NATIONAL POLICE ACCOUNTABILITY PROJECT; JULIAN; AMERICAN CIVIL LIBERTIES UNION FOUNDATION; AMERICAN CIVIL LIBERTIES UNION FOUNDDATION OF VIRGINIA; RICHMOND TRANSPARENCY AND ACCOUNTABILITY PROJECT,

Amici Supporting Appellee.

Appeal from the United States District Court for the Eastern District of Virginia, at Richmond. John A. Gibney, Jr., Senior District Judge. (3:21-cr-00042-JAG-1)

Argued: March 14, 2025 Decided: August 1, 2025

Before NIEMEYER, HARRIS, and BERNER, Circuit Judges.

Reversed and remanded by published opinion. Judge Niemeyer wrote the opinion, in which Judge Harris and Judge Berner joined.

Page 2ARGUED: Jacqueline Romy Bechara, OFFICE OF THE UNITED STATES ATTORNEY, Alexandria, Virginia, for Appellant. Rohiniyurie Tashima, GOODWIN PROCTER, LLP, Washington, D.C., for Appellee. ON BRIEF: Jessica D. Aber, United States Attorney, Shea M. Gibbons, Assistant United States Attorney, Erik S. Siebert, Assistant United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Richmond, Virginia, for Appellant. Patrick L. Bryant, Alexandria, Virginia, Laura J. Koenig, OFFICE OF THE FEDERAL PUBLIC DEFENDER, Richmond, Virginia; Brian T. Burgess, Kaley Preciado, GOODWIN PROCTER LLP, Washington, D.C., for Appellee. Lauren Bonds, Keisha James Eliana Machefsky, Devontae W. Torriente, NATIONAL POLICE ACCOUNTABILITY PROJECT, Kansas City, Kansas; Jill Collen Jefferson, JULIAN, Hattiesburg, Mississippi, for Amici National Police Accountability Project and JULIAN. Geri Greenspan, Eden B. Heilman, AMERICAN CIVIL LIBERTIES UNION FOUNDATION OF VIRGINIA, Richmond, Virginia; Jenn Rolnick Borchetta, Brandon Buskey, AMERICAN CIVIL LIBERTIES UNION FOUNDATION, New York, New York, for Amici American Civil Liberties Union Foundation and American Civil Liberties Union Foundation of Virginia. Maisie Osteen, LEGAL AID JUSTICE CENTER, Richmond, Virginia, for Amicus Richmond Transparency and Accountability Project.

Page 3NIEMEYER, Circuit Judge:

The district court dismissed Keith Moore’s indictment for the illegal possession of a firearm because, as it found, the traffic stop leading to Moore’s arrest was made because Moore was Black. The government appeals.

When Richmond, Virginia, police officers observed a vehicle with the temporary license tag number 11134Y, they concluded that it was a fake tag because, within the preceding hours, they had stopped two other vehicles with the identical temporary tag number. The officers activated the lights on their patrol car to institute a stop, but the driver, Keith Moore, failed to stop and instead fled, running several stop signs before crashing into a curb. Moore then exited the vehicle, leaving the engine running and the door open, and ran on foot. After the officers caught and detained him, they observed a gun on the floorboard of Moore’s vehicle. Following a records check that revealed that Moore had previously been convicted of a felony, they arrested him for the illegal possession of a firearm. A federal grand jury subsequently indicted Moore for violation of 18 U.S.C. § 922(g)(1).

Moore filed a motion to suppress, contending that the traffic stop had violated his Fourth Amendment rights and that the officers interrogated him in violation of his Miranda rights. During a hearing on the motion, Moore presented evidence that all traffic stops made that day by the officers who arrested him were made in predominantly Black neighborhoods, suggesting that the police had targeted him because he was Black, in violation of the Equal Protection Clause of the Fourteenth Amendment.

