bishops just yet)
14...Rad8 15.g5 Ne8 16.h4 White is completely dominating, and 16...Nd6 can be met by:
17.Bxd6 exd6 18.h5ƒ
14.Rhe1
14.h4 and 14.Kb1 both allow 14...e5! with good counterplay.
14.Qe2 e6 15.g4 Nd5÷ is also decent for Black.
14...e6
White’s last move prevented ...e5, but Black can play on the light squares as well.
15.g4 Nb6÷
Keeping the ...Nc4 threat alive.

10...Nbd7
10...h5 can be met by 11.Ne5 Bxe2 12.Qxe2 with good chances to open the kingside, for
instance: 12...Re8 (12...Nd5?! 13.Ne4±; 12...Nbd7 13.Nc4 Qb8 14.Qxe7²) 13.g4ƒ

11.Ne5! Nxe5
11...Bxe2 12.Qxe2 Qe6 (12...h5 13.g4ƒ) 13.h5 Nxh5 14.Nxd7 Qxd7 15.g4 Nf6 16.f3ƒ gives
White fine attacking chances.

12.dxe5 Qxd2† 13.Bxd2

359
13...Ne4
After 13...Bxe2?! 14.exf6 Bxd1 15.fxg7 Kxg7 16.Nxd1± White’s minor pieces should prevail
over the rook and pawn.

14.Bxg4 Nxf2 15.Bf3 Nxd1 16.Rxd1 Bxe5 17.Re1

A rook plus two pawns would usually be at least a match against bishop and knight, but here
White’s initiative comes into the equation as well.

17...Bf6 18.h5 Rfd8 19.Ne4ƒ

With such finely coordinated pieces, White is doing well.

B22) 9...Nbd7!?N

360
 This flexible move seems like a natural attempt to improve on the previous variation. Black
may play ...c6 if it is needed, but he also retains the useful option of ...c5, while also preventing a
quick Ne5 by White.

10.Kb1!
It is best to avoid committing the h-pawn for now. If it goes to h3, than Black could simply
exchange on f3 and play ...c6; and if White goes for an attack with h2-h4, then ...c5 offers Black
enough counterplay. Therefore we should keep both options open, and moving the king to b1 is
useful in all scenarios.

10.h3 Bxf3 11.Bxf3 c6 transposes to the Grigoriants – Kruchev game, as discussed in the note to
White’s 10th move in the previous variation, where it was shown that Black could have obtained
a fine position with a suitable improvement.
If 10.Bf4 then 10...Qb6! 11.Ne5 Bxe2 12.Qxe2 c5! undermines the centre with good effect.

10.h4!? c5! is another line showing why it was useful for Black to avoid committing to ...c6 too
soon. Play may continue 11.h5!? (11.Kb1 Rfc8 12.h5 Bxh5 does not seem like an improvement
for White) 11...Bxh5 12.Rh4 e6 13.Rg1 with a double-edged game; although this could be
interesting to analyse in more detail, I can’t help feeling that Black has made good use of the
flexible nature of his 9th move here, so I think we should reply with a flexible move of our own.
Now it is Black’s turn to come up with a move that will be useful in both the h2-h3 and h2-h4
scenarios.

10...Rad8
This seems as good a choice as any, although it is not the most useful move in the positions
after h2-h3. Here are some other options:

361
10...Nd5 11.Nxd5 Qxd5 12.h3 gives Black an unpleasant choice between retreating his bishop
and exchanging on f3.

10...Nb6 is met by 11.Bf4ƒ when the queen cannot go to the preferred b6-square as in the note
above.

10...c5 leads to a more open centre, and so 11.h3! Bxf3 12.Bxf3 should give White a lasting edge
with the two bishops.

10...c6 11.Bf4!? (11.h4!? also makes sense but the text move seems easier) 11...Qb4 12.Ne5
Bxe2 13.Qxe2 and White is ready to start rolling on the kingside. A sample continuation is:
13...Rad8 14.h4 Nxe5 15.Bxe5 h5 16.f3 b5 17.g4 Qc4 18.Qe3ƒ

Finally, 10...Bxf3 11.Bxf3 c6 leaves White practically a tempo up on the normal h2-h3 lines,
which obviously makes a difference. Among other things, 12.Bf4 Qb4 13.a3 is a good option,
when 13...Qa5 allows White to put the extra Kb1 move to concrete use with:

14.Nd5! Qd8 (14...Qxd2 is the lesser evil although 15.Nxe7† Kh8 16.Rxd2± leaves White a
clear pawn up) 15.Bc7 Qe8 16.Nxe7†! Qxe7 17.Rhe1+–

362
11.h3!
11.h4 would be the wrong idea as it invites 11...c5, when the rook on d8 will be useful.

11...Bxf3
11...Be6?! is met by 12.d5! Bf5 13.Bh6 when White is ready to start expanding on the
kingside, while Black has just wasted a bunch of tempos with his light-squared bishop.

12.Bxf3 c6 13.h4
13.Bf4 Qb4 14.Bc7!? is an attempt to trap the queen on b4, but it somehow manages to
wriggle out after 14...Rc8 15.a3 Qc4 16.Be2 Qe6 17.Bf4 Nb6 and the queen will settle back on
d7.

13...h5
13...Nb6 14.h5 Nc4 15.Qc1! shows the usefulness of the earlier Kb1. Meanwhile on the
kingside, White is dominating.

14.Rhe1
Simply centralizing and preparing Ne4, with g2-g4 to follow; there is no need to hurry the
attack.

14...Nb6
14...e6 15.Ne4 Nxe4 16.Rxe4± sees White’s plan develop smoothly.

15.Ne4 Nxe4 16.Rxe4 Rd7 17.g4±

White’s strategy is working out perfectly.

363
 B3) 8...a6

This has been Black’s most popular choice – not that there have been all that many games from
this position so far.

9.Bf4!
Good timing. If Black was given time to play ...b5, his queen would have a perfect square
available in b6, but now he must make a decision where either choice has a downside. We will
start by analysing B31) 9...Qd8, which has been Black’s usual choice so far, but which leaves the
queen passively placed. We will then move on to B32) 9...Qb6!, which seems more challenging,
even though ...b5 won’t be possible for a while.

B31) 9...Qd8

364
10.h4!
There is no reason to refrain from launching our attack. Black’s most logical replies are B311)
10...Bg4 and B312) 10...h5.

A good example showing the ease of White’s play continued 10...Re8 11.Bh6 Bg4 12.Be2 Nbd7
13.Bxg7 Kxg7 14.Ne5 Bxe2 15.Qxe2 e6 16.h5± when Black struggled to find any counterplay
in Saric – Mrdja, Mali Losinj 2016.

Black also has a hard time after:

10...b5 11.h5!
In the race between attacks, White has a clear head start.
11...Nxh5N
After 11...b4 White moved the knight to a4 in Yilmaz – Kurajica, Rijeka 2010, but
12.Ne2!N is an improvement, as the knight can play a useful role in the kingside attack. For
instance: 12...Qd5 (12...Nxh5? is worse in view of 13.Bh6 Qd5 14.Bxg7 Kxg7 15.Nf4
Nxf4 16.Qxf4 h5 17.Ne5+–) 13.Kb1 Nxh5 14.Bh6ƒ
12.Bh6 Bg4 13.Be2 b4 14.Bxg7 Kxg7 15.Ne4 Qd5

16.Ng3!
Simply ignoring the a2-pawn and focusing on the target.
16...Nd7
16...Qxa2 17.Nxh5† gxh5 18.Qg5† Kh8 19.Bd3ƒ
17.Nxh5† gxh5 18.Kb1 Nf6 19.Ne5ƒ
Black continues to have a hard time on the kingside.

B311) 10...Bg4

365
11.Be2 c5
I checked a couple of other options:

11...b5 rarely combines well with ...Bg4. After 12.Ne5 Bxe2 13.Nxe2 White’s pieces are well
placed for further action on the kingside, whereas Black has only created weaknesses on the
queenside.

11...Nc6 prevents Bh6 temporarily, but we can just prepare for it with: 12.Kb1! (12.Bh6? Bxf3
13.Bxf3 Qxd4! 14.Qxd4 Bxh6† is the tactical trick we must avoid; 12.h5!? on the other hand is
interesting, but there is no need for White to force things) 12...h5

13.Ne5! Bxe2 14.Nxc6 bxc6 15.Qxe2 Rb8 16.f3±

366
12.Bh6!N
12.h5!? Bxh5 13.Bh6 gave White a strong initiative in Darini – Miroshnichenko, Bandar-e
Anzali 2016, but the text move seems both stronger and more practical, as it involves fewer
concrete lines to analyse.

12...Nc6
12...Bxf3? 13.Bxf3 Qxd4 is too greedy, as 14.Qg5! leaves Black with insurmountable
problems.

The crude 12...Bxh6? 13.Qxh6 Bxf3 14.Bxf3 cxd4 also comes unstuck after 15.h5+– when the
attack is too strong.

Another nice line is: 12...cxd4 13.Bxg7 Kxg7 14.Nxd4 Bxe2

367
15.Nf5†! gxf5 16.Qg5† Kh8 17.Rxd8 Rxd8 18.Nxe2+–

13.Bxg7 Kxg7 14.d5 Bxf3 15.Bxf3 Nd4

16.h5! Nxf3 17.gxf3 Nxh5 18.Ne4ƒ

Clearly Black’s position is no fun, with ideas of Qc3†, Ng3 and Rxh5 all floating about.

B312) 10...h5

This way Black holds up the attack for now but White still controls the centre, and can aim for
a future g2-g4 break and/or pressure against g6.

11.Ne5 Nbd7!?N
This untested move seems like Black’s best try.

368
The obvious alternative is: 11...b5 12.Bh6 Bb7N (after 12...c5 13.Bxg7 Kxg7 14.dxc5! it was too
hard to get the c5-pawn back in Brkic – Plenkovic, Zagreb 2014) 13.Bd3 Nbd7

14.f3± White can consider g2-g4 next, or even transferring the c3-knight to f4, while it is not
clear how Black will get any counterplay. (White could also consider 14.Nxg6!? fxg6 15.Bxg6ƒ
but such ideas are not really necessary.)

12.Bd3
Not the only plan, but it seems good enough.

12...c5
12...Nxe5 13.dxe5 Ng4 14.Qe2 Qd4 15.Bg3 Qb6 16.Rhe1± is also great for White.

13.Nxf7!

369
 Often in these positions, I note that such possibilities are interesting but not really necessary.
However, this is one of those times when the flashy sacrifice is essential to make the most out of
White’s dynamic advantages.

13...Rxf7
13...Kxf7? 14.Bc4† catches the king.

14.Bxg6 Nf8
14...Rf8 15.Bh6 leaves Black’s pieces absolutely stuck, and after 15...e6 16.Rh3 cxd4 17.Qg5
Bxh6 18.Qxh6 Qe7 19.Rg3 Ng4 20.Bxh5+– the contest is over.

15.Bxf7† Kxf7 16.d5±

Materially, Black might be considered okay, but his weak king and White’s excellently placed
pieces leave us in no doubt that White is doing well.

B32) 9...Qb6!

This is much more challenging than the previous variation, yet it remains virtually untested – I
analysed it as a novelty, before a late variation check turned up an online game from April 2020.
Evidently, having just played ...a6, most players are too emotionally attached to the plan of ...b5
to consider blocking the b-pawn. However, we saw in the ...Qd8 lines that ...b5-b4 was not a big
deal, and White could comfortably start playing on the kingside. With the queen on b6, Black
threatens to create strong central pressure with ...Nc6, followed either by an immediate ...Rd8 or
by developing the light-squared bishop first, with a rook coming to the d-file shortly after. It
should also be noted that, despite the abandonment of the ...b5 plan, the ...a6 move is far from a
wasted tempo, because it provides a retreat square on a7 for the queen, which could be vital after
Ne5-c4 for example.

370
10.Ne5!?
Preparing Nc4 seems like the most principled continuation, despite the fact that the queen has a
ready-made escape square on a7.
10.h4?! plays right into Black’s hands, as 10...Bg4! 11.Be2 Nc6!„ creates powerful pressure
against d4.

I would also like to mention a less forcing alternative, which leads to a tense middlegame
situation:
10.Be2!?N 10...Nc6!
10...Rd8 11.Ne5! gives White good prospects, for instance: 11...Nbd7 (11...Be6 could be a
better try although I still like White after 12.Bf3 or 12.Rhe1) 12.Nc4 Qa7 13.Rhe1 b5
(13...c5? 14.dxc5 and the d8-rook is clearly misplaced: 14...Qxc5 15.Na4+–] 14.Na5 Qb6
15.Nc6!± White is doing great, as 15...Qxc6 runs into 16.Bf3.
11.Ne5 Be6!
It is not clear how White can best create problems for his opponent. It’s still a tense
middlegame position, but I don’t see an obvious way to avoid simplifications that would
make Black’s life easier. For instance:

12.Bf3
12.Nxc6 Qxc6 13.Bf3 Nd5 gives Black good chances to equalize through simplifications.
12...Nxe5 13.Bxe5 Qa5
13...c6 is also possible. Either way, Black intends ...Nd5 with further exchanges, and it is still
not easy to launch a kingside attack.

We will analyse B321) 10...Nc6 and B322) 10...Nbd7N.

10...c6?! seems too passive and 11.h4ƒ gives White easy play on the kingside, since Black is no
longer able to put d4 under pressure.

371
 B321) 10...Nc6

11.Nc4 Qa7 12.d5

Planning Be3 next. Black’s position doesn’t look pretty but he does have ...e6/...c6 and ...b5-b4
coming quickly. But still, the black pieces are too passive and White just needs to play normal
moves to maintain control of the position:

12...Nd8N
This is a natural improvement over 12...b5? 13.dxc6 bxc4, when White missed the most
accurate continuation:

14.Be3!N (14.Bxc4 Qc5 at least enabled Black to improve his queen, although he was still
clearly worse in D. Kopylov – Caspi, Internet 2020) 14...Qb8 15.Bxc4 White is close to winning:

372
he has an extra pawn and his pieces are much more active. An essential detail is that the
otherwise strong 15...Nxe4 can be met by 16.Bxf7†! when everything works out well for White.

13.Be2 b5

14.Be3! Qb8 15.Ne5 b4

15...e6 16.dxe6 Bxe6 17.f4² leaves White clearly more active.

16.Na4 Ne4 17.Qd4 Nd6 18.f4

White continues to control more space and Black has yet to find any serious counterplay.

18...Nf5
18...a5 19.h4 h5 20.g4ƒ gives us a strong attack.

18...f6 19.Nc6 Nxc6 20.dxc6ƒ and White’s activity is more relevant than the doubling of the
pawns.

18...e6 gives White more than one good option: 19.Bf3!? (19.h4!? Nf5 20.Qd2 exd5 21.h5ƒ also
looks tempting) 19...Qb5 20.Nc5

373
20...Nf5 This is less annoying than you might think, as after 21.Qd2 Nxe3 22.Qxe3± the knights
dominate.

19.Qd2 Nxe3 20.Qxe3 Qa7 21.Qxa7 Rxa7 22.Rhe1²

The loss of the dark-squared bishop does not really dampen White’s initiative, as the rest of his
pieces are clearly more active than their counterparts.

B322) 10...Nbd7N

Unlike the previous variation, Black abandons the idea of a piece attack on the d4-pawn in
favour of aiming to exchange it with ...c5.

11.h4!

374
 11.Nc4 does not carry much impact here, and 11...Qa7 12.Be2 c5„ gives Black decent play.

11.Be2 c5! again sees Black carry out his plan. After 12.Na4 Qa7 Black seems to have
successfully undermined White’s centre, and the h4-h5 plan is too slow.

11.f3 could be a slight improvement on the line above, but 11...c5 12.Na4 Qa7 again makes it
difficult for White to find an advantage. For instance: 13.Nxd7 (or 13.dxc5 Nxc5 14.Be3 Qb8!
and the tactics turn out okay for Black) 13...Nxd7 14.dxc5 Nxc5 15.Be3 b6 16.Nxc5 bxc5
17.Bh6 Bxh6 18.Qxh6 Qb6 The queen will come back to defend the kingside with ...Qf6.

11...c5
11...h5 weakens Black’s kingside structure, giving White some additional tactical prospects
after: 12.f3 c5 13.Na4 Qa7 Now White can utilize the inclusion of h2-h4 and ...h5 to his
advantage with:

14.dxc5! (14.Bc4!? b5 15.Nxg6 bxc4 16.Nxe7† Kh7 looks tempting, but after 17.Nf5 Nb6
18.Nxg7 Nxa4 or 17.d5 Qb7 it is not clear if White can claim any advantage) 14...Nxc5 15.Be3
b6 (15...Qb8? does not work here due to 16.Nxg6! Nxa4 17.Nxe7† Kh8 18.Bf4 Qa7 19.Be5
Nh7 20.Qf4+–) 16.Nxc5 bxc5 17.Bh6ƒ White has various attacking ideas including Qg5, Bd3
and g2-g4. And even if Black manages to exchange queens, he may still have problems in a
future endgame with the split queenside pawns, especially if White can trade bishops and
establish his knight on c4.

12.Na4 Qb4
12...Qa7 is well met by 13.h5! Nxh5 (13...cxd4? is even worse due to 14.hxg6 hxg6 15.Bh6+–)
14.Bh6± and White’s attack is too fast.

13.Qxb4

375
 13.Nxc5 Qxd2† 14.Bxd2 Nxe5 15.dxe5 Ng4=

13...cxb4
An interesting endgame. If Black consolidates, he would be just fine, but he still has serious
development issues to solve.

14.Nxd7 Nxd7
14...Bxd7 is well met by 15.Nb6 Rad8 16.Bc4! when it is difficult for Black to get coordinated,
because the otherwise desirable ...Bc6 runs into d4-d5.

15.Be2!
15.g3 enables Black to solve his problems with: 15...e5! 16.dxe5 Nxe5 17.Nb6 Bg4! 18.Nxa8
Bxd1 19.Kxd1 Rxa8=
The text move threatens Bf3, which would paralyse Black’s queenside, while better preparing
against the following central break.

15...e5
This is the critical move – Black needs to try and clear some space for his pieces. At the same
time, it all seems like dodgy stuff: Black barely hangs on with a bunch of tricks, and even then
he falls short of full equality.

15...b5 16.Bf3 Ra7 17.Nc5² maintains the pressure.

15...h5 16.Bf3 Ra7 17.Rhe1 Bf6 18.g3 gives Black an unpleasant endgame, as he still has to
solve some fundamental problems. Here are a couple of brief sample lines:

376
18...b5 (18...b6 19.Bg2 Bb7 20.d5± threatens Be3) 19.Nc5 Nxc5 20.dxc5± With a dangerous
passed pawn.

16.dxe5 b5
In the event of 16...Nxe5?! 17.Nb6 Rb8 18.Nd5 Black must give up a pawn, as 18...a5? leads
to bigger losses after:

19.Ne7† Kh8 20.Rd5!+–

377
17.Bg4!
17.Bf3 Ra7 18.e6 looks tempting, but after 18...fxe6 19.Be3 Rc7 20.Nb6 Ne5 Black is close to
equalizing, with ...Nc4 coming up next.

17...Bxe5
17...bxa4 18.Bxd7± with an extra pawn.

17...Nxe5 18.Bxc8 bxa4 enables White to exert lasting pressure after:

19.Bb7 Ra7 20.Bd5² The two bishops should ensure that White can continue to pose problems
for Black. An important point is that Black does not seem to have any concrete way to cause
problems using his queenside pawns. For instance, ...a3 can always be met by the simple bxa3,
and c2 is easy to defend. Meanwhile it will be hard to arrange any ...Nc4 ideas due to the bishop
on d5. Of course Black can continue playing solidly in various ways, but the bishops offer White

378
good chances for a long-term edge.

18.Bh6!
18.Bxe5 Nxe5 19.Bxc8 bxa4 20.Bb7 Ra7= is nothing special for White with only one bishop
instead of two.

18...Re8
18...Bg7?! enables White to win a pawn in more favourable circumstances after: 19.Bxg7
Kxg7 20.Bxd7 Bxd7 (20...bxa4 21.Bxa4±) 21.Nb6 Bg4 22.f3 Rab8

23.Nd5 Be6 24.Nxb4±

19.Bxd7 Bxd7 20.Nb6 Bg4 21.f3 Rab8 22.Nd5 Be6 23.Nxb4

White has won a pawn. I think the bishop pair should offer Black reasonable defensive
chances, but the onus is clearly on him to equalize.

379
 Conclusion

5...g6 is an interesting move which has become quite popular in recent years. I find 6.Bg5!? Bg7
7.Qd2 to be the most appealing reply, as White plays with a clear purpose and forces his
opponent to think about the possibility of a kingside attack.

We analysed two main approaches by Black, beginning with 7...a6!?, delaying castling for the
time being. After 8.Bf4 Qd8 we analysed the aggressive 9.0-0-0, having in mind the sacrifice of
a pawn with h4-h5, followed by an exchange sacrifice if Black captures with the knight. It may
not guarantee a theoretical advantage, but attacking players may well find such positions
appealing. The subtler 9.Be2!? is my main recommendation: if Black presents a clear target with
9...0-0, we go on the offensive with 10.0-0-0N intending h2-h4 and so on; and if 9...b5, we
switch strategies with 10.b4!N, intending to attack on the queenside.
Next we looked at 7...0-0, when 8.0-0-0 leads to a further branching. Both 8...Bf5 and 8...Bg4
9.Be2 lead to rich play with many subtleties. Often in such positions, White does well to drop the
bishop back to f4 to force Black’s queen to move, but it is important to time this correctly, and to
adapt our strategy according to whether the queen goes to d8 or b6. The main line of 8...a6
9.Bf4! leads to just such a situation: the relatively passive 9...Qd8 should be met be the
aggressive 10.h4!, but 9...Qb6! puts Black in a more suitable position to obtain counterplay, so
10.Ne5!? is more appropriate. I like White’s chances in all these lines, although the theory is still
being mapped out so we can expect plenty of developments.

380
A) 6...Bf5 208
B) 6...Be6!? 7.Bf4 Qd8 8.Qd2 Nbd7 9.0-0-0 209
B1) 9...Nxe5 210
B2) 9...g6 10.Bc4!? Bxc4 11.Nxc4 212
B21) 11...Bg7 212
B22) 11...Nb6N 213
C) 6...Nbd7 7.Nc4 Qc7 8.Qf3 Nb6 9.Bf4 216
C1) 9...Qd8 10.Be5 216
C11) 10...Nbd5 217
C12) 10...Bg4 11.Qg3 219
C121) 11...h5?! 219
C122) 11...Be6 221
C13) 10...Be6 11.Ne3!? 224
C131) 11...Nbd5 225
C132) 11...g6!? 227
C2) 9...Qd7 10.Nxb6 axb6 11.0-0-0 231
C21) 11...Qd8 232

381
 C22) 11...e6 12.Bc4! 235
C221) 12...Nd5 235
C222) 12...Be7 13.g4!?N 237
C2221) 13...b5 239
C2222) 13...Nd5!? 241

1.e4 d5 2.exd5 Qxd5 3.Nc3 Qd6 4.d4 Nf6 5.Nf3 c6

This move isn’t as flexible as 5...g6, but I consider it more solid. To me, 5...c6 seems more in
the spirit of the ...Qd6 Scandinavian, although that is just my own feeling. Even though the text
move lags behind 5...a6 in terms of all-time popularity, 5...c6 has been the more fashionable of
the two moves in recent years – although lately, 5...g6 has become trendier still. Sergey Tiviakov
has played 5...c6 in a huge number of games, contributing many ideas along the way. I also
noticed that Magnus Carlsen introduced an early novelty in one of the critical lines to easily
neutralize Fabiano Caruana at the 2016 Olympiad. The text move has taken a slight dip in
popularity recently, at least in over-the-board games, but there is no obvious reason why.

6.Ne5
This active move prepares Bf4 or Nc4, according to Black’s response.
6.h3!? is a worthy alternative which has yet to be explored in great detail. If you follow the
most recent games, White seems to be doing well, but I believe many improvements should be
possible for Black.

We will analyse A) 6...Bf5, B) 6...Be6!? and C) 6...Nbd7.

A) 6...Bf5

This just seems to present the bishop as an inviting target for White.

7.Bc4
7.g4 will be met by either ...Bc8 or ...Be6, and while White may still have a good position, I
don’t feel comfortable about making such a committal pawn move so early in the game. If we
can provoke ...e6 before playing g2-g4, it’s a different story because we can eliminate the light-
squared bishop by force.

7...e6 8.g4 Bg6 9.h4 Nbd7 10.Nxd7

This simple approach works fine.

382
10.Qe2!? is an interesting alternative. 10...Qxd4 11.Nxd7 Kxd7 12.Bg5! was seen in Poetz –
Maurer, Graz 2011, when Black got too greedy by gobbling the g-pawn, but 12...Ke8!N would
have been more challenging. White has plenty of play for a pawn and perhaps an advantage can
be found here too, but these complications just don’t seem necessary to me.

10...Qxd7

11.h5 Be4 12.Nxe4 Nxe4 13.c3

White has a typical edge with the bishop pair, along with the kingside pawns which are well
placed to assert dominance. Of course Black will castle on the queenside to avoid a direct attack,
but White can still aim for a long-term edge without too many worries.

13...0-0-0
13...Nf6N could be a slight improvement although White’s chances remain preferable after:
14.Qf3 Nd5 (or 14...Be7 15.Bd2 0-0-0 16.0-0-0 h6 17.Kb1 Qc7 18.g5 hxg5 19.Bxg5²) 15.Bd2
0-0-0 16.Bb3! Bd6 17.c4²

383
14.Qf3!N
This seems quite disruptive for Black.

14.Qe2 Nf6 15.Bd2 Bd6 16.0-0-0 was seen in Svidler – Adams, Frankfurt 1999, and now after
16...Qc7!N it seems like Black’s position is becoming more or less acceptable. He can consider
...Nd5 or ...Bf4 next, and the ...c5 break could be an option too.

14...Nf6?!
14...Nd6 15.Bb3² leaves the knight completely out of place for most of Black’s future plans.

14...f5 15.Be3² leaves Black with a weakened kingside.

The text move is more ambitious, as Black hopes to achieve full coordination with the knight
coming to d5 later, but Black runs into concrete problems after:

15.Bg5! h6
15...Be7 16.h6! is even worse for Black.

16.Bxf6 gxf6 17.Qxf6 Bd6 18.0-0-0±

With a safe extra pawn for White.

B) 6...Be6!?

384
 Ignoring the Bf4 threat is a rare but interesting approach. It may seem like a concession for
Black to lose additional time getting his queen chased around – but if White doesn’t know what
he is doing, Black may simply complete his development and equalize.

7.Bf4 Qd8
7...Nd5 gives White many ways to continue but the simplest approach seems to be to avoid
exchanges and continue to make Black suffer for his slow development with: 8.Ne4 Qd8 9.Bg3
Bf5 (9...Nd7 10.Ng5±)

10.Bd3N (10.Ng5!? f6 was unnecessarily messy in Kitze – Krysztofiak, Magdeburg 2012)

10...e6 (10...Bxe4 11.Bxe4²) 11.0-0 Nb4 (11...Be7 12.c4²) 12.Nc4 Nxd3

385
13.Ncd6† Bxd6 14.Nxd6† Kf8 15.cxd3ƒ White should be able to exploit his superior
development and safer king to provoke concessions of some kind before Black has a chance to
get coordinated and exploit his superior structure.

8.Qd2
Intending to continue in the most active and logical way with 0-0-0, while leaving Black to
solve his development issues.

8...Nbd7
8...g6 9.0-0-0 Bg7 (9...Nbd7 leads straight to variation B2 below) 10.Bh6! Bxh6 (10...0-0
11.h4ƒ) 11.Qxh6 Nbd7 12.Bc4 Bxc4 13.Nxc4 transposes to variation B21.

9.0-0-0
We will now analyse these two main options: B1) 9...Nxe5 and B2) 9...g6.

386
9...Nb6?! was once played by Tiviakov but he never repeated it. 10.Be2 (10.h4!?N is an
interesting option to discourage the kingside fianchetto) 10...g6 Now in Soumya – Tiviakov,
Chennai 2012, 11.h4!N would have been strong. Presumably White was worried about the
kingside becoming blocked after 11...h5, but then 12.Rhe1 leaves Black precariously placed.

B1) 9...Nxe5 10.Bxe5 Qa5

Black plays in rather a concrete manner, hoping for simplifications.

11.a3 Nd5
I also considered:
11...0-0-0
The king feels rather vulnerable on the queenside, and Black has a lot of development woes
to deal with. Nevertheless, an accurate move is needed to make the most out of White’s
chances:
12.Kb1!N
Thwarting the main plan of ...Nd5, and making it hard for Black to continue development.
12.Be2 Nd5!N would enable Black to force a queen exchange (rather than the disastrous
12...Ng4? 13.Bg3 g5 14.d5!+– as occurred in Stupavski – Majeric, Bosnjaci 2008).
12...g6
If 12...Nd5 13.Nxd5 Black must play 13...Qxd5 to avoid losing a pawn, and after 14.b3
Black has to deal with annoying threats such as Bc4 followed by Qa5.

387
13.Qc1!
White has plenty of other ways to get an advantage, but I like the idea of targeting the
queen, which is short of squares.
13...Rd7
13...Bg7?? allows White to execute his big threat: 14.b4 Qb6 15.Na4+–
14.b4 Qd8 15.b5!
Using the time gained to launch an attack.
15...c5 16.Na4 cxd4 17.b6! a6
17...axb6 18.Qb2+–
18.Qb2 Bg7 19.Bxa6! bxa6 20.Rxd4+–

12.Bc4!N
White must be precise to avoid a queen exchange, which would solve most of Black’s
problems.

388
Thus, 12.Ne4 was pretty harmless in Erdelyi – Boricsev, Hungary 2006.

12...f6
12...Nxc3 13.Bxe6 is obviously not desirable for Black.

13.Qe1! fxe5
13...Kf7 is met by 14.Bg3 when Black will not be ready to take on c3 for a while, and White
can regroup with Qe2, Rhe1 and so on. Clearly, it will not be long before the king on f7 starts to
feel uncomfortable.

14.Nxd5 Qxe1 15.Nc7† Kd7 16.Bxe6† Kxc7 17.Rhxe1 exd4 18.Rxd4

After a few forced moves, we reach a pretty dismal endgame for Black. The best he can do is:

18...Rd8 19.Rxd8 Kxd8 20.Rd1† Kc7 21.Rd7† Kb6 22.f4 g6 23.g4±

Sooner or later, Black will have to give up the e7-pawn and suffer for a long time in the hope
of making a draw.

B2) 9...g6

389
 Keeping the ...Nxe5 exchange in reserve seems like a better bet. Now White must make an
important decision.

10.Bc4!?
Generally speaking, exchanging minor pieces is not such a great idea for White. However, the
bishop on e6 was becoming slightly annoying due to ideas such as ...Nd5, or ...Nxe5 followed by
...Qa5. After this exchange, Black’s options are considerably reduced – for example, the knight
on c4 will prevent ...Qa5 outright, while a hasty ...Nd5 will run into tactical problems, as shown
in the note to Black’s 11th move below.

10.d5!? was played in a few games but it seems like Black is actually fine after 10...Bxd5
11.Nxd5 Nxd5 12.c4, and now 12...N5b6N 13.Nxd7 Nxd7 14.g3 e6 seems like best play. White
has decent compensation for the pawn but it’s hard to see how he will break through, as Black is
solid enough.

10...Bxc4
10...Nd5N gives us a few good options, including 11.h4!? Nxf4 12.Qxf4 Bxc4 13.Nxc4 h5
14.Kb1 Bg7 15.Ne4 which seems pleasant for White.

11.Nxc4
Black has two logical moves: B21) 11...Bg7 and B22) 11...Nb6N.

11...Nd5?? allows a neat trick: 12.Nxd5 cxd5

390
13.Nd6†! exd6 14.Rhe1† Be7 15.Bxd6+–

B21) 11...Bg7

This position was first reached in Kanovsky – S. Kasparov, Teplice 2013. I suggest:

12.Bh6N
Trading bishops seems like the simplest way to go.

12...Bxh6
12...0-0 invites 13.h4 with attacking chances. If 13...b5 then White can flick in 14.Bxg7 Kxg7
before 15.Ne3, when the knight is perfectly placed to support g2-g4 ideas. Of course, it may also
jump away to allow the queen to go to h6 when needed. 15...h5 (15...Qa5 doesn’t cause any

391
worries for White after 16.Kb1±) 16.g4ƒ White has excellent attacking chances.

13.Qxh6 Qc7 14.Rhe1

14.d5!? is a more aggressive approach, immediately trying to exploit Black’s unsafe king. My
analysis continues: 14...cxd5 15.Ne3 Nb6 (15...e6? loses to 16.Nexd5! exd5 17.Rhe1† Kd8
18.Nxd5 Nxd5 19.Rxd5 Kc8 20.Re7 Rd8 21.Rd3!+–)

16.Ncxd5 Nbxd5 17.Nxd5 Nxd5 18.Rxd5 Qc4 19.Rhd1 Qxa2 20.Qg7 Rf8 21.b3!² The queen is
barely able to do anything on a2, while White dominates the board. 21...Qa6 (21...Rc8?
22.Qc3!+– is a nice detail) 22.Kb1ƒ A lot of technical difficulties remain, but White is clearly in
the driver’s seat – and Black has already had to sidestep a few landmines just to get this far. So
this approach certainly deserves consideration – but it’s not required, as White seems to have a
stable edge anyway.

14...0-0-0 15.f4 Kb8 16.Ne5²

White keeps the upper hand in a one-sided position. It is hard to imagine ...Nxe5 ever being a
good idea for Black, so he will have to keep defending the f7-pawn. Meanwhile, the ...c5 break
seems hard to arrange, as it will only expose the black king further. So White can just continue
regrouping his pieces and slowly build up the pressure.

B22) 11...Nb6N

392
 This seems like a natural attempt to improve Black’s play. We have already seen that
developing the bishop to g7 gave White an easy initiative with Bh6, so it is logical for Black to
leave the bishop on f8 for a moment and prod the knight on c4.

12.Ne5!?
This is the most ambitious approach.

A decent alternative is:

12.Ne3
This allows a little too many simplifications for my taste, but White still keeps a slight
initiative with a better pawn structure.
12...Nfd5
I also considered: 12...Bg7 13.h4 Nfd5 14.Ncxd5 Nxd5 15.Be5!? Simplifications generally
favour Black – but here, specifically because the black king is still not too comfortable, we
can afford to exchange the minor pieces and still be able to cause problems with ideas like
h4-h5. 15...Bxe5 (15...f6 16.Bg3²) 16.dxe5 Nxe3 17.Qxe3 Qb6 18.Qf4 The black king is
still quite uncomfortable and, with h4-h5 or e5-e6 coming up, he will continue to suffer for
a while.
13.Ncxd5 Nxd5
13...cxd5 14.h4 h5 15.Be5 f6 16.Bf4² leaves Black’s position looking shaky, and g2-g4 will
be annoying too.
14.Be5
Provoking a weakening pawn move.

393
14...f6 15.Bg3 b5
15...Bg7 16.c4 Nxe3 17.Qxe3² leaves White completely dominating the centre, as well as
having h4-h5 ideas in the air.
16.Qe2²
Even if Black avoids any major problems in the near future, his position doesn’t seem
comfortable with ...f6 having been played. White can patiently improve his position with Kb1,
Rhe1 and so on, maintaining some pressure.

12...Nfd5 13.Ne4!
Black was actually threatening ...f6, as can be seen after 13.h4?! f6! 14.Nd3 Nc4 15.Qe1
(15.Qe2 Nxb2!) 15...Nxf4 16.Nxf4 and now 16...Bh6! is slightly annoying.

13.Kb1!? is a reasonable option which removes the possibility of the pin at the end of the above
sequence – but the text move seems stronger and more purposeful.

13...f6
This seems more or less forced, as 13...Bg7?! invites 14.Bh6±.

14.Nd3
I don’t see how Black can be happy with his structure after committing to the ugly and
weakening ...f6 move, despite the absence of any serious danger in the short term.

394
14...Nc4
14...Bg7 15.Nec5² sees White target the weak e6-square.

The text move is Black’s best chance to cause problems. Although he can jump around a bit,
eventually his tricks will run out of steam:

15.Qe2 Qa5 16.Kb1 Qb5 17.Bd2 0-0-0

The king will still be somewhat vulnerable here, and the e6-square remains a big weakness.
White just needs to regroup on the queenside to avoid any tricks.

18.b3 Nxd2†
Some beautiful tactics may arise for White after:
18...Qb6? 19.Bc1!
The strongest and fanciest continuation.
The simple 19.c3 would be better for White as well.
19...Qxd4

395
20.Qg4†! Kb8
20...f5 21.Nd6†+– picks up the queen.
21.Ndc5! Nc3†!
Clever stuff, but White has a crushing counter-blow.
22.Nxc3! Qxg4 23.Rxd8† Kc7

24.Rd4!
Safeguarding the rook and utilizing the possibility of a fork on e6 to win the knight.
24...Qf5 25.Rxc4 b5!
Black’s resources are still not exhausted – he threatens to win material, but White can
prevail by targeting the king.
26.g4! Qf3 27.Rd1! bxc4

396
28.Ne6†!
Weaving a mating net.
28...Kb7
Or 28...Kb6 29.Na4† Ka5 30.Rd7!+– and the king is caught.
29.Rd7† Ka6 30.Nc5† Ka5 31.N3a4
It is mate in a maximum of five more moves.

19.Qxd2

19...Qb6 20.c3²
White has an obvious advantage due to his extra space and the weakness of e6.

C) 6...Nbd7

397
7.Nc4
This is an automatic reply, avoiding the knight trade while gaining time against the queen.

7...Qc7 8.Qf3 Nb6 9.Bf4

After a few standard moves, we reach our first major crossroads in this variation. C1) 9...Qd8
and C2) 9...Qd7 are almost identically popular, although they have quite different characteristics.
The former keeps a lot of tension on the board, whereas the latter just aims to simplify with
...Qg4, but in the meantime places the queen rather awkwardly.

C1) 9...Qd8

The possibility of ...Bg4 prevents us castling for now, and Black may also develop the bishop
to e6 in the future.

398
10.Be5
Black has tried a few options from here, but they tend to be quite similar to each other. His
main concern is development – it is still not entirely clear how he should proceed on the
kingside. If he plays ...e6 he blocks his light-squared bishop, and White can counter with Qg3 to
prevent ...Be7. So ...g6 is a more reasonable plan, but it requires suitable preparation to avoid a
tactical mishap.

Another feature of the position is that Black has less space – so, as usual, he would like to
exchange a few pieces to ease his congestion and reduce White’s attacking potential. He would
love to exchange queens of course – but more realistically, the knights can be targeted. From
White’s perspective, we should make sure Black does not get to simplify too easily. Another idea
that I found particularly helpful across these lines was to try and exploit ...g6 with an immediate
h2-h4, when replying with ...h5 clearly weakens Black’s kingside, while allowing the pawn to
get to h5 may also be unpleasant for him.

We will analyse these three main options: C11) 10...Nbd5, C12) 10...Bg4 and C13) 10...Be6.

The immediate 10...g6?? is a blunder. Black actually got away with it and went on to win in A.
Chapman – Heinzel, Lieme 2010, but the result would have been different if White had spotted:

11.Nd6†!N+– Winning a rook.

10...h5!?N is currently untested but there is a good chance it will be played at some point, if only
because it is one of the engine’s top choices. I suggest 11.Bd3 Bg4 12.Qg3 h4 13.Qf4 with
pretty easy development and a slight edge for White. For instance:

399
13...Nxc4 (13...Be6 14.Nxb6 axb6 15.h3²) 14.Bxc4 Qd7 15.h3 Bf5 16.0-0 e6 17.Rfe1 Be7
18.Bb3 Nd5 19.Qf3 Nxc3 20.bxc3² Black’s king faces an uncertain future.

C11) 10...Nbd5

We have already noted that simplifications tend to help Black. However, this way of going about
it is rather time-consuming, and White does well to keep developing.

11.Bd3!
An almost-novelty! The idea is just to develop quickly and flexibly.

Obviously 11.0-0-0?? would lose material to 11...Bg4, and 11.Be2?! would needlessly allow a
bishop exchange after the same move. These considerations explain why 11.h3 has been White’s
most popular choice, but it is too slow for my liking.

A more tempting alternative is:

11.Nxd5!? Qxd5
11...cxd5 12.Ne3 e6 13.Bd3 was promising for White in Emms – A. Ledger, Southend
2006; for instance, after something like 13...Be7N 14.0-0 0-0 15.Qh3 Black is under
pressure.
After the text move, the queen does not have an ideal square to go to.
12.Qe2
12.Qc3!?N 12...Ne4 13.Qd3 looks a bit artificial although it does keep the queen from
being exchanged, so this may deserve attention.
12...Be6

400
13.Ne3!?N
After 13.c3 Qe4 White’s position remains slightly preferable and he went on to win a fine
technical game in Gashimov – Epishin, Ourense 2009. Personally I prefer to keep the
queens on, but a pleasant endgame is also not a bad outcome from the opening.
13...Qa5† 14.c3 Nd7 15.Bg3 g6 16.h4 h5
It is not entirely clear how White should develop his slight initiative, although the computer
prefers White’s position and the whole line could certainly be investigated in more detail. With
that being said, let’s return to my main recommendation.

11...Nxc3N
This seems like the most consistent sequel to Black’s previous move.

11...Bg4? is pointless here, as 12.Qg3± just leaves the bishop unstable.

It is also worth mentioning that 11...g6?? allows a variation on a familiar tactical theme:

401
12.Nd6†! exd6 13.Nxd5+– White does not win a whole rook this time, but his advantage is
decisive all the same.

11...Be6 doesn’t help with Black’s development woes at all. In Tsvetkov – Novoselov, Nizhnij
Tagil 2014, White should have continued 12.0-0N 12...g6 (12...Nxc3 13.bxc3±) 13.Nxd5 Bxd5
14.Qe2 with an excellent game, for instance:

14...Bxc4 (14...Be6 15.f4±; 14...Bg7 15.Nd6†!) 15.Bxc4 Bg7 16.f4 0-0 17.f5ƒ With a great
game for White in all these lines.

12.bxc3

402
12...Qd5 13.Qg3!?
White has other ways to maintain an edge, but I always prefer to keep the queens on in these
lines unless we have a good reason not to.

13...Ne4!?
Pawn-grabbing may seem risky, but it is hard to suggest a good alternative.

For instance, 13...g6 14.0-0 Bg7 15.f4! gives White a huge initiative at no material cost.

14.Qe3 Nxc3 15.0-0 f6

15...e6?! is met by 16.Qe1 Nb5 17.a4! Nxd4 18.Rd1+– when White’s lead in development is
too hot to handle.

403
16.Bg3 e6 17.Rfe1ƒ
With superb play for a mere pawn.

C12) 10...Bg4

This has been played in a lot of games and Houska even goes so far as to call it the only move
for Black, but it doesn’t make much sense to me, as forcing the queen away from f3 seems to
serve White in most scenarios. However, there are certain concrete ideas connected with it,
which will be discussed later.

11.Qg3
There are two moves that deserve attention. C121) 11...h5?! has been the most popular choice
but I am not impressed by Black’s scheme. Instead, C122) 11...Be6!? seems like the best chance
to justify Black’s play, as it takes away certain options that White would have had against
10...Be6 (which we will analyse later, in variation C13 on page 224).

C121) 11...h5?!

Chasing the queen causes White no real problems, while making it highly unlikely that Black
will be able to castle kingside in the future.

12.h3
12.h4?! Nfd5 gives Black more chances of justifying his play.

The rare 12.Bd3!? has scored quite well in a few games, but the text move works just fine.

12...h4

404
13.Qe3!
13.Qf4 has been tested in some high-level games, but 13...Be6 14.Ne3 (or 14.Nxb6 axb6)
14...Nbd5 gives Black the simplifications that he craves.

13...Be6 14.Nxb6 axb6

14...Qxb6 doesn’t challenge White at all. Now we can simply play 15.0-0-0 as there is no open
a-file to worry about, while the black king will not feel comfortable on either flank. For instance:

15...0-0-0N (15...g6?! made White’s task easier after 16.Bxf6 exf6 17.Bc4± in Huschenbeth –
Dranov, Bonn 2011) 16.a3² The last move not only safeguards the queenside, but also prepares
to meet 16...g6?! with 17.Na4! Qa5 18.Nc5± when the black queen is dangerously close to
getting trapped after b2-b4.
Black has managed to draw some old correspondence games after the text move, but in a
practical game I would regard Black’s position as uncomfortable. His development is far from

405
complete and, even more worryingly, the pawn on h4 does not seem to serve any purpose, other
than acting as a target for White.

15.Bd3 b5 16.0-0 b4

17.Ne4!N
17.Ne2 also gives White the more pleasant game, but it has allowed Black to survive in the
correspondence games with some weird trickery which the text move avoids.

17...Nxe4
17...Nd5? 18.Qf3 f6 19.Nc5+– is an example of why Black should exchange pieces when he
gets the chance.

18.Qxe4 Bd5
18...Qd5 19.Qe1!± hits the b4-pawn.

19.Qe1 e6

406
20.a3! Qg5
20...bxa3? 21.c4+– is White’s idea of course.

21.f3 f6 22.Bh2²/±
Black’s b4-pawn is under fire, his light-squared bishop is short of squares and his king remains
vulnerable in the centre.

C122) 11...Be6

At first glance it seems strange to provoke Qg3 before putting the bishop on e6. After all, with
the queen on f3, White couldn’t even castle because of the possibility of ...Bg4. However, the
subtle point is that with the queen on f3, White would like to meet ...Be6 with Ne3 (as shown in
variation C13 below); but with the queen having been nudged to g3, the same plan does not work

407
as well.

12.Nxb6!
Compared to the version with the queen on f3 we need a different approach, and this is the
right way to exploit the fact that Black has essentially given us a free tempo by provoking Qg3.

12.Ne3 Nfd5! revealed Black’s idea in Iordachescu – Ardelean, Iasi 2014. With the queen on g3,
there is a serious threat of ...f6 and ...g5, trapping the dark-squared bishop. This enables Black to
force simplifications.

12.b3!?N is an interesting alternative, when 12...Nbd5 (or 12...g6 13.h4 Nh5!? 14.Qe3 Bg7
15.Be2 Nxc4 16.bxc4 Qa5) 13.Be2 b5 (13...g6) 14.Nd2 Qa5 leads to double-edged play. This
could be explored in more detail but I consider the main line more promising.

12...axb6

13.Bd3
13.0-0-0!? would be a provocative choice, inviting 13...Bxa2!? 14.b3 e6 15.d5! (but not
15.Kb2? Qe7!) with insane complications. It’s not clear if White has any real advantage though,
so it doesn’t seem especially worthwhile to analyse this in depth – especially as Black has a more
solid alternative in 13...Nd5 if he prefers to avoid the craziness.

13...g6
This seems like Black’s best way to continue development.

13...Nd5 14.Nxd5 Qxd5 15.0-0² and it is hard to see how Black will finish his development
other than by resorting to the weakening ...f6 move.

408
13...b5?!
This seems like a luxury that Black can ill afford.
14.0-0 g6
Having castled, we don’t have the option of h4-h5, but there are other ways of attacking.

15.d5!
This is the strongest and most elegant way of exploiting Black’s lack of development.
15.f4!? gives White an improved version of the main line below, so this would have been
attractive if it were not for the existence of an even better alternative.
15...Bxd5 16.Nxb5! cxb5 17.Bxb5† Bc6 18.Bxc6† bxc6 19.Rad1!
Bringing the last piece into play with tempo, before smashing through to the black king.
19...Qc8 20.Bxf6 exf6 21.Rfe1† Be7

22.Rxe7†!
A simple combination to seal the victory.

409
22...Kxe7 23.Qd6† Ke8 24.Re1† Qe6 25.Rxe6† fxe6 26.Qxe6† Kf8 27.Qxf6† Kg8 28.b4+–

After the text move I would like to deviate from the existing correspondence praxis with:

14.f4!?N
I want to attack with 0-0 and f4-f5, and the advantage of delaying castling is that White more
or less rules out ...Bg7 because of the immediate f4-f5.

14.h4!? is similar to the approach we will see in variation C132 on page 227, but here I like it
slightly less because Black will be able to meet the ‘threat’ of h4-h5 with the cold-blooded
...gxh5. Moreover, 0-0-0 is not so appealing here due to the open a-file, so the h4-h5 plan makes
it harder to find a suitable home for the white king. My analysis continues 14...b5! (14...Bg7
15.h5ƒ; 14...h5 15.0-0 Bg7 16.f4ƒ and f4-f5 will be disruptive) 15.h5 gxh5! and it is hard to
prove anything for White.

14...Qd7
Controlling the key f5-square.

14...Bg7?! just invites 15.f5±.

14...b5 is simply met by 15.0-0 when it is not clear what Black is doing, as any further advance
of the b-pawn simply drives White’s knight towards the kingside where it supports his attacking
plans.

I also checked a more forcing alternative: 14...Nh5!? 15.Qe3 Bg7 (15...f6 is met by 16.Bxf6! of
course) 16.f5 gxf5 17.0-0 Bxe5 18.Qxe5 Nf6 19.Bxf5

410
19...Ra5!? (19...Bxf5 20.Rxf5²) 20.d5 Bxd5 21.Rad1 Rg8 22.g3© Black’s position is holding
together but his king is a long way from feeling safe, and overall White’s initiative is good value
for a mere pawn.

15.0-0 Nh5
Black still cannot develop normally with 15...Bg7? because of 16.f5!.

16.Qf2 Bg7
16...f6 17.d5! makes life tricky for Black, for instance: 17...fxe5 (17...Bxd5?! 18.Nxd5 Qxd5
19.Rad1 Qf7 20.Qxb6± is extremely dangerous) 18.dxe6 Qd4 19.fxe5 Qxf2† 20.Rxf2

411
20...Ng7 21.Rf7!² Black still has to figure out a way to regain the pawn, as the immediate
21...Nxe6 allows: 22.Rxh7!±

17.Rad1 0-0

18.f5 Bxf5 19.Bxf5 Qxf5 20.Qxf5 gxf5 21.Rxf5 Nf6

21...Bxe5? 22.Rxe5± and White wins a pawn.

22.a3!?²
White definitely has an easier game, and can follow up with d4-d5(-d6) and/or Rd3-g3. Black
has decent defensive chances of course, but his task is not easy.

C13) 10...Be6

412
11.Ne3!
This possibility has already been discussed in the analogous position with the queen on g3. In
the present version with the queen on f3, avoiding (or at least delaying) the knight trade enables
us to maintain the pressure on Black.

11.Nxb6 isn’t so great. Compared to the variation analysed immediately before this, White is
essentially a tempo down, as the queen would clearly prefer to be on g3. Black has more than
one decent continuation but the simplest is 11...Qxb6!?, and if 12.Be2N 12...Nd7= Black targets
both the e5-bishop and the d4-pawn.

We will analyse two main options: the simplifying C131) 11...Nbd5 and the more subtle C132)
11...g6!?.

11...Nfd7?! 12.Bg3 Nf6 has been played a couple of times but such a loss of time will not go
unpunished. 13.0-0-0 g6 Now in Willemze – S. Kasparov, Hoogeveen 2012, White overlooked a
bone-crunching refutation of Black’s play:

413
14.d5!N 14...Nbxd5 15.Ncxd5 cxd5 16.Bb5† Bd7 17.Nxd5+–

11...Nfd5 12.Ncxd5 and now a couple of games have continued with 12...Nxd5, immediately
transposing to variation C131 below. Instead 12...Bxd5N is playable but it hardly seems like an
improvement in view of:

13.Qg3 f6 Black will have to play this sooner or later in order to continue with kingside
development. 14.Bc7!? Qd7 15.Bd3² Black is behind in development and his kingside has been
weakened, so White has no cause for complaint.

C131) 11...Nbd5

414
 This is the most popular move. Trying to simplify the position by exchanging one or both pairs
of knights seems to make sense – but at the same time it puts more of a focus on Black’s
kingside pieces which remain undeveloped.

