Portland Homicide

Problem Analysis
2019-2021

Portland Police Bureau

California Partnership for Safe Communities

1
Table of Contents

1. PROBLEM ANALYSIS: INTRODUCTION

2. CONTEXT AND TREND DATA
3. DEMOGRAPHICS, CRIMINAL HISTORY, AND CRIMINAL
JUSTICE SYSTEM INVOLVEMENT
4. GUN VIOLENCE INCIDENT ANALYSIS
5. HIGH RISK GROUPS AND NETWORKS
6. SHOOTING DENSITY ANALYSIS
7. DISCUSSION OF FINDINGS AND RECOMMENDATIONS

2
 Problem Analysis:
Introduction and Overview
 Introduction and Overview
• A “problem analysis” is a set of data collection and analysis
exercises designed to support the implementation of violence
reduction strategies, including Ceasefire. This methodology has
been developed and refined over the last 20 years, and used in
dozens of cities nationally.
• This analysis establishes a common understanding of the local
violence problem that guides and informs the work of civic,
community, and criminal justice leaders to reduce violence.
• The problem analysis identifies the groups and individuals within a
community who are at greatest risk of violence and helps tailor an
intervention to reduce that risk.
• Though the methodology is informed by research, the problem
analysis is primarily a practice document with implications for local
policy.

4
 Introduction and Overview

• This analysis examined all homicides in the City of Portland from

January 2019 – June 2021 (n = 117), and all nonfatal injury
shootings between January 2019 – December 2021 (n = 314).

• The 117 homicides involved 214 unique victims or identified

suspects, and the 314 shootings involved 462 unique identified
victims or suspects. This analysis examines the characteristics of
these incidents and the involved individuals, including demographics
and criminal justice system involvement.

• The analysis also includes information on groups, gangs and

networks at high risk of involvement in violence, including their size,
relationships, turfs, and activities as well as the spatial concentration
of violence throughout the city.

5
Social Networks Bring Risk
Research increasingly shows that the concentration of violence within
social networks is a key factor in addressing gun violence.

• The risk of future gun shot victimization is highly influenced by the

composition of an individuals’ social network.

• The more someone is connected to individuals who are victims or

perpetrators of shootings, the higher their risk.

• The average time to a retaliation shooting once someone is “infected”

is about 125 days.

Consistent with the national research, we found that social

networks are a key predictor of risk of being involved in gun
violence in Portland.

6
Context and Trend Data
 Portland Homicides: 2002 - 2021
100

90
88

80

70

57
60

50

40 36
29 26
30 27 26 26
20 20 22 22 20 24
20 20 20 year average = 28
19
20
14 14

10

0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2016 2017 2018 2019 2020 2021

8
 Portland Homicides: 2012 - 2021
100

88
90

80

70
57
60

50

40 36
10 year average = 34

30 26 24 26

20
20 14 14

10

0
2012 2013 2014 2016 2017 2018 2019 2020 2021

9
 Portland Homicide Rate (per 100,000 population):
2002 - 2021
16.0
Portland

14.0 Oregon 13.5

United States
12.0

8.8
10.0

8.0

6.0 5.3
5.0
4.1 4.2 4.4
3.7 3.7 3.7 4.7 3.7 3.8
4.0 3.4 3.9 3.4 4.7
2.3 2.2
2.0

0.0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2016 2017 2018 2019 2020 2021

10
 Portland Nonfatal Shootings: 2019 - 2021
400

350 334

300

250
218

200

150

98
100

50

0
2019 2020 2021

11
 Portland Homicide Rate (per 100,000 population):
Peer Comparison Cities

2019 Homicide Population Median Unemployment Percent Below

Rate Density Household Rate Poverty Line
(per 100,000 pop.) (per sq. mi) Income

Portland, OR 4.4 4,821 $73,159 5% 13%

Atlanta, GA 13.9 3,732 $64,179 7% 19%

Denver, CO 9.2 4,590 $72,661 4% 12%

Minneapolis, MN 10.7 7,876 $66,068 5% 18%

Nashville, TN 12.1 1,346 $62,087 5% 14%

San Francisco, CA 4.5 17,345 $119,136 5% 10%

12
 Portland Homicide Rate (per 100,000 population):
2010 – 2020 with Peer Comparison Cities
35.0
Portland
Atlanta

30.0 Denver
Minneapolis
Nashville
25.0 San Francisco

20.0

15.0
13.5

10.0 8.8

5.3
4.4
5.0 3.9 3.4 3.3 3.7
2.3
4.2 3.8
2.2
0.0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021