Page 4The court authorized further investigation and briefing, and after conducting several hearings and receiving testimony from both parties, including expert testimony, it dismissed Moore's indictment on the ground that it was the product of selective law enforcement against Black drivers. While the court found that the police officers had probable cause to stop Moore and that there was no evidence of their “invidious or bad faith,” it nonetheless relied on statistical evidence that Black drivers were 5.13 times more likely to be stopped in Richmond than White drivers, viewed in the context of Richmond’s history of racial discrimination, to conclude that Moore had “successfully shown both the discriminatory effect and discriminatory purpose elements required for [a] selective enforcement claim.”

After a thorough examination of the record, we conclude that the evidence was insufficient to show that Moore’s stop and arrest were the product of racially discriminatory purpose, in violation of the Equal Protection Clause. Accordingly, we reverse and remand with instructions to reinstate the indictment.

While on patrol on December 5, 2020, in the Fourth Precinct of Richmond, Virginia, four officers with the Richmond Police Department encountered three different cars displaying the same temporary tag number, 11134Y. The officers were members of the Fourth Precinct’s Focus Mission Team, a unit whose primary goal was getting guns and drugs off the streets of Richmond.

Page 5First, at approximately 5:45 p.m., the officers stopped a white Cadillac sedan because its headlights were off. During the stop, the officers asked the driver whether he had any weapons or drugs in the car, shined flashlights throughout the car, and patted down the front-seat passenger after he had exited the car. When the officers entered the vehicle’s temporary tag number, 11134Y, into a database, the database failed to return a result, indicating that the tag was fake. The driver then volunteered to park the car rather than continuing to drive it, but the officers let him go on his way with a warning.

About an hour later, the same officers stopped a second car displaying a temporary tag with the same number, 11134Y. One of the officers asked the driver about her tag and informed her that another car had been displaying a license tag with the same number. The officers nonetheless believed her explanation and also let her go on her way.

Yet a few hours later, at around 9:45 p.m., the officers observed a white Chrysler 300 pull out from a gas station and abruptly stop when the driver saw the officers’ vehicle. One of the officers observed that the car had a temporary license tag, again with the same number, 11134Y, that they had now seen twice before on other vehicles. The officers activated the lights on the patrol car to initiate a stop, but this driver failed to stop. Instead, he fled, running several stop signs before crashing into a curb. The driver then exited his car, leaving it running and the door open, and ran on foot. The officers chased him down and detained him. When they observed a gun on the floorboard of the car and determined that the driver, Keith Moore, had previously been convicted of a felony, they arrested him.

Page 6A federal grand jury returned an indictment charging Moore in one count with possession of a firearm after having been convicted of a felony, in violation of 18 U.S.C. § 922(g)(1).

Moore filed a motion to suppress, arguing that the traffic stop violated his Fourth Amendment rights and that the officers impermissibly interrogated him before and after providing him with Miranda warnings. The suppression motion also noted that the neighborhood where Moore was arrested had “almost exclusively black residents” and claimed that the officers with the Fourth Precinct’s Focus Mission Team had been “looking for traffic infractions that night in hopes that the traffic stops” would reveal evidence of more serious crimes.

The district court held an evidentiary hearing on Moore’s motion in July 2021, and the government presented testimony from the four officers involved in the stop, as well as two others who had assisted their pursuit of Moore. Moore also presented evidence that the Focus Mission Team had conducted traffic stops on the day of his arrest exclusively in a predominantly Black neighborhood and that Black drivers were, on average, pulled over more frequently than White drivers. At the conclusion of the hearing, Moore requested an opportunity to submit supplemental briefing, and the district court granted the request.

