Tracy Conley v. United States

In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 20‐2439
TRACY CONLEY,
Petitioner‐Appellant,
v.
UNITED STATES OF AMERICA,
Respondent‐Appellee.
____________________
Appeal from the United States District Court for the
Northern District of Illinois, Eastern Division.
No. 1:18‐cv‐07122 — Sharon Johnson Coleman, Judge.
____________________
ARGUED MAY 26, 2021 — DECIDED JULY 21, 2021
____________________
Before EASTERBROOK, ROVNER, and HAMILTON, Circuit
Judges.
HAMILTON, Circuit Judge. Tracy Conley has invoked 28
U.S.C. § 2255 to vacate convictions arising from his participa‐
tion in a “fake stash house” sting orchestrated by the Bureau
of Alcohol, Tobacco, Firearms, and Explosives (ATF). Conley
asserts that his convictions were obtained unlawfully through
racially selective law enforcement and outrageous govern‐
ment conduct, in violation of his Fifth Amendment equal
2 No. 20‐2439
protection and due process rights, respectively. The district
court denied Conley’s motion, and we affirm. We disagree
with the district court’s decision to require “clear and con‐
vincing” evidence for Conley’s selective enforcement claim,
but his evidence cannot meet even the less‐demanding stand‐
ard of preponderance of the evidence. This circuit has not rec‐
ognized a defense for “outrageous government conduct,” and
even if we did, ATF’s conduct in Conley’s case would not sat‐
isfy the standard other circuits have applied. Regardless of the
wisdom of this fake stash house sting, it did not violate Con‐
ley’s constitutional rights.
I. Background
Our opinion affirming Conley’s convictions on direct ap‐
peal sets out the facts of his case. United States v. Conley, 875
F.3d 391 (7th Cir. 2017). Here we focus on the facts relevant to
his selective enforcement and outrageous conduct theories.
On November 1, 2011, Conley found himself stranded at a gas
station without money for gas. There he ran into three ac‐
quaintances—Anthony Adams, David Flowers, and Anwar
Trapp—who successfully recruited Conley to help rob a co‐
caine stash house later that day. Adams, David, and Trapp
had themselves been recruited recently for the robbery by Da‐
vid’s brother, Myreon Flowers.
Myreon had been tipped off a week earlier about the sup‐
posed stash house by a man claiming to be a disgruntled cou‐
rier for a drug cartel. The courier told Myreon that the stash
house contained fifty kilograms of cocaine, and that to steal it,
Myreon should assemble an armed team to overpower the
cartel guards. Myreon jumped at the opportunity, bragging
about how easy the robbery would be compared to others he
had pulled off. Myreon quickly assembled a crew, recruiting
No. 20‐2439 3
his brother David Flowers and their cousins Anwar Trapp
and Dwayne Jones. David also recruited Anthony Adams,
who in turn suggested recruiting Conley.
The supposed cartel courier was an undercover ATF agent.
There was no stash house or real drugs, just a convincing ruse
designed to ensnare Myreon and his crew in a conspiracy to
steal fifty kilograms of cocaine at gunpoint. It turns out the
FBI had originally focused on Myreon through an investiga‐
tion into the Belizean Bloods, a Chicago street gang. Confi‐
dential sources had told the FBI that Myreon was an active
member of a Belizean Bloods armed robbery crew, as well as
a murder suspect. Based on that information, the FBI had re‐
ferred Myreon to the ATF as a potential target for a stash
house sting, and the ATF obliged.
On November 1, Myreon’s group, including Conley, met
in Adams’s basement to plan the robbery. Conley not only
agreed to participate but volunteered for a frontline role as
one of two armed robbers who would confront the guards
and steal the cocaine. Conley even asked the group whether
they should kill the cartel courier (i.e., the undercover agent)
but fortunately was told not to harm him. Later that day, the
group met the undercover agent at a secluded parking lot
near the supposed stash house. Conley and the others were
dressed in dark clothing, wearing latex gloves, and equipped
with walkie‐talkies and a toolbox containing three guns for
the robbery. Myreon told Conley, Adams, and another mem‐
ber to get into a van that would take them to the stash house
for the robbery. As soon as they were all in the van, the un‐
dercover agent gave the arrest signal.
4 No. 20‐2439
A jury convicted Conley of conspiring and attempting to
possess with intent to distribute more than five kilograms of
cocaine, in violation of 21 U.S.C. §§ 841(a)(1) and 846; pos‐
sessing a firearm in furtherance of a drug trafficking offense,
in violation of 18 U.S.C. § 924(c)(1)(A); and being a felon in
possession of a firearm in violation of 18 U.S.C. § 922(g)(1).
The district court sentenced Conley to fifteen years in prison,
which included a ten‐year mandatory minimum based on the
large amount of non‐existent drugs he agreed to steal—an
amount the ATF made up just to trigger that ten‐year mini‐
mum.
On direct appeal, we affirmed the denial of Conley’s mo‐
tion for acquittal. United States v. Conley, 875 F.3d 391 (7th Cir.
2017). We found that sufficient evidence supported his con‐
victions, id. at 397–402, and that he could not argue entrap‐
ment because the ATF recruited neither Conley nor those who
recruited him. Id. at 402.
Conley then filed his § 2255 motion in the district court
seeking to set aside his convictions on two new grounds. First,
he asserts that federal agents intentionally targeted him for
the sting because he is Black, in violation of the equal protec‐
tion component of the Fifth Amendment. Second, Conley as‐
serts that even though he cannot show entrapment, the ATF’s
conduct in executing this fake stash house sting was so outra‐
geous that prosecuting him violated his due process rights.
The district court denied Conley’s § 2255 motion but granted
a certificate of appealability on both claims. Conley v. United
States, No. 18 C 7122, 2020 WL 4226676, at *8 (N.D. Ill. July 23,
2020).
Judge Coleman denied Conley’s selective‐enforcement
claim by relying exclusively on then‐Chief Judge Castillo’s
No. 20‐2439 5
denial of a nearly identical claim brought by fake stash house
defendants in United States v. Brown, 299 F. Supp. 3d 976 (N.D.
Ill. 2018). See Conley, 2020 WL 4226676, at *4–6. Judge Cole‐
man certified the issue for appeal, however, “based on the un‐
certainty of the applicable evidentiary standard” governing a
selective‐enforcement claim in this circuit. Id. at *6. As for out‐
rageous government conduct, Judge Coleman denied Con‐
ley’s claim because this circuit has repeatedly rejected the ex‐
istence of such a defense. Id. at *3–4. Again, however, Judge
Coleman certified Conley’s appeal on this issue: “If there ever
was a situation in which the conduct of law enforcement
agents was so outrageous that a criminal defendant’s due pro‐
cess rights have been violated, this is it.” Id. at *4.
While this appeal was pending, Judge Coleman granted
Conley’s motion for compassionate release under 18 U.S.C.
§ 3582(c)(1)(A)(i). Conley’s release from prison does not re‐
move our jurisdiction over his § 2255 motion or render it
moot. A movant under § 2255 must be “in custody.” 28 U.S.C.
§ 2255(a). But Conley “was in custody when he filed [his] mo‐
tion” for § 2255 relief, “and that is all that is required….”