13
 Summary Observations:
• From 2019 – 2021, the City of Portland experienced a:
• 144% increase in counts of homicide incidents
• 241% increase in counts of nonfatal injury shootings

• Relative to five selected peer comparison cities,* Portland

experienced the largest increase in homicide rate from 2019 – 2021:
• Portland: 207% increase
• Minneapolis: 104% increase
• Atlanta: 54% increase
• San Francisco: 53% increase
• Denver: 47% increase
• Nashville: 23% increase

*Minneapolis, Atlanta, Nashville, and Denver selected using the Federal Reserve Bank of Chicago’s Peer City Identification Tool. San
Francisco selected based on comparable homicide rate trends

14
 Victims and Suspects of
Homicides and Shootings:
Demographics
Victims and Suspects of Homicides & Shootings:
Sex and Race
Homicides Nonfatal Shootings All Victims
and
Victims Suspects Victims Suspects Suspects Portland
(n=126) (n=88) (n=385) (n=77) (n=676) Population
Sex

Male 83.3% 90.9% 84.2% 94.8% 86.1% 49.6%

Female 16.7% 9.1% 15.8% 5.2% 13.9% 50.4%

Race

Asian 3.2% 0.0% 2.4% 1.3% 2.1% 8.7%

African-American 38.9% 41.2% 49.2% 57.9% 47.2% 5.9%

Hispanic 7.9% 9.4% 12.3% 6.6% 10.5% 9.8%

Caucasian 46.0% 45.9% 32.5% 30.3% 36.5% 75.3%

Other 4.0% 3.5% 3.7% 3.9% 3.7%

16
Victims and Suspects of Homicides & Shootings:
Age
Homicides Nonfatal Shootings All Victims
and
Victims Suspects Victims Suspects Suspects
(n=126) (n=88) (n=385) (n=77) (n=676)
Age

17 and under 3.2% 9.2% 4.7% 7.8% 5.3%

18-24 23.2% 31.0% 26.2% 29.9% 26.7%

25-34 26.4% 31.0% 28.6% 37.7% 29.5%

35-44 20.8% 12.6% 24.4% 19.5% 21.7%

45-54 13.6% 6.9% 9.4% 2.6% 9.1%

55 and older 12.8% 9.2% 6.8% 2.6% 7.7%

Mean Age 36.8 31.5 33.1 29.1 32.9

17
 Age:
All Known Victims and Suspects (n = 676)
January 2019 – June 2021
55 and older 17 and under
52 (8%) 36 (5%)
45-54
61 (9%)

18-24
180 (27%)

35-44
146 (22%)

25-34
199 (29%)
18
 Summary Observations:
Characteristics of Those Involved in Violence
• Victims and suspects of homicides and shootings in Portland are
primarily African-American (47.2%) and Caucasian (36.5%) men
(86.1%).
• The overall average age of victims and suspects is 32.9, and about
78% are between the ages of 18-44.
• Victims tend to be slightly older than suspects, and those involved in
homicides as victims or suspects are more likely to be white.
Suspects are slightly more likely to be male when compared to
victims.
• Relative to victims and suspects of homicides and shootings 2015-
2019, there are no statistically significant differences in the race,
sex, or age of victims and suspects 2019-2021.

19
 Victims and Suspects of
Homicides and Shootings:
Criminal Justice System Involvement
Victims and Suspects:
Criminal Justice System Involvement
Homicides Nonfatal Shootings All
Victims
Victims Suspects Victims Suspects and
(n=126) (n=88) (n=385) (n=77) Suspects
(n=676)
Known to the CJ system prior to the incident 51* 63 267 59 440
(71.6%) (69.4%) (76.6%)
Of those known to the CJ System:
Average age 40.1 33.2 33.9 30.7 33.8

Average number of prior arrests 8.1 5.9 8.6 8.0 8.1

Average number of prior felony arrests 3.9 2.9 4.7 4.3 4.3

Prior probation/post-prison supervision 58.8% 54.0% 63.7% 60.3% 61.4%

Active probation/post-prison supervision 3.9% 9.5% 10.5% 10.3% 9.1%

Prior incarceration 45.1% 58.7% 65.5% 65.5% 62.5%

Convicted of felony 43.1% 57.1% 58.1% 67.2% 57.9%

*Prior criminal involvement was inconsistently available due to limitations in statewide access to criminal histories for
homicide victims.