12.Ncxd5!
12.Nexd5 has been equally popular but 12...Bxd5 leaves White’s queen without an ideal
square, since 13.Qg3 allows 13...Ne4 with further simplifications, while also keeping Black’s
bishop relatively well placed.

12...Nxd5
12...cxd5? leads to trouble after 13.Bb5†, for instance: 13...Bd7 14.Bxd7† Qxd7 15.Bxf6 gxf6
16.Nxd5±

12...Bxd5
This should be met by a simple improvement:

415
13.Qf4!N
13.Nxd5 Qxd5 is harmless at best, as has been demonstrated in numerous games.
13.Qg3? must be avoided due to 13...Ne4! 14.Qf4 Qa5† 15.c3 Nxc3.
The queen may look a bit odd on f4 but Black can’t really exploit it, and we can soon castle
followed by kicking Black’s bishop away with c2-c4.
13...g6
13...Nd7?! is pointless in view of 14.Bc7.
13...Nh5?! 14.Qh4 leaves Black with nothing better than 14...Nf6 when White has several
good options, for instance: 15.0-0-0±
I also considered 13...Be6 14.Be2 Nd5 15.Nxd5 Qxd5 16.0-0 and Black still has to solve
the problem of developing his kingside.
14.Be2
14.0-0-0 Nh5! 15.Qh4 f6 would be annoying, so we develop the bishop while guarding the
h5-square.
14...Bg7

416
15.0-0
15.0-0-0!? is also possible, but White clearly has the more pleasant game thanks to the plan
of c2-c4 and d4-d5 anyway, so why not go for the version where our king is absolutely
safe?
15...0-0 16.c4 Be6 17.Rad1 Ne8
17...Qa5?! only makes matters worse after: 18.d5 cxd5 19.cxd5 Bd7 20.b4 Qd8 (20...Qb6
21.d6+–) 21.d6±
18.d5 cxd5 19.Nxd5²
Black still suffers from a lack of manoeuvring space for his pieces, and White intends to
increase the pressure with Bf3 next.

13.Bd3
Simple development works well. Black will almost certainly have to play ...f6 to develop
further, which creates serious weaknesses and gives White rather easy play.

417
13...f6
13...Nxe3?! 14.Qxe3± makes it even harder for Black to develop. He still needs to spend
another tempo on ...Qd7 or ...Bd5 before he can play ...f6, and only then continue with ...g6 and
so on. Meanwhile, White is completely developed and ready to start creating threats.

14.Bg3 g6N
This seems like as good a try as any.

14...Nb4 15.Be4 Bd5 16.0-0 was excellent for White in Donev – Cochet, Bern 1997.

14...Bf7 15.0-0 e6 occurred in Dekic – Radlovacki, Sombor 2010, when 16.Rfe1²N would have
been the most logical way to maintain an indisputable advantage.

15.0-0 Bg7 16.Rfe1 Bf7

16...0-0?! enables White to win a pawn: 17.Qe4 Bf7 18.Nxd5 Bxd5 19.Qxe7±

17.Nxd5 Qxd5
17...Bxd5?! 18.Qe3± just seems like a worse version for Black.

18.Be4 Qd7 19.Rad1

19...f5
19...0-0 20.d5 cxd5 21.Bxd5 Bxd5 22.Rxd5² maintains White’s edge in a one-sided position.
19...Rd8 20.d5! cxd5 21.c4 0-0 22.Bxd5² is similar.

20.Bd3 0-0 21.Be5 Bxe5 22.dxe5 Bd5

418
 22...Qc7? 23.e6± is not something Black should allow.

23.Qf4 Qc7 24.h4²

Black has a difficult defence ahead.

C132) 11...g6!?

Despite being far less popular than the previous variation, this move seems more logical to me.
Black prioritizes the completion of his kingside development with ...Bg7 and ...0-0, while
maintaining the options of simplifying with ...Nbd5 or even dropping the knight back to d7. So if
White just continues playing without any concrete ideas, Black will happily complete his
development and then start simplifying with one knight or the other moving to d5, avoiding the
developmental problems seen in the previous variation.

419
12.h4!N
12.d5!? leads to some forced lines but Black’s position was shown to be fine after 12...Nbxd5
13.Nexd5 Bxd5 14.Nxd5 Qa5† 15.c3 Qxd5 16.Qxd5 cxd5 17.Bb5† Kd8 18.Bxf6 exf6= in
Komodo – Sting, engine game 2015.
I also considered 12.Bd3N but after 12...Bg7 13.0-0 0-0 it is not clear how White should
continue.

12...h5
This move has certain drawbacks which will be discussed shortly – but allowing the h-pawn to
advance is no picnic for Black either.
12...Bh6 seems rather strange, and 13.h5 Bxe3 14.fxe3² is pleasant for White.

12...Bg7 13.h5! Nxh5

13...gxh5 seems risky in view of: 14.Qg3 Kf8 (14...Rg8 15.Bxf6 exf6 16.0-0-0±) 15.Ne4ƒ
14.Rxh5
14.0-0-0© is also possible.
14...f6
There are many interesting possibilities here; but theoretically, it seems simplest to show
that White can reach a pawn-up endgame after:

15.Qg3 fxe5 16.Rxh7

16.Rxe5 Bf7 seems okay for Black.
16.dxe5!? 0-0 17.Rh4 certainly deserves attention, but the text move is simplest from a
theoretical perspective.
16...Rxh7 17.Qxg6† Kd7 18.Qxh7 Qg8 19.Qxg8 Rxg8

420
20.dxe5 Bxe5 21.0-0-0†²
Black’s bishop pair gives him defensive chances, but White can obviously press with the extra
pawn.

Returning to our main line, Black’s last move stops the h4-pawn in its tracks but at the same time
represents a serious concession on the kingside, particularly if Black goes on to castle there.
Thus White’s decision to advance the h-pawn has been justified, and now we can prepare f4-f5
to target the black kingside.

13.Qg3
13.Bd3 Nbd7! 14.0-0-0 Qa5 is somewhat annoying.

13...Bg7
13...Nbd7 is not too helpful here because of 14.0-0-0 when ...Nxe5 isn’t really possible. (I

421
don’t see much point in the immediate 14.Bc7 Qc8 when we cannot play f2-f4.) If Black goes
14...Qa5 then we have a nice idea available:

15.Bc7! Suddenly this becomes a lot more annoying for Black. A sample line is 15...b6 16.Bc4
Bxc4 17.Nxc4 Qb4 18.Nd6†! exd6 19.Rhe1† Be7 20.Bxd6+– and White crashes through.

14.0-0-0
An interesting alternative is:
14.Bd3!?
In some ways this is the more flexible approach: there doesn’t appear to be any need to
commit the king to the queenside, and perhaps White can even combine the f4-f5 idea with
0-0 in some lines. But it also gives Black some extra options with:
14...Nbd7!
By targeting the bishop on e5, Black exposes the drawback of White’s decision to delay
castling.

422
 14...Kf8 enables White to justify his previous move with: 15.Ne2 Nbd5 (15...Nfd5?!
16.Bxg7† Kxg7 17.Qe5† Nf6 18.Nf4±) 16.Nxd5 Nxd5 17.0-0!? White’s king will be safer
here than on the queenside, and after something like 17...Bxe5 18.Qxe5 Rg8 19.Rfe1²
Black remains under pressure.
15.0-0-0
15.Bc7?! Qc8 just leaves the bishop uselessly placed while making f2-f4 impossible.

15...Nxe5
15...0-0 16.f4ƒ and 15...Qa5 16.Bc7! Nb6 17.a3 both seem promising for White.
16.dxe5 Ng4
16...Nd7? 17.f4± leaves White in full control.
17.Bf5 Qc8 18.Bxg4 Bxg4
Black should avoid 18...hxg4? 19.Ne4±.
We have reached an interesting position where it is worth trying to find something for
White. My efforts continued with:
19.f3 Be6 20.f4 Qc7
20...0-0?! proves to be too risky after: 21.Qg5! Qe8 (even worse is 21...Qc7? 22.g4+–)
22.f5 gxf5 23.Nxf5 Bxf5 24.Qxf5±

423
21.Ne4
21.Qf3 can be met by 21...Bd7! (but not 21...0-0 22.g4) 22.Ne4 0-0-0 23.Ng5 Rhf8= when
Black seems fine.
21...0-0
21...Bxa2? is punished by: 22.b3 a5 23.Nf5!+–
22.Ng5 Bxa2 23.b3 a5 24.e6 a4 25.exf7† Rxf7 26.Nxf7 Kxf7 27.Nc4
White may be able to claim some advantage with machine-like precision, but I would be
worried by the open queenside and general messiness of the position.

14...0-0
14...Nbd5 cause no disruption to White’s plans because after 15.Ncxd5 Black can’t recapture
with the knight. 15...Bxd5 (after 15...cxd5 16.Bd3 White continues in a similar vein while the
d5-pawn just blocks all of Black’s play) 16.Bd3 0-0 17.c4 Be6 18.f4ƒ

14...Kf8!? is an interesting idea – the king is less vulnerable than after castling, and Black can
hope for an improved version of the ...Nd5 simplifications with the g7-bishop now defended.
Nevertheless, Black’s position is still far from safe and White keeps the better chances after:
15.Bd3 Nbd5 (15...Nbd7?! 16.f4±; 15...Nfd5 16.Ne4 Nxe3 17.fxe3ƒ)

424
16.Ncxd5 (16.Ne2 is unconvincing after 16...Nxe3 17.fxe3 Ng4!) 16...Nxd5 (16...cxd5 17.f4ƒ)
17.Rhe1² Black’s position remains awkward, especially without the rooks connected.

15.Bd3
Preparing the typical plan of f4-f5.

15...Qc8
In the event of 15...Nbd7 16.f4 Bg4 White can consider 17.f5!? (17.Rde1 should also be
pleasant for White) 17...Bxd1 18.Rxd1 Bh6 19.Bf4 Bxf4 20.Qxf4© with a promising attack for
the exchange.

Black’s other obvious try is:

15...Nbd5 16.Ncxd5 Nxd5 17.Rhe1
Simply preparing to recapture on e3 with the rook.

425
17...a5
17...Bxe5?! looks too risky due to: 18.dxe5 Qe8 (18...Qc7? 19.Bxg6+–) 19.f4‚

18.Bxg7 Kxg7 19.Nc4!

Simple and strong.
There are some fancy tactical ideas involving Nxd5 and Re5 which should also lead to some
advantage, but such antics are completely unnecessary.
19...Bd7
19...Qc8? loses to: 20.Qe5† Kg8 21.Bxg6+–
20.Ne5²/±
Black will have a hard time avoiding a calamity on the kingside, and can only dream about the
notion of counterplay.

16.Kb1!
16.f4 Ng4! prevents any further advance of the f-pawn for the time being.

426
Fortunately White does not need to hurry with f4-f5, and can take his time to improve his king
and develop the h1-rook before launching any attack. Meanwhile, Black’s pieces are rather
passive and there isn’t much he can do to improve his position.

16...a5
16...Nbd7 17.Rhe1 Nxe5 18.dxe5 Nd7 19.f4ƒ is good for White.

17.Rhf1 Nbd7
17...a4? 18.f4+– already leaves Black without a defence.

18.f4
The direct approach seems most natural, although it could be interesting to explore the subtle
18.Be2!?, anticipating the arrival of a black knight on g4.

18...Nxe5 19.dxe5 Ng4

20.f5 Nxe3 21.Qxe3 Bxf5 22.Bxf5 gxf5 23.Qg5ƒ

Black could be looking at a difficult choice between safeguarding his kingside at the expense
of the f5- and h5-pawns, or vice versa. He certainly has defensive chances, but White remains in
the driver’s seat.

C2) 9...Qd7

427
 It looks a bit weird to block the c8-bishop; but as stated earlier, Black’s idea is to have the
option of simplifying with ...Qg4.

10.Nxb6
10.0-0-0 Qg4! 11.Qxg4 Bxg4 12.f3 Be6= was a simple demonstration of Black’s scheme in Lu
Shanglei – Kotronias, Moscow 2016.

10...axb6 11.0-0-0
Castling is the right way to develop White’s opening initiative, despite the open a-file.

We have reached an important juncture, where C21) 11...Qd8 and C22) 11...e6 are the main
options to consider.
11...Qg4?! has been played quite often but it makes no sense here because 12.Qe3 leaves
Black’s queen looking silly, and if 12...Qe6 White breaks through with:

428
13.d5! Nxd5 (in the event of 13...Qxe3† 14.Bxe3 Nxd5 15.Nxd5 cxd5 16.Rxd5 Bd7 17.Bb5
Bxb5 18.Rxb5+– Black’s queenside will collapse by the time he develops his kingside pieces,
Pap – Wortmann, Heusenstamm 2018) 14.Nxd5 cxd5 15.Qb3!? Ra5 16.Be3 Bd7 17.Bxb6+–
Black’s lack of development proved to be his undoing in Kravtsiv – Kulkarni, Mumbai 2015.

C21) 11...Qd8

It looks strange to move the queen yet again, but Black reasons that the queen has already
served its purpose on d7 by provoking the knight trade on b6. As we will see, after following up
with ...e6, there are certain advantages to having the queen on d8 instead of d7.

12.h3
White needs to address the threat of ...Bg4, and 12.Qe3 Nd5 13.Nxd5 Qxd5 14.c4 Qe6 does
not offer much.

429
12...e6 13.Bc4 Bd6
This seems to me like the best attempt to justify Black’s 11th move.

13...Be7?!
This allows a sacrificial breakthrough on a seemingly well defended square:

14.d5!N
14.g4 was played in Grischuk – Rapport, Internet (blitz) 2017, when 14...0-0!?N would
have been best; the position resembles the later variation C222 but this could be a slightly
improved version for Black, as h2-h3 hardly seems necessary. This is all academic though,
as the text move is much stronger.
14...cxd5
14...Nxd5 15.Nxd5 just leads to the same thing after further exchanges on d5.
14...exd5 15.Bxd5 Bd7 is an independent possibility, but 16.Bb3 0-0 17.Rhe1± is obviously
great for White.
15.Bxd5! exd5 16.Nxd5 Nxd5
16...Bd7 17.Nc7† Kf8 18.Rhe1!± leaves Black in big trouble.
17.Rxd5 Bd7 18.Rhd1 Rxa2

430
19.R1d3!
Ensuring that any check along the first rank will not be a problem.
19...0-0 20.Rxd7 Qe8
20...Ra1† 21.Kd2 Bb4† (or 21...Qe8 22.Qe3) 22.c3 Qf6 23.Bg3 just leads to an even worse
endgame for Black.
21.Qe3!±
Avoiding any tricks and forcing a queen exchange in the near future. In the resulting endgame,
White will soon pick up a pawn on the queenside, with excellent winning chances.

If Black wishes to develop his bishop to e7, he should first include:

13...b5 14.Bb3 Be7
This is the recommendation of Kotronias in The Safest Scandinavian.

15.Be5

431
 Unlike the 13...Be7?! line above, the d4-d5 sacrifice is a non-starter here, as Black’s queen
will have the b6/a5 escape route available. However, there are other ways to put Black’s
position under pressure.
15...0-0 16.Qg3!N
A simple improvement over the strange 16.Bxf6?! Bxf6 which was played in A. Horvath –
J. Horvath, Hungary 2017.
16.Ne4!?N is a good alternative. For instance, 16...Nd5 (16...Nxe4 17.Qxe4±) 17.Qg3 f6
18.Bd6 b6 (18...Bd7 19.Rhe1 Be8 20.Bxe7 Qxe7 21.Qd6 Rf7 22.Rd2±) 19.Rhe1 Ra7
20.Bxe7 Qxe7 21.Nd6± and White keeps the upper hand.
16...Nh5

17.Qh2!
Kotronias gives 17.Qg4 Nf6 18.Qg5 Ne8! 19.Qg3 Qa5! followed by a long line where
Black seems to be okay. The text move seems like a natural improvement.
17...Nf6
Black will have to bring the knight back into play at some point.
17...f6? 18.Bc7 Qd7 19.g4+– traps it.
18.Rhe1 Nd5 19.Ne4²/±
White is better mobilized and has good control over the dark squares. Next he may creep up
the board with the h-pawn or play Bd6 to trade off the better of Black’s bishops. Black is
reasonably solid, but he is still a long way from equalizing.

432
14.Bg5!?N
In the similar position with Black’s queen on d7 and White’s pawn on h2, 14.Be5 would be a
good move, but here the presence of the queen on d8 means there is no real threat to take on f6.
Black benefitted from this by means of 14...b5 15.Bb3 b4 16.Ne4 Nxe4 17.Qxe4 0-0 in Krejcir –
Barclay, corr. 2018; the piece exchanges ease the congestion in Black’s position, and it is not so
easy for White to develop an initiative after these simplifications.

This led me to suggest the text move as an improvement. The idea is simply to avoid exchanging
the bishops, and start gaining space on the kingside. If Black plays ...h6, it will create a hook
which we can exploit with g4-g5. And if the bishop drops back to e7, we can happily exchange
on f6 when it suits us, as neither of Black’s bishops will have any great prospects in the resulting
position, while White retains the easy plan of advancing on the kingside.

14...Be7
14...h6 15.Bd2! leaves Black vulnerable to the typical plan of g4-g4. Play may continue:
15...Nd5 (15...0-0 seems risky in view of 16.g4 Nd5 17.Ne4ƒ) 16.Kb1!?

433
 Avoiding any annoying checks. A sample continuation is 16...b5 17.Bb3 Nxc3† (17...b6
18.g4ƒ) 18.Bxc3 Qf6 19.Qe2² and White is ready to start advancing the g- and h-pawns.

15.h4
15.Bxf6!? is another option to avoid the possibility of ...Nd5 – but I don’t see any special
reason to avoid it.

15...Nd5!?
15...0-0 16.Bxf6! Bxf6 17.Ne4ƒ gives White straightforward play on the kingside.

16.Ne4
Avoiding an unwanted knight trade while continuing with the plan to advance on the kingside.

434
16...b5 17.Bb3 0-0
17...b6 18.Qg3! creates annoying threats such as Bxe7 and Qxg7, as well as Bxe7 and Nd6†.
So 18...f6 is natural, and after 19.Bf4² White is obviously better.

18.g4 b6
18...f6 The e6-pawn becomes really weak. Black’s hopes rest on the idea ...b6 and ...c5, but we
can deal with it by 19.Bd2 b6 20.Nc3 and ...Nxc3 would be a disaster for Black, and without it,
he can’t really create threats with ...c5.

19.Bxe7 Qxe7 20.g5 c5

20...f5 21.gxf6 Nxf6 22.Ng5ƒ looks unpleasant for Black.

21.Nf6†! gxf6 22.Bxd5 exd5 23.gxf6 Qe4

24.Rhg1† Kh8 25.Qg3 Bg4 26.Qxg4 Qxg4 27.Rxg4 Rxa2 28.Rg5±

Black has avoided being mated, but his only reward is a miserable endgame to defend.

C22) 11...e6

435
12.Bc4!
This is an important part of our scheme. The bishop may not seem too impressive aiming at e6,
but it will do a perfect job of securing the queenside after dropping back to b3. This gives White
a relatively free hand to focus on expanding and developing his initiative on the kingside.
12.Be5 has been played more commonly but I am not a fan of it, for two reasons:
a) 12...Nd5 has been the most frequent reply, when Black will have to follow up with ...f6,
which clearly creates some weaknesses of course. At the same time, White has to switch to a
completely different plan on the kingside than what I had envisaged. And based on the games
that have been played so far, it is far from clear if White has any real claim to an advantage.
b) 12...Be7 has been a rare choice but it seems the simplest solution for Black. In that case
White has nothing better than 13.Bc4, which is similar to our main lines, but there are more
useful ways of spending a tempo than by moving the bishop to e5.

Black’s two most logical moves are C221) 12...Nd5 and C222) 12...Be7.

12...Bd6 doesn’t help Black at all due to 13.Be5, and if 13...Bxe5 14.dxe5 Nd5 15.Ne4 Black’s
weak dark squares look rather horrible.

12...b5 13.Bb3 Be7 14.g4!?N transposes to variation C2221 on page 239.

C221) 12...Nd5

436
 It is crucial to notice that White can unexpectedly exploit Black’s slow development with:

13.Nxd5!
13.Be5 has been seen often, especially if we count the games that arrived here via the 12.Be5
move order. However, I am not wholly convinced by White’s prospects in the resulting position,
as explained in the note to our previous move.

I briefly considered 13.g4!?N, by analogy with variation C222 below. However, 13...Nxf4
14.Qxf4 Bd6 gives Black a better version, as White has to spend time moving his queen to a
suboptimal square. For instance: 15.Qf3 (in the event of 15.Qe3 Qc7! 16.Kb1 b5 17.Bb3 0-0
18.h4 b6 19.g5 Bb7 Black is ready for ...c5, and if 20.g6 hxg6 21.h5 Bf4! White’s attack does
not quite work) 15...b5 16.Bb3 0-0 17.g5 b6 18.h4 Bb7 Here too, ...c5 is coming and White is
not quite in time with his attack; the queen on f3 is also placed rather unfortunately.

13...exd5
The alternative is:
13...cxd5 14.Bb3
White’s queenside is even more secure here than in the ...exd5 structure. Still, there are a
few subtleties that will enable us to make the most out of White’s chances.

437
14...b5!?N
14...Bd6 was played in E. Andreev – Bartos, Teplice 2013, and now 15.c3!N 15...0-0 16.h4
b5 17.a3± would have kept White’s queenside super-solid, and it’s only a matter of time
before the black king comes under fire.
The text move (which, by the way, briefly transposes to another game) seems like a natural
attempt to improve on the above line. Black hurries to advance the pawn to b4, smothering
White’s queenside and enabling attacking plans such as ...b6 followed by ...Ba6-c4.
15.Qg3!
Obstructing the further development of Black’s kingside.
15...b4N
15...f6 16.h4 h5 was played in Chigaev – Khairullin, Sochi 2016, when 17.Rd3!N± seems
like the most attractive of several promising continuations, with ideas of playing on the c-
and/or e-files.
The text move looks natural enough but White can create serious problems with:

438
16.Kb1 b6 17.Rc1!
Intending to attack along the c-file. For instance:
17...Ba6 18.c3! Bc4 19.cxb4
Black has no time to cause any damage on the queenside.

14.Rde1† Be7 15.Bd3! 0-0

Black does not have time for 15...Rxa2? because 16.Kb1 Ra8 17.Bg5 f6 18.Bxf6! (18.Re3!? is
also great for White) 18...gxf6 19.Re3 Kd8 20.Rhe1 Re8 21.Bxh7 is crushing.

16.Qh5 f5
16...g6 is well met by 17.Qe5! Qe6 (or 17...Re8 18.Kb1) 18.Qc7 Bd8 19.Qd6 Qd7 20.h4 and
Black continues to feel the heat.

439
17.g4!?N
17.Kb1 Bf6 18.Be5 Qf7 19.Qh3 Bxe5 20.Rxe5 Bd7 21.Rhe1 g6 22.f4² seems rather pleasant
for White as well, even though Black managed to hold a draw with careful defence in Rubio
Doblas – Kreindl, corr. 2012.

17...Bd6 18.Bxd6
18.Be5!? could also be considered.

18...Qxd6 19.Rhg1 g6 20.Qh4

20...Qf6
20...fxg4 21.Re7 h5 22.Re5! maintains the pressure.

440
21.g5 Qd6 22.Kb1²
White’s initiative is far from over – and he can combine it with threats to invade along the e-
file, as well as on the dark squares.

C222) 12...Be7

13.g4!?N
Before explaining the thinking behind this new move, here is a brief look at why I was less
than satisfied with the alternatives:

13.Be5 has been the usual choice but it should convert to 13.h4, and I don’t see any point in
hurrying to commit the bishop.

13.h4 0-0 14.h5 Nd5 15.Be5 (15.Nxd5 exd5 16.Bb3 b5 makes it difficult for White to advance
further on the kingside; and 15.g4 runs into 15...Nxf4 16.Qxf4 e5! 17.Qxe5 Bd6 18.Qe2 Qxg4=)
15...b5 16.Bb3 b4 17.h6 g6 18.Ne4 f6 19.Bh2

441
19...b6! 20.Kb1 Ba6 21.Rhe1 f5 22.Nd2 Bb5! Black had sufficient counterplay in Gburek –
Peschke, corr. 2013. If he succeeds in exchanging the light-squared bishops, his pressure along
the a-file could become annoying.

In most of the games, White has advanced his pawn to h6 as in the example above, but it’s not
clear to me if this is much of an achievement. Instead, I want to construct the kind of attacking
set-up on the kingside which would more commonly be associated with the Open Sicilian. Thus,
we start by advancing a pawn to g5, which serves the purpose of restricting Black’s pieces as
well as gaining White space to develop further threats on the kingside. Next we will play h4-h5,
followed by a timely g5-g6 to open a file or two.
In response to g4-g5, Black will of course answer with ...Nd5. When that happens, we should
simply ignore it, because neither of the possible exchanges (...Nxf4 or ...Nxc3) should
particularly bother us. It might seem counterintuitive to allow the dark-squared bishop to be
traded off for a knight, but our knight can be equally effective in supporting the attack via e4.
Once we have established the pawn on g5, Black will be in something of a stranglehold, so the
exact move order will not be so important thereafter. For instance, while it is possible to plough
straight ahead with h4-h5 and g5-g6, White may also consider playing more slowly with Kb1.
Meanwhile, it is not clear how Black should proceed. One possible plan is ...b5 followed by ...b6,
hoping for ...c5-c4, but it is quite tricky to implement. Alternatively, he could delay castling and
perhaps even aim for ...0-0-0, but this is also easier said than achieved.

We will analyse C2221) 13...b5 and C2222) 13...Nd5!?.

Let’s briefly consider what happens if Black rather cooperatively plays 13...0-0 and gives White
a clear target: 14.g5 Nd5 15.h4 White will proceed in essentially the same way as in variation
C2221 below, with the added possibility of meeting ...b5 with Bd3 if the attacking benefit
outweighs the weakening of the queenside. Thus, it feels more natural for Black to kick the

442
bishop sooner rather than later.

13...e5 is also of no concern, as after 14.Bxe5 Qxg4 15.Qe3 the open files are only going to hurt
Black.

C2221) 13...b5 14.Bb3

14...0-0
14...b4 15.Ne4 Nxe4 16.Qxe4 b6 17.g5 Bb7 18.Rhg1 g6 (or 18...0-0-0 19.Be5) 19.Be5 is
obviously pleasant for White.

15.g5 Nd5 16.h4

Black’s only real way to fight back appears to be connected with ...b6 and ...c5-c4, but he
needs to solve several problems before he can execute that plan.

16...Nxf4
Swapping off the bishop seems as good as anything. Here are some other possibilities:

16...b4 17.Ne4 b6 can be punished by:

443
18.Be5! Ba6 19.Rdg1 Clearly White is too fast. The main threat is Nf6†, but even the slower
plan of pawns to h5 and g6 is enough to crush Black.

16...f6 is well met by: 17.g6! hxg6 (17...h6 does not lose immediately but the pawn on g6 is
obviously a huge pain, and White will have many ways to continue his attacking plans,
beginning with 18.Bd2ƒ. Even more importantly, Black has no good way of threatening anything
on the queenside.)

18.Rdg1! Qe8 19.Ne4 White has an obviously strong initiative.

In the event of 16...b6 17.h5 Bb7 White should proceed with 18.Rhg1, removing the rook from
harm’s way after a subsequent ...c5. Play could continue:

444
18...Nxf4 19.Qxf4 Bd6 20.Qg4 Qc7 (or 20...c5 21.dxc5 bxc5 22.Nxb5) 21.g6ƒ The rook is well
placed to support White’s attacking plans.

17.Qxf4

17...Bd6
17...b6 18.Ne4 Bb7 is too slow due to 19.Nf6†! and White wins.

18.Qf3!
18.Qg4 needlessly exposes the queen, and 18...Qc7 19.Kb1 e5! gives Black enough activity.

I also checked 18.Qe3 Qc7 19.Kb1 b6 20.Ne4 Bf4! 21.Qf3 Bb7 when Black is just in time; he
will play ...c5 next, which is enough to deal with any threats involving knight jumps.

445
18...b6
18...b4 19.Ne4 just helps White to improve his knight.

18...Qc7?! proves to be too slow after 19.Ne4 Bf4† 20.Kb1 b6 when Black is a full tempo down
compared to the 18.Qe3 line noted above. Unsurprisingly, a single move makes a huge
difference to White’s attack, as shown after:

21.h5 Bb7 22.h6! If Black had the extra tempo of ...c5, he would have been able to eliminate the
knight and block the kingside with ...g6. But as things stand, 22...g6 23.Rh4!+– is deadly.

19.Ne4 Bb7 20.Qe3!

20.h5 c5„ gives Black enough play.

20.Qg4!? is a reasonable alternative though. Play may continue 20...Qc7 21.Nxd6 Qxd6 22.h5
c5 23.dxc5 Qxc5 24.Rhe1 Bd5 25.Bxd5 exd5 when Black has avoided succumbing to a kingside

446
attack, but after 26.a3 White may still be able to claim some sort of edge due to his superior
pawn structure.

20...Qc7
20...Bc7 21.g6!ƒ would be dangerous for Black.

21.Nxd6 Qxd6

22.h5 c5
22...Qc7 runs into 23.g6! and Black is not quite in time for ...c5. Therefore Black is forced to
spoil his queenside structure.

23.dxc5 Qxc5 24.Rh4 Qxe3† 25.fxe3²

In addition to targeting Black’s weak queenside pawns, White can also throw in a timely h5-h6
to weaken Black’s kingside and back rank.

C2222) 13...Nd5!?

447
 This is a more subtle attempt to deal with White’s scheme. The main difference here is that
Black maintains the option of long castling. This seems a bit odd, but if White isn’t able to do
much immediately then Black will be solid enough.

14.Kb1!
The blunt 14.g5 b5 15.Bb3 Nxf4 16.Qxf4 can be met by a sudden change of plans from the
previous variation: 16...Qd6! 17.Qe3 b6! Preparing long castling. 18.h4 Bb7 19.h5 0-0-0÷
Black’s position will be hard to crack.

14.Be5!? could be worth considering, but I like the flexibility of the text move and I am not
concerned by the possibility of the bishop being swapped off.

14...b5
14...0-0 15.g5 b5 16.Bb3 Nxf4 17.Qxf4² is similar to the previous variation.

15.Bb3 b6
Obviously we should also consider:
15...Nxf4 16.Qxf4
At first it may seem good for Black to eliminate the dark-squared bishop, but the knight was
also an excellently placed piece for him. With that gone, it will make White’s play quite
smooth and easy. The general plan remains the same: advance with g4-g5 and use that pawn
to build a kingside attack.
16...Qd6
The slow 16...b6? is refuted by 17.d5! and White smashes through.
16...0-0 17.Ne4 b6 18.g5ƒ is our familiar attacking plan.
16...Bd6 17.Qf3 0-0 18.Ne4ƒ also looks pleasant for White.

448
17.Qe3 b6?!
Trying to prepare long castling is an important plan to consider. However, Black’s position
comes unglued after:
18.d5! Qc5
18...cxd5 19.Nxd5 exd5 20.Bxd5 Ra7 21.Bxf7† Kxf7 22.Rxd6 Bxd6 23.Qf3† and White is
able to clean up the queenside pawns as well: 23...Ke7 24.Re1† Kd8 25.Qc6! Rd7
26.Qxb6† Bc7 27.Qxb5+–
19.dxc6 Qxc6 20.Nd5 Bc5 21.Qc3 Bb7
21...0-0? 22.Ne7†+– wins the queen.
21...Rg8 22.Qg3 Ra7 23.Rhe1ƒ leaves Black in a mess.

22.Nf4 0-0 23.Nh5 f6

23...e5!? may be the lesser evil although 24.Qxe5± is simply a free pawn for White.
24.Nf4±

449
 White has strong pressure against e6, and the plan of attacking with the g- and h-pawns has not
gone away.

16.Be5
Provoking a commitment: Black will either have to place his king in the firing line or weaken
his position with ...f6.
16.h4!? is also possible, intending g4-g5 and so on. Black may reply with 16...Nxf4 17.Qxf4
Bb7 when, unlike the lines in the previous note, he has managed to exchange on f4 and develop
his bishop to b7 without allowing a deadly d4-d5 break. Even so, after 18.Ne4 White’s play still
seems clearly easier.

16...f6
16...0-0 17.g5 Bb7 (or 17...f6 18.gxf6 Bxf6 19.Qg3ƒ) 18.h4 gives White easy attacking
prospects, for instance:

450
18...Nxc3† (18...c5 19.dxc5 bxc5 20.Nxd5 c4 21.Nf6† Bxf6 22.Qe3±) 19.Qxc3 Bd6 20.g6‚

17.Bg3

17...0-0
17...Bb7?! is not really a viable option due to 18.Rhe1 followed by Qe2.

18.h4ƒ
White has the makings of a promising attack.

Conclusion

This chapter brings our coverage of the Scandinavian Defence to an end, by dealing with the
popular and solid option of 5...c6. After 6.Ne5 we started by checking 6...Bf5, which enables
White to hunt down the bishop and claim a relatively simple edge, followed by 6...Be6!? 7.Bf4
Qd8 8.Qd2 Nbd7 9.0-0-0, when White should be able to put his development advantage to good
use, although Black’s solid set-up should not be underestimated.

6...Nbd7 is by far the most popular continuation, when 7.Nc4 Qc7 8.Qf3 Nb6 9.Bf4 gives Black
a choice of queen moves.

First we looked at 9...Qd8, when 10.Be5 led to a further split. 10...Nbd5 is time-consuming, and
White does well with 11.Bd3 and fast development. 10...Bg4 11.Qg3 gives Black a choice: the
popular 11...h5?! seems suspicious to me, but the unusual 11...Be6 is an interesting attempt to
exploit the shifting of White’s queen from f3 to g3; however, with 12.Nxb6! we use that shift to
our advantage. On the other hand, if Black plays 10...Be6 a move earlier then 11.Ne3! should be

451
preferred, as this fits in better with the queen on f3. Then 11...Nbd5 is an attempt to simplify, but
Black’s kingside development remains slow. I believe 11...g6!? is a better try, but 12.h4!N is a
useful innovation.

We then looked at the other important queen move 9...Qd7, when the possibility of ...Qg4 means
we should change tack with 10.Nxb6 axb6 11.0-0-0. Then 11...Qd8 is not stupid, but moving the
queen yet again is rather slow, and I found some nice ways for White to build up on the kingside.
Instead 11...e6 is the main line, when 12.Bc4! is best. White should be ready to meet 12...Nd5
with 13.Nxd5!, with some concrete ideas after either recapture. 12...Be7 is the other main move,
when 13.g4!?N is a promising new idea, aiming for a Sicilian-style attack.

452
A) 1...d6 2.d4 Nf6 3.Nc3 c6 4.Nf3 245
A1) 4...Bg4!? 245
A2) 4...Qa5 247
B) 1...g6 2.d4 249
B1) 2...Nf6 250
B2) 2...d6 252
B3) 2...Bg7 3.Nc3 253
B31) 3...Nc6 255
B32) 3...d5!? 257
B33) 3...c6!? 4.Be3 d5 5.e5! 260
B331) 5...Nh6 260
B332) 5...Qb6 262
B333) 5...f6 263
B34) 3...d6 266

453
1.e4
In this chapter, we begin our investigation of the Pirc/Modern complex by dealing with some
of Black’s unusual options after A) 1...d6 and B) 1...g6.

A) 1...d6 2.d4 Nf6

2...e5 3.Nf3 takes us back to the Philidor, as covered in my first volume. There are other legal
moves of course, but nothing to warrant serious attention in an opening book.

3.Nc3

3...c6
This is known as the Czech Pirc. Apart from Black’s first and second moves being the same, it
has little in common with the actual Pirc. Black rarely fianchettoes his dark-squared bishop and

454
often plays an early ...e5, so in some ways it is more closely related with the Philidor.

Speaking of the Philidor, 3...e5 was already covered in my first volume; and 3...g6 is the Pirc of
course.

3...Nc6 seems like the only other move worth mentioning, when 4.Nf3 transposes to the
Nimzowitsch Defence – see variation E2 of Chapter 17 on page 405.

4.Nf3
White has a pleasant choice; among other options, 4.f4 is a popular and promising alternative.
However, the text move seems like the most convenient fit with the rest of our repertoire.

We will now analyse A1) 4...Bg4!? and A2) 4...Qa5.

As before, there are other legal moves, but none of them require special attention. For instance:

4...g6 5.Be3 Bg7 6.Qd2 transposes to Chapter 12.

4...Nbd7?! seems weird. At the minimum, White can play 5.a4, when 5...e5 6.Bc4 steers back to
the Philidor, but it also looks tempting to try and punish Black’s move order with 5.e5.

4...b5 is not really the point behind Black’s previous move – at least not at such an early stage.
5.Bd3 gives White an easy initiative, with e4-e5 a tempting possibility; there is no need for us to
go any deeper.

A1) 4...Bg4!?

This has become quite popular recently in internet blitz events (also via the move order 1.d4 d6

455
2.Nf3 Bg4 3.e4 Nf6 etc.). The appeal seems to come from the fact that Black can continue with
simple moves such as ...e6, ...Be7, ...d5 and ...Nbd7, which are ideal for faster time controls,
whereas White has to make all the tough decisions, since White has many different ways to
continue.

5.h3 Bh5
5...Bxf3 6.Qxf3 would just give White a solid edge without much to worry about.

6.Qe2 e6 7.g4 Bg6

8.d5!?
This seems to me like the simplest continuation.

8.h4 has been much more common but I don’t see any need to rush with fixing the kingside
structure. After something like 8...h5 9.g5 Nfd7 10.Bh3 Be7 11.0-0 White has a promising game
due to the nice plan of Ne1, followed by either f2-f4 or Nd3-f4, although careful attention is
needed to ensure Black will not find some way to prevent White’s kingside advances. Without
these Ne1 plans, Black would probably be just fine.

8...cxd5
8...exd5? is a bad idea due to 9.exd5† Be7 (or 9...Qe7 10.Be3±) 10.Nh4 0-0 11.Nxg6 hxg6
12.Be3± when Black lags behind in both development and space.

8...Nbd7 seems positionally risky due to 9.dxe6 fxe6 10.g5 Nh5 11.Be3² when Black’s structure
has been permanently damaged.

9.exd5 e5

456
 This position was reached in Tregubov – Nogueiras Santiago, Linares 1996, when 10.h4 was
played. I would prefer:

10.Nh4!N
After eliminating the light-squared bishop, White should have an excellent version of a
Sicilian.

10...Be7 11.Nxg6 hxg6 12.Be3²

Intending 0-0-0. Black faces a dilemma: it is hard for him to arrange short castling, as White is
completely dominating the kingside; but if the rook on h8 cannot get into the game, Black has
little hopes of a successful offensive on the queenside.

A2) 4...Qa5

457
 This is the more thematic choice for the Czech Pirc, but exposing the queen like this has
obvious drawbacks.

5.Bd2 e5
5...Qb6 achieves nothing after 6.a4, when taking on b2 loses the queen, so it is not clear what
Black is doing.

The text move has been Black’s most popular choice. Allowing the Nd5 discovered attack seems
brave, but White needs to be clever to exploit it in an effective way.

6.dxe5
6.Bc4 sets up some Ng5 ideas but 6...exd4! spoils the fun. 7.Nd5 Qc5 8.Nxf6† gxf6 may still
offer White good compensation, but such measures are not required.

6.Nd5 Qd8 7.Nxf6† Qxf6 8.Bg5 Qg6 may well be good for White but it leads to messy
complications, whereas the text move leads to an advantage by relatively simple means.

6...dxe5

7.Bc4!
It is hard for Black to defend against the Ng5 and Nd5 knight jumps.

7...Qc5
This seems like the only way to avoid instant problems, but the queen becomes awkwardly
placed.

7...Qc7? invites 8.Ng5 of course.

458
7...h6 comes unstuck after: 8.Nb5! Qd8 (or 8...Qb6 9.Nxe5+–)

9.Nxe5 Bc5 (9...cxb5? 10.Bxf7† Ke7 11.Bb4† leads to a quick mate) 10.Bxf7† Ke7 Black
somehow went on to win from this disastrous position in Gardner – Gensler, Chicago 1989, but
11.Bb3N 11...cxb5 12.Ng6† would have been easily winning for White.
7...Be7
This was played in Mishugin – Kondratski, Bila Tserkva 2004. A big improvement is:

8.Ng5!N 8...0-0
8...Rf8 9.Nd5 Qd8 10.Nxf6† Bxf6 11.Nxh7+– is hardly an improvement for Black.
9.Nd5 Qd8 10.Nxf6† Bxf6 11.h4!
With a crushing attack coming up. Here are a few illustrative lines:
11...h6
11...g6 12.Qf3 Nd7 13.0-0-0 Qe7 14.Qg3 and White will soon break through, for instance:

459
 14...b5 (14...h5 15.Nxf7+–) 15.Bb3 a5 16.h5 a4 17.hxg6 hxg6 18.Qh2+–

12.Qh5 Qe7 13.0-0-0

13.Qg6 Bxg5 14.hxg5 Be6 15.Bxe6 Qxe6 16.Qxe6 fxe6 17.gxh6+– is also adequate.
13...Be6
13...Nd7 14.Qg6+–
14.Nxe6 fxe6 15.Qg4 Re8 16.Bxh6+–
A final alternative is:
7...Bg4 8.h3
8.Nd5 Qd8 9.Ne3 Bh5 10.g4!? Bg6 11.Nf5 also looks quite promising but the text move is
simpler.

8...Bh5
8...Bxf3 9.Qxf3 Nbd7 occurred in Reig Albero – Balada Moreno, Mislata 1995, when 10.0-
0-0N± would have been great for White.

460
9.Nd5 Qd8 10.Nxf6† Qxf6
10...gxf6 was seen in Ledezma Borregales – Massari, Barquisimeto 2012, and now the
simple 11.g4N 11...Bg6 12.Qe2 gives White a strategically won game due to the dead
bishop on g6.
11.g4 Bg6 12.Bg5 Qd6 13.Qxd6 Bxd6 14.0-0-0
White’s fast development combined with the potential for a mate on d8 leads to the win of a
pawn.

14...Bc7
14...Be7 15.Bxe7 Kxe7 16.Nxe5± also leaves White a clear pawn up.
15.Nxe5 b5
In Talmazan – Rakhmangulov, Ilichevsk 2008, the most accurate continuation would have
been:
16.Bb3N±
White can remove the bishop on g6 whenever he feels like it, and Black has no compensation
for the missing pawn.

461
8.Qe2!N
White preferred 8.Bb3 in Baumegger – Stuhlik, Leiner 1996, but the text move seems more
accurate, as White prepares long castling without delay.

8...Be7
8...b5 9.Bb3 Be7 enables White to exploit the loosening of Black’s queenside with: 10.Be3
Qd6 11.a4! b4 12.Rd1 Qc7

13.Qc4! (13.Nb1² would give White a pleasant position but the text move leads to more tangible
gains) 13...Be6 14.Nd5 Qb7 15.Nxe5 cxd5 16.exd5 Nxd5 17.Rxd5 0-0 18.0-0±

9.Be3 Qa5

462
10.Ng5! Rf8
10...0-0? runs into the tactical problem of 11.Bxf7† Rxf7 12.Qc4.

After the text move Black avoids any immediate refutation but the longer-term future of his king
is far from rosy. White can choose between options such as 11.Bd2, with a long-term advantage,
and the sharper 11.h3!?, as favoured by the engine.

B) 1...g6 2.d4

The main lines of the Modern Defence will follow later. For now though, we will look at a
variety of offbeat lines and move orders which can occur after B1) 2...Nf6, B2) 2...d6 and B3)
2...Bg7.

B1) 2...Nf6

463
 This provocative move has been used by several top players – often in blitz, but now and again
at classical time limits as well.

3.Nc3
3.e5 Nh5 actually seems to be holding up okay for Black, who will play ...d6 next. The text
move seems simpler.

3...d5
This seems like the only serious way to avoid converting to normal paths with 3...d6.

4.e5 Ne4!?
This seems like the critical variation; it has been played several times by Rapport, as well as by
Carlsen in a blitz game.

The more common 4...Nh5 gives White a pleasant game after: 5.Be3 Ng7 (5...c6 can be met by
6.Qd2 b5 [6...Ng7 7.Bh6²] 7.Nf3 and it’s not clear what Black can do besides ...Ng7)

464
6.Bh6!? This seems simplest to me. Once the dark-squared bishops get exchanged, White is
more or less guaranteed a lasting edge. 6...Ne6 7.Bxf8 Kxf8 In Ochoa – Jaramillo, Manizales
2017, 8.Nf3N 8...Kg7 9.Qd2² would have been most logical.

5.Nce2!?
5.Nxe4 dxe4 may offer White chances for an edge, but I prefer to avoid the exchange and
leave Black’s knight as a target.

5...f6

6.c3!?N
I like this useful waiting move. Somehow the following alternatives were not quite working the
way I would have wanted:

465
6.f3 Ng5 7.Be3 (7.Bxg5 fxg5 8.h4N 8...gxh4 9.Rxh4 c5 10.c3 Nc6 gives Black enough
counterplay) 7...Nf7 8.exf6 exf6 and Black was okay in D. Radovanovic – Ostojic, Mosko 2015.

6.h4 fxe5 7.f3 Nd6 8.dxe5 Nf7 9.f4 c5!?N (9...Bg4 can be met by 10.Nh3!?N [10.Qd3 Qd7
11.Ng3 e6 12.Nf3 c5 13.Ng5 Nc6 14.c3 0-0-0÷ was not at all clear in Harikrishna – Vakhidov,
Dubai 2014] 10...c5 11.Nf2² when I like White) 10.h5 g5 seems pretty messy.

6...fxe5
Against other moves, White should be able to count on an improved version of the 6.f3 line
noted above. For instance:

6...c5 7.f3 Ng5 8.dxc5 Ne6 9.exf6 exf6 10.Be3² and Black’s structure has been weakened.

6...Bg7 7.f3 Ng5 8.Bxg5 fxg5 9.h4 gxh4 10.Rxh4 c5 Having already defended his centre with
c2-c3, White can play 11.f4² with a fine position.

7.f3 Nd6 8.dxe5 Nf7 9.f4

Compared to the 6.h4 line noted above, here we have played c2-c3 instead of h2-h4. The
benefit is that we are not committed to all-out aggression on the kingside, and can instead aim for
a positional edge.

9...c5 10.Nf3 Nc6

10...Bg4 11.Ng3 Nc6 12.Be2² also seems pleasant for White.

11.Be3 e6
It is hard for Black to do without this move indefinitely, but now of course the light-squared
bishop will find itself restricted for a long time.

466
Other moves such as 11...b6 can simply be met by 12.Ng3².

12.Ng3 Bd7 13.Be2 Qc7

In the event of 13...Be7 14.0-0 0-0 15.c4 d4 16.Bd2² White will benefit from the e4-square.

Parking the king on the queenside seems like Black’s most ambitious approach, though it also
carries some risk.

14.0-0 0-0-0 15.b4!?

This is not immediately necessary, but it fits the requirements of the position.

15...cxb4 16.cxb4 Bxb4 17.Rb1©

The open lines leading towards Black’s king should offer White a lasting initiative for the
pawn.

B2) 2...d6

467
 There is nothing out of the ordinary about this move, but we should consider the slightly
unusual follow-up whereby Black delays ...Bg7.

3.Nc3 a6!?
3...Bg7 leads to normal lines of course.

In the event of 3...c6 4.Be3 I can think of no good reason for Black to avoid ...Bg7 or ...Nf6 in
the near future, so a transposition to a later chapter is highly likely.

The text move is more interesting, as Black still has the option of converting to normal lines with
...Bg7 at any moment, and White has to be quite careful to avoid being tricked into a line falling
outside of the recommended repertoire. As we will see in Chapters 15 and 16, when Black goes
for the ...a6 set-up with the bishop already on g7, my main idea is to put pressure on Black with a
timely e4-e5. So as long as you keep that plan in mind, you should have no problem.

4.Nf3!
Straight away we control the e5-square, rather than opting for 4.Be3 b5 when we are at risk of
being ‘move-ordered’ as in the note to move 6 in the main line below.

4...b5 5.Bd3 Bb7

5...Bg7 6.Be3 converts to Chapter 15.

468
6.e5!N
We are playing in exactly the same spirit as in the more common variations with ...Bg7 having
already been played. Now it is hard for Black to complete development, and we don’t even need
to spend a tempo on Be3.

6.Be3 invites 6...Bg7 with a transposition to Chapter 16, but 6...Nd7! discourages our e4-e5 plan
(White can still consider it, but it is no more than an unclear pawn sacrifice). Thus, Black has
genuinely benefited from his weird move order delaying ...Bg7.

6...dxe5
6...Nd7? 7.e6 fxe6 8.Ng5± is horrible for Black – the lack of the ...Bg7 move proves costly as
there is no ...Nf8 defence available.

6...Bg7 gives White the option of transposing to variation A of Chapter 16 with 7.Be3, but 7.0-0!
gives us a better version of that line, as the bishop has other squares available and the e5-e6 plan
could be even stronger here.

7.dxe5 e6
7...Nd7 gives White a pleasant choice: 8.Be4² keeps an easy positional edge while 8.e6!? is a
perfectly reasonable sacrifice guaranteeing fine compensation.

After the text move White has many ways to continue. One sensible option is:

469
8.a4!? b4 9.Bg5 Be7 10.Ne4²
White clearly has the initiative and Black’s position doesn’t feel solid enough.

B3) 2...Bg7

This is the usual move of course.

3.Nc3
I find this move order the most suitable for reaching our desired set-up. In the rest of the
chapter we will analyse the quirky options of B31) 3...Nc6 and B32) 3...d5!?, followed by the
somewhat more common B33) 3...c6!? and finally a few non-standard options after the normal
B34) 3...d6.

470
Other options exist of course, but in most cases White gets an excellent position with natural
moves. Here are a few examples:

3...b6 4.Be3 Bb7 5.Qd2

It’s not clear what Black wants to do next. Most games from here have featured some sort
of Hippo set-up, with moves like ...d6, ...e6, and even ...h6 (in order to play ...Ne7 without
allowing Bh6) – but this seems really passive, and White has many good ways to play
against it. For instance:

5...e6 6.0-0-0 d6
6...h6 can be met in the same way, but White may also consider 7.d5!?N 7...exd5 8.exd5 d6
9.h4 Nd7 10.Nf3 Ngf6 11.Bb5± when the weakness of the h6-pawn makes it hard for
Black to castle.
7.h4 h6 8.h5 g5 9.f4 gxf4 10.Bxf4 Nd7 11.Nf3 a6
11...Ne7 12.Kb1 is similar and may transpose after a few moves.
12.Kb1 b5
In Guseinov – Kamsky, Khanty-Mansiysk (blitz) 2013, White could have capitalized on his
advantage with:

471
13.Bd3!N 13...Ne7 14.g4!+–
Intending g4-g5. Obviously there are some more moves to make, but Black is essentially
doomed because the kingside is opening up and he is unable to evacuate his king without making
serious concessions.

3...a6
This resembles variation B2 above, although it feels even stranger for Black to try and delay
...d6 for any meaningful length of time, as the b8-knight will usually want to go to d7 at
some point. Once again, we just need to keep in mind that Black may (and indeed is quite
likely to) transpose to normal lines by playing ...d6 pretty soon.
4.Be3
Unlike in variation B2, Black has already developed his bishop to g7 here, so there is no
reason for us to refrain from developing our bishop.
4...b5 5.Nf3
5.Qd2 d6! converts to a line outside our repertoire. Remember, when Black plays ...a6 we
avoid an early Qd2.

472
5...Bb7 6.Bd3 Nf6
6...d6 converts to Chapter 16; see page 359.
6...e6 is another version of the Hippo. White has excellent chances after 7.Qd2, followed by
the plan from the note on the previous page, or perhaps that shown in variation A1 of
Chapter 15 on page 340, according to circumstances.
7.e5 Nd5 8.Nxd5 Bxd5 9.a4 c6 10.h4 d6 11.h5 Nd7 12.h6 Bf8 13.0-0±
Haslinger – Vereggen, Dieren 2015.

