Notice: This opinion is subject to formal revision before publication in the Atlantic and Maryland Reporters. Users are requested to notify the Clerk of the Court of any formal errors so that corrections may be made before the bound volumes go to press.

DISTRICT OF COLUMBIA COURT OF APPEALS

No. 23-CF-0196

RAKEEM WILLIS, APPELLANT,

V.

UNITED STATES, APPELLEE.

Appeal from the Superior Court

of the District of Columbia

(2019-CF1-007968)

(Hon. Michael Ryan, Trial Judge)

(Hon. Milton C. Lee Jr., Motions Judge) (Argued September 24, 2025 Decided November 20, 2025)

Stefanie Schneider, Public Defender Service, with whom Jaclyn Frankfurt and Alice Wang, Public Defender Service, were on the briefs, for appellant.

David B. Goodhand, Assistant United States Attorney, with whom Edward R. Martin, United States Attorney at the time the brief was filed, and Chrisellen R. Kolb, Ariel Dean, and Michael Spence, Assistant United States Attorneys, were on the brief, for appellee.

Before EASTERLY and SHANKER, Associate Judges, and RUIZ, Senior Judge.

SHANKER, Associate Judge: Late one night in January 2019, Sean Shuler, Javon Abney, and Tyrik Hagood were shot and killed on the 1500 block of Fort Davis Place in Southeast, Washington, D.C. Appellant Rakeem Willis, along with

2 his codefendant Jonathan Winston, was charged with three counts of first-degree (premeditated) murder for the deaths of the three men. During pretrial proceedings in Superior Court, the motions court admitted the testimony of the government’s historical cell site location information (CSLI) expert over objections from Mr. Winston. 1 Following a jury trial, Mr. Willis was convicted of all three counts of first-degree murder. 2 Mr. Willis appeals these convictions on three grounds.

First, Mr. Willis asserts that the motions court erred in its analysis under Motorola Inc. v. Murray, 147 A.3d 751, 756 (D.C. 2016) (en banc), and Federal Rule of Evidence 702 when admitting the expert testimony of Federal Bureau of Investigation (FBI) Special Agent Billy Shaw, a member of the Bureau’s Cellular Analysis Survey Team (CAST). Specifically, he contends that the motions court failed to properly determine whether Special Agent Shaw’s testimony was the product of reliable principles and methods and to consider whether Special Agent Shaw reliably applied the methodology to the facts of this case. Second, Mr. Willis argues that the motions court erred when it found that the government’s notice regarding Special Agent Shaw’s expert testimony satisfied Superior Court Rule of

1

Because the judge who presided over the motions proceedings and the judge who presided over trial were different, we refer in this opinion to the “motions court” and the “trial court” as appropriate.

2

The trial court granted Mr. Winston’s motion for a judgment of acquittal at the conclusion of the government’s case in chief.

3 Criminal Procedure 16. Third, Mr. Willis contends that the trial court abused its discretion in declining to disqualify a juror based on indications that the juror was sleeping or was otherwise inattentive throughout trial.

We hold that the motions court’s Motorola/Rule 702 analysis fell short of what is required by both Rule 702 and our case law. We also conclude that this error was not harmless. Because we find reversible error on this issue, we decline to reach Mr. Willis’s other arguments. Accordingly, we reverse and remand for further proceedings consistent with this opinion.

I. Background

The evidence at trial included the following. Two witnesses who lived on the 1500 block of Fort Davis Place testified that, on the night Messrs. Shuler, Abney, and Hagood were murdered (January 26, 2019), they heard gunshots around 10:00 p.m. and saw two “silhouette[s]” shooting into a vehicle. As the shooters drove away, one got out of their vehicle and stood over Mr. Shuler’s body and “mess[ed]” with it. The car then drove off and neither witness was able to identify the make of the car or its license plate number. When police responded, they found Mr. Shuler’s body in the middle of the road and the bodies of Mr. Abney and Mr. Hagood seated in a parked car with the engine running and the windows and

4 driver’s door open. A medical examiner concluded that all three men died from multiple gunshot wounds.