21
 Victims and Suspects of Homicides & Shootings:
Criminal Justice System Involvement
100%

90%

80%

70%
61.4% 62.5%
60% 57.6%

50%

40%

30%

20%

9.1%
10%

0%
Prior Probation Active Probation Prior Incarceration Convicted of Felony

22
Victims and Suspects:
Criminal Histories – Average Prior Offenses
Homicides Nonfatal Shootings All
Victims
Victims Suspects Victims Suspects and
Suspects

Of those known to the CJ System: (51) (63) (267) (59)

Armed violent offenses 0.6 0.2 0.2 0.2 0.2
Unarmed violent offenses 1.0 1.4 1.8 1.6 1.6
Nonviolent firearm offenses 0.5 0.6 0.7 0.9 0.7
Property offenses 1.8 1.7 2.0 2.3 2.0
Drug offenses 2.4 1.0 2.1 1.7 1.9
Disorder offenses 1.9 1.3 2.0 1.4 1.9
Sex industry offenses 0.3 0.0 0.1 0.0 0.0
Fraud offenses 0.4 0.1 0.2 0.2 0.2
Other offenses 1.0 0.5 1.2 0.8 1.1
Total 9.2 6.8 10.4 9.2 9.6

23
 Victims and Suspects of Homicides & Shootings:
Criminal Histories – Average Prior Offenses
2.5

2.1
2.0 2.0 2.0
2.0

1.6
1.5 1.5
1.5 1.4

1.2

1.0

0.7
0.6 0.6

0.5

0.2 0.2 0.2

0.1 0.1
0.0
0.0
Armed Unarmed Nonviolent Property Drug Disorder Sex Industry Fraud Other
violent violent Firearm
All Victims All Suspects

24
Summary Observations:
Criminal Justice System Involvement of Victims and Suspects
of Homicides & Shootings

• About 70% of victims and suspects had prior criminal justice system
involvement.

• Just over 61% of victims and suspects had prior probation or post-
prison supervision and about 9% were on active probation or post-
prison supervision.

• About 58% of victims and suspects had a prior felony conviction.

25
Summary Observations:
Criminal Justice System Involvement of Victims and Suspects
of Homicides & Shootings
• On average, homicide victims had been arrested 8.1 times for 9.2
different criminal offenses and shooting victims has been arrested
8.6 times for 10.4 different criminal offenses prior to the incident.

• Homicide suspects had been arrested an average of 5.9 times for

6.8 different criminal offenses and shooting suspects had been
arrested 8.0 times for 9.2 different criminal offenses prior to the
incident.

• Most prior arrests were for property, drug, disorder, and unarmed
violent offenses, respectively.

26
Summary Observations:
Criminal Justice System Involvement of Victims and Suspects
of Homicides & Shootings
• Victims and suspects from 2019-2021 were significantly more likely to
have been on prior probation/post-prison supervision (61%) when
compared to victims and suspects from 2015-2019.

• Victims and suspects 2019-2021 were significantly less likely to have

been on active probation/post-prison supervision (9%) at the time of the
incident when compared to victims and suspects from 2015-2019
(18%).
• However, this difference may emerge from data limitations wherein we
could not collect victim/suspect supervision status from detectives in
unreviewed cases.

• Overall, victims and suspects 2019-2021 had similar arrest histories

and criminal justice system involvement to victims and suspects 2015-
2019.

27
Incident Analysis & Group Involvement
Data Note:
• The City of Portland experienced a rapid increase in fatal and nonfatal
shootings beginning in the summer of 2020 amidst a decrease in PPB
staffing and resources. PPB personnel resources were also necessarily
reallocated to respond to ongoing protest activity.

• Due to these staffing and resource constraints, data on shooting incidents

and investigative activity throughout 2020-2021 is limited.

• The PPB faces continued resource limitations and sustained increases in

gun violence that have restricted their ability to pull and review detailed
reports that would have been necessary for a thorough evaluation of all
gun violence incidents for this project.
Data Note:
• The PPB accommodated our research team however they could, but this
data is limited in that the motives for many homicides and shootings
remain unknown.

• Beyond incident motives, former/current members of the GVRT/ECST

dedicated time to review cases and share their knowledge about
group/gang associations.