In his supplemental brief, Moore argued for the first time that the indictment should be dismissed because the police officers had violated his Fourteenth Amendment right to equal protection by targeting him and other Black residents of the Fourth Precinct, using pretextual traffic stops to search their cars. While Moore acknowledged that he had to Page 7show, in order to state a selective-enforcement claim under the Equal Protection Clause, that the officers’ action had a discriminatory effect and was motivated by a discriminatory purpose (quoting United States v. Mason , 774 F.3d 824, 829 (4th Cir. 2014)), he contended that he had satisfied both prongs. He argued first that the vast disparity in the rates at which the Richmond Police Department stopped Black drivers as opposed to White drivers constituted evidence of a 'discriminatory effect on black drivers in Richmond,' and second, that “[t]he stark numbers and heavy-handed policing . . . of minority neighborhoods in Richmond” justified “an inference of discriminatory intent.” The government responded to the supplemental brief, arguing that Moore’s selective-enforcement claim was untimely and that, in any event, his evidence fell “far short of [the] high standard for proving racially animated selective law enforcement.”

At a hearing following this briefing, the court suggested that Moore’s motion should be expanded and that “a more fulsome examination of the statistical data” would be helpful. It directed Moore to consult with an expert.

Moore thereafter retained Dr. Eli Coston, an assistant professor in the Department of Gender, Sexuality, and Women’s Studies at Virginia Commonwealth University. Dr. Coston prepared a report analyzing traffic stops that the Richmond Police Department conducted from July 1, 2020, through December 6, 2020. Those data, which had been collected pursuant to the Virginia Community Policing Act, indicated that of the 2,500 relevant traffic stops conducted in Richmond, 77% involved Black drivers, while 14.16% involved White drivers. Based on those numbers, Dr. Coston concluded that “Black drivers were 5.13 times more likely to be stopped” than White drivers. The data, according to Dr.

Page 8Coston, also revealed a “somewhat weak” but still “important” relationship between race and the likelihood of being pulled over, as well as the eventual outcome of a given stop. Dr. Coston noted, however, that the report could not determine whether any 'specific traffic stop was the result of racial bias or racial profiling.”

Dr. Coston also prepared a set of maps that examined the racial demographics of the areas of Richmond where traffic stops were most frequently conducted. Based on the maps, Dr. Coston stated that the stops of Black drivers were more likely to be clustered near one another, while the stops of White drivers were “distributed throughout the city and less likely to show as clusters.” In addition, the few clusters of White-driver stops that existed occurred in predominantly White neighborhoods, whereas the clusters of Blackdriver stops were 'distributed throughout the city' and not limited to majority-Black neighborhoods.

In addition to Dr. Coston, Moore also retained a second expert, Dr. Marvin Chiles, an Assistant Professor of African American History at Old Dominion University. Dr. Chiles prepared a report in which he detailed the history of racial segregation in Richmond, focusing on the period from 1865 through the 1980s. He noted that although '[a] combination of federal law and local activism [had] helped curb systemic residential segregation between the 1960s and 1980s,' the stain of Richmond's segregated past “remains etched into the city’s landscape,” with most “working-class and underclass blacks [still living] in racially homogenous enclaves in Richmond.' Dr. Chiles also described how the Richmond Police Department had targeted and overpoliced Black citizens over the years. He noted, for example, that an effort to revitalize downtown Richmond in thePage 9 1980s led to the development of a 'new police strategy, as dictated by black political leaders,” of increased “policing of black neighborhoods.” Dr. Chiles’s report contained no relevant information specific to the Richmond Police Department that post-dated 1989.

The government retained its own expert, Dr. Michael Smith, a professor of criminology and criminal justice at the University of Texas at San Antonio, who challenged Dr. Coston's statistical analysis. In his report, Dr. Smith stated that Dr. Coston's conclusions were 'unsupported by the data and analysis' and relied 'upon elementary statistical techniques that no experienced social scientist familiar with racial disparity research in policing would rely on to draw inferences of bias.' He explained that “examining the raw percentages of drivers stopped by race is meaningless unless, at a minimum, those percentages can be compared against an estimate of the population of drivers in the jurisdiction of interest' and that, in addition, 'the population of persons who live in an area often serves as a poor representation of persons who drive in an area or who are at risk for being stopped by the police.” Dr. Smith also critiqued Dr. Coston for using statistical methods that could not control for confounding variables like poverty or residence in a high-crime area, noting that '[p]olice typically deploy more officers to neighborhoods with higher crime rates” and that such “police deployment variables often correlate with race and ethnicity in America’s urban neighborhoods.”