3...c5 4.Nf3
Our simplest approach is to invite a transposition to the Accelerated Dragon. The present
position has actually been covered under the 1.e4 c5 2.Nf3 g6 3.d4 Bg7 move order in 1.e4
vs The Sicilian III – but looking at the position again, I decided to mention a few points of
interest which were missing from my previous coverage.
4...Nc6?!
Black should really prefer 4...cxd4 when 5.Nxd4 reaches the Accelerated Dragon, which
was covered in 1.e4 vs The Sicilian II.
4...Qa5 gives White a pleasant choice. 5.d5!? is a good move which I analysed before, but
perhaps it’s simpler to go for 5.Be3 Nf6 6.Qd2 cxd4 7.Nxd4 Nc6 8.f3 d6 9.0-0-0 when
White has a great version of a Dragon position. Not only did we avoid Bc4 (which is
usually necessary to prevent ...d5); but in the lines without Bc4, such an early ...Qa5 is not
considered the best way for Black to continue.
5.d5!
5.Be3 Qa5 has been analysed in 1.e4 vs The Sicilian III; but Black should of course take the
opportunity to convert to an Accelerated Dragon with 5...cxd4 6.Nxd4.
Although converting to a normal variation would be convenient in some ways, I think
White should really be looking to punish Black’s dubious move order.
5...Nd4

473
6.d6!?
White’s position looks incredibly promising and easy to play.
6...Qb6
6...e6 7.Nxd4 cxd4 8.Nb5± leaves Black in a mess.
6...exd6N 7.Nxd4 cxd4 8.Nb5 Qb6 transposes to 6...Qb6.
6...a6 7.Nxd4 cxd4 8.Nd5 exd6 9.Bf4± was great for White in Nadzamova – Koval,
Liptovsky Mikulas 2017.
7.Nxd4 cxd4 8.Nb5 exd6

9.Qf3! Ne7 10.Qa3 0-0 11.Bf4±

Kochetov – Volobuev, Moscow 2008.

B31) 3...Nc6

474
4.Be3
4.d5 is possible but I would prefer not to commit to it at this stage, as after 4...Nb8 Black
seems to be doing okay in most lines.

4...d5!?
This surprising lunge seems like the only serious way to give Black’s move order unique
significance.

4...d6 transposes to variation B34 on page 266.

5.Nxd5!
Unlike the 3...d5 variation which we will get to shortly, here Black doesn’t have an easy life
after getting the pawn back.

5.exd5 Nb4 6.Bb5† may look tempting, but after 6...Bd7 7.Bc4 Bf5 8.Bb3 Nf6 Black actually
seems to be doing fine.

5...e6 6.Nc3 Nxd4

6...Bxd4?! has not been played, and the exchange of dark-squared bishops should in principle
be favourable for White: 7.Bxd4 Qxd4 If the queens stayed on, Black’s dark-square weaknesses
could render his king unsafe. But after the queen exchange, White can still exploit Black’s lack
of development: 8.Qxd4 Nxd4 9.0-0-0 e5 (9...c5 10.e5±)

475
10.Nf3! Nxf3 11.Nb5! Rb8 12.Nxc7† Ke7 13.gxf3±

7.Qd2 e5
7...Nc6 is hardly an improvement in view of 8.Qxd8† Kxd8 (8...Nxd8 9.Nb5+–) 9.0-0-0†
when Black still needs to solve some awkward development issues: 9...Bd7 10.Nf3 h6 11.Bb5

476
11...Nge7?? 11...Kc8N is essential although after 12.Rd3± there is no doubt as to who has won
the opening battle) 12.Rd2 Kc8 13.Rhd1 Be8 14.Bc5 Bf8 15.Ne5+– Gomez Anadon –
Fernandez Lopez, Aragon 2007.

8.0-0-0
Black has no easy way to complete development. For instance:

8...c6N
Covering d5, but losing additional time.

8...Be6 was played in Mester – Voros, Hungary 2017, when 9.f4!Nƒ would have been unpleasant
for Black.

8...Ne7 allows 9.Bh6 0-0 (9...Bxh6 10.Qxh6±) 10.h4ƒ when White has an easy plan of attacking

477
on the kingside.

9.Nge2 Nxe2† 10.Qxe2 Qe7?

10...Qc7 is necessary although 11.Qc4± is obviously great for White.

The text move is what Black would like to play – but it runs into a tactical refutation.

11.Nb5! cxb5 12.Qxb5† Bd7 13.Qxb7 Rd8

My illustrative line ends with a final tactical stroke:

14.Rxd7! Rxd7 15.Bb5 Nf6 16.Rd1+–

Black pays the price for his slow development.

B32) 3...d5!?

478
 I have seen this move used quite effectively as a surprise weapon. If White doesn’t play in the
most principled ways, Black can often obtain a fine game. Fortunately for us, it doesn’t require
too much preparation to make life unpleasant for Black.

4.exd5 Nf6 5.Bc4 Nbd7

Black should bring the knight to b6 without delay.

If he starts with 5...0-0, then 6.Nge2 Nbd7 7.Nf4 Nb6 8.Bb3± simply preserves White’s extra
pawn without much fuss.

6.Bg5
We should be ready to eliminate the knight on f6 as soon as the other one arrives on b6.

479
6...Nb6
6...0-0 7.Bb3 Nb6 8.Bxf6 Bxf6 transposes to the note on 8...0-0 below.

7.Bxf6 Bxf6 8.Bb3 a5

This is Black’s most popular approach, looking to provoke some kind of queenside weakness.

8...c6
This just leads to a worse endgame for Black.
9.dxc6 Qxd4 10.Qxd4 Bxd4 11.Nge2

11...Bxc3†
I also checked 11...Be5!?N, when 12.0-0-0 (12.f4 Bd6 is not so clear after 13.cxb7 Bxb7
with some practical compensation, or 13.Ne4 bxc6 14.Nxd6† exd6) 12...bxc6 13.Rhe1 Bg7
(or 13...Bg4 14.f3) 14.Na4 seems like a straightforward route to an edge for White.
12.Nxc3 bxc6 13.0-0-0²
Quesada Perez – Rocha, Tromso (ol) 2014.

8...0-0 9.Nf3!N
This works in much the same way as in our main line below, except that here the a-pawns
remain untouched.
9.Nge2 a5 will soon transpose to one of the lines I prefer to avoid.
9...Bg4 10.h3 Bxf3 11.Qxf3 Bxd4 12.0-0-0

480
12...Bxc3
12...Bg7 13.h4 h5 gives Black a much worse version of our main line, as now he has no
counterplay on the queenside. Thus, after 14.g4! hxg4 15.Qxg4 Bxc3 16.bxc3 a5 17.h5+–
he gets smashed on the kingside.
13.Qxc3 a5 14.a3 Qd6 15.h4 h5 16.Kb1 Nd7 17.g4ƒ

9.a3!
I also considered 9.a4 0-0 10.Nf3!? (10.Nge2 c6 11.dxc6 bxc6 12.0-0 is the usual line)
10...Bg4 11.h3 Bxf3 12.Qxf3 Bxd4 13.0-0-0 Bxc3 14.Qxc3 Qd6 when it would be much nicer
for White to have the pawn on a3. The way things stand, Black has good chances to create
counterplay with ...Nd7, ...b5, ...Nc5 and so on.

9...a4 10.Ba2 0-0

White has played 11.Nge2 in a few games but I prefer:

481
11.Nf3!N 11...Bg4
11...c6 12.dxc6 bxc6 13.0-0 Bg4 14.Ne4 leaves Black a pawn down for not much
compensation. For instance:

14...Bxf3 15.Nxf6† exf6 16.Qxf3 Qxd4 17.c3 Qe5 18.Qxc6±

12.h3 Bxf3 13.Qxf3 Bxd4 14.0-0-0

White has a simple and straightforward plan of attacking on the kingside.

482
14...Bg7
14...Bxc3 15.Qxc3± gives Black no real chances on the queenside – the pawn on a4 is only a
weakness. Thus, White will dominate in all areas.

15.h4!
15.g4!? Qd6 gives Black chances to create some annoying threats. A sample line is 16.Ne4
(16.h4 Qc5) 16...Qd7 (16...Qe5 17.c3) 17.h4 Qb5 18.Nc3 Qc5 19.Rd3 c6 when White may still
be better with precise play, but matters are not easy.

15...h5
15...Qd6 is too slow: 16.h5 Qf6 (16...Qc5 17.hxg6 hxg6 18.Ne4 Qb5 19.c3 Nc4 20.Bxc4
Qxc4 21.Rh4+–) 17.Qg3ƒ

16.g4 hxg4 17.Qe3!

17.Qxg4 Bxc3 18.bxc3 Qd6 is not clear at all.

17.Qd3 c6! (17...Qd6 18.h5 g5 19.Kb1ƒ) 18.h5 Nxd5 19.hxg6 e6!÷ is also messy.

17...Qd6 18.h5 Qe5 19.hxg6

19.d6 g5 (19...Qxe3† 20.fxe3 cxd6 21.hxg6 Bxc3 22.gxf7† Rxf7 23.Bxf7† Kxf7 24.bxc3²)
20.dxe7 Rfe8÷ and once again the outcome is uncertain.

19...Qxe3† 20.fxe3 Bxc3

483
21.d6!?
21.gxf7† Kxf7 22.bxc3 also looks slightly favourable for White.

21...cxd6 22.gxf7† Rxf7 23.Bxf7† Kxf7 24.bxc3²

White’s structure may not be pretty, but with good technique the extra exchange should offer
some winning chances.

B33) 3...c6!?

4.Be3
It seems easiest to utilize the set-up that we play against ...d6 lines.

4.f4 is a alternative, when 4...d5 5.e5 is promising for White. Black could switch plans with

484
4...d6 when we are out of our repertoire, but if you have some knowledge of the Austrian Attack
you will know that the set-up with an early ...c6 is nothing special for Black, so this could also be
considered.

4.Nf3 maintains compatibility with our repertoire after 4...d6. However, 4...d5 is another branch
of theory. 4.Bc4 is the other main move, when 4...d6 5.Qf3 e6 6.Nge2 Nd7 is the main line.
Both of these paths are quite different from the main Pirc/Modern lines, and quite complex in
their own way. After the text move, on the other hand, it seems that we can get quite a pleasant
version of a typical structure.

4...d5 5.e5!
The key idea is to play against the g7-bishop. The big difference between this and the 4.f4 d5
5.e5 line is that White remains more flexible here, keeping the dark-squared bishop open and
maintaining more options, especially against set-ups involving ...h5 and ...Nh6. Black may also
provoke f2-f4 with ...f6, but this weakens Black’s kingside to some extent and restricts his
options. In general, we should be able to keep Black’s counterplay to a minimum and keep
playing against the bad g7-bishop.

5.Qd2 is another option that tries to make use of the early Be3, but after 5...dxe4 6.Nxe4 Nd7
Black seems to be doing quite well if he follows typical plans with ...Ngf6, ...Nb6 and so on,
intending to simplify the position.

We will analyse three main options: B331) 5...Nh6, B332) 5...Qb6 and B333) 5...f6.
5...h5 6.Qd2 Nh6 7.0-0-0 Nf5 8.Bf4 transposes to the first main line below.

B331) 5...Nh6

485
6.Qd2 Nf5 7.Bf4
7.0-0-0 also seems quite promising for White, but there is no need to give up the bishop at all.

7...h5 8.Nf3 Nd7

8...Bh6 9.Bxh6 Nxh6 10.h3!N± leaves a lot of nice squares on the kingside for White’s pieces,
which can be exploited with any of Ng5, Ne2-f4 or even Qf4. Meanwhile, Black’s knight will
not even be safe on f5.

9.Bd3
9.e6!? looks quite promising too, but it is not required.

9...Nf8 10.0-0-0N
A couple of other moves have been tried but I like this one, having a specific follow-up in
mind.

10...Ne6
10...b5 11.Kb1 a5 is nothing to worry about, and 12.Ne2± is excellent for White.

10...Nh6 is well met by 11.Ne2 Ne6 12.Kb1 Nxf4 13.Nxf4± and e5-e6 is coming up.

11.Bxf5!
This is critical – especially since it leads to a queen sacrifice!

11.Ne2 Nxf4 12.Nxf4 Bh6 helps Black: his pieces are no longer too passive, and ...Ng7-e6 may
follow.

11...Nxf4

486
 11...gxf5 12.Ne2 is excellent for White because he can keep playing against Black’s bad
bishops and pawn weaknesses.

12.Bxc8 Bh6
The next few moves are forced.

13.Bxb7 Ne6 14.Bxc6† Kf8 15.Qxh6† Rxh6 16.Bxd5

White has two powerful minor pieces and three healthy pawns for the queen, so it barely
counts as a sacrifice.

16...Rc8
After 16...Nc7 17.Bxa8 Qxa8 18.d5± the pawn will soon get to d7.

17.Bxe6 fxe6 18.Ng5 Rc6 19.Nce4±

Black is still in the game because of his extra queen, but White has a dominating position and
can play for a win with no real risk, as long as he is not in too much of a hurry. White’s ideas
include slowly expanding on the queenside with b2-b3 and c2-c4, and/or preparing d4-d5, while
Nc5 can be thrown in at any moment.

B332) 5...Qb6

487
 Trying to create some discord in White’s queenside.

6.Na4
I am not wildly keen about putting the knight on a4 but it doesn’t change things too much, and
later it may return to the centre via c5-d3.

6...Qa5†
The position is pretty similar after:
6...Qc7 7.Qd2 Nd7
7...f6N 8.f4 does not really help Black, for instance: 8...Nh6 9.Nf3 Nd7 10.h3 b5 11.Nc3
Nb6 12.Bd3 0-0 13.0-0 Nf5 14.Bf2 Bh6 15.g4 Ng7 16.Be3±
This position has been reached in a few games. I suggest:

8.Bf4!?N

488
 This seems like a simple and effective way to make use of the fact that f2-f4 has not been
played.
8.h4 is not so clear after: 8...f6 9.f4 Nh6 10.h5 Nb6 11.Nc5 Nc4÷
8...Qd8
8...e6 would leave Black’s bishop forever stuck behind the pawns, while also preventing a
future ...Nf8-e6, so it is definitely not something he wants to do.
9.h4 f6 10.Nf3
The e5-pawn is easily maintained without any need for f2-f4. It’s not clear how Black will
continue developing, and White has a strong initiative with h4-h5 coming up.

7.c3
It’s important that Black cannot cut off the white knight with ...b6, since his queen would be
trapped.

7...f6
If Black prepares ...b6 with 7...Qc7, White has time to bring the knight to a nice square with
8.Nc5 and later Nd3, or possibly Nb3.

8.f4 Nh6 9.Nc5

9.Nf3 is also possible, and should lead to similar positions.

9...Qb6
9...Ng4 10.Bd2 does not help Black, for instance: 10...Qc7 11.Be2 Nh6 12.Nf3 b6 13.Nd3 0-0
14.0-0±

9...0-0!?N may be the best that Black can do, although 10.Nf3 Qc7 11.Be2 b6 12.Nd3² is
pleasant for White.

489
10.Qd2 a5 11.Nf3 a4

12.Be2 Qd8 13.h3 b6 14.Nd3 Na6 15.g4±

We have been following Bologan – Iordachescu, Dresden 2006. White is building up in a
menacing fashion on the kingside, whereas Black has spent several tempos on the queenside with
no apparent benefit.

B333) 5...f6 6.f4

White’s pawn centre is robust enough; and with accurate play, we should be able to utilize the
space advantage to develop a kingside attack.

6...Nh6 7.Qd2 0-0 8.0-0-0

490
 Castling immediately seems simplest and most flexible.
With that being said, it is worth pointing out that White should have the better chances even
after playing a slightly inaccurate move order:
8.Nf3 Bg4 9.Be2 Nf5
White’s move order has given Black this additional way of arranging his pieces. I will show
a few more moves in order to illustrate a positional theme, which will crop up again in my
main recommended line below.
10.0-0-0 Bh6?!
Objectively this is not the best move, but in a way it is the most important option to
consider, as it seems like Black’s only serious attempt to create immediate tactical threats. If
Black plays more slowly, then White should be able to count on a solid edge.
11.g3 Nd7

12.Nh4!N
12.Kb1? Bxf3 13.Bxf3 fxe5 14.dxe5 Nxe5µ illustrates Black’s threat.
12.exf6?! does not lose material but is a clear positional concession, and 12...exf6 was
pleasant for Black in Choisy – Poley, Agneaux 2004.
12...Bxe2
12...Nxh4 13.Bxg4 is also excellent for White.
13.Nxf5 Bxd1 14.Nxh6† Kg7 15.Nxd1 Kxh6 16.f5† Kg7 17.Bh6† Kh8 18.fxg6 hxg6 19.Bxf8
Qxf8 20.Rf1±
Having seen how White can deal with the central pressure, even when Black seemed to have
arranged his pieces optimally, let’s return to our main line where we avoid having an early Nf3
being met by ...Bg4.

491
8...Ng4
8...e6 invites 9.Nf3 since ...Bg4 is no longer an option. After the further 9...c5 10.dxc5 Nc6
11.Bb5 Ng4 12.Bxc6 bxc6 13.h3 Nxe3 14.Qxe3± Black’s bishops were rather subdued, so she
had no real compensation for the pawn in Skripchenko – Rudolf, Kishinev 2005.

8...fxe5 9.fxe5

In general, this exchange just simplifies White’s task by clarifying the pawn structure nice
and early. However, it is worth emphasizing an important finesse in the move order, which
is mirrored in the main line below.
9...Ng4
I will mention two options. First we will see why a seemingly natural move is slightly
inaccurate, before moving on to the correct continuation.

492
a) 10.Nf3 Nxe3 11.Qxe3
White has chances to fight for an advantage here, but Black obtains more counterplay than I
would wish to allow after:

11...Bg4!
It is worth mentioning that trying to attack on the queenside with 11...b5?! is the wrong
idea, as White’s attack tends to be much faster. For instance: 12.Kb1 Na6 13.h4 Bg4
14.Be2 Qd7 15.Nh2!?N (15.Ng5 Nc7 16.h5ƒ was also promising for White in Tan –
Petrov, Ascot 2014) 15...Bxe2 16.Nxe2 Nc7 17.h5± Black’s queenside play has not even
started and White’s knights are ready to trample all over the kingside.
12.Be2 e6 13.Kb1
13.Ne1 Bxe2 14.Nxe2 is met by: 14...Qh4! Activating the other bishop. 15.Kb1 Nd7
16.Nf3 Bh6=
13.h4!?N is interesting but Black obtains adequate play with 13...c5! intending ...Nc6, when
it is hard to keep the centre intact, and the g7-bishop has found something to do again.
(13...Nd7?! allows White to obtain the upper hand with: 14.Ne1! Bxe2 15.Nxe2 c5 16.Kb1
Qe7 [16...cxd4 17.Nxd4 Qe7 18.Nd3±] 17.Nd3 c4 18.Nf2±)
13...Nd7 14.Ng5 Bxe2 15.Nxe2

493
15...Bh6!
Once Black activates or exchanges this bishop, nothing is wrong with his position anymore.
16.h4 Qe7 17.Nf4 Bxg5 18.hxg5 Rf5=
Black was fine in Dominguez Perez – Anton Guijarro, Linares 2013.

b) Instead, 10.Kb1!N is a simple improvement which mirrors the main line below. Black
struggles to find a good role for his light-squared bishop, so there is simply no reason to allow it
to pin a knight on f3.

9.Kb1!
Just as in the note above, this king move makes the most sense. Delaying ...fxe5 keeps a bit
more tension in the position but Black essentially has all the same problems, especially when it
comes to finding suitable roles for his bishops.

494
9...Nxe3
9...Bh6 is an extra option due to delaying ...fxe5, but 10.g3² keeps control.

I also considered:
9...b5N
Planning ...Nd7-b6, with some counterplay on the queenside. Maybe if Black had a couple
of spare tempos he would have been able to develop a meaningful initiative; but as things
stand, he is too slow to create any real problems for White.
10.Nf3 Nd7
In the event of 10...Nxe3 11.Qxe3 Nd7 12.h4! Nb6 (or 12...h5 13.Bd3+–) 13.h5+– White is
too fast on the kingside.

11.Bg1!
11.h3 Nxe3 12.Qxe3 Nb6 is not so clear.
11...Nb6
11...Bh6? loses material after 12.e6 (or 12.h3).
11...fxe5 does not help Black, as 12.dxe5 leaves him vulnerable to Nxd5 tricks, as well as
the usual kingside advances.
12.h3 Nh6 13.g4
White is completely dominating on the kingside and it is hard for Black to do much on the
other flank. A few illustrative lines are:
13...b4
13...Nc4 14.Bxc4 bxc4 15.Be3 Nf7 16.Na4±
14.Ne2 Nc4 15.Qe1 a5

495
16.Nc1! Qb6 17.Nd3 Rb8 18.Nc5±
White is doing great on the queenside as well.

10.Qxe3 fxe5 11.fxe5 Bg4

11...c5 12.dxc5 gives Black nothing because the d5-pawn is hanging too.

11...b5 is met by the typical 12.h4 Na6 (or 12...h5 13.Bd3±) 13.h5± when the attacking race can
only have one winner.

12.Be2!N
12.Nf3?! was played in Sun Chin – Hemmat, Montevideo 2016, but we already know from the
earlier notes that this would be a positional mistake.

In the lines where an early Nf3 was answered by ...Bg4, we saw that White would often play

496
Be2 and move the knight again in order to exchange bishops – so to exchange them without
losing time represents an obvious improvement.

12...Qd7
In the event of 12...Bxe2 White gets a clear advantage with either knight recapture;
13.Ncxe2!? would be my slight preference, so that both knights will be able to join in the
kingside massacre.

13.h4!?‚
13.h3 Bxe2 14.Ncxe2!? is also good, with similar ideas as in the note above. Either way, White
has every chance of success on the kingside.

B34) 3...d6 4.Be3

4...Nc6

If Black does not wish to play a main line with 4...Nf6, 4...c6 or 4...a6 (all of which will be
analysed in later chapters), the text move is the most interesting option. Black intends a quick
...e5, but we can conveniently meet this plan by maintaining the tension and castling on the
queenside.

The only other option worth mentioning is:

4...Nd7!?
White has to be careful with the move orders here. As long as you know what you are
aiming for against both the ...a6 and ...c6 systems, the solution is easy.
5.Nf3!
5.Qd2 would fit in with our repertoire after a subsequent ...c6, but 5...a6!? leaves us unable

497
 to carry out the plan recommended in Chapter 15.
5...a6
5...c6 6.Qd2 converts to variation A of Chapter 13 on page 308.
Obviously Black has other legal moves available, but they aren’t really in the spirit of the
Pirc/Modern and there is no real need to prepare against such ideas.
6.Bd3
We have transposed to variation A of Chapter 15 on page 340.

5.Qd2 e5 6.Nge2 Nf6

Black has nothing better, for instance:

6...f5 7.exf5 Bxf5 gives White a pleasant choice: 8.d5 Nce7 9.Ng3² gave him a solid positional
plus in Palac – Kosanski, Makarska Tucepi 1995; and maintaining the tension with 8.0-0-0!?
looks good as well.

6...Nge7 7.0-0-0 0-0 8.d5 Nb8 loses time for Black, and after 9.f3 White is ready to advance on
the kingside.

7.f3 0-0
7...exd4 8.Nxd4 0-0 9.0-0-0 is a simple transposition.

8.0-0-0 exd4
It is hard to suggest a good way for Black to maintain the tension. For instance, after something
like 8...a6 then 9.d5 Ne7 10.g4 b5 11.Ng3± gives White easy kingside play, while Black is a
long way from threatening anything on the queenside.

9.Nxd4

498
 We have transposed to a variation of the Philidor which has already been analysed in my 1.e4
repertoire – see page 550 of 1.e4 vs The French, Caro-Kann & Philidor. I will just mention a
couple of options for Black which were omitted from my previous analysis.

9...Nxd4 10.Bxd4 Be6 11.g4 c5

11...c6 12.h4ƒ

11...a6 12.g5 Nh5 13.Bxg7 Nxg7 14.h4 Qe7 15.f4± Fercec – Jurkovic, Zagreb 2010.

12.Be3 Qa5 13.Bh6

13.Kb1 is too slow in view of: 13...b5! 14.Bh6 b4 15.Nd5 Nxd5 16.exd5 Bxd5 17.Qxd5
Bxh6=

13...Bxh6
Black’s other moves were covered in my previous work, but the text move was not mentioned.
It has been played in quite a few games so I decided it would be worth including it here – but
with accurate play White is close to winning.

14.Qxh6 b5
14...Bxa2 15.h4 Be6 16.h5 Qc7 occurred in Hennigan – Westerinen, Gausdal 1995. Out of
several winning continuations, I prefer 17.Ne2N+– intending g2-g4 or Nf4 next.

15.Bxb5
15.Kb1? is completely the wrong idea, as becomes clear after: 15...Rfe8! 16.h4 b4 17.Nd5
Nxd5 18.exd5 Bxd5µ

15...Rab8

499
16.h4!N 16...Rxb5 17.Nxb5 Qxb5 18.h5 Rb8 19.b3
White is close to winning: his king is completely safe, and Black will have to make serious
concessions in order to avoid being mated on the kingside.

Conclusion

We began our Pirc/Modern coverage by considering the slightly unusual 1.e4 d6 2.d4 Nf6 3.Nc3
c6, which is conceptually related to the Philidor. 4.Nf3 is a good choice for us, after which
4...Bg4!? invites us to advance the kingside pawns and swap off Black’s light-squared bishop,
which should offer White an edge; we then looked at 4...Qa5 followed by ...e5, which is a
characteristic plan for the Czech Pirc, but which enables White to gain a useful initiative by
using the queen as a target.

We then turned to 1...g6 2.d4, when there are a number of slightly offbeat variants of the Modern
Defence. 2...Nf6 is an interesting one, when 3.Nc3 d5 4.e5 Ne4 5.Nce2 is my suggestion,
supported by 5...f6 6.c3!?N which seems like a useful novelty – although Black’s position is
generally not so bad in this under-rated sideline. We then turned to 2...d6, which usually
transposes to normal lines but could lead to some interesting move-order subtleties if Black
meets 3.Nc3 with 3...a6 followed by a quick ...b5 while leaving the bishop on f8. As long as you
remember to develop with 4.Nf3 rather than Be3, you should have no trouble maintaining
compatibility with our repertoire from later chapters – and perhaps even obtaining an improved
version of it if Black continues to leave the bishop on f8.

Next is the most common 2...Bg7 3.Nc3, when Black has an array of options. 3...d5!? is an
interesting gambit; after capturing the pawn and playing a few standard moves, I found some
useful subtleties, such as meeting ...a5 with a2-a3 (rather than the more common a2-a4) and
developing the knight to f3 rather than e2, all with a view to castling on the queenside and

500
launching an attack. Another important option is 3...c6!?, when 4.Be3 d5 5.e5! seems like a
convenient reply; Black has several options but in all cases White maintains control. It is worth
reiterating that, following 5...f6 6.f4, White can make the most of his position by castling long
and delaying the development of the king’s knight to f3, at least until such time as Black does
not have the convenient option of ...Bg4. And finally, the most common 3...d6 usually leads to
the main lines of later chapters; but should Black wish to deviate with an early ...Nc6 and ...e5,
White should only be too happy to convert to a Philidor variation where we have a huge kingside
attack.

501
A) 6...Bg4 275
B) 6...Qa5 277
C) 6...0-0 279
C1) 7.h3!? 279
C2) 7.Bd3 283
D) 6...b5 7.Bd3 Nbd7 288
D1) 8.0-0 291
D2) 8.Ne2! 294
D21) 8...c5!?N 296
D22) 8...Qc7 297
D23) 8...Bb7 299
D24) 8...0-0 9.0-0 Qc7N 10.h3 300
D241) 10...e5 301
D242) 10...c5 302

502
1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3
I prefer to put the bishop on this square on move 4 against both the Pirc and Modern move
orders. On the next turn, we can choose between Qd2 and Nf3, according to what Black does.

4...Bg7
A relevant alternative is:
4...c6
Delaying ...Bg7 makes some sense against plans involving f2-f3 – but against the set-up
with Qd2, Nf3 and Bd3 that we are choosing, it should not affect things too much.
5.Qd2 b5 6.Bd3 Nbd7
Black has made some natural developing moves but there is not much else that he can do to
continue delaying ...Bg7. There have been some games involving ...e5 and ...Be7 – but to
me, the bishop just feels better placed on g7 instead of e7. It is important to recognize that
these structures after ...b5 and ...e5 tend to resemble the Spanish, especially after the Ne2-
g3 plan that I often recommend. In the Spanish, it is typical to reroute the e7-bishop to g7,
so it is hard to believe that Black should avoid developing it directly to g7 when he has the
chance. A final point, which will become clearer as we work through the next few chapters,
is that we are usually happy to see the combination of ...b5 and ...e5 being played, as we end
up with convenient targets on the queenside after dxe5 – so the trickiest move orders tend to
be those where Black avoids committing to one or the other of those moves.
7.Nf3 e5
7...Bg7 transposes to variation D.
As explained above, by committing to ...b5 and ...e5, Black is essentially giving us what we
want. The only caveat to this is that Black’s dark-squared bishop exerts more influence on
the queenside when it avoids going to g7, so White’s queenside initiative may not pack the
same punch as in similar positions with the bishop on g7 – but even so, I think the negative
features of Black’s chosen move order outweigh the positives.

503
8.dxe5!
8.0-0 is sensible but I see no reason to give Black the choice between 8...Bb7 and 8...exd4!?
9.Nxd4 Bb7.
8...Nxe5!?N
8...dxe5 9.h3² has occurred in many games, with a significant plus score for White. It would
be possible to analyse variation after variation from here, but I don’t see too much value in
doing so. Most Pirc players avoid this position these days, and we will reach the same pawn
structure in some other lines – so as long as you have a reasonable grasp of the general
plans, you will have every reason to be happy from here.
The text move is worth mentioning as the engine suggests it as an improvement, but White
keeps the upper hand after:
9.Nxe5 dxe5 10.f3²
With a lead in development and a convenient plan of targeting Black’s weaknesses on the
queenside.

5.Qd2
5.Nf3 is a popular move order which may well transpose, but it is more flexible to maintain the
option of f2-f3, just in case. It may seem as if the early Qd2 implies that 0-0-0 is on the way, but
this is not the case, because the queen usually belongs on d2 even after Nf3, Bd3 and 0-0 have
been played.
While we are on the subject of move orders, I would like to emphasize the importance of
playing as flexibly as possible. One of the advantages of the Pirc/Modern complex is the sheer
variety of options and move orders available to Black. This can be really annoying for some
White players to deal with, especially if they don’t have a well-constructed repertoire. That’s
why I consider it especially important for White to keep the options of castling on either side. In
general, I am not so keen on committing to 0-0-0 when Black avoids ...0-0 and arranges a quick
...b5 (with either ...c6 or ...a6) – which happens to be one of the most popular plans for Black. So

504
when Black plays ...b5 before castling, we will revert to short castling. As we will see, White’s
set-up with 0-0 has more sting to it when Black has played ...c6 and ...b5, as we will be able to
strike with a timely a2-a4.

5...c6
This move is the defining theme of the chapter, so of course we take it as the main line. Before
discussing it in detail, we will quickly deal with a few alternatives.

5...a6 is a major option which will be covered in Chapter 14.

5...Ng4?! achieves nothing after 6.Bg5 h6 7.Bh4, when our bishop easily evades capture and
Black’s knight has been misplaced. If Black follows up with ...g5, he weakens himself even
more.

5...Nc6?! would not be so bad if White had already committed to Nf3, but 6.f3 e5 7.Nge2 shows
the advantage of White’s flexible move order. We have transposed to variation B34 from the end
of the previous chapter, which in turn is likely to transpose to a dubious version of a Philidor, all
of which is great news for us.

The only other option worth analysing is:

5...0-0?!
With this, Black commits his king too quickly, and we should not pass up the opportunity to
play for a straightforward attack on the kingside.

6.0-0-0
It is worth starting with this move just to rule out the possibilities of ...e5 or ...c5.
6...c6
6...a6 7.f3 b5 8.Bh6‚ followed by h2-h4 gives White something resembling a Yugoslav

505
 Attack against the Dragon, but with the advantages of a powerful pawn centre and no open
c-file for Black.
7.f3 b5 8.g4!?
The text move is criticized by Vigus in a chapter of Dangerous Weapons: The Pirc and
Modern, but it actually turns out rather well for White.
8.Bh6 and 8.Kb1!? are also promising; it is hard to say exactly which continuation is best,
but it’s clear to me that White has a pleasant choice.
8...Qa5
8...b4 9.Nce2 Qa5 10.Kb1 c5 is another possible attempt to find counterplay, but 11.Bh6
Nc6 (11...Bxh6 12.Qxh6 Nc6 13.h4ƒ; 11...Ba6 12.Bxg7 Kxg7 13.d5ƒ) 12.Bxg7 Kxg7
13.d5 Ne5 14.Ng3 c4 15.g5± left White in control in Soeder – Full, Massbach 2001.
9.Kb1 Be6
Vigus recommended this but overlooked White’s reply:

10.Nd5!
Killing Black’s attack and leading to a favourable endgame in one way or another.
10...Qxd2
Black can insist on keeping the queens on with 10...Qd8? 11.Nxf6† exf6 12.h4 Re8 as seen
in Dang – H. Nguyen, Ho Chi Minh City 2009, but 13.h5N 13...g5 14.d5+– leaves him with
a disgusting position.
11.Nxe7†!
It may look risky to put the knight in harm’s way, but it soon becomes obvious that White is
in no danger of losing material.
11.Nxf6† exf6 is nothing special.
11...Kh8 12.Rxd2!N
A simple improvement over 12.Bxd2, as played in Widmer – Pinol Fulgoni, Pontresina
2000.
12...Re8 13.g5 Rxe7

506
 Moving the knight to h5 or d7 would be pointless due to 14.d5 or even 14.Nxc6!?.
14.gxf6 Bxf6 15.h4 h5 16.Nh3
16.Bf4 also seems promising.
16...Nd7 17.Ng5±
White’s powerful pawn centre makes him the clear favourite, irrespective of the queen
exchange.

6.Nf3
Clearly, with this move we give up on the f2-f3 set-up, but not necessarily on the plan of
attacking on the kingside, which can still be quite potent with the knight on f3. For the time
being we will remain flexible, awaiting further information from Black’s side before choosing
the most appropriate plan.

6.0-0-0 is an important option of course, but my general feeling is that Black should be doing
quite well if he avoids castling and focuses on the queenside with moves like ...Nbd7, ...b5,
...Nb6 and ...b4, which raises some annoying problems for White. Should you choose to
investigate this in more detail, be warned that the computer often prefers White in these lines; but
as the practical games show, things are actually a lot more complicated.

A tempting alternative is:

6.Bh6!?
This was recommended by John Shaw in the first of his Playing 1.e4 books, published in
2016 – also by Quality Chess of course, so it was natural for me to explore a different
direction; and besides, I had some nice ideas in mind when choosing the Nf3 set-up. Still,
there is no denying that the text move is quite challenging.
6...Bxh6 7.Qxh6 Qa5
Shaw does a good job of dealing with Black’s alternatives; in this brief summary, I will
focus on the critical main line.

507
8.Bd3 c5
8...Nbd7 9.Nf3 c5 10.0-0-0 gives White a promising version of a Dragon.
9.d5 Nbd7
9...c4 10.Bxc4 Nxe4 (10...Qc5 runs into 11.Bb5†) 11.Nge2 seems good for White.

10.Nf3 c4 11.Bxc4 Qc5

Setting up a neat tactic – but not necessarily solving Black’s problems.
12.Bd3 Qxf2† 13.Kxf2 Ng4† 14.Ke2 Nxh6 15.h3
This endgame is still being debated, with Motylev – Levin, Yaroslavl 2019, a relatively recent
example. As the engines have got stronger, they seem to be finding ways of putting Black’s
position in this variation under more pressure than seemed possible a few years ago; so if for any
reason you want to explore another direction for White, this would seem like a good place to
look.

Returning to my main recommendation, we will analyse four options in detail: A) 6...Bg4, B)

6...Qa5, C) 6...0-0 and D) 6...b5.

As usual, 6...Ng4 is of little use to Black, because White can simply play 7.Bg5, when any
further chasing of the bishop will only weaken Black’s position.

The only other option worth discussing is:

6...Nbd7!? 7.Bd3
7.0-0-0 is too early for my taste, for reasons already explained.
7.Bh6!? is worth considering, with a possible transposition to one of the 6.Bh6 lines from
the note to White’s previous move.
7...e5
This seems like the most consistent follow-up to Black’s last move.
8.Bh6

508
 I briefly considered 8.dxe5 dxe5 9.Bc4 but after 9...b5 10.Bb3 Qe7 11.0-0-0 a5 White does
not seem to have anything, despite the vague feeling that the active pieces might conjure
something.
It is worth mentioning that 8.0-0-0 Ng4! 9.Bg5 f6 enables Black to hunt down the valuable
bishop, as 10.Bh4?? Bh6–+ would be embarrassing. So please remember that when the king
lands on c1, the ...Ng4 idea might gain in strength.

8...Bxh6
8...0-0 9.Bxg7 Kxg7 10.0-0-0 b5 actually transposes to the later variation C2; see page 285
for the continuation.
9.Qxh6 exd4 10.Nxd4 Qb6 11.Nce2
Avoiding 11.Nb3? Qxf2†!.

11...c5
11...Ng4 occurred in Pichot – Flores, Buenos Aires 2015, when 12.Qg7N 12...Rf8 13.h3

509
 would have been good for White.
11...Qa5† 12.c3 Qh5 is possible, but 13.Qd2² leaves Black with a weak d6-pawn, and weak
on the dark squares generally.
11...Qxb2N is more ambitious but 12.0-0© gives White obvious compensation for the pawn.
The text move was played in Boguszlavszkij – Valdes Romero, Harkany 2002. I would
suggest:
12.Nb5N 12...a6 13.Nbc3²
Black still cannot castle, and the d5-square is an inviting target for the white knights.

A) 6...Bg4

7.Nh4!?
7.Be2 followed by h2-h3 is perfectly possible of course. I think that in the ensuing positions
after Black exchanges on f3, White should have some kind of long-term edge – but of course it
isn’t much, and a lot of Pirc players seem to be content playing such positions.

The text move is more ambitious.

7...b5 8.f3!?N
Vigus only considers 8.Bd3 when covering White’s last move in a note in Chess Developments
– The Pirc.

8.h3N 8...Bc8!? 9.Bd3 Nbd7 10.Nf3 is similar to the main lines we will analyse after 6...b5, but
in this version White has played h2-h3 slightly earlier than I would like. (It is also worth noting
that 10.f4 can be met by 10...e5! when I think Black is doing okay.)

8...Bc8 9.g4

510
 White’s set-up may appear a little strange, but the analogous position with the knight on g1 is
actually pretty common. In our version, Black has essentially wasted two tempos. Of course, one
might argue that the knight on h4 is an impediment to White’s attack – but if Black castles, then
Nf5 ideas start to appear.

9...Nbd7
9...0-0 10.0-0-0ƒ makes White’s life easier.

It seems to me that Black’s only really challenging approach is to avoid ...0-0 altogether, instead
favouring a plan such as ...Nbd7-b6, ...b4, ...a5 and so on. However, it turns out that White can
deal rather conveniently with those ideas, and the knight on h4 might even come in handy.

10.0-0-0 Nb6 11.Bh6 Bxh6 12.Qxh6 b4 13.Ne2 a5

511
14.Nf4! g5!
14...a4 15.g5! Nfd7 16.Nhxg6! shows the value of the knight on h4.

I also considered 14...e6 15.g5! Ng8 16.Qg7 Qxg5 17.Qxh8 Qxf4† 18.Kb1 Qxh4 19.Qxg8†
Ke7 20.Qg7± when White has won the exchange for a pawn.

15.e5!
But not 15.Qxg5?? h6!–+ when the queen will be caught.

15...gxf4
15...dxe5 16.dxe5 Qc7 is beautifully refuted by:

17.Nfg6! fxg6 18.Nxg6 Rg8 19.Nxe7! Qxe7 20.exf6 Qf7 21.Bb5 cxb5 22.Rhe1† Be6 23.Rd6+–

16.exf6

512
16...Be6
16...exf6 17.d5! Nxd5 18.Rxd5 cxd5 19.Bb5† Bd7 20.Re1†+– is another attractive kill.

17.fxe7 Kxe7 18.Qxf4²

For the time being White is a pawn up, and the knight will be excellent on f5.

B) 6...Qa5

This can be regarded as a prophylactic measure against an early 0-0-0 or Bh6. However, as we
will see in various lines throughout the chapter, I prefer short castling in many cases, so Black’s
scheme does not worry me too much.

7.h3

513
 It is also worth considering:
7.Bd3!?
Attempting to save time by avoiding h2-h3 altogether. It is hard to say how beneficial this
is, and the main line seems sufficient to get an edge, but I will present some lines anyway to
allow you to make up your own mind.
7...0-0
Immediate attempts to utilize the g4-square get nowhere:
a) 7...Ng4 8.Bg5 is, as usual, pointless for Black.
b) 7...Bg4 8.Nh4 0-0N (8...Qh5 9.g3 does not help Black; 8...e5 9.h3 Be6 10.f4 exf4
11.Bxf4± A. Ivanov – Yedidia, Warwick 1998) 9.h3 Bc8 10.f4²
8.0-0 Nbd7
Other possibilities include:
a) 8...Qc7 loses time, and 9.a4!? a5 10.Rad1 Nbd7 11.Be2 b6 12.Ng5 Bb7 13.f4± looks
excellent for White.
b) 8...Bg4N 9.Nd5 Qxd2 10.Nxf6† followed by 11.Nxd2 offers White a slight edge.
c) 8...e5 9.dxe5 dxe5 10.h3 offers White a nice positional edge, for instance: 10...Qc7
(10...Nbd7 actually transposes to our main line below, where 11.Nd5 is the continuation)
11.a4 Nbd7 12.Bc4±

9.Nd5 Qd8 10.Nxf6† Nxf6 11.h3 Qc7

This position was reached in Wirthensohn – Welz, Germany 1993. Here I would suggest:
12.c3N²
White has a typical slight edge, although the removal of a pair of knights gives Black a bit
more breathing room.

7...0-0 8.Bd3
With Black’s queen on a5 and having already committed to h2-h3, we will of course be
castling on the kingside.

514
8...Nbd7 9.0-0 e5
I also checked:
9...Qc7
Black’s position appears solid of course, and it may not seem immediately clear how White
is supposed to improve. We do have a few interesting plans though.
10.a4!?
10.Ne2 e5 11.c3 d5! is just fine for Black.
10...a5

11.Ng5!N
11.d5 was played in J. van Foreest – Bachmann, Stockholm 2016, but this plays into
Black’s hands, as it gives him much more scope for activity.
11...h6
If Black ignores White’s play, for instance with 11...b6, then 12.f4ƒ gives White a menacing

515
 presence in the centre, with e4-e5 a serious possibility.
11...e5 leads to an opening of the centre, which will benefit White, especially as the g5-
knight will be a huge pain for Black to deal with: 12.dxe5 dxe5 (perhaps 12...Nxe5 should
be preferred although 13.Be2 definitely favours White) 13.Bc4± Black seems severely tied
down. Next White can either go Rad1 and Qd6, or open the kingside with f2-f4.
12.Nf3 Kh7 13.Nh2 e5 14.f4ƒ
The annoying knight has been kicked away from g5, but having the pawn on h6 instead of h7
weakens Black’s kingside considerably.

10.dxe5! dxe5
The rare 10...Nxe5 is preferred by the computer but the position after 11.Be2N remains
slightly more pleasant for White, for instance: 11...Nxf3† 12.Bxf3 Rd8 13.Rad1 Be6 14.Bd4²

11.Nd5 Qd8
11...Qxd2 12.Ne7† Kh8 occurred in Chandana – Khegay, New Delhi 2007, when 13.Nxd2²/±
would have left White in control.

12.Nxf6† Nxf6

516
13.a4!
Black must either allow the pawn to advance and gain further space, or weaken his own
queenside with ...a5.

13...Qc7
13...Be6 14.a5² gives a nice bind on the queenside.

13...a5 gives White a few good options, including the surprising 14.b4!?, when the splitting of
White’s queenside pawns does not matter a great deal, as his pieces are so well placed. For
instance: 14...Re8 15.bxa5 Qxa5 16.Qxa5 Rxa5 17.Nd2 Ra8 (17...Be6 18.Rfb1±) 18.Nc4²

14.a5 Nh5
This position was reached in Cech – Biolek, Czech Republic 2007. I recommend:

15.Rfd1N 15...Rd8 16.Qc3 Nf4 17.Bf1²

White keeps a pleasant edge, just by playing simple chess.

C) 6...0-0

517
 It is still not clear if White will want to castle on the kingside or queenside, so I prefer to keep
things flexible for now. If Black opts for ...b5, we can just follow the same plans as in variation
D below. However, this is generally not what Black intends with an early ...0-0. After all, if he
intended to play ...b5 anyway, he may as well have delayed castling in order to be better placed
against any 0-0-0 plans. Black’s move order comes into its own when White castles on the
kingside, in which case Black can go for moves like ...Nbd7, ...e5 and ...Re8, perhaps combined
with ...Qc7, ...b6 and so on. In other words, he can avoid playing ...b5 altogether, which makes
his position harder to crack open with the a4/c4 pawn levers. Thus, in the upcoming variations
we will only consider this type of plan from Black, since the plans with ...b5 will be covered
under variation D.

We will consider C1) 7.h3!? followed by the more common C2) 7.Bd3.

C1) 7.h3!? Nbd7

7...b5 8.Bd3 Nbd7 9.0-0 converts to a position reached on page 291, via the move order 6...b5
7.Bd3 Nbd7 8.0-0 0-0 9.h3; it’s not my main suggestion for White, but you can find it analysed
in depth anyway, as I think it could still be worth considering.

518
8.Bh6
After 8.Bd3 e5 White does not have time to arrange Ne2 and c2-c3, which is what we ideally
want.

White can also switch to a positional plan with 8.a4!?, preparing against the possibility of ...e5 –
but the trouble is deciding what we should do if Black avoids this move. For instance: 8...Qc7
(8...e5 9.dxe5 dxe5 10.Qd6! suits White perfectly) 9.Be2!? b6 10.0-0 Bb7 11.Rfd1 (11.a5 b5
12.a6 Bc8÷; perhaps 11.Ng5!? e5 12.dxe5 dxe5 13.Bc4 could be considered) 11...Rad8! Black
seems solid enough; he has avoided any e4-e5 threats and has also freed the a8-square for the
bishop to deal with a5-a6 ideas.

8...e5
This is a typical response when the bishop arrives on h6.

8...b5 9.Bxg7 Kxg7 10.e5 Ne8 11.h4!‚ was troublesome for the black king in Koliaei – Ghari
Zadeh, Iran 1993.

8...Qc7 is so slow that we can even afford to lose a tempo outright with 9.h4!?. A plausible
continuation is 9...Bxh6 10.Qxh6 Ng4 as seen in Abela – Dos Ramos, Mallorca 2004, and here I
would suggest:

519
11.Qd2N 11...h5 12.Ng5 b5 13.f3 Ngf6 14.0-0-0 Nb6 15.g4ƒ White’s attack seems a lot faster
than Black’s.

9.0-0-0!? Qe7 10.Bxg7 Kxg7 11.g4 b5

11...h6 can be met by 12.g5N 12...hxg5 13.Nxg5² and White’s attack seems faster.

11...Re8 was played in Bakhmatov – Fridman, Recklinghausen 2002. At first this move appears
peculiar, but I think Fridman’s idea was to prepare ...b5 when, compared to the main line below,
White’s ideas involving dxe5 and Qd6 will not carry as much impact. Still, White could have
obtained a fine position with:

520
12.g5!N 12...Nh5 13.d5! Essentially forcing a blocked centre. 13...c5 14.h4 a6 15.Ne2!
Intending Ng3, when White will have a free initiative on the kingside.

We have been following Hauge – Alekseev, Eilat 2012. My improvement is:

12.g5!N 12...Nh5
12...Ne8 13.h4 Nb6 14.Ne2! (14.h5 is premature due to 14...Bg4) 14...Nc4 15.Qc3± and
White remains in control.

12...b4 13.gxf6†
13.dxe5?! dxe5 14.gxf6† Qxf6 15.Bg2 bxc3 16.Qxc3 Rb8 puts Black under much less
pressure than the version without the exchange on e5.
13...Qxf6 14.Bg2
14.Be2?! bxc3 15.Qxc3 Rb8 16.h4 Qf4† 17.Kb1 Nb6 turns out well for Black.

521
14...bxc3 15.Qxc3

15...Rb8
15...Qf4† 16.Kb1 Qxe4 is pointless in view of 17.Rhe1±.
15...Bb7 16.h4 h5 17.Ng5ƒ gives White good chances on the kingside.
15...c5 16.dxe5 Nxe5 17.h4 Nxf3 18.Qxf3² maintains some pressure.
16.b3 Qe7
16...exd4 17.Qxd4² leaves d6 weak.
17.h4 h5 18.Bh3²
Both sides have certain weaknesses, but White’s active pieces make him the slight favourite.

13.dxe5 Nxe5
13...dxe5? runs into 14.Qd6! Qxd6 15.Rxd6± when Black must lose material.

We have reached a critical position.

522
14.Be2!
14.Ne1 b4 15.Na4 f5! 16.f4 Nf7÷ seems okay for Black.

14.Bg2 can be met by 14...Nc4!? (14...b4 15.Ne2 Ba6 16.Ng3 Nxg3 17.fxg3 Rad8 18.h4ƒ)
15.Qd4† Kg8 16.Bf1 Ne5 and White does not seem to have anything better than repeating:
17.Bg2=

At first I thought the text move looked great for White; but unfortunately, Black seems to be able
to hold on with extremely precise play. I’m still not completely convinced that Black is okay
here, but it is hard to prove otherwise.

14...b4! 15.Na4 Ba6! 16.Qxd6

16.Bxa6 Nxf3 17.Qe3 Nxg5 18.Be2 Qxe4 19.Bxh5 Qxe3† 20.fxe3 gxh5 21.Rxd6 Ne4
22.Rxc6 Rac8 reaches a balanced endgame.

523
16...Rae8! 17.Bxa6
17.Qxe7 Rxe7 18.Nd4 yields no advantage after: 18...Bxe2 19.Nxe2 Nf3 20.Nd4 Nxg5 21.f3
Rc8 22.Nc5 Nf4 23.h4 Nge6=

17...Nxf3 18.Nc5
I also considered:
18.h4 Qxe4!
18...Qxd6 19.Rxd6 Rxe4 20.Rxc6 Rxh4 (20...Rfe8 21.b3 Re1† 22.Rxe1 Rxe1† 23.Kb2
Nxh4 24.Rc7±) 21.Rxh4 Nxh4 22.Rc7 Nf3 23.Rxa7 Nxg5 24.Rb7 Ne4 25.Rxb4 Nxf2
26.Nc3± leaves White the clear favourite in the upcoming pawn race.
19.Nc5 Qf4† 20.Kb1
20.Qxf4 Nxf4 21.Rd6 looks tempting but Black’s active knights maintain the balance:
21...Ne2† 22.Kb1 Ned4 23.Nd3 Rd8 24.Rxd8 Rxd8 25.Nxb4 Rd6 26.Nd3 Nf5=

524
20...Qxd6 21.Rxd6 Nxh4 22.Nd7
22.a3?! Nf5 23.Rxc6 bxa3 24.b4 Re7 is more dangerous for White than for Black.
22...Rh8 23.Nf6 Rd8
23...Nxf6? 24.gxf6† Kg8 25.c4 Nf5 26.Rxc6± would be perfect for White.
24.Rxc6 Nf4 25.Rc4 h6!
Black is doing just fine.

18...Qxd6 19.Rxd6 Nxg5 20.Rxc6 Nf4!

In the event of 20...Nxe4 21.Nxe4 Rxe4 22.Rc7 Nf6 23.Rxa7 Rf4 24.Bd3 Rxf2 25.Rb7² White
would be happy.

21.h4
21.Rc7?! Ngxh3 22.Rxa7 Nxf2 23.Rf1 N2h3 24.Rb7 g5 25.Rxb4 h5 is dangerous for White, as
Black’s knights are perfectly placed to support the advancing g-pawn.

