Skip navigation

United States Sentencing Commission Jan. 2019 - Intra-City Differences in Federal Sentencing Practices, 2005-2017

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
United States Sentencing Commission
January 2019

INTRA-CITY DIFFERENCES IN
FEDERAL SENTENCING PRACTICES
Federal District Judges in 30 Cities, 2005 - 2017

REPORT OF THE UNITED STATES SENTENCING COMMISSION

INTRA-CITY DIFFERENCES IN
FEDERAL SENTENCING PRACTICES
Federal District Judges in 30 Cities, 2005 - 2017

WILLIAM H. PRYOR JR.
Acting Chair
RACHEL E. BARKOW
Commissioner
CHARLES R. BREYER
Commissioner
DANNY C. REEVES
Commissioner
PATRICIA K. CUSHWA
Ex Officio
DAVID RYBICKI
Ex Officio

KENNETH P. COHEN
Staff Director
BRENT E. NEWTON
Deputy Staff Director

JANUARY 2019

TABLE OF CONTENTS

Table of Contents
INTRODUCTION & KEY FINDINGS
5	

Introduction

7	

Key Findings

METHODOLOGY
9	

The Commission’s 2012 Study

11	

Refining the 2012 Study

14	

Case Exclusions

16	

Impact of Exclusions

18	

Measuring Judicial Discretion

20	

Graphical Presentation of the Results

FINDINGS & CONCLUSIONS
25	

Overall Findings

29	

Conclusions

1

Intra-City Differences in Federal Sentencing Practices
January 2019

Table of Contents
Appendices

APPENDIX A

Minneapolis
Salt Lake City

Boston
Chicago

Denver
Saint Louis

Detroit

Manhattan

Cleveland
Columbus

Philadelphia
Pittsburgh

San Diego
Phoenix

Oklahoma City

Baltimore
Alexandria

Los Angeles

Charlotte

Memphis
Atlanta

Dallas

San Antonio

San Juan

Houston

Tampa
Miami

2

|

2012 Intra-District Analysis

	

31	

Bubble Plot Example

	

32	

Scatter Plot Example

APPENDIX B

Seattle

Portland

San Francisco

31	

33	

City-By-City Results
34	
Alexandria
35	Atlanta
36	
Baltimore
37	
Boston
38	
Charlotte
39	
Chicago
40	Cleveland
41	
Columbus
42	
Dallas
43	Denver
44	
Detroit
45	
Houston
46	
Los Angeles
47	
Manhattan
48	
Memphis

https://www.ussc.gov/topic/research-reports

49	
50	
51	
52	
53	
54	
55	
56	
57	
58	
59	
60	
61	
62	
63	

Miami
Minneapolis
Oklahoma City
Philadelphia
Phoenix
Pittsburgh
Portland
Saint Louis
Salt Lake City
San Antonio
San Diego
San Francisco
San Juan
Seattle
Tampa

TABLE OF CONTENTS

APPENDIX C
65	

Caseload Compositions
66	
Alexandria
67	Atlanta
68	
Baltimore
69	
Boston
70	
Charlotte
71	
Chicago
73	Cleveland
74	
Columbus
75	
Dallas
76	Denver
77	
Detroit
78	
Houston
79	
Los Angeles
81	
Manhattan
84	
Memphis

85	
86	
87	
88	
89	
90	
91	
92	
93	
94	
95	
96	
97	
98	
99	

Miami
Minneapolis
Oklahoma City
Philadelphia
Phoenix
Pittsburgh
Portland
Saint Louis
Salt Lake City
San Antonio
San Diego
San Francisco
San Juan
Seattle
Tampa

APPENDIX D
101	

Weighting Analyses

	

101	

Differences in Caseload in a Single Time Period

	

102	

Differences in Caseload Over Time Periods

APPENDIX E
105	

Statistical Outliers and Standard Deviation
119	
Minneapolis
108	Atlanta
120	
Oklahoma City
109	
Chicago
121	
Philadelphia
110	
Baltimore
122	
Phoenix
111	Cleveland
123	
Pittsburgh
112	
Columbus
124	
Portland
113	
Dallas
125	
Salt Lake City
114	
Detroit
126	
San Diego
115	
Los Angeles
127	
San Juan
116	
Manhattan
128	
Tampa
117	
Memphis
118	
Miami
3

Congress included
three provisions of the
Sentencing Reform Act in
response to its concerns
about unwarranted
sentencing disparities.

4

|

https://www.ussc.gov/topic/research-reports

INTRODUCTION

INTRODUCTION

Federal Judges in 30 Cities
2005 - 2017

	

This report examines variations in sentencing

	

In its 2012 Booker Report, the Commission’s

practices—and corresponding variations in sentencing

analysis

outcomes—in the federal courts since the Supreme

differences among district judges in a majority of

1

Court’s 2005 decision in United States v. Booker.

the 94 federal judicial districts. The analysis focused

The United States Sentencing Commission analyzed

on the judges’ rates of non-government sponsored

the sentencing practices of federal district judges

below range sentences. Such sentences result from

in 30 major cities located throughout the country

downward “departures” pursuant to commentary

to determine the extent of the judges’ variations in

or policy statements in the Guidelines Manual or

imposing sentences in relation to the city average.

downward “variances” that are outside of the guidelines

	

This report is the second in a series of reports

framework pursuant to the courts’ consideration of

updating the analyses and findings of the Commission’s

the factors in 18 U.S.C. § 3553(a), without a motion

2012 Report on the Continuing Impact of United States

by the government.5 The Commission’s 2012 report

v. Booker on Federal Sentencing.2 The first such update,

concluded that “sentencing outcomes increasingly

issued in 2017, focused on demographic differences at

depend[ed] upon the judge to whom the case [was]

the national level in federal sentencing.3 Its multivariate

assigned.”6

analysis concluded that the increases in demographic

	

differences in sentencing that had occurred during

compared judges’ individual sentencing practices

the first seven years after Booker—including a higher

to average sentencing practices within their same

average sentence for Black males compared to White

city. Focusing on the average guideline minimum of

4

males—persisted in the subsequent five-year period.

the guideline range for each judge’s caseload, the

This second update focuses on judges’ sentencing

Commission determined whether each judge on

practices at the city level.

average sentenced below or above the guideline

It addresses intra-city

of

sentencing

data

showed

increasing

In its current analysis, the Commission

variations in sentencing practices; it is not intended to

minimum and by how much.

The Commission

address inter-city variations.

calculated each judge’s average percent difference

5

Intra-City Differences in Federal Sentencing Practices
January 2019

from the guideline minimum, whether positive or

	

negative, in their individual cases and then compared

related to a main reason for the Sentencing Reform

the judges’ average percent differences to the average

Act of 1984, which largely still governs federal

percent difference for all the judges in their city.

sentencing today, after certain modifications made by

	

The current study is based on a broader

the Supreme Court in Booker.7 The Act was the result

range of cases than the 2012 study—not only non-

of a widespread bipartisan concern that unwarranted

government sponsored below range sentences but also

sentencing disparities existed in the federal judicial

many sentences within and above the guideline range.

system.8

It also considered some, but not all, government-

differences among the districts) as well as within the

sponsored below range sentences. The cases studied

same courthouse.9 This report concerns the extent of

by the Commission for this report accounted for 49.2

the latter type of disparities since 2005.

percent of all cases in the 30 cities from fiscal year

	

2005 to fiscal year 2017. They were a representative

sentencing

sample of all cases nationally during those same years.

different provisions of the Act—two directed to

	

A key benefit of comparing how different

the Commission and a third directed to sentencing

judges in the same city sentenced is that they

judges—intended to reduce the extent of disparities.10

generally had similar caseloads in terms of offense

Congress instructed the Commission to pay “particular

types and offender characteristics as the result of the

attention”

random assignment of cases among the judges in that

disparities in creating guidelines.11

city.

Therefore, differences in sentencing practices

Supreme Court’s decision in Booker rendered the

among those judges generally are not explained by

guidelines advisory—thereby providing courts with

differences in the composition of their caseloads. The

more discretion in sentencing than under the pre-

methodology used by the Commission for this report

Booker guidelines—the Court did not invalidate those

is further discussed at pages 9 through 24.

three provisions. Indeed, the Court specifically stated

	

that it believed that the post-Booker advisory guideline

The Commission’s updated analysis is directly

Such disparities were both regional (e.g.,

In response to its concern about unwarranted
disparities,

to

avoiding

Congress

included

unwarranted

three

sentencing

Although the

system would “promote uniformity in the sentencing
process” and thus help avoid unwarranted sentencing
disparities.12

6

|

https://www.ussc.gov/topic/research-reports

KEY FINDINGS

KEY FINDINGS

Findings of the Commission’s
30-City Analysis
	

Overall increases in sentencing differences among judges in 30
major cities since 2005 are consistent with the Commission’s findings in
its 2012 Booker Report regarding intra-district sentencing differences—
demonstrating that differences persist, 13 years after Booker and six years
after the 2012 analysis.

•	 The Commission’s current analysis measured judges’ average percent differences from the guideline
minimums in their cases in relation to their city’s average during three periods between 2005 and
2017.* It demonstrated a clear increase in the extent of differences in sentencing practices in a
majority of the cities studied following the Supreme Court’s 2005 decision in Booker and continuing
after the Court’s 2007 decisions in Gall and Kimbrough. The overall trend continued, although to lesser
extent, in the six years following the last period analyzed in the Commission’s 2012 Booker Report.

•	 Not all of the 30 cities experienced the same changes in differences in sentencing practices since 2005.
In some cities, particularly the ones with the largest number of judges, the increases in differences
were substantial. However, in other cities, the increases were modest, and a few cities experienced
decreases in the extent of sentencing differences among their judges since 2005.

•	 In most cities, the length of a defendant’s sentence increasingly depends on which judge in the
courthouse is assigned to his or her case.

* The Booker Period (January 12, 2005—December 9, 2007); the Gall Period (December 10, 2007—September 30, 2011);
and the Post-Report Period (October 1, 2011—September 30, 2017).

7

By controlling for offense
and offender characteristics
through random case
assignment, differences in
sentencing practices among
a city’s judges will primarily
be attributable to individual
judges’ different degrees
of sentencing discretion
exercised in their cases.

8

|

https://www.ussc.gov/topic/research-reports

METHODOLOGY

METHODOLOGY

The Commission’s 2012
Study of Sentencing Differences

	

In 2012, the Commission examined sentencing

	

An example of a bubble plot, taken from

differences among judges within each of the 94 federal

the 2012 Booker Report, appears in Appendix A

judicial districts. Its report used “bubble plots” to show

(p. 31).	

how judges within the same district differed in how

	

often they imposed sentences below the guideline

district (in the example in Appendix A, the Northern

range without a request from the government.13

District of Illinois).

Each bubble represented a single judge in a
The relative size of a bubble

depicted the size of a judge’s caseload compared to
The 2012 analysis covered four time periods:
•	 the Koon Period (the Supreme Court’s 1996
decision in Koon v. United States14 until the
enactment of the PROTECT Act on April 30,
2003);
•	 the PROTECT Act Period (April 30, 2003,

the sizes of the caseloads of other judges in the same
district. The larger the bubble, the larger a judge’s
share of the district’s caseload.

The location of a

bubble on the vertical axis of the graph showed the
rate of that judge’s non-government sponsored below
range sentences for the relevant time period.
	

The bubble plots for the 94 districts showed

through the Supreme Court’s 2004 decision in

“variation in the rates of non-government sponsored

Blakely v. Washington,15 which foreshadowed

below range sentences among judges within the same

the Supreme Court’s 2005 decision in United

district ha[d] increased in most districts since Booker,

States v. Booker16);

indicating that sentencing outcomes increasingly

•	 the Booker Period (January 12, 2005, to the
Gall and Kimbrough decisions17); and

depend[ed] upon the judge to whom the case [was]
assigned.”18

In particular, the 2012 report found

that “in two-thirds of [the 94] districts the spread in

•	 the Gall Period (the period following the Gall

the rates of non-government sponsored below range

and Kimbrough decisions through the end of

sentences” among judges in the same district increased

fiscal year 2011).	

from the Booker Period through the Gall Period.19
9

Intra-City Differences in Federal Sentencing Practices
January 2019

	

In addition to the bubble plots for each district, in a specific district may not account for differences

the Commission also produced corresponding scatter in caseload and other factors among the different
plots showing the average extent of reduction below divisions or cities within that district that could partially
the average guideline minimum in the cases in which explain the differences in sentencing practices and
judges imposed a non-government sponsored below corresponding sentencing outcomes. 	
range sentence. An example of a scatter plot—also 	

The 2012 district-level study was limited in

concerning the Northern District of Illinois—appears two other respects. First, it analyzed every district
in Appendix A (p. 32). In the scatter plots, individual judge and magistrate judge “who sentenced at least
judges in each district were represented by triangles; one offender” during the relevant time periods.22
unlike in the bubble plots, the size of each judge’s Because there was no minimum caseload requirement
caseload was not depicted by the size of the triangles. for inclusion of a judge in the analysis, some judges’
The scatter plots showed significant differences in the actual overall sentencing practices may not have been
judges’ average extents of reduction below the average reflected in the data. Second, as noted above, the
guideline minimum in those cases, yet the differences 2012 methodology looked solely at the rate and extent
existed in each of the four periods studied.20
	

of non-government sponsored below range sentences.

In discussing the limitations in its district- Although the rate of such sentences has steadily

level study, the 2012 report noted that, although increased after Booker, they still only constitute around
“judges within the same district generally are more one-fifth of all sentences today.23

A more robust

likely than judges across districts to preside over analysis would include a broader range of judges’ cases.
similar cases to the extent that the district’s cases are
randomly distributed among the judges,” “the caseload
composition and prosecutorial practices . . . may differ
across divisions within the same district.”21 As a result,
a comparison of the sentencing practices of all judges

10

|

https://www.ussc.gov/topic/research-reports

	

METHODOLOGY

METHODOLOGY

Refining the 2012 Study
Study of Cities vs. Districts

Figure 1. Periods Analyzed in the 30-City Study
Jan. 12, 2005 – Dec. 9, 2007

Booker Period

	

Oct. 1, 2011 – Sept. 30, 2017

Dec. 10, 2007 – Sept. 30, 2011

Post-Report Period

Gall Period

For its current report, the Commission has

	

The

refined

methodology

makes

other

created a refined methodology to analyze differences

changes responsive to the limitations noted by the

in sentencing practices in the three periods since the

Commission in its 2012 study. To be considered in the

guidelines became advisory—the Booker Period, Gall

current analysis, a sentencing judge had to impose a

Period, and six-year period following the publication of

minimum of 50 sentences during at least one of the

the 2012 Booker Report (called the “Post-Report Period,”

periods studied, in order to make it more likely that the

using data from fiscal years 2012-2017).24 The refined

judge’s cases were representative of his or her overall

methodology narrows the bubble plots’ focus on intra-

sentencing practices. A city had to have at least three

district differences by looking at sentencing differences

judges in each period who met this threshold. There

at the city level—that is, differences in sentencing

was no requirement, however, that those judges be

practices among district judges with chambers and

the same for each period studied.25 In addition, the

courtrooms in the same city.

analysis included more than just cases in which judges

11

Intra-City Differences in Federal Sentencing Practices
January 2019

imposed sentences below the guideline minimum

practices may differ depending on the city or division

without a request from the government (discussed

within a district. Therefore, comparing the sentencing

further below).

practices of an entire district’s judges may result in

	

The study of different sentencing practices of

findings of differences in sentencing practices and

judges within a single city offers a distinct advantage

sentencing outcomes that could be explained in part by

over the study of sentencing differences of judges

differences in offense types and offender characteristics

in larger geographical areas, including federal judicial

or different prosecutorial practices associated with

districts. When the sentencing practices of federal

different parts of the district.

judges in a single city are compared, the analysis benefits

	

from random distribution of cases from the same

analyzed the sentencing practices of district judges

pool of cases, which are subject to generally uniform

located in large metropolitan areas in 30 federal judicial

prosecutorial practices (e.g., charging practices). That

districts throughout the country.28 The cities represent

distribution is a function of random case assignment

each of the major regions of the United States—the

generally used in the federal judicial system in each

Southeast, the Northeast, the Southwest, the Midwest,

major city or division of a district in which a courthouse

the West29—as well as 11 of the 12 federal judicial

is located.26 Analyzing different sentencing practices

circuits.30

among district judges in the same city is akin to a

densely populated areas of the United States.31

“natural experiment.”27

	

Assuming a city’s judges’

As part of this analysis, the Commission

Each city is located in one of the most

The total number of cases from the 30 cities

caseloads are each large enough as well as randomly

over the three periods analyzed by the Commission

assigned, all judges in a city overall will sentence

was 143,589. The total number of district judges who

similarly situated offenders with respect to offense

imposed sentences in those cases was 413 (see p. 15).

types and offender characteristics.

By controlling

Judges who imposed 50 or more sentences in more

for offense type and offender characteristics through

than one period were counted only once in calculating

random case assignment, differences in sentencing

that total. Judges are not identified by name in the

practices among a city’s judges will primarily be

Commission’s study. Appendix C lists only identification

attributable to individual judges’ different degrees of

numbers assigned to each judge by the Commission for

sentencing discretion exercised in their cases.

research purposes.

	

Conversely, when judges within an entire

judicial district are analyzed, the offense types and
offender characteristics in the judges’ caseloads may
differ substantially depending on the different divisions
or cities within the district. Furthermore, prosecutorial

12

|

https://www.ussc.gov/topic/research-reports

METHODOLOGY

Figure 2. Cities Analyzed in the Refined Study

Seattle

Portland

Minneapolis
San Francisco

Salt Lake City

Boston
Chicago

Denver
Saint Louis

Detroit

Manhattan

Cleveland
Columbus

Philadelphia
Pittsburgh

Baltimore
Alexandria

Los Angeles
San Diego

Oklahoma City

Phoenix

Charlotte

Memphis
Atlanta

Dallas

San Antonio

San Juan

Houston

Tampa
Miami

1.	 Alexandria (Eastern District of Virginia);32

16.	 Miami (Southern District of Florida);

2.	 Atlanta (Northern District of Georgia);

17.	 Minneapolis (District of Minnesota);

3.	 Baltimore (District of Maryland);

18.	 Oklahoma City (Western District of Oklahoma);

4.	 Boston (District of Massachusetts);

19.	 Philadelphia (Eastern District of Pennsylvania);

5.	 Charlotte (Western District of North Carolina);

20.	 Phoenix (District of Arizona);

6.	 Chicago (Northern District of Illinois);

21.	 Pittsburgh (Western District of Pennsylvania);

7.	 Cleveland (Northern District of Ohio);

22.	 Portland (District of Oregon);

8.	 Columbus (Southern District of Ohio);

23.	 Saint Louis (Eastern District of Missouri);

9.	 Dallas (Northern District of Texas);

24.	 Salt Lake City (District of Utah);

10.	 Denver (District of Colorado);

25.	 San Antonio (Western District of Texas);

11.	 Detroit (Eastern District of Michigan);

26.	 San Diego (Southern District of California);

12.	 Houston (Southern District of Texas);

27.	 San Francisco (Northern District of California);

13.	 Los Angeles (Central District of California);
14.	 Manhattan (Southern District of New York);
15.	 Memphis (Western District of Tennessee);

28.	 San Juan (District of Puerto Rico);
33

29.	 Seattle (Western District of Washington); and
30.	 Tampa (Middle District of Florida).

13

Intra-City Differences in Federal Sentencing Practices
January 2019

METHODOLOGY

Case Exclusions
Focusing on Cases in Which
Meaningful Sentencing Discretion Can Be Measured
	

Another refinement of this study is its

range, which together constitute around half of all

inclusion of all types of cases in which judicial

sentences.34 The Commission’s consideration of these

sentencing discretion can be measured. In the 2012

additional categories of cases allows for a fuller study

study, the Commission focused on cases in which

of sentencing discretion.

courts sentenced defendants below the guideline

	

range without a motion from the government. The

dataset in other ways in order to focus on sentencing

current study analyzes those cases as well as cases

discretion.

in which courts imposed sentences outside of the

cases that did not allow for a meaningful assessment

guideline range based on a government motion (with

of sentencing discretion, or in which it was difficult to

two exceptions discussed below).

accurately measure the percent difference from the

In addition, the

Commission’s current analysis includes cases in which

The Commission, however, narrowed the
Specifically, the Commission excluded

guideline minimum.

judges imposed sentences within or above the guideline

The Commission excluded two categories of cases because they do not allow
for a meaningful assessment of judges’ sentencing discretion:

EXCLUSION 1

Cases in which a statutory mandatory minimum penalty equaled or exceeded the

EXCLUSION 2

Cases in which a court departed downwardly based on a defendant’s “substantial

otherwise applicable guideline minimum (“mandatory minimum trumps”);35 and

assistance” to the government or as part of a “fast-track” program (insofar as such
departures require the government to file motions requesting them and, when filed,
the motions are almost always granted).36

14

|

https://www.ussc.gov/topic/research-reports

METHODOLOGY

SUMMARY OF ANALYSIS

30
CITIES

413

The analysis refines
the dataset to
allow for a fuller
study of sentencing
discretion.

JUDGES

13
YEARS

143,589
CASES

The Commission excluded three additional types of cases because of
difficulties in accurately measuring the percent difference from the guideline
minimum in such cases:

EXCLUSION 3

Cases with one or more counts of conviction under a statute requiring a mandatory
sentence of imprisonment to run consecutively to any sentence imposed under the
guidelines (e.g., 18 U.S.C. § 924(c));37

EXCLUSION 4

Cases in which the guideline minimum was life imprisonment or cases in which a
court varied or departed upwardly from a guideline range of a term of months to a
life imprisonment sentence (as there is no meaningful way to measure the extent of
a departure or variance below a life term or the extent of an upward departure or
variance to a life term, as a life term is not a fixed term of months); and

EXCLUSION 5

Cases in which the guideline minimum was less than ten months (which, after 2010,
necessarily fell in Zones A or B of the Sentencing Table).38

15

Intra-City Differences in Federal Sentencing Practices
January 2019

METHODOLOGY

Impact of Exclusions

	

After excluding these five categories of cases,

	

The Commission applied the same case

along with cases for which incomplete sentencing

exclusions to the national caseload during the 13-year

documentation was submitted to the Commission,39

study period, which showed that the 143,589 cases

149,363 of the total 291,763 cases from the 30

from the 30 cities were a representative sample of the

cities during fiscal years 2005 to 2017 remained for

national caseload.41 In terms of caseload composition

analysis.40 An additional 5,774 cases were excluded

(guideline types), the 30 cities’ caseloads (combined)

because they were handled by judges who did not

resembled the national caseload composition during

meet the minimum 50-case requirement per period

the 13-year period, as shown in the figure below.

for inclusion in the Commission’s analysis—bringing
the total number of cases in the Commission’s 30-city
dataset to 143,589 (49.2% of the 291,763 cases).

Figure 3. Comparison of Caseload Composition After Exclusions
National Caseload (Exclusions Applied)
FY 2005 – FY 2017

Combined 30-City Caseload (Exclusions Applied)
FY 2005 – FY 2017

(N = 478,833)

(N = 143,589)

§2B1.1

§2B1.1

9.8%

§2D1.1

§2K2.1

§2L1.2

All Other

16

|

33.5%

13.2%

17.8%

25.7%

https://www.ussc.gov/topic/research-reports

13.1%

§2D1.1

§2K2.1

§2L1.2

All Other

31.0%

12.7%

14.2%

29.0%

METHODOLOGY

Figure 4. Cases Remaining in 30-City Study After Exclusions
Federal Offenders Sentenced from FY 2005 through FY 2017
291,763
The Five Methodology Exclusions
(n=131,647)

160,116
Incomplete Documentation
(n=10,753)

149,363
50-Case Threshold Not Met
(n=5,774)

No Mandatory Minimum Penalty
(n=124,129 or 86.4%)

Mandatory Minimum Below Guideline Range
(n=19,460 or 13.6%)

143,589 Cases
Included in 30-City Analysis

Impact of Exclusions on the Effect of
Prosecutorial Charging Decisions
	

The exclusion of many cases with statutory

	

In 13.6 percent of the cases in the

mandatory minimum penalties reduced the effect that

Commission’s dataset for this study (19,460 of 143,589

prosecutorial charging practices could have on the

cases), offenders were subject to statutory mandatory

Commission’s analyses in this report. Although random

minimum sentences. In all of those cases, however,

case assignment theoretically should make the effect

the statutory minimum fell below the applicable

of prosecutorial charging decisions equally spread

guideline

among all judges in a city, some prosecutors may file

discretion to sentence below the guideline minimums

superseding indictments containing charges carrying

(down to the statutory minimums). In most of those

statutory mandatory minimum penalties after they

cases, courts sentenced well above the statutory

learn that their cases have been assigned to certain

mandatory minimum sentence even as they departed

judges in a courthouse who are perceived as unduly

or varied below the guideline range,43 suggesting that

lenient. In such cases, the filing of the superseding

prosecutorial charging decisions did not constrain the

charges could constrain the judge’s sentencing

courts’ sentencing discretion in such cases.

discretion.

42

minimum.

Therefore,

judges

retained

The Commission’s methodology has

reduced the potential influence of this factor.

17

Intra-City Differences in Federal Sentencing Practices
January 2019

METHODOLOGY

Measuring Judicial Discretion
Judges’ Average Percent Difference from Guideline Minimums

	

The Commission’s city-level analysis focused

	

The guideline minimum was chosen as the

on the average percent difference between the

baseline for analysis because of the gravitational pull it

guideline minimums and the sentences imposed in each

tends to have on sentences. The Supreme Court has

judge’s cases. For each case, the guideline minimum

directed district courts to consider the guideline range

and the actual sentence imposed were determined, and

as the “benchmark” and “starting point” in the post-

a percent difference between the two was calculated

Booker federal sentencing process and also to “remain

(see Figure 5 on the next page). For example, if the

cognizant” of it during all three steps of the “Booker

guideline minimum in a case was 63 months and the

three-step process” used at federal sentencing.44 In

judge imposed a sentence of 39 months (24 months

a majority of cases in which judges impose sentences

below the guideline minimum), the percent difference

within the applicable guideline range, they impose

in that case was -38.1 percent. Conversely, if the

the guideline minimum (58.4% of cases from fiscal

judge imposed a sentence of 87 months in a case with

years 2005 through 2017). Furthermore, in the cases

a guideline minimum of 63 months (24 months above

analyzed by the Commission for this report, when

the guideline minimum), the percent difference was

judges imposed a sentence outside of the guideline

38.1 percent. All of a judge’s cases’ percent differences

range, they departed or varied below the range nearly

were then added together and divided by the total

21 times as often as they departed or varied above the

number of his or her cases, which yielded an average

range.45 In those cases with below range sentences,

percent difference for that judge. For some judges, the

judges were required by Supreme Court precedent to

average percent difference was a positive percentage

use the guideline minimum as the “starting point” in

(meaning that, on average, his or her sentences were

deciding how far to depart or vary below the guideline

above the guideline minimums), while for others it was

range. The guideline minimum is thus the focal point

a negative percentage (meaning that, on average, his or

in the advisory guidelines system and, for that reason,

her sentences were below the guideline minimums).

the baseline for the Commission’s analysis.