The district court held an evidentiary hearing on Moore's selective-enforcement claim on July 18 and 19, 2022, and again on October 28, 2022, to admit additional evidence. While on the stand, Dr. Coston testified that 'when we look at large scale patterns what we can say is whether or not Black people are disproportionately stopped at Page 10higher rates [and] disproportionately experience adverse outcomes when they are stopped.” But Dr. Coston clarified, 'I am not making conclusions about [the] causation' of the disparity and certainly not opining that 'Mr. Moore was stopped because he was an African-American.”

After conducting these numerous hearings, the district court issued two memorandum opinions. In the first, dated November 13, 2023, the court granted in part and denied in part Moore's motion to suppress. The court denied Moore's Fourth Amendment claim, noting that “Moore’s temporary tags had the same number as those of two other drivers the officers had stopped that night” and explaining that “[b]ecause the officers suspected Moore had violated a traffic law,” they were legally justified in stopping Moore's vehicle. With respect to Moore's Miranda claim, the court held that some statements should be suppressed but that others could be admitted.

But in the second opinion, dated February 12, 2024, the district court granted Moore’s motion to dismiss the indictment for selective enforcement in violation of the Equal Protection Clause, concluding that Moore had adequately supported his contention that Richmond Police Department 'officers selectively stop Black people, and that this selective enforcement led to his current charges.” The court began by denying motions filed by the government to exclude the testimony of Dr. Coston and Dr. Chiles. It then turned to the merits, stating: “Moore must prove, by a preponderance of the evidence,” that the Richmond Police Department’s “enforcement process ‘had a discriminatory effect and Page 11that it was motivated by a discriminatory purpose.’” (Quoting Cent. Radio Co. v. City of Norfolk , 811 F.3d 625, 634-35 (4th Cir. 2016)).

The court found that the testimony of Moore’s expert witnesses established that the Richmond Police Department’s traffic stops produce a discriminatory effect , stating that “[t]he statistics provided in this case make abundantly clear the disparate impact of traffic stops on Black drivers in Richmond.” And the court determined that the same evidence sufficed to show discriminatory purpose as well. It acknowledged that 'Moore ha[d] presented no evidence of the four officers' invidious or bad faith,' observing that the officers had, rather, “pulled Moore over for having suspicious temporary tags.” But the court concluded that it could still infer discriminatory purpose from '(1) evidence of a consistent pattern of actions by the Richmond Police Department that disparately impacts Black drivers in Richmond; and (2) the history of discrimination by the Richmond Police Department and in Richmond itself.” (Cleaned up). The court emphasized that “Moore’s statistics evince that [Richmond Police Department] police officers stop Black drivers at a rate that far exceeds the rate at which they stop white drivers.” And it observed that Moore had supported his claim with evidence of 'Richmond's racially segregated and discriminatory history.” This evidence included the fact that three of Richmond’s four police precincts “overlap Black neighborhoods,” while a fourth “aligns exactly with the boundaries of the white section of Richmond.”

In sum, the court found that '[t]he data showing that [the Richmond Police Department] stops Black drivers five times as often as it stops white drivers, coupled with the evidence of Richmond’s history of discrimination, reveals that [the Richmond Police Page 12Department’s] discriminatory purpose contributed to its officers’ decision to stop Moore.” Thus, it held that Moore had established that the Richmond police had selectively enforced traffic laws against Black drivers, including Moore himself, in violation of the Equal Protection Clause. On that basis, it dismissed the indictment against him.

The government filed a motion for reconsideration, arguing that the court had improperly inferred that the Richmond Police Department “had orchestrated an intentionally discriminatory campaign to stop more Black drivers” and that “the practical effect of [its] opinion may well be to discourage [the Richmond Police Department] . . . from actively policing the most violent areas of our City.” The district court denied the government’s motion, explaining that it had “dismissed this case because the evidence showed that [Richmond Police Department] officers disproportionately pull over Black drivers” and declaring that “[t]he time has come for this practice to end.”