• While it is difficult to draw conclusions about incident motives in this

dataset, we are confident that this data is accurate in reflecting group and
gang involvement in homicides and nonfatal shootings.
Group Member Involved Homicides

Unknown
43 (37%) Yes
48 (41%)

No
26 (22%)
 Group-Involved Victims & Suspects:
Homicides

Victims (n=126)) Suspects (n=88)

Yes
Yes 22 (25%)
35 (28%)

No No
58 (46%) 39 (44%)

Unknown
Unknown
27 (31%)
33 (26%)
Homicide Cause of Death

Total Group-Involved
(n=117) (n=48)

Frequency Percent Frequency Percent

Gunshot 88 75.2% 46 52.3%

Stabbing 13 11.1% 1 7.7%

Blunt trauma 7 6.0% 0 -

Strangulation 3 2.6% 0 -

Other 3 2.6% 1 33.3%

Unknown 3 2.6% 0 -
Group Member Involvement:
Gun Homicides (n = 88)

Unknown
29 (33%)

Yes
46 (52%)

No
13 (15%)
 Homicide Circumstances and Group Involvement
Total % Group Member
(n=117) Involved
Frequency Percent Frequency Percent
Instant dispute 16 13.7% 7 43.8%
Personal dispute 15 12.8% 6 40.0%
Group-related dispute 8 6.8% 8 100%
Domestic dispute 6 5.1% 2 33.3%
Robbery/Carjacking 4 3.4% 2 50.0%
Burglary/Larceny 2 1.7% 0 -
Mental illness 2 1.7% 0 -
Drug robbery/drug dispute 2 1.7% 1 50.0%
Other 4 3.4% 1 25.0%
Unknown 18 15.4% 9 50.0%
Case Not Reviewed* 40 34.2% 12 30.0%

*A large number of cases were investigated by detectives who had retired by the time this analysis was conducted. A lack of
resources and personnel in Portland prevented us from being able to review these incidents.
Group Member Involved Nonfatal Shootings

Unknown Yes
144 (46%) 145 (46%)

No
25 (8%)
 Group-Involved Victims & Suspects:
Nonfatal Shootings

Victims (n=385)) Suspects (n=77)

No
No 21 (27%)
108 (28%) Yes
135 (35%)
Yes
35 (46%)

Unknown
Unknown 21 (27%)
142 (37%)
 Shooting Circumstances and Group Involvement
Total % Group Member
(n=314) Involved
Frequency Percent Frequency Percent
Personal dispute 44 14.0% 24 54.5%
Instant dispute 36 11.5% 14 38.9%
Group-related dispute 28 8.9% 28 100%
Robbery/Carjacking 9 2.9% 1 11.1%
Drug robbery/drug dispute 9 2.9% 3 33.3%
Domestic dispute 6 1.9% 4 66.7%
Burglary/Larceny 1 0.3% 0 -
Other 8 2.5% 1 12.5%
Unknown 101 32.2% 42 41.6%
Case Not Reviewed* 72 22.9% 28 38.9%

*A large number of cases were not assigned to assault detectives for follow-up due to a lack of solvability factors and
resources. Limited resources and personnel in Portland prevented us from being able to pull reports and review these
incidents.
 Summary Observations:
Incident Analysis & Group Involvement
• Homicides from 2019-2021 were significantly more likely to be gun
homicides (75%) when compared to homicides 2015-2019 (60%).
• Just under half of all homicides in Portland from January 2019 – June
2021 involved group or gang members as either victims or suspects or
both (41-78%).
• More than half of the 88 gun homicides in Portland from January 2019 –
June 2021 involved group or gang members as either victims or suspects
or both (52-85%).
• We use the term “group-involved” because it:
▪ captures the full variety of gangs, sets and crews characteristic of
Oregon cities; and
▪ includes not just group-on-group conflicts but also incidents in
which someone’s association with a group increases their risk of
violence.

39
 Summary Observations:
Incident Analysis & Group Involvement
• Homicides in Portland fall into many categories of circumstances, but the
largest are ongoing personal disputes and instant disputes.
• This is closely followed by violence resulting from group/gang members
engaged in ongoing group-related conflicts.
• Group and gang members are also involved in a significant share of other
crimes that result in homicides, such as other types of drug robberies and
drug-related disputes.

• Overall, 0.1% of Portland’s population was involved in a homicide or shooting

from 2019-2021.
• However, approximately 18.2% of the estimated population of group/gang-
involved individuals in Portland was involved in a homicide or shooting from
2019-2021.
• Over the study period there were approximately 227 group involved
individuals directly involved in a homicide or a shooting, indicating that a
very small and very high risk population is driving a significant portion
of the gun violence in Portland.

40
High Risk Groups and Networks
 Summary Observations:
Groups, Gangs, and Networks at High Risk of Violence

• The Portland Police Bureau identified 30 groups and gangs currently active
and at significant risk of involvement in violence as of 2021.
• These groups and gangs have an estimated currently active membership
of 1000-1495 members (0.16-0.23% of Portland’s population).
• These groups tend to be primarily African-American, though there are also
several Hispanic groups and White groups.
• Of these 30 groups, 10 Portland groups were associated with the greatest
amount of homicide cases (43) from January 2019 – June 2021. These 10
groups have an estimated currently active membership of 570-815
members.
• Homicides and shootings associated with these groups are driven by
ongoing group conflicts but also by members using violence to resolve
personal conflicts.