18

|

https://www.ussc.gov/topic/research-reports

METHODOLOGY

Figure 5. Calculating Percent Difference from Guideline Minimum
EXAMPLE:

Average Percent Difference Versus
Average Sentence Length
	

While

other

researchers

have

Guideline Minimum
63 months

studied

24-month Difference
-38.1%

differences in average sentence length among judges
in the same city,46 the Commission focused on average
percent differences above or below the guideline

39 months

minimum because that measure is less sensitive to

Sentence Imposed

differences in the judges’ caseload compositions and,
thus, is a better gauge of sentencing discretion. As

	

shown in Appendix C (p. 65), the judges in the majority

based on average sentence lengths, the Commission’s

of the 30 cities had generally similar caseloads

analysis is informative of differences in sentence length

compared to the other judges in their cities. However,

in similar cases. Judges’ average percent differences

even with random case assignments, some judges had

from the guideline minimums in their cases are, of

caseload compositions that differed somewhat from

course, necessarily associated with their average

other judges in the same city. Such differences can

sentence lengths. For example, assuming two judges’

have a significant impact on average sentence length

average guideline minimums each reflected the national

per judge. For instance, if one judge in a city drew

average of 60 months,48 but their average percent

a disproportionately large number of drug-trafficking

differences were 10.0 percent and -40.0 percent, they

or firearms cases, which tend to have much higher

would impose average sentences of 66.0 months and

guideline minimums (and correspondingly higher

36.0 months, respectively—a difference of 30 months

sentences) on average than other common offense

(or 2-½ years). If those two judges had courtrooms in

types,47 that judge’s average sentence could be higher

the same city and each drew a sufficiently large number

than other judges’ sentences in the same city for

of cases through random case assignment to result in

reasons unrelated to how that judge exercised his or

comparable offender and offense characteristics in

her post-Booker sentencing discretion. As discussed

their caseloads, that difference in average sentences

in Appendix D (p. 101), however, uneven distributions

would indicate that similarly situated defendants were

of offense types among judges in some of the 30

receiving substantially different sentences within the

cities did not substantially affect the average percent

same courthouse.

Although the Commission’s analysis is not

differences of those judges.

19

Intra-City Differences in Federal Sentencing Practices
January 2019

METHODOLOGY

Graphical Presentation of the Results
Figure 6. Graphical Depiction of the 30-City Analysis

Judges above

35 to 39.99%

the city’s average.

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

City’s
total spread
over one
time period.

Each judge in a city is represented by
a bar on the graph.

5 to 9.99%

In this example, 24 judges are included

0 to 4.99%

in the analysis of this time period.

-0.01 to -4.99%
-5 to -9.99%

Judges below

-10 to -14.99%

the city’s average.

-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

	

-40.0

-30.0

-20.0

-10.0

0.0
City
Avg
-28.1%

10.0

20.0

30.0

40.0

City’s average percent difference from the city’s average guideline minimum.

The complete results of the Commission’s 30-

minimum for a period) is represented by a bar on the

city study appear in Appendix B (p. 33). As shown in

graph. The bars show the sentencing practices of each

Figure 6 above, a graph for each city uses horizontal

judge in relation to the city average. Bars are in either

bars to show differences in sentencing practices among

the positive or negative halves of the graph depending

district judges in each city studied (for each time

on judges’ average percent differences in relation to

period). Each judge in a city (who met the 50-case

the city average.

20

|

https://www.ussc.gov/topic/research-reports

METHODOLOGY

	

Red-colored bars represent judges whose

	

The total spread or range measures the

average percent differences are above the city’s

absolute percentage difference between the two

average, while blue-colored bars represent judges

judges at opposite ends of the bar graphs,50 while the

whose average percent differences are below the

standard deviation measures the overall extent of all the

city’s average. The darker shades of each color, the

city’s judges’ differences in sentencing practices. The

further the percentages are away from the city average,

standard deviation is a separate “measure of spread,

positive or negative. A color key shows what range of

dispersion or variability of a group of” datapoints.51 The

percentages (in 5% bands) that each shade of each

larger the standard deviation, the greater the dispersion

color represents (e.g., “-25 to -29.99%”).

or variability among the datapoints in the dataset. For

	

Each graph contains one, and usually two,

cities where there were less than five judges in any

important datapoints relevant to differences in

period, only the total spread is reported because four or

sentencing practices among the city’s judges for each

less judges are too few for reporting the city’s standard

time period: (1) the difference between the judge

deviation. For cities with five or more judges in all

with the largest positive average percent difference

three periods, both the total spread and the standard

in relation to the city average and the judge with the

deviation are reported. Twenty-seven of the 30 cities

largest negative average percent difference in relation

had at least five judges in all the periods.52 Appendix

to the city average in a given time period (referred to

E (p. 105) contains a discussion of the relevance of the

as the total spread or range);

and (2) the standard

standard deviation to the Commission’s analysis, as

deviation for all judges’ average percent differences in

well as a discussion of statistical outliers in the analysis.

49

relation to the city average for the same time period.

21

Intra-City Differences in Federal Sentencing Practices
January 2019

METHODOLOGY

Graphical Presentation

Depicting a Single Time Period

	

An example of the bar graph (in a single time

period) is depicted in Figure 7 to the right.
	

In the case of Chicago in the Post-Report

Period, the city’s average percent difference from the

CHIThis means, on
guideline minimum was -28.1 percent.
Post

average, judges in Chicago imposed sentences 28.1
percent below the guideline minimums in their cases
in the most recent period. The judge represented by

Figure 7. Chicago-Based Federal Judges in the
Post-Report Period
POST-REPORT PERIOD
Judges: 24
Cases: 2,951
Spread: 49.5
Standard Deviation: 10.6

the bar with the darkest red shade was the farthest
from the city’s average in the positive direction (with
an average percent difference 32.4 percentage points
higher than the city average). The judge represented
by the bar with the darkest blue shade was the farthest
from the city’s average in the negative direction (with
an average percent difference 17.0 percentage points
lower than the city average). The total spread of 49.5 is
the absolute percentage difference between those two
judges’ average percent differences.53 The standard
-40.0
deviation of 10.6 is the measure of the dispersion from

the city average for all 24 judges in the analysis.

22

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

City
Avg
0.0
-28.1%

10.0

20.0

30.0

40

METHODOLOGY

Figure 8. Chicago-Based Federal Judges in
All Periods

METHODOLOGY

CHI
Booker

Graphical Presentation
Depicting All Time Periods

	

BOOKER PERIOD
Judges: 18
Cases: 1,341
Spread: 31.6
Standard Deviation: 7.9

The bar graphs for all 30 cities in Appendix B

(p. 33) show all three time periods together, so that
changes in the total spread and standard deviation
from one period to the next can be seen.
	

An example of bar graphs for a single city over

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-10.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-28.1%

10.0

20.0

30.0

40.0

CHI

all three periods is depicted on the right in Figure
Gall8.
This graphical presentation for Chicago shows that
the total spread increased from 31.6 in the Booker
Period to 42.7 in the Gall Period to 49.5 in the PostReport Period. During the same three periods, the
standard deviation increased from 7.9 to 11.0 but

GALL PERIOD
Judges: 22
Cases: 2,033
Spread: 42.7
Standard Deviation: 11.0

then decreased slightly to 10.6.

-40.0

CHI
Post

-30.0

-20.0

-10.0

POST-REPORT PERIOD
Judges: 24
Cases: 2,951
Spread: 49.5
Standard Deviation: 10.6

-40.0

-30.0

-20.0

-10.0

23

Intra-City Differences in Federal Sentencing Practices
January 2019

Figure 9. Oklahoma City-Based Federal Judges in
All Periods

METHODOLOGY

OKC
Booker

BOOKER PERIOD

Graphical Presentation
Depicting All Time Periods

	

Judges: 7
Cases: 436
Spread: 32.1
Standard Deviation: 10.2

In contrast to Chicago, Oklahoma City is an

example of a city whose total spreads and standard
deviations consistently decreased during the three
periods.
	

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-0.7%

10.0

20.0

30.0

40.0

City
0.0Avg
-7.6%

10.0

20.0

30.0

40.0

City
0.0Avg
-19.3%

10.0

20.0

30.0

40.0

Its total spread decreased from 32.1 in the
OKC

Booker Period to 14.8 in the Gall Period to 6.9 inGall
the
Post-Report Period. During the same three periods,
the standard deviation decreased from 10.2 to 5.2 and
then to 2.7.

GALL PERIOD
Judges: 7
Cases: 800
Spread: 14.8
Standard Deviation: 5.2

-40.0

-30.0

-20.0

-10.0

OKC
Post

POST-REPORT PERIOD
Judges: 6
Cases: 1,369
Spread: 6.9
Standard Deviation: 2.7

-40.0

24

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

FINDINGS

Results for all 30 cities, as
depicted in the bar charts
for the three periods
(including the total
spreads and standard
deviations), are contained
in Appendix B.

FINDINGS

Overall Findings of the
30-City Analysis
City-by-City Changes Across Periods
Booker Period to Gall Period	
	

Consistent with the findings of the Commission’s

2012 Booker Report, the city-level analysis demonstrated
an overall increase in differences in sentencing practices
from the Booker Period to the Gall Period. Specifically,
as shown in Appendix B (p. 33), 23 of the 30 cities had
increases in their total spreads, and 22 of 27 cities (those
with at least five judges in all three periods) had increases
in their standard deviations. Seven cities had decreases
in their total spreads, and five cities had decreases in the

23 CITIES

INCREASES IN TOTAL SPREAD

7 CITIES

DECREASES IN TOTAL SPREAD

22 CITIES

INCREASES IN STANDARD DEVIATION

5 CITIES

DECREASES IN STANDARD DEVIATION

standard deviations.54 Not all cities experienced increases
or decreases in both measures. Some had an increase in
one measure and a decrease in the other measure.
Gall Period to Post-Report Period
	

This trend continued from the Gall Period to

the Post-Report Period, but at a slower rate. As shown
in Appendix B (p. 33), 20 of the 30 cities had increases
in their total spreads.

Sixteen of the 27 cities (those

with at least five judges in all periods) had increases in
their standard deviations, although the magnitude of the
increases was noticeably less than the magnitude of the
increases from the Booker Period to the Gall Period. Ten
cities had decreases in their total spreads, and 11 had

20 CITIES

INCREASES IN TOTAL SPREAD

10 CITIES

DECREASES IN TOTAL SPREAD

16 CITIES

INCREASES IN STANDARD DEVIATION

11 CITIES

DECREASES IN STANDARD DEVIATION

25

Intra-City Differences in Federal Sentencing Practices
January 2019

decreases in their standard deviations.55 Just as in the
prior period change, not all cities experienced increases
or decreases in both measures. Some had an increase
in one measure and a decrease in the other measure.
Booker Period to Post-Report Period
	

Of the 30 cities, five—Boston, Cleveland,

25 CITIES

Columbus, Oklahoma City, and Saint Louis—saw net

INCREASES IN TOTAL SPREAD

decreases in their total spreads from the Booker Period

5 CITIES

to the Post-Report Period, while four (four of the
same five, with the exception of Cleveland) saw net
decreases in their standard deviations as well. The rest
of the cities saw net increases in their total spreads and
standard deviations from the Booker Period to the PostReport Period.

DECREASES IN TOTAL SPREAD

23 CITIES

INCREASES IN STANDARD DEVIATION

4 CITIES

DECREASES IN STANDARD DEVIATION

Largest and Smallest Total
Spreads and Standard Deviations
Post-Report Period
	

Looking at the most recent period, the Post-

PHILADELPHIA

Report Period, the city with the largest total spread

LARGEST SPREAD

was Philadelphia (63.8), and the city with the smallest

OKLAHOMA CITY

total spread was Oklahoma City (6.9). The city with the
largest standard deviation was Houston (14.6), and the
city with the smallest standard deviation was Oklahoma
City (2.7).

SMALLEST SPREAD

HOUSTON

LARGEST STANDARD DEVIATION

OKLAHOMA CITY

SMALLEST STANDARD DEVIATION

26

|

https://www.ussc.gov/topic/research-reports

FINDINGS

Changes for All 30 Cities
Combined Across Periods
	

A similar trend of overall increasing sentencing

differences is seen with respect to the average total
spread and average standard deviation for all 30 cities
combined (except for the three cities with fewer than
five judges for at least one period, for which standard

Seattle

Portland

Minneapolis
San Francisco

deviations were not reported). As reflected in Figure
10 below, average total spreads for all 30 cities in the

Salt Lake City

Boston
Chicago

Denver
Saint Louis

Detroit

Manhattan

Cleveland
Columbus

Philadelphia
Pittsburgh

Los Angeles
San Diego
Phoenix

three periods grew from 18.2 in the Booker Period

Oklahoma City

Charlotte

Memphis
Atlanta

Dallas

to 23.7 in the Gall Period to 27.6 in the Post-Report

Baltimore
Alexandria

San Antonio

San Juan

Houston

Tampa

Period. Average standard deviations for the 27 cities

Miami

(those with at least five judges in all three periods) grew
from 5.8 in the Booker Period to 7.7 in the Gall Period
to 8.3 in the Post-Report Period.

Figure 10. Averages for All 30 Cities Combined

Booker Period

Gall Period

Post-Report Period

Average Total Spread

18.2

23.7

27.6

Average Standard Deviation

5.8

7.7

8.3

27

This report’s findings
of overall increasing
differences in sentencing
practices among judges
within the same cities
are consistent with intradistrict findings from the
Commission’s 2012 report.

28

|

https://www.ussc.gov/topic/research-reports

CONCLUSIONS

CONCLUSIONS

Differences in Sentencing Practices
Among Federal Judges

	

Although the trend of increasing differences among judges slowed after 2011, the increasing

differences in sentencing practices first reported at the district level in the Commission’s 2012 Booker Report
generally persist to this day, even within the same courthouse. In particular, the Commission finds that:

CONCLUSION 1

From the Booker to Gall Periods, 23 of the 30 cities had increases in their total
spreads, and 22 of 27 cities (those with at least five judges in all three periods)
had increases in their standard deviations. From the Gall to the Post-Report
Periods, 20 of the 30 cities had increases in their total spreads, and 16 of the 27
cities (those with at least five judges in all periods) had increases in their standard
deviations, although the magnitude of the increases was less than the magnitude
of the increases from the Booker Period to the Gall Period.

CONCLUSION 2

In terms of the overall changes during the 13 years, from the Booker Period to the
Post-Report Period, 25 of the 30 cities saw a net increase in their total spreads and
23 cities of the 27 with reported standard deviations saw a net increase in their
standard deviations.

CONCLUSION 3

Considering all 30 cities together as a representative sample of the country as a
whole, the average total spreads for all 30 cities in the three periods increased
from 18.2 in the Booker Period to 23.7 in the Gall Period to 27.6 in the Post-Report
Period. The average standard deviations for the 27 cities (those with at least five
judges) grew from 5.8 to 7.7 to 8.3 during the same three periods.

CONCLUSION 4

In most cities, the length of a defendant’s sentence increasingly depends on which
judge in the courthouse is assigned to his or her case.

29

The Commission’s 2012
Booker Report depicted
differences among
judges within districts
using bubble plots and
scatter plots.

30

|

https://www.ussc.gov/topic/research-reports

APPENDIX A: 2012 Intra-District Analysis

APPENDIX A

2012 Booker Report Analysis

Bubble Plot Example

31

Intra-City Differences in Federal Sentencing Practices
January 2019

Scatter Plot Example

32

|

https://www.ussc.gov/topic/research-reports

APPENDIX B: City-by-City Results

APPENDIX B

City-By-City Results

Seattle

Portland

Minneapolis
San Francisco

Salt Lake City

Boston
Chicago

Denver
Saint Louis

Detroit

Manhattan

Cleveland
Columbus

Philadelphia
Pittsburgh

Baltimore
Alexandria

Los Angeles
San Diego
Phoenix

Oklahoma City

Charlotte

Memphis
Atlanta

Dallas

San Antonio

San Juan

Houston

Tampa
Miami

	

The results for the 30 cities, as depicted in

Rather, the Commission’s analysis in this report is

the bar charts for the three periods with information

offered to compare judges within each city to one

about the total spreads and standard deviations, are

another and also to compare the extent of sentencing

contained in this appendix. As noted at the outset of

differences in each city over time.

this report, the 30 cities’ results are not offered for
the purpose of comparing the cities to each other.

33

Intra-City Differences in Federal Sentencing Practices
January 2019

ALEX
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 5
Cases: 746
Spread: 8.5
Standard Deviation: 3.5

ALEXANDRIA-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF VIRGINIA

The Commission analyzed 3,263 cases from Alexandria
-40.0

across the three periods.
ALEX
From the Booker Period to the Gall Period, Alexandria’s
Gall

total spread increased from 8.5 to 21.0, and its
standard deviation increased from 3.5 to 7.9.
From the Gall Period to the Post-Report Period,
Alexandria’s total spread increased from 21.0 to 22.1,

-30.0

-20.0

-10.0

City
Avg
0.0
-4.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-17.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-29.3%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 1,058
Spread: 21.0
Standard Deviation: 7.9

and its standard deviation decreased from 7.9 to 7.5.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

ALEX
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 7
Cases: 1,459
Spread: 22.1
Standard Deviation: 7.5

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

34

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

ATL
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 12
Cases: 995
Spread: 19.6
Standard Deviation: 5.5

ATLANTA-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF GEORGIA

The Commission analyzed 4,642 cases from Atlanta
-40.0

across the three periods.
ATL
From the Booker Period to the Gall Period, Atlanta’s
Gall

total spread increased from 19.6 to 31.8, and its
standard deviation increased from 5.5 to 7.2.
From the Gall Period to the Post-Report Period,
Atlanta’s total spread decreased from 31.8 to 26.9, and

-30.0

-20.0

-10.0

City
Avg
0.0
-1.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-8.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.5%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 14
Cases: 1,504
Spread: 31.8
Standard Deviation: 7.2

its standard deviation increased from 7.2 to 9.6.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%

-10.0

ATL
Post

POST-REPORT PERIOD
Judges: 15
Cases: 2,143
Spread: 26.9
Standard Deviation: 9.6

-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

35

Intra-City Differences in Federal Sentencing Practices
January 2019

BALT
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 6
Cases: 618
Spread: 18.7
Standard Deviation: 6.7

BALTIMORE-BASED
FEDERAL JUDGES
DISTRICT OF MARYLAND

The Commission analyzed 3,354 cases from Baltimore
-40.0

across the three periods.
BALT
From the Booker Period to the Gall Period, Baltimore’s
Gall

total spread increased from 18.7 to 24.5, and its
standard deviation increased from 6.7 to 8.2.
From the Gall Period to the Post-Report Period,
Baltimore’s total spread decreased from 24.5 to 19.7,

-30.0

-20.0

-10.0

City0.0
Avg
-5.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-12.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-15.0%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 968
Spread: 24.5
Standard Deviation: 8.2

and its standard deviation decreased from 8.2 to 6.7.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

BALT
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 9
Cases: 1,768
Spread: 19.7
Standard Deviation: 6.7

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

36

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

BOS
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 12
Cases: 830
Spread: 32.2
Standard Deviation: 9.1

BOSTON-BASED
FEDERAL JUDGES
DISTRICT OF MASSACHUSETTS

The Commission analyzed 3,619 cases from Boston
-40.0

across the three periods.
BOS
From the Booker Period to the Gall Period, Boston’s
Gall

total spread increased from 32.2 to 35.2, and its
standard deviation increased from 9.1 to 11.3.
From the Gall Period to the Post-Report Period,
Boston’s total spread decreased from 35.2 to 28.7, and

-30.0

-20.0

-10.0

City
Avg
0.0
-13.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-20.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-27.8%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 11
Cases: 1,177
Spread: 35.2
Standard Deviation: 11.3

its standard deviation decreased from 11.3 to 8.9.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

BOS
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 11
Cases: 1,612
Spread: 28.7
Standard Deviation: 8.9

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

37

Intra-City Differences in Federal Sentencing Practices
January 2019

CHAR
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 4
Cases: 709
Spread: 8.9

CHARLOTTE-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF NORTH CAROLINA

The Commission analyzed 2,527 cases from Charlotte
-40.0

across the three periods.

-30.0

-20.0

-10.0

City
0.0Avg
1.1%

10.0

20.0

30.0

40.0

-10.0

City
Avg
0.0
-1.9%

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

CHAR
From the Booker Period to the Gall Period, Charlotte’s

total spread increased from 8.9 to 14.1.

Gall

From the Gall Period to the Post-Report Period,
Charlotte’s total spread increased from 14.1 to 19.4.

GALL PERIOD
Judges: 3
Cases: 667
Spread: 14.1

Because Charlotte did not have at least five judges in
all three periods, the standard deviations for the three
periods are not reported.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

CHAR
Post

POST-REPORT PERIOD

5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%

Judges: 3
Cases: 1,151
Spread: 19.4

-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

38

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

City
Avg
0.0
-7.7%

APPENDIX B: City-by-City Results

CHI
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 18
Cases: 1,341
Spread: 31.6
Standard Deviation: 7.9

CHICAGO-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF ILLINOIS

The Commission analyzed 6,325 cases from Chicago
-40.0

across the three periods.

-30.0

-20.0

-10.0

City
Avg
0.0
-10.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.8%

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

CHI
From the Booker Period to the Gall Period, Chicago’s
Gall

total spread increased from 31.6 to 42.7, and its

GALL PERIOD

standard deviation increased from 7.9 to 11.0.
From the Gall Period to the Post-Report Period,
Chicago’s total spread increased from 42.7 to 49.5, and
its standard deviation decreased from 11.0 to 10.6.

Judges: 22
Cases: 2,033
Spread: 42.7
Standard Deviation: 11.0

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

CHI
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 24
Cases: 2,951
Spread: 49.5
Standard Deviation: 10.6

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-28.1%

39

Intra-City Differences in Federal Sentencing Practices
January 2019

CLEV
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 9
Cases: 924
Spread: 22.7
Standard Deviation: 6.6

CLEVELAND-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF OHIO

The Commission analyzed 2,908 cases from Cleveland
-40.0

across the three periods.
CLEV
From the Booker Period to the Gall Period, Cleveland’s
Gall

total spread decreased from 22.7 to 16.6, and its
standard deviation decreased from 6.6 to 5.2.
From the Gall Period to the Post-Report Period,
Cleveland’s total spread increased from 16.6 to 19.9,

-30.0

-20.0

-10.0

City
Avg
0.0
-4.8%

10.0

20.0

30.0

40.0

City
0.0Avg
-6.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-12.2%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 8
Cases: 744
Spread: 16.6
Standard Deviation: 5.2

and its standard deviation increased from 5.2 to 7.5.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

CLEV
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 6
Cases: 1,240
Spread: 19.9
Standard Deviation: 7.5

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

40

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

COL
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 5
Cases: 282
Spread: 15.2
Standard Deviation: 6.0

COLUMBUS-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF OHIO

The Commission analyzed 1,672 cases from Columbus
-40.0

across the three periods.
COL
From the Booker Period to the Gall Period, Columbus’s
Gall

total spread decreased from 15.2 to 12.4, and its
standard deviation decreased from 6.0 to 5.0.
From the Gall Period to the Post-Report Period,
Columbus’s total spread decreased from 12.4 to 8.7,

-30.0

-20.0

-10.0

City
Avg
0.0
-10.4%

10.0

20.0

30.0

40.0

City
Avg
0.0
-19.4%

10.0

20.0

30.0

40.0

City
Avg
0.0
-24.6%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 6
Cases: 463
Spread: 12.4
Standard Deviation: 5.0

and its standard deviation decreased from 5.0 to 3.1.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

COL
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 7
Cases: 927
Spread: 8.7
Standard Deviation: 3.1

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

41

Intra-City Differences in Federal Sentencing Practices
January 2019

DAL
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 8
Cases: 660
Spread: 22.0
Standard Deviation: 6.6

DALLAS-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF TEXAS

The Commission analyzed 4,071 cases from Dallas
-40.0

across the three periods.
From the Booker Period to the Gall Period, Dallas’DAL
total
Gall

spread increased from 22.0 to 33.0, and its standard
deviation increased from 6.6 to 10.9.
From the Gall Period to the Post-Report Period, Dallas’
total spread increased from 33.0 to 40.3, and its

-30.0

-20.0

-10.0

City
Avg
0.0
1.7%

10.0

20.0

30.0

40.0

City
0.0Avg
-3.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-2.1%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 904
Spread: 33.0
Standard Deviation: 10.9

standard deviation increased from 10.9 to 12.5.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

DAL
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 7
Cases: 2,507
Spread: 40.3
Standard Deviation: 12.5

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

42

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

DEN
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 6
Cases: 751
Spread: 4.9
Standard Deviation: 2.3

DENVER-BASED
FEDERAL JUDGES
DISTRICT OF COLORADO

The Commission analyzed 3,417 cases from Denver
-40.0

across the three periods.
DEN
From the Booker Period to the Gall Period, Denver’s
Gall

total spread increased from 4.9 to 12.7, and its standard
deviation increased from 2.3 to 4.4.
From the Gall Period to the Post-Report Period,
Denver’s total spread increased from 12.7 to 17.9, and

-30.0

-20.0

-10.0

City
Avg
0.0
-4.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-7.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-13.9%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 1,182
Spread: 12.7
Standard Deviation: 4.4

its standard deviation increased from 4.4 to 6.3.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

DEN
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 9
Cases: 1,484
Spread: 17.9
Standard Deviation: 6.3

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

43

Intra-City Differences in Federal Sentencing Practices
January 2019

DET
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 11
Cases: 680
Spread: 34.4
Standard Deviation: 9.5

DETROIT-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF MICHIGAN

The Commission analyzed 3,901 cases from Detroit
-40.0

across the three periods.
DET
From the Booker Period to the Gall Period, Detroit’s
Gall

total spread decreased from 34.4 to 34.1, and its
standard deviation increased from 9.5 to 10.6.
From the Gall Period to the Post-Report Period,
Detroit’s total spread increased from 34.1 to 47.7, and

-30.0

-20.0

-10.0

City
Avg
0.0
-9.6%

10.0

20.0

30.0

40.0

City
Avg
0.0
-21.5%

10.0

20.0

30.0

40.0

City
0.0Avg
-23.3%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 12
Cases: 1,011
Spread: 34.1
Standard Deviation: 10.6

its standard deviation increased from 10.6 to 11.1.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

DET
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 18
Cases: 2,210
Spread: 47.7
Standard Deviation: 11.1

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

44

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

HOU
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 10
Cases: 1,148
Spread: 23.3
Standard Deviation: 7.9

HOUSTON-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF TEXAS

The Commission analyzed 6,417 cases from Houston
-40.0

across the three periods.
HOU
From the Booker Period to the Gall Period, Houston’s
Gall

total spread increased from 23.3 to 36.5, and its
standard deviation increased from 7.9 to 12.6.
From the Gall Period to the Post-Report Period,
Houston’s total spread increased from 36.5 to 38.4,

-30.0

-20.0

-10.0

City
0.0Avg
4.1%

10.0

20.0

30.0

40.0

City
0.0Avg
-1.2%

10.0

20.0

30.0

40.0

City
0.0Avg
-6.9%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 11
Cases: 2,411
Spread: 36.5
Standard Deviation: 12.6

and its standard deviation increased from 12.6 to 14.6.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

HOU
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%

Judges: 11
Cases: 2,858
Spread: 38.4
Standard Deviation: 14.6

-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

45

Intra-City Differences in Federal Sentencing Practices
January 2019

LA
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 15
Cases: 1,273
Spread: 29.1
Standard Deviation: 7.6

LOS ANGELES-BASED
FEDERAL JUDGES
CENTRAL DISTRICT OF CALIFORNIA

The Commission analyzed 6,791 cases from Los
Angeles across the three periods.