From the district court’s order dismissing the indictment dated February 12, 2024, and order denying the government’s motion for reconsideration dated March 11, 2024, the government filed this appeal.

The sole issue before us is whether Moore provided sufficient evidence to show that his traffic stop and arrest on December 5, 2020, were the result of racially selective law enforcement, in violation of the Equal Protection Clause of the Fourteenth Amendment.

The Equal Protection Clause “prohibits selective enforcement of the law based on considerations such as race.” Whren v. United States , 517 U.S. 806, 813 (1996). In Whren , Page 13the Supreme Court acknowledged that a constitutional violation would occur were “police officers [to] decide which motorists to stop based on decidedly impermissible factors, such as the race of the car’s occupants.” Id. at 810. It made clear that 'the constitutional basis for objecting to intentionally discriminatory application of laws is the Equal Protection Clause, not the Fourth Amendment.” Id. at 813. It is now well established that to prevail on such a claim, a claimant must prove both (1) that the officers' enforcement had a discriminatory effect and (2) that it was motivated by discriminatory purpose . See United States v. Mason , 774 F.3d 824, 829 (4th Cir. 2014) (quoting United States v. Armstrong , 517 U.S. 456, 465 (1996)); United States v. Hare , 820 F.3d 93, 98–99 (4th Cir. 2016).

We have held selective-enforcement claims to a demanding standard of proof because such claims “ask a court to exercise judicial power over a special province of the Executive.” Mason , 774 F.3d at 829-30 (cleaned up). Inasmuch as such a claim poses 'the great danger of unnecessarily impairing the performance of a core executive constitutional function,” the party bringing the claim “must demonstrate clear evidence of racially animated selective law enforcement. This standard is intended to be a demanding and rigorous one.” Id. at 830 (emphasis added) (cleaned up); see also Hare , 820 F.3d at 98–99.

In sum, Moore must carry the burden of showing, by clear evidence, that his traffic stop had a discriminatory effect and was motivated by a discriminatory purpose or intent.

The government contends that Moore failed to carry his burden as to both prongs. It focuses mainly, however, on Moore’s failure to show discriminatory purpose in making the stop. Noting the lack of evidence that race played any role in Moore’s traffic stop, the Page 14government argues that the statistical evidence that Moore presented was insufficient to overcome that evidentiary gap. It claims that the statistical evidence lacked an appropriate benchmark for comparison between Black and White drivers and that Moore’s historical evidence left off in 1989, making it irrelevant to a stop conducted in 2020. The government reasons that to conclude that the statistical evidence presented in this case was sufficient to overcome the lack of evidence of discriminatory intent “would mean that any Black driver stopped by [a Richmond Police Department] officer could prove discriminatory purpose based on the exact same statistical and historical evidence, regardless of the actual motives of the officer who stopped his car.” The consequence would be to jeopardize the ability of Richmond Police Department officers to enforce traffic laws against Black drivers in Richmond.

Moore defends the district court’s decision, pointing to the substantial “circumstantial evidence” that he was stopped pretextually and “on the basis of his race.” He points to the practices of the Fourth Precinct's Focus Mission Team in patrolling predominantly Black neighborhoods and in habitually employing pretextual stops as an excuse to ferret out more serious crimes, like gun and drug offenses. Additionally, he highlights the statistical evidence that showed that while Black and White residents each make up roughly one-half of Richmond’s population, Black drivers were the subject of 77% of the traffic stops and thus were “5.13 times more likely to be stopped” than White drivers. Finally, Moore contends that the history of discrimination in Richmond supplies important context through which to understand the statistical evidence.

