dissenting:
I respectfully dissent. I would hold that the trial court erred by denying Johnson’s motion to suppress, and, because the erroneously admitted evidence was not harmless beyond a reasonable doubt, I would reverse and remand for a new trial.1
*372The central question in this case is whether the Philadelphia police had reasonable, articulable suspicion to believe that Johnson “might be armed and pose a danger to their safety before the frisk occurred.” Maj. Op. 363. The majority holds that they did. Id. To analyze the correctness of the majority’s conclusion, I first consider when the frisk commenced and then assess what facts were known to the officers at that time.
I.
When did the frisk begin? I agree with the majority that no decision of this court has decided the question of “when a frisk commences.” And I agree with the majority that we decide that question as a matter of law. Thus, we must look elsewhere as we apply our own logic and experience to the cases that touch upon the issue. The majority relies on a New York case,2 an Oregon statute,3 and a dictionary4 to conclude that “it makes little conceptual sense to regard a frisk ... as having begun before the police begin to pat down the exterior of a person’s body.” Maj. Op. 367. In the very same paragraph, however, the majority, citing two cases from the Supreme Court of the United States,5 cogently observes that “one may ask whether taking hold of a person by the elbow or shoulder to move him into a better position for a frisk is functionally — or legally — part of the patdown itself, rather than of the antecedent detention leading to a frisk.” Id. It is hard to imagine how it is not. We are splitting hairs much too finely if we pretend that an officer who necessarily touches a suspect to move him into position for a frisk has not thereby commenced the frisk, even though the actual “pat-down” of the suspect’s clothing does not occur until seconds or minutes later, after the officer has “positioned” the suspect by moving him to the spot where he wants him to be. Common sense dictates that the frisk is one continuous event that begins, at the latest, when the officer physically moves a suspect into position for a patdown, and it is logically inseparable from the patdown itself. As the government urged at oral argument, “a frisk has to be intentional touching for the purpose of locating a weapon.” To say that the “intentional touching” does not include positioning the suspect is nonsensical, and none of the authorities cited by the majority (from New York, Oregon, or the dictionary) hold to the contrary. Rather, each defines that portion of a frisk that includes the patdown of the suspect’s outer clothing, but none states that the frisk cannot, as a matter of law, begin earlier. I would hold, as a matter of law, that the frisk in this case began when Officer Harvey moved Johnson into position.
The record in this case amply supports the conclusion that the initial touching of Johnson was part and parcel of the frisk. The touching of Johnson was not necessary to prevent his escape or to preserve the safety of the officers during the investigation; rather, its sole purpose was to effectuate the frisk. Had Officer Harvey actually felt a weapon while manipulating Johnson into position, I would have no *373trouble holding that the actual “frisk” had begun. Thus, I find no legally significant distinction between Officer Harvey’s physical manipulation of Johnson and the “frisk” itself.6
Although the government argues that the “frisk” did not begin until Officer Harvey patted down the outside of Johnson’s clothing, it has provided no authority to support its position, and I have found scant support. Furthermore, our prior cases suggest that the frisk began, at the latest, when Officer Harvey manipulated Johnson’s body to effectuate the frisk. See Jackson v. United States, 805 A.2d 979, 987 (D.C.2002) (“[B]y the time [appellant] was asked to turn around (apparently in preparation for a frisk), and the officer touched [appellant’s] jacket, the police crossed the critical line between consent and coercion.” (citation and internal quotation marks omitted)); Powell v. United States, 649 A.2d 1082, 1087 (D.C.1994) (opinion of Sullivan, J.) (finding that factors that manifested themselves after the suspect was ordered to assume the position and after the decision to frisk had been made should not have been included in the reasonable suspicion determination); id. at 1090 (Farrell, J., concurring) (considering only facts known to the officers before “ordering [appellant] to place his hands on the car and patting him down”); see also United States v. Christian, 187 F.3d 663, 670 (D.C.Cir.1999) (“[W]e assess a Terry search from the standpoint of the moment of the stop ... not from the subsequent period in which the officer begins to take protective measures.”).