Torzala v. United States, 545 F.3d 517, 521 (7th Cir. 2008), citing
Spencer v. Kemna, 523 U.S. 1, 7 (1998). Moreover, despite his
release from prison, Conley is still subject to a three‐year term
of supervised release, which is “a form of custody,” Clarke v.
United States, 703 F.3d 1098, 1101 (7th Cir. 2013), and carries
collateral consequences sufficient to prevent his motion from
being moot. See Spencer, 523 U.S. at 7 (“An incarcerated con‐
vict’s (or a parolee’s) challenge to the validity of his conviction
always satisfies the case‐or‐controversy requirement, because
the incarceration (or the restriction imposed by the terms of the
parole) constitutes a concrete injury, caused by the conviction
6 No. 20‐2439
and redressable by invalidation of the conviction.”) (empha‐
ses added).
II. Controversy Surrounding Fake Stash House Stings
Before turning to the merits, we provide some background
on the ATF’s controversial and now‐abandoned practice of
conducting fake stash house stings in Chicago. As we noted
in Conley’s direct appeal, there is a “substantial body of criti‐
cism of similar stash house cases both from this circuit and
others.” Conley, 875 F.3d at 402. We have described the tactic
as “‘tawdry,’ noting in particular how these operations are
‘directed at unsophisticated, and perhaps desperate defend‐
ants’ like Conley who easily take the all‐too‐tempting bait put
out for them by the government.” Id. at 402–03, quoting United
States v. Lewis, 641 F.3d 773, 777 (7th Cir. 2011).
These stings are troubling in several ways. First, the agents
in these stings blindly encourage the initial target to rope in a
large group of co‐conspirators who otherwise may have
stayed out of trouble. See United States v. Washington, 869 F.3d
193, 225 (3d Cir. 2017) (McKee, J., dissenting) (defendant “was
not even the intended target of this operation. Despite his
criminal past, [he] was not necessarily destined to commit fu‐
ture crimes…. ‘we donʹt want the government pushing [crim‐
inals] back into a life of crime.’”), quoting United States v. Kin‐
dle, 698 F.3d 401, 415–16 (7th Cir. 2012) (Posner, J., dissenting),
on reh’g en banc sub nom. United States v. Mayfield, 771 F.3d
417 (7th Cir. 2014).
Second, the vast majority of defendants caught in these
stings have been poor people of color. See Brown, 299 F. Supp.
3d at 986 (“These sting operations have used tremendous
public resources to investigate and prosecute a large number
No. 20‐2439 7
of principally minority individuals for fictitious crimes.”); see
also United States v. Black, 750 F.3d 1053, 1057 (9th Cir. 2014)
(Reinhardt, J., dissenting from denial of rehearing en banc)
(“In this era of mass incarceration, in which we already lock
up more of our population than any other nation on Earth, it
is especially curious that the government feels compelled to
invent fake crimes and imprison people for long periods of
time for agreeing to participate in them….”).
Third, these stings give law enforcement free rein to ma‐
nipulate sentences by setting imaginary drug amounts. See
United States v. Briggs, 623 F.3d 724, 729–30 (9th Cir. 2010)
(criticizing ability to manipulate sentences with fake drug
amounts).
Fourth, economic analysis suggests that fake stash house
stings may actually facilitate the illegal drug trade. They make
real stash houses more secure by deterring potential robbers.
See Kindle, 698 F.3d at 416 (Posner, J., dissenting) (“The oper‐
ators of stash houses would pay law enforcement to sting po‐
tential stash house robbers.”). These problems have led jurists
to question whether fake stash house stings comport with
“the pride we take in presenting American criminal justice as
a system that treats defendants fairly and equally.” United
States v. Flowers, 712 F. App’x. 492, 511 (6th Cir. 2017) (Stranch,
J., concurring).
Amid this controversy, many defendants in Chicago have
challenged the ATF’s fake stash house stings as racially selec‐
tive. Sitting en banc in 2015, we laid out principles for grant‐
ing discovery in these cases. United States v. Davis, 793 F.3d
712, 719–23 (7th Cir. 2015) (en banc). Various stash house cases
then progressed on different timelines. In December 2017, the
matter culminated in a unique two‐day evidentiary hearing
8 No. 20‐2439
in the Northern District of Illinois before nine district judges
presiding over twelve similar stash house cases. Statistical ex‐
perts testified to whether the ATF’s fake stash house program
was racially discriminatory. See Brown, 299 F. Supp. 3d at 993.
Following that hearing, Judge Castillo issued a careful
opinion denying defendants’ selective‐enforcement claims
due to flaws and limits in the statistical analysis performed by
the court’s appointed expert, Columbia University Law Pro‐
fessor Jeffrey Fagan. Id. at 1001–13. The opinion criticized the
ATF’s fake stash house stings as a matter of policy, id. at 984–
85, and it took pains to clarify that the court was not deciding
whether these stings “are honorable or fair,” but only
“whether law enforcement agents violated the Due Process
Clause of the Fifth Amendment in pursuing these investiga‐
tions.” Id. at 986.
This history forms the backdrop of Conley’s appeal. In
particular, Judge Castillo’s analysis in Brown is critical here
because Conley relies on the statistics in the Fagan Report to
prove his selective‐enforcement claim. We now turn to the
merits of that claim.
III. Selective Enforcement
Conley asserts that the ATF targeted him for the sting be‐
cause he is Black. Conley did not raise this selective‐enforce‐
ment claim until his § 2255 motion. But we may consider it
because, as the district court noted, the government chose not
to raise procedural default as a defense to this claim. See Per‐
rone v. United States, 889 F.3d 898, 904 (7th Cir. 2018) (“Proce‐
dural default is a waivable defense.”).
Because Conley himself was not the initial target of the
sting, his theory is that the ATF first targeted Myreon Flowers
No. 20‐2439 9
because of his race, with further hopes that Myreon would re‐
cruit other Black people like Conley as co‐conspirators. If
proven, that would amount to purposeful discrimination in
violation of the Fifth Amendment’s equal protection compo‐
nent. See Whren v. United States, 517 U.S. 806, 813 (1996)
(“[T]he Constitution prohibits selective enforcement of the
law based on considerations such as race.... [T]he constitu‐
tional basis for objecting to intentionally discriminatory ap‐
plication of laws is the Equal Protection Clause.”); Chavez v.
Illinois State Police, 251 F.3d 612, 635 (7th Cir. 2001) (same).1
Racially selective law enforcement is a quintessential
equal protection violation. One of the Supreme Court’s earli‐
est equal protection opinions granted habeas corpus relief to
petitioners who alleged selective enforcement of building
codes against Chinese‐owned laundries in San Francisco. Yick
Wo v. Hopkins, 118 U.S. 356 (1886). The Yick Wo petitioners
showed that authorities had denied the applications of two
hundred Chinese people for permits to operate laundries in
wooden buildings but granted the applications of eighty non‐
Chinese people “under similar conditions.” Id. at 374. The
Court made clear that, in such a situation, law enforcement
can be “directed so exclusively against a particular class of
persons … with a mind so unequal and oppressive as to
amount to a practical denial by the state of that equal protec‐
tion of the laws which is secured to the petitioners … by the
broad and benign provisions of the fourteenth
1 The Equal Protection Clause of the Fourteenth Amendment applies
only to the States, but it has been “reverse‐incorporated” to apply to the
federal government as a component of the Fifth Amendment’s Due Pro‐
cess Clause. Bolling v. Sharpe, 347 U.S. 497, 498–500 (1954) (racially segre‐
gated public schools in District of Columbia violated Fifth Amendment).