Page 15We begin with the main issue that the government has presented, whether the evidence showed that the Richmond police officers who stopped Moore were motivated by a discriminatory purpose, i.e. , whether they had racial animus. This requires proof “that the decisionmaker . . . selected or reaffirmed a particular course of action at least in part 'because of,' not merely 'in spite of,' its adverse effects upon an identifiable group.' Pers. Adm’r of Mass. v. Feeney , 442 U.S. 256, 279 (1979); see also Hare , 820 F.3d at 100 n.6. Since the officers as individuals, and not the Richmond Police Department, made the decision to stop Moore, it is the officers’ motivation that must be examined. Indeed, Moore only challenges their decision to stop his car and not any police department policy or order. See McCleskey v. Kemp , 481 U.S. 279, 292-93 (1987) ('[T]o prevail under the Equal Protection Clause, [a party] must prove that the decisionmakers in his case acted with discriminatory purpose”).

The evidence shows that when the officers observed a vehicle bearing a temporary license tag with the number 11134Y, they had probable cause to believe that the tag was fake because they had stopped two other vehicles with the same temporary tag number only hours before. No one disputes that knowingly displaying a fake license tag violates the law. See Va. Code Ann. § 46.2-612(B)(1). And while the officers merely gave warnings to the first two drivers whom they had stopped, Moore’s circumstances were different. Moore did not stop when required to do so, but rather fled. Following a car chase, Moore crashed into a curb and abandoned his car, leaving his door open and engine running and thus enabling the officers to see a gun lying on the floorboard. No aspect of Page 16this sequence suggests that race played a role in the decision to initiate the traffic stop or to arrest Moore for the illegal possession of a gun.

In light of this evidence, the district court appropriately found that the officers had probable cause to stop Moore and that there was no evidence that they made their decision based on race. The court stated, “Moore has presented no evidence of the four officers’ invidious or bad faith.” Indeed, based on its review of video footage of the traffic stop, the court observed that “the police essentially treated him with great respect.”

Based on these facts of record and the district court’s finding, we conclude that there is a complete absence of evidence that the officers acted with discriminatory purpose. And because Moore's equal protection claim requires clear evidence that he was stopped because he was Black, it should fail for that reason alone.

The district court, however, did not limit its analysis to the question whether Moore was stopped because of his race. It chose to consider circumstantial evidence relating to police stops in Richmond more generally, remarking that “there is a problem in Virginia that needs to be addressed.' The court thus undertook to treat Moore's motion as a programmatic challenge to the fairness of traffic stops made by the Richmond Police Department. And in doing so, the court drew on its personal observations:

[Y]ou know, I have been sitting up here now for ten years looking at these cases and I have never had a case with a white driver. Never in ten years. . . . And when I drive in other places in the City, I frequently see stops with young African-American men standing outside their cars. And I just think it is an intolerable situation. It is intolerable that the City puts up with it.

With these observations, the court lamented, “One would think that Virginia’s citizens would cry out in protest over this situation, but they don't.' Consequently, the court Page 17announced, “The time has come for this practice to end.” Thus, despite finding no evidence of racial animus on the part of the police officers who stopped Moore, the court invited Moore to present more generalized evidence about traffic stops in Richmond, and based on Moore’s statistical and historical evidence, it found that the Richmond Police Department acted with a racially discriminatory purpose. Evaluating that evidence rather uncritically, the court explained:

Moore’s statistics evince that RPD police officers stop Black drivers at a rate that far exceeds the rate at which they stop white drivers. This correlation, on its own, does not prove causation. But that does not mean evidence of a correlation is per se irrelevant.

* * * In analyzing the statistics in this case, Dr. Coston never asserted that race caused a particular stop. Instead, Dr. Coston determined the relationship between a driver’s race and the primary result of a traffic stop.

* * * To further support his claim, Moore cites to Richmond’s racially segregated and discriminatory history. . . . Dr. Chiles testified that the historical segregation of the City of Richmond is the product of “the tide of history.”

* * * The data showing that RPD stops Black drivers five times as often as it stops white drivers, coupled with the evidence of Richmond's history of discrimination, reveals that RPD’s discriminatory purpose contributed to its officers’ decision to stop Moore. Moore therefore succeeds in meeting his burden of showing discriminatory purpose.