-40.0

LA
From the Booker Period to the Gall Period, Los Angeles’
Gall

total spread increased from 29.1 to 29.3, and its
standard deviation increased from 7.6 to 8.0.
From the Gall Period to the Post-Report Period, Los
Angeles’ total spread increased from 29.3 to 42.9, and

-30.0

-20.0

-10.0

City
Avg
0.0
-14.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-18.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-26.7%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 18
Cases: 2,407
Spread: 29.3
Standard Deviation: 8.0

its standard deviation increased from 8.0 to 11.8.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

LA
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 20
Cases: 3,111
Spread: 42.9
Standard Deviation: 11.8

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

46

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

MAN
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 32
Cases: 2,899
Spread: 44.8
Standard Deviation: 9.2

MANHATTAN-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF NEW YORK

The

Commission

analyzed

11,197

cases

from
-40.0

Manhattan across the three periods.
MAN
From the Booker Period to the Gall Period, Manhattan’s
Gall

total spread decreased from 44.8 to 34.5 and its
standard deviation decreased from 9.2 to 8.7.
From the Gall Period to the Post-Report Period,
Manhattan’s total spread increased from 34.5 to 59.1,

-30.0

-20.0

-10.0

City
Avg
0.0
-14.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-25.5%

10.0

20.0

30.0

40.0

City
Avg
0.0
-36.3%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 31
Cases: 3,435
Spread: 34.5
Standard Deviation: 8.7

and its standard deviation increased from 8.7 to 12.9.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

MAN
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 38
Cases: 4,863
Spread: 59.1
Standard Deviation: 12.9

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

47

Intra-City Differences in Federal Sentencing Practices
January 2019

MEM
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 3
Cases: 534
Spread: 14.0

MEMPHIS-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF TENNESSEE

The Commission analyzed 2,914 cases from Memphis
-40.0

across the three periods.
MEM
From the Booker Period to the Gall Period, Memphis’

total spread increased from 14.0 to 17.4.

Gall

From the Gall Period to the Post-Report Period,
Memphis’ total spread increased from 17.4 to 18.6.

-30.0

-20.0

-10.0

City
Avg
0.0
-5.8%

10.0

20.0

30.0

40.0

-10.0

City
Avg
0.0
-6.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-13.4%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 4
Cases: 980
Spread: 17.4

Because Memphis did not have at least five judges in
all three periods, the standard deviations for the three
periods are not reported.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

MEM
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%

Judges: 5
Cases: 1,400
Spread: 18.6

-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

48

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

MIA
Booker

KEY FINDINGS

MIAMI-BASED FEDERAL
JUDGES

BOOKER PERIOD
Judges: 13
Cases: 2,074
Spread: 16.0
Standard Deviation: 4.1

SOUTHERN DISTRICT OF FLORIDA

The Commission analyzed 9,467 cases from Miami
-40.0

across the three periods.
From the Booker Period to the Gall Period, Miami’sMIA
total
Gall

spread increased from 16.0 to 17.1, and its standard
deviation increased from 4.1 to 5.2.
From the Gall Period to the Post-Report Period,
Miami’s total spread increased from 17.1 to 29.9, and

-30.0

-20.0

-10.0

City
Avg
0.0
-4.3%

10.0

20.0

30.0

40.0

City
0.0Avg
-6.9%

10.0

20.0

30.0

40.0

City
0.0Avg
-10.0%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 12
Cases: 2,859
Spread: 17.1
Standard Deviation: 5.2

its standard deviation increased from 5.2 to 9.0.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

MIA
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 15
Cases: 4,534
Spread: 29.9
Standard Deviation: 9.0

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

49

Intra-City Differences in Federal Sentencing Practices
January 2019

MINN
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 6
Cases: 606
Spread: 13.7
Standard Deviation: 5.1

MINNEAPOLIS-BASED
FEDERAL JUDGES
DISTRICT OF MINNESOTA

The

Commission

analyzed

2,510

cases

from
-40.0

Minneapolis across the three periods.
MINN
From the Booker Period to the Gall Period, Minneapolis’
Gall

total spread increased from 13.7 to 20.5, and its
standard deviation increased from 5.1 to 6.9.
From the Gall Period to the Post-Report Period,
Minneapolis’ total spread decreased from 20.5 to 15.0,

-30.0

-20.0

-10.0

City
0.0Avg
-7.9%

10.0

20.0

30.0

40.0

City
Avg
0.0
-15.6%

10.0

20.0

30.0

40.0

City
Avg
0.0
-18.7%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 908
Spread: 20.5
Standard Deviation: 6.9

and its standard deviation decreased from 6.9 to 5.3.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

MINN
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%

Judges: 6
Cases: 996
Spread: 15.0
Standard Deviation: 5.3

-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

50

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

OKC
Booker

KEY FINDINGS

OKLAHOMA CITY-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 7
Cases: 436
Spread: 32.1
Standard Deviation: 10.2

WESTERN DISTRICT OF OKLAHOMA

The Commission analyzed 2,605 cases from Oklahoma
-40.0

City across the three periods.
OKC
From the Booker Period to the Gall Period, Oklahoma
Gall

City’s total spread decreased from 32.1 to 14.8, and its
standard deviation decreased from 10.2 to 5.2.
From the Gall Period to the Post-Report Period,
Oklahoma City’s total spread decreased from 14.8 to

-30.0

-20.0

-10.0

City
Avg
0.0
-0.7%

10.0

20.0

30.0

40.0

City
0.0Avg
-7.6%

10.0

20.0

30.0

40.0

City
0.0Avg
-19.3%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 800
Spread: 14.8
Standard Deviation: 5.2

6.9, and its standard deviation decreased from 5.2 to
2.7.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

OKC
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%

Judges: 6
Cases: 1,369
Spread: 6.9
Standard Deviation: 2.7

-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

51

Intra-City Differences in Federal Sentencing Practices
January 2019

PHI
Booker

KEY FINDINGS

PHILADELPHIA-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 10
Cases: 534
Spread: 33.8
Standard Deviation: 9.6

EASTERN DISTRICT OF PENNSYLVANIA

The

Commission

analyzed

2,852

cases

from
-40.0

Philadelphia across the three periods.
PHI
From the Booker Period to the Gall Period, Philadelphia’s
Gall

total spread increased from 33.8 to 53.7, and its
standard deviation increased from 9.6 to 13.5.
From the Gall Period to the Post-Report Period,
Philadelphia’s total spread increased from 53.7 to 63.8,

-30.0

-20.0

-10.0

City
Avg
0.0
-11.9%

10.0

20.0

30.0

40.0

City
Avg
0.0
-14.6%

10.0

20.0

30.0

40.0

City
0.0Avg
-18.4%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 16
Cases: 1,004
Spread: 53.7
Standard Deviation: 13.5

and its standard deviation increased from 13.5 to 13.6.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

PHI
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%

Judges: 18
Cases: 1,314
Spread: 63.8
Standard Deviation: 13.6

-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

52

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

PHX
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 10
Cases: 1,843
Spread: 8.7
Standard Deviation: 2.8

PHOENIX-BASED
FEDERAL JUDGES
DISTRICT OF ARIZONA

The Commission analyzed 7,915 cases from Phoenix
-40.0

across the three periods.
PHX
From the Booker Period to the Gall Period, Phoenix’s
Gall

total spread increased from 8.7 to 25.7, and its standard
deviation increased from 2.8 to 7.1.
From the Gall Period to the Post-Report Period,
Phoenix’s total spread increased from 25.7 to 27.5, and

-30.0

-20.0

-10.0

City
Avg
0.0
-3.7%

10.0

20.0

30.0

40.0

City
Avg
0.0
-7.4%

10.0

20.0

30.0

40.0

City
Avg
0.0
-10.1%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 11
Cases: 2,330
Spread: 25.7
Standard Deviation: 7.1

its standard deviation increased from 7.1 to 7.9.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

PHX
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 13
Cases: 3,742
Spread: 27.5
Standard Deviation: 7.9

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

53

Intra-City Differences in Federal Sentencing Practices
January 2019

PITT
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 9
Cases: 629
Spread: 19.7
Standard Deviation: 5.9

PITTSBURGH-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF PENNSYLVANIA

The Commission analyzed 2,770 cases from Pittsburgh
-40.0

across the three periods.
PITT
From the Booker Period to the Gall Period, Pittsburgh’s
Gall

total spread increased from 19.7 to 32.5, and its
standard deviation increased from 5.9 to 9.4.
From the Gall Period to the Post-Report Period,
Pittsburgh’s total spread decreased from 32.5 to 19.8,

-30.0

-20.0

-10.0

City
0.0Avg
-1.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-13.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.9%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 10
Cases: 854
Spread: 32.5
Standard Deviation: 9.4

and its standard deviation decreased from 9.4 to 6.9.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

PITT
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 10
Cases: 1,287
Spread: 19.8
Standard Deviation: 6.9

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

54

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

PORT
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 5
Cases: 586
Spread: 8.3
Standard Deviation: 3.3

PORTLAND-BASED
FEDERAL JUDGES
DISTRICT OF OREGON

The Commission analyzed 3,314 cases from Portland
-40.0

across the three periods.
PORT
From the Booker Period to the Gall Period, Portland’s
Gall

total spread increased from 8.3 to 14.7, and its standard
deviation increased from 3.3 to 5.8.
From the Gall Period to the Post-Report Period,
Portland’s total spread increased from 14.7 to 24.6,

-30.0

-20.0

-10.0

City
0.0Avg
-11.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-21.6%

10.0

20.0

30.0

40.0

City
0.0Avg
-29.2%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 1,017
Spread: 14.7
Standard Deviation: 5.8

and its standard deviation increased from 5.8 to 7.1.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

PORT
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 10
Cases: 1,711
Spread: 24.6
Standard Deviation: 7.1

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

55

Intra-City Differences in Federal Sentencing Practices
January 2019

STL
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 8
Cases: 1,569
Spread: 11.5
Standard Deviation: 4.2

SAINT LOUIS-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF MISSOURI

The Commission analyzed 6,007 cases from Saint
-40.0

Louis across the three periods.
STL
From the Booker Period to the Gall Period, Saint Louis’
Gall

total spread increased from 11.5 to 19.6, and its
standard deviation increased from 4.2 to 5.8.
From the Gall Period to the Post-Report Period, Saint
Louis’ total spread decreased from 19.6 to 10.4, and its

-30.0

-20.0

-10.0

City
Avg
0.0
-2.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-13.4%

10.0

20.0

30.0

40.0

City
Avg
0.0
-20.0%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 8
Cases: 2,026
Spread: 19.6
Standard Deviation: 5.8

standard deviation decreased from 5.8 to 3.6.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

STL
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 9
Cases: 2,412
Spread: 10.4
Standard Deviation: 3.6

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

56

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

SLC
Booker

KEY FINDINGS

SALT LAKE CITY-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 8
Cases: 1,504
Spread: 7.8
Standard Deviation: 3.0

DISTRICT OF UTAH

The Commission analyzed 4,665 cases from Salt Lake
-40.0

City across the three periods.
From the Booker Period to the Gall Period, SaltSLC
Lake
Gall

City’s total spread increased from 7.8 to 14.7, and its
standard deviation increased from 3.0 to 5.6.
From the Gall Period to the Post-Report Period, Salt
Lake City’s total spread increased from 14.7 to 20.1,

-30.0

-20.0

-10.0

City
Avg
0.0
-6.0%

10.0

20.0

30.0

40.0

City
0.0Avg
-15.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-29.0%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 6
Cases: 1,428
Spread: 14.7
Standard Deviation: 5.6

and its standard deviation increased from 5.6 to 5.8.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

SLC
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 9
Cases: 1,733
Spread: 20.1
Standard Deviation: 5.8

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

57

Intra-City Differences in Federal Sentencing Practices
January 2019

SANT
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 3
Cases: 665
Spread: 7.5

SAN ANTONIO-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF TEXAS

The Commission analyzed 5,198 cases from San
-40.0

Antonio across the three periods.
SANT
From the Booker Period to the Gall Period, San Antonio’s

total spread decreased from 7.5 to 7.1.

Gall

From the Gall Period to the Post-Report Period, San
Antonio’s total spread increased from 7.1 to 12.2.

-30.0

-20.0

-10.0

City
Avg
0.0
-6.1%

10.0

20.0

30.0

40.0

-10.0

City
Avg
0.0
-18.3%

10.0

20.0

30.0

40.0

City
0.0Avg
-15.2%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 3
Cases: 1,481
Spread: 7.1

Because San Antonio did not have at least five judges
in all three periods, the standard deviations for the
three periods are not reported.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

SANT
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%

Judges: 4
Cases: 3,052
Spread: 12.2

-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

58

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

SD
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 9
Cases: 2,295
Spread: 13.9
Standard Deviation: 4.5

SAN DIEGO-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF CALIFORNIA

The Commission analyzed 11,960 cases from San
-40.0

Diego across the three periods.
SD
From the Booker Period to the Gall Period, San Diego’s
Gall

total spread increased from 13.9 to 21.3, and its
standard deviation increased from 4.5 to 6.4.
From the Gall Period to the Post-Report Period, San
Diego’s total spread increased from 21.3 to 54.4, and

-30.0

-20.0

-10.0

City
0.0Avg
-6.5%

10.0

20.0

30.0

40.0

City
Avg
0.0
-9.5%

10.0

20.0

30.0

40.0

City
0.0Avg
-23.8%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 14
Cases: 5,288
Spread: 21.3
Standard Deviation: 6.4

its standard deviation increased from 6.4 to 13.0.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

SD
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 16
Cases: 4,377
Spread: 54.4
Standard Deviation: 13.0

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

59

Intra-City Differences in Federal Sentencing Practices
January 2019

SF
Booker

KEY FINDINGS

SAN FRANCISCO-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 6
Cases: 384
Spread: 10.0
Standard Deviation: 4.3

NORTHERN DISTRICT OF CALIFORNIA

The Commission analyzed 1,880 cases from San
-40.0

Francisco across the three periods.
From the Booker Period to the Gall Period,SF San
Gall

Francisco’s total spread increased from 10.0 to 18.3,
and its standard deviation increased from 4.3 to 7.4.
From the Gall Period to the Post-Report Period, San
Francisco’s total spread decreased from 18.3 to 17.9,

-30.0

-20.0

-10.0

City
0.0Avg
-9.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-12.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-23.8%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 5
Cases: 567
Spread: 18.3
Standard Deviation: 7.4

and its standard deviation decreased from 7.4 to 6.1.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

SF
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 8
Cases: 929
Spread: 17.9
Standard Deviation: 6.1

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

60

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

KEY FINDINGS

SAN
JUAN
Booker

BOOKER PERIOD
Judges: 7
Cases: 923
Spread: 9.6
Standard Deviation: 3.6

SAN JUAN-BASED
FEDERAL JUDGES
DISTRICT OF PUERTO RICO

The Commission analyzed 6,460 cases from San Juan
-40.0

across the three periods.
SAN
From the Booker Period to the Gall Period, San Juan’s
JUAN

total spread increased from 9.6 to 20.9, and its standard
Gall
deviation increased from 3.6 to 7.2.
From the Gall Period to the Post-Report Period, San
Juan’s total spread increased from 20.9 to 22.2, and its

-30.0

-20.0

-10.0

City
Avg
0.0
0.7%

10.0

20.0

30.0

40.0

City
Avg
0.0
-4.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
1.2%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 8
Cases: 1,524
Spread: 20.9
Standard Deviation: 7.2

standard deviation increased from 7.2 to 7.4.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

-10.0

SAN
JUAN
Post

POST-REPORT PERIOD
Judges: 9
Cases: 4,013
Spread: 22.2
Standard Deviation: 7.4

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

61

Intra-City Differences in Federal Sentencing Practices
January 2019

SEA
Booker

KEY FINDINGS

BOOKER PERIOD
Judges: 7
Cases: 1,108
Spread: 10.4
Standard Deviation: 3.9

SEATTLE-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF WASHINGTON

The Commission analyzed 3,689 cases from Seattle
-40.0

across the three periods.
From the Booker Period to the Gall Period, Seattle’sSEA
total
Gall

spread decreased from 10.4 to 4.9, and its standard
deviation decreased from 3.9 to 1.6.
From the Gall Period to the Post-Report Period,
Seattle’s total spread increased from 4.9 to 21.1, and

-30.0

-20.0

-10.0

City
Avg
0.0
-20.5%

10.0

20.0

30.0

40.0

City
0.0Avg
-22.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-31.3%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 7
Cases: 1,134
Spread: 4.9
Standard Deviation: 1.6

its standard deviation increased from 1.6 to 7.1.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

SEA
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 7
Cases: 1,447
Spread: 21.1
Standard Deviation: 7.1

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

62

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX B: City-by-City Results

TAMPA
Booker

KEY FINDINGS

TAMPA-BASED FEDERAL
JUDGES

BOOKER PERIOD
Judges: 7
Cases: 1,283
Spread: 11.9
Standard Deviation: 3.7

MIDDLE DISTRICT OF FLORIDA

The Commission analyzed 5,279 cases from Tampa
-40.0

across the three periods.
From the Booker Period to the Gall Period, Tampa’sTAMPA
total
Gall

spread increased from 11.9 to 29.7, and its standard
deviation increased from 3.7 to 9.6.
From the Gall Period to the Post-Report Period,
Tampa’s total spread decreased from 29.7 to 22.7, and

-30.0

-20.0

-10.0

City
0.0Avg
-0.2%

10.0

20.0

30.0

40.0

City
0.0Avg
-4.7%

10.0

20.0

30.0

40.0

City
Avg
0.0
-9.6%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 9
Cases: 1,567
Spread: 29.7
Standard Deviation: 9.6

its standard deviation decreased from 9.6 to 5.9.

-40.0

Legend

-30.0

-20.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-10.0

TAMPA
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 10
Cases: 2,429
Spread: 22.7
Standard Deviation: 5.9

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

63

The Commission
analyzed whether
random assignment of
cases resulted in even
distribution of offense
types among judges.

64

|

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

APPENDIX C

Caseload Compositions of
30 Cities

	

As discussed above, a key reason for

fairly even distribution of the primary four guideline

studying differences in sentencing practices of judges

types—§§2B1.1 (fraud and theft offenses), 2D1.1

within a single city—as opposed to judges in a larger

(drug-trafficking offenses), 2K2.1 (firearm offenses),

geographical region, such as a federal judicial district

and 2L1.2 (illegal reentry offenses)—as well as all

with multiple divisions—is that, as a general practice,

other offense types (as a catch-all fifth category).56

federal district judges within a single city are randomly

This appendix contains data regarding all 30 cities’

assigned cases from the same pool of cases. Assuming

judges’ caseload compositions in each period, as well

a sufficiently large enough number of cases per judge,

as data about each city’s total caseload over the three

random assignment should generally result in each

periods. It shows that for most judges in most cities,

judge in a city having a roughly even distribution of

their caseload compositions were generally similar.

guideline offense types.

The Commission selected

It likewise shows that most cities had comparable

the 50-case minimum per judge requirement for

caseload compositions over the three periods.57 Judges

each period studied for the purpose of increasing the

are identified only by a four-digit number assigned by

likelihood that case assignment resulted in randomly

the Commission for research purposes.

distributed guideline offense types.

	

	

The

Commission

analyzed

each

There

were

some

exceptions,

however.

city’s

As a result, as discussed in Appendix D (p. 101),

sentencing data in each period to determine if random

the Commission conducted weighting analyses to

distribution resulted in a generally even distribution

determine whether those differences substantially

of cases—both among judges in each city in the

contributed to differences in caseload compositions

three periods and across the three periods in each

among the 30 cities’ judges in terms of their average

city with respect to the city’s total caseload. In the

percent differences or corresponding differences in

majority of cities, random assignment resulted in

the 30 cities’ total spreads and standard deviations.

65

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

ALEXANDRIA-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF VIRGINIA
Appendix B: Judge's Caseload by Guideline
ALEXANDRIA
USSG §2B1.1
17.2%
16.7%
19.7%

USSG §2D1.1
41.7%
36.2%
31.5%

18.5%
18.0%
16.7%
15.7%
16.6%

39.7%
42.9%
39.1%
47.0%
41.4%

6.0%
6.2%
2.9%
4.3%
4.8%

9.9%
6.8%
4.6%
5.2%
6.9%

25.8%
26.1%
36.8%
27.8%
30.3%

4298
4618
4947
4995
5909
6058
6194

17.2%
12.9%
16.0%
17.6%
21.8%
12.4%
18.0%

27.2%
37.1%
41.7%
34.3%
29.1%
49.6%
37.1%

7.2%
2.1%
4.0%
8.8%
6.7%
4.7%
6.6%

12.8%
20.7%
16.0%
14.7%
15.8%
13.2%
10.2%

35.6%
27.1%
22.3%
24.5%
26.7%
20.2%
28.1%

Post-Report Period Judges
4298
4618
4947
4995
5909
6058
6194

18.1%
17.9%
10.9%
23.4%
19.0%
19.0%
30.8%

29.5%
40.0%
35.6%
30.1%
25.1%
35.4%
23.7%

8.6%
2.1%
9.7%
4.8%
4.6%
6.1%
7.6%

12.4%
10.4%
7.3%
16.7%
13.8%
10.9%
4.7%

31.4%
29.6%
36.4%
24.9%
37.4%
28.6%
33.2%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4298
4947
5909
6058
6194

Gall Period Judges

66

|

https://www.ussc.gov/topic/research-reports

USSG §2K2.1
4.8%
5.7%
6.2%

USSG §2L1.2
6.7%
14.7%
10.8%

OTHER
29.6%
26.7%
31.8%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

ATLANTA-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF GEORGIA
Appendix B: Judge's Caseload by Guideline
ATLANTA
USSG §2B1.1
18.5%
14.7%
17.9%

USSG §2D1.1
27.8%
24.9%
26.3%

USSG §2K2.1
15.2%
12.4%
13.0%

USSG §2L1.2
8.1%
18.9%
15.7%

OTHER
30.4%
29.1%
27.2%

20.7%
5.5%
21.8%
24.5%
17.6%
13.8%
19.2%
21.5%
19.2%
15.3%
19.6%
21.6%

18.4%
34.2%
20.7%
18.9%
37.3%
26.6%
28.8%
28.0%
34.6%
22.0%
36.3%
25.0%

14.9%
27.4%
14.9%
13.2%
17.6%
14.9%
17.3%
12.9%
12.5%
10.2%
15.7%
11.4%

3.4%
19.2%
8.0%
11.3%
7.8%
4.3%
7.7%
3.2%
3.8%
16.9%
5.9%
13.6%

42.5%
13.7%
34.5%
32.1%
19.6%
40.4%
26.9%
34.4%
29.8%
35.6%
22.5%
28.4%

4045
4219
4515
4982
5071
5243
5324
5627
5981
6027
6082
6210
6217
6312

4.2%
9.6%
17.3%
21.1%
13.9%
7.0%
8.7%
32.0%
25.8%
14.6%
12.3%
20.8%
14.2%
5.3%

26.4%
11.3%
37.8%
18.3%
16.7%
30.1%
40.6%
20.0%
31.1%
18.5%
25.2%
9.4%
25.5%
31.6%

9.7%
18.3%
6.3%
6.4%
15.3%
14.0%
18.8%
18.7%
6.1%
12.1%
8.6%
24.5%
12.1%
19.7%

20.8%
40.0%
15.0%
19.3%
26.4%
21.0%
11.6%
4.0%
15.2%
15.9%
19.6%
26.4%
17.7%
9.2%

38.9%
20.9%
23.6%
34.9%
27.8%
28.0%
20.3%
25.3%
22.0%
38.9%
34.4%
18.9%
30.5%
34.2%

Post-Report Period Judges
4219
4413
4515
4735
4982
5155
5222
5243
5324
5556
5981
6027
6082
6149
6217

11.3%
15.8%
7.4%
20.0%
17.8%
17.7%
36.4%
17.4%
11.8%
21.1%
14.6%
19.3%
19.3%
33.3%
13.1%

12.7%
29.8%
34.6%
23.9%
36.3%
23.9%
13.6%
24.9%
31.4%
10.5%
31.7%
29.2%
24.2%
13.6%
26.8%

7.0%
21.1%
21.0%
11.7%
10.7%
14.4%
9.1%
10.8%
9.8%
12.3%
9.8%
14.3%
13.5%
9.1%
19.2%

56.3%
14.0%
8.6%
15.2%
8.5%
12.9%
19.7%
19.7%
23.5%
36.8%
15.9%
13.0%
12.1%
21.2%
8.9%

12.7%
19.3%
28.4%
29.1%
26.7%
31.1%
21.2%
27.2%
23.5%
19.3%
28.0%
24.2%
30.9%
22.7%
31.9%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4045
4219
4515
5243
5556
5627
5981
6027
6082
6210
6217
6312

Gall Period Judges

67

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

BALTIMORE-BASED
FEDERAL JUDGES
DISTRICT OF MARYLAND
Appendix B: Judge's Caseload by Guideline
BALTIMORE
USSG §2B1.1
9.1%
9.8%
11.9%

USSG §2D1.1
36.4%
36.3%
39.3%

USSG §2K2.1
22.8%
16.6%
15.7%

6.7%
12.7%
6.3%
13.4%
9.4%
6.3%

37.3%
30.4%
45.0%
32.0%
34.9%
37.8%

20.0%
21.6%
21.6%
27.8%
19.8%
25.2%

5.3%
7.8%
7.2%
6.2%
4.7%
7.9%

30.7%
27.5%
19.8%
20.6%
31.1%
22.8%

4240
4246
4296
5069
5848
6166
6200

5.6%
12.2%
7.9%
4.8%
11.7%
25.4%
10.9%

35.2%
36.7%
44.7%
35.3%
41.7%
.
.
27.3%

18.5%
15.4%
15.8%
21.0%
20.9%
.
.
12.7%

5.6%
4.3%
3.9%
4.2%
3.7%
18.6%
14.5%

35.2%
31.4%
27.6%
34.7%
22.1%
55.9%
34.5%

Post-Report Period Judges
4240
4246
4296
4538
4691
4902
5069
5313
5848

10.5%
9.4%
11.0%
10.9%
9.2%
7.6%
14.8%
29.5%
14.4%

33.3%
48.8%
45.4%
42.0%
34.9%
37.4%
41.8%
8.2%
33.0%

12.3%
17.2%
10.6%
15.5%
17.1%
21.0%
17.5%
1.6%
15.4%

3.5%
2.0%
4.1%
4.6%
5.3%
6.7%
2.3%
1.6%
4.6%

40.4%
22.7%
28.9%
26.9%
33.6%
27.3%
23.6%
59.0%
32.6%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4240
4246
4296
5069
5848
6200