II.
What, then, did the officers know at the time the frisk commenced, that is, at the time Officer Harvey moved Johnson into position, that would have given them reasonable, articulable suspicion that he was armed and dangerous? To pass constitutional muster, the officers had to have reasonable, articulable suspicion not simply that Johnson might be engaged in criminal activity, but that he might be armed and dangerous. See Terry v. Ohio, 392 U.S. 1, 27, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968).
The trial court noted several factors that it believed would have given a well-trained officer the requisite suspicion. The trial court relied on the following factors:
1) Johnson was initially hesitant about pulling over;
2) Johnson was looking around, behind and to the side of him, and showed signs of nervousness or hyper-vigilance;
3) Johnson tried to exit his vehicle while Officer Harvey was still in his, having just stopped, and had to be ordered to stay in the car;
4) Having been told once to remain in the car, Johnson again tried to exit the *374car, which Officer Harvey clearly found suspicious and of concern;
5) Johnson provided a false name (which Officer Harvey clearly did not know at the time) and seemed nervous during the stop;
6) Johnson had no license;
7) Johnson said he was returning from an errand to retrieve diapers, but none were visible in the car;
8) Officer Harvey described Johnson as being hyper and moving around a lot in the vehicle; and
9) The vehicle was in “try [and] locate” status, and therefore Johnson would not be able to leave with the vehicle, and must be investigated.
Looking only at the first eight factors, even in their totality, the majority does not hold that “the constitutional strictures of Terry and its progeny had been satisfied.” Maj. Op. 368. Instead, the majority becomes comfortable with its holding only when it adds the “try and locate” status of the vehicle to the first eight factors, seemingly because it concludes that “try and locate” means that Johnson was driving a vehicle that could be regarded as stolen. Id. at 368-69. I, however, conclude that neither the first eight factors alone, nor those factors as augmented by the “try and locate” status of the vehicle, gave the officers reasonable, articulable suspicion that Johnson was armed and dangerous.
The stop was conducted during the day, and no evidence suggested that the area was a “high crime” location. Cf. Umanzor v. United States, 803 A.2d 983, 993 (D.C. 2002) (upholding a stop that occurred at 2:45 a.m. in response to a report of a stabbing involving a car of the same make and model as appellant’s car). At the time of the frisk, Johnson had made no furtive hand movements, and Officer Harvey did not notice any bulge or other indication that Johnson was in fact armed. Cf. In re D.E.W., 612 A.2d 194, 195, 198 (D.C.1992) (upholding a frisk where appellant unambiguously attempted to conceal a weapon). Here, the only basis for the stop was an expired inspection sticker, which did not give rise to any suspicion of criminal activity. Even Johnson’s driving without a license would not give rise to the suspicion that he was armed and dangerous — as the trial court recognized — and Officer Harvey testified that driving without a license is not a crime for which one can be arrested in Pennsylvania.
Although the trial court found that Johnson was reluctant to pull over, no testimony suggested that he attempted to flee, and Officer Harvey acknowledged that Johnson stopped within half a block. Cf. Powell, 649 A.2d at 1085 (opinion of Sullivan, J.) (concluding that the trial court’s ruling that the suspect’s reluctance to stop contributed to reasonable suspicion was erroneous where the suspect pulled over within three quarters of a block). Nor were Johnson’s attempts to exit the car, his ambiguous story about diapers,7 or his nervousness sufficient indicators that he was armed and dangerous.8 Officer Harvey testified that he did not know why Johnson was attempting to exit the car— *375but did not testify that his attempts led him to believe that Johnson was armed and dangerous. Although police officers concerned for their safety during a traffic stop may ask a driver to exit the car, Pennsylvania v. Mimms, 434 U.S. 106, 111, 98 S.Ct. 330, 54 L.Ed.2d 331 (1977) (per curiam), the police may not frisk the driver without specific, articulable suspicion. See United States v. Page, 298 A.2d 233, 237 (D.C.1972). Even when combined with Johnson’s nervousness and “hyperactivity,” these factors fail to indicate why the police felt he was armed and dangerous. See, e.g., Powell, 649 A.2d at 1087 (opinion of Sullivan, J.) (“[HJesitaney and nervousness may not be unusual when people are stopped.”) (quotation marks omitted); United States v. Salzano, 158 F.3d 1107, 1113 (10th Cir.1998) (“[Nervousness is of limited significance in determining reasonable suspicion ... because it is common for most people to exhibit signs of nervousness when confronted by a law enforcement officer whether or not the person is currently engaged in criminal activity.” (citations and quotation marks omitted)).