About an hour after the shooting, in Capitol Heights, Maryland, police responded to a report of a black Lexus on fire. At trial, the government introduced video footage purporting to show that this Lexus was the car used by the shooters on the night of the murders. Police later searched the car and found parts of a Glock 27 handgun, bullets, and cartridge casings. The government’s firearms expert testified that the ammunition recovered from the murder scene did not match the Glock 27.

Phone numbers and related call records connected Mr. Willis to Mr. Shuler. The government introduced evidence indicating that, at the time of the murders, Mr. Willis used two different phone numbers: one with a 443 area code and another with a 240 area code. Call records showed calls between Mr. Shuler’s number and Mr. Willis’s 443 number throughout the afternoon of January 26. The records also revealed calls between Mr. Shuler’s number and Mr. Willis’s 240 number from shortly after 6:00 p.m. until just before 10:00 p.m. that night. The calls between Mr. Shuler’s number and the numbers associated with Mr. Willis stopped around 10:00 p.m. The 443 number linked with Mr. Willis made a call to an IHOP restaurant in Maryland just before midnight.

5

The government presented a historical CSLI expert, FBI Special Agent Shaw, who used this phone activity to chronicle the approximate movement of the relevant phones on the night of the murders. Using the locations of the cell towers that the phone numbers connected with, Special Agent Shaw showed the 240 number linked with Mr. Willis moving towards an area in the vicinity of the shootings shortly before 10:00 p.m. He then showed the 443 number moving northbound on I-295 and converging with an alleged co-conspirator approximately where the burned-out Lexus was later found. The cell site evidence further revealed the 443 number being used just south of the IHOP restaurant where Mr. Willis was later identified on surveillance footage by a law enforcement officer.

The jury found Mr. Willis guilty of three counts of first-degree murder. The trial court sentenced Mr. Willis to 120 years of imprisonment. This appeal followed.

II. Analysis

Mr. Willis raises three issues on appeal, but, as noted above, we need not address two of them. We hold that the motions court improperly exercised its discretion when assessing the reliability of Special Agent Shaw’s expert testimony

6 under Motorola and Federal Rule of Evidence 702 and that this error was not harmless. 3

A. Additional Background

In deciding the admissibility of expert testimony, a court must consider whether:

(a) the expert’s scientific, technical, or other specialized

knowledge will help the trier of fact to understand the

evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data;

(c) the testimony is the product of reliable principles and

methods; and

(d) the expert has reliably applied the principles and

methods to the facts of the case.

3

The jury also found Mr. Willis guilty of one count of fleeing a law enforcement officer. The trial court sentenced Mr. Willis on that conviction to five years in prison to be followed by three years of supervised release. Mr. Willis does not raise an argument related to this count on appeal or suggest that the error in admitting the expert testimony affects it, and we are satisfied that the count “can be logically separated” from the first-degree murder counts. See Gethers v. United States, 556 A.2d 201, 205 (D.C. 1989). Accordingly, we do not disturb Mr. Willis’s conviction for fleeing a law enforcement officer. See id. (“An appellate court may affirm a conviction on one count of an indictment despite reversing another count if . . . the two counts can be logically separated from one another because they relate to separate and distinct crimes with respect to which the jury made separate findings.”).

7 Faltz v. United States, 318 A.3d 338, 348 (D.C. 2024) (quoting Motorola, 147 A.3d at 756); see Fed. R. Evid. 702. 4 “The Rule [ ] elevates the trial judge’s role as gatekeeper; it essentially provides that when a party proffers expert scientific testimony, the trial court must make a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid . . . .” Lewis v. United States, 263 A.3d 1049, 1059 (D.C. 2021) (citation modified). “Rule 702 expressly requires the trial court to determine whether an expert reliably applied principles and methods to the case at hand . . . .” Faltz, 318 A.3d at 348 (citation modified).