Gall Period Judges

68

|

https://www.ussc.gov/topic/research-reports

USSG §2L1.2
6.6%
5.7%
4.0%

OTHER
25.1%
31.6%
29.1%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

BOSTON-BASED
FEDERAL JUDGES
DISTRICT OF MASSACHUSETTS
Appendix B: Judge's Caseload by Guideline
BOSTON
USSG §2B1.1
8.7%
12.9%
14.5%

USSG §2D1.1
39.5%
37.6%
34.6%

USSG §2K2.1
10.2%
12.2%
11.5%

USSG §2L1.2
8.1%
8.0%
9.3%

OTHER
33.5%
29.2%
30.1%

4.7%
10.9%
9.8%
14.1%
6.0%
6.1%
4.7%
9.3%
9.1%
5.5%
9.1%
13.8%

28.1%
43.6%
45.1%
42.4%
53.0%
59.1%
34.4%
38.7%
27.3%
41.8%
29.9%
25.9%

9.4%
12.7%
3.7%
2.4%
8.4%
12.1%
15.6%
10.7%
9.1%
12.7%
15.6%
15.5%

10.9%
5.5%
3.7%
8.2%
6.0%
7.6%
14.1%
5.3%
13.6%
5.5%
10.4%
6.9%

46.9%
27.3%
37.8%
32.9%
26.5%
15.2%
31.3%
36.0%
40.9%
34.5%
35.1%
37.9%

4007
4167
4321
4421
4736
4812
4850
4874
5784
5958
6340

23.7%
11.0%
10.4%
14.1%
12.2%
12.4%
5.7%
11.4%
18.6%
10.4%
14.3%

33.0%
51.0%
34.4%
27.1%
44.9%
43.8%
36.8%
36.0%
21.6%
36.8%
39.3%

12.4%
10.3%
17.7%
20.0%
11.2%
6.6%
11.3%
14.9%
15.5%
8.5%
9.8%

5.2%
6.9%
6.3%
10.6%
9.2%
9.1%
8.5%
7.0%
10.3%
8.5%
7.1%

25.8%
20.7%
31.3%
28.2%
22.4%
28.1%
37.7%
30.7%
34.0%
35.8%
29.5%

Post-Report Period Judges
4007
4135
4321
4421
4736
4812
4850
4874
5784
5958
6340

23.0%
13.7%
14.6%
13.9%
14.5%
14.9%
13.5%
14.9%
14.0%
16.4%
8.8%

34.1%
26.6%
25.5%
39.8%
35.5%
36.3%
32.4%
34.3%
33.5%
32.1%
46.4%

13.3%
13.7%
12.7%
9.0%
10.1%
16.1%
10.1%
6.0%
10.1%
14.9%
8.8%

5.2%
7.9%
7.6%
13.3%
7.2%
8.3%
11.5%
13.4%
7.3%
10.4%
11.6%

24.4%
38.1%
39.5%
24.1%
32.6%
24.4%
32.4%
31.3%
35.2%
26.1%
24.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4007
4167
4321
4421
4443
4736
4812
4850
4874
5784
5958
6340

Gall Period Judges

69

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

CHARLOTTE-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF NORTH CAROLINA
Appendix B: Judge's Caseload by Guideline
CHARLOTTE
USSG §2D1.1
37.7%
35.7%
31.6%

USSG §2K2.1
28.9%
18.7%
18.0%

USSG §2L1.2
8.7%
12.0%
10.9%

OTHER
18.3%
19.2%
26.8%

5.5%
4.7%
9.2%
4.9%

24.8%
40.6%
29.3%
49.8%

34.9%
30.5%
35.4%
19.3%

12.8%
7.0%
11.8%
4.9%

22.0%
17.2%
14.4%
21.0%

4541
5437
5442

20.3%
11.7%
9.9%

30.1%
29.1%
50.8%

16.7%
22.6%
16.8%

11.8%
14.8%
8.9%

21.1%
21.7%
13.6%

Post-Report Period Judges
4541
5437
5442

17.2%
14.1%
5.5%

26.0%
23.1%
49.5%

18.0%
19.6%
15.8%

11.8%
12.5%
7.9%

27.0%
30.7%
21.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4541
5064
5437
5442

Gall Period Judges

70

|

USSG §2B1.1
6.3%
14.4%
12.7%

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

CHICAGO-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF ILLINOIS
Appendix B: Judge's Caseload by Guideline
CHICAGO
USSG §2B1.1
17.4%
21.2%
27.7%

USSG §2D1.1
36.2%
32.1%
29.3%

USSG §2K2.1
5.7%
4.5%
7.9%

USSG §2L1.2
7.8%
14.3%
10.5%

OTHER
33.0%
27.9%
24.6%

21.1%
11.9%
12.2%
18.3%
15.3%
21.3%
11.1%
19.5%
13.7%
8.8%
25.9%
21.2%
13.1%
21.5%
25.6%
19.4%
16.5%
16.9%

26.8%
32.1%
44.6%
39.4%
37.5%
37.1%
48.9%
42.9%
39.7%
29.8%
20.7%
47.0%
42.6%
30.4%
33.3%
19.4%
35.2%
37.3%

4.2%
11.9%
4.1%
4.2%
5.6%
5.6%
6.7%
5.2%
6.8%
.
.
5.2%
1.5%
8.2%
7.6%
3.8%
6.0%
7.7%
4.8%

7.0%
9.5%
9.5%
11.3%
8.3%
6.7%
6.7%
9.1%
.
.
7.0%
8.6%
4.5%
9.8%
10.1%
9.0%
11.9%
6.6%
6.0%

40.8%
34.5%
29.7%
26.8%
33.3%
29.2%
26.7%
23.4%
39.7%
54.4%
39.7%
25.8%
26.2%
30.4%
28.2%
43.3%
34.1%
34.9%

4076
4123
4143
4336
4506
4516
4643
4649
4763
4839
4841
4989
5032
5135
5430
5447
5611
5639
5746
5899
6131
6158

18.3%
29.9%
31.4%
19.8%
19.8%
32.2%
14.0%
12.0%
19.2%
17.0%
14.8%
26.1%
31.8%
9.3%
26.8%
16.4%
18.6%
33.3%
19.3%
19.8%
17.0%
29.6%

44.2%
35.9%
25.5%
39.6%
31.3%
19.5%
32.3%
38.7%
39.4%
40.9%
28.1%
14.5%
21.2%
41.1%
17.9%
28.8%
29.9%
23.4%
38.5%
39.6%
33.0%
19.7%

0.8%
0.9%
5.9%
2.8%
6.3%
10.3%
23.7%
4.0%
2.9%
4.5%
0.8%
8.7%
3.5%
4.7%
3.6%
5.5%
2.1%
2.7%
3.7%
3.1%
1.1%
4.2%

13.3%
10.3%
11.8%
8.5%
17.7%
13.8%
11.8%
16.0%
9.6%
12.5%
15.6%
21.7%
12.9%
13.1%
26.8%
20.5%
16.5%
13.5%
11.0%
15.6%
13.8%
18.3%

23.3%
23.1%
25.5%
29.2%
25.0%
24.1%
18.3%
29.3%
28.8%
25.0%
40.6%
29.0%
30.6%
31.8%
25.0%
28.8%
33.0%
27.0%
27.5%
21.9%
35.1%
28.2%

Post-Report Period Judges
4123
4143
4188
4336
4506
4516
4643
4649

USSG §2B1.1
29.7%
31.1%
24.8%
23.5%
24.8%
32.6%
23.3%
35.5%

USSG §2D1.1
33.1%
14.8%
23.4%
34.6%
39.7%
27.1%
26.2%
16.4%

USSG §2K2.1
5.1%
14.8%
8.8%
8.1%
6.6%
3.5%
16.3%
7.9%

USSG §2L1.2
11.9%
18.0%
9.5%
5.9%
8.3%
12.5%
15.1%
10.5%

OTHER
20.3%
21.3%
33.6%
27.9%
20.7%
24.3%
19.2%
29.6%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4076
4123
4336
4506
4513
4649
4763
4839
4841
4989
5032
5107
5135
5447
5639
5746
5899
6131

Gall Period Judges

71

4143
4336
4506
4516
4643
4649
4763
4839
4841
4989
5032
5135
5430
5447
5611
5639
5746
5899
6131
6158

Intra-City
January 2019

31.4%
19.8%
19.8%
32.2%
Differences
in
14.0%
12.0%
19.2%
17.0%
14.8%
26.1%
31.8%
9.3%
26.8%
16.4%
18.6%
33.3%
19.3%
19.8%
17.0%
29.6%

Federal

25.5%
39.6%
31.3%
19.5%
Sentencing
32.3%
38.7%
39.4%
40.9%
28.1%
14.5%
21.2%
41.1%
17.9%
28.8%
29.9%
23.4%
38.5%
39.6%
33.0%
19.7%

Practices

5.9%
2.8%
6.3%
10.3%
23.7%
4.0%
2.9%
4.5%
0.8%
8.7%
3.5%
4.7%
3.6%
5.5%
2.1%
2.7%
3.7%
3.1%
1.1%
4.2%

11.8%
8.5%
17.7%
13.8%
11.8%
16.0%
9.6%
12.5%
15.6%
21.7%
12.9%
13.1%
26.8%
20.5%
16.5%
13.5%
11.0%
15.6%
13.8%
18.3%

25.5%
29.2%
25.0%
24.1%
18.3%
29.3%
28.8%
25.0%
40.6%
29.0%
30.6%
31.8%
25.0%
28.8%
33.0%
27.0%
27.5%
21.9%
35.1%
28.2%

USSG §2K2.1
5.1%
14.8%
8.8%
8.1%
6.6%
3.5%
16.3%
7.9%
8.8%
5.8%
6.9%
7.6%
6.1%
2.4%
9.0%
3.3%
15.1%
6.5%
6.8%
9.1%
9.3%
10.9%
4.7%
8.9%

USSG §2L1.2
11.9%
18.0%
9.5%
5.9%
8.3%
12.5%
15.1%
10.5%
9.6%
9.1%
8.3%
12.2%
16.7%
10.7%
7.7%
15.7%
11.0%
9.8%
10.2%
12.1%
7.8%
8.9%
9.4%
8.9%

OTHER
20.3%
21.3%
33.6%
27.9%
20.7%
24.3%
19.2%
29.6%
30.9%
32.5%
21.5%
29.0%
21.2%
17.9%
28.8%
24.0%
20.5%
20.3%
23.7%
25.8%
23.3%
35.6%
17.2%
17.2%

CHICAGO-BASED FEDERAL JUDGES continued
Post-Report Period Judges
4123
4143
4188
4336
4506
4516
4643
4649
4760
4763
4839
4841
4989
5032
5135
5430
5486
5611
5639
5694
5746
5969
6106
6131

72

|

USSG §2B1.1
29.7%
31.1%
24.8%
23.5%
24.8%
32.6%
23.3%
35.5%
35.3%
26.0%
28.5%
26.0%
22.7%
20.2%
16.7%
24.8%
28.8%
35.0%
25.4%
22.7%
26.4%
28.7%
30.5%
35.0%

https://www.ussc.gov/topic/research-reports

USSG §2D1.1
33.1%
14.8%
23.4%
34.6%
39.7%
27.1%
26.2%
16.4%
15.4%
26.6%
34.7%
25.2%
33.3%
48.8%
37.8%
32.2%
24.7%
28.5%
33.9%
30.3%
33.3%
15.8%
38.3%
30.0%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

CLEVELAND-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF OHIO
Appendix B: Judge's Caseload by Guideline
CLEVELAND
USSG §2B1.1
15.8%
17.6%
14.7%

USSG §2D1.1
33.9%
21.8%
30.0%

USSG §2K2.1
21.3%
29.8%
24.2%

12.0%
22.6%
33.3%
17.7%
11.3%
13.7%
17.5%
10.7%
11.9%

40.7%
28.3%
23.8%
31.5%
33.0%
30.4%
28.9%
43.0%
36.6%

16.7%
17.0%
17.9%
19.4%
33.0%
26.5%
27.8%
13.4%
21.8%

4471
4688
5086
5235
5408
5489
5591
6049

15.2%
16.3%
25.0%
13.8%
16.0%
28.4%
17.3%
9.2%

26.6%
27.6%
8.7%
29.4%
18.5%
10.2%
30.9%
18.4%

19.0%
32.7%
35.9%
31.2%
33.3%
18.2%
30.9%
35.6%

8.9%
2.0%
6.5%
2.8%
6.2%
2.3%
2.7%
10.3%

30.4%
21.4%
23.9%
22.9%
25.9%
40.9%
18.2%
26.4%

Post-Report Period Judges
4688
5086
5235
5408
5591
6049

13.2%
11.2%
16.2%
16.3%
13.0%
16.4%

34.3%
33.6%
37.1%
25.4%
19.5%
29.0%

21.9%
29.3%
22.8%
21.5%
25.5%
27.1%

1.7%
3.4%
0.8%
2.9%
1.5%
2.8%

28.9%
22.4%
23.2%
34.0%
40.5%
24.8%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4471
4517
4688
5086
5235
5408
5489
5591
6049

Gall Period Judges

USSG §2L1.2
3.0%
5.0%
2.0%

.

2.8%
7.5%
8.3%
4.8%
0.9%
1.0%
2.0%
3.0%

.

OTHER
26.0%
25.8%
29.1%
27.8%
24.5%
16.7%
26.6%
21.7%
28.4%
25.8%
30.9%
26.7%

73

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

COLUMBUS-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF OHIO
Appendix B: Judge's Caseload by Guideline
COLUMBUS
USSG §2B1.1
17.7%
14.9%
12.0%

USSG §2D1.1
36.9%
32.0%
29.8%

USSG §2K2.1
17.0%
10.6%
16.6%

22.6%
16.7%
26.0%
19.2%
7.5%

45.3%
28.3%
26.0%
44.2%
40.3%

18.9%
15.0%
16.0%
11.5%
22.4%

4285
4301
5595
5771
5930
6261

22.0%
13.3%
14.3%
13.8%
9.8%
18.0%

37.8%
41.3%
29.8%
26.3%
27.2%
30.0%

7.3%
6.7%
16.7%
12.5%
9.8%
10.0%

6.1%
10.7%
10.7%
12.5%
10.9%
10.0%

26.8%
28.0%
28.6%
35.0%
42.4%
32.0%

Post-Report Period Judges
4285
4301
5462
5595
5771
5930
6261

7.8%
11.6%
8.4%
16.2%
14.5%
10.8%
15.9%

33.0%
34.1%
26.5%
29.1%
24.8%
31.8%
22.7%

17.9%
17.7%
18.1%
10.8%
14.5%
20.9%
15.9%

6.1%
10.4%
10.8%
7.4%
6.8%
4.7%
10.2%

35.2%
26.2%
36.1%
36.5%
39.3%
31.8%
35.2%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4285
4301
5595
5771
5930

Gall Period Judges

74

|

https://www.ussc.gov/topic/research-reports

USSG §2L1.2
3.5%
10.2%
7.8%

.

3.8%
8.3%
2.0%
3.0%

.

OTHER
24.8%
32.4%
33.9%
9.4%
31.7%
30.0%
25.0%
26.9%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

DALLAS-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF TEXAS
Appendix B: Judge's Caseload by Guideline
DALLAS
USSG §2B1.1
24.2%
24.4%
13.1%

USSG §2D1.1
19.8%
19.5%
33.6%

USSG §2K2.1
13.0%
11.6%
15.1%

USSG §2L1.2
12.1%
15.0%
19.6%

OTHER
30.8%
29.4%
18.6%

17.6%
30.9%
23.7%
18.3%
31.9%
27.0%
24.0%
21.0%

25.3%
17.0%
31.6%
13.4%
12.5%
27.0%
22.7%
8.6%

29.7%
9.6%
9.2%
12.2%
11.1%
9.0%
8.0%
13.6%

6.6%
12.8%
10.5%
18.3%
13.9%
9.0%
8.0%
18.5%

20.9%
29.8%
25.0%
37.8%
30.6%
28.1%
37.3%
38.3%

4121
5058
5525
5775
5879
6281
6349

11.5%
22.7%
34.5%
35.7%
27.3%
22.2%
17.9%

42.7%
13.5%
9.2%
12.4%
14.1%
20.5%
23.4%

7.6%
18.4%
10.6%
13.2%
6.1%
9.4%
13.8%

14.5%
10.6%
13.4%
14.0%
24.2%
16.2%
15.2%

23.7%
34.8%
32.4%
24.8%
28.3%
31.6%
29.7%

Post-Report Period Judges
4121
5058
5525
5775
5879
6281
6349

11.4%
14.6%
16.4%
11.9%
12.8%
10.9%
13.4%

38.2%
29.8%
31.7%
32.2%
37.3%
26.5%
37.2%

15.5%
18.9%
17.6%
17.7%
4.7%
22.4%
11.8%

15.5%
19.2%
16.4%
20.9%
24.5%
21.1%
19.4%

19.4%
17.5%
17.9%
17.4%
20.6%
19.2%
18.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4025
4121
4840
5058
5525
5775
6281
6349

Gall Period Judges

75

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

DENVER-BASED
FEDERAL JUDGES
DISTRICT OF COLORADO
Appendix B: Judge's Caseload by Guideline
DENVER
USSG §2D1.1
16.1%
13.6%
15.8%

USSG §2K2.1
19.0%
17.3%
25.1%

USSG §2L1.2
33.2%
40.6%
27.1%

OTHER
25.7%
22.1%
23.4%

7.2%
8.6%
4.6%
4.2%
3.8%
6.8%

30.4%
10.2%
10.8%
16.0%
15.2%
12.9%

15.2%
18.0%
16.9%
19.4%
22.8%
23.5%

22.5%
41.4%
34.6%
35.4%
34.2%
31.8%

24.6%
21.9%
33.1%
25.0%
24.1%
25.0%

4563
5045
5312
5725
6022
6154
6317

6.5%
6.4%
8.3%
3.6%
3.4%
10.4%
8.0%

9.1%
16.2%
5.2%
19.3%
10.2%
16.7%
16.0%

15.7%
18.4%
15.6%
15.7%
25.9%
16.7%
14.1%

47.4%
38.9%
43.8%
38.6%
45.6%
21.9%
40.4%

21.3%
20.1%
27.1%
22.9%
15.0%
34.4%
21.6%

Post-Report Period Judges
4307
4563
5045
5325
5725
5871
6022
6154
6317

6.5%
6.8%
8.7%
8.8%
11.1%
5.4%
7.8%
13.3%
9.1%

21.2%
8.5%
17.5%
14.0%
11.1%
16.2%
18.3%
6.0%
21.8%

23.4%
30.5%
27.9%
31.6%
22.2%
27.7%
23.7%
16.9%
23.4%

22.3%
25.4%
24.0%
28.7%
28.2%
25.4%
32.0%
28.9%
27.9%

26.6%
28.8%
21.9%
16.9%
27.4%
25.4%
18.3%
34.9%
17.8%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4100
5045
5088
5312
5625
6317

Gall Period Judges

76

|

USSG §2B1.1
6.0%
6.4%
8.6%

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

DETROIT-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF MICHIGAN
Appendix B: Judge's Caseload by Guideline
DETROIT
USSG §2B1.1
17.1%
18.6%
19.6%

USSG §2D1.1
30.7%
33.3%
21.6%

USSG §2K2.1
14.4%
13.1%
25.3%

12.7%
6.6%
24.2%
20.6%
16.7%
26.4%
15.4%
14.0%
23.0%
9.8%
20.0%

23.6%
23.0%
30.6%
23.8%
42.4%
20.8%
50.0%
34.9%
24.6%
32.8%
30.0%

25.5%
16.4%
9.7%
23.8%
10.6%
15.1%
13.5%
9.3%
18.0%
8.2%
11.7%

9.1%
9.8%
8.1%
4.8%
7.6%
9.4%
1.9%
1.2%
6.6%
4.9%
5.0%

29.1%
44.3%
27.4%
27.0%
22.7%
28.3%
19.2%
40.7%
27.9%
44.3%
33.3%

4046
4442
4459
4462
4601
4945
5037
5238
5400
5406
5504
5550

12.9%
6.9%
23.6%
34.8%
20.0%
17.0%
18.3%
26.8%
11.4%
24.2%
22.2%
14.1%

34.1%
44.6%
36.0%
30.4%
47.7%
34.0%
38.0%
28.0%
32.9%
20.0%
25.9%
31.0%

17.6%
13.9%
15.7%
14.5%
9.2%
20.8%
12.7%
12.2%
4.0%
10.5%
17.3%
18.3%

11.8%
3.0%
4.5%
2.9%
3.1%
3.8%
5.6%
8.5%
4.0%
7.4%
2.5%
5.6%

23.5%
31.7%
20.2%
17.4%
20.0%
24.5%
25.4%
24.4%
47.7%
37.9%
32.1%
31.0%

Post-Report Period Judges
4046
4442
4459
4462
4601
4762
5037
5238
5333
5400
5406
5504
5550
5904
5984
6080
6148
6153

18.6%
20.9%
27.2%
15.7%
20.8%
22.1%
10.5%
16.7%
20.8%
23.4%
22.4%
25.4%
25.4%
10.7%
21.2%
19.6%
21.9%
13.4%

16.8%
29.9%
11.6%
27.8%
24.5%
17.2%
36.8%
22.7%
23.4%
27.0%
25.9%
19.8%
26.8%
15.0%
28.5%
12.5%
18.0%
7.6%

22.1%
9.0%
14.3%
13.0%
13.2%
17.2%
14.7%
22.0%
32.5%
21.6%
16.4%
22.2%
16.7%
61.5%
17.9%
41.1%
21.9%
59.7%

2.7%
5.6%
6.1%
12.2%
11.3%
4.9%
6.3%
6.8%
5.2%
1.8%
7.8%
4.0%
5.8%
1.3%
5.3%
5.4%
9.4%
.
.

39.8%
34.5%
40.8%
31.3%
30.2%
38.5%
31.6%
31.8%
18.2%
26.1%
27.6%
28.6%
25.4%
11.5%
27.2%
21.4%
28.9%
19.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4046
4442
4459
4462
4601
5037
5238
5400
5406
5550
5984

Gall Period Judges

USSG §2L1.2
6.0%
5.2%
5.3%

OTHER
31.8%
29.8%
28.2%

77

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

HOUSTON-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF TEXAS
Appendix B: Judge's Caseload by Guideline
HOUSTON
USSG §2B1.1
7.8%
8.0%
11.4%

USSG §2D1.1
20.3%
10.9%
12.0%

USSG §2K2.1
17.2%
8.3%
8.1%

USSG §2L1.2
24.0%
51.2%
49.1%

OTHER
30.7%
21.6%
19.5%

11.6%
8.0%
7.2%
9.4%
5.2%
5.4%
7.3%
3.4%
10.6%
14.0%

24.2%
23.0%
24.7%
28.3%
22.2%
21.3%
14.6%
14.8%
9.4%
12.8%

22.1%
17.0%
18.6%
22.8%
16.3%
10.0%
16.7%
21.6%
22.4%
15.1%

17.9%
17.0%
25.8%
15.7%
14.8%
38.9%
19.8%
29.5%
20.0%
25.6%

24.2%
35.0%
23.7%
23.6%
41.5%
24.3%
41.7%
30.7%
37.6%
32.6%

4104
4166
4281
4370
4615
4674
4861
5111
5846
5857
6081

6.3%
6.5%
6.3%
7.1%
14.7%
9.7%
9.2%
5.8%
3.7%
13.7%
8.5%

9.2%
7.2%
11.7%
7.1%
11.5%
14.0%
5.5%
16.4%
14.6%
15.4%
4.5%

6.7%
7.2%
9.3%
7.1%
7.9%
8.2%
12.4%
9.1%
8.1%
9.4%
6.5%

60.5%
59.7%
46.8%
52.0%
46.6%
48.0%
49.5%
54.5%
52.8%
17.9%
62.3%

17.2%
19.4%
25.9%
26.8%
19.4%
20.1%
23.4%
14.2%
20.7%
43.6%
18.1%

Post-Report Period Judges
4104
4166
4281
4370
4615
4674
4861
5111
5846
5857
6081

16.7%
9.9%
9.2%
10.2%
11.4%
9.6%
12.4%
14.0%
6.7%
18.1%
10.4%

10.4%
13.2%
13.6%
10.2%
10.8%
13.3%
6.8%
11.9%
19.2%
18.8%
5.0%

5.6%
8.0%
6.5%
8.3%
8.5%
7.0%
6.4%
10.4%
9.9%
9.4%
9.6%

51.7%
44.3%
53.1%
53.1%
50.0%
54.2%
59.0%
47.5%
43.8%
6.5%
56.5%

15.6%
24.5%
17.7%
18.1%
19.3%
15.9%
15.4%
16.2%
20.4%
47.1%
18.5%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4104
4166
4281
4370
4615
4674
4861
5846
5857
6081

Gall Period Judges

78

|

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

LOS ANGELES-BASED
FEDERAL JUDGES
CENTRAL DISTRICT OF CALIFORNIA
Appendix B: Judge's Caseload by Guideline
LOS ANGELES
USSG §2B1.1
14.7%
18.2%
21.8%

USSG §2D1.1
25.8%
35.5%
37.0%

USSG §2K2.1
6.8%
5.7%
5.9%

USSG §2L1.2
15.9%
8.3%
7.1%

OTHER
36.8%
32.3%
28.2%

10.6%
16.1%
22.9%
10.8%
12.4%
10.5%
14.7%
12.5%
18.6%
11.5%
16.9%
25.4%
11.2%
19.1%
12.5%

41.5%
12.5%
21.4%
21.6%
25.8%
22.4%
29.3%
33.3%
34.0%
17.7%
32.6%
16.4%
36.0%
11.2%
22.5%

6.4%
8.9%
4.3%
6.9%
9.3%
10.5%
4.0%
10.4%
2.1%
8.3%
4.5%
4.5%
7.9%
9.0%
3.8%

11.7%
16.1%
21.4%
16.7%
17.5%
15.8%
17.3%
12.5%
17.5%
12.5%
18.0%
20.9%
10.1%
16.9%
16.3%

29.8%
46.4%
30.0%
44.1%
35.1%
40.8%
34.7%
31.3%
27.8%
50.0%
28.1%
32.8%
34.8%
43.8%
45.0%

4273
4592
4680
4787
5028
5157
5247
5435
5484
5533
5542
5646
5903
5919
5939
5943
6146
6280

12.7%
21.3%
17.9%
16.7%
11.1%
23.5%
9.0%
18.5%
16.3%
21.0%
28.6%
19.1%
24.6%
16.7%
16.3%
24.2%
18.9%
23.4%