21...Nxe4 22.Nxe4 Rxe4 23.Rc7

525
23...Rd8!? 24.Rxa7
24.b3 Ne2† 25.Kb2 Nc3= also seems okay for Black.

24...Ne2† 25.Bxe2 Rxe2=

The double rook endgame should be a draw, as Black has a lot of activity to compensate for the
missing pawn.

C2) 7.Bd3

This is the more popular and arguably more accurate continuation, keeping the option of
launching the h-pawn without loss of tempo.

7...Nbd7

526
 7...b5 transposes to variation D below.

7...Bg4 is an attempt to exploit the lack of h2-h3, but a good answer is 8.Ng5 (even 8.h4!? could
be considered) and after something like 8...b5N 9.f3 Bc8 10.h4ƒ White is doing great.

8.0-0-0!?
As explained earlier, the 0-0 set-up loses some of its potency when there is no b5-pawn to
attack. On the other hand, since Black has not been in a hurry to advance on the queenside, we
are more justified in castling on that side.

8...b5
I checked a couple of other ideas.

8...Ng4
This is recommended by Marin in Grandmaster Repertoire – The Pirc Defence. I have
already touched on the theme of this move being a serious option when White’s king lands
on c1, but on this occasion I found a nice way of dealing with Black’s idea.
9.Bg5 f6 10.Bf4N
10.Bh4?? Bh6 has been played a few times, so please take care to avoid this trick.
10...e5
My improvement is:

11.h3!
Marin continues 11.dxe5 dxe5 12.Bc4† Kh8 and although there is more that could be said
about what follows, it is irrelevant for our purposes, as the text move just seems more
promising.
11...Nxf2 12.Qxf2 exf4 13.g3! fxg3 14.Qxg3©
White has good attacking chances against Black’s kingside, which has been permanently

527
loosened by the earlier ...f6.

The other move worth mentioning is:

8...e5 9.dxe5 dxe5
As played in Doluhanova – Ajrapetjan, Alushta 2002. A good continuation is:

10.Bc4!?N
This position is surprisingly unpleasant for Black as he struggles to complete development.
Typical moves such as ...Qc7 or ...Qe7 are met by Qd6, which just leads to a worse
endgame for Black.
White could play even more ambitiously with 10.h4!?N, when Black can’t really afford to
just play ...h5, as the position after Bc4 will be much worse, with Ng5 coming up. Thus, the
best reaction is 10...b5! (10...h5? 11.Bc4±; 10...Ng4 11.Bg5ƒ) 11.h5 Nxh5 (11...Ng4
12.Bg5ƒ) when we reach a position that would be interesting to explore in more detail, but I
cannot see an easy way to break Black’s defences.
10...b5
10...Qe7 is met by: 11.Qd6 Qxd6 12.Rxd6²
11.Bb3 Ng4
11...a5 12.Nxe5± is just a free pawn.
11...Bb7 12.g4! with a strong attack.
11...Qc7 again allows 12.Qd6! Qxd6 13.Rxd6 Bb7 14.Rhd1 with an excellent queenless
position for White.
12.Bg5 Qb6 13.h3 Ngf6

528
14.Qd6
14.Be3 is another idea, hoping for 14...Qc7? 15.Qd6±, but 14...Qb7! is not so easy.
14...c5 15.Qxb6 Nxb6 16.Nd5 Nfxd5 17.Bxd5 Nxd5 18.Rxd5 Bb7 19.Rxc5 Bxe4 20.Rxb5²
Black will have to work to prove his compensation for the pawn.

9.Bh6!?
This seems like the most promising approach.
White does not get anything special after:
9.h4 b4!N
9...h5 comes unstuck after: 10.e5N 10...dxe5 11.dxe5 Ng4 12.Bxg6! fxg6 13.e6±
9...Qa5 10.h5 b4 11.Nb1!‚ also works well for White.
10.Nb1
10.Ne2 Qa5 11.Kb1 h5 is another interesting position to analyse further. White has many
possible ideas, but no obvious way to break through on the kingside.

529
10...c5! 11.h5 Ng4
11...Nxh5 12.g4 Nhf6 13.Bh6 looks risky for Black.
12.hxg6 hxg6
Despite the dangerous appearance of the open h-file, it is hard to prove anything on the
kingside, primarily because Black is able to keep his dark-squared bishop alive.
13.Rh4÷
With a complex game ahead. (I also checked 13.Bh6 but found 13...Nxh6! 14.Rxh6 e6!
15.Rdh1 Qf6÷ to be satisfactory for Black.)

9...e5
9...Bxh6? 10.Qxh6 is too dangerous for Black, for instance: 10...e5 (10...Ng4 11.Qh4+–)
11.Ng5!± The f6-knight is stuck and h4-h5 will be painful.

Another example continued 9...b4 10.Ne2 Qa5 11.Kb1 Nb6 12.Bxg7 Kxg7 13.h4 h5 and now in
Ruzicka – Karhanek, Frydek Mistek 2007, White should have played:

530
14.Nc1!N Safeguarding the king before deciding on a plan of action. Play may continue
14...Na4 (14...Ba6 15.Qg5! is awkward for Black) 15.Rde1ƒ when Black does not really have
anything on the queenside (since 15...Nc3† can be met by the calm 16.Ka1) and e4-e5 could
come at any moment, perhaps with Nb3 being thrown in at some point.

10.Bxg7 Kxg7 11.Ne2!

It is useful to improve the knight while incidentally holding up ...b4 and preventing ...Qa5 for
the time being.

11...Qb6
This is recommended by Kornev in A Practical Black Repertoire with Nf6, g6, d6. I checked
some other moves as well:

11...exd4 12.Nexd4 Bb7 occurred in A. Ivanov – Kirichenko, Kiev 2001, when the natural

531
13.h4Nƒ would have been promising for White.

I also considered 11...Qc7N 12.Ng3 Bb7 13.h4 h5 14.Qg5ƒ when Black has plenty to worry
about on the kingside; as well as 11...a5N 12.Ng3, with similar play.

A final alternative is:

11...Qe7!?N
Keeping the queen close to the kingside seems like a sensible idea to consider.
12.h4 h5 13.Ng5

13...Bb7
13...Nb6 14.f3 Bd7 15.g4 Rh8 16.gxh5 Nxh5 17.f4ƒ sees White’s attack continue.
14.f3 a6 15.g4 exd4
15...c5 comes a bit too late, as White can abandon his centre and focus on the attack with:
16.Ng3!ƒ
16.Nf4!
Putting pressure on Black’s kingside is more important than recapturing the pawn.
16...Rh8 17.gxh5 Nxh5 18.Nxh5† Rxh5

532
19.f4 c5 20.f5 Nf6 21.Rhg1ƒ
Although it would be possible to continue analysing, I am happy with White’s attacking
chances and the whole line just seems more dangerous for Black.

12.h4!?
Kornev mentions 12.dxe5 dxe5 13.Ng3 c5= as played in Straka – J. Bauer, Klatovy 2000,
when Black indeed has no problems.

12...h5
12...c5? allows the typical 13.h5! when the attack is too strong. For instance: 13...c4 14.hxg6
fxg6 15.Qh6† Kg8 16.Ng5 Rf7

533
17.Nc3!! Black’s defences are overwhelmed. My main line continues 17...Rg7 18.Nd5 Nxd5
19.Nxh7 N7f6 20.exd5 Nxh7 21.Bxg6 Qc7 22.Bxh7† Rxh7 23.Qg6† and it’s all over.

13.Ng3

13...c5!?N
This is one of the computer’s top suggestions and it mirrors the note to move 12, where Black
was doing fine. However, the inclusion of h2-h4 and ...h5 makes a difference.

13...exd4? 14.Nf5† is crushing.

The passive 13...Nh7?! was played in Weinzettl – Schlosser, Austria 2010, when 14.Be2!N
would have been strong. After 14...Ndf6 15.Ng5 Qxd4 16.Qxd4 exd4 17.Rxd4 Black has
avoided any major disaster on the kingside, but the resulting endgame will be no fun for him. A

534
possible continuation is:

17...Ne8 18.Nxh7 Kxh7 19.a4 a6 20.a5!?± White remains in control and the a6-pawn has been
fixed as a long-term target.

14.Qg5! Kh7
Black must tread carefully, as 14...Kg8 allows a surprising breakthrough: 15.Nf5! c4

16.g4!!+– Black has no good defence.

15.dxe5 dxe5 16.Nxe5 Nxe5 17.Qxe5 c4 18.Be2 Qxf2

Black wins his pawn back but he is not yet out of the woods.

19.Rd2 Nd7 20.Qg5 Qf6

535
21.Rd5 Qxg5† 22.hxg5²
Black remains under pressure in the endgame.

D) 6...b5

This is by far Black’s most popular continuation. The advantage of this move order is that
Black avoids committing his king too early, which obviously makes it less appealing for us to
commit to a kingside attack with Bh6, h2-h4 and so on. The disadvantage of Black’s move order
becomes apparent when we castle on the kingside though, as the b5-pawn presents us with a
clear target to attack with a2-a4. This is especially apparent in the lines where Black goes ...e5,
as White can simply exchange pawns on e5 and then continue playing on the queenside with a2-
a4 and so on. Black also has other plans, such as trying to get ...c5 in, which is possible because
of the pressure on e4. We will look to counter such ideas by regrouping our pieces with Bd3,
Ne2 and Ng3, perhaps with c2-c3 if the d4-pawn requires support.

536
7.Bd3
Before going into the exact details of the variations here, I would like to point out the main
ideas we will be looking to achieve. Although the attacking plan of Bh6 and h4-h5 still exists,
and could lead to interesting play, the key idea that interests me is to carry out the
aforementioned regrouping with Ne2-g3 along with h2-h3, getting ready to meet ...c5 with c2-
c3. This is clearly inspired from various Ruy Lopez variations, and I rather like this set-up for
White. It doesn’t put immediate pressure on Black on the kingside; but with such a well-fortified
pawn centre, the onus would be on Black to try and get some counterplay. The structure after
...c5 and c2-c3 will be discussed in more depth later; for now, it suffices to say that I like White’s
set-up aesthetically and intuitively, and believe it represents a promising version of a Spanish
structure for White. One explanation is that we did not spend too much time with the light-
squared bishop: instead of going via Bb5-a4-b3-c2-d3 as in the main Ruy Lopez lines, the bishop
has gone to d3 in just one tempo! Of course there are other differences and we should be cautious
about oversimplifying – but still, it makes some intuitive sense.
If Black plays ...e5 instead of ...c5, then c2-c3 isn’t as good because it would allow Black to
carry out mass simplifications with ...d5. Instead, White should react with dxe5, leading to a new
structure where the b5-pawn can be targeted most effectively with a2-a4 and perhaps c2-c4 as
well, when White’s bishops should exert powerful pressure on the queenside.

7...Nbd7
This seems like the most natural and consistent choice.

7...0-0 might tempt us to revert to attacking on the kingside, but it seems simplest just to play
8.0-0, when 8...Nbd7 9.Ne2 takes us to variation D24 on page 300.

7...Ng4 8.Bg5 just provokes weaknesses from Black, for instance: 8...h6 (8...f6 9.Bh4 e5 10.h3
Nh6 11.a4 b4 12.Ne2 a5 13.c3± was great for White in Kulaots – Kessler, Deizisau 2002) 9.Bh4

537
9...0-0 (9...Qc7 10.h3 Nf6 11.e5±) 10.0-0 g5 11.Bg3 e5 12.dxe5 Nxe5 13.Nxe5 dxe5 14.a4 b4
15.Nd1² Black’s position was full of holes in Jaracz – Krasenkow, Dzwirzyno 2004.

The only other significant option is:

7...Bg4 8.e5!
8.Nh4 is often a good move, but on this occasion it allows 8...e5! 9.h3 exd4 10.Bxd4 Be6÷
as seen in Zapata – Gallegos, Willemstad 2001.
8...b4
8...dxe5 9.Nxe5± leaves Black’s bishop looking silly.
8...Nfd7 is well met by: 9.exd6 Bxf3 (9...exd6N is not much of an improvement: 10.Bg5 f6
[10...Bf6 11.Bh6±] 11.Bh6 0-0 12.0-0-0±) 10.gxf3 exd6 11.h4 Black’s pieces are all kind of
misplaced, while the spoilt structure on the kingside only helps White to conduct an easy
attack there. 11...Nb6 12.0-0-0 N8d7 13.Bg5± Kharchenko – Borzov, Kiev 2006.
9.Ne4 Nxe4
9...Nd5 10.Bh6 is not much fun for Black either.
10.Bxe4

538
10...Bxf3N
10...d5 11.Bd3 Bxf3 12.gxf3 a5 13.h4 Nd7 14.h5± occurred in Leko – Beliavsky, Madrid
1998, which is not something Black should be looking to repeat.
The text move has been suggested as a possible attempt to improve Black’s play, but it
basically just amounts to trading a bad middlegame for a depressing endgame.
11.Bxf3 dxe5 12.0-0-0 exd4 13.Bxd4 Qxd4 14.Qxd4 Bxd4 15.Rxd4 a5 16.Rhd1 0-0 17.Rd8

17...e6
17...Na6 is met by 18.Rxa8 Rxa8 19.Rd7 and although good technique is still needed, White
is essentially winning.
The text move was suggested by Vigus in Chess Developments – The Pirc. The English FM
evaluates the endgame as holdable for Black, which doesn’t sound all that inspiring anyway
– but as it happens, I think White can break the defence by means of:
18.Rxf8†!

539
 18.R1d6 Na6 19.R8d7 Nc5 20.Rc7 e5 21.Rcxc6 Ne6 is mentioned by Vigus, but there is no
need for us to allow the knight to escape.
18...Kxf8 19.Rd8† Ke7 20.Rh8

20...Kd6
20...h5? 21.Bxc6 Ra6 22.Bb7 wins.
21.a4! bxa3 22.bxa3 h5

23.h4!+–
Black will succumb to zugzwang, as 23...Ke7? loses to 24.Bxc6 and 23...Kc7 allows 24.Rf8
when Black’s kingside collapses.

540
 We have reached a popular position where there are some important move-order subtleties. I
will present my analysis of the natural D1) 8.0-0 before moving on to the much rarer D2)
8.Ne2!, which I believe to be more accurate.

8.e5 is too soon, and 8...dxe5 9.dxe5 Ng4 10.e6 fxe6 11.Be4 b4 12.Na4 Nxe3 13.Qxe3 Qc7÷
was messy in Kovalev – Nepomniachtchi, Moscow 2016.

8.0-0-0 Qa5 9.Kb1 Ng4! looks promising for Black. As mentioned earlier, castling on the
queenside is pretty much the last thing we should be doing when Black plays an early ...b5 and
avoids committing his own king to the kingside.

8.h3 has actually been the most popular choice of all, and it could transpose to certain lines in
our repertoire. However, I do not consider ...Ng4 a threat at this point, and if White is going to
delay castling then I regard Ne2 as a more useful move; see variation D2 for further comments
about the merits of this approach.

D1) 8.0-0

This natural move was my original intention, until I discovered the following refinement for
Black:

8...Qc7!
This has hardly ever been played via the present move order, but the resulting position has
often occurred via transposition by way of the Modern Defence. To appreciate the usefulness of
this move, we should first familiarize ourselves with Black’s alternatives:

8...Bb7? prevents Ne2 – but it allows a lethal blow instead 9.e5! dxe5 10.dxe5 Ng4 11.e6 fxe6

541
12.Ng5± and it’s clear why moving the c8-bishop was bad.

Another obvious candidate is:

8...0-0 9.h3
9.Ne2! would be a nice improvement, and we will discuss this position in variation D24
below, where we arrive via the more accurate 8.Ne2! move order.
The text move was my original intention. After most moves, White can follow up with Ne2-
g3 to reach our desired set-up. However, Black can make life trickier with:
9...Bb7! 10.e5!?
10.Ne2 would be met by 10...c5! when White doesn’t have time for both c2-c3 and Ng3.
Therefore it is necessary to go into messier territory.
10...Ne8
White has to reckon on the annoying threat of ...c5, with ...Bxf3 in the air. This is a far cry
from the smooth positional edge I had in mind with the Ne2-g3/c2-c3 plan. Nevertheless, I
did find an interesting novelty.

11.e6!?N
After all other moves, Black seems to be doing quite well.
11...fxe6 12.Ng5!?
12.Qe2 appeared promising, until I ran into some concrete challenges after: 12...Qc8!
13.Bg5 (or 13.Ne4 c5 14.c3 e5! 15.dxe5 Rxf3! 16.Qxf3 d5) 13...c5 14.Qxe6† Rf7÷
12...Nc7 13.f4
13.Qe2 is another idea but 13...e5! 14.Nxh7 exd4 seems to work out well for Black.
The text move reaches a position which would be fascinating to explore further. I definitely
prefer White, but it is hard to draw a definite conclusion. A few brief lines are:

542
13...Nf6
13...h6?! 14.Nf3 only weakens Black’s kingside.
13...c5!? is interesting though.
14.Rae1
14.Qe2 is another idea.
14...b4
14...Qd7 15.Qe2 a5 16.Nce4 Nfd5 17.Nxh7 looks dangerous for Black.
14...Qe8 15.Qe2 h6 16.Nf3 maintains some initiative for White.
15.Ne2 a5 16.c4 bxc3 17.bxc3 Ba6 18.c4 d5÷
The position remains complex and double-edged.

Let’s return to 8...Qc7!. The main advantage of this move is that it is now surprisingly hard to
reach our desired set-up with Ne2-g3.

543
9.Bh6!?
This is White’s most popular continuation and it is certainly playable. However, one of my
aims in this repertoire has been to avoid the Bh6 trade when we have already castled on the
kingside, because now our attacking plans on the queenside (as could occur after ...e5, when we
react with dxe5, a2-a4 and so on) lose a lot of their strength, simply because the bishop on e3
exerts more influence on the queenside than its counterpart on g7.
Of course, White can instead opt for the standard plan of Ne2-g3, intending to meet ...e5 with
c2-c3, reaching a Spanish-type position. However, it’s not obvious to me if White will get a
particularly good version of the Spanish. For many players this would be a perfectly acceptable
outcome from the opening, but it depends on whether you feel comfortable there or not.
Assessing the position objectively, it seems to me that Black is doing okay in such positions, as I
show in the lines below. So I eventually decided that the c2-c3 plan works best against ...c5 – as
once Black has committed to this move, the Spanish approach works out slightly better for
White.

9.h3 Bb7! leads to a familiar problem: whenever White plays Ne2, Black will respond with ...c5,
when White does not have time for both c2-c3 and Ng3.

Of course, our other main option is:

9.Ne2 Bb7
9...0-0 10.a4 might be annoying for Black.
10.c3!?N
10.Ng3 Ng4! leads to unclear complications, as the bishop doesn’t have enough squares to
survive.
10.h3 runs into the typical 10...c5! when White cannot stabilize his centre.
The text move is a natural attempt to consolidate, but Black has a few interesting ways to
create a mess – mostly exploiting the fact that he has a spare tempo due to not having
castled yet.

544
10...Ng4!
Another interesting option is: 10...c5!? 11.Ng3 c4 12.Bc2 Ng4 13.Bg5 Incidentally, a
similar position will be reached on page 301 (see the note on 9...Bb7N), but here Black has
played ...Qc7 instead of ...0-0. In general, ...Ng4 ideas tend to work better when Black has
not castled, as Black can seriously consider advancing his kingside pawns to eliminate the
dark-squared bishop, without worrying so much about exposing his king. Play may
continue: 13...h6 14.Bh4 (14.h3!? has been played in similar situations and is worthy of
further analysis, but I am generally not a fan of this idea – although perhaps it’s just a
matter of taste) 14...e6!? Black will follow up with ...Ngf6, and White will be forced to give
up the dark-squared bishop.
11.Bg5
11.Bf4 is well met by 11...e5 when having the queen on c7 proves especially useful –
otherwise we could simply play Bg5 and provoke further weaknesses on the kingside.
Instead after 12.Bg3 0-0 White’s bishop is misplaced. Remember, against ...b5 and ...e5, our
plan was always to play dxe5 and then play on the queenside. But with the bishop on g3,
this becomes much less effective, as White is no longer preventing ideas such as ...Nc5.
11...h6

545
12.h3!?
12.Bh4 Ngf6! leaves the bishop awkwardly placed.
12...hxg5 13.hxg4 Nb6 14.Qxg5 0-0-0 15.Qd2 e5
This could be analysed in more detail, but I think Black’s bishop pair and open h-file add up to
decent compensation for the pawn.

9...0-0 10.Ne2 e5!

As mentioned before, this is a good choice because the dxe5 plan is not so great for White
without a bishop on e3.

11.c3
Just to reiterate the point, 11.Bxg7 Kxg7 12.dxe5 dxe5= leaves White unable to exert
meaningful pressure on the queenside, so Black is fine.

11...Re8
We have reached a relatively normal-looking Spanish structure. I wouldn’t actually mind
playing this as White, as I feel quite at home in the Ruy Lopez – but at the same time, I wouldn’t
mind playing it as Black either. It is up to the reader to decide what he thinks of these positions,
but overall I consider them less promising than my main recommendation, so I will just add a
few brief lines.

12.Bxg7 Kxg7 13.Ng3

13.a4 seems okay for Black after 13...bxa4! 14.Rxa4 exd4, targeting the e4-pawn.

546
13...Nb6!?N
This seems better than 13...Nf8, when 14.a4! forces Black to compromise on the queenside
structure.

14.b3
14.Nh4 d5!= seems fine for Black.

14.h3 can be met by 14...a5 when it’s not easy to see what White is doing next. He isn’t really in
a position to attack on the queenside, and 15.Nh2 (or some other move with the same knight)
invites 15...d5!.

14...Qe7!
Both 14...a5 and 14...Bg4 allow 15.Nh4!, intending Rae1 and f2-f4, with excellent attacking
chances.

The text move is directed against the Nh4 plan, as now Black can simply meet it with ...Ng8 (or
...Nfd7), forcing the knight to retreat.

547
15.a4
15.h3 can be met by 15...a5, and if 16.a4 bxa4 17.bxa4 Ba6= Black is fine.

15...bxa4 16.bxa4 Bg4

White has no more than a tiny advantage, if that.

D2) 8.Ne2!

Even though I am confident in my decision to recommend this move and award it an

exclamation mark, there is still a part of me that worries that things are becoming a little too
subtle. In order to convey the advantages of this move order as clearly as possible, I will recap
the general thought process that led me here, while summarizing the drawbacks of the alternative
move orders. It will of course help if you actually look through the previous variations – but I am

548
sure that some readers prefer just to skip ahead to the main recommended move. In any case, I
hope the following explanation will give you a clear understanding of my reasoning.

Originally, I started analysing 8.0-0 0-0 9.h3, intending Ne2-g3 plans. I was quite happy with
those lines, until I turned to 9...Bb7!, which essentially prevents the ideal Ne2-g3 set-up and
forces White to enter messy complications.
This led to the natural improvement 8.0-0 0-0 9.Ne2!, avoiding the ...Bb7 problem. And since
...Ng4 isn’t a big deal, this enables me to get my ideal position with Ng3, meeting ...c5 with c2-
c3. This still needs to be analysed – but the main battle of move orders seemed to have been
won.
However, I then realized that Black can be subtle as well. Thus, after 8.0-0, instead of castling,
Black can play 8...Qc7!. The idea is that if White continues to try and sneak in Ne2-g3 without
playing h2-h3, then Black can do better in the ...Ng4 lines by avoiding ...0-0. And if White
spends a tempo on h2-h3, then ...Bb7 is an annoying reply, since Ne2 will run into ...c5 before
White can consolidate his centre. Thus, 8...Qc7 forces White to switch plans with early Bh6. In
that case, we get to carry out the desired Ne2-g3 manoeuvre, but the inclusion of Bh6 leads to
some subtle differences after ...e5. If White’s bishop was on e3, then dxe5 would be an excellent
reply, followed by attacking on the queenside – but with the bishop already on h6, this plan loses
its impact. So instead, White should meet ...e5 with c2-c3, which leads to an interesting game –
but if we are to convert to a Spanish-like structure, I would rather do so in a way that gives
White some kind of obvious advantage as compared to a typical Spanish position.

All this brings us to the proposed move order, where we play Ne2-g3 earlier than almost anyone
has done in practice so far. The advantages to this approach are:
1) We remain on track for the most important positional plan of this chapter, namely that of
meeting ...e5 with dxe5 followed by a2-a4, with an easy attack on the queenside.
2) We are ready to meet ...c5 with c2-c3, and ...Bb7 with Ng3, keeping the central pawns safe.
If Black follows up with ...c5 we could reach a typical Spanish structure, but with Black having a
pawn on c5 rather than c6, which helps White in a few ways: the b5-pawn is weaker, and d4-d5
could be a tempting option to block the b7-bishop out of play.
3) Unlike variation D1 above, White keeps the option of long castling. Attacking on the
kingside is by no means my main goal from here – although as we will see, there are certain
scenarios where it can work well.

Of course, there could still be some players who prefer not to concern themselves with such
extreme move-order subtleties – in which case I refer you back to the previous section on 8.0-0.

I analysed four main options for Black: D21) 8...c5!?N, D22) 8...Qc7, D23) 8...Bb7 and D24)
8...0-0.

8...e5 was played in Fressinet – Andriasian, Riyadh (blitz) 2017, and now 9.dxe5!N followed by

549
a2-a4 would be consistent with our strategy. It is likely that both sides will castle in the near
future, and we will deal with such positions under variation D24 below.

8...Ng4N should be met by: 9.Bf4!N (9.Bg5 is also possible, but I am not sure about the position
after 9...h6 10.Bh4 Puuska – Lehtinen, Tampere 2004, 10...0-0N where I feel that the bishop on
h4 is far from ideal. It’s true that Black will have to waste time retreating with ...Ngf6, and
perhaps we can exploit it somehow with a quick e4-e5; but I would prefer to avoid having my
bishop stuck on h4 in the first place.)

9...e5 10.Bg5 f6 11.Bh4² This time we also have our bishop stuck on h4, but in a more
favourable scenario: the ...e5 and ...f6 moves have left a number of holes in Black’s kingside,
and he also has to be on the lookout for a2-a4 ideas, creating additional weaknesses.

D21) 8...c5!?N

This could be an interesting attempt to exploit White’s unusual move order.

9.c3 Bb7 10.Ng3 c4

10...Ng4 allows 11.Bxb5, and after 11...Nxe3 12.fxe3 I do not really believe in Black’s
compensation.

11.Bc2
In a way, Black has saved time by avoiding ...a6, but releasing the tension with ...c5-c4 is a
concession, so Black needs to do something to justify his play. The critical continuation looks to
be:

550
11...Ng4!? 12.Bf4! e5 13.Bg5 f6 14.Bh4 0-0
14...Nb6 15.0-0 keeps an edge for White. (15.a4!? bxa4 [15...a6 16.a5±] 16.Rxa4 Nxa4
17.Bxa4† Kf8 18.0-0© is a fun try too, but not required.)

15.0-0
White should keep things simple, rather than venture 15.a4, when 15...bxa4! 16.Rxa4 d5 is
messy.

The structure has changed dramatically and the resulting position is not trivial to assess.
However, I think that once White regroups his pieces, his superiority should become evident.
The plan is simple: the f1-rook will go to e1 (or perhaps d1) and the g3-knight will go via f1 to
e3. Please also note that, despite the uncomfortable appearance of the bishop on h4, White is in
no danger of losing a piece.

551
15...Nb6
15...g5? 16.Nf5 gxh4 17.h3+– would be a positional disaster for Black.

15...Nh6 16.Rfe1 Nf7 17.Nf1² is similar to the main line.

16.Rfe1 Nh6 17.Nf1 Nf7 18.Ne3²

The dark-squared bishop is still not ideally placed, but everything else about White’s position
is rather nice. There is a lot of scope to continue manoeuvring the minor pieces. Meanwhile,
Black’s play on the queenside isn’t so clear; and the fact that he has released the central tension
by playing ...c5-c4 so early definitely favours White.

D22) 8...Qc7

This move order gives White a pleasant choice.

9.Bh6!?
This aggressive move is a ‘bonus option.’ I haven’t analysed it exhaustively because there is no
special reason to deviate from our normal approach – but as far as I can see, this is quite
promising for White as well, so it is worth considering if you are in the mood for attacking.

9.h3 is the simpler choice, leading to the usual scenarios where the knight goes to g3 and we
meet ...c5 with c2-c3, or ...e5 with dxe5. The early ...Qc7 does not change much, since Black
would surely castle anyway, with a likely transposition to variation D24.

9...0-0
The big difference from the 8.0-0 lines is of course that we have the h2-h4 idea at various
points.

552
After 9...Bxh6 10.Qxh6 the queen on h6 will be extremely annoying for Black; there is not even
any special need for White to follow up with h4-h5. 10...c5 (10...Bb7 11.c3 c5 12.Nd2²) 11.c3
cxd4 12.cxd4 Qa5† 13.Nd2²

10.Bxg7!?
This is the simple choice, which offers White a pleasant edge without getting embroiled in
complications.

10.h4!?
This is the maximalist approach, leading to sharp play after:
10...Bxh6 11.Qxh6 Ng4!
11...c5? proves too slow after 12.h5 c4 13.Ng5! (13.hxg6 fxg6 14.Nf4 cxd3 15.Ne6 Qa5†
16.Kf1 Rf7 17.Nfg5„) 13...cxd3 14.c3!+– when White’s attack decides.
12.Qg5
12.Qd2 h5 13.Ng3 e5 (13...Ndf6 14.Ng5; 13...Ngf6 14.0-0-0; 13...Bb7 14.Nxh5) 14.c3
might be nice for White too – but I think White gets a better version of this structure in the
main line with 10.Bxg7.
12...Ndf6
We are nearing the critical phase of the opening.

553
13.h5
13.Ng3!? also deserves attention.
13...e5!
This could be a fascinating position to analyse in more detail. From a repertoire perspective,
there are simpler ways for White to obtain the kind of advantage we can be happy with; at the
same time, I would not be especially surprised if deeper analysis revealed a large or even
decisive advantage for White with this sharper approach.

10...Kxg7

11.h4! h5
11...c5? 12.h5 c4 13.hxg6 cxd3 14.Qh6† Kg8 15.gxh7† Kh8 16.cxd3 gives White three pawns
for a piece plus a fierce initiative.

554
12.0-0
12.Ng3 could also be considered, although after 12...e5 I am not sure if White has anything
better than 13.0-0.

12...e5
The position strongly resembles the 8.0-0 Qc7! line – something that I went to great pains to
explain why I was trying to avoid. However, the crucial difference is the placement of the pawns
on h4 and h5. I would argue that Black’s ...h5 is a much bigger weakness, as is typical in the
Spanish. For one thing, it gives the g5-square for the white knight. But more crucially, in lines
with f2-f4 for instance, Black’s king will be much more exposed, and tricks involving Nxh5(†)
could start springing up all over the place. Another advantage is that ...exd4 would leave Black’s
kingside structure more vulnerable than it would have been with the pawn on h7, so we can
manage without spending a tempo on c2-c3. I have included a few brief lines to show how these
ideas could play out.

13.a4!
13.c3² is fine but the text move is more ambitious and energetic.

13...exd4?
13...bxa4 14.Rxa4² would be the lesser evil for Black.

14.Nexd4 bxa4 15.Nf5†!

A straightforward exploitation of the weaknesses created by ...h5.

15...gxf5 16.Qg5† Kh7 17.e5! Nxe5

17...dxe5 18.Bxf5† Kh8 19.Qh6† Kg8 20.Ng5+– leads to the same result.

555
18.Nxe5 dxe5 19.Bxf5† Bxf5 20.Qxf5† Kg7 21.Qg5† Kh7 22.Qxf6+–
Black has avoided being mated but his position is in ruins.

D23) 8...Bb7

This slightly tricky move makes it harder to convert to one of our main lines. However, we can
get a good game by exploiting the other advantage of White’s move order, namely the fact that
we have not yet castled.

9.Bh6!?
9.h3 c5! reveals Black’s main threat, and we don’t have time to reinforce both of the central
pawns.

9.Ng3 guards the e4-pawn but invites 9...Ng4! when the bishop will not have the g3-square on
which to hide.

Several years ago, I actually reached this position myself, after transposing from a Modern
Defence (with 8...Ngf6 being Black’s last move). I opted for 9.c3, after which 9...e5 10.Bh6 0-0
11.Bxg7 Kxg7 led to something resembling the earlier variation D1 in Negi – Gashimov,
Germany 2012. White’s position is okay – perhaps a little better – but it’s nothing to get excited
about.

9...Bxh6
It is important to note that 9...0-0? is simply not possible because, after 10.Bxg7 Kxg7, as
played in Ogloblin – M. Perez, Buenos Aires 1999, White can play:

556
11.e5!N And after the knight retreats, h2-h4 will give White a crushing attack.

10.Qxh6
The queen is a major annoyance for Black.

10...e5
This seems like the only sensible move, as Black tries to expel the queen with ...Ng4 next.

The immediate 10...Ng4? allows 11.Qg7 of course.

10...Qa5† 11.c3 c5 12.d5 Rg8 (12...c4 13.Bc2 Nc5 14.Nd2±) 13.0-0± left Black’s king
awkwardly placed in the centre in Bouquet – Komodo, engine game 2013.

10...Qb6 11.c3 c5 12.Nd2 a5 13.0-0 is another example where the queen on h6 is a pain for
Black to deal with. In a practical game this would be horrible; and even against machine-accurate
defence, White won convincingly in Komodo – Bouquet, engine game 2013.

11.c3
An alternative plan is to simplify the centre and then focus on trying to exploit Black’s
weakened queenside by means of: 11.a4 a6 12.dxe5!? (12.c3² is similar to the main line below
and could easily transpose) 12...dxe5 13.0-0 Ng4 14.Qd2 0-0 15.b4 Of course, these kinds of
ideas would work much better if we still had the dark-squared bishop, as a bishop on e3 would
be a lot more influential than the one on g7 – but even so, the position remains slightly more
pleasant for White.

557
11...Ng4
11...exd4?! 12.cxd4 Qa5† achieves nothing after 13.Nd2±.

12.Qd2 0-0 13.0-0 Ngf6 14.Ng3 Re8

The position resembles the one after 13.Ng3 in the earlier variation D1. White’s set-up is
actually identical, but there are a couple of differences on Black’s side. Firstly, he has played
...Bb7 instead of ...Qc7, which isn’t a big deal, as both are normal developing moves. More
significantly, Black’s king is on g8 instead of g7, and it is White to move instead of Black – so
Black has essentially spent a tempo on ...Kg8, which is hardly the most useful move and may
even harm his chances slightly, albeit not in any grave way.

15.a4 a6 16.Rfe1²
In general I rather like playing closed Spanish positions, and this seems like a pleasant version
for White.

D24) 8...0-0

558
 Even though the extreme rarity of my recommended move order means that there is no
established theory on it, I think the text move can be considered the clear main line; it is the most
obvious developing move, and also leads to a high probability of transposing to existing games
in which White played Ne2 a bit later.

9.0-0
This seems accurate, as ...Ng4 is not a real threat.

The slow 9.h3 invites 9...c5 10.c3 Bb7 11.Ng3 d5! with interesting counterplay for Black.

Of course it is also possible to try and exploit Black’s castling in some smart way, keeping the
option of a kingside attack – but honestly, I was just interested in reaching the present position
while avoiding the 8.0-0 Qc7! move order from variation D1. I only found one example of this
exact position on the database, but we are likely to transpose to more games after Ng3 and h2-h3
are played.

9...Qc7N
This seems like the most natural and flexible choice.
9...e5?! 10.dxe5 dxe5 11.a4± enables us to carry out our desired plan with ease.

The one game I found on the database continued: 9...a6 10.h3 e5 11.dxe5 dxe5 12.a4 Bb7
13.Ng3 Qe7

559
14.c4! b4 15.c5 a5 16.Qc2 White had an excellent game in Asis Gargatagli – Carrasco Martinez,
Linares 2016; we will see more of this type of position in variation D241 below.

9...Bb7N threatens ...c5, but we are in time to reinforce the centre. 10.Ng3 c5 11.c3 c4!? This is
the only serious attempt to exploit White’s delay in playing h2-h3; but releasing the central
tension is a concession of sorts. (The immediate 11...Ng4 allows 12.Bxb5 when Black will not
have enough compensation for the pawn; and if 11...a6 12.h3 the position resembles variation
D242 below, and may transpose after a subsequent ...Qc7.) 12.Bc2 Ng4

13.Bf4! e5 14.Bg5 f6 15.Bh4 We have transposed to variation D21 on page 296.

10.h3
With this, we suddenly transpose to quite a lot of games. White’s entire set-up is designed to
deal with the two main types of pawn structures arising after D241) 10...e5 and D242) 10...c5.

560
Moves like 10...Bb7 and 10...a6 do not really change much, as 11.Ng3² continues White’s plan.

D241) 10...e5

This is a fairly common idea for Black, but here White is perfectly placed to simplify in the
centre and then attack on the queenside with the help of the strong bishops.

11.dxe5!
11.c3 d5! is completely fine for Black.

11...Nxe5
Exchanging the knights takes some of the strain off Black’s position, although the problems
with his queenside structure remain.

The more popular 11...dxe5?! makes White’s task easier: 12.a4 Bb7 13.Ng3 a6

561
14.c4!± As seen in Adams – Lanchava, London/Crowthorne 2006, and several other games. This
is pretty much White’s dream opening outcome for the set-up I have been advocating.

12.Nxe5 dxe5

13.a4 a6!?N
13...Be6 14.axb5 cxb5 15.Qa5 gave Black immediate problems with the queenside. 15...Rfc8
16.Qxc7 Rxc7 17.Nc3 Bc4 18.Ra5² Saric – Van Wely, Wijk aan Zee 2015.

14.Qc3!?²
14.f3 can be met by 14...Be6 when c2-c4 is not so easy to achieve, so it seems best to abandon
that plan.

The text move prepares Nc1-b3-c5, while the e4-pawn may be bolstered by f2-f3 whenever it

562
becomes necessary. White has a clear edge on the queenside, and it is not clear if Black will be
able to find much counterplay.

D242) 10...c5 11.c3

11...Bb7
11...e5 and 11...a6 have been played, but in both cases 12.Ng3 leads to similar play as the lines
below, and transpositions are quite likely.

12.Ng3 a6
12...e5 is an attempt to save a tempo, but after 13.dxe5 Nxe5 14.Nxe5 dxe5 15.Bxb5 Nxe4
16.Nxe4 Bxe4 17.Qe2 Black is struggling with the split queenside pawns, while White’s bishops
will be placed perfectly on e3 and c4. A practical example continued:

563
17...Rfb8 (17...Rab8N 18.Bc4 Bc6 19.b3 Rbd8 20.Rfd1² also sees White keep control) 18.Rad1
(18.Bc4 was also good for White in Gopal – Nolte, Kolkata 2012) 18...Qb7 19.Ba6 Qc6
(19...Qxb2?? 20.Rd8†+–) 20.f3 Bd5 21.b3²/± Akopian – Mamedov, Ningbo 2011.

13.Bh6 e5
13...e6 has been played a few times but it’s not clear to me what Black wants to do next.
Meanwhile, White has many ways to improve his position, for instance: 14.Rfe1 (there is also
14.a4!? with the idea of bothering Black on the queenside) 14...Rfd8 15.Rad1 Rac8 16.Bb1²
White has centralized all his pieces before committing to any specific plan, while ensuring that
Black will not get any serious counterplay.

After the text move, the position starts to take on the form of a Ruy Lopez.

14.a4!?N
This is the most ambitious approach, trying to provoke ...c4. If Black plays this move without
White having committed to d4-d5, it should count as a significant achievement for White, who
gets to control the centre completely.

Another valid approach is:

14.d5
I consider this quite promising as well – although this may be partially because I just enjoy
playing these closed Spanish positions in general. An important point is that the b7-bishop
is dead for now, and Black will most likely have to reroute it, perhaps via c8-d7. It is not
easy for White to break through on either side, but he does at least have the makings of a
decent attack on the kingside. Here are a few illustrative lines:
14...c4 15.Bc2 Nc5
a) 15...Bxh6? is suicidal in view of 16.Qxh6 Kh8 17.Ng5 followed by f2-f4, with a

564
 crushing attack.
b) 15...Kh8 was played in Olszewski – Jaracz, Karpacz 2008, when White went for an
overly sophisticated approach in retreating his bishop to e3. Although this seems good
enough to retain some edge, I would recommend a more direct attacking approach with:
16.Bxg7† Kxg7 17.Nh4! Kh8 (17...Ng8 18.f4 f6 19.f5ƒ) 18.f4 exf4 19.Rxf4 Ng8 20.Raf1²
c15...Qb6N can be met in much the same way: 16.Bxg7 Kxg7 17.Nh4 (17.Qg5 Kh8)
17...Kh8 18.Kh1 a5 19.f4²

16.Bxg7 Kxg7 17.Nf5†

17.Nh2!?N is a good alternative, preparing f2-f4.
17...Kg8
17...Kh8 18.Qh6 Rg8 19.Ng5 Raf8 20.f4ƒ
Now in So – Ghasi, Internet (rapid) 2019, White went for the tempting knight jumps to h6
and g5, but did not achieve anything special. Instead I would recommend:
18.Qh6N 18...Ne8 19.Ne3 f6 20.Nd2ƒ
With ideas of f2-f4; and if Black opts for a queen trade on g7, White can change tack with b2-
b3/b4, with good chances to press Black’s queenside in the endgame.

14...Qb6
Defending the b5-pawn seems best. Here are some other lines I analysed:

14...c4?! 15.Bc2± gives White exactly what he wants.

14...Rfe8?! simply allows us to eat a pawn: 15.Bxg7 Kxg7 16.axb5 axb5 17.dxe5 dxe5
18.Bxb5±

14...Bxh6?! 15.Qxh6 endangers Black’s king for no good reason: 15...cxd4 16.cxd4 exd4
(16...bxa4 17.Ng5 Qb6 18.dxe5 dxe5 19.Rxa4±) 17.axb5 axb5 18.Nxd4±

565
Finally, it is important to note that Black cannot get away with:
14...d5?
Such a move, if tactically viable, could provoke mass simplifications resulting in a drawish
symmetric structure. However, there are times when we are able to use our kingside
resources to deal with such breaks, as is the case here:
15.dxe5! Nxe4
This is the move Black wants to play, but it fails for tactical reasons.
15...Nxe5 is objectively the lesser evil, but the forcing sequence of 16.Nxe5 Qxe5 17.Bxg7
Kxg7 18.f4 Qe7 19.e5 Ne4 20.Bxe4 dxe4 21.Qe2± reaches a position where White clearly
has the superior minor piece, as well as attacking chances with a timely f4-f5.
16.Bxe4 dxe4

17.Nf5!!
A typical idea in various Spanish lines! Black has no good defence, for instance:
17...Bxh6
17...gxf5 18.Qg5+–
17...exf3 18.Bxg7+–
18.Qxh6 gxf5 19.Ng5+–

15.Bxg7 Kxg7

566
16.b4!
A typical idea in such positions.

16...cxb4
16...exd4 17.cxd4 c4 looks tempting, but after 18.Bc2 it will be a long time before Black’s
protected passer becomes useful, and in the meantime White enjoys a mobile central majority
and attacking chances on the kingside. A plausible continuation is: 18...d5 19.e5 Ne4 20.Qe3
Nxg3 (20...f5 21.Ne2±) 21.fxg3ƒ

17.cxb4 bxa4
17...exd4 is met by 18.axb5 axb5 19.Ra5!² when White’s pressure on the queenside is still
apparent, and Black’s extra pawn is clearly only temporary.

18.d5!
The b7-bishop is dead again, and the open queenside files give White an easy initiative.

567
18...a5 19.Rxa4 axb4 20.Rxb4 Qa5 21.Rxb7 Qxd2 22.Nxd2 Nc5 23.Rb6 Nxd3 24.Nc4²
Despite the simplifications, the weakness of the d6-pawn means that Black is still far from
equalizing.

Conclusion

This chapter has introduced our repertoire against the Pirc Defence, 4.Be3 Bg7 5.Qd2 c6 6.Nf3
being my recommended move order (with 6.Bh6 being noted in passing as a promising
alternative). We started by addressing a couple of interesting sidelines: after 6...Bg4 then
7.Nh4!? followed by f2-f3 gives White nice prospects; and 6...Qa5 can be met by some
combination of Bd3, h2-h3 and 0-0, when Black’s queen is not so well placed.

We then moved on to the main lines, where the crux of the matter is that Black will almost
always play ...b5 and/or ...e5 at some point. I will not repeat all the earlier waffle about the
various move-order subtleties. It suffices to say that after 6...0-0, both 7.h3!? and 7.Bd3 lead to
interesting positions where White has yet to decide where to castle; I slightly favour the latter
move, but judge for yourself.

Finally we turned to the most popular 6...b5 7.Bd3 Nbd7, when 8.Ne2! is an important
refinement in White’s move order, followed by some combination of Ng3, h2-h3 and 0-0, with
the exact move order depending on what Black does. If ...e5 comes, then dxe5 followed by a2-a4
gives us exactly what we want; and if Black favours ...c5, then c2-c3 will keep the centre under
control.

568
569
A) 5...Nd7 6.Nf3 308
A1) 6...e5 308
A2) 6...Qc7!? 309
B) 5...b5 6.Bd3 Nd7 7.Nf3 312
B1) 7...Nb6 313
B2) 7...Bb7 314
B3) 7...Qc7 317

1.e4 g6 2.d4 Bg7

Delaying this move with 2...d6 3.Nc3 c6 4.Be3 b5 adds nothing to Black’s set-up. At some
point, he will have to transpose to either this chapter with ...Bg7, or to the previous chapter with

570
...Nf6.

3.Nc3 d6 4.Be3 c6
Unlike the comparable Pirc set-up covered in the previous chapter, here Black avoids going
...Nf6. The idea is obvious: Black prevents a possible bishop trade with Bh6, which is a useful
idea in many lines when ...Nf6 is played. At first, it seems like there is not much of a downside
to leaving the knight on g8 for a while: in the equivalent Pirc lines, the extra pressure on the e4-
pawn exerted by the knight on f6 was not really critical in most lines.
For the most part, White should follow the same general plan of development as in Chapter 12,
which means a transposition could occur at any moment if Black develops the knight to f6. If
you have not already looked through that chapter, you may find it useful to familiarize yourself
with the plans I have described there, as in this chapter I will focus mostly on the differences,
rather than reiterating the various plans. In brief, the set-up I recommend revolves around setting
up the minor pieces with Nf3, Bd3 and Ne2-g3, being ready to meet ...c5 with c2-c3. A lot of
players seem to want to charge ahead on the kingside with f2-f3, g2-g4, h2-h4 and so on; but as I
explained previously, I am not a fan of committing to that too quickly. Nevertheless, even having
played Nf3 (thus forgoing the option of f2-f3), it is still possible to revert to kingside aggression
by launching the h-pawn when the situation warrants it, so we should always keep that in mind.
Another typical idea to remember is that if Black plays both ...b5 and ...e5, White can simply
exchange with dxe5, and after ...dxe5 we can use the b5-pawn as a target with a2-a4 and perhaps
c2-c4 as well. Black will typically have to respond with ...bxa4 or ...bxc4, either of which will
leave him with an ugly pawn structure.

5.Qd2
As usual in the Modern, transpositional possibilities abound. We will consider two main
approaches: the first involving A) 5...Nd7 where Black delays ...b5 for a while, followed by B)
5...b5.

5...Qc7
This is another of the weird move orders that Black may adopt. The only idea I can see
behind it is to meet Nf3 with ...Bg4. Based on our repertoire, the simplest option would be:
6.Bd3
After 6.Nf3 Bg4 Black would be doing decently if he swapped off his light-squared bishop.
We can of course prevent this with something like 7.Nh4 followed by h2-h3, which is a
fairly typical idea, but I’m not sure if the knight on h4 would be all that useful.
6...b5
Now White has a pleasant choice.

571
7.a4!?
7.Nce2 is perfectly fine, when we will almost certainly convert to one of the lines from
Chapter 12 after a subsequent ...Nf6.
7...b4 8.Nce2
White will follow up with Ng3, Nf3 and so on, with the same general set-up as in the main
lines, and all the subsequent plans and ideas should be the same as well. In general, Black
usually doesn’t want to commit to ...b4 so easily, so this might turn out to be a slightly improved
version.

A) 5...Nd7

6.Nf3
With ...Bg4 no longer an option, there is no reason to delay this move.

572
We will analyse A1) 6...e5 and A2) 6...Qc7!?.

6...b5 converts to variation B.

A1) 6...e5

This move has been played a fair bit, including by a few GMs, but it doesn’t make much sense to
me. White has several possible ways to get a pleasant position, and I’ll just highlight the one that
seems most natural to me.

7.0-0-0 Qe7 8.h4 Ngf6

8...h5 is a big commitment and White can just enjoy having the g5-square for the knight. 9.Bc4
Ngf6 (9...f6 looks horribly ugly) 10.Ng5 0-0 11.dxe5 dxe5 (11...Nxe5 12.Bb3±) 12.Qd6 Qxd6
13.Rxd6±

9.Bh6 Bxh6 10.Qxh6 b5N

This seems like a better try for counterplay than 10...Qf8, when Black had nothing better than
meekly returning with the queen to e7 on the very next move in Jaracz – Jansen, Lubawka 2008.

11.h5 b4
11...Ng4 12.Qd2± leaves Black’s knight without an obvious purpose.

11...Bb7 12.hxg6 fxg6 13.Qd2± and White will benefit from the open h-file.

Finally, 11...Nxh5?! runs into some tactical trickery:

573
12.Bxb5! Black should admit that he is much worse and struggle on with 12...Bb7. If he grabs
the material with 12...cxb5? then 13.Nd5 Qd8 14.Rxh5! gxh5 15.Qxd6+– is crushing.

12.Na4 Nxh5
12...Nxe4 can be met by 13.dxe5 dxe5 14.hxg6 fxg6 15.Re1 Nxf2 16.Rh4ƒ when Black’s
position looks extremely shaky.

13.Bc4! Nhf6
13...Ndf6 14.dxe5 dxe5 15.Ng5 Rf8 16.Nxh7 Nxh7 17.Qxh7± leaves Black far too exposed.

14.dxe5 Nxe5 15.Nxe5 dxe5

574
16.f3 Be6 17.Bxe6 Qxe6 18.b3©
Next White will play Nb2-c4, and the black position contains more than enough weak pawns
and squares to be targeted. White is only one pawn down and it should be easy enough to recover
that deficit, considering the state of Black’s queenside – and the safety of his king will not
improve easily.

A2) 6...Qc7!?

At first this move does not seem like any more than a device for transposing to some other line,
but there are some subtleties.

7.Bd3
7.0-0-0!? is interesting as well, although 7...b5 threatens to convert to the type of scenario we

575
generally try to avoid, namely where White commits to 0-0-0 early, while Black keeps his king
in the centre and starts an early offensive on the queenside. White avoided this with 8.e5!? in
Naiditsch – Rapport, Wijk aan Zee 2014, and although this could be an interesting avenue to
explore, my impression is that the complications are not as favourable for White as they may
appear at first glance. Moreover, I don’t see any need to sacrifice a pawn and generally play so
concretely when the rewards are unclear.

7...e5
7...b5 converts to variation B3 on page 317.

8.0-0-0!N
It seems a little strange that this move has never been played before; but then again, there have
not been all that many games in which Black has used the present move order, and most players
seem to associate an early Bd3 with kingside castling in these lines. There are two reasons why I
prefer the opposite flank here. Firstly, lines with 0-0 are most effective when Black has already
played ...b5, as then we can simply exchange on e5 and focus on the queenside weaknesses.
More importantly, with Black already having played ...e5, he cannot respond to 0-0-0 with the
straightforward attacking plan of ...b5, as he will be smashed in the centre. Thus, we have no
reason to refrain from long castling and a direct assault. True, Bd3 is arguably a waste of a move
– but Black has not exactly been conservative with his use of tempos either.

8...Ngf6
8...b5? 9.dxe5 dxe5 10.Bxb5 is crushing.

9.h4 0-0
9...exd4 10.Bxd4 and 9...Ng4 10.h5 Nxe3 11.Qxe3 both seem promising for White.