10 No. 20‐2439
amendment….” Id. at 373. Today, Yick Wo is seen as providing
historical roots not only for selective enforcement claims but
also for selective‐prosecution claims. See United States v. Arm‐
strong, 517 U.S. 456, 464–66 (1996) (discussing Yick Wo as one
source of selective‐prosecution claims).
Selective prosecution occurs when, from among the pool
of people referred by police, a prosecutor pursues similar
cases differently based on race. Selective enforcement occurs
when police investigate people of one race but not similarly‐
situated people of a different race. Hence, with selective en‐
forcement, “the constitutional problem … precede[s] the
prosecutor’s role.” Davis, 793 F.3d at 720. It does not matter if
prosecutors then pursue each case equally because the pool of
defendants itself was racially selected. Id.
As equal protection claims, both selective prosecution and
selective enforcement require proof “that the defendants’ ac‐
tions had a discriminatory effect and were motivated by a dis‐
criminatory purpose.” Chavez, 251 F.3d at 635–36, citing Per‐
sonnel Administrator of Mass. v. Feeney, 442 U.S. 256, 272–74
(1979), Arlington Heights v. Metro. Housing Dev. Corp., 429 U.S.
252, 264–66 (1977), and Washington v. Davis, 426 U.S. 229, 239–
42 (1976). A plaintiff must show discriminatory purpose “in
his case.” McCleskey v. Kemp, 481 U.S. 279, 292 (1987). And dis‐
criminatory purpose “implies more than … awareness of con‐
sequences. It implies that the decisionmaker … selected or re‐
affirmed a particular course of action at least in part ‘because
of,’ not merely ‘in spite of,’ its adverse effects upon an identi‐
fiable group.” Id. at 298, quoting Feeney, 442 U.S. at 279.
No. 20‐2439 11
A. Standard of Proof
The first issue we confront is the standard of proof Conley
must meet to prove his selective‐enforcement claim. The dis‐
trict court granted a certificate of appealability on this claim
“because reasonable jurists could debate that his claim should
have been resolved in a different manner based on the uncer‐
tainty of the applicable evidentiary standard.” Conley, 2020
WL 4226676, at *6. We review this legal issue de novo. Martin
v. United States, 789 F.3d 703, 705 (7th Cir. 2015). The govern‐
ment argues that Conley must meet the same “clear evidence”
standard that governs selective‐prosecution claims. See Arm‐
strong, 517 U.S. at 465 (“In order to dispel the presumption
that a prosecutor has not violated equal protection, a criminal
defendant must present ‘clear evidence to the contrary.’”),
quoting United States v. Chemical Foundation, 272 U.S. 1, 14–15
(1926). Conley counters that the reasons for Armstrong’s clear‐
evidence standard are specific to prosecutors, so that selec‐
tive‐enforcement claims against police need be proven by
only a preponderance of the evidence. The district court
agreed with the government’s view, as did Judge Castillo in
Brown. We disagree, however, and hold that racially selective‐
enforcement claims must be proven by a preponderance of
the evidence.
Our analysis proceeds as follows. First, in Part III‐A‐1, we
explain Armstrong’s clear‐evidence standard and why its ra‐
tionale does not extend to selective‐enforcement claims. In
Part III‐A‐2, we explain why McCleskey v. Kemp does not re‐
quire a heightened standard of proof. Finally, in Part III‐A‐3,
we draw connections to other doctrines showing that a pre‐
ponderance standard fits better into the broader legal land‐
scape.
12 No. 20‐2439
1. Armstrong Does Not Set the Standard of Proof
a. Armstrong’s Clear‐Evidence Standard
Equal protection claims generally must be proven by a
preponderance of the evidence. See Radentz v. Marion County,
640 F.3d 754, 756 (7th Cir. 2011) (holding in § 1983 equal pro‐
tection case that “plaintiffs first must demonstrate by a pre‐
ponderance of the evidence that they were the victims of in‐
tentional discrimination”), citing Williams v. Seniff, 342 F.3d
774, 788 n.13 (7th Cir. 2003) (“the same standards for proving
intentional discrimination apply to Title VII and § 1983 equal
protection”); see also Harris v. Arizona Indep. Redistricting
Comm’n, 136 S. Ct. 1301, 1307 (2016) (using “more probable
than not” standard for equal protection redistricting claim);
Hunter v. Underwood, 471 U.S. 222, 225 (1985) (citing with ap‐
proval circuit court’s statement that “plaintiffs must prove by
a preponderance of the evidence that racial discrimination
was a substantial or motivating factor”).
Selective‐prosecution claims, however, must be proven by
“clear evidence.” Armstrong, 517 U.S. at 465. In Armstrong, the
Supreme Court reversed a grant of discovery in a selective‐
prosecution case despite evidence that “every defendant in
every crack‐cocaine prosecution filed by a particular United
States Attorney’s office and assigned to the public defender
was black.” Davis, 793 F.3d at 719. The Supreme Court held
that, because a “selective‐prosecution claim asks a court to ex‐
ercise judicial power over a ‘special province’ of the Execu‐
tive,” the plaintiffs needed further evidence showing that
comparators of another race were not prosecuted before a
court could even order discovery against the prosecutor’s of‐
fice. Armstrong, 517 U.S. at 464, 468.
No. 20‐2439 13
The Court derived that discovery requirement from a
heightened “clear‐evidence” standard governing the merits
of a selective prosecution claim. Id. at 468 (“The justifications
for a rigorous standard for the elements of a selective‐prose‐
cution claim thus require a correspondingly rigorous stand‐
ard for discovery in aid of such a claim.”). Armstrong’s “clear‐
evidence” standard is based in turn on the need to preserve
prosecutorial discretion and on the “presumption of regular‐
ity” that applies to federal prosecutors as officers of the Exec‐
utive Branch:
The Attorney General and United States Attor‐
neys retain “‘broad discretion’” to enforce the
Nation’s criminal laws. Wayte v. United States,
470 U.S. 598, 607 (1985) (quoting United States v.
Goodwin, 457 U.S. 368, 380, n. 11 (1982)). They
have this latitude because they are designated
by statute as the President’s delegates to help
him discharge his constitutional responsibility
to “take Care that the Laws be faithfully exe‐
cuted.” U.S. Const., Art. II, § 3; see 28 U.S.C.
§§ 516, 547. As a result, “[t]he presumption of
regularity supports” their prosecutorial deci‐
sions and, “in the absence of clear evidence to
the contrary, courts presume that they have
properly discharged their official duties.”
United States v. Chemical Foundation, Inc., 272
U.S. 1, 14–15 (1926)….
* * *
In order to dispel the presumption that a
prosecutor has not violated equal protection, a
criminal defendant must present “clear
14 No. 20‐2439
evidence to the contrary.” Chemical Foundation,
supra, at 14–15.