42
 Groups Involved in Three or More Homicides
9

6 4
3
5

4
1
3 4
3 1 1
2 4 4
3 3 3 3
1 2 2
1 1
0

Victims Suspects Both

43
Groups Involved in Two or Fewer Homicides
2

1 1

1 2 2 2 2

1 1 1 1 1 1 1 1

Victims Suspects

44
Groups Involved in Three or More Nonfatal Shootings
30

25

20 8 5

7
15 6
3
10
18 3
16
13 12 3
5 11
1
7 2 2
5 5 4 4 1 1 1 1 1 1
2 2 2 2 2 2 2 2
0

Victims Suspects Both

45
Group Conflicts and Alliances

46
Group Conflicts and Alliances

47
Group Conflicts and Alliances

48
Violent Crime Density
Portland Homicides: January 2019 – June 2021

50
Portland Homicide Density: January 2019 – June 2021

51
Portland Nonfatal Shootings: January 2019 – June 2021

52
Portland Nonfatal Shootings: January 2019 – June 2021

53
Discussion of Findings and
Recommendations
 Key Findings
• Since the last analysis Portland’s homicide and shooting rate has risen dramatically.
• This sharp increase is the largest amongst its peer cities– a 207% increase in homicide
rate from 2019-2021.

• Homicides were significantly more likely to be gun homicides (75%) from 2019-2021
when compared to homicides from 2015-2019 (60%).

• During this spike the dynamics of homicides and shootings and demographics of
those involved did not change significantly:
• Mainly Caucasian and African-American males, with an average age of 33, and more
than half involved in gangs/groups.

• Less than 6% of victims and suspects in shootings and homicides are under the age
of 18.
• The majority of victims and suspects are between the age of 18-34, with 25-34 being the
largest age group.

• Portland’s gun homicides are mostly group member involved (52%), and 46% of
shootings are are also group member involved.
• This is likely a conservative estimate: group member involvement was unknown for 33%
of gun homicides and 46% of nonfatal shootings.

55
 Key Findings (2)
• Because of the inability to review a significant proportion of homicide and shooting
cases, conclusions about incident motives are unclear.

• Victims and suspects have extensive prior criminal justice system involvement:
• About 70% of victims and suspects had been arrested before their involvement in the homicide
or nonfatal shooting. Of those:
• 58% had prior felony convictions
• Victims had committed an average of 9 prior criminal offenses
• Suspects had committed an average of 7 prior criminal offenses

• Disproportionate Impact: African-American males continue to be the most

disproportionately impacted by shootings and homicides.
• While African-Americans comprise only 5.9% of Portland’s population, 47% of victims and
suspects of homicides and shootings were African-American.
• While 37% of victims and suspects were Caucasian, and they are 75% of Portland’s
population.

56
 Recommendations
• Central Recommendation: Develop a violence intervention strategy that focuses on
the majority of the shootings and homicides. In Portland the majority of the shootings
and homicides are gang/group member involved

• Focus on the individuals that are at the very highest risk of being involved in
shootings and homicides as victims and suspects. In Portland these are the
characteristics of those at very highest risk:

• African-American males age 18-34

• Caucasian males age 18-34
• Well known to the criminal justice system, especially those with felonies
• Prior or current probation/post prison supervision
• Involved in an active high-risk network/group
• Has been previously shot or socially connected to a recent shooting or
homicide victim

57
 Recommendations (2)
• When implementing strategies to reduce gun violence utilize available data to
focus on the largest known driver of gun violence.

• Utilize evidence-based practices and plan data collection procedures that

facilitate ongoing assessment and strategic adjustments.

• If the goal of public safety strategies is to reduce gun violence in the near term,
invest in and focus on the people that are at highest risk now.

• Obtain technical assistance from research partners to assist with design and
implementation.

• Work with probation/parole on risk assessments for those most likely to be

involved in gun violence.

58
Acknowledgements
 The CPSC and Dr. Lisa Barao would like to thank the Office of Mayor Ted
Wheeler, Assistant Chief Jami Resch, Lt. David Michaelson, the Enhanced
Community Safety Team (ECST), PPB Detectives Division and Strategic
Services Division for their assistance in obtaining the data needed to complete
this report.

www.thecapartnership.org