576
I also considered:
9...h5
As usual, this is a huge commitment.
10.dxe5!
10.Ng5 Ng4 11.Bc4 0-0 is not so clear.
10...dxe5?!
10...Nxe5 is safer although White has more than one good continuation. For instance,
11.Be2!?² and Black has to worry about the d6-pawn.
11.Bc4! Ng4
11...0-0 12.Qd6! is a typical idea that I have used at various points in these lines: after
12...Qxd6 13.Rxd6± White has by far the more active pieces.

12.Bxf7†! Kxf7 13.Ng5† Ke8

13...Ke7 14.Nd5†! cxd5 15.Qxd5 is killing, for instance: 15...Rf8 16.Qe6† Kd8 17.Nf7†
Rxf7 18.Qxf7+–
14.Ne6 Qb8 15.Nxg7† Kf7 16.Nf5 gxf5 17.exf5‚
The black king’s survival chances are slim at best.

577
10.h5!
10.dxe5 dxe5 11.h5 Nxh5 does not work so well, as 12.g4 could simply be met by 12...Nf4.

10...Nxh5
Grabbing the pawn looks risky, but White would have had a strong attack anyway.

11.g4 Nhf6
11...Nf4 12.Bxf4 exf4 13.Qxf4± is excellent for White.

After the text move, White needs to conduct the attack precisely to punish Black. The right way
is:

12.Bh6! Nxg4 13.Bxg7 Kxg7 14.dxe5 dxe5

578
 In the event of 14...Ndxe5 15.Nxe5 dxe5 16.f3 Nf6 17.Qh6† Kg8 White breaks through with:

18.f4! exf4 19.e5! Qxe5 20.Rde1 Qd4 21.Ne4+–

15.Nh4!ƒ
This looks horrifying for Black. A few lines are:

15...Rg8
15...Nc5 16.Be2 Nf6 17.Rdg1 Rg8 (17...Ng8 18.Qg5 Qe7 19.Qh5! Nf6 20.Qh6†+–) 18.Qg5
Qe7 19.f4‚

16.Rdg1 Nc5 17.Nf5† Bxf5 18.exf5 Nf6 19.Bc4 Kf8

Black is having to play ‘only moves’ just to survive, and still the attack continues.

579
20.Qh6† Rg7 21.fxg6 hxg6?!
Black would like to keep his pawn cluster intact, but it leads to a worsening of his situation.

21...fxg6 is preferable although 22.f4! exf4 23.Rf1ƒ keeps up the pressure.

22.Qh8† Ng8

23.Rh7! Rxh7 24.Qxh7

With the crushing threat of Rxg6.

B) 5...b5

580
 When Black gets his queenside counterplay underway this early, we will of course avoid
castling for the time being.
6.Bd3 Nd7 7.Nf3
After a few natural moves, Black now has a choice, the main options being B1) 7...Nb6, B2)
7...Bb7 and B3) 7...Qc7.

7...Ngf6 transposes to line D of the previous chapter, where 8.Ne2! was recommended.

7...a6
This is similar to variation B2, and White can follow the same plans.
8.Ne2!? e5
8...c5 9.c3 Bb7 10.Ng3 simply transposes to variation B2.
8...Nb6 has been played in multiple games but I see little point to this move, other than the
possibility of ...Bg4 being annoying. Therefore I recommend 9.h3, and if 9...Nc4 10.Bxc4
bxc4 11.0-0 Black’s pawn on c4 is an obvious target.
After the text move we can of course play c2-c3 with a normal position, but I think an even
more attractive approach is:

581
9.dxe5!?N
Following the plan I have been reiterating throughout these variations: if Black plays both
...b5 and ...e5, then White can simplify in the centre and focus on putting pressure on
Black’s queenside.
9...dxe5 10.a4 Bb7 11.c4 Ngf6 12.h3²
White has a pleasant game with pressure on the queenside. Please note that it is useful that we
still have the dark-squared bishops on the board, as the e3-bishop does a great job of preventing
annoying ideas like ...Nc5.

B1) 7...Nb6

8.a4
Forcing Black to commit himself in one way or another.

582
8.h3 Nf6 9.Ne2 Nc4! seems okay for Black.

8...b4
8...Nc4?! is never played, and indeed after 9.Bxc4 bxc4 10.0-0 Nf6 11.Qe2± Black’s light-
squared bishop will be perpetually stuck behind its own pawns.

9.Ne2 a5

10.h3 Nf6 11.0-0 0-0 12.Ng3 Ba6

12...c5N 13.c3 resembles our usual lines, and if 13...c4 14.Be2² White is happy.

12...Nbd7 was played in Ponomariov – Mamedov, Moscow (blitz) 2010, when 13.Bh6!N±
would have given White an excellent version of a Pirc, as here Black has wasted time moving his
knight to b6 and back.

The text move seems logical, and was played in Zawadzka – Leszner, Zgierz 2018. However, it
is not clear to me if the impending bishop exchange really helps Black, as the cxd3 recapture will
expose the c6-pawn as a fresh weakness.

13.Bh6!?N
13.Bxa6N 13...Rxa6 14.Bh6 also seems quite pleasant for White.

583
13...d5
13...Bxd3 14.cxd3²

14.Bxg7 Kxg7 15.e5 Ne4 16.Qe2²

Depending on how the play develops, White may develop an initiative on either flank, or even
both simultaneously.

B2) 7...Bb7

8.Ne2!?
Just continuing in the same spirit as in the similar positions from Chapter 12.

8...a6

584
 This is the most common continuation, preparing ...c5.

8...Ngf6 converts to variation D23 of the previous chapter on page 299, where we saw 9.Bh6!?
working well; such possibilities show the great flexibility behind White’s move order.

8...c5 has not yet been played, probably with good reason, as 9.dxc5! would be a safe pawn grab.
(9.c3 c4 10.Bc2 Ngf6 11.Ng3 Ng4 12.Bg5 h6 13.Bh4 Ngf6 14.a4 a6 15.0-0 leads to interesting
play, but is totally unnecessary.)

9.Ng3 c5

10.c3
We have once again achieved our dream central structure, so the onus is on Black to create
counterplay. We could simply castle next, and I would still like White’s position – but as we will
see, we may strive to exploit the fact that we still have a rook on h1.

10...Ngf6
10...h5!? pre-empts the h4-h5 plan and prevents Bh6 for the time being, but also creates serious
weaknesses for Black – notably the g5-square. Therefore we can simply switch plans with 11.h3,
preventing any ...h4-h3 ideas, before castling short and plonking a knight on g5. Play could
continue:

585
11...h4 (it is important to note that 11...Ngf6?! 12.e5! cxd4 13.cxd4 dxe5 14.dxe5 Nd5 15.e6
Ne5 16.exf7† Kxf7 17.Ng5†! Kg8 18.Rd1± turns out great for White) 12.Ne2 cxd4 13.cxd4
Ngf6 14.Ng5² White enjoys a stable edge.

10...e5
This leads to a typical Spanish-style position. White should be somewhat better after closing
the centre with d4-d5 – I particularly like to do this in the Spanish after Black has
committed his bishop to b7. However, with Black’s knight still on g8, he could consider
putting it on e7 and preparing ...f5. Thus, I prefer to maintain the central tension for the
moment, so my suggestion would be:

11.a4!
Trying to provoke ...c4.
11...cxd4

586
 Playing 11...c4 with a pawn still on d4 counts as a mini-victory for White, as 12.Bc2²
leaves Black unable to fight against the central pawns.
11...Qb6 is legal, but the queen can hardly feel safe while placed on the diagonal of the e3-
bishop.
12.cxd4 exd4 13.Bxd4 Ngf6 14.axb5 axb5 15.0-0 0-0

16.b4!²
White has clear targets on b5 and d6, while Black’s position remains mostly tied down. There
isn’t much hope for ...d5 and there is no obvious way to simplify in the near future, so Black
looks likely to remain passive.

11.Bh6!?
The simplest option is 11.h3 0-0 12.0-0, in the same spirit as variation D242 of the previous
chapter. However, White is completely justified in playing for higher stakes.

587
11...0-0
This seems the natural reply, but White can continue to use the fact that he hasn’t castled yet.
I also checked:
11...Bxh6 12.Qxh6 cxd4 13.cxd4 Qa5†
And now a simple improvement is:

14.Ke2!N
Black’s troubles with the king in the centre are more extreme than White’s.
14...Rg8!?
14...d5 15.e5 Ne4 16.Rhc1 leaves White in control.
14...e5 15.Rhd1 exd4 16.Kf1 is also unpleasant for Black.
15.Rhd1 g5!? 16.Nxg5 Qb4

17.Nf3 Qxb2† 18.Qd2 Qxd2† 19.Nxd2

The queen exchange has solved certain issues for Black, but new problems have emerged. His

588
queenside is vulnerable, with White planning Nb3-a5 and/or a2-a4, perhaps also rerouting the
g3-knight towards the queenside at some point. White’s centre is strong, and he can further
consolidate it with f2-f3.

12.Bxg7
12.h4 can be met by 12...Bxh6! 13.Qxh6 cxd4 14.cxd4 e5! when h4-h5 is not such a threat
because of ...Ng4.

12...Kxg7

13.h4! cxd4
The obvious alternative is:
13...h5?!
With accurate play, we can bust through the defences.
14.e5!
14.Nf5† is less accurate due to: 14...gxf5! 15.Qg5† Kh7! 16.Qxf5† Kh8 17.e5 dxe5
18.dxe5 e6 19.Qg5 Nh7! 20.Qh6 f5÷
14.0-0 would be slightly more pleasant for White, although Black has a marginally
improved version of the main line below, as he has some additional options besides ...cxd4.
This is immaterial though, as White can do better.
14...dxe5

589
15.Nf5†! gxf5 16.Qg5† Kh8
16...Kh7 17.Bxf5†! is the reason why it was necessary to play e4-e5 before Nf5†. 17...Kh8
18.Qh6† Kg8 19.Rh3+–
17.Qh6† Kg8 18.Rh3 e4
18...f4 19.dxe5 wins for White.
19.Rg3† Ng4

20.Qg5†!
Gaining a tempo and winning after:
20...Kh8 21.Qxh5† Kg7 22.Qxf5 Ndf6 23.Rxg4† Nxg4 24.Qxg4† Kh8 25.Bxe4+–

14.cxd4

590
14...h5!?
This looks risky, but the exchange on d4 changes some of the tactical details from the previous
note. However, once Black commits to this move, the g5-square becomes a serious weakness on
the kingside, and we can make Black’s life unpleasant by simply plonking the knight on the
ready-made outpost, while keeping ideas like e4-e5 in reserve. We also have the option of a2-a4
at any moment, combining threats on both flanks.

15.0-0²
15.e5!? dxe5 16.Nf5† gxf5 17.Qg5† Kh8 18.Qh6† Kg8 19.Rh3 might be good for White too,
but the inclusion of the pawn trade on d4 gives Black some extra possibilities, and there doesn’t
seem to be any need to get involved in this kind of messiness.

B3) 7...Qc7

591
8.Ne2!
White’s ideas are essentially the same as the ones discussed in great detail in the previous
chapter. Black can hardly profit from delaying ...Nf6, as he is quickly running out of useful
moves.

8.0-0 Ngf6 converts to the 8.0-0 Qc7! line, as discussed in variation D1 of the previous chapter
on page 291. I explained my reasons for avoiding it in great detail there, although it’s certainly
playable.

8.h4 can also be considered, but 8...Ngf6 9.Bh6 Bxh6 10.Qxh6 b4 11.Ne2 c5÷ seems reasonable
for Black.

8...Bb7
If Black wishes to leave the knight on g8, this seems like the most natural choice.

8...Ngf6 would convert to variation D22 of the previous chapter on page 297.

8...Rb8
This has been played in a few games but it seems pointless. We will meet ...c5 with c3
anyway, and ...Nb6-c4 is never going to be a serious threat.
9.Ng3 Nb6
9...c5 10.c3 does not change much.
9...Ngf6 10.Bh6 Bxh6 11.Qxh6 c5 12.c3 cxd4 13.cxd4 Qa5† 14.Ke2 is another scenario
where both kings are centralized but Black suffers from it more than White.

592
 If ...Nc4 was going to be a problem, we could easily play b2-b3 and make the knight on b6
look silly. I don’t consider it a threat at all though.
10.0-0 Nc4
10...Bg4 11.Ng5 h6 12.h3 Bc8 13.Nf3± leaves White in control.
11.Bxc4 bxc4 12.b3 cxb3 13.axb3²
White has the better pawn structure and a generally excellent position.

9.Ng3 Ngf6
9...a6 continues to avoid committing the knight. Black’s usual reason for doing this is to avoid
White’s kingside aggression with Bh6 and h4-h5, but we saw in the previous chapter that
White’s Ne2-g3 plan is usually connected with short castling anyway, so we should be perfectly
happy with 10.0-0 and so on. (If White really wants to, he can also consider 10.a4!? in order to
keep the h4-h5 attacking ideas alive for a little longer, but it’s not necessary.)

593
9...h5 is another move that we should be happy to see, as Bh6 and h4-h5 was not really our main
plan anyway. So we can simply play 10.h3 and follow our usual plans, with the added bonus that
Black’s kingside and especially the g5-square have been permanently weakened.

10.h3
This is the simplest continuation for our purposes. We are on the verge of transposing to a
variation from the previous chapter, after both sides castle and Black plays ...e5 and/or ...c5.
Note the following alternatives:

a) 10.0-0 is slightly inaccurate, as it transposes to a note in the previous chapter on page 292,
where it was pointed out that the strong 10...Ng4! complicates matters.
b) On the other hand, 10.Bh6!? seems quite promising, for instance: 10...Bxh6 (10...0-0 11.h4
gives White a strong attack) 11.Qxh6 c5 12.c3 cxd4 13.cxd4 Qa5† 14.Qd2 Qxd2† 15.Kxd2²
White has a strong centre and Black’s queenside weaknesses will give him problems in the
endgame.

Conclusion

This was not the longest or most complicated chapter, as most of White’s ideas can be clearly
understood in the light of the previous chapter. When you know what you are aiming for when
...Ng8-f6 is played, it is not too hard to proceed along parallel lines when the knight stays on g8.

After 1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 c6 5.Qd2, we started by considering 5...Nd7 6.Nf3. If
Black follows up with 6...e5, we can forget about the previous chapter and simply play 7.0-0-0,
with fine prospects on the kingside no matter what Black does. It’s a similar story after 6...Qc7!?
7.Bd3 e5, when 8.0-0-0!N is a nice innovation. In the previous chapter I was not keen on
committing to long castling too early, as Black was able to keep the centre closed and start his
own queenside attack, but once he has played ...e5 the centre is too volatile for such plans to
work.

We then turned to the more common 5...b5, when 6.Bd3 Nd7 7.Nf3 is our choice. I analysed a
few possibilities just to be thorough, but the general picture is the same: White essentially
follows the same regrouping plan with Ne2-g3 as shown in the previous chapter, and a
transposition is quite likely after ...Ngf6 – and in certain lines, White can also consider Bh6
and/or h4-h5 as a promising bonus option.

594
A) 6...0-0 7.Bh6 Bg4!? 8.0-0-0 Nc6 9.h4!N 321
A1) 9...Bh5 322
A2) 9...e5 323
B) 6...b5 7.Bh6 0-0 8.Bd3 Nc6! 324
B1) 9.d5!? 326
B11) 9...Bxh6!?N 326
B12) 9...Ne5 327
B13) 9...Nb4!?N 10.h4 331
B131) 10...Nxd3† 331
B132) 10...e5! 333
B2) 9.Bxg7 334

595
1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7
4...a6 has no special significance. White should play 5.Nf3, and after 5...Bg7 6.Qd2 we
continue as normal. If Black tries to get creative by leaving the bishop on f8 and playing 5...b5?!,
then 6.e5! Ng4 7.Bf4± is unpleasant for him.

5.Qd2 a6
Just like the ...c6 variations we have analysed, the ...a6 set-up comes in two main flavours: the
Pirc version with the knight on f6, which we analyse in this chapter, and the Modern version
where the knight stays on g8 for longer, which we will cover in the following two chapters. In
both cases, Black plans a quick ...b5; but unsurprisingly, the placement of the knight on the
kingside makes a big difference to the character of the play. Just as with the ...c6 set-ups, we will
cover the Pirc version first.

6.Nf3
As you may expect, we will look to exploit the early ...Nf6 move by playing an early Bh6 and
launching a kingside assault. Against the ...c6 set-ups I was not so keen on committing to this at
an early stage, but against ...a6 there are some differences. To take a simple example, a quick
...b5-b4 can be met by Nd5 here, so Black will have to develop his counterplay by more
sophisticated means.

We will start by analysing lines with A) 6...0-0 where Black delays ...b5, followed by the main
lines with B) 6...b5.

6...Ng4?! has been played in a bunch of games but 7.Bg5 leaves Black without a good follow-
up. A GM game continued 7...h6 8.Bh4 g5 9.Bg3 Nf6 10.h3 b5 11.Bd3 Bb7 and now in
Arakhamia-Grant – Speelman, London 2015, White could have obtained a clear advantage with
12.e5!N.

596
 A) 6...0-0

7.Bh6 Bg4!?
This seems like the most interesting of Black’s independent possibilities.

7...b5 is by far the most popular continuation, and it will be covered under the 6...b5 move order
in variation B.

7...c5 8.Bxg7 Kxg7 9.0-0-0 cxd4 10.Nxd4 seems like a particularly nice version of a Sicilian
‘Dragadorf’ for White, as he has swapped off the Dragon bishop without needing to spend time
on f2-f3.

The only other possibility worth mentioning is 7...Nc6!? 8.0-0-0 b5 (8...Bg4 converts to 7...Bg4;
8...e5? simply loses a pawn after 9.Bxg7 Kxg7 10.dxe5 dxe5 11.Qxd8 Nxd8 12.Nxe5±) and
now 9.d5!? resembles the later variation B1, the only difference being that White has played 0-0-
0 instead of Bd3, which doesn’t seem like a bad change and could even prove helpful in some
lines.

8.0-0-0 Nc6
This position has been reached in a few games. I found a promising new idea:

597
9.h4!N
9.Bxg7 Kxg7 10.Qe3 was a tad more pleasant for White in Shomoev – Fedoseev, Taganrog
2015, but we can be more ambitious.

9.d5!?N 9...Na7 10.h4 Nb5 11.Bxg7 Kxg7 12.h5 could be interesting to analyse further, but the
text move just seems stronger.

I analysed two main tries for Black: A1) 9...Bh5 and A2) 9...e5.

9...d5? is pointless due to 10.h5‚ when Black does not even have time to capture the e4-pawn, as
he is already threatened with immediate annihilation on the kingside.

9...Bxf3 10.gxf3 does not help Black either, for instance: 10...Nh5 (10...e5 11.Bxg7 Kxg7
12.dxe5 dxe5 13.Qe3 followed by h4-h5 with a crushing attack; 10...e6 11.h5 Nxh5 12.Be2‚)
11.Bxg7 Kxg7 12.d5!±

9...b5 is well met by: 10.h5! Bxh5 11.Bxg7 Kxg7

598
12.Be2! The open h-file is just too much, for instance: 12...b4 13.Nd5 Nxe4 14.Qe3 Nf6
15.Nxf6 exf6 16.d5 Re8 (16...Ne5 17.Nd4+–) 17.Qd2±

A1) 9...Bh5

Black prevents the h4-h5 advance, but only temporarily.

10.Bxg7 Kxg7 11.Be2 e5

11...e6 12.Kb1! is a strong move, the main idea of which is revealed after: 12...b5 13.e5 dxe5
14.dxe5+–

11...b5 12.e5 dxe5 13.Nxe5 is also unpleasant for Black: 13...Bxe2 (13...Nxe5? 14.dxe5 Qxd2†
15.Rxd2+–) 14.Nxc6 Qd6 15.Nxe2 Qxc6 16.h5‚

599
12.dxe5!? dxe5 13.Qe3

13...Qb8
13...Qe7 runs into 14.Nd5± of course.

14.Nd2 Bxe2
There is no time for 14...Qa7? due to 15.Qxa7 Rxa7 16.f3 when Black’s bishop is caught.

15.Qxe2 h5 16.Nb3
Intending f2-f3 followed by g2-g4. White can swap off the f6-knight with Nd5 at any moment,
which will make the attack more potent. Black is also not helped by the remoteness of his queen
from the kingside.

A2) 9...e5

600
 Countering in the centre seems logical enough – but we can ignore it.

10.h5! Bxh5
Alternatives are no better, for instance:

10...exd4 11.hxg6 dxc3 12.gxh7† Nxh7 13.Qf4 cxb2† 14.Kb1 Qf6 15.Qxg4± leaves White with
the much safer king.

10...Bxf3 11.gxf3
11.hxg6!? is also promising.
11...Nxh5 12.dxe5
12.Bc4!? is also strong, but one good option is enough.
12...dxe5 13.Qe3

601
13...Nd4
13...Qe7 and 13...Qf6 are clearly not comfortable because of Nd5; and other queen moves
would allow an instant kill on the kingside.
14.Bc4 Kh8 15.Bxg7† Kxg7 16.f4! Nxf4 17.Nd5 b5
In the event of 17...Nxd5 18.Qh6† Kf6 it may look as if Black could be getting away, but
19.c3! Nb6 20.cxd4 Nxc4 21.dxe5†± gives White excellent winning chances.
18.Nxf4 exf4
18...bxc4 19.Ng2 h5 20.c3±
19.Qxf4 bxc4 20.Qh6† Kf6 21.Qe3 c5 22.c3²
Black is not dead yet, but the position is clearly unpleasant for him due to his overexposed
king.

11.dxe5 dxe5

12.Qg5! Qe7
Inserting 12...Bxh6 13.Qxh6 does not really help Black: 13...Ng4 (13...Qe7? 14.Ng5+–)
14.Rxd8 Nxh6 15.Rxf8† Kxf8

602
16.Nd2! Black has no good way to save his h5-bishop from being trapped with f2-f3 and g2-g4.
The best he can do is 16...Nd4 17.f3 Kg7, but White keeps the upper hand with: 18.Nd5! There
is no need to hurry with the g-pawn. (18.g4 Bxg4 19.fxg4 Nxg4÷) 18...b5 19.c3 Ne6 20.Ne3±

13.Bxg7 Kxg7

14.Nh4! Kh8 15.f3©

Black’s bishop is basically dead on h5, and White clearly has overwhelming compensation for
the pawn.

B) 6...b5

603
 This is by far Black’s most popular choice, especially when we factor in transpositions
whereby Black castles and then follows up with 7...b5.

7.Bh6 0-0
This is practically forced.
7...Bxh6? 8.Qxh6 b4 (8...Bb7 9.e5 b4 10.Ne2 Bxf3 11.gxf3± was great for White in Hebden –
Pein, Bunratty 2016) 9.Nd5! leaves Black with a most unpleasant choice:

9...Nxd5 (9...Nxe4 was no better after 10.Bd3 Nf6 11.Nxf6† exf6 12.0-0± in Boudier –
Mataigne, La Fere 2007) 10.exd5 Nd7 White obviously stands much better and there is nothing
wrong with 11.Qg7, which has been played in several games – but White can play that move
virtually any time he wishes, so for now I like the idea of opening a second front with 11.a3!N±
even more.

604
8.Bd3
White should take a moment to safeguard his centre, as 8.h4 b4 9.Nd5 Nxe4 10.Qf4 e6! works
out well for Black.

8...Nc6!
Black already has to be rather precise. Other moves could quickly land him in trouble, for
instance:

8...e5?! would have been an excellent move, if it were not for the fact that White can simply
capture the given pawn with 9.Bxg7 Kxg7 10.dxe5 dxe5 11.Nxe5! as in Mortazavi – Banas,
Cappelle-la-Grande 1991.

8...c5?! has been played in several games but 9.dxc5 dxc5 10.Bxg7 Kxg7 11.e5± gives White an
easy initiative.

8...Bg4?! 9.e5! Nfd7 10.Bxg7 Kxg7 occurred in Weinzettl – Nyback, Belgrade 2002, when
White could have caused big problems with:

605
11.h4!N 11...h5 (11...Bxf3 12.gxf3 dxe5 13.h5! Rh8 14.Be4 c6 15.dxe5 Nxe5 16.h6†+–)
12.Ng5±

Finally, 8...Bb7 has been played in a lot of games but 9.Bxg7 Kxg7 10.e5! poses Black
unpleasant problems. For instance: 10...Ng8 (or 10...Nfd7 11.h4ƒ with an easy attack)

11.h4! dxe5 (11...h5 12.e6!+–) 12.dxe5 Bxf3 (12...c5? 13.h5 was horrible for Black in
Rathnakaran – Rahman, Al Ain 2015) 13.gxf3 Nc6 14.Qe3 Qd4 15.Be4 Qxe3† 16.fxe3±

It is worth analysing two continuations for White: B1) 9.d5!? and B2) 9.Bxg7.

9.e5 allows a nice tactical resource: 9...dxe5 10.dxe5 Bxh6 11.Qxh6 Ng4 12.Qf4 Ncxe5!
13.Nxe5 Qd6! and Black is doing just fine.

606
 B1) 9.d5!?

We have a further split, with B11) 9...Bxh6!?N, B12) 9...Ne5 and B13) 9...Nb4!?N all
deserving attention.

9...Nb8?! is too slow, and 10.h4! e6 11.h5 gives White a strong attack.

B11) 9...Bxh6!?N 10.Qxh6 Ng4 11.Qd2 Nce5 12.Nxe5 Nxe5

12...dxe5 13.f3 Nf6 14.h4 h5 transposes to variation B12 below.

At first, Black does not seem to be doing badly at all: he has exchanged a few pieces and is
ready to open the centre with ...e6, so it is not clear if White would be justified in starting a
kingside attack. However, the knight on e5 is a little awkward, and White has not committed to
anything aggressive on the kingside. Instead, 0-0 could be more attractive, in connection with a2-

607
a4 to create some targets on the queenside.

13.Be2 e6
13...c6?! is premature due to 14.f4±.

13...b4 just helps White to regroup the knight to a better square, and after 14.Nd1 a5 15.Ne3
White has good prospects on the kingside. For instance: 15...Ba6 (15...Bb7 16.0-0-0! e6 17.h4ƒ)
16.Bxa6 Rxa6

17.f4 Nd7 18.0-0-0 Nc5 19.Qd4 Qb8 20.h4 a4 21.h5‚ White is clearly ahead in the attacking
race.

14.0-0!?
From a theoretical perspective, I think White can be satisfied with this modest continuation, as
he maintains a slight pull without giving Black much counterplay.

14.h4 b4! 15.Nd1 exd5 16.h5 Qf6 is messy, and White’s attack seems unlikely to succeed.

14.a3!? is the most ambitious move, intending 0-0-0 followed by launching an attack with h2-h4.
It doesn’t seem easy for Black to get play on the queenside, so he should try to open the centre
with: 14...exd5 (14...Bb7 15.0-0-0 exd5 16.exd5 Qf6 17.h4ƒ looks good for White)

608
15.Qxd5! (15.Nxd5 Bb7 16.0-0-0 Bxd5 17.exd5 Qf6= would be fine for Black) 15...Be6 16.Qd2
White’s position is slightly more pleasant, with a kingside attack still on the agenda.

14...b4
Alternatives are no better:

14...f5 enables White to keep control with 15.f4 Nf7 16.exf5 exf5 17.Bf3 followed by Ne2-d4.
14...Bb7 15.b4!? (15.f4 Nd7 16.Bf3 is a good alternative) 15...exd5 16.exd5 c6 17.a4² with
pressure on the queenside.

15.Nd1 exd5 16.exd5 a5 17.a3 Rb8

17...bxa3?! 18.Rxa3± worsens Black’s structure for no good reason.

18.axb4 axb4 19.b3²

609
 The b4-pawn is a weakness requiring constant defence. Meanwhile White has many ways of
improving his position: the knight may head for any number of squares via b2, e3 or even f2
(should White feel inclined to play f2-f4); and the a1-rook may go to a4 to target the weak pawn,
or perhaps to a7 – especially if Black’s queen moves away from the defence of c7.

B12) 9...Ne5

10.Nxe5 dxe5 11.h4!

11.Bxg7 Kxg7 was played in Kazakov – Chernin, Lvov 2000, but I find this less appealing, as
12.h4N could then be met by 12...h5÷ of course.

11...Bxh6N
It seems pretty logical to make this exchange and then immediately expel the queen with
...Ng4.

I also considered:
11...e6 12.h5! Nxh5N
12...Bxh6 13.Qxh6 Ng4 14.Qd2 transposes to the 13...e6 14.h5 line in the notes to the main
line below, and was the actual move order of the Gindi – Henrichs game quoted there.

610
13.Be2! b4 14.Nd1 Nf4!?
This seems risky, but 14...exd5 15.Bxh5 gxh5 16.Bxg7 Kxg7 17.exd5 Bf5 18.Ne3± is no
fun for Black either.
15.Bxg7 Kxg7 16.g3 Nxe2 17.Qh6† Kf6 18.Ne3
18.Kxe2? exd5 would squander at least some of White’s advantage.
18.Qh4†!? is a strong alternative but it is not required, as the text move is simpler.

18...Nd4
18...Ke7 19.Kxe2± gives White a huge initiative.
19.c3! Nf3†
19...Nb5? 20.0-0-0+– leaves Black defenceless.
20.Ke2 Bb7 21.Rad1!±
The knight is trapped, so White regains the temporarily sacrificed piece while Black’s king
remains extremely vulnerable.

611
12.Qxh6

12...Ng4
This seems like the most challenging continuation.

12...b4 13.Nd1 e6 14.Ne3 exd5 15.exd5² favours White, since 15...Nxd5?? 16.Nxd5 Qxd5
17.h5 ends in disaster for Black.
I also considered:
12...e6 13.h5 b4
In the event of 13...Qe7 14.f3 b4 15.Ne2 a5 16.0-0-0 White’s attack seems more potent, for
instance: 16...exd5 17.exd5 Ba6 18.g4 Bxd3 19.Rxd3 a4 20.g5 Nxh5 21.Ng3 f5 22.Nxh5
gxh5 23.g6±

14.hxg6

612
 Retreating the knight immediately gives Black the additional option of ...Ng4, which may
not be a big deal – but it seems easiest to rule it out entirely.
14...fxg6
14...bxc3? is impossible due to: 15.g7! Re8 16.d6! cxd6 17.g4+–
15.Nd1 exd5 16.Ne3! Qe7
16...dxe4? 17.Bc4† is no good for Black.

17.Nxd5
17.exd5!? Qg7 18.Qg5 is a more ambitious continuation which keeps the position more
complicated, but the text move seems fine too.
17...Nxd5 18.exd5 Qg7 19.Qxg7† Kxg7 20.a3²
Black’s disconnected structure ensures White of a pleasant endgame.

13.Qd2

613
13...h5
13...e6 14.h5 transposes to a game which we will follow momentarily. 14...b4 15.Ne2!?N (I
slightly prefer this to 15.Nd1 as played in Gindi – Henrichs, Biel 2017) 15...a5 16.f3 Nf6
17.Qg5 exd5 18.0-0-0 Ba6 (18...a4 19.hxg6 fxg6 20.Qxe5±)

19.Qxe5 Bxd3 20.Rxd3 Re8 21.Qg5 Qe7 22.hxg6 fxg6 23.exd5²

14.f3 Nf6

15.Qh6!
The idea behind this can be seen from the line 15.g4 Kg7! 16.0-0-0 (16.g5 Nd7÷) 16...Qd6
17.Rdg1 Rh8 (alternatives include 17...e6 18.Ne2 and 17...Bd7 18.Ne2 hxg4 19.h5) 18.gxh5
Nxh5 19.Ne2 e6 when matters remain unclear.

The text move prevents ...Kg7 entirely, and White is ready to launch the attack with g2-g4

614
whenever he wants to. There is no hurry, because Black can’t really dislodge the queen from h6.

15...e6
15...b4 is simply met by 16.Nd1 or 16.Ne2, when White’s knight manoeuvres to a better
location.

15...Qd6 looks natural but after 16.0-0-0 it is not clear what Black’s queen is doing, and all his
plans seem to be too slow. For instance: 16...b4 17.Ne2 a5 18.g4 Bxg4

19.Rdf1! (19.fxg4 Nxg4 20.Qd2 Nf2„) 19...Bd7 20.Ng3‚

16.g4! b4
16...exd5?! 17.Nxd5 Nxd5 18.exd5 Qf6 19.gxh5+– wins at least one pawn for no
compensation.

17.Nd1 exd5 18.Ne3! dxe4

17...Re8 18.Ne3 Qd6 19.0-0-0 Qf8 20.Qg5ƒ gives White a strong attack without any hassle.

The text move is more critical but White will emerge victorious with some precise moves:

615
19.Bc4! exf3 20.g5!
Taking on g6 immediately is not clear at all, but delaying it in this way leads to a crushing
attack.

20...Nh7 21.Rd1 Bd7

This horrible self-pinning move can be explained by the fact that 21...Qe7 loses immediately to
22.Nd5.

22.Qxg6† Kh8 23.Qxh5+–

B13) 9...Nb4!?N

This untested move may well be Black’s best.

616
10.h4
One of the points of Black’s move order is that 10.Bxg7 can be met by: 10...Nxd3†! (rather
than 10...Kxg7? 11.Be2!±) 11.cxd3 Kxg7 12.h4 h6÷

Attempting to ram Black with the h-pawn is the most natural attacking try. I analysed two main
responses: B131) 10...Nxd3† and the more accurate B132) 10...e5!.

10...Bg4 is another typical reaction to h2-h4, but White is well placed to meet it: 11.Bxg7 Nxd3†
12.cxd3 Kxg7

13.Nd4! Now Nc6 is an annoying possibility, and after 13...Bd7 14.h5 (14.Nce2!? is a
reasonable alternative) 14...Nxh5 15.0-0-0 White’s initiative must be worth at least a pawn.

B131) 10...Nxd3† 11.cxd3

617
11...Bxh6
11...e6 12.Bxg7 Kxg7 13.h5 Nxh5 14.g4 Qf6 15.0-0-0 Qxf3 16.gxh5ƒ also gives White nice
play on the kingside.

12.Qxh6 Ng4 13.Qd2 e6

13...h5 is met by 14.Ng5! when White’s attack will continue unhindered. For instance:
14...Bd7 15.f3 Nf6 16.g4 (16.0-0-0!?ƒ) 16...c6 17.dxc6 Bxc6

18.Ne2! Followed by Ng3 and Black’s kingside will crumble.

618
14.dxe6 Bxe6 15.h5
15.Ng5!? would also be worth considering. Opening the h-file seems more natural though.

15...Ne5
15...Qf6 can be met by 16.Ng5! and if 16...Ne5 (16...h6 17.Nxe6 fxe6 18.f3 Ne5 19.hxg6± is
also no fun for Black) 17.hxg6 hxg6 18.Nh7+– it transpires that the queen was not so well
placed on f6.

16.Nd4 Qf6
16...Bd7 17.0-0-0± seems too slow for Black.

17.f3!
Preparing Nxe6. It is useful to compare the immediate exchanges on e6 and g6 to see why this

619
preliminary move is needed.

17.0-0-0? b4! 18.Nce2 Bxa2µ would be completely the wrong direction to take.

17.Nxe6 fxe6! (17...Qxe6?! 18.0-0-0 favours White after 18...c6 19.Kb1± or 18...b4 19.Nd5 c6
20.Nc7 Qxa2 21.Nxa8 Rxa8 22.Qc2±)

18.f3 (18.hxg6 Qxf2†! is a vital detail) 18...g5÷

Inserting 17.hxg6 fxg6 before 18.Nxe6 Qxe6÷ is also less convincing than I would like. An
important point is that 19.0-0-0 is no longer possible because of:

19...b4 20.Nd5 c6 21.Nc7 Qxa2 22.Nxa8 b3!–+ The open f-file makes all the difference
compared to the analogous position from the 17.Nxe6 line noted above.

620
17...g5
17...c5 18.Nxe6 fxe6 19.hxg6 Qxg6 20.0-0-0² looks pleasant for White due to his more secure
king.

18.Nc2!
White keeps the options of castling on either side, and at the right time he can choose between
them. Meanwhile, the knight will be ideally placed on e3, threatening Nd5 and keeping ideas
like d3-d4 in reserve.

18...c5 19.Ne3 Nc6 20.0-0-0!?

20.0-0 also deserves attention.

20...Nd4 21.Kb1²

B132) 10...e5!

621
 I was unable to prove anything against this move, although there are certainly enough ideas to
keep the game interesting.

11.Ne2
11.Bxg7 is met by 11...Nxd3† (but not 11...Kxg7? 12.Be2!±) 12.cxd3 Kxg7 when it is hard to
get any real venom into White’s attack.
11.dxe6 Nxd3† 12.cxd3 Bxh6 13.Qxh6 fxe6 should also be safe enough for Black. For
instance:

14.Ne2 Qe7 15.Ng3 Qg7!= (Avoiding 15...e5 16.h5 or 15...Ng4 16.Qg5 Qxg5 17.hxg5, both of
which give White more cause for optimism.)

11...Nxd3† 12.cxd3 Bxh6 13.Qxh6

622
13...Ng4 14.Qd2 h5÷
White can certainly continue hoping for something on the kingside but I feel like Black’s
defences should be solid enough, so I will leave things here and invite interested readers to carry
out their own investigations from this point.

B2) 9.Bxg7 Kxg7

Initially, I was looking for various alternatives for White here, especially with a view to getting
a kingside attack flowing with h4-h5, while preventing Black’s counterplay in the centre.
However, as Daniel Fernandez pointed out during our discussions on the Pirc, White has the
opportunity to gain a slight edge in the endgame, using a relatively forcing sequence which has
already been demonstrated in practice.

623
10.e5
Both 10.d5!? and the rare 10.Nd5!? offer interesting ways to go about attacking on the
kingside, but the position remains complicated.

10...dxe5 11.dxe5 Ng4 12.0-0-0 b4

12...Ngxe5? is never played because of: 13.Nxe5 Nxe5 14.Qe2 Nxd3† 15.Rxd3 Qe8 16.Nd5±

13.Rhe1!
Rather a shocking move at first, but it has become a relatively common occurrence in recent
years.

Vigus only mentions the passive 13.Nb1?! in the relevant chapter of the Dangerous Weapons
book.

13.Ne4?! Qd5 is not really working for White either.

13.h3!? bxc3N 14.Qxc3 Nxf2 15.Be4 is interesting but only equal with accurate play, so we will
focus on the strongest move.

13...Rb8!?
13...bxc3? 14.Qxc3 Qd5 15.Be4 Qxa2 16.Bxc6 was much better for White in Sulava –
Kazakov, Evry 2008, and a few other games.

13...Be6?! can be met by the simple 14.Nb1 when the inclusion of Rhe1 and ...Be6 makes a huge
difference to the evaluation. For instance, after 14...Qd5 the rook on e1 makes 15.Be4 possible,
and 15...Qc5 16.Bxc6 Qxc6 17.h3 Nh6 18.Qxb4± left White an easy pawn up in Rodriguez Vila
– Tristan, Villa Constitucion 2018.

624
I also considered:
13...Bf5!?N
Not a bad novelty, but White can reach a favourable endgame without too much difficulty.

14.h3!
14.Na4?! is a weird engine suggestion, when 14...Bxd3 15.cxd3 Qd5= seems fine for
Black.
14...bxc3
14...Nh6?! 15.Ne2±
15.Qxc3 Nxf2 16.Bxf5 Nxd1 17.Rxd1 Qb8 18.Be4
A tempting alternative is 18.e6†!? f6 19.Rd7 gxf5 20.Qxc6 Re8 21.Nd4² when White will
almost certainly restore approximate material equality by taking on f5 and c7, and his pieces
remain the more active.
18...Nb4 19.e6† f6 20.Qc5 Nxa2† 21.Kb1 Qb4 22.Qxb4 Nxb4 23.Bxa8 Rxa8

625
24.Rd7 Re8 25.c3 Nc6 26.Rxc7 Ne5 27.Nd4²
Black has a difficult defence ahead.

14.Na4 Qd5 15.b3 Ncxe5 16.Nxe5 Nxe5 17.Bf1

17.Bxa6 is not so easy after: 17...Qxg2! 18.Rxe5 Qxh2 (18...Bxa6 19.Qf4±) 19.Qd4 Bxa6
20.Re6† Kg8 21.Rxa6 Rbd8÷

17...Qxd2† 18.Rxd2

My first impression of this position was that Black should be able to somehow give up his
extra pawn to convert to a drawish endgame; but as Daniel pointed out, it isn’t so simple for
Black to do that. The trouble is that there are many weaknesses on the queenside, and White’s
rooks are excellently placed to utilize them, for instance with Rd5-c5. Giving back a pawn will
not really help Black to safeguard these.

18...Kf6!?N
This seems like the trickiest option which forces White to play accurately to retain the
advantage.

I also considered:
18...Nc6 19.Rd5 a5!?N
a) 19...e6 20.Rc5 Bb7 21.Bxa6 occurred in Kosteniuk – Tan Zhongyi, Huaian (rapid) 2016,
with White maintaining a pleasant edge after winning back the pawn.
b) 19...Nd8 20.Rde5 Ne6 21.Bc4 Kf6 22.Bxe6! should lead to a classic knight-versus-
bishop position, with the knight having nice squares and targets to jump around and attack.
For instance: 22...Bxe6 23.Rc5 Rfd8 24.Rc6 Rd5 (24...Rd6 25.Rxc7²; 24...Kg7 25.Nc5±)
25.Rxa6 Rd6 26.Ra5²
20.Rc5 Bb7 21.g3

626
21...e6!? 22.Bg2 Ne7 23.Bxb7 Rxb7 24.Rxa5 Nd5 25.Re4²
The knight is excellent on d5 but Black’s pawns are still weak, and the white rooks have some
nice squares available to continue putting pressure on the queenside targets. Thus, White keeps a
slightly preferable endgame.

19.Rd4!
Black seems to solve his problems after:
19.Rd5 Nd7 20.Bxa6 Nb6!
Forcing the exchanges of as many minor pieces as possible.
20...Bxa6? 21.Rxd7 Bb5 22.Nc5± would be ideal for White.
21.Nxb6 Rxb6 22.Bd3
22.Bxc8 Rxc8 23.Rc5 Re6 24.Rxe6† Kxe6 25.c3 bxc3 26.Kc2 Kd6 27.b4 Ra8 28.Kb3 e5
29.Rxc3 e4 30.b5 f5 would be drawish.
22...Ba6 23.Be4 Bb7 24.Rc5

627
24...Re6!
24...Bxe4 25.Rxe4 Rc8 26.Kb2± offers White much better winning chances, as his kingside
structure remains intact and all the rooks remain on the board. To win such a position is
obviously not easy, but Black will continue to suffer for a long time; White will just play
c2-c3 at some point, then start advancing the a-pawn, while keeping both of the black rooks
tied to the defence of the c-pawn.
25.f3 Bxe4 26.Rxe4 Rxe4 27.fxe4 Rc8=
Black should not be in any real danger from here.

19...Bf5
I checked a few other possibilities:

19...Nd7 20.Bxa6! highlights the advantage of having the rook on d4 instead of d5: without
...Nb6 to attack the rook and force the right kind of simplifications, Black doesn’t have much.

628
19...a5 20.Rde4 Nc6 21.Rc4 Bd7 22.Nc5 Be8 23.g3© once again leaves Black struggling with
development. Meanwhile, I doubt that he will be able to hold on to the extra pawn for too long,
so White’s pressure will remain without much risk.

19...Re8 defends the e7-pawn. I suggest: 20.f4 (20.Rde4 Nd7 would justify Black’s last move)
20...Nc6 (20...Nd7? 21.Bxa6±; 20...Ng4 21.h3 Nh6 22.g4©) 21.Rc4 Nd8 22.Rxc7

White regains the pawn, and after something like 22...Ne6 23.Rc6 Kg7 24.g3 he retains some
pressure without much risk.

20.Rd5 Nc6
20...Nd7 21.Bxa6 Nb6 runs into 22.Rc5! Nxa4 23.bxa4 Ra8 24.Bb5± when the doubled a-
pawns are not really weak – unlike Black’s b- and c-pawns.

629
21.Rc5 Nd4 22.Rxc7²
Black has gained a bit of activity for his pieces, but White should be able to keep this under
control while continuing to target Black’s pawn weaknesses.

Conclusion

The Pirc with ...a6 is an interesting choice which offers both sides slightly different challenges
and opportunities compared with the ...c6 lines. After 4.Be3 Bg7 5.Qd2 a6 6.Nf3 we started by
considering lines with 6...0-0 7.Bh6 where Black avoids converting to the main lines by
advancing his b-pawn. 7...Bg4!? seems like the most interesting way of doing so, but 8.0-0-0
Nc6 9.h4!N is an important novelty which gives White excellent attacking chances according to
my analysis.

6...b5 7.Bh6 0-0 is the main line, when 8.Bd3 Nc6! forces us to make an important decision.
9.d5!? is a tempting choice which is well worth considering, as it forces Black to make some
difficult decisions and the best response has yet to be played. I analysed 9...Bxh6!?N and
9...Ne5, both of which give White good chances for an advantage if he knows what he is doing.
However, 9...Nb4!?N 10.h4 followed by the accurate 10...e5! seems adequate for Black. Still, I
imagine 9.d5!? may appeal to many players.

Finally, I presented a different approach, angling for a favourable endgame. The main line
continues 9.Bxg7 Kxg7 10.e5 (10.d5!? and 10.Nd5!? have been noted as more aggressive
possibilities which could be worth exploring) 10...dxe5 11.dxe5 Ng4 12.0-0-0 b4 and now
13.Rhe1! is the key move, leading to a positional sacrifice of the e-pawn after a rather forcing
line. In the ensuing endgame, White enjoys greater mobility for his pieces and numerous weak
squares and pawns in the opponent’s camp, which should add up to more than enough
compensation for the pawn.

630
A) 5...Nd7!? 6.Bd3 340
A1) 6...e6 340
A2) 6...Ngf6 342
A3) 6...c5 343
B) 5...b5 6.Bd3 Nd7 344
B1) 7.e5 345
B2) 7.h4! h6 8.e5 e6! 9.Qd2!N Bb7 10.0-0-0 Ne7 11.Kb1! dxe5 12.dxe5 Nd5 13.Nxd5
Bxd5 14.h5 g5 15.Nxg5! 347
B21) 15...hxg5 350
B22) 15...Nxe5! 16.Ne4 Qe7! 352
B221) 17.Nc5 Ng4 353
B2211) 18.f3 353
B2212) 18.c3 355
B222) 17.Bc5 356

631
1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6
The ...a6 Modern is a popular choice, especially thanks to the influence of Swedish GM Tiger
Hillarp Persson, who wrote about his pet system in Tiger’s Modern (2005) followed by the
heavily updated The Modern Tiger (2014). Like in some of the ...c6 lines, Black is ready to
expand on the queenside with ...b5, but here he retains the option of striking at the centre with
...c5 without loss of tempo.

5.Nf3
5.Qd2 would not be a good fit with the scheme I have in mind. In the event that Black
continues to leave the knight on g8, I have some ideas involving an early attack in the centre,
often involving e4-e5, for which we need the knight on f3.

Black has two main options: A) 5...Nd7!? and B) 5...b5.

5...Nf6 6.Qd2 takes us straight back to the previous chapter.

5...Bg4?! 6.h3 Bxf3 7.Qxf3 e6 is not a good idea for Black.

632
 It is true that trading off the light-squared bishop for the knight on f3 is a common theme in the
Pirc/Modern complex, but at this stage it is premature. Usually, it works better when White has
already played Qd2, in which case he must either accept doubled pawns or spend time on
Be2xf3. Also, the exchange tends to work better for Black when White has already castled short
(or is clearly planning to do so). Here, White can simply proceed with 8.0-0-0 followed by
playing on the kingside unhindered, as seen after 8...Nd7 9.h4 h5 10.Qg3² in Vocaturo –
Cordova, San Cristobal 2012.

A) 5...Nd7!?

This tricky move order may or may not transpose to more common paths.

6.Bd3
This is the right way to maintain compatibility with our repertoire in the event of an early ...b5.

633
We will analyse three main set-ups: A1) 6...e6, A2) 6...Ngf6 and A3) 6...c5!?.

6...b5 converts to variation B.

A1) 6...e6

This version of the ‘Hippo’ seems excessively passive for Black.

7.Qd2 h6 8.0-0-0 Ne7

9.e5!?
9.h4 is a good alternative but the text move is most direct. My editor took the opportunity to
test an earlier draft of this repertoire in a few online correspondence games, and it is worth
including one such game from the present position to show how Black’s set-up can be violently
broken down.

9...d5 10.h4 b6?

This is a surprising error for a correspondence game, although preparing ...c5 would be the
most natural idea for a human player over the board. Objectively Black should take some kind of
precautionary measure on the kingside, but this would be a clear admission that Black’s opening
strategy has not worked.

11.h5! g5 12.Nxg5! c5
12...hxg5 13.Bxg5 Bb7 can be met calmly by 14.Ne2 c5 15.h6 Bf8 16.c3+– when Black is
defenceless on the kingside.

634
13.Nxf7!
White would keep some advantage with 13.Nf3 or even 13.Nce4!? dxe4 14.Nxe4 cxd4
15.Bxd4, but there is no need for such ideas when we can take a sledgehammer to the black
kingside.

13...Kxf7 14.f4
Black is a piece for two pawns up, but he is hopelessly uncoordinated and powerless against
the advance of the kingside pawns.

14...Kg8 15.g4 cxd4 16.Bxd4 Nc6 17.Bf2 Nc5 18.Bg6 Bb7 19.Rhe1 Qe7

20.Nxd5!
There are other ways to get the job done, but this is the most efficient and stylish.

635
20...exd5 21.Bh4 Qd7 22.Bf5 Qc7 23.Qxd5† Kf8 24.Qc4 a5 25.e6
Black resigned in Greet – Valovoj, corr. 2019.

A2) 6...Ngf6

This hasn’t been played much but is worth considering, as it’s no longer a simple matter of
transposing to the previous chapter with the knight’s arrival on f6. We needn’t worry though, as
White has a pretty good version of a Pirc.

7.Qd2
I am happy to play in the spirit of the previous chapter, as Bd3 is always a useful move
whereas the early ...Nbd7 has some drawbacks: Black loses certain options such as ...Nc6 and
...Bg4; and in the event of e4-e5, Black’s knight will not be able to retreat to d7.

7...0-0
7...Ng4 occurred in M. Brunello – Muse, Berlin 2018, when 8.Bf4!N would have been best,
and after 8...e5 9.Bg5 f6 10.Bh4 Black is left with significant weaknesses.

7...e5N has not been played here although it immediately transposes to a few games. 8.dxe5 dxe5
occurred in Nebolsina – Van der Zalm, Vlissingen 2019, and now 9.0-0-0N 9...b5 10.h4! h5
11.Ng5 would have been a good way to develop White’s initiative.

8.0-0-0 c5
Once again, 8...Ng4N is nothing to worry about: 9.Bg5 c5 10.h3 cxd4 11.Nxd4 Ngf6 12.h4ƒ

8...b5 was played in Atakisi – M. Gurevich, Izmir 2002, when White missed a good opportunity

636
to seize the initiative:

9.e5!N 9...Ng4 10.Bg5 dxe5 (10...f6 11.exf6 exf6 12.Bf4±) 11.Be4! Rb8 12.dxe5 Ngxe5
13.Nxe5 Bxe5 14.h4‚ With a virulent attack for a mere pawn.

9.dxc5!? Nxc5
9...dxc5 10.h4ƒ gives White fine prospects on the kingside.

10.Bxc5!
This seems simplest.

10...dxc5 11.e5 Ng4

In the event of 11...Nd7 12.e6! fxe6 13.Bc4 Kh8 14.Bxe6 Qb6 Black isn’t losing a piece
because of the ...Rd8 resource, but 15.Qe3 Ne5 16.Nxe5 Bxe6 17.Rhe1± gives White a pleasant
advantage all the same.

637
12.Qe2±
In Bok – Reilly, Bangkok 2016, Black’s bishop pair was less important than White’s initiative
and threats against the queen.

A3) 6...c5!?