517 U.S. at 464–65.
b. Armstrong’s Rationale Does Not Extend to Selective
Enforcement
The reasons underlying Armstrong’s clear‐evidence stand‐
ard for selective‐prosecution claims simply do not extend to
selective‐enforcement claims brought against police. As the
above passage shows, Armstrong’s “clear‐evidence” standard
invokes language from Chemical Foundation, a bedrock pre‐
sumption‐of‐regularity case. The presumption of regularity is
a deference doctrine that limits judicial scrutiny of procedures
and motivations supporting certain Executive Branch deci‐
sions. The presumption is driven by separation‐of‐powers
concerns, which increase as courts venture closer to core ex‐
ecutive activity. The presumption has been applied to differ‐
ent degrees depending on “the Court’s assessment of the rel‐
evant decisionmaking scheme across several dimensions: pro‐
cedural fairness, accountability, and degree of discretion, as
well as complexity….” Note, The Presumption of Regularity in
Judicial Review of the Executive Branch, 131 Harv. L. Rev. 2431,
2432–33 (2018). The presumption of regularity is an analytic
tool, not an excuse to rubberstamp any and all executive ac‐
tion as lawful absent clear evidence to the contrary.
In Armstrong, the court applied the presumption to prose‐
cutorial decision‐making. And in Reno v. American‐Arab Anti‐
Discrimination Committee, the Court extended that logic to fed‐
eral deportation decisions, which, like prosecutorial deci‐
sions, implicate core executive strategy that courts should not
over‐scrutinize. 525 U.S. 471, 490–91 (1999). The Court has not
No. 20‐2439 15
extended the presumption of regularity, however, to street‐
level police investigations. Few police officers are high‐level
executive officers. And while their decisions certainly reflect
law enforcement priorities, judicial inquiry into their motives
is routine. We therefore see no persuasive reason to extend
Armstrong’s heightened standard to selective‐enforcement
claims.
Our en banc opinion in Davis took this limited approach
to Armstrong. Like this case, Davis arose in the context of a fake
stash house sting. But the issue there concerned Armstrong’s
discovery standard. The district court had granted broad dis‐
covery against the prosecutor’s office, ATF, FBI, and others
based on the fact that the prosecution had “brought at least
twenty purported phony stash house cases, with the over‐
whelming majority of the defendants named being individu‐
als of color.” Davis, 793 F.3d at 719. The government argued
that, under Armstrong, that was an insufficient showing of ra‐
cial discrimination to authorize discovery.
We agreed with respect to discovery from the prosecutor’s
office, but we disagreed with respect to discovery from the
ATF and FBI for selective enforcement because “the sort of
considerations that led to the outcome in Armstrong do not
apply to a contention that agents of the FBI or ATF engaged
in racial discrimination when selecting targets for sting oper‐
ations, or when deciding which suspects to refer for prosecu‐
tion.” Id. at 721. “Armstrong was about prosecutorial discre‐
tion,” and unlike prosecutors, “Agents of the ATF and FBI are
not protected by a powerful privilege or covered by a pre‐
sumption of constitutional behavior.” Id. at 720. On the con‐
trary, police officers are regularly called before courts to jus‐
tify their tactics. Id. at 720–21. Accordingly, Davis found that
16 No. 20‐2439
Armstrong’s requirement to show comparators before grant‐
ing discovery did not apply to the plaintiffs’ selective‐enforce‐
ment claims.
The government in this appeal asks us not to extend Da‐
vis’s reasoning when it comes to the standard of proof. But
given that Armstrong imposed its “rigorous standard for dis‐
covery” only to match the correspondingly “rigorous stand‐
ard for the elements of a selective‐prosecution claim,” 517 U.S.
at 468, the distinction we drew in Davis for discovery natu‐
rally extends to the standard of proof. Indeed, we recognized
in Chavez that Armstrong does not govern at least one aspect
of proving discriminatory effect in a selective‐enforcement
case. There, we affirmed summary judgment on a selective‐
enforcement claim on the merits without suggesting a stand‐
ard of proof higher than preponderance applied. Rather, we
rejected plaintiffs’ need to satisfy Armstrong’s burden to iden‐
tify comparators to prove discriminatory effect on the merits.
We did so for the same reasons we rejected the need for com‐
parators at the discovery stage in Davis: “[T]he instant case
involves police conduct, not prosecutorial discretion, and is in
a civil, not criminal, context. This case is thus not like Arm‐
strong.” Chavez, 251 F.3d at 640.
2. McCleskey Did Not Set the Standard of Proof
The government argues that McCleskey v. Kemp, 481 U.S.
279 (1987), provides an independent ground for applying a
clear‐evidence standard, one that relies on a distinction be‐
tween criminal and civil cases rather than a comparison be‐
tween police and prosecutors. This argument comes from a
sweeping line of dicta in McCleskey: “Because discretion is es‐
sential to the criminal justice process, we would demand ex‐
ceptionally clear proof before we would infer that the discretion
No. 20‐2439 17
has been abused.” 481 U.S. at 297 (emphasis added). The gov‐
ernment seems to argue that this language in McCleskey im‐
poses a clear‐and‐convincing evidence standard on any claim
that “challenges decisions at the heart of the State’s criminal
justice system.” Id.It appears the district court agreed with this broad reading
of McCleskey. Judge Coleman denied Conley’s selective‐en‐
forcement claim based on Judge Castillo’s reasoning in Brown.
And in Brown, Judge Castillo wrote that McCleskey “unfortu‐
nately” required him to extend Armstrong’s clear‐evidence
standard to selective‐enforcement claims. Brown, 299 F. Supp.
3d at 996 & n.14. But McCleskey did not hold broadly that any
constitutional claim challenging core criminal justice activity
is subject to a heightened standard of proof. This is immedi‐
ately apparent once one recognizes the many claims and de‐
fenses raised by accused defendants that are governed by a
preponderance standard even though they challenge activity
at the heart of the criminal process.
For instance, Batson claims for racially motivated peremp‐
tory challenges by prosecutors can be proven by a preponder‐
ance of the evidence. See Johnson v. California, 545 U.S. 162, 170
(2005) (at final step of Batson inquiry, judge must decide
“whether it was more likely than not that the challenge was
improperly motivated”). Likewise, when seeking to suppress
evidence (a remedy that often leads to dismissal), accused de‐
fendants need prove by only a preponderance of evidence
that police abused their investigative discretion in violation of
the Fourth Amendment. See Franks v. Delaware, 438 U.S. 154,
156 (1978) (if “perjury or reckless disregard is established by
the defendant by a preponderance of the evidence, and … the
affidavit’s remaining content is insufficient to establish
18 No. 20‐2439
probable cause, the search warrant must be voided and the
fruits of the search excluded….”). See also Part III‐A‐3‐c, be‐
low, noting other affirmative criminal defenses provable by a
preponderance of evidence.
Accordingly, rather than reading this passage in McCles‐
key as announcing a broad new standard that would have im‐
plicitly unsettled many doctrines, we understand McCleskey’s
call for “exceptionally clear proof” as part of the Court’s crit‐
icism of the petitioner’s insufficient statistics in that case.