Like the trial court, the majority relies heavily on the fact that the car was in “try and locate” status and that, as a consequence, Officer Harvey believed that the car might in fact have been stolen. Putting aside that the car should no longer have been in “try and locate” status (Johnson’s girlfriend, who had initially reported the car overdue, had called police to report its return), nothing about “try and locate” status indicates that the operator of the car might be armed and dangerous. The record indicates that a car in “try and locate” status was a car that the owner had lent to a friend or relative and that had not been promptly returned. The owner did not want it reported stolen, though after ten days the car was supposed to be upgraded to “stolen” in the system. Officer Harvey testified that frequently cars that are due to be upgraded to “stolen” are not upgraded properly, and the majority contends that this gave Officer Harvey reason to believe that the car was in fact “stolen,” despite its “try and locate” status.9 Even assuming that the driver of the “try and locate” car had indeed borrowed the car for longer than the owner expected, and further assuming that the car was overdue to be upgraded to “stolen,” neither indicates that the car was somehow taken by force or that the driver might be presently armed and dangerous. To the contrary, it indicates that the car was willingly lent to the driver. Cf. Thomas v. United States, 553 A.2d 1206, 1206-08 (D.C.1989) (explaining that a ski mask could be indicative of the presence of weapons because ski masks are common in armed holdups).
The majority distorts the nature of “try and locate” status to get the officers over the hurdle of the need to establish reasonable, articulable suspicion. First, the majority seems to take solace in the fact that, had Johnson been stopped in the District of Columbia and found to have been engaged in the unauthorized use of a vehicle in violation of D.C.Code § 22-3215 (2001), the officers would have had probable cause to arrest him — and perforce to frisk him— because he would have been engaged in a felony. Maj. Op. 369 n. 4. Johnson was not stopped in the District of Columbia, however, and his conduct in Pennsylvania amounted to no more than a misdemeanor, *376see 18 Pa. Cons.Stat. § 1104(2) (1990), and the testimony at trial confirmed that the officers had no intent to arrest him.
The majority further hypothesizes a “potentially lengthy ... need for the police to keep [Johnson] in their presence while they investigated his identity and connection with the car.” Maj. Op. 869. This seems a bit of an expansion of Officer Harvey’s testimony, which was simply that when he encounters a car in “try and locate” status, he is trained to “remove the driver from the vehicle and recover the vehicle for the owner.” After removing the driver, he has to investigate the driver, meaning that he checks for open warrants and confirms whether the driver has a valid license. Driving a car in “try and locate” status is not a crime, and drivers are free to go after they are investigated. There is nothing in the record to suggest that the investigation need be particularly lengthy. And, absent more, such an investigation cannot, as a matter of course, include a frisk for weapons. The Supreme Court has emphasized that while officers face legitimate and weighty safety concerns when conducting traffic stops, those concerns do not, in and of themselves, create reasonable, articulable suspicion to frisk occupants of vehicles. See Knowles v. Iowa, 525 U.S. 113, 117, 119 S.Ct. 484, 142 L.Ed.2d 492 (1998) (“[W]hile the concern for officer safety in this context may justify the ‘minimal’ additional intrusion of ordering a driver and passengers out of the car, it does not by itself justify the often considerably greater intrusion attending a full field-type search.”). To protect themselves, officers may conduct a brief Terry frisk “wpon reasonable suspicion that [drivers or passengers] may be armed and dangerous.” Id. at 118, 119 S.Ct. 484 (emphasis added). Thus, the fact that officers were required to remove Johnson from the vehicle and conduct an investigation cannot justify a frisk, where nothing suggested that Johnson was armed and dangerous. See Powell, 649 A.2d at 1088 (opinion of Sullivan, J.) (noting that even Mimms does not allow for an “automatic frisk” of citizens pulled over for minor traffic violations).