Before trial, Mr. Willis’s codefendant, Mr. Winston, moved to exclude the historical cell site report and proposed testimony of the government’s expert, Special Agent Shaw. 5 In Mr. Winston’s view, the notice of the expert and accompanying

4

Federal Rule of Evidence 702 was amended in December 2023, after Mr. Willis’s trial. Fed. R. Evid. 702 (Apr. 24, 2023, eff. Dec. 1, 2023). The revised version requires a court to determine whether “the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.” Fed. R. Evid. 702(d). The intent of the amendment was “to emphasize that each expert opinion must stay within the bounds of what can be concluded from a reliable application of the expert’s basis and methodology.” Fed. R. Evid. 702 advisory committee’s notes to 2023 amendments. “Nothing in the amendment imposes any new, specific procedures. Rather, the amendment is simply intended to clarify that Rule 104(a)’s requirement applies to expert opinions under Rule 702.” Id.

5

The government’s original expert was FBI Special Agent Lynda Thomas, and she is the subject of Mr. Winston’s motion. The government notified the

8 report did not satisfy the requirements of Rule 16 and the methodology in the report was insufficiently reliable to be admitted under Motorola and Rule 702. The thrust of Mr. Winston’s argument was that the government’s expert would testify to the exact coverage area of the relevant cell towers, which, he alleged, is not possible based solely on historical call records. In opposition, the government argued that its cell site expert would opine neither on the exact location of the phones at issue nor on the exact coverage areas of the cell towers. Because Mr. Winston had, in the government’s view, grounded his motion on an inaccurate presumption about the expert’s opinions, his arguments were “essentially moot.”

The motions court, ruling from the bench at a status hearing, concluded that the proposed expert testimony and report were based on reliable principles and would be admitted, albeit with certain limitations. The court declined to conduct a hearing under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), because it had previously heard testimony by similar experts in approximately a dozen other cases and had found the principles underlying the testimony reliable. The court repeatedly stated that cross-examination was the appropriate tool for meeting the force of the expert testimony. To address Mr. Winston’s concerns, the court also indicated that it would instruct the jurors about how to process and weigh defendants in June 2022 that Special Agent Shaw would be replacing Special Agent Thomas. That substitution is not pertinent to Mr. Willis’s claims on appeal.

9 expert testimony. The court explained that, in its view, Rule 702’s gatekeeping function is designed to allow “more expert evidence in” and the rejection of expert testimony is “the exception rather than the rule.” Specifically with respect to Rule 702(d), the motions court stated that “vigorous cross examination, presentation of contrary evidence including [a] contrary witness, [and] careful instruction on the burden of proof are the traditional means” of addressing the issue of reliable application. Accordingly, the court denied Mr. Winston’s motion.

B. Standard of Review

We review a decision to admit or exclude expert testimony for abuse of discretion. 6 Faltz, 318 A.3d at 347. “In reviewing for abuse of discretion, we must determine whether the decision maker failed to consider a relevant factor, whether the decision maker relied upon an improper factor, and whether the reasons given reasonably support the conclusion.” Bishop v. United States, 310 A.3d 629, 641 (D.C. 2024) (citation modified). In addition, the trial court must make “an informed choice drawn from a firm factual foundation.” Brooks v. United States, 993 A.2d 1090, 1093 (D.C. 2010) (citation modified). Although “the concept of exercise of

6

As noted above, only Mr. Willis’s codefendant, Mr. Winston, challenged the admissibility of Special Agent Shaw’s expert testimony. The government, however, has not argued that Mr. Willis failed to preserve this claim, and, at oral argument, it conceded that Mr. Winston’s objections served to preserve the issue for both defendants.

10 discretion is a review-restraining one,” id. (citation modified), “there is an important tradeoff for giving the trial court such latitude: that court must take no shortcuts; it must exercise its discretion with reference to [a]ll the necessary criteria.” Ibn-Tamas v. United States, 407 A.2d 626, 635 (D.C. 1979). Consequently, “the appellate court must not affirm a ruling premised on trial court discretion unless the record clearly manifests either (1) that the trial court has ruled on each essential criterion, or (2) that the trial court, as a matter of law, had but one option.” Id. (citation modified).