(Citation omitted).

We conclude that the court's expansion of the inquiry was unjustified by the evidence and that the evidence, even if taken at face value, was insufficient to show a discriminatory purpose - on the part of either the arresting officers or the Richmond Police Department as a whole.

Page 18First, the statistical evidence concerning the relationship between race and the likelihood of a traffic stop was not strong. Dr. Coston described the relationship between the variables, as indicated by results of statistical analyses known as chi-square tests, as “important” but “somewhat weak.” Moreover, the disparities themselves did not paint a materially different picture. While the fact that Black drivers constituted 77% of the drivers stopped in Richmond may have been striking at first blush, the figure represented only a naked statistical disparity, and both this court and the Supreme Court have declined to infer a violation of the Equal Protection Clause from statistical disparities far starker than any presented here. See, e.g. , Feeney , 442 U.S. at 270 (98% of those eligible for a benefit were men); Armstrong , 517 U.S. at 459 (100% of the office’s 24 drug conspiracy prosecutions closed in a given year involved Black defendants); Hare , 820 F.3d at 99 (100% of defendants prosecuted in stash-house sting cases in Maryland were Black); United States v. Venable , 666 F.3d 893, 903 (4th Cir. 2012) (87% of defendants charged with specific firearm offenses in a division were Black); United States v. Olvis , 97 F.3d 739, 742 (4th Cir. 1996) (more than 90% of those indicted in the area for crack cocaine trafficking were Black). Although statistical disparities might bear some relevance to a broader question, they generally cannot, as Dr. Coston conceded, show that any one decision was motivated by racial animus. Thus, a finding of discriminatory intent or purpose based solely on naked statistical disparities “remains the exceedingly rare exception to the general rule.” United States v. Thorpe , 471 F.3d 652, 661 (6th Cir. 2006); see also Olvis , 97 F.3d at 746 (noting the “general rule” that statistical evidence is inadequate to prove discriminatory purpose in cases implicating 'discretionary judgments essential to the criminal justice process'Page 19 (cleaned up)). Yet, the district court relied mainly on this naked statistical disparity to conclude that Moore was stopped because he was Black.

Second, the basis on which Dr. Coston compared White and Black drivers was also weak. While Dr. Coston compared the frequency with which drivers of each race were stopped in a 6-month window in 2020, this type of data could say little about whether any specific driver was stopped because he was Black, in part because Dr. Coston provided no data about the proportion of drivers in Richmond who were Black. Thus, if one were to show, hypothetically, that 65% of the drivers stopped in Jonesville were Black when 65% of the drivers in Jonesville were Black, the 65%-of-drivers-stopped statistic establishes nothing, because there is no disparity. We have repeatedly noted that “raw data” lacking “an appropriate basis for comparison” cannot satisfy any element of an Equal Protection claim. Olvis , 97 F.3d at 745; accord Hare , 820 F.3d at 99–100; Venable , 666 F.3d at 903. And we have specifically disclaimed reliance on statistics that fail to account for 'the number of blacks who were actually committing [a specific type of crime] or whether a greater percentage of whites could have been prosecuted for such crimes.” Olvis , 97 F.3d at 745. To be sure, we have noted that the absence of an appropriate benchmark is not necessarily dispositive with respect to selective-enforcement claims, mindful that such claims are not as conducive to data retention as are selective-prosecution claims. See Hare , 820 F.3d at 100–01. But the lack of any sound basis for comparison still weighs against finding clear evidence of animus in this case. See Mason , 774 F.3d at 830.