Authorities relied on by the majority all exhibited a specific, articulable reason why the police thought that the suspect was armed and dangerous. In particular, the majority cites State v. Cowart, 862 So.2d 225, 231-32 (La.Ct.App.2003), where the suspect was pulled over for an expired inspection sticker, had no license, exited the vehicle as the officer approached, and was pacing around the vehicle with his hands near his waist. Although the facts are very similar to the instant case, one crucial fact sets it apart — in Cowart, the officer, who was alone, testified that he was unable to see what the suspect was holding because of the way he was holding his hands, and thus feared for his safety. Id. at 229, 231. Here, Officer Harvey pointed to nothing that actually made him (or would make a reasonable officer) believe Johnson might have been armed and dangerous. In fact, he testified that at the time he ordered the frisk, he did not believe Johnson was armed.
Even considering the totality of the circumstances, I would hold that there was no reasonable, articulable suspicion to frisk Johnson. Although the totality of the circumstances may have been puzzling or ambiguous — in fact they may have caused Officer Harvey to suspect that something was amiss — such “inchoate suspicion” is not sufficient to warrant a frisk. Terry, 392 U.S. at 27, 88 S.Ct. 1868; see also Singleton v. United States, 998 A.2d 295, 300-01 (D.C.2010) (“[T]o be articulable, there must be specific evidence — not merely conclusions — that led the officer to suspect criminal activity in a particular circumstance.”). The totality of the circum*377stances failed to indicate why a reasonable officer would suspect that Johnson was armed and dangerous. See In re R.M.C., 719 A.2d 491, 496 (D.C.1998) (holding that there was no reasonable, articulable suspicion for a frisk despite suspect’s walking unnaturally, clutching his rib cage, and acting nervously, because there was no evidence to indicate a weapon and no report of criminal activity outside of a curfew violation); Anderson v. United States, 658 A.2d 1036, 1040 (D.C.1995) (no reasonable, articulable suspicion despite suspect walking quickly away from officers, answering questions evasively, and putting his hands in his pockets despite being asked not to, because the officer “did not observe criminal activity, was not responding to a report of criminal activity, nor was he following up on an informant’s tip” and “[t]here was no bulge or object being concealed that the officer could think was a weapon”).
Because the various factors discussed above do not provide a reasonable, articu-lable suspicion that Johnson was armed and dangerous, I would hold that the trial court erred by denying Johnson’s motion to suppress the fruits of Johnson’s frisk.
III.
Because I believe that Officer Morales’s weapon and ammunition were erroneously admitted, I would reverse Johnson’s conviction unless the government can demonstrate that its admission was harmless beyond a reasonable doubt. See Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967). To meet this “exacting standard,” the “properly admitted evidence against the defendant must be ‘overwhelming,’” and “[t]he government must show that there is no ‘reasonable possibility that the evidence complained of might have contributed to the conviction.’ ” Ellis v. United States, 941 A.2d 1042, 1048-49 (D.C.2008) (quoting McCoy v. United States, 890 A.2d 204, 212 (D.C.2006); Chapman, 386 U.S. at 23, 87 S.Ct. 824). “Indeed, the ‘inquiry [under Chapman ] ... is not whether, in a trial that occurred without the error, a guilty verdict would surely have been rendered, but whether the guilty verdict actually rendered in this trial was surely unattributable to the error.’ ” Ellis, 941 A.2d at 1048-49 (quoting Sullivan v. Louisiana, 508 U.S. 275, 279, 113 S.Ct. 2078, 124 L.Ed.2d 182 (1993)) (emphasis in original).