48.1%
36.8%
39.8%
28.9%
49.8%
20.0%
41.9%
41.7%
28.6%
46.8%
25.5%
30.6%
27.0%
23.1%
30.8%
43.0%
26.8%
39.4%

4.4%
7.7%
5.7%
8.8%
3.4%
4.3%
5.2%
4.6%
6.1%
3.2%
5.1%
5.8%
7.1%
9.3%
7.6%
3.6%
5.5%
4.3%

4.4%
7.7%
4.9%
7.0%
6.8%
8.7%
14.2%
5.6%
13.6%
6.5%
10.2%
18.5%
3.2%
14.8%
7.0%
5.5%
2.4%
5.3%

30.4%
26.5%
31.7%
38.6%
29.0%
43.5%
29.7%
29.6%
35.4%
22.6%
30.6%
26.0%
38.1%
36.1%
38.4%
23.6%
46.5%
27.7%

Post-Report Period Judges
4273
4592
4632
4680
4787
4913
5028
5053
5157
5166
5247
5435
5484
5542

USSG §2B1.1
20.3%
24.3%
12.3%
31.7%
15.4%
17.8%
20.3%
20.7%
32.6%
25.5%
27.9%
25.0%
15.0%
10.7%

USSG §2D1.1
38.6%
29.7%
37.0%
32.5%
48.5%
45.4%
42.9%
39.6%
22.0%
45.1%
34.7%
35.0%
38.1%
38.1%

USSG §2L1.2
5.2%
6.8%
8.2%
5.7%
6.2%
6.5%
6.6%
5.4%
7.1%
4.9%
2.7%
3.3%
9.7%
7.7%

OTHER
32.0%
31.8%
26.0%
26.8%
26.2%
22.2%
26.9%
22.5%
34.8%
17.6%
28.4%
33.3%
30.1%
37.5%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4273
4453
4592
4710
4787
4922
5028
5157
5247
5435
5484
5542
5755
5919
5943

Gall Period Judges

USSG §2K2.1
3.9%
7.4%
16.4%
3.3%
3.8%
8.1%
3.3%
11.7%
3.5%
6.9%
6.3%
3.3%
7.1%
6.0%

79

5943

Gall Period Judges
4273
4592
4680
4787
5028
5157
5247
5435
5484
5533
5542
5646
5903
5919
5939
5943
6146
6280

12.5%

22.5%

3.8%

16.3%

45.0%

12.7%

48.1%

17.9%
16.7%
11.1%
23.5%
9.0%
18.5%
16.3%
21.0%
28.6%
19.1%
24.6%
16.7%
16.3%
24.2%
18.9%
23.4%

39.8%
28.9%
49.8%
20.0%
41.9%
41.7%
28.6%
46.8%
25.5%
30.6%
27.0%
23.1%
30.8%
43.0%
26.8%
39.4%

4.4%
7.7%
5.7%
8.8%
3.4%
4.3%
5.2%
4.6%
6.1%
3.2%
5.1%
5.8%
7.1%
9.3%
7.6%
3.6%
5.5%
4.3%

4.4%
7.7%
4.9%
7.0%
6.8%
8.7%
14.2%
5.6%
13.6%
6.5%
10.2%
18.5%
3.2%
14.8%
7.0%
5.5%
2.4%
5.3%

30.4%
26.5%
31.7%
38.6%
29.0%
43.5%
29.7%
29.6%
35.4%
22.6%
30.6%
26.0%
38.1%
36.1%
38.4%
23.6%
46.5%
27.7%

USSG §2B1.1
20.3%
24.3%
12.3%
31.7%
15.4%
17.8%
20.3%
20.7%
32.6%
25.5%
27.9%
25.0%
15.0%
10.7%
26.7%
25.8%
14.5%
18.7%
22.9%
21.5%

USSG §2D1.1
38.6%
29.7%
37.0%
32.5%
48.5%
45.4%
42.9%
39.6%
22.0%
45.1%
34.7%
35.0%
38.1%
38.1%
27.7%
35.2%
39.6%
34.4%
38.9%
42.3%

USSG §2L1.2
5.2%
6.8%
8.2%
5.7%
6.2%
6.5%
6.6%
5.4%
7.1%
4.9%
2.7%
3.3%
9.7%
7.7%
17.5%
9.3%
6.3%
11.0%
2.9%
6.0%

OTHER
32.0%
31.8%
26.0%
26.8%
26.2%
22.2%
26.9%
22.5%
34.8%
17.6%
28.4%
33.3%
30.1%
37.5%
24.3%
24.2%
31.4%
32.1%
28.6%
22.1%

Intra-City Differences
in Federal Sentencing
Practices
21.3%
36.8%
January 2019

LOS ANGELES-BASED FEDERAL JUDGES continued

Post-Report Period Judges
4273
4592
4632
4680
4787
4913
5028
5053
5157
5166
5247
5435
5484
5542
5646
5903
5919
5939
6146
6282

80

|

https://www.ussc.gov/topic/research-reports

USSG §2K2.1
3.9%
7.4%
16.4%
3.3%
3.8%
8.1%
3.3%
11.7%
3.5%
6.9%
6.3%
3.3%
7.1%
6.0%
3.9%
5.5%
8.2%
3.8%
6.9%
8.1%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

MANHATTAN-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF NEW YORK
Appendix B: Judge's Caseload by Guideline
MANHATTAN
USSG §2B1.1
13.5%
17.8%
21.7%

USSG §2D1.1
40.2%
38.0%
43.2%

USSG §2K2.1
9.1%
7.1%
7.4%

USSG §2L1.2
12.2%
14.6%
3.9%

OTHER
25.0%
22.4%
23.8%

4169
4173
4178
4215
4226
4341
4436
4551
4626
4789
4838
4909
5173
5183
5231
5261
5355
5507
5522
5590
5635
5669
5681
5720
5727
5778
5822
6071
6084
6188
6305
6329

18.6%
12.1%
18.9%
13.4%
24.6%
3.2%
19.1%
6.3%
12.5%
10.7%
12.8%
11.2%
8.3%
9.4%
10.0%
18.5%
12.9%
5.9%
21.4%
15.5%
24.2%
12.7%
10.1%
10.5%
11.7%
17.6%
11.1%
7.9%
8.1%
17.2%
12.0%
13.3%

40.2%
51.5%
40.6%
41.1%
34.4%
64.5%
32.6%
46.9%
30.0%
29.3%
31.4%
35.7%
33.3%
38.5%
52.5%
29.6%
38.8%
56.4%
38.8%
60.6%
32.3%
35.4%
59.7%
41.0%
45.6%
25.5%
31.5%
36.6%
22.6%
29.0%
40.0%
47.6%

8.2%
5.3%
9.4%
6.3%
8.2%
1.6%
15.7%
.
.
16.3%
8.0%
11.6%
12.2%
10.0%
8.3%
2.5%
9.3%
10.3%
8.9%
6.1%
.
.
7.3%
10.1%
6.2%
11.4%
5.8%
13.7%
3.7%
19.8%
32.3%
6.5%
12.0%
7.6%

8.2%
14.4%
11.3%
6.3%
13.1%
12.9%
15.7%
4.7%
12.5%
6.7%
15.1%
10.2%
21.7%
18.8%
13.8%
12.0%
10.3%
11.9%
12.2%
9.9%
8.1%
11.4%
10.1%
11.4%
17.5%
5.9%
13.0%
9.9%
21.0%
11.8%
16.0%
20.0%

24.7%
16.7%
19.8%
33.0%
19.7%
17.7%
16.9%
42.2%
28.8%
45.3%
29.1%
30.6%
26.7%
25.0%
21.3%
30.6%
27.6%
16.8%
21.4%
14.1%
28.2%
30.4%
14.0%
25.7%
19.4%
37.3%
40.7%
25.7%
16.1%
35.5%
20.0%
11.4%

Gall Period Judges

USSG §2B1.1
17.6%
14.2%
21.0%
18.6%
18.5%
13.2%
18.5%
16.2%
27.9%
13.2%
19.7%
12.6%
7.5%
5.1%
15.4%
19.5%
19.6%

USSG §2D1.1
42.9%
37.8%
25.2%
27.1%
23.1%
39.7%
45.2%
35.4%
29.1%
43.9%
29.1%
36.2%
41.8%
45.8%
49.6%
28.9%
44.6%

USSG §2K2.1
7.6%
11.8%
8.4%
8.5%
6.2%
4.4%
4.0%
6.1%
8.1%
4.4%
11.1%
5.5%
14.9%
15.3%
7.7%
5.5%
1.8%

USSG §2L1.2
11.2%
15.7%
23.5%
22.0%
20.0%
16.9%
10.5%
20.2%
11.6%
9.6%
9.4%
18.9%
14.9%
20.3%
13.7%
10.2%
16.1%

OTHER
20.6%
20.5%
21.8%
23.7%
32.3%
25.7%
21.8%
22.2%
23.3%
28.9%
30.8%
26.8%
20.9%
13.6%
13.7%
35.9%
17.9%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges

4169
4173
4178
4190
4215
4226
4436
4469
4551
4626
4761
4909
5164
5183
5231
5261
5355

81

4909
5173
5183
5231
5261
5355
5507
5522
5590
5635
5669
5681
5720
5727
5778
5822
6071
6084
6188
6305
6329

Intra-City
January 2019

11.2%
8.3%
9.4%
10.0%
Differences
in
18.5%
12.9%
5.9%
21.4%
15.5%
24.2%
12.7%
10.1%
10.5%
11.7%
17.6%
11.1%
7.9%
8.1%
17.2%
12.0%
13.3%

Federal

35.7%
33.3%
38.5%
52.5%
Sentencing
29.6%
38.8%
56.4%
38.8%
60.6%
32.3%
35.4%
59.7%
41.0%
45.6%
25.5%
31.5%
36.6%
22.6%
29.0%
40.0%
47.6%

Practices

12.2%
10.0%
8.3%
2.5%
9.3%
10.3%
8.9%
6.1%
.
.
7.3%
10.1%
6.2%
11.4%
5.8%
13.7%
3.7%
19.8%
32.3%
6.5%
12.0%
7.6%

10.2%
21.7%
18.8%
13.8%
12.0%
10.3%
11.9%
12.2%
9.9%
8.1%
11.4%
10.1%
11.4%
17.5%
5.9%
13.0%
9.9%
21.0%
11.8%
16.0%
20.0%

30.6%
26.7%
25.0%
21.3%
30.6%
27.6%
16.8%
21.4%
14.1%
28.2%
30.4%
14.0%
25.7%
19.4%
37.3%
40.7%
25.7%
16.1%
35.5%
20.0%
11.4%

USSG §2K2.1
7.6%
11.8%
8.4%
8.5%
6.2%
4.4%
4.0%
6.1%
8.1%
4.4%
11.1%
5.5%
14.9%
15.3%
7.7%
5.5%
1.8%
4.4%
6.7%
5.6%
11.8%
4.7%
5.4%
4.3%
8.9%
6.6%
8.7%
9.1%
6.1%
5.2%
8.1%

USSG §2L1.2
11.2%
15.7%
23.5%
22.0%
20.0%
16.9%
10.5%
20.2%
11.6%
9.6%
9.4%
18.9%
14.9%
20.3%
13.7%
10.2%
16.1%
11.0%
17.8%
11.1%
10.2%
13.5%
13.9%
11.4%
19.5%
19.0%
13.5%
9.8%
20.2%
11.7%
15.4%

OTHER
20.6%
20.5%
21.8%
23.7%
32.3%
25.7%
21.8%
22.2%
23.3%
28.9%
30.8%
26.8%
20.9%
13.6%
13.7%
35.9%
17.9%
20.6%
18.9%
14.8%
12.6%
18.2%
7.2%
30.0%
18.7%
32.2%
26.9%
30.8%
14.0%
27.3%
25.7%

MANHATTAN-BASED FEDERAL JUDGES continued

Gall Period Judges
4169
4173
4178
4190
4215
4226
4436
4469
4551
4626
4761
4909
5164
5183
5231
5261
5355
5507
5522
5590
5635
5669
5681
5720
5727
5778
6071
6084
6188
6305
6329

82

|

USSG §2B1.1
17.6%
14.2%
21.0%
18.6%
18.5%
13.2%
18.5%
16.2%
27.9%
13.2%
19.7%
12.6%
7.5%
5.1%
15.4%
19.5%
19.6%
28.7%
14.4%
24.1%
27.6%
16.9%
15.1%
20.0%
16.3%
11.6%
25.0%
27.3%
15.8%
14.3%
12.5%

https://www.ussc.gov/topic/research-reports

USSG §2D1.1
42.9%
37.8%
25.2%
27.1%
23.1%
39.7%
45.2%
35.4%
29.1%
43.9%
29.1%
36.2%
41.8%
45.8%
49.6%
28.9%
44.6%
35.3%
42.2%
44.4%
37.8%
46.6%
58.4%
34.3%
36.6%
30.6%
26.0%
23.1%
43.9%
41.6%
38.2%

APPENDIX C: Caseload Compositions

MANHATTAN-BASED FEDERAL JUDGES continued

Post-Report Period Judges
4169
4173
4178
4190
4215
4226
4424
4469
4626
4699
4761
4866
4885
4909
4969
5164
5212
5228
5231
5261
5273
5355
5507
5522
5590
5635
5669
5681
5727
5778
5896
5971
6071
6188
6208
6241
6305
6329

USSG §2B1.1
13.9%
36.5%
22.2%
17.4%
39.7%
15.2%
6.7%
25.9%
12.0%
24.7%
23.5%
10.9%
19.1%
18.0%
27.9%
41.4%
16.2%
14.7%
26.4%
19.4%
33.0%
36.6%
6.2%
13.5%
23.0%
13.8%
25.9%
25.6%
27.0%
26.8%
23.2%
24.8%
19.1%
21.7%
23.0%
20.4%
23.9%
16.5%

USSG §2D1.1
51.0%
28.4%
38.1%
38.4%
27.8%
44.4%
57.8%
34.8%
54.4%
48.1%
41.4%
54.5%
52.8%
50.6%
31.1%
28.8%
30.6%
44.1%
38.8%
55.8%
32.1%
36.1%
62.3%
54.6%
40.5%
62.5%
36.2%
38.1%
33.3%
37.8%
41.1%
43.6%
40.9%
42.0%
33.9%
48.7%
39.8%
62.4%

USSG §2K2.1
8.8%
2.7%
6.3%
8.7%
6.6%
11.7%
15.6%
12.7%
3.2%
11.7%
8.6%
7.9%
10.1%
10.1%
1.6%
6.3%
8.7%
7.4%
10.7%
2.4%
8.5%
4.4%
8.0%
7.6%
6.8%
1.3%
3.4%
6.0%
10.6%
9.8%
14.3%
6.8%
7.8%
7.0%
4.4%
8.6%
8.8%
4.6%

USSG §2L1.2
3.1%
2.7%
7.9%
2.3%
7.3%
4.1%
3.3%
5.7%
4.8%
2.6%
4.3%
5.0%
2.2%
4.5%
7.4%
4.5%
2.9%
1.5%
5.0%
1.8%
.
.
6.6%
1.2%
2.7%
4.1%
1.3%
5.2%
3.0%
5.0%
3.7%
.
.
0.8%
5.2%
8.9%
3.8%
4.6%
3.5%
2.8%

OTHER
23.2%
29.7%
25.4%
33.1%
18.5%
24.6%
16.7%
20.9%
25.6%
13.0%
22.2%
21.8%
15.7%
16.9%
32.0%
18.9%
41.6%
32.4%
19.0%
20.6%
26.4%
16.4%
22.2%
21.6%
25.7%
21.1%
29.3%
27.4%
24.1%
22.0%
21.4%
24.1%
27.0%
20.4%
35.0%
17.8%
23.9%
13.8%

83

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

MEMPHIS-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF TENNESSEE
Appendix B: Judge's Caseload by Guideline
MEMPHIS
USSG §2B1.1
20.8%
15.1%
14.9%

USSG §2D1.1
20.6%
21.4%
28.6%

USSG §2K2.1
30.5%
36.3%
33.6%

20.1%
17.1%
25.1%

21.3%
17.1%
23.5%

31.6%
32.6%
27.4%

2.3%
3.9%
2.2%

24.7%
29.3%
21.8%

4021
4684
4871
6221

15.3%
20.1%
17.1%
25.1%

15.9%
29.0%
21.6%
17.6%

41.2%
30.5%
39.8%
35.6%

6.5%
4.6%
1.9%
3.8%

21.2%
22.9%
20.5%
27.0%

Post-Report Period Judges
4021
4684
5103
5130
6221

20.8%
11.4%
13.6%
17.3%
13.9%

25.2%
29.4%
28.5%
24.9%
34.0%

31.2%
37.9%
33.9%
36.5%
26.5%

2.4%
1.8%
1.2%
1.0%
2.1%

20.4%
19.5%
22.7%
20.3%
23.5%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4684
4871
6221

Gall Period Judges

84

|

https://www.ussc.gov/topic/research-reports

USSG §2L1.2
2.8%
4.0%
1.7%

OTHER
25.3%
23.2%
21.2%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

MIAMI-BASED FEDERAL
JUDGES
SOUTHERN DISTRICT OF FLORIDA
Appendix B: Judge's Caseload by Guideline
MIAMI
USSG §2B1.1
15.7%
26.0%
25.1%

USSG §2D1.1
49.0%
35.7%
33.3%

USSG §2K2.1
5.0%
5.7%
7.4%

USSG §2L1.2
5.9%
6.9%
5.9%

OTHER
24.3%
25.8%
28.2%

16.5%
11.4%
17.8%
17.7%
15.5%
19.9%
23.7%
13.5%
13.4%
22.2%
4.2%
8.8%
15.3%

48.9%
54.4%
37.8%
55.8%
49.7%
32.6%
40.5%
53.4%
54.5%
40.0%
60.5%
59.3%
54.5%

2.7%
3.8%
10.8%
3.9%
6.5%
2.8%
4.0%
4.1%
4.3%
2.8%
7.9%
2.2%
6.8%

3.2%
1.3%
6.5%
6.6%
6.5%
10.6%
2.9%
6.1%
5.9%
7.2%
6.8%
8.8%
4.5%

28.7%
29.1%
27.0%
16.0%
21.9%
34.0%
28.9%
23.0%
21.9%
27.8%
20.5%
20.9%
18.8%

4141
4237
4348
4835
4889
5142
5191
5392
5463
5467
6206
6328

24.4%
17.4%
17.8%
21.9%
23.5%
37.5%
33.5%
27.5%
37.1%
21.3%
18.8%
27.6%

38.7%
36.8%
34.1%
44.6%
46.6%
28.2%
28.3%
31.9%
32.4%
24.6%
39.5%
40.1%

6.5%
6.0%
6.5%
7.2%
3.0%
6.0%
4.7%
7.7%
6.3%
4.5%
5.9%
4.6%

4.1%
6.5%
10.3%
5.2%
4.7%
4.3%
7.3%
6.6%
5.9%
9.4%
11.8%
6.9%

26.3%
33.3%
31.3%
21.1%
22.1%
23.9%
26.2%
26.4%
18.4%
40.2%
24.0%
20.7%

Post-Report Period Judges
4141
4237
4348
4635
4835
4889
4894
5191
5463
5467
5887
6185
6206
6308
6328

39.2%
16.7%
28.8%
21.9%
24.8%
20.9%
28.5%
24.2%
27.1%
21.5%
32.8%
13.5%
16.7%
32.4%
27.7%

29.6%
30.6%
33.3%
29.8%
37.2%
37.0%
33.8%
37.9%
34.1%
27.5%
33.3%
41.0%
28.4%
31.0%
35.0%

2.4%
8.1%
6.4%
8.4%
5.4%
6.1%
8.2%
10.6%
5.3%
8.8%
7.0%
8.5%
10.2%
9.5%
10.0%

3.2%
12.4%
2.0%
7.9%
2.8%
4.5%
4.7%
6.1%
5.8%
10.1%
4.3%
12.0%
11.2%
5.2%
3.5%

25.6%
32.3%
29.6%
31.9%
29.7%
31.6%
24.7%
21.2%
27.8%
32.1%
22.7%
25.0%
33.5%
21.9%
23.8%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4141
4237
4348
4835
4889
5142
5191
5392
5463
5467
6206
6324
6328

Gall Period Judges

85

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

MINNEAPOLIS-BASED
FEDERAL JUDGES
DISTRICT OF MINNESOTA
Appendix B: Judge's Caseload by Guideline
MINNEAPOLIS
USSG §2B1.1
11.2%
11.9%
14.1%

USSG §2D1.1
49.0%
44.3%
36.0%

USSG §2K2.1
10.7%
7.8%
12.4%

9.2%
19.1%
4.4%
11.4%
11.7%
9.8%

41.8%
43.6%
55.6%
53.3%
51.4%
48.9%

15.3%
10.0%
14.4%
6.7%
10.8%
7.6%

4.1%
2.7%
5.6%
4.8%
4.5%
7.6%

29.6%
24.5%
20.0%
23.8%
21.6%
26.1%

4081
4452
4964
5116
5495
5530
6322

6.3%
9.8%
19.5%
9.2%
15.8%
10.9%
10.7%

58.5%
38.2%
35.6%
53.3%
36.6%
46.5%
43.0%

7.7%
8.9%
6.8%
9.2%
5.5%
8.9%
9.1%

8.5%
8.1%
12.7%
8.3%
4.9%
12.9%
9.1%

19.0%
35.0%
25.4%
20.0%
37.2%
20.8%
28.1%

Post-Report Period Judges
4081
4452
4964
5495
5530
6322

12.3%
11.3%
11.0%
13.3%
19.6%
16.2%

40.1%
37.3%
21.3%
37.5%
39.9%
37.1%

11.8%
15.3%
16.2%
11.7%
8.2%
11.9%

5.3%
6.2%
11.8%
10.2%
5.1%
5.7%

30.5%
29.9%
39.7%
27.3%
27.2%
29.0%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4081
4964
5116
5495
5530
6322

Gall Period Judges

86

|

https://www.ussc.gov/topic/research-reports

USSG §2L1.2
4.8%
8.8%
7.0%

OTHER
24.3%
27.2%
30.4%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

OKLAHOMA CITY-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF OKLAHOMA
Appendix B: Judge's Caseload by Guideline
OKLAHOMA CITY
USSG §2B1.1
21.3%
17.9%
16.1%

USSG §2D1.1
29.8%
24.3%
27.5%

USSG §2K2.1
11.0%
16.5%
12.2%

29.4%
19.6%
21.7%
16.1%
32.7%
26.2%
6.6%

17.6%
19.6%
27.5%
37.5%
25.5%
21.3%
53.9%

13.2%
19.6%
14.5%
12.5%
5.5%
6.6%
6.6%

4063
4523
4621
4814
5334
5597
5770

16.9%
18.2%
15.6%
23.3%
13.1%
17.5%
21.0%

24.3%
39.4%
20.8%
21.8%
21.3%
13.2%
27.4%

19.1%
9.5%
16.7%
17.3%
18.0%
17.5%
19.4%

13.2%
8.8%
19.8%
12.0%
13.1%
12.3%
4.8%

26.5%
24.1%
27.1%
25.6%
34.4%
39.5%
27.4%

Post-Report Period Judges
4063
4523
4621
4814
5334
5597

17.0%
8.2%
16.4%
17.5%
16.1%
21.4%

27.4%
24.7%
34.0%
23.5%
28.9%
25.2%

11.3%
14.7%
9.8%
11.0%
10.6%
15.8%

16.1%
16.5%
19.1%
23.0%
17.0%
11.5%

28.3%
35.9%
20.7%
25.0%
27.5%
26.1%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4063
4523
4621
4814
5334
5471
5770

Gall Period Judges

USSG §2L1.2
2.8%
12.3%
17.1%

.

.

5.9%
2.9%
5.4%
3.6%
1.6%

.

.