McCleskey asked the Court to infer intentional racial discrim‐
ination in his specific case from a broad statistical study show‐
ing that Georgia juries more readily imposed the death pen‐
alty on Black defendants. The Court thus noted that “McCles‐
key’s statistical proffer must be viewed in the context of his
challenge. McCleskey challenges decisions at the heart of the
State’s criminal justice system” that “necessarily require[] dis‐
cretionary judgments.” 481 U.S. at 297. Given that discretion,
if McCleskey wished to rely solely on statewide statistics, they
would need to be “exceptionally clear” to infer that his spe‐
cific jury sentenced him to death because he was Black and
not because of the “legitimate and unchallenged explanation
for the decision”—that he committed a death‐eligible murder.
Id.In other words, McCleskey clarified that, even under a pre‐
ponderance standard, if “a legitimate and unchallenged ex‐
planation … is apparent from the record,” it takes “exception‐
ally clear proof” in the other direction to push the case across
the fifty‐yard line. Id. McCleskey does not independently re‐
quire a heightened standard of proof here simply because
Conley challenges decisions made in the criminal justice sys‐
tem.
No. 20‐2439 19
3. Support from Related Areas of Law
Situating this issue within the larger fabric of the law fur‐
ther confirms that a preponderance standard should govern
Conley’s selective‐enforcement claim. Specifically, the Su‐
preme Court’s distinction between absolute prosecutorial im‐
munity and qualified immunity, its aversion to clear‐evidence
standards in civil cases, and its use of preponderance stand‐
ards for other criminal defenses all bolster the case for apply‐
ing a preponderance standard here.
a. Civil Immunity for Prosecutors and Police
To begin, the same reasons for treating selective enforce‐
ment differently than selective prosecution also justify the dif‐
ferent levels of immunity that apply to prosecutors and police
in § 1983 and Bivens actions. Unlike police, prosecutors enjoy
absolute civil immunity from damages for unconstitutional
actions taken in their prosecutorial role. Imbler v. Pachtman,
424 U.S. 409, 427 (1976). The primary justification for prosecu‐
torial immunity is that it preserves core executive discretion,
just as in Armstrong: “A prosecutor is duty bound to exercise
his best judgment both in deciding which suits to bring and
in conducting them in court. The public trust of the prosecu‐
tor’s office would suffer if he were constrained in making
every decision by the consequences in terms of his own po‐
tential liability in a suit for damages.” Id. at 424–25. On the
other hand, when prosecutors conduct police‐like activity
outside their prosecutorial role, the rationale for absolute im‐
munity vanishes and they enjoy only qualified immunity, just
like police officers. Kalina v. Fletcher, 522 U.S. 118, 126 (1997)
(“When a prosecutor performs the investigative functions
normally performed by a detective or police officer, it is ‘nei‐
ther appropriate nor justifiable that, for the same act,
20 No. 20‐2439
immunity should protect the one and not the other.’”), quot‐
ing Hampton v. City of Chicago, 484 F.2d 602, 608 (7th Cir. 1973).
This reasoning is consistent with our view that Arm‐
strong’s clear‐evidence standard similarly should not extend
to selective‐enforcement claims, which challenge police inves‐
tigations, not prosecutorial decisions. An accused defendant
who secures release due to racially selective prosecution can‐
not then seek damages from the prosecutor. But a defendant
who proves selective enforcement can sue the police. And in
such a case, qualified immunity would likely not protect the
police because the plaintiff has already proven racial animus.
See Taylor v. Ways, 999 F.3d 478, 490 (7th Cir. 2021) (“Any rea‐
sonable official … would have known that intentional racial
discrimination … was unconstitutional.”). Since prosecutorial
immunity and Armstrong both turn on similar, prosecutor‐
specific reasons, the stark gap in civil relief available in selec‐
tive‐prosecution cases versus selective‐enforcement cases
supports a similar gap in the standards for proving such
claims in the criminal context.
b. Clear‐and‐Convincing Evidence Standard Disfa‐
vored in Civil Cases
Habeas corpus cases like this are civil cases. See, e.g., Ban‐
ister v. Davis, 140 S. Ct. 1698, 1705 (2020) (“The Federal Rules
of Civil Procedure generally govern habeas proceedings.”),
citing Fed. Rule Civ. Proc. 81(a)(4); Browder v. Director, Depʹt
of Corr. of Illinois, 434 U.S. 257, 270 (1978) (analogizing habeas
proceedings to “other civil proceedings”). And in civil cases,
the Supreme Court has developed a presumption in favor of
preserving preponderance standards in the face of attempts
to impose “clear‐and‐convincing evidence” standards. See
Ramirez v. T&H Lemont, Inc., 845 F.3d 772, 777 (7th Cir. 2016)
No. 20‐2439 21
(describing “‘the presumption that the burden of proof in fed‐
eral civil cases is proof by a preponderance of the evidence,’
… a presumption reinforced by the Supreme Court’s repeated
rejection of more demanding evidentiary burdens in the civil
setting”), quoting Yi v. Sterling Collision Centers, Inc., 480 F.3d
505, 507 (7th Cir. 2007). In rejecting a “clear‐and‐convincing”
standard under Title VII, the Supreme Court has stressed that
one of the “[c]onventional rules of civil litigation … is that
parties … need only prove their case by a preponderance of
the evidence.” Price Waterhouse v. Hopkins, 490 U.S. 228, 253
(1989).
Due to this presumption, the “imposition of even severe
civil sanctions … has been permitted after proof by a prepon‐
derance of the evidence.” Herman & MacLean v. Huddleston,
459 U.S. 375, 389–90 (1983) (rejecting clear‐and‐convincing
standard for securities fraud); accord, Halo Electronics, Inc. v.
Pulse Electronics, Inc., 136 S. Ct. 1923, 1934 (2016) (same for en‐
hanced patent damages under 35 U.S.C. § 284); Octane Fitness,
LLC v. ICON Health & Fitness, Inc., 572 U.S. 545, 557–58 (2014)
(same for patent fee‐shifting under 35 U.S.C. § 285); Grogan v.
Garner, 498 U.S. 279, 286–87 (1991) (same for bankruptcy dis‐
charges); United States v. Regan, 232 U.S. 37, 48–49 (1914) (pre‐
ponderance standard suffices in civil suits for acts exposing
party to criminal prosecution); Ramirez, 845 F.3d at 776–80
(conduct supporting dismissal of civil case as sanction need
be proven by only a preponderance of evidence).
“Exceptions to this standard are uncommon, and in fact
are ordinarily recognized only when the government seeks to
take unusual coercive action … against an individual.” Price
Waterhouse, 490 U.S. at 253, citing Santosky v. Kramer, 455 U.S.
745, 756 (1982) (clear and convincing evidence needed to
22 No. 20‐2439
terminate parental rights); Addington v. Texas, 441 U.S. 418, 427
(1979) (same for involuntary commitment); Woodby v. INS, 385
U.S. 276, 285 (1966) (same for deportation); Schneiderman v.
United States, 320 U.S. 118, 122, 125 (1943) (same for denatu‐
ralization).