This is rather a critical option. Black aims to convert to a Dragon set-up, and if White just
continues developing normally, his pieces will not be ideally placed for the resulting structure.

7.Bc4!N
Despite being the top engine suggestion, this has never been played. Most players tend not to
analyse such unusual move orders in depth, and moving an already developed piece would seem
counterintuitive during a game. Amusingly, the resulting position actually has arisen a few times

638
with White to move, yet nobody noticed that there was a forced win available with Ng5!.

7...e6
7...Nb6 is refuted by 8.Bxf7†! Kxf7 9.dxc5 Nd7 10.Ng5† Ke8 11.Ne6 Bxc3† (11...Qa5
12.Nxg7† Kf7 13.Ne6!+–) and now a nice touch is:

12.Bd2!+– Winning easily.

7...cxd4 allows White to show another clever point behind the previous move: 8.Ng5! e6 (or
8...Ne5 9.Bxd4 e6 10.f4 Bf6 11.Qe2 Bxg5 12.fxe5ƒ with a massive lead in development)
9.Bxd4 Black has a huge weakness on d6, which makes his position rather unpleasant.

8.dxc5! dxc5

8...Nxc5 is met by 9.Bd4± when once again Black has no compensation for the weakness of
the d6-pawn.

Now we should prevent any further hopes for Black’s activity with:

639
9.a4!
9.Qd6 appears tempting but after 9...b5! 10.Be2 Bb7 11.0-0-0 Ngf6 12.e5 Bf8 13.Qd2 Nd5
14.Ne4 Qc7÷ Black is very much in the game.

9...Ne7 10.Qd6 0-0 11.0-0²

White has a favourable pawn structure and can look to increase the pressure with Rfd1 and
perhaps a knight manoeuvre via d2 to c4.

B) 5...b5

This is the main line by far, and clearly the most consistent choice with Black’s early ...a6.

6.Bd3

640
 We have reached a critical juncture. Black’s two main options are 6...Bb7 (covered in the next
chapter), and 6...Nd7, which lead to similar types of position with potential for transpositions,
and we will treat them similarly too. Originally, my plan had been to meet both of these moves
with 7.e5; this central advance, restricting the knight on g8, is really the crux of how White can
try to exploit Black’s otherwise extremely flexible move order. However, Black seems to have
enough resources against both versions of the 7.e5 plan: see variation B1 below for the version
with Black’s knight on d7, and variation A of the next chapter for the version with the bishop on
b7. This led me to search for an improved version of the e4-e5 plan, which will be explained
shortly.

6...Nd7
This move will be our main focus for the rest of this chapter.

A risky, provocative alternative is:

6...Nf6 7.e5 Ng4
7...Nfd7?! pushes the provocation too far in view of: 8.e6!N 8...fxe6 9.Ng5 Nf6 10.h4‚
8.Bf4
If White has time for h2-h3 he will be doing great, so Black should try to force the play
somehow.
8...b4
8...Bb7 is met by 9.h3± of course.
8...Nc6 9.Qe2 b4 10.Be4 Bd7 11.Nd5± is also excellent for White.
8...c5 is the only other serious try, but 9.Be4 Ra7 10.dxc5 dxe5 11.Qxd8† Kxd8 12.Bg3±
gives White an obvious advantage in activity and coordination.

9.Na4!N
9.Ne4 is not a bad move but after 9...Nc6 (9...dxe5N is a valid alternative) it was not so
simple for White to maintain the e5-pawn in Christensen – Skytte, Denmark 2013.

641
9...Bd7
9...dxe5 10.dxe5 Nc6? 11.Be4+– shows the advantage of leaving the e4-square vacant.
9...Nc6 10.h3 dxe5 11.Bg3!± also works out well for White.
10.c3
10.b3!? is an interesting alternative, ensuring that the queen will not be tied to the defence
of the knight. Black could damage our pawn structure by exchanging on a4, but White’s
initiative and strong light-squared bishop would seem more important.
10...0-0 11.Be4ƒ
White has an active game and the “knight on the rim” is not so bad – especially if the pawns
are exchanged on e5, in which case the knight could come to the dominating c5-square.

Returning to the main line of this chapter, we will start by analysing my initial attempt with B1)
7.e5, before moving on to my recommendation of B2) 7.h4!.

The most popular continuation has been 7.a4 b4 8.Ne2 but this has been extensively tested and
analysed, and I think Black is basically okay.

B1) 7.e5

7...e6!
I was unable to find anything special for White against this solid choice. Here is a quick
summary of Black’s alternatives:

7...Bb7 is the most popular move as well as Hillarp Persson’s recommendation; we will consider
this under the 6...Bb7 move order, in variation A1 of the next chapter.

7...dxe5? 8.Be4 Rb8 9.dxe5 is unpleasant for Black.

642
7...Nb6!?
This is a rare but interesting way to avoid the e5-e6 sacrifice. Still, White can cause some
problems with:
8.Qd2!N
Preventing ...Nh6.
After 8.0-0 Nh6!? the knight is surprisingly well placed, and 9.Re1 0-0 10.Qe2 Bb7
11.Rad1 Qc8 12.h3 Nf5 was fine for Black in Blomqvist – Hillarp Persson, Gothenburg
2015.
8...Bb7 9.0-0-0 e6 10.h4 h6
10...Ne7 11.h5ƒ
10...b4 11.Ne4±
10...h5 is obviously unpleasant for Black because of the weak g5-square: 11.Bg5±

11.Ne4 Nd5 12.h5!

Intending a sacrificial breakthrough, not unlike that which we saw in variation A1.
12...g5 13.Nfxg5 hxg5
In the event of 13...Nxe3 14.Nxf7! Kxf7 15.fxe3 Ke8 16.Rhf1ƒ Black will do well to
survive the middlegame.
14.Bxg5 Bh6
The alternative is 14...f6 15.Nxf6† Bxf6 (15...Ngxf6 16.h6! Bxh6 17.exf6+–) 16.exf6
Ngxf6 17.f4© when White’s powerful bishops and the open black king ensure that Black’s
life will be hard.
15.f4! Bxg5 16.fxg5

643
 Black is stuck for a good move.
16...Qe7 17.Nf6† Ngxf6 18.exf6ƒ
White has more than enough play for the sacrificed piece, and will keep a strong initiative even
if Black opts for a counter-sacrifice on f6.

8.Bf4!?N
8.Ne4 occurred in So – Carlsen, Internet (bullet) 2017, when 8...dxe5N would have led to
messy complications.

The text move is a valid idea to try and provoke ...d5, after which White can maintain his centre
with Ne2 and c2-c3, thereby giving him a nice version of a French structure. However, I am not
so keen on the idea of moving my bishop again so soon, which makes me sceptical about
exploring this too much.

644
8...Bb7!?
8...dxe5 might be okay for Black too, although 9.Nxe5 Nxe5 10.Bxe5! (10.dxe5 Bb7=) still
poses some problems.

9.exd6 c5„
This could be an interesting direction for further analysis. Nevertheless, my general impression
is that Black has decent play.

B2) 7.h4!

This rare move is my main recommendation. Hillarp Persson does not cover it. Essentially, the
idea is to soften Black’s kingside in order to play a more potent version of the e4-e5 idea from
the previous variation.

7...h6
This is the only move that we need to analyse seriously in this chapter.

7...Bb7 is an important option which transposes to variation B4 of the next chapter.

7...h5?! 8.e5 gives White a clearly improved version of the 7.e5 line, due to the weakness of
Black’s dark squares, especially g5.

7...Ngf6?! can also be ruled out due to: 8.e5 dxe5 9.dxe5 Ng4 10.e6 fxe6 11.Qe2±

7...c5 8.h5 b4 9.Ne2 cxd4 was played in Polzin – Gipslis, Dresden 1998, when 10.Nexd4Nƒ (or
10.h6!?N) would have been promising; Black has several ideas to worry about, including h5-h6,
Ng5 (eyeing e6) and a2-a3.

645
Finally, I also checked:
7...b4
This only causes weaknesses for Black in the short run – and unlike variation B3 of the next
chapter where 6...Bb7 has been played instead of ...Nd7, here Black will not threaten the
e4-pawn with ...Ngf6 next.
8.Ne2 Ngf6
8...h6 seems like the other obvious try to deal with h4-h5, but 9.Qd2 c5 10.dxc5 dxc5 11.0-
0-0 gives White an easy initiative.
9.e5
9.h5!? Nxh5 10.e5 is also worth considering.

9...Nd5
9...Ng4 looks risky in view of 10.e6 fxe6 11.Nf4 with Ng5 and possibly h4-h5 coming up.
9...dxe5 10.dxe5 Nd5 11.e6 fxe6 12.h5 also gives Black problems on the kingside.
10.e6 fxe6 11.h5
In all these lines, Black is still in the game, but he is clearly under pressure on the kingside.

8.e5 e6!

646
 Other moves are legal of course, but it seems to me that Black’s options mirror those we
considered after 7.e5 in the previous variation, and here Black will just be heading for a worse
version due to having weakened his kingside with ...h6. Even without h2-h4 and ...h6 having
been played, I prefer White’s chances, so it seems to me that the text move is the only option
which demands serious analysis.

9.Qd2!N
White has many playable moves but it is important to come up with a purposeful plan. Black
intends ...Bb7 and ...Ne7, perhaps followed by ...Nf5 or ...Nb6-d5, and ...dxe5 could come at
any moment. In any case, Black has some decent squares for his pieces, and later in the game his
fianchettoed bishops could show their full potential. Thus, White needs to act soon. I find the d2-
square a much more natural choice than e2 for the queen, because we are targeting h6. It is
important to appreciate that we should not worry much about ...Bxf3 or losing the e5-pawn, as
those things involve opening files which we can use for attacking purposes. The other advantage
of the queen on d2 is that we are preparing a kingside breakthrough with h4-h5, intending to
meet ...g5 with Nxg5 – once we have completed development of course.

I only found a single game from this position, in which White played:
9.Bf4
Mirroring the earlier variation B1, White moves the bishop for a second time in order to put
pressure on the d6-pawn, but his pieces don’t seem ideally placed for that.
9...Bb7!?N
The game continued with 9...d5?! which completely justified White’s last move, as after
10.Ne2 Black was left with a passive version of a French structure in Starostits – Diaz
Castro, Seville 2018.
9...dxe5 10.Nxe5 Nxe5 (or 10...Bb7 11.Qe2 Ne7 12.0-0-0 0-0) 11.dxe5 Bb7 12.Qe2 Qd4
13.Bg3 is another possible direction that would need to be analysed further.

647
10.exd6 c5
I haven’t analysed this in much detail but I like the way Black’s pieces are developed, and
intuitively the pawn sacrifice feels quite nice for Black. Thus, I do not feel inclined to explore
this Bf4 plan further.

9...Bb7 10.0-0-0 Ne7

10...dxe5 11.Nxe5! Nxe5 12.dxe5 Bxe5 13.f4 Bg7 14.Qf2ƒ gives White lovely open files and
attacking chances for a mere pawn.

11.Kb1!
An important subtlety! I started by analysing the thematic sacrifice with 11.h5 g5 12.Nxg5
hxg5 13.Bxg5 but found that it was not quite working. Therefore we start by playing a typical
improving move which is useful in all scenarios.

11...dxe5
It is hard to suggest anything better. Black has a few natural-looking knight moves available,
but he is almost in a funny kind of zugzwang, making it hard for him to avoid our kingside
breakthrough. Here are some examples to show what I mean:

11...Nb6?! just runs straight into the breakthrough we have been aiming for: 12.h5 g5

648
13.Nxg5! hxg5 14.Bxg5 dxe5 (14...Rf8 15.Ne4±) 15.h6 Bf8 16.Bf6±

11...Nf5 12.h5! gives Black an unpleasant choice: 12...Nxe3 (12...g5 allows the familiar 13.Bxf5
exf5 14.Nxg5 dxe5 [14...hxg5 15.Bxg5 f6 16.h6+–] 15.dxe5 Nxe5 16.Qc1 Qe7 17.Nf3±)
13.Qxe3 g5 Black no longer has to worry about the dark squares, but White can switch to
attacking on the light squares with:

14.exd6 cxd6 15.d5! e5 16.Nd4± Making full use of the queen on e3.

11...Nd5 12.Nxd5 Bxd5 allows us to continue with the same theme: 13.h5 g5

649
14.Nxg5! hxg5?! (Black’s best try is 14...dxe5! when 15.dxe5 transposes to our main line, as
continued on the next page) 15.Bxg5 Qc8 16.f4! c5 (16...Rf8 17.f5‚) 17.h6 Bf8 18.f5 cxd4
19.fxe6 Nxe5 20.Bf6±

12.dxe5 Nd5
This seems the most natural way to continue Black’s development.

12...c5?! is too slow, and 13.Be4± gives White easy play.

12...Nxe5 13.Nxe5 Bxe5 14.f4! Bg7 15.h5 gives White everything he wants, with 15...gxh5
16.Rxh5 b4 17.Na4ƒ yielding more than enough compensation for the pawn.

12...Bxf3 13.gxf3 Nxe5 14.Qe2 also gives White a potent initiative, for instance:

650
14...Nxd3 (14...Nd7 15.h5 g5 16.Be4 c6 17.f4ƒ opens things up further) 15.Rxd3 Qb8 16.Ne4©
White is ready for h4-h5, meeting ...g5 with f3-f4, and Black’s king will not be safe anywhere.

12...Nc6 is an attempt to keep things more solid, but we can still cause plenty of problems with:
13.Qe2 Qe7 (13...Ncxe5 can be dealt with in the usual way: 14.Nxe5 Bxe5 15.Be4 Bxe4
16.Nxe4 Qe7 17.f4 Bg7 18.h5©) 14.Be4

14...0-0-0 (Black also fails to find safety after 14...0-0 15.h5 g5 16.Bxg5! hxg5 17.h6 Bxe5
18.Rxd7 Qxd7 19.Nxe5 Nxe5 20.Bxb7±) 15.a4! b4 16.Bxc6 Bxc6 17.Qc4 Bb7

651
18.Ne4 Nxe5 19.Nxe5 Bxe5 20.Nc5© Once again, White’s initiative is worth more than a pawn.

13.Nxd5 Bxd5 14.h5 g5 15.Nxg5!

Once again, this sacrificial breakthrough is key.

We will now analyse B21) 15...hxg5 and B22) 15...Nxe5!.

B21) 15...hxg5 16.Bxg5 Qc8

Grabbing the knight seems excessively risky for Black.

17.h6

652
17...Bxe5
17...Bf8 18.Qf4 Qb7 19.f3 c6 20.Bf6 Nxf6 21.Qxf6 Rxh6 22.Rxh6 Bxh6 23.Qxh6± simply
leaves White a pawn up.

18.Rhe1 Bf6
18...c5 19.Bf5!? sets the full power of White’s army on Black’s fragile centre. Play may
continue: 19...Qc6 (19...Bc6 20.f4 Bf6 21.Bxe6+–; 19...Qc7 20.f4 Bd4 21.c3±) 20.f4 exf5
21.fxe5 Nb6 22.Bf6 Kd7 White has a few good continuations but the most accurate is:

23.a4! bxa4 (23...Rh7 24.a5+–) 24.Qf4 c4 25.g4!+– Black’s position is collapsing, as 25...fxg4
allows 26.Bxh8 followed by Qxf7†.

19.Bxf6 Nxf6
For the time being White only has one pawn for the sacrificed piece, but the h-pawn is

653
dangerous and Black is horribly uncoordinated, so it is only a matter of time before White either
wins back the piece or hurts Black in some other way.

20.Qf4 Qd8 21.g4 Ng8 22.Be4

22.h7 enables Black to escape to an endgame which may well be tenable: 22...Qd6!?
23.hxg8=Q† (23.Qg5 Ne7 24.Qg7 0-0-0 25.Qxf7 Nc6÷) 23...Rxg8 24.Qxd6 cxd6 25.Bh7 Rg7
26.Rxd5 Rxh7 27.Rxd6 Ke7 Black has decent chances to hold the double rook endgame a pawn
down, although White can certainly press.

22...c6
22...Ne7 23.Bxd5 Nxd5 24.Qe5+– wins back the piece while keeping a huge initiative.

23.g5 Ne7 24.b3 a5 25.Qe5 Kd7 26.c4±

Winning back the piece. The game goes on, but White’s far-advanced pawns make him the
clear favourite.

654
 B22) 15...Nxe5!

This seems like Black’s best bet. He is still a long way from safety, but at least the game is not
all one-way traffic like in the variation above.

16.Ne4
With the knight sac having been declined, the position has stabilized to a certain degree.
Black’s king will have trouble finding a safe haven on either flank, and in general his position
seems replete with weaknesses, the h6-pawn and c5-square being two examples. At the same
time, Black’s pieces are extremely active, so his defensive and counterattacking resources should
not be underestimated.

16...Qe7!
16...Qd7 seems less accurate due to 17.Nc3 0-0-0 18.Nxd5 Qxd5 19.Qc1!² when the weak h6-
pawn keeps the h8-rook tied up, so Black is unable to fight for the central files the way he would
like to.

16...Nxd3 can be met by 17.cxd3!?, and if 17...Bxe4 18.dxe4 Qxd2 19.Rxd2 White has a clearly
favourable endgame because of the weak pawns on c7 and h6.

We have reached another critical position where White has a number of tempting possibilities. I
have analysed B221) 17.Nc5 and B222) 17.Bc5 in some detail.

There are a couple of other options which I looked at but found less attractive. Here are some
brief details:
17.Nc3 Bb7!

655
 17...0-0-0 gives White some pressure after: 18.Nxd5 Rxd5 19.Qa5 Qd6 20.Be2 Nc4
(20...Rd8 21.Rxd5 exd5 22.Rd1²) 21.Rxd5 exd5 22.Bxc4 dxc4 (22...bxc4 23.Rd1 Rd8
24.g4²) 23.a4 Qf6 24.Bc1 Qc6 25.Re1²
18.Be2 Nd7
18...Rd8 19.Qc1²

19.Bf4
19.Bf3 Bxf3 20.gxf3 0-0-0 also seems okay for Black.
19...0-0-0 20.Qe3 e5 21.Rhe1 Qe6 22.Bg3 Nb6÷

I also checked:
17.Rhe1
This forces Black to navigate some pitfalls – but if he does so successfully, White has
nothing special.
17...Nxd3
17...0-0-0?! 18.Qa5 Bb7 19.Nc5 Nc6 20.Qd2± is unpleasant for Black.
18.cxd3 f5!
18...0-0-0? 19.Nc5+– and the queenside threats are too strong.
19.Nc3
19.Bc5 Qf7 20.Nc3 0-0-0÷

656
19...Bb7!?
Black seems to be doing fine. (19...0-0-0 20.Nxd5 Rxd5 21.Rc1 Qd6 22.Rc2! Rxd3 23.Qa5
Rd1† 24.Rxd1 Qxd1† 25.Rc1 Qd3† 26.Ka1 might give White some pressure though.)

B221) 17.Nc5

Even though I was ultimately unable to prove anything for White after the text move, it
certainly warrants consideration as it poses Black certain problems. The main idea is to impede
...0-0-0.

17...Ng4
Going after the dark-squared bishop is Black’s best idea, as the g7-bishop will gain in value.

17...Qf6 18.Qc1 Nxd3 19.Rxd3 0-0-0 20.f3² seems pleasant for White.

657
17...Nxd3 18.cxd3ƒ leaves the knight nicely placed on c5. Black will be forced to castle on the
kingside sooner or later, after which Rhg1 and g4-g5 will be annoying for him.

Here I analysed two interesting options: B2211) 18.f3 and B2212) 18.c3.

In the event of 18.Be4 Rd8 19.Qe2 Nxe3 20.Qxe3 Bxe4 21.Nxe4 0-0 22.g4 Rxd1† 23.Rxd1 f5÷
there are too few minor pieces for White to be able to have any meaningful initiative.

B2211) 18.f3 Nxe3 19.Qxe3

Black faces the usual difficulty regarding what to do with his king.

19...Qf6
19...0-0 20.Rhg1! is both blunt and effective: 20...Qf6 21.c3 Rfd8 22.g4 Qxf3 23.Qc1!ƒ With a
nasty threat of g4-g5.

20.c3
20.Qc1 Qe7!? seems okay for Black. If White does not wish to repeat moves he can try
21.Ne4 but then 21...0-0-0÷ leaves White less than ideally placed to attack on the queenside.

20...b4 21.Bc2!
21.Ne4 Qe7 22.c4 Bxe4 23.Qxe4 Rd8= seems fine for Black.

21...c6
Black must avoid 21...bxc3?? 22.Rxd5 of course.

658
22.Ba4
22.Ne4 Qe7 23.c4 again proves harmless after: 23...Bxe4 24.Bxe4 0-0 25.g4 Rad8=

22.Rxd5!? cxd5 23.Nd7! is a cool idea but Black should be fine after: 23...Qe7 24.Ba4

24...0-0-0 (24...Qd6!? 25.Rh4 bxc3 26.Rg4 Rg8÷ also keeps things balanced, but in a more
complicated way) 25.Qa7 White has nothing better than forcing a perpetual: 25...Rxd7 26.Qa8†
Kc7=

22...0-0 23.Nd7 Qe7!

23...Qf5† is riskier in view of 24.Bc2 Bxa2† 25.Kc1 Qa5 26.Qe4 f5 27.Qxb4 Qxb4 28.cxb4
Rfe8 29.b3 when Black’s bishop is stuck on a2, and White can just go about the rest of his life as
usual.

659
24.Nxf8 bxc3 25.Rxd5
I also considered 25.Ng6 Qb4 26.Bb3 Bxb3 27.axb3 Qxb3 28.Qe2 fxg6 29.Qc2 Qb5 30.Rd7
gxh5 31.Rhd1 c5= when Black has too many pawns, so I doubt that White can get more than a
perpetual somewhere around here.

25...Qb7 26.Bb3 exd5 27.Ne6!

A good try, but Black is okay after the mostly forced continuation:

27...Qe7 28.Re1 Re8 29.Nxg7 Qxe3 30.Rxe3 Rxe3 31.Nf5 Re1† 32.Kc2 cxb2 33.Kxb2

33...Re2† 34.Kc3 Rxg2 35.Nxh6† Kf8

I would take White’s side of this endgame, but Black should not be in any real danger.

B2212) 18.c3

660
 After seeing the previous variation, the idea of the text move should be clear. The general
character of the play is similar, but here we avoid Black’s concrete counterplay with ...Qf6 and
...b4. The trade-off is that Black gets a bit more freedom to choose between different options –
although in some ways this makes life harder, especially in an unfamiliar position. For instance,
it is not obvious where Black should put his king. He also needs to decide whether to eliminate
the e3-bishop while he has the chance, or allow it to move to d4.

18...0-0-0!?
This leads to richly double-edged play.

18...Nxe3 19.Qxe3 0-0 is met by 20.Rhg1!ƒ with the straightforward plan of g4-g5.

Black can also consider:

18...0-0 19.Bd4
19.Rhg1 b4! 20.cxb4 a5÷ offers Black decent counterplay.
19.f3 Nxe3 20.Qxe3 is met by 20...b4! 21.c4 Bc6 when ...Qf6 will be annoying, and
White’s g4-g5 plan does not quite come in time.
19...e5 20.Rhe1
20.Qe2 Rfe8 21.Qxg4 exd4 22.cxd4 Rad8© gives Black decent play for the pawn.

661
20...Qh4 21.Be3 Nxe3 22.Qxe3 Qxh5 23.g4
23.Rh1 Qg4 24.f3 Qf4 25.Qf2 Rfd8 seems satisfactory for Black.
23...Qg5 24.Qxg5 hxg5 25.Be4 Bxe4† 26.Nxe4©
White has a bit of pressure, although it might just fizzle out after further exchanges.

19.Bd4
19.Nxa6 Ne5! gives Black a decent initiative for the pawn.

In the event of 19.f3 Nxe3 20.Qxe3 Rd6÷ Black’s bishop pair compensates for the weak
structure and things could go either way, which is not the kind of outcome I am aiming for.

19...Ne5
19...Bxd4?! 20.cxd4 gives White excellent chances on the queenside.

662
20.Qe3
20.Be2 Bxg2 doesn’t quite work for White; and 20.f3 Nxd3 21.Qxd3 Rhg8 22.Nxa6 e5 is
messy.

20.a4!? deserves to be analysed further. The position remains complex, with many possibilities
for both sides on every move.

20...Nxd3
20...Qd6 allows 21.Be2² when, with White’s bishops saved, we have got what we were hoping
for.

21.Rxd3 Bxd4
21...e5?! 22.f4± leaves the e5-pawn pinned in two directions.

21...f6 22.Nxa6 is complicated, but the extra pawn should come in handy for White.

22.cxd4 Qg5

23.f4!?
23.Qxg5 hxg5 24.f3 reaches an interesting endgame but Black seems solid enough.

23...Qf5 24.Ka1
The position remains complex and double-edged; I would tend to favour White slightly
because of his safer king, but Black seems to have enough resources. As with all these lines,
further analysis is recommended. (24.g4!? Qxg4 25.Rc1 Rd6 26.Ra3 Bb7 is yet another possible
direction.)

663
 B222) 17.Bc5

Finally, we consider a natural move which leads to a relatively forcing line for the next few
moves.

17...Qd7 18.Qf4 0-0-0 19.Nf6 Bxf6 20.Qxf6 Bxa2†

21.Kc1!
21.Kxa2 Qd5† 22.Kb1 Qxc5 would be fine for Black.

21...Nxd3† 22.Rxd3 Bd5

White clearly has compensation for the pawn. My initial impression was that Black must be
okay in the long run; but as I delved deeper, I realized his task is not easy. The problem is not
just that Black’s king is slightly exposed; the h6-pawn is also in trouble.

664
23.Rhd1
Threatening Rxd5.

23...Kb7
23...Qc6? does not help because of 24.Rxd5! when the loose rook on h8 makes all the
difference.

24.Be3
The h6-pawn cannot be defended; and after regaining the pawn, White should have a better
game without much risk.

24...Rde8
24...b4 is a natural move to consider but Black’s queenside play does not amount to much
after: 25.Bxh6 Qc6 (25...Qa4? does not threaten anything, and 26.Bg7 is winning for White)
26.Rd4 a5

27.Bg7! Bb3 (27...Rhe8 28.R1d3! prepares to deal with Black’s queenside threats, leaving the h-
pawn unstoppable) 28.Qf3! Qxf3 29.gxf3 Rxd4 30.Rxd4 Rh7 31.h6 gives White excellent
winning chances due to the awful position of Black’s rook.

25.Bxh6 Qd8
25...Qe7 is possible although it is not clear if the endgame makes Black’s life easier: 26.Qxe7
Rxe7 27.Bg5 Ree8 28.Rh1 Bxg2 29.Rh4 Rh7 30.Bf6 Bd5 31.b3²

26.Qf4 f5

665
27.Bg5 Qd6 28.Qh4²
Material is level and Black’s pieces are tied up due to the h-pawn. White may look to exploit
this by switching the focus to an attack on Black’s king; and endgames will not be easy for Black
either.

Conclusion

4...a6 remains a popular choice in the Modern Defence. Having developed the bishop to e3, we
refrain from Qd2 for the time being, in favour of the flexible 5.Nf3. From there, we started by
analysing the tricky 5...Nd7!?, when 6.Bd3 is our choice. It is well worth knowing that 6...c5
should be met by the counterintuitive 7.Bc4!, when the threat of Ng5 should force Black to
compromise his position in one way or another.

5...b5 6.Bd3 Nd7 is the main line of the chapter, at which point I included my analysis of 7.e5 as
a starting point, while noting that 7...e6! seems satisfactory for Black. Once this is understood,
the rationale for recommending 7.h4! h6 8.e5 becomes clear: White is once again playing for a
central initiative while restricting the movement of the knight on g8 – but with the pawns on h4
and h6, Black’s kingside has been softened up. Once again, 8...e6! seems like Black’s best
reaction, when 9.Qd2!N 9...Bb7 10.0-0-0 Ne7 11.Kb1! is the most accurate continuation,
playing a high-quality waiting move. Black would like to improve one of the knights on d7 or e7,
but virtually any such knight move invites h4-h5, followed by a powerful knight sac on g5. The
critical line seems to be 11...dxe5 12.dxe5 Nd5 13.Nxd5 Bxd5 14.h5 g5 15.Nxg5!, at which
point 15...hxg5 seems a bit too risky, but 15...Nxe5! 16.Ne4 Qe7! is more challenging. In that
case I found 17.Nc5 a fascinating move to analyse, and there are plenty of ways in which Black
could land in trouble here; but ultimately, 17.Bc5 seems like the more practical choice, leading
after a few forcing moves to a middlegame with opposite-coloured bishops, where White enjoys
a lasting edge without much risk.

666
667
A) 7.e5 360
A1) 7...Nd7 360
A2) 7...c5! 363
B) 7.h4! 365
B1) 7...h6 366
B2) 7...Nf6 8.e5 368
B21) 8...dxe5 369
B22) 8...Ng4!?N 372
B3) 7...b4!?N 8.Ne2 Nf6! 9.Ng5!? 377
B31) 9...h6 378
B32) 9...e6 380
B4) 7...Nd7 8.h5 382
B41) 8...c5 382
B42) 8...Ngf6! 9.h6 Bf8 10.Bg5!N 384
B421) 10...c5 385
B422) 10...b4 387

668
1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.Nf3 b5 6.Bd3 Bb7
Just as in the related 6...Nd7 variation, I will start by presenting some analysis of the blunt A)
7.e5, which was my original idea, before moving on to my main recommendation of B) 7.h4!N.

A) 7.e5

I made some interesting discoveries when analysing this move. I actually rather like White’s
chances after the most common A1) 7...Nd7, but the extremely rare A2) 7...c5! is stronger –
although there are still opportunities to create problems for Black here as well.

In variation B1 of the previous chapter, 7...e6 was mentioned as a strong move in the analogous
position with ...Nd7 having been played instead of ...Bb7. Here it does not work so well, as
8.Ne4 (or 8.a4!? b4 9.Ne4) seems quite unpleasant for Black, due to the possibility of Bg5. With
the knight on d7, Black could have countered this idea by capturing the e5-pawn.

7...Nc6N is worth a quick mention. 8.Qe2! is an excellent response, and if 8...dxe5 9.dxe5 Nxe5
10.Nxe5 Bxe5 11.0-0-0 Qc8 12.Bxb5† axb5 13.Qxb5†± Black’s position is a mess.

A1) 7...Nd7

669
 This is by far Black’s most popular response, especially if we count transpositions via the
6...Nd7 7.e5 Bb7 move order – which is Hillarp Persson’s recommendation in The Modern
Tiger.

8.e6
This has been a unanimous choice so far, but that does not mean it is the only reasonable move.
I can’t resist sharing another more subtle pawn sacrifice:
8.Qe2!?N
This could pack a certain punch as well, especially considering its surprise value. Here are
some lines:
8...dxe5
8...e6 9.Ne4 f5 10.exf6 Ngxf6 11.Bg5² leads to a nice position at no material cost.
8...Bxf3 9.Qxf3 dxe5 10.Be4 Rc8 11.0-0-0ƒ gives White a decent initiative for a pawn.
9.Nxe5

670
9...Bxg2
Alternatively, there is 9...Ngf6 10.0-0-0 and now:
a) 10...b4 11.Na4 0-0 12.h4ƒ
b) 10...e6 11.h4 Qe7 12.h5ƒ
c) 10...0-0 11.h4ƒ
d) 10...c5 11.Nxd7 Nxd7 12.Be4! Bxe4 13.Nxe4 c4 14.h4 Nf6 15.Bg5²
10.Rg1 Bb7 11.0-0-0 Ngf6 12.h4
Grabbing the g2-pawn was obviously dangerous for Black, but it requires some accuracy to
make the most of White’s chances.
12...0-0 13.h5! Nxh5

14.Qxh5!! gxh5
14...Nxe5 15.dxe5 gxh5 16.Bh6 f5 17.Bxg7 Qe8 18.Rg5! Rf7 19.Rdg1 wins.
15.Bh6 Nxe5 16.Rxg7† Kh8 17.Rxh7† Kg8 18.dxe5 f5

671
19.Bxf8! Qxf8 20.Rxh5 Qg7 21.Bxf5 Kf8 22.f4 Bc6 23.Ne4+–

8...fxe6
We have reached a typical-looking position after White’s thematic pawn sacrifice. The
compensation is real, although Black’s position is tough to break down and it also contains a fair
share of latent dynamic potential. Many strong GMs have played this way with Black, and Tiger
also recommends it in his book. I found a promising way forward for White though.

9.Ng5 Nf8 10.0-0 Nf6

10...h6?! 11.Bxg6†! Nxg6 12.Nxe6 Qd7 13.Nxg7† Kf7 14.Nh5 saw White regain the pawn
with interest in Svidler – McNab, Gothenburg 2005.

10...Nh6 11.a4 b4 12.Ne2 (12.Na2!? is another interesting direction) 12...Nf5 13.c3© has been
played a few times; the black king will continue to be a problem.

672
11.f4!?
11.Re1 has been played in the great majority of games, and 11.a4!? is an interesting alternative
which Tiger analyses in his book.

The text move is a rare choice, which I find puzzling, as Black’s position looks really
uncomfortable to me. I guess most players have wanted the rook on the e-file with a view to
winning back the e6-pawn at some point, but I really like the idea of clamping down on the
centre and establishing a significant space advantage. Even without any direct attack, Black has a
problem with the long-term safety of his king. If Black just sits passively, White will continue to
dominate and the knight on g5 will be especially annoying. On the other hand, there is no easy
way to generate counterplay for Black; he can chase the knight away with ...h6, but this will only
weaken the g6-point and Black’s king.

11...Qc8!?N
11...Qd7 has been played a few times, but after 12.Qe2 h6 (12...Rc8 13.b4©) 13.Nf3© White
kept the more pleasant game in Cabarkapa – Dragojlovic, Novi Sad 2018, for reasons explained
in the previous paragraph.

11...h6 12.Nf3© does not change much.

11...b4 was played in Fargere – Dussol, Le Port Marly 2009, when White moved the attacked
knight to e4. Although the game continuation was interesting, I would prefer to avoid
unnecessary simplifications with 12.Ne2N, and after 12...Qd7 13.c3! we exploit Black’s 11th
move by opening up the queenside. A few lines are:

673
13...Qc6 (13...Nd5 14.Bf2©; 13...bxc3 14.Nxc3©) 14.Nf3 bxc3 15.Rc1 Nd5 16.Bd2 Nd7
17.Nxc3 Nxc3 18.bxc3© The black king still lacks a safe shelter on either side.

12.Qe2 b4
I also considered 12...c5!? 13.dxc5 dxc5 14.b3 N6d7 15.Bd2 b4 16.Na4 h6 17.Nf3 Bxa1
18.Rxa1© when White has plenty of compensation for the exchange, especially in a practical
game.

13.Nd1 Nd5 14.c3!? bxc3 15.bxc3 c5

15...Nxe3 16.Nxe3 c5 would be a worse version for Black due to: 17.Nc4! Qc7

674
18.f5!! exf5 19.Rab1 There are too many open lines and tactical tricks for Black’s position to
withstand. For instance: 19...Bc6 20.Bxf5! gxf5 21.Qh5† Kd7 22.Nb6†+–

16.dxc5 Nxe3 17.Nxe3 Qxc5 18.Kh1©

The position is opening up, so king safety is becoming more of a concern for Black.

A2) 7...c5!

675
 Surprisingly, Black seems to be doing okay after this risky-looking move – although a lot of
accurate moves are needed to justify this assessment.

8.e6!
Neither 8.dxc5 dxe5÷ nor 8.exd6 cxd4 9.Nxd4 Qxd6= should worry Black.

8...f5!

This idea was suggested to me by GM Daniel Fernandez, and later it was tested in a game
which will be referenced shortly. It may seem strange to leave the pawn untouched on e6, but the
advantages are many: Black keeps the position relatively closed; he is poised to continue
developing smoothly; and the e6-pawn will require protection by White.

I am more optimistic about White’s chances after the alternative:

8...fxe6 9.dxc5 Nf6N
This seems like a natural improvement over 9...Bxc3†?! 10.bxc3 Nf6 11.0-0 (or 11.a4!?Nƒ)
when the loss of Black’s dark-squared bishop was more relevant than the doubled c-pawns
in Perelshteyn – Piasetski, Reykjavik 2017.
9...Qd7N 10.Ng5 Nf6 is a simple transposition; and 9...Nc6 10.Ng5ƒ seems promising for
White.
10.Ng5 Qd7

676
11.h4! 0-0
11...Nc6 12.h5 gxh5 13.Qe2! Ne5 (13...0-0-0 14.Nf7±) 14.0-0-0ƒ
12.h5! h6 13.Nf3
13.Bxg6 is not so clear after: 13...b4! (rather than 13...hxg5? 14.h6 Bh8 when Black is
busted by: 15.Qd3! Bxg2 16.Bh7† Nxh7 17.Qg6† Bg7 18.Qxg7#) 14.Ne2 Nc6 15.Ne4
Ng4÷
13...b4 14.Ne2 g5

15.Bxg5! Bxf3
15...hxg5 16.h6 Bh8 17.Nxg5+–
16.gxf3 hxg5 17.h6 Bh8 18.h7† Kf7 19.Qd2 Nc6 20.Qxg5 Ke8 21.0-0-0ƒ
This analysis could be expanded of course, but in general I like White’s chances.

9.dxc5 dxc5

677
 9...Nf6? would lead to interesting play if White took on d6, but 10.Be2!± essentially refutes
Black’s idea.

10.Be2!N
10.Bxc5 Qc7 11.Be3 Nf6 gave Black more than enough play for the pawn in Kosteniuk –
Lagno, Kazan 2019.

The text move is a natural improvement. I still think Black’s position is basically okay – but that
does not necessarily mean we should close off this avenue entirely. There are dangers for both
sides, and Black in particular has to memorize some ridiculously tough computer moves to
maintain the balance. I say “memorize” because it would be practically impossible for a human
player to navigate the following complications accurately over the board.

10...Qa5 11.h4 Nf6

11...Bxc3†? 12.bxc3 Qxc3† 13.Bd2+– is no good for Black.

12.h5!

678
12...h6!!
This is one of those moves that the human mind can scarcely comprehend.

12...Nxh5? is met by the brutal 13.Rxh5 gxh5 14.Ng5+– when White’s attack is too strong.

12...b4 leads to interesting complications after: 13.h6! 0-0! (13...Bf8 14.Nb1+–) 14.hxg7 Rd8
15.Qc1 bxc3 16.b3 Nc6 17.Ng5

17...Kxg7! (17...c4? is refuted by 18.Nxh7! Nxh7 19.Bd4!!+–; but not 19.Bb6? when 19...Qe5!
defends) 18.Nf7 Rd4 19.Bc4 Rxc4 20.bxc4²

13.Nh4!?
The basic idea of Black’s last move is that 13.hxg6 can be met by 13...Nc6ƒ when White is a
pawn up but has no immediate way through on the kingside, and Black’s pieces are extremely

679
active.
13.Ne5!? b4 14.Nc4 Qc7 15.Na4 0-0 16.hxg6 Nc6 17.Bxh6 Nd4 18.Qd2 f4 19.Bxf4 Qc6÷ is
another insanely complicated line where the engine somehow demonstrates equality everywhere,
but in a human game anything could happen.

13...b4 14.Nb1 f4 15.Bxf4 g5 16.Nf5 gxf4 17.Nxg7† Kf8 18.Nf5 Nc6„

Again we have a weird situation where Black’s kingside has been partially dismantled, but his
active pieces produce some kind of dynamic balance.

19.Qc1 b3† 20.Kf1

20.Nc3?! Nd4³

20...Rg8÷
The position remains roughly equal among engines. For humans, the whole line seems like one
big minefield for both sides. Overall, it did not quite tick all of the boxes for my recommended
repertoire – but if you enjoy such adventures, these lines could be well worth considering, and
there are many possible avenues for independent analysis.

B) 7.h4!

680
 After realizing that 7.e5 was not quite working the way I would have liked, I tried to come up
with other ideas that may still exploit Black’s move order – in particular the fact that the prospect
of e4-e5 is still on the table. Of course, the situation in the present position goes hand in hand
with variation B2 of the previous chapter, in which Black also had a decent way of meeting the
immediate 7.e5; but we saw that 7.h4! posed more problems, while keeping the e4-e5 idea in
reserve. My recommendation in the present variation is conceptually the same, and it also
provides a convenient way of dealing with transpositions, as Black could play ...Nd7 and ...Bb7
on moves 6 and 7 in either order.

The h2-h4 advance is a common occurrence in many Pirc/Modern lines, but often it doesn’t
change much because the threat of h4-h5 can be easily parried by any of ...h6, ...h5 or ...Nf6.
However, in this instance we are primed for the e4-e5 advance as well, so if Black reacts in any
of the three ways just mentioned, he will have to reckon on 8.e5 again. We saw in the 7.e5 lines
that Black was barely able to reach a satisfactory position with precise play; so if he commits to a
move such as ...h6 or ...h5, creating additional weaknesses on the kingside, then White will enjoy
a clearly improved version of the e4-e5 plan. A more challenging idea is to bring the knight to f6,
either immediately or after a preliminary ...b4. That way, Black stops h4-h5 for the moment
while developing a piece – but White will chase the knight with e4-e5, after which e5-e6
sacrifices are in the air. Finally, Black could simply allow h4-h5, focusing on development with
...Nd7 and ...Ngf6, while keeping the ...c5 break in reserve. Still, having the pawn on h5 creates
kingside tension which favours White, and a timely h5-h6 could prove annoying for Black.

Having sketched out the main plans and ideas, we will analyse four main options: B1) 7...h6, B2)
7...Nf6, B3) 7...b4!?N and B4) 7...Nd7.

7...h5?! 8.e5 just gives Black a worse version of variation B1, as he has not only loosened the g6-
pawn but also left a gaping hole on g5 without gaining anything in return.

681
Black might also just ignore the kingside with 7...c5, but then 8.h5 cxd4 9.Nxd4 Nf6 10.h6 gives
White an amazing version of an Open Sicilian.

B1) 7...h6

I found one game with this move, Neumann – Taufratshofer, Bad Woerishofen 2016, when
White responded with 8.Ne2 – but you already know that I have a different concept in mind.

8.e5!N 8...e6
This seems like the only sensible move.

If Black tries to follow the recipe from variation A2 with 8...c5?, the placement of the pawns on
h4 and h6 makes a huge difference after 9.e6!, when Black will not survive for long.

9.Ne4!
9.Be4 d5 10.Bd3 c5! sees Black obtain counterplay before White can carry out the desired
regrouping with Ne2.

9.Qd2!? is an interesting try, angling for a transposition to variation B2 of the previous chapter,
which could happen at any moment if and when ...Nd7 is played. However, Black seems to be
able to avoid this and reach a safe position with: 9...Ne7 10.0-0-0

682
10...Nd5! 11.Nxd5 (11.Kb1 Nxe3 12.Qxe3 Qe7=) 11...Bxd5 12.Kb1 Nc6! Even at this late
stage, 12...Nd7 would have transposed to the previous chapter – but after the text move, the
sacrifices on g5 are no longer as strong, and Black seems to have an okay position otherwise.

9...Nc6
Sacrificing a pawn with ...Nd7 does not make much sense here, so Black needs to find another
way to develop.

9...dxe5 10.Nxe5 leaves Black with a big hole on c5, and the f7- and g6-pawns look weak too;
meanwhile, it is hard to get rid of the e5-knight.

The other main candidate is:

9...Ne7 10.Bf4 Nf5
10...dxe5 11.Bxe5 causes a lot of problems on the dark squares.
10...Bxe4 11.Bxe4 d5 12.Bd3 c5 13.c3 gives White a superb version of a French. Black has
no real hope of challenging the centre and he already has weaknesses on the kingside. It
would be a rare gift for White to obtain the c3-d4-e5 pawn chain, the bishop pair and
excellent piece coordination without making any concession in return.

683
11.g4
Not the only decent move, but it seems the simplest.
11...Nxd4 12.Nxd4 dxe5 13.Nxe6 fxe6 14.Be3©
Black’s extra pawn is of little value, his dark-squared is dead for the foreseeable future, his
king is unsafe and his pawn structure is riddled with weaknesses.

10.c3!
10.Qd2 dxe5 11.Nc5 does not quite work because of 11...exd4!.

Instead we strengthen the centre while preventing ...Nb4 and preparing to meet ...dxe5 with Nc5.

10...Nge7
10...dxe5 11.Nc5! Qc8 12.Nxe5 Nxe5 13.dxe5 Bxe5 gives White a tremendous initiative for
the pawn after:

684
14.f4 Bg7 15.h5±

After the text move, we can borrow an idea from the previous chapter with:

11.h5 g5 12.Nfxg5! dxe5

12...hxg5 gives White a huge initiative after 13.h6!?, or simply 13.Bxg5.

13.Nxe6! fxe6 14.Qg4 Rg8 15.Nc5 Bf8

15...Bc8 16.Bh7+– and the defence collapses.

16.Qf3 Qc8 17.Be4 exd4 18.cxd4©

Black looks horribly tied up. A sample line is:

685
18...Rb8 19.Nxb7 Qxb7 20.Rc1 Nd5 21.Bxd5 exd5 22.0-0 Ne7 23.Bf4 Qc8!?
This seems like the only chance, as after 23...Rc8 24.Rfe1 c6 25.Re6+– the attack is too strong.

24.Rxc7 Qf5 25.Re1 Rd8 26.g3±

White keeps some initiative for the sacrificed piece and is likely to pick up a third extra pawn
soon.

B2) 7...Nf6

This seems provocative, as Black just invites the e5-e6 plan. At the same time, it prevents h4-
h5 for the time being while developing another piece; besides, devotees of the ...a6 Modern are
not known for backing down from a confrontation. Thus, the text move is undoubtedly an
important option to analyse further. If White cannot pose concrete problems quickly, then things
might just turn out fine for Black.

686
8.e5
We will analyse B21) 8...dxe5 and B22) 8...Ng4!?N.

B21) 8...dxe5 9.dxe5 Nd5N

This seems like an improvement on a correspondence game in which White soon built up a
deadly initiative:
9...Ng4 10.Qe2 Qc8 11.0-0-0 h5 12.Bg5 0-0 13.Be4 e6
Black’s moves have all been engine favourites – at a relatively low search depth. The game
should serve as a cautionary tale against superficial assessments in complicated positions,
especially when king safety is at stake.

14.Nh2! Nxe5 15.g4!

Black is essentially doomed already.
15...b4 16.gxh5! bxc3 17.h6 Bh8 18.h7† Kg7 19.h5
By now, Black (and his engine) realized the hopelessness of the situation and he resigned in
Cortigiani – Bonoldi, corr. 2018. A human continuation might be:
19...Nbd7
19...cxb2† 20.Kb1 could occur at any point over the next several moves, without helping
Black at all.
19...Bxe4 20.Qxe4 Nbc6 21.f4+– is crushing.

687
20.Nf3!
One way or another, White will break through to the key squares on the kingside. A sample
continuation is:
20...Ng4 21.Rdg1 Bxe4 22.Qxe4 f5 23.Qxe6 Qe8 24.Be7 Rf6 25.h6† Nxh6 26.Qe3+–
Black will not survive for much longer.

10.Bd2!
10.Qe2 Nxc3 11.bxc3 c5!? bothers me a bit; I considered 12.Rd1 Qc8 13.Bf4 but found
13...h6! 14.0-0 0-0 15.Be4 Bxe4 16.Qxe4 Nc6= to be satisfactory. Thus, I prefer to avoid
doubled c-pawns for the time being – although as we will see in the next note, we may still go
for this structure in a slightly different setting.

10...Nxc3
10...c5 can be met by 11.Qe2 when White continues developing quickly.

688
I also considered:
10...Nc6 11.Qe2 Nxc3 12.bxc3!?
12.Bxc3 b4 13.Bd2 Nd4 14.Nxd4 Qxd4 15.f4 Qxb2 16.0-0 seems like good compensation
for White, but Black may be able to return the extra pawn and maintain some sort of
balance.

12...Qd7
12...h5 13.a4 0-0 (13...bxa4?! 14.0-0± will give White a lot of pressure on both flanks) and
now 14.Rb1!² is the simplest way to win the pawn. (14.axb5 axb5 15.Rxa8 Bxa8 16.Bxb5
Qd5 17.e6 also seems better for White, although Black has some compensation, so this
would need to be explored further.)
13.h5 Qg4 14.Rh4 Qxg2 15.0-0-0 g5 16.Rg1 Qxg1† 17.Nxg1 gxh4

18.Be4!

689
 Normally two rooks would be considered at least the equal of a queen, but here White is
completely dominating. Black’s rooks are doing nothing, and White’s queen will coordinate
with the minor pieces to attack the numerous targets.
18...0-0-0
18...h6 19.f4 Rg8 20.Qg2 Kd7 21.Bh7+–
19.Qf3 Na5 20.Bxb7† Nxb7 21.Qxf7 Rhg8 22.Qe6† Kb8 23.Nf3+–

11.Bxc3

11...Qd5
11...Nc6 12.a3 Qd5 13.Qe2± keeps everything under control.

12.Qe2 Nd7 13.h5 Nc5 14.h6!?

The pawn sacrifice with 14.0-0-0 Nxd3† 15.Rxd3 Qxa2 16.b3 looks tempting, but things are
still complicated after:

690
16...Bh6†! 17.Nd2 e6 18.f4 Rg8!÷ The last move is important. (18...Bd5? 19.Rxd5 exd5
20.e6+–; 18...Bxf4? 19.hxg6 fxg6 20.Qg4!+–; and 18...g5?! 19.Ne4 Bxe4 20.Qxe4 0-0 21.Kd2
Qa3 22.Bb4± all turn out well for White.)

14...Nxd3†
We should meet 14...Bf8 with 15.0-0-0! when, without the ...Bh6† option, the pawn sac idea
from the previous note would be much stronger.

15.cxd3 Bf8 16.Rh4!? c5

16...0-0-0? runs into a beautiful refutation: 17.Rd4! Qe6 18.Rxd8† Kxd8 19.Ng5 Qd5

691
20.Nxf7†! Qxf7 21.e6 Qg8 22.Qe5+–

17.Bd2!
Bringing the bishop to g5 in order to exploit Black’s weak dark squares all over the kingside.

17...Rd8
17...e6 runs into 18.Bg5 Be7 19.Bxe7 Kxe7 20.Qe3 when Black’s weak dark squares can be
easily exploited.

17...0-0-0 18.a4 gives Black trouble on both flanks, and if 18...Qxd3 19.Ng5± Black must make
a concession one way or another.

18.d4 c4 19.Bg5
Black has an amazing light-squared bishop and an ideally placed queen, but he has serious

692
trouble developing, and his king could come under pressure if and when any files get opened. It
is especially unclear how Black is supposed to develop the f8-bishop and the h8-rook, whereas
White’s plans should be much more straightforward.

19...Rc8 20.Kf1 Rc6

I considered two other plans:

20...c3?! 21.bxc3 Rxc3 opens things up a little, which is not good news for the black king.
22.Kg1 e6 23.Rg4 Threatening Nd2-e4 etc. 23...Qc4 24.Qe1

24...Rc2 (In the event of 24...Bxf3 25.gxf3 Black has lost the one good thing about his position!
25...Qc6 26.d5 exd5 27.e6 fxe6 28.Bf6+–) 25.Nd2±

20...Kd7 21.Rd1 Kc7 is more sensible, but White keeps the upper hand after:

693
22.Rg4 Kb8 23.Nd2²

21.Re4! Rc7
21...Re6 is met by 22.Re3 when Black’s rook is stuck too – which stops his kingside
development plans completely.

22.Rg4!
22.Re3 e6÷ would make it harder for White to get organized. The rook is excellent on g4,
because it safeguards d4 and g2, thus enabling the knight to start manoeuvring, most likely via
d2-e4, although e1-c2-e3/b4 might also be tempting. The position could be analysed further of
course, but it seems to me that White’s chances are preferable.

B22) 8...Ng4!?N

694
 Although the previous variation led to some complex stuff, I consider the text move the critical
continuation, as it is very much in the spirit of the ...a6 Modern to create tension and provoke the
e5-e6 advance.