OTHER
35.1%
29.1%
27.2%
33.8%
41.2%
33.3%
28.6%
32.7%
44.3%
32.9%

87

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

PHILADELPHIA-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF PENNSYLVANIA

Appendix B: Judge's Caseload by Guideline
PHILADELPHIA

USSG §2B1.1
16.1%
15.3%
19.4%

USSG §2D1.1
22.8%
22.2%
22.2%

USSG §2K2.1
16.3%
16.5%
11.0%

USSG §2L1.2
8.1%
10.2%
9.2%

OTHER
36.7%
35.8%
38.1%

17.2%
24.0%
5.5%
23.5%
18.9%
17.6%
7.3%
10.0%
9.4%
28.8%

18.8%
14.0%
38.2%
9.8%
20.8%
13.7%
20.0%
40.0%
28.3%
25.0%

20.3%
10.0%
14.5%
23.5%
15.1%
21.6%
14.5%
20.0%
13.2%
9.6%

6.3%
6.0%
12.7%
9.8%
9.4%
2.0%
14.5%
.
.
9.4%
9.6%

37.5%
46.0%
29.1%
33.3%
35.8%
45.1%
43.6%
30.0%
39.6%
26.9%

4022
4324
4570
4651
4788
4854
4886
5051
5160
5433
5583
5956
6275
6311
6356
6367

20.0%
8.6%
11.9%
22.6%
6.9%
9.0%
16.9%
9.8%
18.9%
16.4%
20.9%
9.7%
16.1%
5.7%
19.4%
28.0%

18.2%
25.9%
20.3%
18.9%
15.5%
23.9%
32.3%
16.4%
13.5%
28.8%
23.9%
17.7%
29.0%
26.4%
19.4%
24.0%

20.0%
15.5%
16.9%
13.2%
24.1%
16.4%
7.7%
16.4%
17.6%
16.4%
17.9%
27.4%
22.6%
17.0%
9.7%
8.0%

10.9%
15.5%
15.3%
11.3%
8.6%
4.5%
4.6%
16.4%
8.1%
11.0%
11.9%
16.1%
4.8%
11.3%
9.7%
5.3%

30.9%
34.5%
35.6%
34.0%
44.8%
46.3%
38.5%
41.0%
41.9%
27.4%
25.4%
29.0%
27.4%
39.6%
41.9%
34.7%

Post-Report Period Judges
4022
4324
4570
4623
4641
4651
4788
4854
4886
5042
5051
5287
5433
5583
5956
6275
6356
6367

27.3%
21.2%
30.2%
21.2%
14.6%
14.3%
11.8%
16.0%
5.7%
20.3%
16.1%
18.4%
21.5%
25.3%
17.4%
13.3%
21.7%
28.6%

18.2%
20.0%
19.8%
24.2%
25.6%
19.6%
11.8%
10.7%
13.2%
26.6%
21.4%
22.4%
21.5%
28.0%
27.9%
32.0%
37.3%
14.3%

7.6%
12.9%
7.3%
4.5%
11.0%
14.3%
12.9%
9.3%
11.3%
12.5%
17.9%
14.5%
12.7%
10.7%
9.3%
14.7%
7.2%
10.7%

15.2%
9.4%
10.4%
6.1%
8.5%
19.6%
12.9%
13.3%
11.3%
7.8%
8.9%
2.6%
6.3%
12.0%
7.0%
8.0%
2.4%
7.1%

31.8%
36.5%
32.3%
43.9%
40.2%
32.1%
50.6%
50.7%
58.5%
32.8%
35.7%
42.1%
38.0%
24.0%
38.4%
32.0%
31.3%
39.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4022
4651
4886
5160
5444
5451
5583
6275
6311
6356

Gall Period Judges

88

|

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

PHOENIX-BASED
FEDERAL JUDGES
DISTRICT OF ARIZONA

Appendix B: Judge's Caseload by Guideline
PHOENIX
USSG §2D1.1
21.5%
21.2%
27.0%

USSG §2K2.1
14.1%
7.8%
9.6%

USSG §2L1.2
12.9%
30.1%
23.7%

OTHER
47.9%
37.1%
35.2%

3.4%
3.3%
4.5%
7.9%
3.4%
3.7%
3.2%
2.6%
2.4%
0.9%

23.3%
16.3%
15.8%
18.9%
23.9%
20.2%
24.7%
28.8%
15.9%
25.7%

12.5%
14.4%
14.9%
11.6%
12.9%
14.9%
24.1%
12.2%
9.8%
12.8%

13.6%
12.0%
13.4%
10.0%
14.0%
8.5%
14.6%
17.9%
8.5%
14.7%

47.2%
54.1%
51.5%
51.6%
45.8%
52.7%
33.5%
38.5%
63.4%
45.9%

4148
4156
4497
4757
4933
5249
5347
5804
5917
6224
6253

3.7%
4.1%
4.7%
8.7%
3.7%
2.8%
3.9%
2.5%
3.5%
1.3%
3.1%

24.8%
34.2%
17.0%
18.3%
11.4%
20.0%
17.5%
35.4%
22.7%
19.3%
11.5%

6.4%
5.2%
7.5%
8.3%
8.2%
8.8%
10.1%
7.7%
4.6%
8.7%
9.2%

21.1%
26.4%
29.2%
27.0%
38.8%
32.0%
30.7%
27.0%
38.1%
28.0%
26.5%

44.0%
30.1%
41.5%
37.8%
37.9%
36.4%
37.7%
27.4%
31.2%
42.7%
49.6%

Post-Report Period Judges
4033
4148
4156
4546
4757
5249
5347
5351
5804
5917
6121
6224
6253

1.6%
3.2%
3.7%
3.6%
4.5%
4.9%
5.3%
4.7%
5.6%
9.2%
4.0%
3.4%
5.7%

37.3%
10.6%
22.9%
27.4%
29.0%
21.1%
24.5%
24.5%
34.1%
29.5%
33.8%
13.7%
21.7%

5.8%
7.4%
10.4%
4.9%
12.1%
12.7%
11.5%
9.4%
8.1%
11.6%
8.6%
7.4%
12.6%

27.0%
64.9%
22.1%
24.7%
22.7%
26.1%
20.7%
26.9%
17.6%
17.9%
20.2%
22.9%
27.0%

28.3%
13.8%
41.0%
39.5%
31.6%
35.2%
38.0%
34.4%
34.5%
31.8%
33.3%
52.6%
33.0%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4148
4497
4757
4933
5249
5347
5804
5917
6224
6253

Gall Period Judges

USSG §2B1.1
3.6%
3.8%
4.5%

89

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

PITTSBURGH-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF PENNSYLVANIA
Appendix B: Judge's Caseload by Guideline
PITTSBURGH
USSG §2B1.1
16.9%
15.2%
15.5%

USSG §2D1.1
34.3%
33.8%
40.9%

USSG §2K2.1
19.9%
17.1%
9.9%

USSG §2L1.2
1.3%
1.8%
1.2%

OTHER
27.7%
32.1%
32.5%

16.9%
15.9%
11.8%
12.8%
19.5%
17.2%
23.5%
14.5%
18.8%

35.6%
37.7%
21.6%
50.0%
27.3%
31.0%
39.7%
42.1%
18.8%

16.9%
21.7%
37.3%
14.1%
16.9%
20.7%
22.1%
10.5%
25.0%

.
.
.

30.5%
24.6%
29.4%
21.8%
33.8%
28.7%
11.8%
31.6%
37.5%

4334
4678
4877
5136
5386
5572
5671
5682
5975
6251

14.3%
27.3%
16.9%
16.4%
8.6%
8.1%
26.3%
14.1%
8.2%
13.0%

33.0%
26.1%
27.3%
48.2%
25.9%
34.9%
19.7%
38.5%
46.9%
29.0%

23.1%
19.3%
18.2%
9.1%
21.0%
19.8%
19.7%
10.3%
15.3%
17.4%

.

Post-Report Period Judges
4334
4877
5085
5136
5386
5477
5671
5682
5975
6251

15.4%
18.2%
14.4%
13.0%
18.6%
14.0%
16.3%
20.3%
11.0%
10.0%

39.2%
43.2%
50.0%
47.4%
31.9%
43.0%
32.6%
33.2%
33.0%
55.0%

9.2%
6.1%
9.3%
11.0%
8.8%
10.3%
14.0%
9.9%
14.7%
8.3%

.

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4180
4334
4678
5136
5386
5572
5671
5682
5975

Gall Period Judges

90

|

https://www.ussc.gov/topic/research-reports

.

.
.
.
.

.

.

1.3%
2.6%
2.3%
2.9%
1.3%

2.3%
8.6%
2.3%
2.6%
2.9%

1.4%
1.9%
4.4%
0.9%
1.2%
1.0%
1.8%

.
.
.

.
.
.
.
.
.

.
.

.

29.7%
25.0%
37.7%
26.4%
35.8%
34.9%
34.2%
34.6%
29.6%
37.7%
36.2%
31.1%
26.3%
26.6%
36.3%
31.8%
36.0%
35.6%
39.4%
26.7%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

PORTLAND-BASED
FEDERAL JUDGES
DISTRICT OF OREGON
Appendix B: Judge's Caseload by Guideline
PORTLAND
USSG §2B1.1
13.1%
10.1%
7.6%

USSG §2D1.1
25.6%
26.6%
26.1%

USSG §2K2.1
21.8%
16.9%
21.0%

17.6%
14.7%
11.3%
10.8%
14.0%

29.7%
14.7%
33.8%
26.6%
21.7%

20.3%
19.6%
21.8%
19.4%
27.1%

8.1%
9.8%
4.9%
6.5%
7.0%

24.3%
41.2%
28.2%
36.7%
30.2%

4051
4068
5274
5445
5459
5687
5973

5.4%
14.7%
10.7%
8.7%
2.2%
11.2%
16.0%

18.5%
13.3%
20.7%
34.3%
52.9%
22.9%
18.7%

16.3%
13.3%
19.3%
19.2%
12.5%
16.6%
18.2%

15.2%
20.0%
8.0%
9.3%
17.6%
15.6%
13.4%

44.6%
38.7%
41.3%
28.5%
14.7%
33.7%
33.7%

Post-Report Period Judges
4051
4117
4536
4988
5274
5445
5459
5687
5973
6162

8.7%
8.6%
10.5%
6.4%
6.8%
8.2%
2.3%
11.8%
6.9%
.
.

25.2%
21.4%
24.3%
30.4%
28.8%
31.4%
34.9%
22.4%
27.1%
.
.

11.5%
26.7%
21.5%
20.1%
25.4%
27.3%
19.8%
19.7%
26.1%
2.6%

18.2%
6.4%
7.2%
4.4%
4.2%
0.3%
24.4%
3.9%
.
.
83.1%

36.4%
36.9%
36.5%
38.7%
34.7%
32.8%
18.6%
42.1%
39.9%
14.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4068
5274
5445
5687
5973

Gall Period Judges

USSG §2L1.2
7.0%
13.6%
10.5%

OTHER
32.4%
32.7%
34.8%

91

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

SAINT LOUIS-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF MISSOURI
Appendix B: Judge's Caseload by Guideline
SAINT LOUIS
USSG §2B1.1
12.8%
13.6%
19.0%

USSG §2D1.1
42.2%
29.3%
24.5%

USSG §2K2.1
23.5%
21.8%
27.4%

14.7%
12.1%
14.7%
17.9%
13.7%
12.6%
6.2%
10.2%

44.8%
40.2%
44.1%
34.6%
41.6%
44.4%
45.0%
39.8%

22.8%
21.5%
23.2%
24.1%
25.7%
21.3%
24.4%
25.9%

2.2%
4.7%
3.3%
5.6%
3.1%
2.9%
6.2%
4.6%

15.5%
21.5%
14.7%
17.9%
15.9%
18.8%
18.2%
19.4%

4078
4751
4806
5225
5549
5759
5965
6043

14.4%
11.9%
14.3%
14.5%
13.0%
12.1%
15.8%
11.5%

25.5%
25.2%
34.6%
29.0%
28.2%
34.7%
29.1%
27.9%

23.6%
22.4%
18.0%
20.2%
21.4%
19.1%
23.1%
30.3%

3.0%
1.7%
0.5%
1.5%
0.4%
1.3%
1.6%
6.6%

33.6%
38.8%
32.7%
34.7%
37.0%
32.8%
30.4%
23.8%

Post-Report Period Judges
4078
4149
4751
5225
5549
5759
5824
5926
5965

24.1%
13.5%
31.9%
16.0%
16.3%
19.5%
26.2%
17.4%
15.9%

25.8%
30.4%
16.3%
19.8%
22.3%
27.6%
16.1%
27.3%
24.6%

24.9%
27.6%
13.5%
30.2%
19.3%
25.6%
37.6%
34.5%
27.5%

0.9%
1.4%
0.7%
0.9%
1.2%
1.7%
2.0%
0.3%
1.9%

24.4%
27.0%
37.6%
33.0%
41.0%
25.6%
18.1%
20.5%
30.1%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4078
4751
4806
5225
5549
5759
5965
6043

Gall Period Judges

92

|

https://www.ussc.gov/topic/research-reports

USSG §2L1.2
4.0%
1.8%
1.2%

OTHER
17.6%
33.6%
27.8%

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

SALT LAKE CITY-BASED
FEDERAL JUDGES
DISTRICT OF UTAH
Appendix B: Judge's Caseload by Guideline
SALT LAKE CITY
USSG §2B1.1
6.6%
7.6%
9.0%

USSG §2D1.1
25.9%
28.0%
30.3%

USSG §2K2.1
24.7%
25.7%
25.6%

USSG §2L1.2
20.2%
7.9%
5.4%

OTHER
22.6%
30.8%
29.7%

4.9%
12.9%
9.4%
7.5%
5.9%
4.0%
5.5%
7.3%

29.2%
12.9%
33.6%
19.4%
23.5%
29.6%
24.7%
19.5%

25.1%
22.4%
21.7%
25.4%
23.5%
26.1%
24.0%
27.1%

18.4%
34.1%
16.8%
26.9%
17.6%
21.3%
17.5%
21.4%

22.5%
17.6%
18.4%
20.9%
29.4%
19.0%
28.4%
24.8%

4109
4550
4775
4983
5615
6343

8.0%
7.5%
16.0%
5.5%
6.6%
7.4%

20.5%
28.9%
22.2%
32.4%
24.5%
33.3%

24.0%
27.5%
22.2%
30.5%
24.1%
23.8%

9.1%
7.5%
16.0%
5.9%
8.0%
6.8%

38.4%
28.6%
23.5%
25.8%
36.8%
28.6%

Post-Report Period Judges
4109
4550
4775
4870
4983
5615
6074
6130
6343

9.4%
9.9%
17.6%
7.8%
14.3%
7.1%
9.5%
1.4%
6.7%

26.2%
22.1%
24.7%
33.3%
23.6%
23.0%
30.0%
35.6%
47.5%

20.8%
35.1%
29.4%
20.8%
25.7%
29.4%
21.6%
34.2%
20.2%

4.7%
5.3%
4.7%
6.3%
4.3%
5.6%
8.1%
1.4%
4.0%

38.9%
27.5%
23.5%
31.8%
32.1%
34.9%
30.7%
27.4%
21.5%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4109
4242
4550
4775
4781
4983
6165
6343

Gall Period Judges

93

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

SAN ANTONIO-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF TEXAS
Appendix B: Judge's Caseload by Guideline
SAN ANTONIO
USSG §2D1.1
38.2%
28.1%
31.9%

USSG §2K2.1
19.5%
14.0%
11.0%

USSG §2L1.2
12.0%
32.4%
34.3%

OTHER
23.3%
19.7%
17.2%

5.0%
9.1%
6.4%

41.2%
35.7%
38.1%

21.1%
18.3%
19.5%

10.1%
13.5%
12.3%

22.6%
23.5%
23.7%

5364
5531
5742

4.0%
5.5%
8.0%

30.1%
23.8%
31.6%

15.7%
12.8%
13.8%

27.0%
40.5%
27.4%

23.2%
17.4%
19.2%

Post-Report Period Judges
5364
5405
5531
5742

7.2%
4.7%
3.9%
6.1%

32.9%
26.6%
33.9%
31.0%

11.4%
9.1%
11.3%
11.0%

29.5%
41.1%
35.2%
34.9%

18.9%
18.5%
15.7%
17.0%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
5364
5531
5742

Gall Period Judges

94

|

USSG §2B1.1
6.9%
5.8%
5.6%

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

SAN DIEGO-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF CALIFORNIA
Appendix B: Judge's Caseload by Guideline
SAN DIEGO

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4062
4491
4572
4810
5404
5455
5474
5539
6373

Gall Period Judges
4062
4491
4559
4572
4810
5059
5474
5539
5709
5714
5751
5901
6222
6373

Post-Report Period Judges
4012
4062
4491
4559
4572
4810
5059
5269
5539
5709
5714
5751
5901
6086
6222
6373

USSG §2B1.1
1.3%
2.3%
4.0%

USSG §2D1.1
21.0%
19.3%
36.9%

1.4%
0.4%
2.0%
1.3%
2.2%
3.2%
0.8%
1.4%
1.3%

36.2%
23.1%
22.0%
25.3%
24.2%
17.9%
33.1%
27.1%
3.5%

4.3%
1.8%
1.5%
3.2%
1.2%
1.2%
3.4%
2.1%
4.8%
1.9%
4.0%
1.9%
1.9%
1.7%

19.0%
27.4%
12.8%
21.1%
23.3%
19.8%
26.8%
24.0%
17.2%
20.4%
18.7%
19.5%
21.4%
.
.

6.7%
1.8%
1.8%
7.4%
3.5%
2.8%
2.0%
4.3%
10.3%
9.1%
3.1%
9.1%
2.5%
4.8%
2.6%
.
.

32.4%
53.2%
54.1%
25.6%
41.0%
42.5%
26.4%
42.9%
32.2%
23.0%
45.2%
36.4%
33.1%
16.2%
36.5%
.
.

USSG §2K2.1
0.4%
2.3%
3.3%

.

.
.

.

.

.

0.5%
0.5%
0.9%
1.1%
0.8%
1.0%

3.0%
1.1%
2.9%
3.2%
1.7%
2.9%
6.3%
1.8%
2.4%
0.9%
2.5%
4.2%
2.6%

3.2%
2.4%
4.8%
5.5%
3.3%
3.0%
2.5%
1.7%
4.8%
2.0%
4.5%
5.0%
7.6%
2.6%

USSG §2L1.2
21.7%
40.8%
23.3%

OTHER
55.6%
35.4%
32.4%

.

24.2%
21.4%
35.5%
19.7%
30.8%
24.2%
13.9%
25.7%
19.2%

37.7%
55.1%
40.0%
52.8%
41.8%
54.7%
51.5%
44.8%
76.0%

.

39.2%
34.7%
45.5%
42.6%
35.1%
37.8%
24.4%
37.2%
36.3%
37.0%
36.0%
39.5%
39.8%
76.7%

34.5%
35.0%
37.3%
29.9%
38.6%
38.2%
39.0%
34.9%
39.3%
39.8%
38.8%
34.9%
34.2%
21.7%

20.0%
9.2%
23.7%
23.7%
18.9%
20.8%
39.6%
18.6%
23.0%
21.2%
23.4%
7.3%
21.1%
17.1%
20.6%
82.3%

37.8%
35.8%
17.9%
38.5%
31.1%
30.6%
29.1%
31.7%
32.8%
41.8%
26.4%
42.7%
38.3%
54.3%
37.6%
17.7%

.

.

.

.

95

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

SAN FRANCISCO-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF CALIFORNIA
Appendix B: Judge's Caseload by Guideline
SAN FRANCISCO
USSG §2B1.1
6.5%
13.8%
12.7%

USSG §2D1.1
26.8%
24.9%
27.2%

USSG §2K2.1
17.7%
12.2%
18.2%

USSG §2L1.2
10.7%
22.0%
8.7%

OTHER
38.3%
27.2%
33.2%

3.3%
6.2%
6.7%
7.8%
13.7%
2.8%

28.3%
24.6%
25.0%
32.5%
15.7%
31.0%

23.3%
21.5%
5.0%
14.3%
21.6%
21.1%

10.0%
16.9%
8.3%
9.1%
7.8%
11.3%

35.0%
30.8%
55.0%
36.4%
41.2%
33.8%

4040
4194
4644
4962
5483

14.3%
11.1%
14.5%
10.9%
18.3%

25.5%
21.2%
19.7%
37.0%
18.3%

14.3%
12.1%
12.0%
13.8%
8.7%

12.2%
25.3%
23.1%
21.7%
27.0%

33.7%
30.3%
30.8%
16.7%
27.8%

Post-Report Period Judges
4040
4418
4574
4644
4962
5483
5554
5847

13.8%
13.3%
14.8%
15.3%
12.7%
7.4%
14.9%
10.1%

36.2%
36.1%
27.8%
26.8%
24.5%
27.3%
29.7%
19.3%

12.1%
20.5%
16.5%
19.7%
18.9%
13.2%
25.7%
18.3%

12.1%
3.6%
13.0%
6.4%
6.1%
15.7%
5.4%
9.2%

25.9%
26.5%
27.8%
31.8%
37.7%
36.4%
24.3%
43.1%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4194
4644
4698
4962
5483
5786

Gall Period Judges

96

|

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

SAN JUAN-BASED
FEDERAL JUDGES
DISTRICT OF PUERTO RICO
Appendix B: Judge's Caseload by Guideline
SAN JUAN
USSG §2D1.1
53.2%
27.0%
23.0%

USSG §2K2.1
7.6%
6.4%
24.5%

USSG §2L1.2
11.8%
3.3%
2.3%

OTHER
23.3%
59.7%
47.6%

1.4%
1.4%
2.2%
9.5%
2.3%
2.6%
6.3%

64.2%
65.8%
47.8%
53.2%
45.7%
52.3%
42.2%

4.7%
2.7%
9.7%
9.5%
9.3%
5.2%
10.9%

13.5%
9.6%
17.2%
7.3%
15.5%
7.7%
17.2%

16.2%
20.5%
23.1%
20.5%
27.1%
32.3%
23.4%

4017
4203
4262
4656
5229
5341
6078
6144

2.9%
2.2%
0.8%
1.4%
2.3%
6.0%
5.9%
6.3%

33.6%
33.8%
29.2%
27.1%
24.7%
25.6%
17.0%
32.8%

0.7%
7.4%
5.3%
12.5%
8.8%
10.1%
3.0%
5.3%

1.5%
9.6%
1.1%
5.6%
3.3%
3.6%
3.0%
2.1%

61.3%
47.1%
63.6%
53.5%
60.9%
54.8%
71.2%
53.4%

Post-Report Period Judges
4017
4203
4262
4656
4778
5229
5341
6078
6144

4.2%
2.5%
2.3%
1.3%
2.4%
2.6%
3.3%
2.7%
3.2%

32.9%
19.9%
26.1%
24.6%
10.4%
19.2%
24.9%
20.9%
24.3%

19.2%
24.5%
19.5%
25.5%
43.2%
23.0%
29.1%
21.1%
26.7%

2.8%
2.1%
1.9%
2.2%
3.2%
1.5%
1.8%
3.0%
3.0%

40.8%
51.0%
50.3%
46.4%
40.8%
53.8%
40.8%
52.2%
42.8%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4017
4136
4203
4262
4656
5229
6078

Gall Period Judges

USSG §2B1.1
4.1%
3.5%
2.6%

97

Intra-City Differences in Federal Sentencing Practices
January 2019

Judge's Caseload by Guideline

SEATTLE-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF WASHINGTON
Appendix B: Judge's Caseload by Guideline
SEATTLE
USSG §2B1.1
13.5%
13.5%
13.3%

USSG §2D1.1
41.2%
39.7%
39.0%

USSG §2K2.1
9.0%
10.8%
15.8%

USSG §2L1.2
11.9%
14.3%
6.6%

OTHER
24.3%
21.8%
25.3%

13.3%
13.5%
9.9%
12.1%
19.2%
12.3%
13.7%

39.0%
33.3%
38.9%
39.6%
40.4%
43.9%
51.6%

5.7%
6.3%
17.3%
6.6%
8.8%
10.7%
5.9%

17.1%
20.6%
7.4%
6.0%
12.4%
8.0%
17.0%

24.8%
26.2%
26.5%
35.7%
19.2%
25.1%
11.8%

4150
4185
5011
5515
5891
6150
6193

15.5%
18.0%
10.5%
15.0%
13.1%
12.0%
13.9%

29.6%
36.1%
42.4%
38.9%
40.4%
44.8%
40.6%

7.7%
26.2%
9.3%
7.8%
11.6%
10.9%
10.9%

16.9%
1.6%
12.8%
19.8%
14.1%
12.0%
15.3%

30.3%
18.0%
25.0%
18.6%
20.7%
20.3%
19.3%

Post-Report Period Judges
4150
4185
5011
5515
5891
6150
6193

11.8%
20.4%
15.3%
11.5%
11.2%
9.5%
19.9%

37.7%
23.9%
34.4%
42.7%
50.8%
43.6%
29.5%

12.7%
25.7%
17.7%
13.8%
15.2%
15.3%
15.7%

5.7%
8.8%
7.9%
5.5%
4.1%
6.2%
9.6%

32.0%
21.2%
24.7%
26.5%
18.8%
25.5%
25.3%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4185
5011
5018
5515
5891
6150
6193

Gall Period Judges

98

|

https://www.ussc.gov/topic/research-reports

APPENDIX C: Caseload Compositions

Judge's Caseload by Guideline

TAMPA-BASED
FEDERAL JUDGES
MIDDLE DISTRICT OF FLORIDA
Appendix B: Judge's Caseload by Guideline
TAMPA
USSG §2B1.1
7.6%
10.7%
14.5%

USSG §2D1.1
60.1%
47.9%
49.8%

USSG §2K2.1
7.2%
9.4%
11.5%

USSG §2L1.2
10.2%
12.9%
7.8%

OTHER
14.9%
19.0%
16.4%

5.9%
14.1%
4.1%
11.0%
8.1%
2.9%
7.1%

60.6%
51.3%
68.5%
61.0%
59.3%
63.4%
55.5%

8.0%
7.0%
4.1%
3.8%
3.7%
11.4%
12.6%

10.6%
9.5%
9.5%
10.4%
12.6%
10.9%
8.8%

14.9%
18.1%
14.0%
13.7%
16.3%
11.4%
15.9%

4303
4720
5025
5171
5236
5439
5798
6115
6334

9.5%
17.6%
9.3%
21.2%
8.2%
12.4%
19.2%
6.7%
6.7%

44.1%
42.0%
48.5%
31.8%
46.2%
55.9%
21.9%
56.0%
57.1%

14.7%
10.8%
8.9%
1.5%
11.1%
6.2%
13.7%
8.1%
7.1%

12.3%
14.2%
11.8%
33.3%
13.9%
9.6%
11.0%
10.0%
12.4%

19.4%
15.3%
21.5%
12.1%
20.7%
15.8%
34.2%
19.1%
16.7%

Post-Report Period Judges
4303
4720
5025
5171
5236
5311
5439
5798
6115
6334

11.0%
18.6%
14.2%
17.0%
11.6%
1.5%
19.2%
15.0%
14.6%
11.8%

54.5%
52.3%
60.1%
43.9%
49.2%
14.9%
48.2%
43.0%
46.6%
56.1%

11.0%
9.8%
7.9%
12.5%
17.4%
3.0%
12.0%
10.6%
13.0%
11.8%

8.9%
6.4%
4.7%
4.8%
6.6%
73.1%
2.9%
7.2%
9.7%
4.5%

14.7%
12.9%
13.0%
21.8%
15.1%
7.5%
17.8%
24.2%
16.2%
15.9%

Booker Period Total
Gall Period Total

Post-Report Period Total

Booker Period Judges
4303
4720
5025
5236
5439
6115
6334

Gall Period Judges

99

There were certain cities
where judges in the
city did not have similar
caseload compositions or
where a city’s caseload
composition changed
significantly from one
period to the next.

100

|

https://www.ussc.gov/topic/research-reports

APPENDIX D: Weighting Analyses

APPENDIX D

Weighting Analyses

	

As discussed earlier, the Commission chose

	

In order to test the sensitivity of differences

to measure differences in judges’ sentencing practices

in caseload composition on judges’ average percent

by comparing their average percent differences

differences—both among a city’s judges during a

from the guideline minimums in their cases rather

given time period and also across time (in terms of a

than comparing their average sentence lengths. As

city’s overall caseload)—the Commission conducted

explained, the extent of differences among judges

weighting analyses.58

regarding sentence length is much more sensitive to
variations in caseload composition than the extent of
differences in average percent difference. Appendix C
shows that, for a majority of the 30 cities, most judges’
caseload compositions in each city were generally
similar compared to the other judges in their same city.
Appendix C also shows that most cities had generally
similar overall caseload compositions across the periods.
There were certain cities, however, where most judges
in the city did not have similar caseload compositions
(e.g., Chicago), or where a city’s caseload composition
changed significantly from one period to the next (e.g.,
Houston, from the Booker to Gall Periods).

Differences in Caseload Among a
City’s Judges in a Given Time Period
	

The first weighting test assigned all judges in a

city in a given time period the same weighted caseload
composition by using the city’s overall caseload
composition for each judge. The analysis then applied
each judge’s actual average percent differences for each
of the four primary guidelines (§§2B1.1, 2D1.1, 2K2.1
and 2L1.2) and for all other guidelines (combined into a
fifth category) to the judges’ weighted caseloads. Each
judge’s overall average percent difference based on his

101

Intra-City Differences in Federal Sentencing Practices
January 2019

Figure 11. Weighted Analysis for All 30 Cities Combined
Booker Period

Gall Period

Post-Report Period

Actual Spread

18.2

23.7

27.6

Weighted Spread

18.0

23.1

27.7

Actual Standard Deviation

5.8

7.7

8.3

Weighted Standard Deviation

5.8

7.5

8.4

or her weighted caseload composition was substituted

	

This weighting analysis demonstrates that any

for that judge’s overall average percent difference

differences in caseload composition among judges in

based on his or her actual caseload composition. Total

the 30 cities during all three periods had very little effect

spreads and standard deviations based on the judges’

on the two key measures of differences in sentencing

weighted average percent differences for each city in

practices.

each period were calculated and then compared to the
the actual total spreads and standard deviations for
each of the three time periods.
	