C. Discussion

Mr. Willis contends that the motions court abused its discretion in admitting Special Agent Shaw’s testimony under Motorola and Federal Rule of Evidence 702. He argues that, while the court understood that Rule 702 is the governing standard, it failed to “actually apply it.” In particular, he asserts that the motions court did not properly determine whether Special Agent Shaw’s testimony was the product of reliable principles and methods or “whether Agent Shaw ‘reliably applied’ the methodology to the facts of this case, as required by Rule 702(d).” Instead, the court, according to Mr. Willis, “presumed that all CAST agents use the same principles and methods to reach their conclusions” and that their “testimony is always admissible.” These “presumption[s],” argues Mr. Willis, are both “factually and legally flawed.” Mr. Willis further claims that the motions court relied too heavily on the tools of the adversarial system, like cross-examination, as justification for

11 inquiring no further into the reliability issue. He contends that this was improper because Daubert requires the court to make a preliminary assessment of reliability before the jury weighs the evidence. This error, Mr. Willis asserts, was not harmless and requires reversal.

In response, the government argues that the motions court properly identified the data supporting Special Agent Shaw’s proposed testimony and the methodology he employed, finding them both sufficient and reliable, which satisfies Rule 702. The government additionally contends that Special Agent Shaw’s testimony was correctly admitted under Rule 702 because the court acted within its discretion when declining to hold a Daubert hearing after finding the expert testimony clearly admissible. Historical CSLI analysis qualifies as such evidence, according to the government, because it has been found to be sufficiently reliable, with some limitations, by numerous state and federal courts. Because the court imposed appropriate limitations on Special Agent Shaw’s testimony, the government asserts, no further assessment was needed.

1. The Court Failed to Properly Exercise

Its Discretion Under Rule 702.

As an initial matter, we face some challenges in assessing whether the motions court properly analyzed the admissibility of Special Agent Shaw’s testimony. After

12 discussing Mr. Winston’s motion to exclude Special Agent Shaw at a status hearing, the motions court did not issue a written order setting forth its bases for admitting the expert testimony. This court is thus left to parse the court’s statements at the hearing to determine whether the court conducted a proper Motorola analysis. 7 See, e.g., District of Columbia v. Facebook, Inc., 340 A.3d 1, 11 (D.C. 2025) (describing the difficulties appellate courts face when they are forced to review hearing transcripts as the primary source of a court’s Motorola/Rule 702 analysis); Ealey v. Ealey, 596 A.2d 43, 46 (D.C. 1991) (“A trial judge must make findings of fact and conclusions of law with respect to every material issue that is raised; otherwise meaningful appellate review cannot occur and this court must remand the case or the record.”); cf. Faltz, 318 A.3d at 348 (explaining that a conclusory statement in an order denying a motion to exclude an expert witness necessitated a remand for further analysis as required under Motorola).

7

We do not suggest that written rulings are always required. Cf. Ibn-Tamas, 407 A.2d at 636 n.17 (“This is not to suggest that a trial court must articulate reasons, let alone correct reasons, for every evidentiary ruling. In most instances, the appellate court will be able to ‘infer the reasoning upon which the trial court made its determination.’” (citation omitted)). They are, however, advisable where, as here, a court must analyze the admissibility of evidence under a multi-factor test, in order to avoid a situation in which “the appellate court cannot tell from the record whether the judge sufficiently addressed each criterion required for the ruling.” Id.

13

To the extent we can glean the court’s thinking from the hearing transcript, the court’s statements reveal some analysis under Motorola—and Mr. Willis conceded at oral argument that the motions court’s analysis of Rules 702(a) and (b) was sufficient—but fail to show a sufficient assessment of all the elements of Rule 702.

Rule 702, which we adopted in Motorola, requires the court to consider all four of the rule’s factors when assessing proposed expert testimony. Fed. R. Evid. 702; see Motorola, 147 A.3d at 757-58. Notably absent from the discussion at the motions hearing is an analysis of Special Agent Shaw’s proposed testimony under Rule 702(d). Rule 702(d), at the time of the hearing, “expressly require[d] the court to determine whether ‘the expert has reliably applied the principles and methods to the facts of the case.’” Motorola, 147 A.3d at 757 (quoting Fed. R. Evid. 702(d)). While the motions court here clearly understood the standard it was supposed to employ, there is no indication in the record that it determined whether Special Agent Shaw reliably applied the methodology used by CAST agents to the facts of this case. Instead, the court immediately turned to the tools of the adversarial process as the means to make the reliability determination. This was improper. See Daubert, 509 U.S. at 592-93 (holding that the trial judge must make “a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether that reasoning or methodology properly can be applied to the