9.e6
9.Qe2 would be met by 9...h5 when Black has successfully dealt with the h4-h5 threat, and a
future e5-e6 will be no problem due to ...f5, with great prospects for Black.

9...f5!
9...fxe6? runs into 10.Ng5 Nxe3 11.fxe3± with all kinds of threats on the kingside.

Although this exact position has not yet been tested in practice, the idea of meeting e5-e6 with
...f5 is known from some other variations of the ...a6 Modern. Indeed, we already saw just such
an example in the earlier variation A2, where Black is doing alright. In general, positions with
this pawn configuration are complex and tricky to assess. Given time, Black will coordinate his
pieces and generate counterplay, or even just attack the e6-pawn directly, so it is important to
have a clear sense of what you are doing with White. I do believe the e6-pawn to be a strength
rather than a weakness, provided we use it to put immediate pressure on Black.

10.h5
The slower 10.Ne2 does not yield anything special after: 10...c5! (but not 10...0-0? 11.h5+– or
10...Nc6?! 11.c3±) 11.c3 (11.Nf4 cxd4) 11...Nc6 (after 11...Nxe3 12.fxe3 the elimination of the
dark-squared bishop doesn’t really help Black, since his bishop doesn’t have much to do at this
stage of the game) 12.h5 Qa5!? (12...cxd4 13.cxd4 Nb4 is another possible direction) Black
seems to be doing quite well.

10...Qc8

695
 Black should wage war on the e6-pawn.

11.d5!
This move looks obvious, but there are various dynamic and tactical points which are
necessary to justify it.

11...Nxe3
Black should eliminate the bishop in preparation for undermining White’s centre with ...c6.

The immediate 11...c6?! allows 12.Bd4! when White should be able to break through on the
kingside before Black can destroy the centre pawns. For instance: 12...Bxd4 13.Nxd4 Rg8
14.hxg6 hxg6 15.Bxf5 gxf5 16.Nxf5 cxd5 It may seem like Black is getting away with it (since
17.Qxg4?? loses to 17...Qxe6†), but White can shatter that illusion with:

17.Ng7†! Kd8 18.Qxg4 Black ends up a pawn down with a dreadful king position. 18...Qc4 is
his best option, but after 19.Qxc4 dxc4 20.Nf5+– White should win.

A more challenging alternative is:

11...b4
Targeting the knight, which is the only defender of the d5-pawn. However, with accurate
play, White can break through on the kingside.
12.Bd4!
This is strongest, although two alternatives are worth mentioning:
a) 12.Na4 is the computer’s initial preference but 12...Bxd5! 13.Bxf5 Nxe3! (13...Bxf3
14.Qxf3 gxf5 15.Qxa8 0-0 16.0-0-0 Qxe6 17.Kb1 is good for White) 14.fxe3 Bc6 15.hxg6
h6 16.c4 0-0© reached an extremely messy position; White is materially ahead, but Black’s
activity is suddenly much better and I think the tables could easily be turned.
b) 12.Ne2 Bxd5 13.Nf4 Nxe3 14.fxe3 Bxe6 15.Ng5 Bg8 16.0-0 is wildly complicated,

696
 although the engine offers a typical “0.00”. There could be a lot more to explore here, but it
seems irrelevant as the recommended move is clearly stronger.

12...Bxd4 13.Nxd4 bxc3 14.hxg6 Bxd5 15.Bxf5 Nf6 16.Rh6!!

16.Rxh7 seems tempting but after 16...Rg8! (rather than 16...Nxh7? 17.g7 Rg8 18.Qh5†
Kd8 19.Qxh7+–) things remain messy.
The text move is an amazing resource: Black can do absolutely nothing against g6-g7 and
Rxf6.

16...cxb2
16...Nc6 17.g7 Rg8 18.Rxf6+– is similar.
17.Rb1 c5 18.g7 Rg8 19.Rxf6 exf6 20.Qh5† Kd8 21.e7†+–

12.fxe3

697
12...c6 13.Nd4!
White has other playable moves but I really like the concept behind this one – White just wants
to maintain the pawn on e6 and is willing to sacrifice a pawn, or sometimes even a piece, to
accomplish that objective. If we can finish developing while maintaining the pawn on e6, it will
be hard for Black to develop harmoniously – and there could of course be some tactical
opportunities along the way.

13...Kd8!?
This surprising move is an engine suggestion which definitely deserves to be analysed. I
checked a couple of other possibilities:

13...cxd5 14.hxg6 hxg6 15.Rxh8† Bxh8 allows a simple tactic:

16.Bxf5! Kd8 17.Bxg6 (17.Bg4!? is a double-edged alternative which could also be considered)

698
17...Bxd4 18.Qxd4 Qxe6 19.Bh5² White keeps an edge due to his safer king and the weakness
of the d5-pawn.

13...c5!? 14.Nf3 c4 15.Be2 Qc5 16.Qd2 b4 reaches one of those situations where maintaining
the d5- and e6-pawns is everything, and we should have no qualms about sacrificing material to
do so:

17.Na4! Qa5 (17...Qb5 18.0-0-0! Qxa4 19.Bxc4 transposes; and 17...Qxd5? 18.Qxd5 Bxd5
19.Nb6± turns out badly for Black) 18.Bxc4! Qxa4 19.0-0-0© It is possible to analyse this
further, but I am happy to stop here as White’s compensation seems overwhelming.

14.0-0!
The main point of Black’s last move was to take the sting out of White’s attack after opening
the h-file (compare the 13...cxd5 line in the previous note, where the possibility of a check on h5
made all the difference). Therefore I suggest abandoning that idea for now, in favour of
developing as quickly as possible and fortifying the vital e6-pawn.
14.Qd2 looks like a tempting way to implement the same kind of strategy but with long
castling. However, I was unable to find anything convincing against the strong reply: 14...c5!
15.Nde2 (15.Nf3 b4 16.Na4 Bxd5 is good for Black) 15...c4 16.e4 Qc5! (16...cxd3 17.cxd3 fxe4
18.Rc1!ƒ would be nicer for us)

699
17.Rc1!? a5!?÷ Somehow, in this bizarre position White lacks a convincing way to break
through, and some kind of weird dynamic balance is in effect. An illustrative line is 18.hxg6
hxg6 19.Rxh8† Bxh8 20.Qh6 Bxc3† 21.Nxc3 Nd7 22.exd7 cxd3 23.cxd3 Qg1†= when White
will most likely have to force a perpetual with Qf8† and Rc7† in the near future.

14...cxd5
Compared to the previous note, here we are much better placed to meet 14...c5 with: 15.Nxf5!
(15.Nde2!? c4 16.Bxf5! gxf5 17.Rxf5© is an interesting variation on the same theme) 15...gxf5
16.Rxf5 c4 17.Be4

17...Qc5 Black has other moves available but the general character of the position will remain
the same, so I will just give a brief illustrative line. 18.Qf3 b4 19.Na4 Qb5 20.h6! Bxh6 21.b3©
White clearly has a lot of compensation for the piece.

700
15.Nce2 Nc6

16.Qe1!©
Keeping the e6-pawn alive. Next we can start opening the queenside with a2-a4, and if ...b4
then c2-c3 and so on. Meanwhile, Black is only one pawn up, and he will need to make some
difficult choices about how to complete development and position his pieces, not least his king.

16...Bh6
16...Qc7 17.a4 b4 18.c3 bxc3 19.Rc1!© shows White’s plan in action.

16...Kc7 17.a4 b4 18.c3 b3 19.Nf4 Ne5 20.Qe2ƒ also looks promising for us.

17.Nf4 Nxd4 18.exd4 g5

701
19.Nh3 f4 20.a4ƒ
The pawn structure is among the weirdest you are ever likely to see – but it seems clear that
Black is going to have a hard time of it. By the way, Black’s problems are not limited to having
an unsafe king in the middlegame. There is also some potential for endgame domination – for
instance, if the major pieces are exchanged then White’s bishop may settle on g4, with c2-c3
securing the d4-pawn, after which Black’s dark-squared bishop will be caged on the kingside.

B3) 7...b4!?N

Usually Black does not want to rush with this move, as the pawn on b4 may be more of a
weakness than a strength; there are several well-known lines in which White even spends a
tempo provoking it with a2-a4. However, in this specific situation, White has made an unusual
commitment with the early h2-h4; so if Black can neutralize the combination of h4-h5 and e4-e5
threats, it could justify the early lunge with the b-pawn. There is no doubt that by forcing the
knight to move away from c3, Black reduces the potency of the e4-e5 threats.

8.Ne2
8.Nd5!? is an interesting alternative, and the plausible continuation of 8...e6 9.Nf4 (9.Nxb4??
a5 traps the knight) 9...Nf6 10.Ng5 would transpose to variation B32 below. Without going into
too many details, both move orders offer certain independent possibilities for both sides (meeting
8.Nd5!? with 8...Nd7!? is just one such example on Black’s side), and I slightly prefer the e2-
route; but you could certainly investigate the alternative if it interests you.

8...Nf6!
Developing quickly – and with the knight on e2, the e5-e6 ideas seem less dangerous.
8...h6?! is too passive, and 9.h5 g5 10.a3!± exposes the b4-pawn as a weakness.

702
9.Ng5!?
I like this move more than the usual e5-e6 plan, as it generates more unusual problems for
Black. The key justification of this concept is that we can meet ...h6 with a blunt sacrifice on f7,
after which all of White’s pieces will be nicely placed to exploit Black’s weaknesses. If Black
refrains from ...h6, the knight on g5 can become quite annoying for him, especially as we will
soon be getting in h4-h5 as well.

9.Ng3 Ng4 10.Bd2 c5 11.c3 h5 seems okay for Black.

9.e5
This is another critical line of course, but it’s not my first choice so I won’t analyse it in
great detail.
9...dxe5
Black should avoid 9...Ng4 10.e6 f5 11.h5+– when the knight on e2 is perfectly placed to
go to f4 and wreak havoc on the kingside.
10.dxe5 Ng4 11.e6 fxe6 12.Nf4 Qd6 13.Ng5 Nxe3 14.fxe3 Nd7
14...Nc6 15.Qf3 Bxb2 16.Rd1 0-0-0 17.Nf7±

15.0-0 Bh6!?
15...Bxb2 16.Rb1 Be5 17.Qe2©
16.Nfxe6 Bxg5 17.Nxg5 Ne5 18.Qe2 0-0-0 19.Rad1 Qb6
The position should be close to equal, although White still has some chances to press.

Returning to our main line, Black’s most important replies are B31) 9...h6 and B32) 9...e6.

White has a simpler path to an edge after:

9...Nbd7 10.h5 e5
10...c5 11.h6 Bf8 12.c3² gives White an ideal pawn centre and the pawn on h6 is extremely
annoying for Black in the long run. Among other things, it makes it practically impossible

703
 to dislodge the knight from g5.
11.h6
This seems simplest to me.
It is also worth considering options such as 11.hxg6 hxg6 12.Rxh8† Bxh8 13.d5, fixing the
g5-square as a stable stronghold for the knight.
11...Bf8

12.Qd2 a5
12...Be7 13.Qxb4± is just a free pawn.
12...Ng4 is well met by: 13.Bc4! d5 14.exd5 Nb6 15.Bb3 Nxd5 16.0-0-0ƒ
13.Bc4 d5 14.exd5 Nxd5 15.0-0-0ƒ
Black is playing a dangerous game with his king still in the centre.

B31) 9...h6

704
10.Nxf7! Kxf7 11.Nf4
Just as prescribed earlier. For now, White only has one pawn for the piece, but Black’s king is
vulnerable and the e6- and g6-points are weak.

11...g5!?
This seems critical. Other moves are less challenging, for instance:

11...e5 12.dxe5 dxe5 13.Bc4† Ke8 14.Ne6 Qxd1† 15.Rxd1±

11...Nc6 12.e5 dxe5

13.Bc4†! e6 14.Nxe6 Qc8 15.Nxc7† Ke7 16.Nxa8 Bxa8 17.Qe2±

11...Nbd7 12.e5 dxe5 gives White more than one good option:

705
13.Bc4† This would be simplest. (13.Bxg6† Kg8 14.Ne6 Qb8 15.Rh3 Bd5 16.Bf5!ƒ is a good
alternative though) 13...Nd5 14.Bxd5† Bxd5 15.Nxd5² Black’s position is full of weaknesses.

12.Nh3!
12.Bc4† looks tempting but 12...d5 13.Nxd5 Bxd5 14.exd5 Nbd7 leaves White without an
easy way to prove anything. A few sample lines are: 15.d6† Kf8 16.dxe7† (16.hxg5 hxg5
17.Rxh8† Bxh8 18.Qf3 exd6 19.0-0-0 Qe8 20.Be6 Nb6 21.Rh1 Bg7 22.Qf5 Qb5÷) 16...Qxe7
17.0-0 Nb6 18.Bb3 Nfd5 19.Bd2 Bf6÷

12...Kf8
12...gxh4? 13.e5+– is crushing.

12...Kg8? is not much better due to: 13.e5 dxe5 14.hxg5 hxg5 15.dxe5 Nd5 16.Bf5!+–

12...g4 13.Nf4 h5 looks natural but White has a great way to strengthen the attack:

706
14.c3! Threatening a nasty check on b3. (There is no reason to hurry with 14.e5, when 14...Nd5
15.Bg6† Kg8 16.Nxh5 Qd7÷ is messy) 14...e6 (14...Nxe4 15.Qb3†+–) 15.e5 Nd5 16.Bg6†
Kg8 17.Nxh5± The g4-pawn is under fire too; and because Black has played ...e6, he can’t
defend it.

13.e5 dxe5 14.dxe5 Qd5!?

14...Nd5? 15.Qf3† Kg8 16.Bf5+– is murderous.

14...Nfd7 is met by 15.f4!, threatening to open the f-file, and if 15...Bxg2 16.Rg1 Bxh3
17.Qf3+– White wins.

15.exf6 Bxf6 16.f3 Nc6 17.Be4 Qe5!

Avoiding 17...Qxd1† 18.Rxd1±.

707
18.0-0
This seems the simplest.

18.Rb1 maintains the material balance but 18...Na5 19.Qe2 Bxe4 20.fxe4 Qg3†÷ is rather
messy.

18.Qc1!? is a safe alternative which is worth considering, even though we don’t need to
exchange queens this early. After 18...Rd8 19.Nf2 Qxb2 20.Qxb2 Bxb2 21.Rd1 Bc3† 22.Ke2
Black is a pawn up but his pawns are still weak, and his king is not completely safe either. Thus,
at the very least, White should be able to win back the pawn and still have a better structure.

18...Rd8 19.Qe1
Black may or may not pick up a pawn, but his king will remain vulnerable regardless.

19...Nd4
19...Qxb2 20.Qf2©

20.Bxd4 Qxd4† 21.Nf2 Qb6 22.Bxb7 Qxb7 23.Ne4

Black remains under unpleasant pressure.

B32) 9...e6

By bringing the queen to e7, Black tries to keep things more solid on the kingside.

10.Nf4 Qe7
It is too late for 10...h6??, when White has a few possible routes to victory, 11.Ngxe6! fxe6
12.e5+– being the most convincing.

708
11.h5 Bh6!
Preventing the pawn from reaching h6 is the most challenging approach. However, we will be
ready to meet it with a sacrificial surprise.

11...h6? 12.Nxf7 is not how Black should do it.

11...Nbd7 12.h6
I find White’s position distinctly preferable after landing the pawn on this square and
forcing the bishop to retreat to f8, as further development seems to be much harder for
Black.
12...Bf8 13.Qe2
Just continuing development. and preparing to castle.
13...Ng8
It is hard to think of another decent plan for Black.
13...a5 14.0-0-0 0-0-0 loses to 15.Ba6+– when the queenside threats are too strong.
14.Nf3 e5
14...Bxh6 15.Nxe6! fxe6 16.Bxh6 Nxh6 17.Rxh6± leaves Black with no compensation for
his pawn weaknesses.
14...c5 15.c3² does not really change much.
15.dxe5 dxe5 16.Nd5 Bxd5 17.exd5 e4
Now White has a pleasant choice:

18.Bd4
The more aggressive 18.0-0-0!? exf3 19.Qxf3© offers strong compensation for the sacrificed
piece.
18...exf3 19.Qxe7† Nxe7 20.Bxh8 fxg2 21.Rg1²
With a somewhat favourable endgame.

709
12.hxg6! Bxg5 13.e5! Bxf4
13...dxe5 14.dxe5 Bxf4 15.Bxf4 Ne4 gets busted by:

16.Qg4! fxg6 17.Be3 Nc5 18.Bxg6† Kd7 19.0-0-0†+–

14.Bxf4 fxg6 15.Bg5 Nbd7

White has a number of tempting ways to continue. A simple and flexible approach is:

710
16.Qe2!?©
White will play exf6 when he feels like it, after which the power of the dark-squared bishop
should ensure excellent compensation for Black’s extra pawn – that is, in the event that White
decides not to win it back by taking on g6. Meanwhile, Black has weaknesses on both flanks and
it is hard to imagine his king finding a comfortable home.

B4) 7...Nd7

This exact move order has hardly ever been played, although it immediately transposes to a
few other games where 6...Nd7 7.h4 Bb7 occurred.

8.h5
We will analyse B41) 8...c5 followed by the more challenging B42) 8...Ngf6!.

711
 B41) 8...c5 9.h6 Bf6

9...Bxd4 10.Nxd4 cxd4 11.Bxd4 Ngf6 just gave Black a dubious Dragon without the dark-
squared bishop in Bruzon Batista – Franco Ocampos, Linares 2001.

10.d5!
This structure seems pretty unpleasant for Black. With the bishop clumsily placed on f6, Black
will have to play ...e6 to develop the g8-knight at some point, when dxe6 will create clear
weaknesses which we can target. Just as in variation A1 of the previous chapter, I was able to
update my analysis using one of the online correspondence games in which my editor tested out
my ideas.

10...e6
Black will have to play this move at some point, so he may as well do it now, rather than wait
for White to castle and centralize the rooks.

11.dxe6 fxe6 12.Qd2 Qb6 13.0-0!

Preparing some possible pieces sacs, all of which turn out well.

13...Ne7
13...b4 14.Nd1 gives White the useful option of Bc4; and the knight can emerge on e3 later.

13...d5 is met by 14.exd5 exd5 15.Rfe1 0-0-0 16.Bf4± when Black’s whole position is unstable.

Finally, trying to evacuate the king with 13...0-0-0 14.b4 Qc7 comes unstuck after:

712
15.a4! cxb4 16.Ne2 Bxa1 17.Rxa1± With a huge attack for the sacrificed exchange.

14.Ng5!
Practically forcing Black to open the centre.

14...d5 15.exd5 exd5 16.Rfe1 0-0

16...b4 17.Bf4! bxc3 18.bxc3 is a surprising piece sac, which works because White’s pieces are
overwhelmingly active. A sample continuation is:

713
18...0-0 19.Re6 Qd8 20.Rb1 Bc6 21.c4! d4 22.Rd6!+– The threats include Ne6 and Qe2-e6†,
and Black has no way to keep his position together.

16...d4 forks two pieces but White once again obtains overwhelming compensation: 17.Bf4 dxc3
18.bxc3 0-0 19.Re6 Qd8

20.Be4! Bxe4 21.Nxe4± Followed by Rd1 or Re1. Black suffers from numerous pins and
hanging pieces; so one way or another, White will soon be able to win back the piece while
keeping an extra pawn, plus an advantage in activity.

17.Bf4 c4 18.Be4!
Other moves are decent but this seems most precise.

714
18...dxe4
18...d4?! is too loosening, and 19.Bxb7 Qxb7 20.Nce4 Nf5 21.Ne6+– sees White’s knights
swarm in.

19.Qxd7 Qc6 20.Qh3 e3 21.Rxe3±

White eventually converted the extra pawn to victory in Greet – Valovoj, corr. 2019.

B42) 8...Ngf6!

This is the more challenging move, which was played by another of Andrew’s opponents.

9.h6
Once again, pushing the bishop back to its original square seems best.

715
9.hxg6 hxg6 10.Rxh8† Bxh8 11.Ng5 is another idea, but 11...Bg7 12.f4 e5 seems okay for
Black.

9...Bf8
Black has wasted time with his bishop, but White’s centre is under pressure.

10.Bg5!N
White should be willing to sacrifice a pawn, and this seems like the best way to do it. Let me
show you why I was not quite satisfied by White’s alternatives:

10.Ng5 safeguards the e4-pawn, but Black will just play 10...e6 (or 10...e5!?) and complete
development, and it’s not clear how White can make something happen.

10.e5 dxe5 11.Nxe5!? (11.dxe5 is met by 11...Bxf3! 12.Qxf3 Nxe5 when Black is completely
fine) seems tempting but after 11...Bxg2 12.Rg1 Bb7 it is hard to make use of White’s lead in
development.

10.Qe2!? threatens e4-e5, and after 10...b4 White is ready to sacrifice a pawn with: 11.Nd5
Nxd5 12.exd5 Bxd5

13.a3 White has decent compensation, but after 13...bxa3 14.Rxa3 e6 15.Bxa6 (15.c4 Bb7 and
15.0-0 Be7 are also nothing special) 15...Rb8 Black was able to finish development and equalize
in Greet – Bujdak, corr. 2019.

My recommended improvement is based on a similar type of pawn sacrifice, but with Bg5
having been played instead of Qe2, which seems more useful. We will analyse B421) 10...c5
followed by the critical B422) 10...b4.

716
10...Rc8 11.0-0 c5 is too slow due to:

12.e5! dxe5 13.dxe5 Bxf3 14.Qxf3 Nxe5 15.Qe2 Nxd3 16.Bxf6+–

B421) 10...c5

11.d5!
11.e5? cxd4! is no good for White, while 11.dxc5 Nxc5÷ seems playable but nothing special.

After the text move we reach the same pawn structure as in variation B41 above, but here
Black’s bishop is on f8 rather than f6, which has both pros and cons. Either way though, Black’s
position is strategically risky. Ideally he would have liked to counter d4-d5 with ...c6; but as
things stand, he faces a tough dilemma over which pawn(s) to move. If he wants to develop the
dark-squared bishop then ...e6 will be required, but this will lead to a much shakier pawn

717
structure. Alternatively, he can advance on the queenside with ...c4 and ...b4 and target the e4-
pawn, but such ideas carry obvious risks when the f8-bishop and h8-rook lie dead and buried.

11...c4
11...e6 12.dxe6 fxe6 13.e5! dxe5 14.Qe2 Qc7 15.0-0-0 gives White great play, for instance:

15...Be7 (15...0-0-0 16.Be4 Be7 17.Bxb7† Kxb7 18.a4±) 16.Bxf6 Bxf6 (16...Nxf6 17.Ng5!+–)
17.Be4± With the perfect knight-versus-bishop situation.

11...Qc7!?
With this flexible move, Black keeps both of the aforementioned pawn moves in reserve.
12.b3

12...e6
12...0-0-0?! is met by 13.0-0 intending a2-a4, when the queenside does not feel at all like a

718
 safe place for Black’s king.
In the event of 12...e5!? 13.a4! b4 14.Ne2 the closed position ensures that Black doesn’t
face immediate problems. However, the central pawn structure resembles the Spanish main
lines, but White already has his pawn on h6, giving him a huge space advantage on the
kingside. This makes it hard for Black to castle on the kingside, as he could be hit by ideas
such as knight sacrifices on f5, or a simple pawn break with f2-f4. Black’s pieces don’t have
enough space to defend the kingside properly. Meanwhile, Black isn’t really in a position to
go ...0-0-0 either, as White has the easy option of c2-c3 to open the queenside.
13.a4!
In the event of 13.dxe6 fxe6 it is not so easy to exploit the e6 weakness. A possible
continuation is 14.a4 (14.Ne2? would be desirable to attack e6, but there is the small matter
of 14...c4! trapping the bishop) 14...c4! 15.bxc4 b4 16.Ne2 Be7© when Black is getting
quite nicely coordinated.
13...b4 14.dxe6! fxe6 15.Ne2
Thanks to White’s accurate move order over the past couple of moves, Nf4 is a strong and
annoying threat.

15...Be7
After 15...0-0-0?! 16.c3 bxc3 17.Nxc3 d5 18.Qe2± Black’s king can hardly feel
comfortable.
15...Nxe4?! is risky, and 16.Nf4 e5 17.Bxe4 Bxe4 18.0-0! Qb7 (18...exf4 19.Re1 Qb7
20.Nd2+–) 19.Nd2ƒ gives White tremendous play.
16.Nf4 Nf8 17.0-0²
Black was not in time to arrange ...0-0, and his position contains a number of loose pawns and
weak squares.

12.Be2 b4 13.Na4

719
13...Qa5
13...Nxe4 14.Qd4 Nef6 15.Bxc4± looks great for White.

14.Bxf6!
14.b3 is less accurate due to 14...Nxe4 15.Bxc4 Nxg5 16.Nxg5 Ne5 when Black will be able
to free himself with ...e6 next.

14.Nd2?! Qxa4 15.Nxc4 is a fancy sacrificial idea, but I am not sure if White has much after:

15...Ne5! Forcing White to go after the a8-rook, at the expense of the initiative. (15...Nc5 is
conveniently met by 16.f3! and Black will remain tied down) 16.Nb6 Qa5 17.Nxa8 Nxe4
18.Qd4 Bxa8 19.Qxe4 Bxd5©

14...Nxf6 15.b3

720
 Under normal circumstances, trading our dark-squared bishop for the knight on f6 would be a
serious positional concession; but with Black’s bishop currently buried on f8, all we really have
to worry about is keeping it that way and being ready for a possible attempt by Black to free
himself with ...e6 or ...e5.

15...Nxe4
Opening the position with 15...e5?! 16.dxe6 fxe6 17.Bxc4 seems too dangerous for Black.

16.Qd4 Nf6 17.Nb6 Rd8!?

17...Rb8 18.Bxc4 leaves Black with no real option but to head for a clearly inferior endgame
with: 18...Qc5 19.Qxc5 dxc5 20.Ne5 Rd8

21.a3! Nxd5 22.axb4 cxb4 23.Nxd5 Bxd5 24.Bxa6±

18.bxc4!

721
 18.Bxc4 Qc5 leaves Black’s rook better placed than in the above line – but even so, White still
has a slightly more pleasant endgame after 19.0-0-0 Qxd4 20.Rxd4, intending to target Black’s
queenside with Be2.
Nevertheless, I like the text move even more. With Black’s rook placed so awkwardly on d8,
we can profit from some concrete tactics.

18...Qc5
18...Rb8 19.0-0-0 Qxa2 20.Rd3ƒ gives Black nothing much on the queenside and Re3 is a
worry for him.

19.c3! bxc3 20.Bd1! Rb8 21.Qxc5 dxc5 22.Ne5±

Followed by Ba4†.

B422) 10...b4

722
 This seems like the critical continuation, by analogy with the Greet – Bujdak game mentioned
earlier.

11.Nd5 Nxd5 12.exd5 Bxd5 13.0-0

The g5-bishop is useful in impeding the development of Black’s kingside, and it seems likely
that Black will have to make a serious positional concession with ...f6. Black’s queenside also
remains vulnerable – it will almost always be possible for White to win the pawn back with a2-
a3 and so on at some point. However, we should not try to regain the pawn too soon, as that may
give Black time to catch up on development and get out of danger. Instead, we should be striving
to make finishing Black’s development as problematic as possible.

13...f6
Despite the weaknesses it creates, it is hard for Black to do without this move.

I also examined:
13...Qc8
This seems too slow.
14.Re1
14.c4!? bxc3 15.bxc3 e6 16.c4 Bxf3 17.Qxf3 Be7 18.Be4 Rb8 19.Bxe7 Kxe7 20.Bc6 gives
interesting compensation; still, it is hard to argue with placing the rook on the open file
leading towards Black’s king.
14...f6
Alternatives are no better, for instance:
a) After 14...Qb7 15.c4! bxc3 16.bxc3 Bxf3 17.gxf3 Black’s development remains a huge
problem, and the temporary weakening of White’s kingside structure is of little
consequence.
b) 14...e6 avoids ...f6 and prepares ...Be7. This may appear solid, but the reality is that

723
 White will simply respond with Bxe7 and Ng5, with a terrible bind. My analysis continues
15.a3! a5 (15...bxa3 16.Rxa3 Be7 17.Bxe7 Kxe7 18.Ng5+–) 16.Qe2 Be7 17.Bxe7 Kxe7
18.Ng5 Nb6 19.Qg4 and Black is absolutely stuck.

15.Be3 Nb6
How else will Black continue developing?
15...e6 16.c4 bxc3 17.bxc3 gives White a strong initiative, with c3-c4 coming soon; and if
17...Nb6 18.Nd2, we transpose to the note to Black’s next move below.
16.Nd2!
Preventing ...Qg4.
16...Qb7
16...e6 17.c4 bxc3 18.bxc3 Be7 19.c4 Bb7 20.Qc2ƒ makes it hard for Black to castle; and
with c4-c5 and d4-d5 ideas constantly in the air, Black is under pressure on all sides.

17.c4! bxc3 18.bxc3 Bxg2 19.c4ƒ

724
 Black is far too underdeveloped to pose any serious threats on the kingside, and d4-d5 is
coming.

14.Be3
Another approach would be to go after the b4-pawn with 14.Bd2, but then 14...e5!? followed
by ...Be7 and ...0-0 sees Black finishing his development quickly. White will need to find some
idea to exploit his temporary lead in development, otherwise Black may consolidate and simply
be fine.

14...e5!
I consider this the critical line, as other continuations give White pretty straightforward play
while Black struggles with development. The text move returns Black’s extra pawn in order to
facilitate a quick ...Be7 and ...0-0. I considered three other ideas:

14...c5 15.dxc5 e6!?

Avoiding the Bxg6† trick.
Black’s last few moves may seem strange; but if he can complete development, it would all
be worth it.
16.cxd6
16.c6!? Ne5 17.Nd4 is messy but seems promising for White too, as Black’s central pawns
feel awkward, and he is still struggling with development. However, the text move is a
simpler route to a pleasant edge.
16...Bxd6

725
17.Be4! Bxe4 18.Qxd6 Qe7 19.Qf4 Bb7
19...Bxf3 20.Qxf3 0-0 21.a3±
20.a3
White has a slight but lasting edge. It should be possible to exert pressure in the centre and/or
on the queenside with a quick Rfd1, and potential endgames after a subsequent Qd6 should be
more pleasant for White as well. Black has a pawn majority on the kingside, but it will be hard to
advance those pawns without leaving weaknesses on the dark squares.

14...e6
This should enable Black to finish development with ...Be7 and ...0-0; but even after those
moves, there will be plenty of threats on the kingside.
15.a3 bxa3 16.Rxa3 Be7 17.c4!
Developing the initiative is more important than picking off the a6-pawn.

726
17...Bb7
17...Bxf3 18.Qxf3 0-0 19.Be4© leaves Black with a position full of weaknesses.
18.Rb3! Qc8
a) 18...Rb8 19.d5! e5 20.Nh4! f5 21.Nxf5 gxf5 22.Bxf5 Nf6 23.Ba7 leaves Black virtually
paralysed and his position is falling apart.
b) 18...Bc6 19.d5! also leads to unpleasantness for Black after: 19...exd5 (or 19...Ba4
20.Qb1! Bxb3 21.Bxg6† Kf8 22.Bxh7 f5 23.Nd4+–) 20.Nd4 Ba4 21.Ne6 Qc8 22.cxd5
Bxb3 23.Qxb3 Kf7 24.f4±
19.Re1 0-0

20.d5!
With this excellent move, we divide Black’s position into two halves in preparation for
smashing up the kingside with Qc2 and Bxg6.
20...exd5
20...e5 gets crushed by: 21.Nh4 f5 22.Nxf5! gxf5 23.Bxf5 Rxf5 24.Qg4†+–
21.Qc2! Kh8
21...dxc4 22.Qxc4† Kh8 23.Rxb7! Qxb7 24.Qe6+– is another nice line.

727
22.c5! dxc5 23.Bxg6 hxg6 24.Nh4 Rg8 25.Bf4!+–
Black’s defences are overloaded.

A final possibility is:

14...Nb6
This doesn’t change too much. Black’s position is solid right now; but the more he delays
developing the kingside, the more suspicious his position becomes.
15.Qe2 Qd7
15...Qc8 16.a3 a5 17.Nd2 is similar.

16.a3!
As usual, 16.Bxa6 is not really in the spirit of the position, and 16...Qg4! 17.Bd3 Bxf3
18.Qxf3 Qxf3 19.gxf3 Nd5= is satisfactory for Black.
16...bxa3

728
 16...a5 17.Nd2 e6 18.c4 bxc3 19.bxc3 Be7 20.c4 Bb7 21.d5! also turns out great for White,
for instance: 21...e5 (21...exd5 22.Bxb6 cxb6 23.Rfe1±) 22.Bxb6 cxb6 23.c5±
17.Rxa3 Qg4
17...a5 18.c4±
18.Bxa6 Bxf3 19.Qxf3 Qxf3 20.gxf3

Even in the endgame, Black’s slow development is a big problem. Play could continue:
20...Kf7 21.Bb5! Rxa3 22.bxa3±
With an advantage in mobility, along with a dangerous passed pawn supported by two
powerful bishops, White has excellent chances of success.

15.dxe5 Nxe5 16.Bxg6†

White can also keep more tension in the centre with:
16.Be2!? c6!
Black should also maintain the tension.

729
 16...Bxf3 17.Bxf3 c6 18.Be2 Be7 19.Rc1© seems extremely pleasant for White, as he has
lasting compensation with the two bishops and Black’s weaknesses.
16...Bb7 17.a3 bxa3 18.Rxa3 Be7 19.Nd4 gives White a fine initiative for a mere pawn.
17.Nd4 Qd7 18.Re1 Be7÷
Despite his apparently strong compensation, it is hard for White to prove anything.

16...hxg6 17.Qxd5 c6
At first it seems like White’s initiative could fizzle out due to the exchanges, but Black has yet
to solve his development woes. The opening of the position also makes it harder to find a safe
place for his king. Moreover, White should keep some initiative even if all the minor pieces are
exchanged, as he can occupy the e-file and/or target the weak black queenside.

18.Qd2!
If 18.Qb3 d5 White’s queen feels misplaced, and 19.Nxe5 fxe5 20.Rfe1 Qh4!÷ is messy.

18.Qd4!? is reasonable though, with similar ideas to the main line.

18...Nxf3†
18...d5 19.Nxe5 fxe5 20.Rae1 Kd7 21.g3 leaves Black’s king a lot more vulnerable.

19.gxf3
Questions remain over what Black should be doing. He can win the h6-pawn with ...g5, but
then his bishop will be rather stuck, while opening the h-file might benefit White after Kg2 and
Rh1. Otherwise, the h6-pawn remains a pain to deal with, while Black’s king will continue to
feel uncomfortable on either side. For instance:

19...d5

730
 19...g5 20.Kg2 Bxh6 21.Rh1 Kf7 (21...a5 22.Qd3! Kf7 23.Qc4†±) 22.Qxb4 Qd7 23.Qc4† d5
24.Qd3²

20.Kg2 Qd7 21.Rh1 Kf7

21...Be7 22.h7 Kf7 23.c4 bxc3 24.bxc3²

22.c4 bxc3 23.Qxc3²

Black remains under pressure in all these lines.

Conclusion

We have reached the end of our Pirc/Modern coverage, having dealt with some fascinating lines
in the 4...a6 Modern after 5.Nf3 b5 6.Bd3 Bb7. I started by analysing 7.e5 – and although it is
not my main recommendation, it definitely deserves consideration. 7...Nd7 is by far the most
common choice, but I like White’s chances after the thematic 8.e6 pawn sac, and even the
untested 8.Qe2!?N looks pretty tricky. The rare 7...c5! is better, intending 8.e6 f5! – but even
then, after 9.dxc5 dxc5 10.Be2!N we can pose serious problems which can only be solved by
machine-like precision.

We then moved on to my main recommendation of 7.h4!, a strong move which benefits from the
added punch of surprise value. The first point to appreciate is that Black’s typical ways of
dealing with this pawn advance – namely 7...h6 and the even more dubious 7...h5?! – allow 8.e5!
with an improved version of the first option we looked at. This leaves three other serious options
for Black.

7...Nf6 is an active choice, although provoking 8.e5 entails certain risks. 8...dxe5 9.dxe5 opens

731
the position slightly, leading to some nice opportunities for White’s better-developed pieces; and
8...Ng4!?N 9.e6 f5! leads to deep complications – but with the pawn on e6 splitting Black’s
position in two, White has some promising ways to keep Black under pressure by sacrificing a
pawn or even a piece in some lines.

7...b4!?N 8.Ne2 Nf6! is an interesting twist on the previous variation. 9.Ng5!? is an attractive
reply though, intending to meet 9...h6 with a blunt sacrifice on f7, giving White excellent
attacking chances against Black’s exposed king and the e6 and g6 weaknesses. 9...e6 is safer, but
10.Nf4 followed by h4-h5 gives White easy pressure on the kingside, with further sacrificial
possibilities lurking at every turn.

Finally we considered 7...Nd7, developing another piece and simply allowing 8.h5 in the hope of
generating pressure against White’s centre. If 8...c5 then 9.h6 is a strong reply, intending 9...Bf6
10.d5! when a subsequent ...e6 will leave Black with many weaknesses. Instead, 8...Ngf6! is
critical, when 9.h6 Bf8 10.Bg5!N improves on the lone correspondence game in this otherwise
completely unexplored battleground. White should have no qualms about sacrificing a pawn, as
he enjoys easy development and a lasting initiative, while Black must make some difficult
decisions just to get his pieces out and figure out what to do with his king.

732
A) 2...f5?! 395
B) 2...g6 397
C) 2...Nf6 398
D) 2...d5 3.exd5 Qxd5 4.Nc3 399
D1) 4...Qa5 400
D2) 4...Qd6 401
E) 2...d6 3.d4 403
E1) 3...e5 404
E2) 3...Nf6 4.Nc3 405
E21) 4...Bg4 406
E22) 4...g6 407

733
1.e4 Nc6
Known as the Nimzowitsch Defence, this is the most popular of Black’s irregular first moves,
and I would regard it as the most respectable.

2.Nf3!
2.d4 is possible of course, but 2...d5! seems to be holding up fine for Black.

The text move is a much simpler option for our purposes, inviting 2...e5, with a transposition to
my next and final volume on 1.e4 e5. Naturally, Black has some independent alternatives: we
will examine A) 2...f5?!, B) 2...g6, C) 2...Nf6, D) 2...d5 and E) 2...d6.

2...a6 would seem a bizarre choice, although after 3.d4 Black can play 3...d6 to reach the 3...a6
sideline in the notes to variation E on page 403. (Black has tried some other moves in practice,
but none of them make sense so we will not waste time on them.)

2...e6 3.d4 d5 4.Nc3 converts to a sideline of the French Defence, which can be found on page
15 of my first volume. (4.e5 is a reasonable alternative, leading to an French Advance Variation
where the early ...Nc6 is less than ideal for Black.)

A) 2...f5?! 3.exf5 d5

This creates obvious weaknesses in Black’s position, although it takes some accurate play to
exploit them.

4.d4 Bxf5 5.Bb5 e6 6.Ne5 Nge7

After a few natural moves, we reach a critical position where White’s best move has hardly
ever been played.

734
7.Nc3!
7.Bg5 has been played much more frequently, but 7...Qd6 8.c3 a6 is not so clear.

The point of the text move is to deal with Black’s idea of ...a6 and ...b5.

7...a6
7...Qd6? 8.Bf4 is horrible for Black.

8.Ba4 Rb8!?
The immediate 8...b5? gets destroyed by: 9.Nxb5! axb5 10.Bxb5 Qd6

11.c3! Black is absolutely stuck, and 11...Ra6!? 12.Bf4! Rb6 13.Qa4 gave White a winning
position in Santo Roman – Palleja, Toulouse 2000.

735
8...Qd6!?N is the only other move that makes any sense, but 9.Qe2! is an excellent reply
(9.Bf4?! b5! is not clear at all). Once again, Black is unable to free himself, for instance:

9...b5 (the preparatory 9...Rb8 runs into 10.g4! Bg6 11.Bf4 with the devastating threat of Nxg6)
10.Nxb5! axb5 11.Bxb5 g6 12.Bg5+– Again, Black is completely tied up.

The text move seems like the best attempt to free Black’s position by preparing ...b5. However, it
is obvious that the inclusion of the moves Nc3 and ...Rb8 helps White’s development along, and
we can use this to launch a kingside assault.

9.Bg5!
Gaining yet another tempo.

9...Qd6 10.f4!
Solidifying the knight on e5 and setting the stage for further kingside advances.

736
10...b5 11.Bb3 Nc8?!
11...h6N 12.Bxe7 Nxe7 is clearly better for White after 13.0-0± or 13.g4!?‚; but this was still
Black’s best chance to stay in the game somehow.

12.g4!N
12.0-0 Be7 13.Qh5† g6 14.Qh6 was somewhat better for White in Kearns – Schulze, Internet
(blitz) 2012, but the text move leads to a vicious attack.

12...Bg6 13.Qd2 Be7 14.Nxc6 Qxc6

15.f5! Bf7
15...exf5 16.0-0+– leads to an overwhelming initiative.

737
16.0-0 0-0 17.f6 gxf6 18.Bxf6 Bxf6 19.Rxf6 Nb6 20.Qh6 Qd6 21.Re1
White has a decisive concentration of forces in the attacking zone. A possible continuation is:

21...Qb4 22.Rd1 Rbe8 23.Rd3 Kh8 24.g5+–

Black is defenceless.

B) 2...g6 3.d4

This is obviously similar to the Modern/Pirc lines examined earlier, but the inclusion of ...Nc6
and Nf3 at such an early stage changes some details.

3...Bg7 4.d5!?
4.Nc3 is actually the simpler option for our purposes, when 4...d6 5.Be3 Nf6 transposes to
variation E22 on page 407.
The text move is not at all essential, but I decided to check a few lines anyway, as it is certainly
tempting to kick the knight.

4...Ne5
4...Nb8 has been played less frequently. 5.Nc3 is a sensible option, but there is also 5.c4!?,
leading to a pleasant version of a King’s Indian; in both cases White should be slightly better,
although the main battle lies ahead.

5.Nxe5 Bxe5 6.Be2

There is no point in lunging forward with f2-f4; Black’s bishop is awkwardly placed on e5, so
he will almost certainly have to waste time with ...Bg7 anyway in the near future.

738
6...d6 7.0-0 Bg7 8.c4 Nf6 9.Nc3 0-0 10.Be3
White’s position seems easy and pleasant to play.

10...c6
10...e5?! 11.c5± gives Black a poor version of a King’s Indian.
10...Nd7 11.Qd2 Nc5 occurred in Donner – Kottnauer, Beverwijk 1962, when 12.Bh6N² would
have been simple and good.

11.Qd2N
White should also be slightly better after other moves, but this new approach seems most
natural to me. A sensible continuation is:

11...Ng4 12.Bf4

739
 12.Bxg4!? Bxg4 13.Bh6² also seems promising, as Black’s remaining bishop will be restricted
by White’s pawn chain.

12...Bd7 13.h3 Nf6

13...Ne5 14.Be3!?± threatens f2-f4.

14.Rfe1²
White keeps a pleasant edge.

C) 2...Nf6

This is recommended by Wisnewski in Play 1...Nc6!, but I think Black is heading for a dodgy
Alekhine.

3.e5
We should accept the invitation to do battle in a Nimzowitsch-Alekhine hybrid opening,
otherwise Black will be able to play ...e5 next.

3...Ng4
3...Nd5 4.d4 d6 converts to variation D of Chapter 1 on page 17.
3...Ng8?! 4.d4 leaves Black with an even more dubious version of the 1...Nf6 2.e5 Ng8 sideline
(which was covered in a note on page 8), as there is no reason for him to commit the knight to c6
so soon.

4.d4 d6
Challenging the centre is the only serious attempt to justify Black’s opening play; but it doesn’t
actually put much pressure on White at all.

740
5.h3 Nh6 6.Nc3!
6.Bb5 a6 7.Bxc6† bxc6 8.0-0² is a simple way to avoid any worries in the centre by removing
the c6-knight – but there is actually no need to give up the bishop just yet.

6...a6
This is the main line although Black’s slow development seems suspicious to say the least.

6...dxe5 7.d5 Nb8 8.Nxe5± gives White a significant lead in development.

7.Bg5 Bf5
Black can hardly be happy with his position after: 7...dxe5 8.d5 Na7 (8...Nd4 9.Nxe5 is close
to winning for White) 9.Nxe5 g6 10.Qd2 Nf5 11.g4 Nd6 12.Qe3

White is overwhelmingly better developed. A good example continued: 12...Nab5 13.Ne2 Bg7
14.Bh6 0-0 15.Bxg7 Kxg7 16.Nf4± Fedeli – Perini, corr. 2008.

741
8.Nh4!
This renders Black’s opening scheme virtually unplayable. Wisnewski does not consider this.

8...Qd7
After 8...Bc8 9.f4 White may appear to be advancing too rashly, but Black is too poorly
developed to do anything about it. For instance: 9...Nf5 (9...dxe5 10.d5±) 10.Nxf5 Bxf5 11.g4
Bd7 12.Bg2±

8...Nb4 is well met by 9.Rc1! (rather than 9.Nxf5 Nxf5 when the threats against the centre are
annoying) 9...dxe5 10.d5 when there is not much that Black can do. For instance:

10...g6 11.a3 Nxc2† 12.Rxc2 Bxc2 13.Qxc2±

9.g4! Bg6 10.d5 Nd8 11.f4

742
 Threatening to win a piece, so Black has no real choice but to weaken himself.

11...e6 12.dxe6 fxe6 13.Nxg6 hxg6 14.Bd3

White had a huge initiative in Jones – V. Smirnov, Sydney 2010.

D) 2...d5 3.exd5 Qxd5

This leads to a version of the Scandinavian which is slightly dubious for Black, but it takes a
certain amount of precision to make the most of White’s chances.

4.Nc3

We will analyse the historically most popular D1) 4...Qa5, followed by the recently more
fashionable D2) 4...Qd6.

4...Qf5 5.d4 Nb4 has been played a few times, but 6.Bb5† c6 7.Ba4± makes Black’s pieces look
silly.

4...Qh5 has been played in over a hundred games – which is bizarre, as 5.Nb5 already forces
5...Kd8, and after 6.Be2± it is obvious that things have gone wrong for Black.

D1) 4...Qa5 5.d4 Bg4

5...Nf6 6.Bb5 Bd7 7.Bd2± is excellent for White.

6.Bb5

743
6...0-0-0
Other moves exist of course, but it is obvious that Black’s entire set-up is geared towards long
castling, so we will focus on that.

6...e6 7.h3 Bh5 8.0-0 0-0-0 9.Bxc6 bxc6 10.Qd3 looks clearly better for White. 10...Bd6 was the
continuation of Surplys – Eliseev, Rimavska Sobota 1992; and now, out of several promising
options, I especially like 11.Nd2!?N intending Nc4.

7.Bxc6 bxc6 8.Qe2!

8.h3 allows 8...Qh5! and Black is doing fine.

8...Kb7
8...Rxd4 9.Be3 Rd6 occurred in Vlahov – Marinic Kragic, Rabac 2004, when 10.h3N (or 10.0-
0N) would have given White more than enough for the pawn.

8...Nf6 9.0-0 e6 10.h3 Bh5 11.g4 Nxg4 12.hxg4 Bxg4 13.Qe4± left Black with no real
compensation for the material deficit in Short – Zakarian, England 2015.

744
9.h3!N
9.Be3 e6 10.h3 Qh5 11.0-0-0 Bxf3 12.gxf3 Nf6÷ was not too convincing in Berzinsh –
Bitinas, Liepaja 2005.

9.b4!? Qf5 10.b5 looks tempting; but after 10...Bxf3 11.gxf3 as seen in Kwiatkowski – Vlassov,
Polanica Zdroj 1993, 11...c5!N 12.dxc5 Qxc5 13.Ne4 Qe5 would have left things unclear.

9...Qh5
9...Bxf3 10.Qxf3 Nf6 11.0-0 simply gives Black a bad position where his weak queenside will
surely tell.

The text move transposes to a couple of games in which the moves h2-h3 and ...Qh5 were played
slightly earlier.

10.d5!
This strike offers White a huge advantage. For instance:

745
10...Rd6
Both 10...Bxf3 11.dxc6† Kxc6 12.gxf3+– and 10...cxd5 11.Qb5† Kc8 12.Nd4+– win almost
effortlessly for White.

10...Rxd5 11.Nxd5 Bxf3 12.gxf3 Qxd5 could be Black’s best chance to prolong the game, but
it’s obvious that White should be winning with decent play.

11.Bf4 Bxf3 12.gxf3 Rf6

13.0-0-0!
White’s attack was decisive in Rabl – Wolf, Moeglingen 1989.

D2) 4...Qd6

746
 In a way, the theoretical state of this variation mirrors the state of the main Scandinavian
Defence, with a ...Qd6 set-up seeming like Black’s most resilient option. But just like variation
D1 above, the inclusion of the early ...Nc6 makes this a worse version for Black.

5.d4
5.Nb5 is too soon because after 5...Qd8 White will not have time to establish the ideal d4/c4
pawn set-up. 6.d4 a6 7.Nc3 Nf6 occurred in Ponomariov – Beerdsen, Germany 2019, when
Black had reached the kind of normal-looking position that Scandinavian players tend to feel at
home in.

5...Nf6
5...a6?! prevents our next move in the main line, but after 6.d5 Ne5 7.Be2± Black will remain
way behind in development.

6.Nb5 Qd8 7.c4!

7.d5 looks tempting but after 7...Nb4 8.c4 e6 it is not obvious how White can keep his centre
intact. 9.Bf4 Na6 10.d6 c6 11.Nc3 Nh5!÷ was adequate for Black in Damaso – Vitor, Santo
Antonio 2003.

7...a6 8.Nc3

747
8...Bg4
8...e6 is too passive, and 9.Be3 gives White a perfect version of the d4-c4 pawn set-up.

9.d5 Ne5 10.Be2 Bxf3

This is more challenging than 10...Nxf3† 11.Bxf3 Qd7 12.0-0 Bxf3 13.Qxf3 g6 14.Re1 Bg7
15.Bg5 0-0 16.Rad1± as seen in Kotenko – Mescheryakov, Astrakhan 2013.

11.Bxf3 Qd7!?N
11...Nxc4? is refuted by 12.Qa4† b5 13.Nxb5! – which explains the text move.

11...e6 occurred in Esquivias Quintero – Gallardo Fernandez, Seville 2017, when 12.0-0N
12...Nxf3† 13.Qxf3± would have given White an easy advantage.

748
12.0-0!
12.Be2 is too slow, and 12...e6 13.0-0 exd5 14.cxd5 Be7= is decent enough for Black.

12...Nxc4
12...e6 13.dxe6 Qxd1 14.exf7† Kxf7 15.Rxd1± makes things easier for White.

Capturing the pawn is the only critical continuation. From an intuitive point of view, it seems
obvious that White’s huge lead in development should yield more than enough compensation. At
the same time, accurate play is needed, otherwise Black will simply play ...g6, ...Bg7 and ...0-0,
at which point he will be fine.

13.Re1!
13.Bf4 g6! 14.Rc1 Nd6 15.Na4 Bg7 16.Nc5 Qf5 17.Bxd6 cxd6 18.Nxb7 0-0 is not so easy.
White should be able to press by using the c6-outpost, especially with his knight; but I think we
can aim for more.

13...Nd6
13...g6? 14.Qd4!+– forks Black’s knights.

14.Na4! g6
14...b6? is too slow, and after 15.Bf4 g6 16.Rc1 Bg7 White has a simple but attractive way of
breaking through:

17.Bxd6 Qxd6 18.Rc6 Qd7 19.d6!+–

15.Nc5 Qc8 16.Qa4†!

Provoking the following pawn move, which will permanently weaken Black along the c-file.

749
16...b5 17.Qd1 Bg7 18.Bf4 Qd8
Black could castle, but in that case White would simply pick up the e7-pawn while maintaining
his other advantages.

19.Nd3!? 0-0 20.Ne5 Re8 21.Rc1±

White’s pieces are tremendously active and it seems highly probable that he will at least restore
material equality by picking up one of Black’s weak pawns sooner or later. Meanwhile it is hard
to find a way for Black to exchange pieces or otherwise relieve the pressure on his position.