An example of this weighting analysis is seen

in the actual and weighted total spreads and standard
deviations for Chicago during the Post-Report Period.
Chicago was a city with judges whose caseload
compositions during that period were not generally
equivalent, as shown in Appendix C (p. 71).

The

Commission’s weighting analysis showed that those
differences did not substantially contribute to the total
spread and standard deviation in Chicago in that period.
The actual total spread and standard deviation were
49.5 and 10.6, while the weighted total spread and
standard deviation were 47.4 and 10.6.
	

The

Commission

conducted

this

same

weighting analysis for all 30 cities in all three periods.
As shown in Figure 11 above, the actual and the
weighted average total spreads for those 30 cities and
the actual and weighted average standard deviations
for the 27 cities with at least five judges in each period
(the minimum number of judges required for a standard
deviation analysis) were very similar.

102

|

https://www.ussc.gov/topic/research-reports

Differences in Caseload Over Time
Periods in a City
	

The Commission used a similar weighting

analysis to determine whether changes in the total
spread and standard deviation were significantly
affected by changes in caseload composition across the
three periods in some of the 30 cities.
	

This evident weighting test carried forward

each judge’s respective caseload composition from a
prior period to the subsequent period in a particular
city (e.g., each judge’s caseload composition in the
Booker Period was carried forward to the Gall Period).
Each judge’s actual average percent differences for the
four primary guidelines (§§2B1.1, 2D1.1, 2K2.1 and
2L1.2) and for all other guidelines (combined into a
fifth category) from the subsequent period were then
applied to that judge’s caseload composition imported
from the prior period.

The judges’ overall average

percent differences for their weighted caseloads were
calculated, and the city’s weighted total spread and
weighted standard deviation for the subsequent period

APPENDIX D: Weighting Analyses

Figure 12. Booker to Gall Period Comparison for All 30 Cities Combined
Booker Period
(Actual)

Gall Period
(Actual)

Gall Period
(Weighted)

Total Spread

17.9

21.7

22.0

Standard Deviation

6.2

7.6

7.8

were then determined using the weighted average

	

The Commission conducted this weighting

percent differences for each judge.59

analysis for all 30 cities for both period changes

	

An example of this weighting analysis is seen

(Booker to Gall, and Gall to Post-Report). As shown in

in the actual and weighted total spreads and standard

Figure 12 above, when comparing the changes from

deviations for Houston in the Booker Period and the

the Booker Period to the Gall Period, the average actual

Gall Period. As shown in Appendix C (p. 78), among

total spread for all 30 cities in the subsequent period

the 30 cities, Houston had one of the most notable

was very similar to the average weighted total spread

changes in caseload composition from the Booker

for all 30 cities in the subsequent period. Similarly,

Period to the Gall Period. This change was the result

the average actual standard deviation for the 27 cities

of a prosecutorial charging policy that substantially

with at least five judges in each period (for which the

increased the percentage of illegal reentry prosecutions,

standard deviations are reported) was very similar to

which on average have one of the highest within-

the average weighted standard deviation.

range rates of all major guideline types.60 Despite the

	

substantial changes in caseload composition from one

next period change—from the Gall Period to the Post-

period to the next,61 the actual and weighted total

Report Period—the average actual and weighted total

spreads and standard deviations in the Gall Period both

spreads and standard deviations also were very similar.

showed noticeable increases from the Booker Period.

Therefore, any differences in caseload composition

The total spread and standard deviation in the Booker

from one period to the next in the 30 cities had very

Period were 23.3 and 7.9. The actual total spread and

little effect on the two measures of differences in

standard deviation in the Gall Period were 36.5 and

sentencing practices.

As shown in Figure 13 below, regarding the

13.2, compared to weighted total spreads of 28.4 and
10.5.

Figure 13. Gall to Post-Report Period Comparison for All 30 Cities Combined
Gall Period
(Actual)

Post-Report Period
(Actual)

Post-Report Period
(Weighted)

Total Spread

20.9

25.3

25.8

Standard Deviation

7.1

8.5

8.6

103

Two important
concepts in statistics—
outliers and standard
deviation—are relevant
to the Commission’s
30-city study.

104

|

https://www.ussc.gov/topic/research-reports

APPENDIX E: Statistical Outliers and Standard Deviation

APPENDIX E

Statistical Outliers and
Standard Deviation Analysis

	

This

appendix

discusses

two

important

	

Using that test, the Commission identified 39

concepts in statistics—outliers and standard deviation—

statistical outliers out of a total 909 different judicial

that are relevant to the Commission’s 30-city study.

sentencing practices in the 30 cities over the three
periods.65 During the Booker Period, there were 11

Statistical Outliers

outliers; during the Gall Period, there were ten outliers;
and during the Post-Report period, there were 18

	

When data is analyzed, a researcher should

outliers.66 Tampa was the only city to have more than

identify whether any point in the dataset is a statistical

one outlier judge during multiple periods (Gall and Post-

outlier—an “[o]bservation [in the dataset] that is far

Report).

removed from the bulk of the data.”

62

Outliers “may

	

In order to show the effect of those outlier

indicate faulty measurements and they may exert undue

judges on both the total spread and standard deviation

influence on summary statistics, such as the mean ….”

measures, the Commission has re-calculated those

63

Although the Commission has no concerns that any

two measures excluding the outlier judges.

datapoints in the 30-city dataset are products of faulty

results—both with and without the outlier judges in the

measurements of any judge’s sentencing practices, the

analysis—are set forth below, using a modification of the

Commission nonetheless has identified judges who

graphical presentation discussed earlier in this report.

qualify as outliers using the most common test for

The total spread and standard deviation after excluding

such identification—the 1.5 x Interquartile Range (IQR)

the outliers are reported in parentheses following the

test.64

total spread and standard deviation before excluding

The

105

Intra-City Differences in Federal Sentencing Practices
January 2019

Figure 14. Chicago-Based Outlier Federal Judges
CHI
the outliers. Outlier judges appear above or below
a
Booker BOOKER PERIOD

red line on the graph. For instance, for Chicago the
two outlier judges—one in the Booker Period and one
in the Post-Report Period—are identified either above
or below red lines superimposed on the graphs to the
right.
	

Judges: 18 (17)
Cases: 1,341 (1,270)
Spread: 31.6 (22.0)
Standard Deviation: 7.9 (6.3)

Graphical representations of the cities with at

least one outlier judge in at least one period are set
forth below. Note that, when removal of an outlier
judge resulted in a city having less than five judges

-40.0

-30.0

-20.0

-10.0

in one or more of the three periods, the standard
CHI
deviation for that city is not reported. As discussed
Gall

City
Avg
0.0
-10.1%

10.0

20.0

30.0

40.0

GALL PERIOD

below, reporting the standard deviation for cities with
less than five judges in any period is not a meaningful
measure of dispersion. Only the total spread will be
reported for those cities with less than five judges in a
period after excluding outliers.
	

Judges: 22
Cases: 2,033
Spread: 42.7
Standard Deviation: 11.0

The Commission has identified the statistical

outliers in order to allow the reader to compare the
graphical presentations of sentencing differences
among judges for each city that has outlier judges—
with and without the outlier judges. Nevertheless, it

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-16.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-28.1%

10.0

20.0

30.0

40.0

must be remembered that outlier judges imposed real
CHI

Post

sentences on real defendants and their sentencing

POST-REPORT PERIOD

practices contributed to the extent of differences in
sentencing practices among judges in their cities. The
Commission’s exclusion of outlier judges from the
alternative analyses in this appendix is not intended to
suggest otherwise.

Judges: 24 (23)
Cases: 2,951 (2,779)
Spread: 49.5 (32.9)
Standard Deviation: 10.6 (8.2)

-40.0

106

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

Standard Deviation
	

In addition to reporting the total spread—i.e.,

	

The standard deviation for each time period

the absolute percentage difference between the

in each city with at least five judges69 was calculated

two judges furthest away the city average, positively

as follows. First, the variance—the statistical term, not

and negatively—the Commission also has reported

the term referring to sentences outside of the guideline

the standard deviation for the 27 cities with at least

range after Booker—was calculated by summing the

five judges in all three periods. Standard deviation

squared deviation from the mean for each judge (i.e.,

is the primary measure of dispersion or variability of

the squared difference between each judge’s average

datapoints compared to the mean67—in the case of the

percent difference and the city’s average percent

30 cities dataset, the dispersion of the judges’ average

difference). The total sum was divided by the number

percent differences from the guideline minimums in

of judges in the city minus one, which yielded the

relation to the city’s average percent difference. The

statistical variance. The standard deviation was then

standard deviation is thus a valid measure of the extent

calculated by taking the square root of that variance.70

of sentencing differences among all the judges in a city
in a given time period. The standard deviation “can
be appropriately understood as the typical distance of
a randomly selected [datapoint] from the mean of the
distribution.”68

107

Intra-City Differences in Federal Sentencing Practices
January 2019

ATL
Booker

BOOKER PERIOD
Judges: 12
Cases: 995
Spread: 19.6
Standard Deviation: 5.5

OUTLIERS

ATLANTA-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF GEORGIA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-1.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-8.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.5%

10.0

20.0

30.0

40.0

outlier judges.
ATL
Gall

In Atlanta, there were two statistical outliers over the
three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding

GALL PERIOD
Judges: 14 (12)
Cases: 1,504 (1,382)
Spread: 31.8 (12.2)
Standard Deviation: 7.2 (3.7)

the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

ATL
Post

POST-REPORT PERIOD

5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%

Judges: 15
Cases: 2,143
Spread: 26.9
Standard Deviation: 9.6

-30 to -34.99%
-35 to -39.99%

-40.0

108

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

CHI
Booker

BOOKER PERIOD
Judges: 18 (17)
Cases: 1,341 (1,270)
Spread: 31.6 (22.0)
Standard Deviation: 7.9 (6.3)

OUTLIERS

CHICAGO-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF ILLINOIS
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-10.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-28.1%

10.0

20.0

30.0

40.0

outlier judges.
CHI

Gallthe
In Chicago, there were two statistical outliers over

GALL PERIOD

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 22
Cases: 2,033
Spread: 42.7
Standard Deviation: 11.0

total spread and standard deviation before excluding
the outliers.

Legend

-40.0

35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%

CHI
Post

POST-REPORT PERIOD

0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%

Judges: 24 (23)
Cases: 2,951 (2,779)
Spread: 49.5 (32.9)
Standard Deviation: 10.6 (8.2)

-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

109

Intra-City Differences in Federal Sentencing Practices
January 2019

CLEV
Booker

BOOKER PERIOD

OUTLIERS

Judges: 9 (8)
Cases: 924 (840)
Spread: 22.7 (12.6)
Standard Deviation: 6.6 (4.7)

CLEVELAND-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF OHIO
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-4.8%

10.0

20.0

30.0

40.0

City
0.0Avg
-6.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-12.2%

10.0

20.0

30.0

40.0

outlier judges.
CLEV
Gall

In Cleveland, there was one statistical outlier over the

GALL PERIOD

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 8
Cases: 744
Spread: 16.6
Standard Deviation: 5.2

total spread and standard deviation before excluding
the outliers.

-40.0

-30.0

-20.0

-10.0

Legend
35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

CLEV
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 6
Cases: 1,240
Spread: 19.9
Standard Deviation: 7.5

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

110

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

COL
Booker

OUTLIERS

BOOKER PERIOD

COLUMBUS-BASED
FEDERAL JUDGES

Judges: 5 (4)
Cases: 282 (222)
Spread: 15.2 (6.0)
Standard Deviation: 6.0

SOUTHERN DISTRICT OF OHIO
In order to show the effect of statistical outliers on
the total spread, the Commission re-calculated that
measure excluding outlier judges.
Because Columbus has less than five judges in atCOL
least

Gall
one period after excluding outliers, the Commission

has not reported the standard deviation after excluding
outliers for this city.
In Columbus, there was one statistical outlier over the
three periods. Outlier judges appear above or below a

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
10.4%

10.0

20.0

30.0

40.0

City
Avg
0.0
-19.4%

10.0

20.0

30.0

40.0

City
Avg
0.0
-24.6%

10.0

20.0

30.0

40.0

GALL PERIOD
Judges: 6
Cases: 463
Spread: 12.4
Standard Deviation: 5.0

red line on the graphs.
The total spread after excluding the outliers is reported
in parentheses following the total spread before
excluding the outliers.
-40.0

-30.0

-20.0

-10.0

Legend
35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

COL
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%

Judges: 7
Cases: 927
Spread: 8.7
Standard Deviation: 3.1

-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

111

Intra-City Differences in Federal Sentencing Practices
January 2019

DAL
Booker

BOOKER PERIOD
Judges: 8 (7)
Cases: 660 (566)
Spread: 22.0 (10.1)
Standard Deviation: 6.6 (4.0)

OUTLIERS

DALLAS-BASED
FEDERAL JUDGES
NORTHERN DISTRICT OF TEXAS
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
1.7%

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

outlier judges.
DAL
Gall

In Dallas, there were three statistical outliers over the

GALL PERIOD

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 7 (6)
Cases: 904 (773)
Spread: 33.0 (15.2)
Standard Deviation: 10.9 (6.0)

total spread and standard deviation before excluding
the outliers.

-40.0

Legend

-30.0

35 to 39.99%

-20.0

-10.0

City
Avg
0.0
-3.0%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%

DAL
Post

POST-REPORT PERIOD

0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%

Judges: 7 (6)
Cases: 2,507 (2,146)
Spread: 40.3 (19.0)
Standard Deviation: 12.5 (7.1)

-35 to -39.99%

-40.0

112

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

City
0.0Avg
-2.1%

APPENDIX E: Statistical Outliers and Standard Deviation

DET
Booker

BOOKER PERIOD
Judges: 11
Cases: 680
Spread: 34.4
Standard Deviation: 9.5

OUTLIERS

DETROIT-BASED
FEDERAL JUDGES
EASTERN DISTRICT OF MICHIGAN
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-9.6%

10.0

20.0

30.0

40.0

City
Avg
0.0
-21.5%

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

outlier judges.
DET
Gall

In Detroit, there were three statistical outliers over the

GALL PERIOD

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 12
Cases: 1,011
Spread: 34.1
Standard Deviation: 10.6

total spread and standard deviation before excluding
the outliers.

-40.0

Legend

-30.0

35 to 39.99%

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%

DET
Post

POST-REPORT PERIOD

0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%

Judges: 18 (15)
Cases: 2,210 (1,727)
Spread: 47.7 (21.2)
Standard Deviation: 11.1 (6.2)

-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
-23.3%

113

Intra-City Differences in Federal Sentencing Practices
January 2019

LA
Booker

BOOKER PERIOD
Judges: 15 (13)
Cases: 1,273 (1,090)
Spread: 29.1 (17.6)
Standard Deviation: 7.6 (5.1)

OUTLIERS

LOS ANGELES-BASED
FEDERAL JUDGES
CENTRAL DISTRICT OF CALIFORNIA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-14.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-18.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-26.7%

10.0

20.0

30.0

40.0

outlier judges.
LA
Gall

In Los Angeles, there were two statistical outliers over
the three periods.

GALL PERIOD

Outlier judges appear above or

below a red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 18
Cases: 2,407
Spread: 29.3
Standard Deviation: 8.0

total spread and standard deviation before excluding
the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

LA
Post

POST-REPORT PERIOD

5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%

Judges: 20
Cases: 3,111
Spread: 42.9
Standard Deviation: 11.8

-30 to -34.99%
-35 to -39.99%

-40.0

114

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

MAN
Booker

BOOKER PERIOD
Judges: 32 (31)
Cases: 2,899 (2,845)
Spread: 44.8 (36.1)
Standard Deviation: 9.2 (7.9)

OUTLIERS

MANHATTAN-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF NEW YORK
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-14.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-25.5%

10.0

20.0

30.0

40.0

City
Avg
0.0
-36.3%

10.0

20.0

30.0

40.0

outlier judges.
MAN
Gall

In Manhattan, there was one statistical outlier over the

GALL PERIOD

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 31
Cases: 3,435
Spread: 34.5
Standard Deviation: 8.7

total spread and standard deviation before excluding
the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

MAN
Post

POST-REPORT PERIOD

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%

Judges: 38
Cases: 4,863
Spread: 59.1
Standard Deviation: 12.9

-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

115

Intra-City Differences in Federal Sentencing Practices
January 2019

MEM
Booker

OUTLIERS

BOOKER PERIOD
Judges: 3
Cases: 534
Spread: 14.0

MEMPHIS-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF TENNESSEE
In order to show the effect of statistical outliers on
the total spread, the Commission re-calculated that
measure excluding outlier judges.
Because Memphis has less than five judges in at
least
MEM

Gall
one period after excluding outliers, the Commission

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-5.8%

10.0

20.0

30.0

40.0

-10.0

City
Avg
0.0
-6.8%

10.0

20.0

30.0

40.0

City
0.0Avg
-13.4%

10.0

20.0

30.0

40.0

GALL PERIOD

has not reported the standard deviation after excluding
outliers for this city.
In Memphis, there was one statistical outlier over the

Judges: 4
Cases: 980
Spread: 17.4

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread after excluding the outliers is reported
in parentheses following the total spread before
excluding the outliers.
-40.0

Legend

-30.0

35 to 39.99%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

-20.0

MEM
Post

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%

POST-REPORT PERIOD

-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

Judges: 5 (4)
Cases: 1,400 (1,162)
Spread: 18.6 (6.8)

-40.0

116

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

MIA
Booker

BOOKER PERIOD
Judges: 13 (12)
Cases: 2,074 (1,983)
Spread: 16.0 (7.8)
Standard Deviation: 4.1 (2.6)

OUTLIERS

MIAMI-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF FLORIDA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-4.3%

10.0

20.0

30.0

40.0

City
0.0Avg
-6.9%

10.0

20.0

30.0

40.0

City
0.0Avg
-10.0%

10.0

20.0

30.0

40.0

outlier judges.
MIA
Gall

In Miami, there was one statistical outlier over the three

GALL PERIOD

periods. Outlier judges appear above or below a red
line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 12
Cases: 2,859
Spread: 17.1
Standard Deviation: 5.2

total spread and standard deviation before excluding
the outliers.

-40.0

Legend
35 to 39.99%

20 to 24.99%
15 to 19.99%

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%

-30.0

MIA
Post

10 to 14.99%
5 to 9.99%
0 to 4.99%
-0.01 to -4.99%

POST-REPORT PERIOD

-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

Judges: 15
Cases: 4,534
Spread: 29.9
Standard Deviation: 9.0

-40.0

-30.0

-20.0

-10.0

117

Intra-City Differences in Federal Sentencing Practices
January 2019

MINN
Booker

OUTLIERS

MINNEAPOLIS-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 6
Cases: 606
Spread: 13.7
Standard Deviation: 5.1

DISTRICT OF MINNESOTA
In order to show the effect of statistical outliers on
the total spread, the Commission re-calculated that
measure excluding outlier judges.
In Minneapolis, there was one statistical outlier over
MINNthe

Gall
three periods. Outlier judges appear above or below
a

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
-7.9%

10.0

20.0

30.0

40.0

City
Avg
0.0
-15.6%

10.0

20.0

30.0

40.0

City
Avg
0.0
-18.7%

10.0

20.0

30.0

40.0

GALL PERIOD

red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding

Judges: 7
Cases: 908
Spread: 20.5
Standard Deviation: 6.9

the outliers.

-40.0

Legend

-30.0

35 to 39.99%

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

MINN
Post

POST-REPORT PERIOD

5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%

Judges: 6 (5)
Cases: 996 (786)
Spread: 15.0 (6.6)
Standard Deviation: 5.3 (3.2)

-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

118

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

OKC
Booker

OUTLIERS

OKLAHOMA CITY-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 7 (6)
Cases: 436 (360)
Spread: 32.1 (14.8)
Standard Deviation: 10.2 (5.8)

WESTERN DISTRICT OF OKLAHOMA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-0.7%

10.0

20.0

30.0

40.0

City
0.0Avg
-7.6%

10.0

20.0

30.0

40.0

City
0.0Avg
-19.3%

10.0

20.0

30.0

40.0

outlier judges.
OKC
Gall

In Oklahoma City, there was one statistical outlier over
the three periods.

GALL PERIOD

Outlier judges appear above or

below a red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 7
Cases: 800
Spread: 14.8
Standard Deviation: 5.2

total spread and standard deviation before excluding
the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

OKC
Post

POST-REPORT PERIOD

5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%

Judges: 6
Cases: 1,369
Spread: 6.9
Standard Deviation: 2.7

-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

119

Intra-City Differences in Federal Sentencing Practices
January 2019

PHI
Booker

BOOKER PERIOD

OUTLIERS

PHILADELPHIA-BASED
FEDERAL JUDGES

Judges: 10 (9)
Cases: 534 (479)
Spread: 33.8 (15.8)
Standard Deviation: 9.6 (5.2)

EASTERN DISTRICT OF PENNSYLVANIA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-11.9%

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

outlier judges.
PHI
Gall

In Philadelphia, there were three statistical outliers

GALL PERIOD

over the three periods. Outlier judges appear above or
below a red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 16 (15)
Cases: 1,004 (937)
Spread: 53.7 (28.3)
Standard Deviation: 13.5 (9.6)

total spread and standard deviation before excluding
the outliers.

Legend

-40.0

35 to 39.99%

-30.0

-20.0

-10.0

City
Avg
0.0
-14.6%

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%

PHI
Post

POST-REPORT PERIOD

5 to 9.99%
0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%

Judges: 18 (17)
Cases: 1,314 (1,239)
Spread: 63.8 (33.8)
Standard Deviation: 13.6 (9.6)

-30 to -34.99%
-35 to -39.99%

-40.0

120

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

City0.0
Avg
18.4%

APPENDIX E: Statistical Outliers and Standard Deviation

PHX
Booker

BOOKER PERIOD

OUTLIERS

Judges: 10 (9)
Cases: 1,843 (1,579)
Spread: 8.7 (5.8)
Standard Deviation: 2.8 (1.9)

PHOENIX-BASED
FEDERAL JUDGES
DISTRICT OF ARIZONA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-3.7%

10.0

20.0

30.0

40.0

City
0.0Avg
-7.4%

10.0

20.0

30.0

40.0

City
0.0Avg
-10.1%

10.0

20.0

30.0

40.0

outlier judges.
PHX
Gall

In Phoenix, there were four statistical outliers over the
three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding
the outliers.

Legend

GALL PERIOD
Judges: 11 (9)
Cases: 2,330 (1,974)
Spread: 25.7 (16.7)
Standard Deviation: 7.1 (4.5)

-40.0

35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%

PHX
Post

POST-REPORT PERIOD

-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

Judges: 13 (12)
Cases: 3,742 (3,600)
Spread: 27.5 (21.9)
Standard Deviation: 7.9 (6.3)

-40.0

-30.0

-20.0

-10.0

121

Intra-City Differences in Federal Sentencing Practices
January 2019

PITT
Booker

BOOKER PERIOD
Judges: 9
Cases: 629
Spread: 19.7
Standard Deviation: 5.9

OUTLIERS

PITTSBURGH-BASED
FEDERAL JUDGES
WESTERN DISTRICT OF PENNSYLVANIA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
-1.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-13.0%

10.0

20.0

30.0

40.0

City
Avg
0.0
-16.9%

10.0

20.0

30.0

40.0

outlier judges.
PITT
Gall

In Pittsburgh, there was one statistical outlier over the
three periods. Outlier judges appear above or below a

GALL PERIOD

red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding

Judges: 10 (9)
Cases: 854 (768)
Spread: 32.5 (12.2)
Standard Deviation: 9.4 (4.0)

the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%

PITT
Post

10 to 14.99%
5 to 9.99%
0 to 4.99%

POST-REPORT PERIOD

-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%

Judges: 10
Cases: 1,287
Spread: 19.8
Standard Deviation: 6.9

-35 to -39.99%

-40.0

122

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

PORT
Booker

BOOKER PERIOD
Judges: 5
Cases: 586
Spread: 8.3
Standard Deviation: 3.3

OUTLIERS

PORTLAND-BASED
FEDERAL JUDGES
DISTRICT OF OREGON
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
-11.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-21.6%

10.0

20.0

30.0

40.0

City0.0
Avg
-29.2%

10.0

20.0

30.0

40.0

outlier judges.
PORT
Gall

In Portland, there was one statistical outlier over the
three periods. Outlier judges appear above or below a

GALL PERIOD

red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding

Judges: 7
Cases: 1,017
Spread: 14.7
Standard Deviation: 5.8

the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%

PORT
Post

POST-REPORT PERIOD

-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%

Judges: 10 (9)
Cases: 1,711 (1,625)
Spread: 24.6 (13.0)
Standard Deviation: 7.1 (4.8)

-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

123

Intra-City Differences in Federal Sentencing Practices
January 2019

SLC
Booker

OUTLIERS

SALT LAKE CITY-BASED
FEDERAL JUDGES

BOOKER PERIOD
Judges: 8
Cases: 1,504
Spread: 7.8
Standard Deviation: 3.0

DISTRICT OF UTAH
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-6.0%

10.0

20.0

30.0

40.0

City
0.0Avg
-15.1%

10.0

20.0

30.0

40.0

City
Avg
0.0
-29.0%

10.0

20.0

30.0

40.0

outlier judges.
SLC
Gall

In Salt Lake City, there were three statistical outliers

GALL PERIOD

over the three periods. Outlier judges appear above or
below a red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 6
Cases: 1,428
Spread: 14.7
Standard Deviation: 5.6

total spread and standard deviation before excluding
the outliers.

-40.0

Legend
35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%

SLC
Post

POST-REPORT PERIOD

-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

Judges: 9 (6)
Cases: 1,733 (1,044)
Spread: 20.1 (1.9)
Standard Deviation: 5.8 (0.7)

-40.0

124

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

SD
Booker

OUTLIERS

BOOKER PERIOD
Judges: 9
Cases: 2,295
Spread: 13.9
Standard Deviation: 4.5

SAN DIEGO-BASED
FEDERAL JUDGES
SOUTHERN DISTRICT OF CALIFORNIA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
-6.5%

10.0

20.0

30.0

40.0

City
Avg
0.0
-9.5%

10.0

20.0

30.0

40.0

10.0

20.0

30.0

40.0

outlier judges.
SD
Gall

In San Diego, there was one statistical outlier over the

GALL PERIOD

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the

Judges: 14
Cases: 5,288
Spread: 21.3
Standard Deviation: 6.4

total spread and standard deviation before excluding
the outliers.

Legend

-40.0

35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%

SD
Post

POST-REPORT PERIOD

0 to 4.99%
-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%

Judges: 16 (15)
Cases: 4,377 (4,315)
Spread: 54.4 (38.3)
Standard Deviation: 13.0 (9.7)

-35 to -39.99%

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
-23.8%

125

Intra-City Differences in Federal Sentencing Practices
January 2019

SAN
JUAN
Booker

BOOKER PERIOD
Judges: 7
Cases: 923
Spread: 9.6
Standard Deviation: 3.6

OUTLIERS

SAN JUAN-BASED
FEDERAL JUDGES
DISTRICT OF PUERTO RICO
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
Avg
0.0
0.7%

10.0

20.0

30.0

40.0

City
Avg
0.0
-4.8%

10.0

20.0

30.0

40.0

City
Avg
0.0
-1.2%

10.0

20.0

30.0

40.0

outlier judges.
In San Juan, there was one statistical outlier

SAN
JUAN
overGallthe

three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding
the outliers.