14 facts in issue”); see also Nease v. Ford Motor Co., 848 F.3d 219, 231 (4th Cir. 2017) (“The district court’s ‘gatekeeping function’ under Daubert ensures that expert evidence is sufficiently relevant and reliable when it is submitted to the jury. Rather than ensure the reliability of the evidence on the front end, the district court effectively let the jury make this determination after listening to [appellant’s] cross examination of [appellee’s] expert.”). While “[v]igorous cross-examination” and other features of the adversarial system remain “the traditional and appropriate means of attacking shaky but admissible evidence,” Daubert, 509 U.S. at 596, it is the court’s duty to make an initial determination about the reliability of the proposed expert’s application of the methodology to the facts at issue, see Motorola, 147 A.3d at 757.

When questions were raised regarding Special Agent Shaw’s application of CAST’s methodology in this case, specifically about the reliability of the graphic displays in the draft reports, the motions court did not engage with the content of the draft reports but instead based its views on past experience. 8 When pressed further by Mr. Winston’s counsel, the motions court sidestepped a Rule 702(d) analysis and

8

The draft report in existence at that time was authored by Special Agent Shaw’s predecessor, Special Agent Thomas. Special Agent Shaw’s final report was not available until July 28, 2022, well after the hearing.

15 again suggested that cross-examination was the means by which to properly assess the reliability of Special Agent Shaw’s work. 9

The court’s approach illustrates the problem the advisory committee was trying to correct with the 2023 amendments to Rule 702(d). When responding to an argument that Special Agent Shaw’s report would subvert the limitations placed on his testimony, the court responded:

So isn’t that something that’s fair for cross examination.

For example, the expert—assuming that the expert goes

beyond what you think is within his knowledge base,

wouldn’t that be appropriate for you to simply use cross

examination to demonstrate . . . . It just seems to me to fit

right in with what we would use as a tool to test the

reliability of the princip[le], but also the reliability of its

application to the data. This view is incorrect. It is the court’s job to determine up front whether an expert can testify “within the bounds of what can be concluded from a reliable application of the expert’s basis and methodology” because “jurors may lack the specialized knowledge to determine whether the conclusions of an expert go beyond what the expert’s basis and methodology may reliably support.” Fed. R. Evid. 702 advisory committee’s notes to 2023 amendments (citation modified). This duty continues

9

When neither party could produce a copy of the draft report to supplement the discussion, the motions court decided to admit Special Agent Shaw’s testimony anyway and placed the burden on the defendants to re-raise with the court objections about particular slides in the report.

16 throughout the expert’s testimony because the court remains in a better position to assess whether the expert has strayed beyond a reasonable application of their methods. See United States v. Garcia, 752 F.3d 382, 396 (4th Cir. 2014) (holding that the district court fulfilled its gatekeeping role initially but that it failed to do so “throughout [the expert’s] testimony” where it was apparent that the expert was not reliably applying their methodology).

At oral argument, the government contended that the motions court satisfied Rule 702(d) because it prohibited Special Agent Shaw from opining about the phone user’s exact location or asserting that phones always connect to the closest cell tower. The motions court’s discussion at the hearing of these so-called “tripwires,” according to the government, indicates an understanding of a decade’s worth of case law regarding CSLI and shows consideration of the application of Special Agent Shaw’s cell site methodology to Mr. Willis’s case. We disagree. These limitations do not reveal whether Special Agent Shaw “reliably applied the principles and methods” behind his testimony to the matter at hand. Faltz, 318 A.3d at 348; see Fed. R. Evid. 702(d). At most, the motions court’s understanding of standard limitations on cell site analysis shows a general grasp of CAST agents’ methodology and likely testimony. But, as the motions court acknowledged at least implicitly through its discussions with the government, these limitations are standard practice for CSLI experts. A general application is not an adequate substitute for determining

17 whether Special Agent Shaw reliably applied CAST’s methodology “to the facts of th[is] case.” Motorola, 147 A.3d at 757.