Here, I would hold that the government cannot meet its burden. Officer Morales’s weapon and ammunition were crucial to the government’s case. Officer Morales was found with his holster empty, and as the prosecutor emphasized during closing argument, the logical conclusion was that the shooter had taken these items. Johnson’s possession of these items — just four days after the shooting — was highly incriminating and likely led the jury to conclude that he had shot Officer Morales.10 As the prosecutor argued to the jury, it was “no coincidence that Walter Johnson just happened to have that gun on him when he got stopped.”
Although the government produced other evidence — including many witnesses who placed Johnson at or near the scene of the shooting — I cannot conclude that the evidence was so “overwhelming” that the verdict here was not influenced by the weapon. None of the government’s witnesses actually witnessed the shooting, and the presence of the gun and ammunition— tangible evidence directly linking Johnson to the crime — likely relieved the jury of the difficult task of considering the accura*378cy of the various identifications. See Benn v. United States, 978 A.2d 1257, 1289 (D.C.2009) (“The identification of strangers is proverbially untrustworthy.”). Further, while both Ferguson and the Philadelphia victim provided compelling evidence against Johnson, our task is not to imagine whether the jury would have convicted Johnson absent the erroneously admitted evidence. Instead, our task is to consider whether or not the erroneously admitted evidence influenced the verdict here. I believe that Officer Morales’s weapon and ammunition were simply too damning for there to be “no reasonable possibility that [they] might have contributed to the conviction.” Ellis, 941 A.2d at 1049 (citation and internal quotation marks omitted).
IV.
Because the trial court erred in denying Johnson’s motion to suppress, and because the admission of Officer Morales’s gun and ammunition was not harmless beyond a reasonable doubt, I would reverse Johnson’s convictions.
. Because I would dispose of the appeal based on the suppression motion, I do not *372address whether or not the trial court abused its discretion in denying the motion for a new trial.
. People v. Batista, 88 N.Y.2d 650, 649 N.Y.S.2d 356, 672 N.E.2d 581, 583 (1996).
. Or.Rev.Stat. § 131.605(3) (2011).
. Webster’s Third New International Dictionary 912 (2002).
. Graham v. Connor, 490 U.S. 386, 396, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989); United States v. Sokolow, 490 U.S. 1, 7, 109 S.Ct. 1581, 104 L.Ed.2d 1 (1989).
. I would not hold that in all cases the inception of a frisk is an officer’s physical contact with a suspect. I am conscious of the potential to create a perverse incentive for officers to prolong the period of time between when they have ordered a suspect to assume a frisk position and when the officer actually contacts the suspect. Cf. United States v. Christian, 187 F.3d 663, 670 (D.C.Cir.1999) (assessing a search under Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), from the moment of the stop because "[o]therwise, we might create a perverse incentive for an arresting officer to prolong the period during which the arrestee is kept in an area where he could pose a danger to the officer” (internal quotation marks omitted)). I am also mindful that certain circumstances not present here may affect the point where police cross the line between "consent and coercion” so critical to Fourth Amendment analysis. Jackson v. United States, 805 A.2d 979, 987 (D.C.2002). My analysis is limited to the facts before us and only holds that, at the latest, the frisk here began when Officer Harvey manipulated Johnson's person.
. Officer Harvey testified that he doubted the story because he did not readily see any diapers in the car, but he also did not thoroughly examine the car and could not see into the trunk.
. Because the “reasonableness of official suspicion must be measured by what the officers knew before they conducted their search,” Florida v. J.L., 529 U.S. 266, 271, 120 S.Ct. 1375, 146 L.Ed.2d 254 (2000) (emphasis added), the fact that Johnson provided a fake name — which Officer Harvey clearly did not know at the time — cannot be considered in determining whether or not the frisk was supported by reasonable, articulable suspicion.
. Johnson's girlfriend put the car into “try and locate” status on either June 5th or June 6th, and Johnson was stopped on June 14th. Thus, even if she had not reported the car returned, the car was not due to be upgraded to "stolen” for at least another day. Officer Harvey’s reliance on the possibility that the car was "stolen” — because its status change might have been overdue — was thus factually incorrect.
. Although Johnson told the police he had purchased the gun, he refused to identify the seller, which the prosecutor argued provided a reason to doubt his story.