Here we have the opposite situation. The government
seeks a clear‐evidence standard to protect its interest in con‐
tinuing severe coercive action (imprisonment) against an in‐
dividual. Viewed through that lens, granting a writ of habeas
corpus that tells a federal official to refrain from unconstitu‐
tionally imprisoning someone is not the sort of relief calling
for a heightened standard of proof for the benefit of the gov‐
ernment. If anything, the Court’s pattern of departing from
the preponderance presumption only to protect individuals
from serious government action points in the opposite direc‐
tion.2
The purpose of selecting a standard of proof is “to allocate
the risk of error between the litigants and to indicate the rela‐
tive importance attached to the ultimate decision.” Herman &
MacLean, 459 U.S. at 389, quoting Addington, 441 U.S. at 423.
Applying a clear‐evidence standard here would inappropri‐
ately “express[] a preference for one side’s interests” in a case
where equally powerful considerations weigh on both sides.
Id. at 390. For the government, concerns about finality and
separation of powers caution against disturbing a federal con‐
viction obtained after a full trial in the district court, on an
2 We address here the standard of proof in first federal motions under
28 U.S.C. § 2255, not the deferential standard of review that usually ap‐
plies when federal courts review state convictions under 28 U.S.C. § 2254,
nor the standard for successive § 2255 petitions under § 2255(h).
No. 20‐2439 23
indictment brought by executive law enforcement agencies.
But for the accused defendant, the preservation of individual
constitutional rights weighs equally in favor of substantial
post‐conviction review. Accordingly, the Supreme Court has
agreed at least in one context that a petitioner’s claim for ha‐
beas relief from a federal conviction can be proven by a pre‐
ponderance of the evidence. See Walker v. Johnston, 312 U.S.
275, 286 (1941) (petitioner alleging deprivation of constitu‐
tional right to counsel “would have the burden of sustaining
his allegations by a preponderance of evidence”). Absent the
additional prosecutorial concerns that motivated Armstrong’s
heightened standard, there is a “balance of interests” here that
“warrants use of the preponderance standard … generally ap‐
plicable in civil actions.” Herman & MacLean, 459 U.S. at 390.3
c. Preponderance Standards Govern Other Affirmative
Criminal Defenses
Even in criminal cases, many affirmative defenses are
provable by a preponderance of the evidence. See, e.g., Dixon
v. United States, 548 U.S. 1, 17 (2006) (duress defense); Cooper
v. Oklahoma, 517 U.S. 348, 355–56 (1996) (government cannot
prosecute if defendant shows he is “more likely than not” in‐
competent). Even the way the entrapment defense operates in
fake stash house cases like this one shows defendants need
3 Even if we thought that habeas corpus relief might call for a higher
standard of proof than a run‐of‐the‐mill § 1983 claim for civil damages,
that approach would produce an awkward tension under Heck v. Humph‐
rey, 512 U.S. 477 (1994). Under Heck, a higher standard of proof in criminal
appeals and habeas corpus cases would in practice mean a higher stand‐
ard for ordinary civil relief too because convicted defendants must have
their convictions invalidated before they can bring selective enforcement
claims for damages. See id. at 486–87.
24 No. 20‐2439
not present “clear” evidence to prove their defenses. Rather,
so long as the defendant puts forth enough evidence to show
that a reasonable jury could find he was entrapped, the judge
must submit the defense to the jury with an instruction that,
to convict, the jury must find beyond a reasonable doubt that
the defendant was not entrapped. Jacobson v. United States, 503
U.S. 540, 549 (1992); United States v. Pillado, 656 F.3d 754, 763
(7th Cir. 2011). So, given that accused defendants can secure
release through other affirmative defenses that use prepon‐
derance standards (or standards even more favorable to de‐
fendants), there should be no concern about applying a pre‐
ponderance standard to Conley’s selective enforcement de‐
fense. For these reasons, we hold that Conley could prove his
selective enforcement claim by a preponderance of the evi‐
dence.4
4 We have considered whether applying a preponderance standard
creates a circuit split requiring circulation under Circuit Rule 40(e), but
conclude it does not. To be sure, the Third, Fourth, Sixth, and Tenth Cir‐
cuits have cited Armstrong’s clear‐evidence standard when assessing se‐
lective‐enforcement claims. See Karns v. Shanahan, 879 F.3d 504, 521 (3d
Cir. 2018); United States v. Mason, 774 F.3d 824, 829–30 (4th Cir. 2014); Gar‐
denhire v. Schubert, 205 F.3d 303, 319 (6th Cir. 2000); United States v. Her‐
nandez‐Chaparro, 357 F. App’x 165, 166 (10th Cir. 2009). These cases, how‐
ever, invoked Armstrong in passing without specifically rejecting a pre‐
ponderance standard. Moreover, the rule in the Sixth and Tenth Circuits
remains unclear because other cases there have applied a preponderance
standard. See United States v. Avery, 137 F.3d 343, 355 (6th Cir. 1997) (“[A]
defendant would have to demonstrate by a ‘preponderance of the evi‐
dence’ that a police officer decided to approach [or pursue] him or her
solely because of his or her race.”), quoting United States v. Travis, 62 F.3d
170, 174 (6th Cir. 1995); York v. Secʹy of Treasury, 774 F.2d 417, 422 (10th Cir.
1985); Barton v. Malley, 626 F.2d 151, 155 (10th Cir. 1980). Regardless, we
also view our decision here as simply reaffirming the logic of Davis, where
No. 20‐2439 25
B. Application
In granting a certificate of appealability on Conley’s selec‐
tive‐enforcement claim, the district court wrote that his claim
may have “been resolved in a different manner” under a pre‐
ponderance standard. See Conley, 2020 WL 4226676, at *6. We
respectfully disagree, however, and find it unnecessary to re‐
mand for new analysis under the preponderance standard.
The parties have briefed the issue, and Conley’s key evi‐
dence—the Fagan Report—has already been thoroughly dis‐
sected in Brown. It is clear on this record that particular as‐
pects of Conley’s individual case prevent him from relying on
the Fagan Report to prove, even by a preponderance of the
evidence, that the ATF targeted him based on race.5
As a general matter, statistics can be “a useful tool” that
can establish discriminatory effect and provide powerful evi‐
dence of discriminatory intent if race can be isolated from
other confounding variables. E.g., United States v. Barlow, 310
F.3d 1007, 1011 (7th Cir. 2002). While few judges are statisti‐
cians, we are tasked with determining whether discrimina‐
tion was afoot, and statistics can be important in that determi‐
nation. Judges must therefore pay close attention to what par‐
ties’ statistics can and cannot prove. When program‐wide sta‐
tistics are the plaintiff’s sole source of evidence, it can be
we have already held en banc that Armstrong does not apply in the selec‐
tive enforcement context. See 793 F.3d at 720–22.
5 Beyond the Fagan Report, Conley’s only other evidence of discrimi‐
nation is comments made by two ATF agents who were not involved in
Conley’s sting. These comments made by other agents in other stings are
of little relevance here. They were thoroughly analyzed in Brown and
found to be insufficient evidence of discriminatory intent. See 299 F. Supp.
3d at 1014–18. We agree with Brown’s analysis of these comments.