Put another way, the motions court understood the “bounds of what can be concluded from a reliable application of [Special Agent Shaw’s] basis and methodology.” Fed. R. Evid. 702 advisory committee’s notes to 2023 amendments. But the record does not show that the court meaningfully assessed whether Special Agent Shaw’s proposed testimony and report fell within those “bounds.” See id. The government is correct that the court made a limited inquiry into the content of Special Agent Shaw’s testimony at the hearing. The court’s query, however, did not go beyond obtaining representations from the government about the content of Special Agent Shaw’s testimony, which is insufficient. See Nelson v. United States, 601 A.2d 582, 590 (D.C. 1991) (holding that courts cannot rely on the “representations of counsel to fill gaps in the record”); Girardot v. United States, 996 A.2d 341, 346 (D.C. 2010) (“In conducting its scrutiny of the [expert] proffer, the trial court must take no shortcuts.”) (citation modified); see also United States v. Daniels, 572 F.2d 535, 540 n.2 (5th Cir. 1978) (“It is questionable whether a trial court should base evidentiary decisions on a prosecutor’s unsubstantiated representations.”). While “there is no particular procedure that the trial court is required to follow in executing its gatekeeping function under Daubert,” Lewis, 263 A.3d at 1060 (citation modified), courts must materially engage with the proffered

18 expert testimony to satisfy Rule 702(d)’s reliability requirement, see Motorola, 147 A.3d at 757.

We find United States v. Morgan, 45 F.4th 192 (D.C. Cir. 2022), instructive. There, the appellant argued that the trial court erred when admitting the testimony of the government’s expert—another member of CAST—after holding a multi-day Daubert hearing. Id. at 198. The expert, rather than using historical CSLI, relied on “drive testing technology” to opine on the approximate location of the defendant’s and victim’s cell phones. Id. at 201-02. When evaluating the trial court’s examination of whether the expert “reliably applied” his methodology to the case, the court of appeals identified parts of the trial court’s order revealing a proper analysis. Id. at 203-04. For example, the trial court checked to see if the expert had assessed whether there had been any changes to the towers or cellular network since the relevant incident before forming his opinions. Id. The trial court also scrutinized whether a second CAST agent had peer reviewed the expert’s work for potential errors. Id. at 204. Satisfied that these details, among others, showed that the trial court had fulfilled its responsibility under Rule 702(d), id. at 203, the appellate court concluded that “the [trial] court did not abuse its discretion in finding [the expert’s] testimony to be sufficiently reliable[,]” id. at 204.

19

We cannot reach the same conclusion here. Courts need not check particular boxes to satisfy Rule 702(d). See Motorola, 147 A.3d at 755 (“[T]he trial judge must have considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable.” (quoting Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 152 (1999))). They must, however, demonstrate, as the district court in Morgan did, that they properly considered whether an expert’s “principles and methods are . . . applied reliably to the facts of the case.” Fed. R. Evid. 702 advisory committee’s notes to 2000 amendments; see Faltz, 318 A.3d at 348. The motions court did not demonstrate such consideration here. Absent that, its Motorola analysis was incomplete. See Faltz, 318 A.3d at 348; Motorola, 147 A.3d at 757.

Consequently, the motions court’s efforts “fell short of its gatekeeping role in admitting expert testimony under Motorola.” See Faltz, 318 A.3d at 348. The court was “expressly require[d]” to determine whether Special Agent Shaw “reliably applied” CAST’s methods to the facts at hand. Motorola, 147 A.3d at 757. The motions court’s statements at the hearing, however, do not allow us to conclude that this analysis was conducted. When questions were raised in that vein, the motions court was not prepared to discuss the report and placed the onus back on the defendants to raise the issue again. A proper analysis under Rule 702(d) requires more. Cf., e.g., Murray v. Motorola, Inc., 339 A.3d 152, 173-74 (D.C. 2025) (“Judge