E) 2...d6 3.d4

Black has tried numerous moves here. We will focus on E1) 3...e5 and E2) 3...Nf6, the latter
move being the more logical choice for Nimzowitsch Defence devotees.

750
3...g6 4.Nc3 Bg7 5.Be3 Nf6 transposes to variation E22 on page 407.

3...Bg4 has been played in lots of games but does not require too much attention. White has a
choice of good replies, the simplest being: 4.Be3 (4.d5 is the maximalist try, based on the fact
that 4...Ne5?! 5.Nxe5! turns out well for White) 4...e5 (4...Nf6 has been by far the most common
choice, when 5.Nc3 transposes to variation E21 on page 406) 5.d5 Nce7 6.h3 Bd7 7.c4 White
has an exceptionally nice King’s Indian, as Black has wasted time.

The only other idea worth mentioning is:

3...a6
Intending ...e5 without allowing Bb5 in reply – but Black is spending a lot of time just to
get a King’s Indian structure.
4.Bd3 e5 5.c3
This seems simplest if you don’t want to rush into a King’s Indian structure. There is not a
lot that Black can do to put pressure on the centre.
5.d5 Nce7 6.0-0 Nf6 7.c4 should also be pleasant for White, as ...a6 just seems like a
wasted tempo in a line which would have been slightly worse for Black anyway.
5...Nf6 6.0-0
How will Black continue developing?

6...Bg4
If 6...Be7 then 7.d5 Nb8 8.c4± really becomes tempting.
7.Be3 d5!?
7...Be7 8.d5 Nb8 9.c4±
8.Nbd2²
White’s lead in development gives him the edge with the position opening up.

751
 E1) 3...e5

4.Bb5!
Technically speaking this position should be categorized as the Steinitz Variation of the Ruy
Lopez (as occurs after 1.e4 e5 2.Nf3 Nc6 3.Bb5 d6 4.d4), so I could have postponed this part
until the next volume – so you can just think of it as an added bonus for this book.

4...exd4
4...Bd7 5.d5 Nce7 6.Bxd7† Qxd7 7.c4² gives White an improved version of a King’s Indian.
The absence of the light-squared bishops clearly helps him – not only because of the general
notion of ‘good/bad’ bishops, but also because Black’s light-squared bishop often serves a
specific role in Black’s future kingside attack. Moreover, White has not castled yet, so he can
play flexibly and perhaps avoid being attacked altogether.

5.Qxd4!?
The more popular 5.Nxd4 should lead to a pleasant edge for White; the choice is a matter of
taste. I happen to like the text move, in connection with rapid development and long castling.

5...Bd7
Other moves exist of course, but in general White obtains a pleasant game by following the
same plan. The whole line is a rare occurrence these days, so I will just focus on the main line.

6.Bxc6 Bxc6 7.Nc3 Nf6 8.Bg5 Be7 9.0-0-0 0-0

752
10.Rhe1
Defending against any ...Nxe4 ideas.

10...h6
10...Nd7 has been the most common choice, but 11.Bxe7 Qxe7 12.Qd2² just seems pleasant
for White.

I also considered: 10...Re8 11.Kb1 Qd7!?N 12.h3 b5 (12...h6 13.Bc1 b5 14.g4 a5 15.g5 hxg5
16.Nxg5ƒ) 13.g4 a5 14.Nd5 Nxd5 15.exd5 Bb7 16.Bxe7 Rxe7 17.Rxe7 Qxe7 Despite the
simplifications, White keeps some pressure after:

18.Qd3 b4 19.Re1 Qd8 20.h4ƒ Black’s bishop is clearly not doing much, and White is
developing a strong initiative.

753
11.Bf4 Nd7

Compared to the 10...Nd7 line mentioned above, Black is playing slightly more ambitiously,
manoeuvring the knight while keeping the bishop pair on the board. The downside is that the
...h6 move weakens the kingside – and with this in mind, I like the regrouping plan found by
White in the following game.

12.Qe3! Re8 13.Nd4 Bf6 14.Nf5 Ne5 15.Bxe5 Bxe5 16.f4 Bf6 17.Qg3±
The duel threats of e4-e5 and Nxh6† led to big problems for Black in Berelowitsch – Voiska,
Achaea 2012.

E2) 3...Nf6 4.Nc3

Black’s two main options are E21) 4...Bg4 and E22) 4...g6.

754
I also checked:
4...e5 5.Bb5!
Once again we convert to the Spanish.
5...Bd7
5...exd4 6.Qxd4!? (6.Nxd4 is a good alternative of course) 6...Bd7 7.Bxc6 Bxc6 transposes
to variation E1 above.

6.Bxc6! Bxc6 7.Qd3

It may seem that White made a concession by exchanging on c6 voluntarily, but now it
becomes hard for Black to maintain the central tension – and after he exchanges on d4,
White obtains an easy edge.
7...exd4 8.Nxd4 Nd7
8...Be7 9.Nf5ƒ
8...Bd7 9.Bg5 Be7 10.0-0-0±
9.Nxc6 bxc6 10.Qa6 Nb6 11.Qb7±
Grischuk – A. Ivanov, Dagomys 2010.

E21) 4...Bg4 5.Be3

755
5...e5
5...e6 has been played in lots of games but I don’t understand the appeal. Black will soon have
to give up the light-squared bishop and he has already given up the centre, so there is no question
that he stands worse after: 6.h3 Bh5 (or 6...Bxf3 7.Qxf3²) 7.d5 exd5 8.exd5 Ne5 9.g4 Bg6
10.Nh4²

6.Bb5
It seems simplest to follow the same kind of recipe as in the ...e5 lines covered above.

With that being said, 6.d5 Ne7 7.Qd2 (or 7.h3) should also favour White. Both sides have
different plans available and I don’t intend to analyse this position in much depth, so I will just
present a sample line:

7...Ng6 8.0-0-0!?N 8...Be7 9.Be2 0-0 10.h4 a6 11.Ne1!? Bxe2 12.Qxe2ƒ

756
6...exd4
6...Nd7?! 7.d5 Ne7? 8.Nxe5!+– is a nice detail.

7.Qxd4 Be7 8.h3!? Bd7

9...Bxf3 gives White an open file for his attack.

9.Bxc6 Bxc6 10.0-0-0 0-0 11.g4²

Delchev – Mestrovic, Zadar 2002.

E22) 4...g6

This is recommended by James Schuyler in The Dark Knight System. If you turn back to page
267, you will see that when Black opted for an early ...Nc6 via a Modern Defence move order, I
recommended meeting it with Nge2 and f2-f3. Here we do not have that option – but still, White
should be able to claim at least a slight edge.

5.Be3
This seems simplest, although 5.Bg5!? looks good too.

5...Bg7 6.Qd2
6.Be2 0-0 7.0-0 a6 8.Qd2 e5= is nothing special for White.

6...0-0
6...Ng4?! 7.Bg5 h6 8.Bh4± simply leaves the knight awkwardly placed.

757
7.0-0-0 Bg4
7...e5? turns out badly for Black after 8.dxe5 dxe5 9.Qxd8 Nxd8 10.Nxe5 Nxe4 as seen in
Cirjan – Maksimovic, Niska Banja 2017, when White missed a fairly simple trick:

11.Nxg6!N 11...Nxc3 12.Ne7† Kh8 13.bxc3 Black must either struggle on a pawn down, or
accept a serious positional disadvantage with: 13...Bxc3 14.Nd5 Be5 15.f4 c6 16.fxe5 cxd5
17.Bg5±

7...a6 8.Bh6 b5 9.Bxg7 Kxg7 is recommended by Schuyler, though he admits that White is
slightly better. I suggest:

10.Qe3N 10...e6 11.h3 b4 12.Ne2 a5 13.g4²

White looks to have the more promising attack.

758
8.Qe2!N
Neither 8.h3 Bxf3 9.gxf3 e5 10.d5 Nd4÷ nor 8.Be2 e5 9.d5 Bxf3 10.Bxf3 Nd4÷ were to my
liking.

It feels a little odd to move the queen again, but it is worth losing a tempo to counter any ...e5
ideas. Next we will eliminate the light-squared bishop with h2-h3, followed by attacking on the
kingside.

8...a6
8...e5 is still legal of course, but 9.dxe5 Nxe5 10.h3! Bxf3 11.gxf3± gives White excellent
play.

9.h3 Bxf3 10.gxf3 b5 11.e5 Nh5 12.f4±

White has a promising initiative and the light-squared bishop has great potential.

Conclusion

We meet the Nimzowitsch Defence with 2.Nf3! when, if Black does not wish to transpose to the
next volume with 2...e5, he can choose from a variety of alternatives. 2...f5?! is the worst of
them, as Black saddles himself with lasting weaknesses. 2...g6 is reasonable, but highly likely to
transpose to a line examined elsewhere. We then looked at 2...Nf6 3.e5 Ng4 4.d4, which just
seems like a dodgy Alekhine from Black’s perspective.

The two most important moves are with the d-pawn. First we looked at 2...d5, when 3.exd5
Qxd5 4.Nc3 seems to offer White slightly better prospects than a normal Scandinavian, as the
knight on c6 can often be targeted by Bb5 or d4-d5 at some point.

759
We finished by analysing 2...d6 3.d4, when the game may take on various forms. If Black plays
...e5 any time over the next few moves, we will counter with Bb5, converting to a Ruy Lopez, or
at least something close to it. The main line is 3...Nf6 4.Nc3, when 4...Bg4 is playable, although
Black will almost automatically stand slightly worse after trading the bishop for the f3-knight.
Finally 4...g6 leads to an interesting version of a Pirc, but I like White’s chances – especially
with the help of the novelty mentioned at the top of this page.

760
A) 1...b6 2.d4 Bb7 3.Nc3 e6 4.Nf3 Nf6 5.Bd3 Bb4 410
A1) 6.Qe2 411
A2) 6.e5!? Ne4 7.0-0 Nxc3 8.bxc3 Bxc3 9.Bg5!?N 414
A21) 9...Bxf3!? 416
A22) 9...Qc8 10.Rb1 417
A221) 10...h6 417
A222) 10...c5 418
A223)10...Nc6 419
B) 1...a6?! 422
C) 1...h6?! 424

761
1.e4
We end the book by looking at a few of Black’s rare and slightly less reputable options. We
start with A) 1...b6, which is just about respectable, followed by B) 1...a6?! and C) 1...h6?!,
which are harder for Black to justify.

1...g5? 2.d4 h6 transposes to variation C.

A) 1...b6 2.d4 Bb7

This is Owen’s Defence – not a brilliant opening, in the sense that it enables White to occupy the
centre and preserve an edge in more than one way – but at the same time, it still presents some
challenges in terms of how to convert White’s slight opening initiative into something tangible.

3.Nc3
A reasonable alternative is 3.Bd3 followed by meeting ...c5 with c2-c3, thus avoiding having
the c3-knight pinned by ...Bb4. This is perfectly acceptable for players who are content to obtain
a slight advantage without getting embroiled in a theoretical debate.

I have chosen to recommend a Nc3 set-up because I consider it the most principled approach. It
also reduces Black’s options – he really needs to go for the ...Bb4 set-up, otherwise White will
have made no compromises at all, so 1...b6 would make no sense. Finally, throughout these 1.e4
books I have always strived to find new ideas where possible, and I was able to make some
interesting discoveries as we will see.

3...e6
3...Nf6 4.Bd3 is likely to convert to normal lines, although White could certainly consider 4.e5
as well.

3...g6 4.Be3 Bg7 has already been covered via 1...g6 – see the note on page 253.

4.Nf3
4.Bd3 followed by Nge2 is a decent alternative, but I personally prefer the knight on f3.

4...Nf6
4...Bb4 5.Bd3 Nf6 leads to the same thing.

5.Bd3 Bb4
As mentioned above, I consider this the only really challenging approach.

762
Black has tried almost every legal move at some point, but I will just mention the most
significant alternative:
5...d5 6.e5
Christian Bauer prefers 6.exd5 in Play 1...b6, but I am happy to convert to a French
structure.
6...Ne4
6...Nfd7 has been a more common choice but 7.Ne2 c5 8.c3 is simply a bad French for
Black, as the bishop is passive on b7 and Black’s queen is deprived of the b6-square.
Sometimes in the French, ...b6 is played with ...Ba6 in mind, but here that would waste
additional time.

7.Ne2!N 7...c5 8.c3 cxd4

8...Nc6 9.0-0² gives White easy play.
9.cxd4!
9.Nexd4² is also good.
9...Bb4† 10.Nd2²/±
One way or another, Black will end up in a dubious version of a French.

A1) 6.Qe2 is the usual line, but the more aggressive A2) 6.e5!? is where I focused most of my
efforts.

A1) 6.Qe2

763
6...d5
Other moves exist but, since this is not my main recommendation anyway, I will just focus on
Black’s most popular and critical continuation.

7.0-0!?
This remarkable move is virtually unknown, though not quite a novelty.

7.exd5 has been the most popular choice but it does not seem to cause Black too much trouble.

7.e5 Ne4 8.0-0!? is a thematic pawn sacrifice but this is not quite the ideal situation for it. Play
continues 8...Nxc3 9.bxc3 Bxc3 when Black should be okay if he keeps the dark-squared bishop
alive. A good example continued: 10.Rb1 Nc6 11.Qe3

11...Bb4! 12.Qf4 h6 13.Bb5 Be7 14.Qg4 g6 15.c3 Qd7 16.Ne1 a6 17.Ba4 b5 18.Bd1 Na5 In

764
Nakamura – Gareev, Saint Louis 2015, Black had completely consolidated and White had to
worry about equalizing. I wanted to show this game as a reference point, as later I will be
recommending a similar type of sacrifice, in an improved scenario where Black will find it
harder to consolidate.

7...Bxc3

8.e5
Black has four possible ways of accepting the pawn sacrifice.

8...Bb4!
Preserving the dark-squared bishop is best, and leads to a strategically complex position.

8...Bxd4?! 9.Nxd4
This should enable White to continue his development rather effortlessly.
9...Ne4
9...Nfd7 10.Qg4 Kf8 11.Bg5 Qe8 12.f4 with a crushing attack.
10.Qg4 Kf8

765
11.b4! Nc6
11...Nd7 12.f3 is awkward for Black.
11...c5 12.f3! cxd4 13.fxe4ƒ opens the f-file to White’s advantage.
12.Nxc6 Bxc6 13.f3 h5 14.Qf4 g5 15.Qe3 d4 16.Qe1 Nc3 17.b5 Bb7 18.Qg3 Rg8 19.h4±

8...Ne4 9.bxc3 Nxc3

This reaches a variant of the 7.e5 line which Black does best to avoid, as White’s dark-
squared bishop may become powerful later.
10.Qe3
Not only has Black easily given up the dark-squared bishop; his awkward knight makes life
even harder for him.

10...Nd7
10...Ne4 11.Ba3©

766
11.Ng5 c5 12.f4!?N
12.Qf4 Qe7 13.Bd2 Ne4 14.Nxe4 dxe4 15.Bxe4 Bxe4 16.Qxe4 0-0 17.c3 was marginally
better for White in Mircea – Taras, corr. 2009, but the dark-squared bishop seems less
significant now.
12...cxd4

13.Qh3!ƒ
With a dangerous initiative for the sacrificed pawns. (13.Qxd4 Ne4 14.Bxe4 dxe4 15.Ba3 Nc5
16.Bxc5 bxc5 17.Qa4† Qd7= would be okay for Black.)

8...Bxb2 9.Bxb2 Nfd7

This leads to something resembling the line above, but the black knight is safer on d7 than
on e4. White has decent compensation but he needs to come up with a plan. Many moves
are possible but I like the idea of building an initiative slowly, just meeting ...c5 with c2-c3.

767
10.Qe3
10.Nd2!? is an interesting alternative, releasing the f-pawn. Here are a few lines: 10...c5
(10...0-0 11.f4! g6 12.Nf3 gives White the makings of a huge kingside initiative) 11.f4
(11.c3 Nc6 12.Rae1 Qc7 13.f4 g6 [13...0-0-0 14.f5±] 14.Nf3 0-0-0 15.Ng5 Rdf8÷) 11...g6
12.c4!?© The position remains complicated. (12.f5 gxf5 13.Bxf5 exf5 14.dxc5 Rf8 15.e6
fxe6 16.Qh5† Ke7 17.Qg5† Ke8 18.Qh5†= only ends in a draw.)
10...c5 11.Qf4
11.a4!? should also give White a decent initiative, with Ba3 and c2-c3 coming up.
11...Nc6 12.c3 Qe7 13.Qg4 g6
13...f6 14.exf6 Nxf6 15.Qh3ƒ
14.Bc1 h6 15.a4©

9.exf6 gxf6
Obviously the other capture would allow Black’s queen to be trapped, so he has to weaken his
structure.

10.c3 Be7
This position is hard to analyse, or assess. Black is a pawn up and there are no immediate
threats, but his pieces are extremely passive, especially the b7-bishop. At the same time, it is not
clear how White will break through. The position deserves further analysis and testing. For now,
I will just present what I see as a logical continuation for White.

11.Nh4!?N
I slightly prefer this to 11.Bf4, as was played in Ludwig – Hille, Germany 2013.

11...h5
Preventing the annoying Qh5.

768
11...f5 is risky, in more than one way. 12.Nxf5!? is certainly tempting, and after 12...exf5 13.Re1
Nc6 14.Qf3© White has full compensation for a piece, although objectively it should only yield
a draw. Instead, 12.Nf3! looks preferable, aiming to exploit the weakness of the e5-square. For
instance, after a subsequent ...Nc6 White has the option of driving the knight away with b4-b5 at
some point, followed by plonking the knight on e5. The game goes on of course, but I like
White’s chances.

12.Re1 Qd7 13.b4 Nc6 14.Bd2

Just waiting for Black to do something.

14...a5
14...0-0-0 seems suspicious, and 15.a4 Kb8 16.a5 Rdg8 17.g3 gives White good attacking
chances.

14...a6 15.a4 does not change much.

15.a3 axb4 16.axb4 Rxa1 17.Rxa1 Kf8÷
The position remains complex and double-edged.

A2) 6.e5!?

769
 This not-so-common move is my main recommendation. The general idea is to sacrifice a
pawn, in a more favourable way than in the Nakamura – Gareev example referred to earlier.

6...Ne4
6...Nd5 7.0-0 will lead to the same thing.

7.0-0 Nxc3
Giving up the dark-squared bishop makes no sense to me:
7...Bxc3?! 8.bxc3
8.Qe2!? is interesting but not needed.
8...Nxc3
Now a small refinement is:

9.Qd2!N

770
 9.Qe1 has been played in all the games so far, and although everyone has responded with
9...Bxf3 (which should transpose to our line after 10.Qxc3N, rather than 10.gxf3 which I
don’t like so much – but that’s another story), the alternative 9...Nd5!?N would at least
deserve consideration, whereas with the queen on d2 this knight retreat is just bad for Black.
9...Bxf3
9...Nd5?! 10.c4 Ne7 11.Ng5± gives White a serious initiative.
10.Qxc3 Bb7 11.f4©
White’s initiative is excellent value for a pawn.

8.bxc3 Bxc3
Declining the sacrifice with 8...Be7 is obviously less critical, and 9.Re1!?N (9.Nd2 d6 10.f4
could be another way; see for instance Nestorovic – Markovic, Senta 2016) 9...d6 (9...0-0 10.h4!
‚) 10.Qe2 Nd7 11.Bf4 0-0 12.Rad1 seems pleasant for White.

9.Bg5!?N
This is my new idea, trying to drive the queen to an awkward square or provoke some other
change in the position. I analysed it myself but later became aware that Bauer actually mentioned
it in his 1...b6 book fifteen years ago! Even though our evaluations differ, I am quite impressed
that the French GM considered it at all.

The alternative is:

9.Rb1
This has been played in all the games so far, and is not a bad move. I will include some
brief lines which I looked at.
9...Nc6 10.Bg5
a) 10.Rb3 Bxd4 (10...Bb4 11.Be4 Be7 is another possible direction) 11.Nxd4 Nxd4 12.Qg4
Nxb3 13.Qxg7 Nxc1 14.Qxh8† Ke7= led to a quick draw in Joecks – Weyrich, Germany
1994.

771
 b) 10.Be3 gives White a slightly worse version of variation A223 below. The position is
still complicated and White has decent compensation; but the black queen is better on d8
than c8, as it gives him options like ...d6, or simply ...Qe7 followed by ...0-0-0.
The text move is a decent try but it offers Black a useful alternative compared to my
recommendation of playing Bg5 a move earlier.
10...f6!
10...Ne7 11.Rb3!? looks quite promising: 11...Bxf3N (11...Ba5 left the bishop sidelined and
White had good compensation in Faas – Turikov, Leningrad 1975; the text move must be
critical) 12.Qxf3 Bxd4 13.Re1© This would need to be analysed further, but I think White
has a powerful initiative for the sacrificed pawns.

11.Be3!?N
11.exf6 gxf6 gave White nothing special in Cuellar – Pena Morales, Lima 2013.
11...Qe7
11...fxe5 12.Ng5 looks dangerous.
12.Rb3 Ba5÷
White still has decent compensation but the worst seems to be over for Black, who is ready for
...0-0-0 next.

We will analyse A21) 9...Bxf3!? and A22) 9...Qc8.

Of course it is also important to consider:

9...f6 10.exf6 gxf6

772
11.Ne5! Qe7
11...h5 12.Bg6† Ke7 13.Qc1 Bxd4 14.Bxf6† Kxf6 15.Qf4† Kg7 16.Qxd4ƒ looks
exceedingly dangerous for Black.
12.Qh5† Kd8 13.Nf7† Kc8 14.Nxh8 Bxa1
14...fxg5 15.Rad1±
15.Rxa1 fxg5

16.Nf7 Nc6 17.c3 a6 18.Nxg5 Kb8 19.Bxh7 Ka7 20.Be4²

White is a pawn up with connected passers on the kingside.

A21) 9...Bxf3!? 10.Qxf3 Qxg5 11.Qxa8

773
 This continuation transforms the position and leads to an interesting situation where Black will
be materially well set, having two pawns for the exchange, but will nevertheless come under
some pressure.

11...Qd8 12.Rad1 Bxd4 13.Rfe1 a5 14.Qe4 Bc5

Black is extremely solid but he can do little except wait. Meanwhile, White has many ways to
probe and put pressure on his opponent. This is especially true with queens on the board, since
the opposite-coloured bishops and rook firepower give White chances to develop a strong attack,
which the engine may underestimate at first.

15.g3 g6
Preparing to castle.

16.Bb5!

774
 Provoking ...c6 and preventing the knight from going to that square.

16...c6
16...0-0 17.Bxd7! Nxd7 18.Qd3± is a nice trick.

17.Bd3 0-0
17...f5 18.Qf4 gives White good chances to prise open a kingside file after further preparations.

Another interesting line is 17...f6 18.Kg2 0-0 19.h4!, and if 19...fxe5 20.Qxe5 Rxf2† 21.Kh3ƒ
the loss of the f2-pawn is less important than the opening of the f-file – especially with Black’s
knight doing absolutely nothing on b8.

18.Kg2 Qe7
18...f5 19.exf6 Qxf6 20.f4 also looks good for White.

19.h4 h5 20.c3 Kg7 21.Be2 Rd8 22.f4²

These moves are not all forced of course, but they seem logical enough. White can continue to
shuffle his pieces about and look for a suitable opportunity to open up the kingside with g3-g4.

A22) 9...Qc8 10.Rb1

This leads to a final split, with A221) 10...h6, A222) 10...c5 and A223) 10...Nc6 the most
logical tries.

10...d5 11.Rb3 Ba5 should enable White to generate a strong initiative with 12.Be3 followed by
Ng5 and so on. (12.Nh4!? intending f2-f4 also looks dangerous.)

775
10...Bd5 seems a bit unnatural but I mention it because Bauer reaches this position in another
line via a different move order. 11.Bd2! Bxd2 12.Nxd2 Qd8 13.Re1© gives White plenty of play
for the sacrificed pawn. A sample continuation is:

13...Bxa2 (13...d6 14.Qg4±) 14.Ra1 Bd5 15.Qg4 g6 16.c4 Bb7 17.Be4 Nc6 18.d5 Na5 19.dxe6
fxe6 20.Rxa5! Bxe4 21.Nxe4 bxa5 22.Nf6† Kf8 23.h4+–

A221) 10...h6

Prodding the bishop is a natural move to consider, but it may only prove to be a weakening of
Black’s kingside.

11.Be3
White’s position has a great deal of attacking potential, as the following lines illustrate.

11...c5
11...Bd5 12.Bd2!? (a decent alternative is 12.Nd2© intending Qg4) 12...Bxd2 13.Nxd2 0-0
(13...Bxa2 14.Qg4 Kf8 15.Ra1 Bd5 16.f4‚) 14.Qg4 Qd8 15.Ne4 Kh8 16.Rbe1 Nc6 (16...Bb7
17.Re3+–)

776
17.Nf6! gxf6 18.Qf4 Kg7 19.Re3 Rh8 20.exf6† Kf8 21.c4± Regaining the piece, with great
attacking chances against Black’s vulnerable king.

12.Nd2
A tempting alternative is 12.Rb3 Bxd4 13.Nxd4 cxd4 14.Bxd4 Nc6 15.Bb2© when White’s
bishops are superb and the queen will join the action next.

12...Bxd4
12...Bxd2 13.Bxd2 Nc6 14.Qg4 Kf8 15.c3© with great play for a pawn.

12...d5 13.Qg4 Kf8 14.f4 (14.dxc5 Nc6 15.f4 is also worth considering) 14...h5 15.Qg5 cxd4
enables White to press ahead with:

16.f5! Nc6 17.f6 g6 18.Bf4ƒ Once again, White’s attack is worth more than the small material

777
investment.

13.Nc4 0-0 14.Qg4

14...Qc6
14...Bxe3 15.fxe3!+– opens the f-file with decisive effect.

15.Qg3ƒ
Black remains under pressure on the kingside.

A222) 10...c5

Bauer gave this move without further analysis, evaluating the position as better for Black.

778
11.Nd2! Bxd2
Either capture on d4 would allow 12.Nc4 with a huge initiative for White.

12.Bxd2 Nc6!
12...cxd4 just opens up the dark squares, and 13.Qg4© gives White good play.

12...h5!? keeps the queen at bay for the time being, but 13.Re1 cxd4 14.Bb4 Qd8 15.Qd2 Bd5
16.Bd6 Nc6 17.Rb5 gives White a strong initiative.

13.Qg4 Kf8
13...Qc7!? is an attempt to bait White into opening the g-file for an attack, but 14.c3© keeps
the tension and ensures that Black’s king will not be safe on either flank.

14.c3

14...h5 15.Qf4 Ne7 16.dxc5 Qxc5 17.c4©

My impression in all these lines is that White has at least enough compensation for the pawn,
and Black will have the more difficult problems to solve.

A223) 10...Nc6

779
 This seems to me like the most natural continuation, as Black continues developing while
keeping the position closed and avoiding the creation of any new weaknesses.

11.Be3
In a note on page 415, we briefly considered the analogous position with Black’s queen on d8.
In the former case, White had fair compensation; but with the queen on the inferior c8-square, I
think it should be more.

11...d5!?
Preparing ...Qd7 and ...0-0-0 seems like as good a plan as any. Here are some alternatives I
considered:

11...Ba6 12.Ng5 Bxd3 13.Qxd3 Ba5 14.c4© with great play for the pawn.

11...h6 feels too slow, and 12.Rb3 Bb4 13.c4 Be7 14.d5 Na5 15.Rc3 exd5 16.cxd5 Bxd5
17.Nd4ƒ gives White tremendous activity.

11...Nb4 is well met by 12.Ng5! Nxd3 13.cxd3 when the awkward placement of Black’s queen
is clearly felt. Play could continue: 13...h6 14.Ne4 Ba5 (or 14...Bxe4 15.dxe4© when Qg4 and
f2-f4-f5 are coming) 15.Qg4ƒ

11...d6 12.Rb3! Ba5 13.exd6 cxd6

Had the queen been on d8, she would have been able to recapture on d6, preventing what
follows.

780
14.d5! exd5 15.Nd4 0-0
15...Nxd4 16.Bxd4 0-0 gets smashed by: 17.Bxh7† Kxh7 18.Qh5† Kg8 19.Bxg7 Kxg7
20.Qg5† Kh7 21.Rg3+–
15...g6 seems like Black’s best, although 16.Bb5 0-0 17.Bh6± leaves him under a great deal
of pressure.
16.Bf5 Qc7

17.Nxc6 Bxc6 18.Bd4!+–

This illustrative line reflects White’s plans perfectly. Almost all his pieces are ready to strike
on the kingside, while Black’s bishops are totally out of play.

781
12.Qc1!
Another nice benefit of having provoked ...Qc8 is that Qa3 becomes an annoyingly powerful
threat which Black must worry about. If we can force Black to play ...Ba5, then White can
simply play c2-c3 and continue building up slowly. Both of Black’s bishops would be stuck on
the queenside and it is not clear how he could make up for this problem.

12.exd6?! cxd6 13.Rb3 is a possible attempt to play the way we did against 11...d6, but here the
c-file has been opened and 13...Na5! is annoying.

12.Rb3 Bb4 13.Ng5 is a decent alternative, intending various queen jumps and/or f4-f5. White
looks to have good compensation, but I like the Qc1 plan even more.

12...Qd7
12...Bb4 13.c4! Be7 14.cxd5 exd5 15.Bb5 leaves Black’s position on the verge of collapse:
15...Qe6

782
16.Bg5! 0-0 17.Bxe7 Nxe7 18.Qxc7+–

12...a5 does not lose immediately but is nevertheless a concession, as Black’s king can no longer
feel safe on the queenside; and ...0-0 was already looking suspicious to say the least. A plausible
continuation is: 13.Qa3 Bb4 14.Qa4 0-0 15.c3 Be7

16.Rfe1 Qd8 17.Qc2 h6 18.Qd2± The threat of Bxh6 forces a further weakening of Black’s
kingside.

13.Qa3 Ba5
There are many possible ways to continue for White. I am extremely optimistic about White’s
chances, and it all boils down to the fact that we don’t need to rush to prove our compensation:
Black’s king will not be safe on either flank, and the dark-squared-bishop looks extremely
awkward on a5.

783
14.Rfc1 f6!?
It seems reasonable to challenge for the centre and see how White reacts before deciding where
to put the king. Here are a few other illustrative lines:

14...Ne7 can be met by: 15.c3 a6 (or 15...0-0-0 16.Nd2 intending Nb3) 16.Nd2 Bc6

17.Qb2! Patiently regrouping. (17.Rd1 Bb5 and 17.Nb3 Bb5 18.c4 dxc4 are not so clear)
17...Bb5 18.Bc2 Nf5 19.Bf4© The comments from the note to move 13 still apply.

Another natural try is:

14...0-0-0 15.c3 f6
15...Kb8 16.Nd2 f6 17.Nb3 fxe5 18.Bb5±
16.Qb2

784
16...Qf7
16...fxe5 17.Bb5! Qd6 18.Bxc6 Bxc6 19.Nxe5± is great for White.
16...a6 prevents the pin but 17.Qe2± produces an awkward attack on a6.
17.Bb5 Qg6
17...Ne7 18.Nd2±
18.Qe2 Kb8 19.Bf4 f5 20.Nd2±
Followed by Nb3.

15.c4!
15.c3 would be a good move in general, but 15...fxe5 16.dxe5 Rf8! leads to some immediate
annoyances in the centre due to the ...Rxf3 threat.

15...dxc4 16.Bxc4©
White’s pieces, especially the rooks, seem well suited to the opening of the position.

B) 1...a6

This is the St. George Defence, a pretty dubious opening.

2.d4 b5
This is the most popular and consistent move, but Black is simply heading for a worse version
of the 1...b6 lines, for two reasons: firstly, he is making two pawn moves rather than one to get
his bishop to b7, giving us a spare tempo to build a lead in development; and secondly, the b5-
pawn is an easy target for us to attack with a timely a2-a4 – especially since White has not even
committed his knight to c3.

2...h6 has been played by Basman in several games, but after 3.Nf3 it is not clear what Black

785
should be doing, as it would be too risky to try and ‘exploit’ White’s last move by playing ...g5.
A sample line is3...d6 4.Nc3 e6 5.Bd3 Nf6 6.0-0 Be7 (6...c5 7.dxc5 dxc5 8.a4±) 7.Qe2± These
moves are not forced of course, and after starting the game with ...a6 and ...h6 Black could try
just about anything – but in general, simple development should give White a great game.

Finally, I considered:
2...d6 3.Nc3 e6
3...g6 converts to variation B2 of Chapter 11 on page 252.
3...b5 only weakens Black’s structure without giving him any additional options. After
something lik4.Nf3 Bb7 5.Bd3 e6 6.0-0 Nf6 7.Qe2 Be7 8.e5 Nfd7 9.a4 b4 10.Ne4±
everything is flowing smoothly for White.
4.Nf3
There is no reason to go for ideas involving Be3, Qd2 and 0-0-0, as this might give Black
chances to justify his set-up with ...b5 and so on. Instead we should simply develop and
castle on the kingside, after which ...b5 will give us new weaknesses to target, and we can
consider e4-e5 whenever it’s appropriate.
4...Nf6 5.Bd3

5...Be7
5...b5 is well met by: 6.e5 dxe5 7.dxe5 Nd5 8.a4 b4 (8...Nxc3 9.bxc3±) 9.Ne4±
6.0-0 0-0 7.Qe2±
White could be ready for e4-e5 any time now – although there is also the option of continuing
development with moves like Bf4, h2-h3 and Rad1.

3.Bd3 Bb7 4.Nf3 e6 5.Qe2 Nf6

The move order can vary, but Black is highly likely to play these moves in one order or another
in the St. George.

786
6.0-0
White has two other equally valid move orders:

6.a4 b4 7.e5 Nd5 8.0-0 should transpose to one of the lines below after 8...c5 or 8...Be7; the
latter was the actual move order of the Jakubek – Chetverik game referred to in the note to
Black’s next move.

6.e5 Nd5 7.0-0 Be7 (7...c5 8.dxc5 Bxc5 transposes to our main line below; 7...Nb4 8.Be4±) 8.a4
bxa4

This occurred in Ievlev – Maeckelbergh, Dos Hermanas 2004, when instead of recapturing on
a4 immediately, 9.c4!N 9...Nb4 10.Be4 would have given White an overwhelming advantage.

6...c5

787
 This has been Black’s usual choice.

The main alternative is:

6...Be7 7.a4 b4
7...bxa4?! 8.e5 Nd5 9.c4 Nb4 10.Be4+–
8.e5 Nd5 9.c4 bxc3 10.bxc3± f5?!
It is understandable that Black wished to gain some breathing room on the kingside, but this
is too weakening.
11.exf6 Nxf6

12.Nbd2!N
Simply bringing all the pieces into play before launching an attack.
12.Bg5 Nc6 13.Ne5 Nxe5 14.dxe5 Nd5 gave White nothing much in Jakubek – Chetverik,
Stare Mesto 2008.
12...Nc6
12...0-0 13.Rb1 also looks horrible for Black.
13.Ne4 Nxe4 14.Bxe4 0-0 15.Qd3 h6 16.Rb1 Rb8 17.d5 Na5 18.c4+–
Black’s position is a shambles.

7.dxc5!
One of the ‘rules’ of these positions is that White should usually meet ...c5 in this way. It may
seem like some kind of concession to allow Black to develop his bishop with tempo, but it is
worth it to obtain a favourable pawn structure with clearly defined plans.

7...Bxc5 8.e5 Nd5 9.a4 b4 10.c4!

It is worth weakening the pawn structure a little in order to swap off Black’s active knight and
clear some lines for attacking purposes.

788
10...bxc3 11.Nxc3 Nxc3 12.bxc3 Qc7

13.Ng5!N
This significantly improves upon 13.Bf4, when 13...f5!÷ was not bad for Black in Roesch –
Chetverik, Metz 2013.

13...Nc6 14.Rb1 Be7 15.f4±

White has a lot of activity and attacking potential. Black is currently unable to castle, and the
attempt to solve that problem only makes matters worse:

15...Bxg5? 16.fxg5 0-0 17.Ba3+–

White wins at least an exchange for starters, as moving the rook from f8 would allow an instant
kill with 18.Bxh7†.

C) 1...h6?! 2.d4 g5?!

Objectively Black’s best approach would be to play some other move that doesn’t create massive
weaknesses on the kingside – but in that case there would have been no reason for the early ...h6.

3.Bd3
White has many ways to continue of course, but the simplest seems to be to prepare Ne2, 0-0
and perhaps c2-c3 to secure the centre, before going on the offensive with f2-f4. The same kind
of plan but with the bishop on c4 would also work well.

3...Bg7
It would seem strange for Black to advance the h- and g-pawns only to leave the bishop on f8,
but here is an example where he did exactly that: 3...d6 4.Ne2 c5 5.c3 Nc6 6.0-0 Nf6 7.Nd2

789
(7.h3!?N prepares Be3, Nd2 and f2-f4, just as in our main line below, to which we could easily
transpose after ...Bg7) 7...Qc7

8.b4!? Playing on the queenside seems like another attractive option. 8...b6 9.bxc5 bxc5 10.Rb1
Bd7 11.Nc4 White was clearly better in Nunn – Basman, Brighton 1980.

4.Ne2 d6
4...c5 5.dxc5 seems the simplest – White gets a massively improved Sicilian with g5 being a
constant target. 5...Qa5† 6.Nbc3 Qxc5 7.Be3 Qa5 In Pritchett – Basman, Torquay 1982, White
should have played:

8.h4!N 8...g4 (8...gxh4 9.Qd2±) 9.Bd2!?± With ideas of Nd5 and Bc3.

5.0-0 c5!?N
5...Nc6 6.c3 e5 has been played a few times but 7.d5 Nce7 8.Ng3 leaves Black’s light squares

790
looking especially sickly. The text move seems like the most natural attempt to make sense of
Black’s position, but the kingside weaknesses are not going away.

6.c3 Nc6 7.Be3 Nf6 8.h3!?

There is no point in allowing our bishop to be hunted down, as occurs after 8.Nd2 Ng4!.

8...Nh5
There is no need to fear 8...g4, as 9.h4 leaves Black with nothing on the kingside. 9...Nh5
10.g3 b6 looks logical, but then 11.Na3 Bb7 12.b4! borrows an idea from the Nunn – Basman
game, with good prospects for White.

8...Qc7 9.Nd2 Bd7 10.f4± looks great for White; and 8...0-0 looks risky in view of 9.Ng3
followed by Nd2 and f2-f4.

9.Nd2 0-0 10.Nb3 b6 11.g4!?

11.Bb5 Bd7 12.f4 is a good alternative.

11...Nf6 12.Ng3
White is dominating on both flanks. His immediate plans include f2-f4 and/or Nf5; and if
Black tries to keep the kingside closed with ...e5, White can establish a space advantage with d4-
d5 and then start prising open the queenside as well.

Conclusion

Owen’s Defence with 1.e4 b6 has been by far the most important and challenging topic of this
final chapter. After 2.d4 Bb7 3.Nc3 e6 4.Nf3 Nf6 5.Bd3 Bb4 I investigated two interesting
directions. 6.Qe2 d5 7.0-0!? Bxc3 8.e5 is a remarkable idea which would likely catch most

791
players by surprise; Black can remain a pawn up in various ways, but in all cases White has quite
promising compensation. We then turned to 6.e5!?, which is where I focused most of my
attention. The critical continuation is 6...Ne4 7.0-0 Nxc3 8.bxc3 Bxc3, when 9.Bg5!?N is a
promising new idea. I investigated a number of variations but did not find clear equality for
Black anywhere, and in a practical game he would have to make a number of tricky decisions.

Finally we looked at the deservedly rare 1...a6?! and 1...h6?!. In each case, Black prepares to
advance the respective knight’s pawn two squares, but he only succeeds in wasting time and
giving White a target to attack.

792
 Variation Index

Chapter 1
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3
A) 4...Bf5!? 9
B) 4...c6 10
C) 4...Nb6!? 5.a4 a5 6.Nc3 11
C1) 6...Bf5 13
C2) 6...g6 7.exd6 14
C21) 7...exd6 14
C22) 7...cxd6 15
D) 4...Nc6!? 5.c4 Nb6 6.e6 fxe6 7.Nc3 g6 8.h4 Bg7 9.Be3 e5 10.d5 Nd4 11.Nxd4 exd4
12.Bxd4 Bxd4 13.Qxd4 e5 14.Qe3 17
D1) 14...Qe7 22
D2) 14...Bf5 24
D3) 14...0-0 25

Chapter 2
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 g6 5.Bc4 Nb6 6.Bb3
A) 6...Nc6 28
B) 6...Bg7 7.Ng5 e6 8.Qf3 30
B1) 8...0-0 9.Qh3 h6 10.Nf3 dxe5 11.dxe5 30
B11) 11...c5 31
B12) 11...Nc6 32
B2) 8...Qe7 9.Ne4 32
B21) 9...h6 33
B22) 9...dxe5 10.Bg5 Qb4† 11.c3 Qa5 12.dxe5!N N8d7 13.Nbd2! Nxe5 14.Qg3
f5 15.0-0-0! 0-0 16.f4 35

793
 B221) 16...Ned7 38
B222) 16...fxe4! 39

Chapter 3
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 Bg4 5.Be2
A) 5...c6 44
B) 5...e6 6.0-0 Be7 7.c4 Nb6 8.Nc3 0-0 9.h3 Bh5 10.Be3! 46
B1) 10...Nc6 48
B2) 10...N8d7 49
B3) 10...d5 11.c5 Bxf3 12.gxf3 Nc8 51
B31) 13.h4!?N 53
B32) 13.f4! 55
B321) 13...Bh4 55
B322) 13...Nc6 57

Chapter 4
1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 dxe5 5.Nxe5
A) 5...Nd7? 6.Nxf7! Kxf7 7.Qh5† Ke6 8.c4 N5f6 9.d5† Kd6 10.Qf7 Ne5 11.Bf4 c5 12.Nc3
a6 13.b4! 64
A1) 13...Qb6 66
A2) 13...b6 68
B) 5...g6 6.Bc4 69
B1) 6...Be6 70
B2) 6...c6 7.0-0 Bg7 8.Re1 72
B21) 8...0-0 72
B22) 8...Nd7!? 75
C) 5...c6 6.Be2 77
C1) 6...g6 77
C2) 6...Nd7 79
C3) 6...Bf5 7.0-0 Nd7 8.Nf3 e6 9.c4 N5f6 10.Nc3 81

794
 C31) 10...Bd6 83
C32) 10...Ne4!? 11.Re1!? 89
C321) 11...Be7N 91
C322) 11...Bd6 93

Chapter 5
1.e4 d5 2.exd5 Nf6 3.d4
A) 3...Nxd5 4.c4 97
A1) 4...Nf6?! 98
A2) 4...Nb6 5.Nf3 99
A21) 5...Bg4 100
A22) 5...g6 6.Nc3 Bg7 7.c5! Nd5 8.Bc4 101
A221) 8...Nxc3 102
A222) 8...c6 104
B) 3...Bg4 4.Bb5†!? 105
B1) 4...c6!? 106
B2) 4...Nbd7 5.Be2 Bxe2 6.Qxe2 Nxd5 7.c4! N5f6 8.Nf3 e6 9.Nc3!? 109
B21) 9...Bd6 111
B22) 9...Bb4!? 113
B221) 10.Bd2!? 114
B222) 10.0-0 115

Chapter 6
1.e4 d5 2.exd5 Qxd5 3.Nc3
A) 3...Qd8 120
B) 3...Qa5 4.d4 123
B1) 4...c6 124
B2) 4...Nf6 5.Bd2 c6 6.Bc4 126
B21) 6...Bg4?! 127
B22) 6...Qb6?! 128

795
 B23) 6...Qc7 131
B24) 6...Bf5 7.Nd5 Qd8 8.Nxf6† gxf6 133
B241) 9.Qe2!? 134
B242) 9.Qf3! 137
B2421) 9...Bxc2N 137
B2422) 9...e6 140

Chapter 7
1.e4 d5 2.exd5 Qxd5 3.Nc3 Qa5 4.d4 Nf6 5.Bd2 Bg4 6.f3
A) 6...Bf5 145
B) 6...Bd7 7.f4!?N 147
B1) 7...e6 148
B2) 7...a6!? 8.Nf3 Qb6 9.Bc4 Nc6 10.Na4! Qa7 11.Nc5 152
B21) 11...Bg4 153
B22) 11...Bf5!? 154
B3) 7...Bg4!? 8.Nf3 e6 156
B31) 9.Bb5†!? 157
B32) 9.h3 158

Chapter 8
1.e4 d5 2.exd5 Qxd5 3.Nc3 Qd6 4.d4
A) 4...c6 164
B) 4...Nf6 5.Nf3 166
B1) 5...Bg4 6.h3 166
B11) 6...Bxf3 166
B12) 6...Bh5 169
B2) 5...a6 6.g3 Bg4 7.h3! 173
B21) 7...Bxf3 175
B22) 7...Bh5 177

796
Chapter 9
1.e4 d5 2.exd5 Qxd5 3.Nc3 Qd6 4.d4 Nf6 5.Nf3 g6 6.Bg5!? Bg7 7.Qd2
A) 7...a6!? 8.Bf4 Qd8 183
A1) 9.0-0-0 184
A2) 9.Be2!? 186
A21) 9...0-0 186
A22) 9...b5 188
B) 7...0-0 8.0-0-0 191
B1) 8...Bf5 192
B2) 8...Bg4 9.Be2 193
B21) 9...c6 194
B22) 9...Nbd7!?N 195
B3) 8...a6 9.Bf4! 197
B31) 9...Qd8 10.h4! 197
B311) 10...Bg4 198
B312) 10...h5 199
B32) 9...Qb6! 10.Ne5!? 200
B321) 10...Nc6 201
B322) 10...Nbd7N 202

Chapter 10
1.e4 d5 2.exd5 Qxd5 3.Nc3 Qd6 4.d4 Nf6 5.Nf3 c6 6.Ne5
A) 6...Bf5 208
B) 6...Be6!? 7.Bf4 Qd8 8.Qd2 Nbd7 9.0-0-0 209
B1) 9...Nxe5 210
B2) 9...g6 10.Bc4!? Bxc4 11.Nxc4 212
B21) 11...Bg7 212
B22) 11...Nb6N 213
C) 6...Nbd7 7.Nc4 Qc7 8.Qf3 Nb6 9.Bf4 216

797
 C1) 9...Qd8 10.Be5 216
C11) 10...Nbd5 217
C12) 10...Bg4 11.Qg3 219
C121) 11...h5?! 219
C122) 11...Be6 221
C13) 10...Be6 11.Ne3!? 224
C131) 11...Nbd5 225
C132) 11...g6!? 227
C2) 9...Qd7 10.Nxb6 axb6 11.0-0-0 231
C21) 11...Qd8 232
C22) 11...e6 12.Bc4! 235
C221) 12...Nd5 235
C222) 12...Be7 13.g4!?N 237
C2221) 13...b5 239
C2222) 13...Nd5!? 241

Chapter 11
1.e4
A) 1...d6 2.d4 Nf6 3.Nc3 c6 4.Nf3 245
A1) 4...Bg4!? 245
A2) 4...Qa5 247
B) 1...g6 2.d4 249
B1) 2...Nf6 250
B2) 2...d6 252
B3) 2...Bg7 3.Nc3 253
B31) 3...Nc6 255
B32) 3...d5!? 257
B33) 3...c6!? 4.Be3 d5 5.e5! 260
B331) 5...Nh6 260

798
 B332) 5...Qb6 262
B333) 5...f6 263
B34) 3...d6 266

Chapter 12
1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7 5.Qd2 c6 6.Nf3
A) 6...Bg4 275
B) 6...Qa5 277
C) 6...0-0 279
C1) 7.h3!? 279
C2) 7.Bd3 283
D) 6...b5 7.Bd3 Nbd7 288
D1) 8.0-0 291
D2) 8.Ne2! 294
D21) 8...c5!?N 296
D22) 8...Qc7 297
D23) 8...Bb7 299
D24) 8...0-0 9.0-0 Qc7N 10.h3 300
D241) 10...e5 301
D242) 10...c5 302

Chapter 13
1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 c6 5.Qd2
A) 5...Nd7 6.Nf3 308
A1) 6...e5 308
A2) 6...Qc7!? 309
B) 5...b5 6.Bd3 Nd7 7.Nf3 312
B1) 7...Nb6 313
B2) 7...Bb7 314
B3) 7...Qc7 317

799
Chapter 14
1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7 5.Qd2 a6 6.Nf3
A) 6...0-0 7.Bh6 Bg4!? 8.0-0-0 Nc6 9.h4!N 321
A1) 9...Bh5 322
A2) 9...e5 323
B) 6...b5 7.Bh6 0-0 8.Bd3 Nc6! 324
B1) 9.d5!? 326
B11) 9...Bxh6!?N 326
B12) 9...Ne5 327
B13) 9...Nb4!?N 10.h4 331
B131) 10...Nxd3† 331
B132) 10...e5! 333
B2) 9.Bxg7 334

Chapter 15
1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.Nf3
A) 5...Nd7!? 6.Bd3 340
A1) 6...e6 340
A2) 6...Ngf6 342
A3) 6...c5 343
B) 5...b5 6.Bd3 Nd7 344
B1) 7.e5 345
B2) 7.h4! h6 8.e5 e6! 9.Qd2!N Bb7 10.0-0-0 Ne7 11.Kb1! dxe5 12.dxe5 Nd5 13.Nxd5
Bxd5 14.h5 g5 15.Nxg5! 347
B21) 15...hxg5 350
B22) 15...Nxe5! 16.Ne4 Qe7! 352
B221) 17.Nc5 Ng4 353
B2211) 18.f3 353
B2212) 18.c3 355

800
 B222) 17.Bc5 356

Chapter 16
1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.Nf3 b5 6.Bd3 Bb7
A) 7.e5 360
A1) 7...Nd7 360
A2) 7...c5! 363
B) 7.h4! 365
B1) 7...h6 366
B2) 7...Nf6 8.e5 368
B21) 8...dxe5 369
B22) 8...Ng4!?N 372
B3) 7...b4!?N 8.Ne2 Nf6! 9.Ng5!? 377
B31) 9...h6 378
B32) 9...e6 380
B4) 7...Nd7 8.h5 382
B41) 8...c5 382
B42) 8...Ngf6! 9.h6 Bf8 10.Bg5!N 384
B421) 10...c5 385
B422) 10...b4 387

Chapter 17
1.e4 Nc6 2.Nf3!
B) 2...g6 397
C) 2...Nf6 398
D) 2...d5 3.exd5 Qxd5 4.Nc3 399
D1) 4...Qa5 400
D2) 4...Qd6 401
E) 2...d6 3.d4 403
E1) 3...e5 404

801
 E2) 3...Nf6 4.Nc3 405
E21) 4...Bg4 406
E22) 4...g6 407

Chapter 18
1.e4

A) 1...b6 2.d4 Bb7 3.Nc3 e6 4.Nf3 Nf6 5.Bd3 Bb4 410

A1) 6.Qe2 411
A2) 6.e5!? Ne4 7.0-0 Nxc3 8.bxc3 Bxc3 9.Bg5!?N 414
A21) 9...Bxf3!? 416
A22) 9...Qc8 10.Rb1 417
A221) 10...h6 417
A222) 10...c5 418
A223)10...Nc6 419
B) 1...a6?! 422
C) 1...h6?! 424

802
Table of Contents
Title Page 3
Key to symbols used & Bibliography 4 5
Preface 5 7
1 Rare Lines 7 9
2 4...g6 27 46
3 4...Bg4 43 75
4 4...dxe5 63 112
5 2...Nf6 96 173
6 2...Qxd5 119 216
7 5...Bg4 144 264
8 3...Qd6 163 300
9 5...g6 181 335
10 5...c6 207 381
11 Early Deviations 244 453
12 Pirc with ...c6 270 502
13 Modern with ...c6 306 570
14 Pirc with ...a6 320 595
15 Modern with ...a6 339 631
16 Main Line with 6...Bb7 359 668
17 1...Nc6 394 733
18 1...b6 and Others 409 761
Variation Index 426 793

803