26 No. 20‐2439
difficult to infer discrimination in a particular case—espe‐
cially where there is another legitimate explanation for the
government’s conduct. See McCleskey, 481 U.S. at 293 n.12
(only in “rare cases [has] a statistical pattern of discriminatory
impact demonstrated a constitutional violation”). Still, we
must be willing to follow statistics where they fairly lead
without fear that they could require “too much justice.”
McCleskey, 481 U.S. at 339 (Brennan, J., dissenting).
Conley’s statistics cannot pass muster, even under a pre‐
ponderance standard. In 2014, Professor Fagan was ap‐
pointed by the court in Brown to conduct a statistical analysis
of whether the federal government’s fake stash house pro‐
gram in Chicago was racially selective. 299 F. Supp. 3d at 992.
In 2016, Professor Fagan issued a comprehensive report ex‐
plaining his findings. The Fagan Report concluded that, even
“after controlling for criminal propensity, race remains statis‐
tically significant, meaning that the ATF is selecting defend‐
ants on the basis of race.” Id. at 1003. Judge Castillo, however,
found that there were “fundamental concerns about the relia‐
bility of Professor Faganʹs findings” that prevented a showing
of discriminatory effect or intent. Id. at 1013.
At least with respect to Conley’s case, we agree. No Chi‐
cago stash house case is more susceptible to the limits of the
Fagan Report. Two critical limits in Professor Fagan’s analysis
apply here. His statistical finding of discrimination depends
on the assumptions that (1) law enforcement officers (not co‐
conspirators), and (2) the ATF Chicago office (not other agen‐
cies) independently selected each stash house defendant.
Conley was not selected by law enforcement at all, let alone
by the ATF. He was brought in by other conspirators who
were themselves chosen by the initial target, Myreon Flowers.
No. 20‐2439 27
And Chicago ATF did not even select Myreon Flowers; the FBI
did. These gaps undermine Conley’s ability to rely on the Fa‐
gan Report. As Judge Castillo explained in Brown:
Critical to Professor Fagan’s entire analysis [is]
an assumption that the ATF selected each of the
94 stash house defendants for participation in
the stings….
As a factual matter, the bulk of the 94 defendants
were not recruited by ATF. Instead, the record
reflects that the stash house investigations be‐
gan with the ATF’s selection of one or more ini‐
tial targets…. The remaining defendants were
recruited by the targets themselves, or by other
members of the conspiracy who were, in turn,
recruited by the targets. In fact, the record re‐
flects that the ATF never even met 32 of the 94
defendants—nearly one‐third of the total de‐
fendant group—until the day of their arrest. Ad‐
ditionally, two of the stash house cases, involv‐
ing a total of 14 defendants, were not initiated
by the ATF at all. Instead, those investigations
were initiated by the Federal Bureau of Investi‐
gation (“FBI”) based on information provided
by its own confidential informants. The record
reflects that the ATF did not become involved in
these two cases until after the selection of the in‐
itial targets had already been made. In the
Court’s view, ATF cannot be said to have “se‐
lected” individuals it never met or individuals
targeted by another governmental agency un‐
der any ordinary meaning of that term.
28 No. 20‐2439
299 F. Supp. 3d at 1003–04 (footnotes and citations omitted).
Conley is one of the thirty‐two defendants who were se‐
lected by co‐conspirators and whose race law enforcement
did not know until his arrest. And the larger Flowers sting of
which Conley was a part was one of the two initiated by the
FBI, not the ATF. Moreover, the FBI individually targeted
Myreon Flowers based on his current armed robbery activity,
not based on the ATF criteria that Professor Fagan used to
construct the comparison group for his analysis. The Fagan
Report’s inference of discrimination, whatever strength it
might have in other cases, is therefore based on two assump‐
tions that simply do not apply to Conley’s case. Professor Fa‐
gan himself acknowledged that correcting for those assump‐
tions would reduce the data to “such a small sample” that it
would be nearly “impossible to conduct a reliable statistical
analysis” that could support an inference of racial discrimina‐
tion. 299 F. Supp. 3d at 1008.
These two inapplicable assumptions undermine the pro‐
bative value of the Fagan Report’s conclusions as applied to
Conley’s case. We thus do not need to review arguments over
other problems with the report that Judge Castillo discussed
in Brown, such as the use of predominantly white LaSalle
County in the comparison pool. See id. at 1009–13. Absent ad‐
ditional evidence showing that the FBI would not have re‐
ferred (or the ATF would not have then pursued) a white tar‐
get situated similarly to Flowers, we could not conclude that,
more likely than not, the government targeted Flowers based
on his race. And even if we assume that Flowers was racially
targeted, to extend the reasoning to Conley we would need to
find further that the ATF affirmatively hoped Flowers would
recruit other Black people as co‐conspirators. When agents
No. 20‐2439 29
targeted Flowers, they were surely aware that, given the con‐
tinued patterns of social and neighborhood segregation in
Chicago, his chosen co‐conspirators would likely include
other Black people. See, e.g., Matthew Bloch, Amanda Cox &
Tom Giratikanon, Mapping Segregation, N.Y. Times (July 8,
2015), https://www.nytimes.com/interactive/ 2015/07/08/us/
census‐race‐map.html (click on Chicago). Yet the “awareness”
of that likely consequence does not prove purposeful discrim‐
ination, Feeney, 442 U.S. at 279, and Conley lacks any other
evidence indicating that the ATF affirmatively intended Flow‐
ers to recruit Black co‐conspirators like Conley. These prob‐
lems “cannot be brushed aside simply because of an underly‐
ing discomfort with” fake stash house stings. 299 F. Supp. 3d
at 1013.
In deciding this appeal, we have not forgotten the long his‐
tory of race in America and in the Chicago region and the
many aspects of that history that contribute to the fact that
“between 2006 and 2013, the defendants prosecuted in the
ATF false stash house cases in this District were 78.7 percent
black, 9.6 percent Hispanic, and 11.7 percent white.” Id. at
1001. But the Equal Protection Clause and its federal Fifth
Amendment corollary provide legal relief only when a party
can prove purposeful discrimination against him by the gov‐
ernment in his case. Our inquiry here is limited to whether
Conley has offered evidence that would support a finding
that it is more likely than not that federal law enforcement
agencies intentionally targeted him for this sting because he
is Black. It does not, despite the district court’s suggestion to
the contrary. So we must affirm dismissal of his selective‐en‐
forcement claim.
30 No. 20‐2439
IV. Outrageous Government Conduct
Conley’s second claim for relief under § 2255 is that ATF’s
fake stash house sting was so outrageous that it violated his
due process rights. This argument, known as the “outrageous
government conduct” defense, stems from an entrapment
case where the Supreme Court suggested that “we may some
day be presented with a situation in which the conduct of law
enforcement agents is so outrageous that due process princi‐
ples would absolutely bar the government from invoking ju‐
dicial processes to obtain a conviction.” United States v. Rus‐
sell, 411 U.S. 423, 431–32 (1973). But the Supreme Court has
never reversed criminal charges due to outrageous govern‐
ment conduct, and neither have we.6
This defense fails for three reasons. First, we have repeat‐
edly rejected the existence of an outrageous conduct defense.