20 Irving meticulously reviewed every expert’s report to determine if their testimony passed muster. He ultimately found that they did not and articulated reasonable bases for each of those determinations . . . act[ing] well within his discretion.”); Morgan, 45 F.4th at 198 (explaining how data from the drive test that the CAST agent employed was run through a proprietary program to generate maps and then “independently revie[ed]” by a second agent “to check whether there were any errors”); United States v. Valencia, 600 F.3d 389, 424 (5th Cir. 2010) (holding that a Rule 702(d) analysis “requires more than a glance at the expert’s credentials; the court must also ensure that the expert has reliably applied the methods in question” (citation omitted)). The failure to properly consider the requirements of Rule 702(d) is an abuse of discretion. See Bishop, 310 A.3d at 641; see also Benn v. United States, 978 A.2d 1257, 1277 (D.C. 2009) (“Th[e] court must take no shortcuts; it must exercise its discretion with reference to all the necessary criteria.” (quoting Ibn-Tamas, 407 A.2d at 635)).

2. The Motions Court’s Error Was Not Harmless.

The remaining question is whether the court’s failure to properly scrutinize the reliability of Special Agent Shaw’s application of his proffered methodology to the facts of this case was harmless. See Smallwood v. United States, 312 A.3d 219, 227 (D.C. 2024) (“This court will affirm the trial court despite a non-constitutional error if the error was harmless . . . .”). “An error is harmless if we can say, with fair

21 assurance, that the judgment was not substantially swayed by the error.” Faltz, 318 A.3d at 348 (citation modified). “[T]he central ‘issue is whether the record eliminates the appellate court’s doubt about whether the error influenced the jury’s decision.’” Smallwood, 312 A.3d at 227 (quoting Moghalu v. United States, 263 A.3d 462, 472 (D.C. 2021)). If an error is found, it is the government’s burden to demonstrate that the error was harmless. See Randolph v. United States, 882 A.2d 210, 223 (D.C. 2005); Morales v. United States, 248 A.3d 161, 182 (D.C. 2021). For nonconstitutional errors, this court has the “authority to find [a] trial court error harmless notwithstanding the government’s failure to claim harmlessness.” Randolph, 882 A.2d at 223. “Nevertheless, this authority must be exercised with restraint.” Id.

The government’s brief includes no argument that the motions court’s error in its Motorola analysis was harmless, nor any citation to Kotteakos v. United States, 328 U.S. 750, 764-65 (1946) (articulating the test for nonconstitutional errors), or any other authority concerning harmless error. “The word harm does not appear in its brief, nor does harmful, harmless, or any other permutation or synonym.” Morales, 248 A.3d at 182. We therefore conclude that the government has forfeited any argument that the motions court’s error in its Motorola/Rule 702 analysis was harmless. See Rose v. United States, 629 A.2d 526, 535 (D.C. 1993) (“It is a basic

22 principle of appellate jurisprudence that points not urged on appeal are deemed to be waived.”).

While we have the discretion to find the trial court error harmless notwithstanding the government’s forfeiture, we exercise this discretion only when harmlessness is “obvious.” Randolph, 882 A.2d at 223. Mr. Willis contends that the government repeatedly highlighted Special Agent Shaw’s testimony during trial, illustrating the “centrality” of this evidence to the government’s case. He also argues that the government presented neither motive evidence nor physical evidence tying him to the shooting, including no fingerprint or DNA evidence connecting him with any physical evidence at the scene of the murders. This resulted, according to Mr. Willis, in an “exceedingly weak” case. The government’s case, he claims, relied almost entirely on inferences to be drawn from Mr. Willis’s purported calls with Mr. Schuler and the alleged locations of Mr. Willis’s phones on the day of the murders.

Mr. Willis’s arguments have at least some force. Because the harmlessness of the motions court’s error is therefore “at least debatable,” we conclude that the government should not be relieved of its forfeiture. See id.

Accordingly, we hold that the admission of Special Agent Shaw’s expert report and testimony in the absence of a sufficient Motorola analysis was error that

23 we cannot deem harmless. Mr. Willis’s murder convictions must therefore be reversed and this case remanded for further proceedings. See Johnson v. United States, 398 A.2d 354, 366 (D.C. 1979) (“[I]f the error had a possibly substantial impact upon the outcome, the case should be reversed.”).

III. Conclusion

For the foregoing reasons, we reverse Mr. Willis’s first-degree murder convictions and remand this case for further proceedings.

So ordered.