Third, the potential for recognized confounding variables to distort the relationship between race and traffic stops greatly undermined the probative value of Dr. Coston's Page 20statistical evidence. Statistics can shed light on the presence of “discriminatory intent if race can be isolated from other confounding variables.” Conley v. United States , 5 F.4th 781, 796 (7th Cir. 2021) (emphasis added). But Dr. Coston acknowledged that the statistical test applied, unlike other available methods, could not control for confounding variables, such as poverty, residence in a high-crime area of town, and existing arrest warrants. That lack of control may not be dispositive in every case, but it is particularly damaging here, where evidence in the record suggested that the Richmond Police Department's traffic stops tended to cluster in high-crime areas, and further evidence indicated that controlling for confounding variables has, in some studies, made racial disparities in traffic-stop outcomes “disappear[].”

Finally, the existence of probable cause to stop Moore makes it even less likely that the traffic stop was motivated by discriminatory purpose. As we have noted, there is a conceptual 'overlap' between Fourth Amendment claims and selective-enforcement challenges. Mason , 774 F.3d at 831. When an officer has probable cause to effect a stop, particularly for an unusual and glaring traffic offense such as the one at issue here, we do not think it inevitable or even likely that the officer’s first thought will concern the race of the suspect, as opposed to, for instance, the offense transpiring before his eyes. Thus, where, as here, officers have probable cause to stop a defendant, it is “even less likely” the defendant’s selective-enforcement claim could succeed. Id. This is particularly so when the record is devoid of any evidence revealing 'significant departures from normal procedures,” which might, if present, suggest that the existence of probable cause was not the actual motive for the stop. See Cent. Radio Co. , 811 F.3d at 635 (quoting Sylvia Dev.

Page 21Corp. v. Calvert County , 48 F.3d 810, 819 (4th Cir. 1995)). Here, the officers observed three cars bearing identical temporary license tag numbers in one night, giving them cause to believe that the drivers were knowingly displaying fake tags, in violation of Virginia Code § 46.2-612(B)(1). The officers thus attempted to stop each of the three cars to question the driver about the tag. This sequence of events is striking only for its normalcy. Indeed, we expect that the public would be surprised had the officers not stopped Moore in the circumstances presented.

For all these reasons, we conclude that Moore’s statistical evidence fails to show, to any degree, that the officers who stopped Moore acted with discriminatory purpose.

Moore nonetheless argues that 'the historical evidence Dr. Chiles proffered provides additional evidentiary support of discriminatory purpose by providing the historical context in which the statistical evidence should be understood.” But Moore’s historical evidence failed to tether past discrimination to this case at all. Indeed, the evidence that he presented does not even broach the 21st century; Dr. Chiles candidly admitted that he did not have any evidence “[b]eyond 1989” regarding the relationship between racial bias and the Richmond Police Department’s practices. This concession renders Dr. Chiles’s testimony virtually irrelevant. As the Supreme Court has observed, “[U]nless historical evidence is reasonably contemporaneous with the challenged decision' - here, the decision to stop Moore - 'it has little probative value.' McCleskey , 481 U.S. at 298 n.20 (emphasis added).

In short, because Moore has challenged a single decision that the district court recognized was not animated by racial prejudice, it is difficult to see the relevance of the Page 22expert testimony offered by Moore to the issue at hand, especially when that testimony concededly did not point to the reason why the officers stopped Moore. Moreover, were we to accept that the data presented by the expert witnesses were sufficient to dismiss Moore's indictment, those same data would, without more, justify dismissing every prosecution arising from the traffic stop of a Black person in Richmond. Such a notion is untenable.

* * *

In addressing Moore’s selective-enforcement claim, the district court should have remained focused on whether Moore was stopped on December 5, 2020, because he was Black. The court, however, out of a more general concern about discriminatory law enforcement in Richmond, invited the inquiry to transform into a broad investigation into the Richmond Police Department’s traffic enforcement practices of a kind more suited for a legislative committee. While we do not dismiss the court’s larger concern, we conclude that the record in this case did not justify dismissing the indictment against Moore for selective enforcement. And as Moore failed to introduce evidence sufficient to show that he was stopped on the basis of race, we need not address his evidence to show discriminatory effect.

We reverse the order of the district court and remand with instructions to reinstate the indictment.

IT IS SO ORDERED.