See United States v. Westmoreland, 712 F.3d 1066, 1071 (7th Cir.
2013) (“[O]ur court has disallowed such a defense.”); United
States v. Stallworth, 656 F.3d 721, 730 (7th Cir. 2011) (“Outra‐
geous government conduct is not a defense in this circuit.”);
United States v. White, 519 F.3d 342, 346 (7th Cir. 2008) (“[T]his
circuit clearly and consistently has refused to recognize any
defense based on … ‘outrageous government conduct.’”);
6Some defendants invoke this due process argument when they can‐
not satisfy the elements of a traditional entrapment defense. For instance,
in Russell, entrapment was unavailable because the defendant conceded
that he may have harbored a predisposition to commit the offense. 441
U.S. at 433. Similarly, in Conley’s direct appeal, we rejected his ability to
argue entrapment. Entrapment “applies only when a government actor
recruits a defendant,” yet Conley was recruited by private co‐conspirators
who themselves were recruited by Myreon Flowers, not police. Conley, 875
F.3d at 402.
No. 20‐2439 31
Alexander v. DeAngelo, 329 F.3d 912, 916 (7th Cir. 2003) (“We
flatly rejected the doctrine in United States v. Boyd, 55 F.3d 239,
241–42 (7th Cir. 1995).”).
Second, even if the defense were available in theory, Con‐
ley procedurally defaulted it. “A claim cannot be raised for the
first time in a § 2255 motion if it could have been raised at trial
or on direct appeal,” absent a showing of cause and prejudice,
or actual innocence. McCoy v. United States, 815 F.3d 292, 295
(7th Cir. 2016), citing Sandoval v. United States, 574 F.3d 847,
850 (7th Cir. 2009). Conley did not raise the defense until this
§ 2255 motion, and the government argues that he has not
shown cause for excusing the default. Conley, however, seeks
to excuse his default on the grounds that (1) the basis for his
claim did not exist when his direct appeal was pending, and
(2) he tried to raise the argument but his lawyer decided
against it.
As to the first argument, “a showing that the factual or le‐
gal basis for a claim was not reasonably available to counsel
… would constitute cause” for excusing procedural default.
Murray v. Carrier, 477 U.S. 478, 488 (1986), citing Reed v. Ross,
468 U.S. 1, 16 (1984). But Conley’s outrageous conduct defense
is not a newly available legal theory. Even in the face of our
“uniform wall of precedent” rejecting the defense, Conley still
needed to raise it to preserve it for potential reconsideration
here. See, e.g., Greer v. United States, 141 S. Ct. 2090, 2099 (2021)
(unpreserved claim is subject to plain‐error review under Fed‐
eral Rule of Criminal Procedure 52 even if precedent fore‐
closed the claim at the time the defendant could have ob‐
jected). Perhaps recognizing this problem, Conley instead
says that the factual basis for his claim was unavailable on di‐
rect appeal because the Fagan Report had not yet been
32 No. 20‐2439
published. We understand Conley’s outrageousness argu‐
ment, however, to rest primarily on the nature of fake stash
house stings as police‐created traps, not on the Fagan Report’s
(disputed) finding that they were racially motivated.
We also reject Conley’s second excuse for defaulting the
claim—that he tried to raise it on direct appeal but was
thwarted by his attorney’s decision to advance other argu‐
ments instead. Attorneys have wide discretion to choose is‐
sues for appeal. Not raising an outrageous conduct defense
made strategic sense given our repeated rejections. An outra‐
geous conduct argument would certainly not have been
“clearly stronger” than the arguments his lawyer raised. Cf.
Shaw v. Wilson, 721 F.3d 908, 915 (7th Cir. 2013) (applying in‐
effective‐assistance‐of‐counsel standard to appellate counsel).
Conley says we nonetheless have excused default in a case
like his where the defendant did all he could to try to raise an
issue. But that case is clearly distinguishable. In Gregory‐Bey
v. Hanks, 332 F.3d 1036, 1044 (7th Cir. 2003), we excused pro‐
cedural default where the petitioner had tried to raise the is‐
sue with his attorneys and courts for fifteen years, and his re‐
quests had been completely ignored—not rejected for a good
strategic reason.
Third, even if we recognized the defense and could reach
its merits, the sting here was not so outrageous as to violate
due process. In the very rare cases where other courts have
reversed convictions due to outrageous government conduct,
law enforcement agencies were much more active in coaxing
the defendants to commit their crimes. For example, in Con‐
ley’s best supporting case, United States v. Twigg, 588 F.2d 373(3d Cir. 1978), the Third Circuit found outrageous conduct
where the DEA “involve[d] itself so directly and continuously
No. 20‐2439 33
over such a long period of time in the creation and mainte‐
nance of criminal operations.” Id. at 379. The DEA not only
purchased most of the supplies for a methamphetamine lab,
but the government informant provided all the expertise and
led the actual process of producing meth in the lab. Id. at 380–
81. All of Twigg’s actions were at the specific direction of the
government informant; Twigg did not provide any expertise,
supplies, or ideas himself. Moreover, the DEA even supplied
the location for the lab after the defendants were having diffi‐
culty finding a place. The agents were thus actively solving
problems and holding the defendants’ hands at every step.
Here, by contrast, the ATF proposed a fake crime and
stood back and watched as Myreon Flowers quickly gathered
a crew, developed a plan, and showed up ready to execute an
armed robbery. Like Twigg, Conley was recruited by another
defendant. But unlike Twigg, Conley was an active partici‐
pant who contributed to the planning process and agreed to
a frontline (i.e., armed) role. Conley was not coaxed along by
government agents at every turn. His case simply does not
exhibit the sort of constant police nudging that outraged the
court in Twigg.
Rather, Conley’s sting closely resembles the ATF sting we
upheld in United States v. Smith, 792 F.3d 760 (7th Cir. 2015),
in which an undercover ATF agent arranged for the defendant
to provide armed security for the transport of cocaine. On ap‐
peal, the defendant argued that the sting constituted outra‐
geous government conduct. We disagreed: “when an individ‐
ual is ready and willing to engage in illegal activity, the fact
that the Government affords him an opportunity to commit
the crime provides no legal impediment to prosecution.” Id.
at 766. That is all the ATF did in Conley’s case. It presented a
34 No. 20‐2439
criminal opportunity, which Conley freely accepted when
other conspirators asked him to join the team.
Even assuming some police tactics could be so outrageous
as to violate due process, such a defense could not simply
“give the federal judiciary a ‘chancellor’s foot’ veto over law
enforcement practices of which it d[oes] not approve. The ex‐
ecution of the federal laws under our Constitution is confided
primarily to the Executive Branch….” Russell, 411 U.S. at 435.
The ATF’s conduct in Conley’s case stopped short of “violat‐
ing that ‘fundamental fairness, shocking to the universal
sense of justice,’ mandated by the Due Process Clause of the
Fifth Amendment.” Id. at 432, quoting Kinsella v. United States
ex rel. Singleton, 361 U.S. 234, 246 (1960).
Because Conley has failed to prove that the ATF violated
his Fifth Amendment equal protection or due process rights,
the district court’s denial of relief under 28 U.S.C. § 2255 is
AFFIRMED.