Legend

GALL PERIOD
Judges: 8
Cases: 1,524
Spread: 20.9
Standard Deviation: 7.2

-40.0

35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%

SAN
JUAN
Post

POST-REPORT PERIOD

-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%

Judges: 9 (8)
Cases: 4,013 (3,486)
Spread: 22.2 (19.9)
Standard Deviation: 7.4 (6.3)

-35 to -39.99%

-40.0

126

|

https://www.ussc.gov/topic/research-reports

-30.0

-20.0

-10.0

APPENDIX E: Statistical Outliers and Standard Deviation

TAMPA
Booker

BOOKER PERIOD
Judges: 7
Cases: 1,283
Spread: 11.9
Standard Deviation: 3.7

OUTLIERS

TAMPA-BASED
FEDERAL JUDGES
MIDDLE DISTRICT OF FLORIDA
In order to show the effect of statistical outliers on both
the total spread and standard deviation measures, the
Commission re-calculated those measures excluding

-40.0

-30.0

-20.0

-10.0

City
0.0Avg
-0.2%

10.0

20.0

30.0

40.0

City
Avg
0.0
-4.7%

10.0

20.0

30.0

40.0

City
Avg
0.0
-9.6%

10.0

20.0

30.0

40.0

outlier judges.
TAMPA
Gall

In Tampa, there were six statistical outliers over the
three periods. Outlier judges appear above or below a
red line on the graphs.
The total spread and standard deviation after excluding
the outliers are reported in parentheses following the
total spread and standard deviation before excluding

GALL PERIOD
Judges: 9 (6)
Cases: 1,567 (1,218)
Spread: 29.7 (6.7)
Standard Deviation: 9.6 (2.3)

the outliers.

Legend

-40.0

35 to 39.99%

-30.0

-20.0

-10.0

Legend

30 to 34.99%
25 to 29.99%
20 to 24.99%
15 to 19.99%
10 to 14.99%
5 to 9.99%
0 to 4.99%

TAMPA
Post

POST-REPORT PERIOD

-0.01 to -4.99%
-5 to -9.99%
-10 to -14.99%
-15 to -19.99%
-20 to -24.99%
-25 to -29.99%
-30 to -34.99%
-35 to -39.99%

Judges: 10 (7)
Cases: 2,429 (1,866)
Spread: 22.7 (3.1)
Standard Deviation: 5.9 (1.1)

-40.0

-30.0

-20.0

-10.0

127

Intra-City Differences in Federal Sentencing Practices
January 2019

Endnotes
1	

543 U.S. 220 (2005).

2	

Report on the Continuing Impact of United States v. Booker on Federal Sentencing (“2012 Booker Report”).

3	
See U.S. Sentencing Comm’n, Demographic Differences in Sentencing: An Update to the 2012 Booker Report
(November 2017).
4	
Id. at 2 (analysis of data from fiscal years 2012 to 2016 found that Black males received a 19.1% higher
sentence on average than White males).
5	
See USSG §1B1.1, commen. (backg’d) (“Subsections (a), (b), and (c) [of §1B1.1] are structured to reflect the
three-step process used in determining the particular sentence to be imposed. If, after step (c), the court imposes a
sentence that is outside the guidelines framework, such a sentence is considered a ‘variance.’); see also United States
v. Rangel, 697 F.3d 795, 801 (9th Cir. 2012), cert. denied, 133 S. Ct. 1294 (2013) (“A ‘departure’ is typically a change
from the final sentencing range computed by examining the provisions of the Guidelines themselves. . . . A ‘variance,’
by contrast, occurs when a judge imposes a sentence above or below the otherwise properly calculated final sentencing
range based on application of the other statutory factors in 18 U.S.C. § 3553(a).”).
2012 Booker Report, at 98.

6	

7	
See Booker, 543 U.S. at 248, 259 (“[W]e conclude that the constitutional jury trial requirement is not compatible
with the Act as written and that some severance and excision are necessary. . . . The remainder of the Act function[s]
independently.”) (citation and internal quotation marks omitted).
8	
Mistretta v. United States, 488 U.S. 361, 366 (1989); see also William W. Wilkins, Phyllis J. Newton, and John R.
Steer, The Sentencing Reform Act of 1984: A Bold Approach to the Unwarranted Sentencing Disparity Problem, 2 Crim. L. Forum
355 (1991). Senator Kennedy, the leading sponsor of the SRA, referred to Congress’s concern over sentencing disparities
as being “the major impetus for sentencing reform.” Edward M. Kennedy, Toward a New System of Criminal Sentencing: Law
with Order, 16 Amer. Crim. L. Rev. 353, 357 (1979).
Report of the Committee of the Judiciary, United States Senate, S. Rep. No. 98-225, 98th Cong. (1st Sess.), at 41-42
9	
(Sept. 14, 1983) (discussing sentencing disparities studies considered by Congress in enacting the SRA); see also Marvin E.
Frankel, Criminal Sentences: Law Without Order 6-7 (1973) (criticizing the “wild array of sentencing judgments [in federal
court] without any semblance of the consistency demanded by our ideal of equal justice” and observing that the type and
length of federal sentences for similar situated offenders “depend[ed] on the judge” drawn by the defendants).
10	

18 U.S.C. § 3553(a)(1)(6); 28 U.S.C. §§ 991(b)(1)(B), 994(f).

11	

28 U.S.C. §§ 991(b)(1)(B), 994(f).

12	
Booker, 543 U.S. at 263; see also id. at 264 (“The system remaining . . ., while lacking the mandatory features that
Congress enacted, retains other features that help to further these objectives [including avoiding unwarranted sentencing
disparities].”).
13	

See 2012 Booker Report, at 98-104.

14	

518 U.S. 996 (1996).

15	

542 U.S. 296 (June 24, 2004).

16	

543 U.S. 200 (June 10, 2005).

17	
2007).

Gall v. United States, 552 U.S. 28 (December 10, 2007); Kimbrough v. United States, 552 U.S. 85 (December 10,

18	

2012 Booker Report, at 98.

19	

Id. at 104.

128

|

https://www.ussc.gov/topic/research-reports

ENDNOTES

20	
Id. at 100. The extent of reduction varied broadly during each period and did not appear to have been affected
by legislation or Supreme Court decisions. Id.
21	

Id. at 98.

22	

Id.

23	
See, e.g., U.S. Sentencing Comm’n, 2017 Sourcebook of Federal Sentencing Statistics S-53 (2018) (non-government
sponsored below range cases constituted 20.1% of caseload in Fiscal Year 2017); 2010 Sourcebook of Federal Sentencing
Statistics 50 (2011) (non-government sponsored below range cases constituted 17.8% of caseload in Fiscal Year 2010);
2006 Sourcebook of Federal Sentencing Statistics 52 (2007)(non-government sponsored below range cases constituted
12.0% of caseload in Fiscal Year 2006).
24	
The “post-report period” discussed in the Commission’s 2017 report on demographic differences spanned fiscal
years 2012 through 2016. See U.S. Sentencing Comm’n, Demographic Differences in Sentencing: An Update to the 2012
Booker Report 6 (2017). The updated data analyses described in this report do not include the Koon and PROTECT Act
periods that were part of the 2012 and 2017 data analyses.
25	
Even if a judge had at least 50 cases in a given period, he or she may not have been included in the analysis of
one or both of the other periods studied depending the size of the judge’s caseloads in those periods.
26	
Random assignment is the general rule in the federal system. See, e.g., S.D.N.Y., Rules for the Division of
Business Among District Judges (2017), Rule 6, http://www.nysd.uscourts.gov/rules/rules.pdf (“In a criminal case, after
an indictment has been returned by the Grand Jury or a notice has been filed by the United States Attorney’s Office of
an intention to file an information upon the defendant’s waiver of indictment, the magistrate judge on duty will randomly
draw from the criminal wheel, in open court, the name of a judge to whom the case should be assigned for all purposes.”);
N.D. Ill. Local Rules (2017), Rule 1 (stating that, as a general matter, “the assignment of cases shall be by lot”), https://
www.ilnd.uscourts.gov/PrintContent.aspx?rid=44; see generally Report of the Proceeding of the Judicial Conference of the
United States 13 (2000) (noting that “all [federal] courts . . . employ random case assignment procedures”). Depending
in the district, there are certain exceptions to this general rule, such as a single judge’s being assigned several “related”
cases (e.g., a large-scale drug-trafficking or fraud prosecution of multiple defendants, using separate indictments) or senior
judges’ being permitted to opt out of certain types of criminal cases. See, e.g., D. Minn., Order for Assignment of Cases,
http://www.mnd.uscourts.gov/cmecf/Order-for-Assignment-of-Cases.pdf (2017).
27	
See, e.g., Crystal S. Yang, Have Interjudge Sentencing Disparities Increased in an Advisory Guidelines Reime? Evidence
From Booker, 89 N.Y.U. L. Rev. 1268, 1297 (2014); Ryan W. Scott, Inter-Judge Sentencing Disparity After Booker: A First Look,
63 Stan. L. Rev. 1, 24 (2010).
28	
Visiting district judges (with one exception, noted below), circuit judges sitting by designation as district judges,
and all magistrate judges were excluded from the analyses. Senior district judges who opted to hear criminal cases were
included in the analyses, assuming they met the minimum 50-case requirement discussed above. A single visiting judge
was included in the analysis because he served as a full-time visiting judge in a city for several years and met the 50-case
minimum for one period.
29	
See National Geographic Society, United States Regions, https://www.nationalgeographic.org/maps/unitedstates-regions/.
30	
Only the D.C. Circuit was not represented in the analysis. The sole district in the D.C. Circuit is the District
for the District of Columbia, whose criminal caseload was too small to allow an analysis of its judges (based on the
Commission’s minimum 50-case per judge requirement).
31	
Some of the larger cities by population were excluded in order to assure better geographic representation by
other cities—e.g., larger cities like Fort Worth and San Jose were excluded because other cities in the same immediate
geographic area had more judges and larger caseloads (i.e., Dallas and San Francisco), while less populous cities like
Saint Louis and Denver were included to assure representation from all regions in the country. The cities were selected
for inclusion in the Commission’s study before the results of the analyses of their judges’ sentencing practices were
conducted.
32	
Although Alexandria is not itself a major city, the federal district courthouse in Alexandria serves the entire
metropolitan area of Northern Virginia, which is a densely populated area. See About Northern Virginia, http://www.
novachamber.org/about-northern-virginia.html (noting that the counties of Northern Virginia, near Washington D.C.,
together have a population of over 2 million people).
129

Intra-City Differences in Federal Sentencing Practices
January 2019

33	
Rather than looking at all federal judges in greater New York City, the analysis is limited to federal judges in
Manhattan. Consideration of all federal judges in the larger city would require two different federal districts—the Eastern
and Southern Districts of New York—to be considered together.
34	
See, e.g., U.S. Sentencing Comm’n, Sourcebook of Federal Sentencing Statistics S-53 (2017) (Table N) (49.1% of all
cases had within range sentences, and 2.9% of cases had above range sentences).
35	
See USSG §§5G1.1(b) (“Where the statutorily authorized minimum sentence is greater than the maximum of the
applicable guideline range, the statutorily required minimum sentence shall be the guideline sentence.”) & 5G1.1(c)(2) (“[T]
he sentence may be imposed at any point within the applicable guideline range, provided that the sentence . . . is not less
than any statutorily required minimum sentence.”). For example, if a defendant’s guideline range before application of a
120-month statutory mandatory minimum sentence was 78-97 months—resulting in a guideline range of 120-120 months
under USSG § 5G1.1(b)—that case was excluded because the court had no sentencing discretion to impose a sentence less
than 120 months.
The Commission also excluded all cases with sentences imposed under the Armed Career Criminal Act (18 U.S.C. §
924(e)) even if those cases did not involve a mandatory minimum trump. Section 924(e) requires a 180-month mandatory
minimum sentence, which is above the otherwise applicable guideline minimums for many armed career criminals and
only 8 months below the guideline minimums for the most serious offenders sentenced under §4B1.4 (after full credit for
acceptance of responsibility under USSC § 3E1.1). See USSG §4B1.4 (Armed Career Criminal) & Ch. 5, Pt. A (Sentencing
Table) (guideline range, after a 3-level reduction for acceptance of responsibility, for defendants with highest offense level
and highest Criminal History Category, i.e., offense level 31 and CHC VI, is 188-235 months). For these reasons, judges’
discretion to sentence below the statutory minimum is so circumscribed that the Commission has excluded all section
924(e) cases from the current study’s analysis.
36	
See 18 U.S.C. § 3553(e); USSG §5K1.1 & 5K.3.1; see also U.S. Sentencing Commission Staff Working Group,
Federal Sentencing Practices: Sentence Reductions Based on Defendants’ Substantial Assistance to the Government, 11 Fed.
Sent’g Rptr. 18, 23, (1998) (noting that substantial assistance motions “were almost always granted” once filed by the
prosecution); L. Felipe Restrepo, To Be Or Not To Be A Cooperating Defendant, Crim. Justice 25 (Winter 1993) (“Practically
speaking, judges rarely deny the [substantial assistance] motion outright ….”).
37	
Until the Supreme Court decided Dean v. United States, 137 S. Ct. 1170 (April 3, 2017), it was unclear whether
district courts in such cases had the authority to vary below the guideline range for another count of conviction (e.g.,
robbery, sentenced under USSG §2B3.1) to account for a perceived excessive total penalty level resulting from a
consecutive statutory mandatory minimum sentence such as that required by section 924(c). Before Dean, some judges
varied on that ground (see, e.g., United States v. Roberson, 573 F. Supp.2d 1040 (N.D. Ill. 2008) (varying from the guideline
minimum of 46 months for a bank robbery conviction to 1 month in order to account for a 84-month consecutive sentence
under § 924(c)), while others believed that they had no such authority to vary on that ground. The percent difference from
the guideline minimum in cases where courts did vary on that ground was usually very substantial. Because the extent
of a below range sentence was often extremely large when a judge varied—and thereby could skew those judges’ average
percent differences from the guideline minimums compared to the average percent difference of judges who did not
believe they could vary on that ground—all cases with a count of conviction under 18 U.S.C. §§ 924(c) or 1028A or other
statutes requiring a consecutive mandatory sentence of imprisonment were excluded from the Commission’s analysis. In
addition, the Commission excluded cases with a single count of conviction under such a statute because a court lacked
authority to depart or vary downwardly in such cases.
38	
Since 2010, when the Commission amended the Sentencing Table to expand Zone B by one offense level, a
case with a guideline minimum of less than 10 months necessarily has fallen in Zone A or Zone B of the Sentencing Table.
Before 2010, cases with guideline minimums of 8 or 9 months fell within Zone C. In order to ensure consistency in the
Commission’s analysis of cases from fiscal year 2005 through fiscal year 2017, the Commission has excluded all cases
with guideline minimums less than 10 months for that entire time period rather than exclude cases based on their zone
designation. The Commission excluded such cases for two reasons. First, because all cases in Zone A have guideline
minimums of 0 months, judges cannot depart or vary below the minimum and, in addition, there is no mathematical way to
calculate the percent difference from the guideline minimum when a court imposes a sentence above the minimum in such
cases. Second, because all of ranges with guideline minimums below 10 months are narrow and their minimums are low
(e.g., 0-6 months, 6-12 months), sentences imposed above or below the guideline minimums have a much greater positive
or negative percent difference than typical sentences above or below the guideline minimums in Zones C and D. The latter
ranges have higher and broader ranges of months (e.g., 12-18 months, 46-57 months, and 121-151 months). Including
cases with guideline minimums below 10 months would thus skew the average percent difference analysis.

130

|

https://www.ussc.gov/topic/research-reports

ENDNOTES

39	
The Commission excluded 10,753 cases because of incomplete sentencing information in those cases. Such
cases were excluded because they did not allow the Commission to make determinations such as the guideline minimum
in a case or whether a statutory mandatory minimum penalty applied.
40	
Of the 291,763 cases, 3.7% were excluded for insufficient documentation. In addition, 15.7% had their
guideline minimums trumped by a statutory mandatory minimum or were subject to a mandatory minimum sentence
under the Armed Career Criminal Act; 26.1% had a substantial assistance or fast-track departure; 5.6% had at least
one mandatory consecutive statutory minimum sentence under a statute such as section 924(c); 0.6% had a guideline
minimum of life imprisonment or involved an upward departure and variance to life imprisonment; and 15.5% had
guideline minimums of less than 10 months. Some cases fell within more than one of these groups, which explains why
the total amount of all these cases combined exceeds the total percentage of excluded cases mentioned above.
41	
After both the exclusions of the five categories of cases and the additional exclusions of cases handled by
judges who did not meet the 50-case minimum and cases with incomplete information, the national caseload resembles
the 30-city caseload during the same 13-year time period. After the same exclusions, the same percentage of all cases,
49.2% (478,833 of 972,648 cases), remained. The percentage of excluded cases (nationally) was as follows: 19.1% had
their guideline minimums trumped by a statutory mandatory minimum or were subject to a mandatory minimum sentence
under the Armed Career Criminal Act; 21.1% had a substantial assistance or fast-track departure; 4.2% had at least
one mandatory consecutive statutory minimum sentence under a statute such as section 924(c); 0.5% had a guideline
minimum of life imprisonment or involved an upward departure and variance to life imprisonment; and 21.8% of cases had
guideline minimums of less than 10 months. The analysis also excluded 6.1% of cases nationally because of incomplete
sentencing information in the documentation submitted to the Commission.
42	

See Yang, supra note 27, at 1324-25.

43	
In over 90% of the 19,460 cases subject to mandatory minimum penalties (17,755, or 91.2%), defendants were
subject to either 60- or 120-month mandatory minimum sentences. The remaining 8.7% of cases had a wide variety of
statutes requiring mandatory minimum sentences ranging from 1 month to 300 months. Of the cases with 60-month
mandatory minimum sentences, the average guideline minimum was 118 months and the average sentence imposed was
97 months (37 months above the statutory mandatory minimum). Of the cases with 120-month mandatory minimum
sentences, the average guideline minimum was 198 months and the average sentence imposed was 171 months (51
months above the mandatory minimum).
44	
See Gall, 552 U.S. at 49, 50 n.6; see also Rosales-Mireles v. United States, 138 S. Ct. 1897, ​
1903-04, 1908 (2018); Molina-Martinez v. United States, 136 S. Ct. 1338, 1349 (2016); Peugh v. United States, 569 U.S.
530, 535, 542, 545 (2013). The three-step Booker process is discussed at USSG §1B1.1, comment. (backg’d).
45	
Of the cases in which judges imposed sentences outside of the guideline ranges, judges departed or varied
downwardly in 95.4% of cases and departed or varied upwardly in 4.6% of cases (a nearly 21:1 ratio between downward
and upward departures and variances).
46	
See, e.g., Scott, supra note 27, at 31-34 (comparing average sentences of judges in Boston after Booker); Paul J.
Hofer et al., The Effect of the Federal Sentencing Guidelines on Inter-Judge Sentencing Disparity, 90 J. Crim. L. & Criminology
239, 286-96 (1999) (comparing average sentences for judges in several cities, both before and after effective date of
federal sentencing guidelines).
47	
See U.S. Sentencing Comm’n, Sourcebook of Federal Sentencing Statistics S-32 (2017) (Table 13) (average sentence
for drug-trafficking cases was 70 months, and average sentence for firearms cases was 71 months; conversely, average
sentence for immigration cases was 12 months and average sentence for fraud cases was 26 months).
48	
In the past two decades, the national average guideline minimum for all federal offenders consistently has been
around five years (60 months). See U.S. Sentencing Comm’n, Federal Alternative-To-Incarceration Court Programs 32 (2017)
(average guideline minimum in 2016 was 59 months); 2012 Booker Report, at 60 (showing guideline minimum slightly
above or slightly below 60 months from 1996 through 2011).
49	

See Michael O. Finkelstein & Bruce Levin, Statistics for Lawyers 22 (2d ed. 2000).

50	
For example, if the judge at the top of the bar graph had an average percent difference of 10.0% and the judge
at the bottom of the bar graph had an average percent difference of -40.0%, the total spread for the city would be 50.0.
51	
Ottavania v. State University of New York at New Paltz, 875 F.2d 365, 371 (2d Cir. 1989) (citation and internal
quotation marks omitted).
131

Intra-City Differences in Federal Sentencing Practices
January 2019

52	

Charlotte, Memphis, and San Antonio each had fewer than five judges in at least one period.

53	
The specific percentages (taken to two decimal points) were 32.43% and 17.02%, which when summed and
rounded equals 49.5.
54	
The following 23 cities had increases in their total spreads from the Booker to Gall Periods: Alexandria, Atlanta,
Baltimore, Boston, Charlotte, Chicago, Dallas, Denver, Houston, Los Angeles, Memphis, Miami, Minneapolis, Philadelphia,
Phoenix, Pittsburgh, Portland, Saint Louis, Salt Lake City, San Diego, San Francisco, San Juan, and Tampa.
The following seven cities had decreases in their total spreads from the Booker to Gall Periods: Cleveland, Columbus,
Detroit, Manhattan, Oklahoma City, San Antonio, and Seattle.
The following 22 cities had increases in their standard deviations from the Booker to Gall Periods: Alexandria, Atlanta,
Baltimore, Boston, Chicago, Dallas, Denver, Detroit, Houston, Los Angeles, Miami, Minneapolis, Philadelphia, Phoenix,
Pittsburgh, Portland, Saint Louis, Salt Lake City, San Diego, San Francisco, San Juan, and Tampa.
The following five cities had decreases in their standard deviations from the Booker to Gall Periods: Cleveland, Columbus,
Manhattan, Oklahoma City, and Seattle.
55	
The following 20 cities had increases in their total spreads from the Gall to Post-Report Periods: Alexandria,
Charlotte, Chicago, Cleveland, Dallas, Denver, Detroit, Houston, Los Angeles, Manhattan, Memphis, Miami, Philadelphia,
Phoenix, Portland, Salt Lake City, San Antonio, San Diego, San Juan, and Seattle.
The following ten cities had decreases in their total spreads from the Gall to Post-Report Periods: Atlanta, Baltimore,
Boston, Columbus, Minneapolis, Oklahoma City, Pittsburgh, Saint Louis, San Francisco, and Tampa.
The following 16 cities had increases in their standard deviations from the Gall to Post-Report Periods: Atlanta, Cleveland,
Dallas, Denver, Detroit, Houston, Los Angeles, Manhattan, Miami, Philadelphia, Phoenix, Portland, Salt Lake City, San
Diego, San Juan, and Seattle.
The following 11 cities had decreases in their standard deviations from the Gall to Post-Report Periods: Alexandria,
Baltimore, Boston, Chicago, Columbus, Minneapolis, Oklahoma City, Pittsburgh, Saint Louis, San Francisco, and Tampa.
56	
The main four guidelines constituted 71.0% of all cases in the Commission’s city-level dataset, while all other
guidelines constituted 29.0% of cases.
57	
Because of rounding to one decimal point, some of the totals for the cities in Appendix C add up to less than
100.0%.
58	
Weighting is a common technique in the social sciences whereby a researcher assigns comparable weights
to different datapoints in an analysis in order to control for unweighted differences. Weighting is most commonly used
in survey methodology. See, e.g., Luke W. Miratrix et al., Worth Weighting? How to Think About and Use Weights in Survey
Experiments, 26 Political Analysis 275 (2018). The Commission used a somewhat similar method to control for differences
in caseload composition among the judges in each city and among the cities’ caseloads over time.
59	
For that weighting analysis, only judges who met the 50-case minimum for two consecutive periods (e.g.,
Booker to Gall, or Gall to Post-Report) were included in order to determine the effect of changes in the city’s caseload
composition from one period to the next. Therefore, it is not possible to compare all three periods to each other because
there were not the exact same judges in all three periods in any of the 30 cities. Therefore, only two periods at a time
(Booker to Gall, and Gall to Post-Report) could be compared based on the judges common to those two periods.
60	
Of all major guideline types, illegal reentry (§2L1.2) had the highest within-range rate from fiscal year 2005 to
fiscal year 2017 (after the case exclusions discussed above):
Average Within-Range Rates (Fiscal Years 2005-2017)
2B1.1: 46.0%
2D1.1: 46.1%
2K2.1: 60.8%
2L1.2: 61.1%
Other: 49.6%

132

|

https://www.ussc.gov/topic/research-reports

ENDNOTES

61	

In the Booker and Gall Periods, Houston had the following percentages of guideline types:
Booker Period			

Gall Period

2B1.1: 7.8%			
2D1.1: 20.3%			
2K2.1: 17.2%			
2L1.2: 24.0%			
Other: 30.7%			

2B1.1: 8.0%
2D1.1: 10.9%
2K2.1: 8.3%
2L1.2: 51.2%
Other: 21.6%

62	

Federal Judicial Center, Reference Manual on Scientific Evidence 291 (3d ed. 2011).

63	

Id.

64	

See, e.g., Nancy Pfenning, Elementary Statistics 96-97 (2011) (discussing the 1.5 x IQR test).

65	
Although there were 413 different judges in the 30-city dataset, many of those judges sentenced in more than
one period. Therefore, the outlier analysis includes a total of 909 unique analyses of judicial sentencing practices in the
dataset during three periods. Note that a judge could be an outlier in one period but not an outlier in a different period.
66	
The cities with at least one outlier judge in at least one period were Atlanta, Chicago, Cleveland, Columbus,
Dallas, Detroit, Los Angeles, Manhattan, Memphis, Miami, Minneapolis, Oklahoma City, Philadelphia, Phoenix, Pittsburgh,
Portland, Salt Lake City, San Diego, San Juan, and Tampa.
67	
298.

See Finkelstein & Levin, supra note 49, at 18-19; see also FJC’s Reference Manual on Scientific Evidence, supra, at

68	

Rebecca M. Warner, Applied Statistics: From Bivariate Through Multivariate Techniques 1114 (2013 2d ed).

69	
The standard deviation is a less meaningful statistic when there are only a few datapoints being analyzed. The
Commission thus only has reported the standard deviation for cities with at least five judges in all three periods.
70	
See Finkelstein & Levin, supra note 49, at 18-19 (discussing the manner in which the standard deviation is
calculated); Warner, supra note 68, at 59 (same). Reporting the standard deviation is preferable to reporting the variance.
The variance, which is the standard deviation squared, is more sensitive to outliers than the standard deviation because
the influence of an outlier is magnified (insofar as the outlier’s average percent difference from the guideline minimum is
squared in calculating the variance).

133

United States Sentencing Commission
WWW.USSC.GOV

THURGOOD MARSHALL FEDERAL JUDICIARY BUILDING
ONE COLUMBUS CIRCLE N.E.
SUITE 2-500, SOUTH LOBBY
WASHINGTON, DC 20002-8002