A jury in the Superior Court convicted the defendant of murder in the first degree on theories of deliberate premeditation and extreme atrocity or cruelty for the shooting death of Francis Stephens. The defendant was also convicted of the armed assault with intent to murder of Jose Astacio, who was shot but not killed, and possession of an unlicensed firearm. A codefendant, Willie Johnson, was acquitted on all charges.1 Represented by new counsel, the defendant filed a motion for a new trial on the ground of ineffective assistance of counsel. The motion was denied after an evidentiary hearing by the same judge who had presided at trial.
In this consolidated appeal, the defendant argues that he should be granted a new trial because his attorney was ineffective in failing to move to suppress an improperly suggestive and unreliable out-of-court identification, in failing to object to the admission of hearsay with the out-of-court identification, in failing to object to the prosecutor’s characterization of the out-of-court identification in closing argument, and in failing to introduce evidence of a third-party confession. In addition, the defendant claims that the judge erred in limiting the jury’s use of exculpatory evidence of third-party culprits, in admitting evidence of the defendant’s participation in drug dealing, and in *592failing to give an alibi instruction. The defendant also claims that he is entitled to reversal of the guilty verdict on the indictment charging armed assault with intent to murder because the evidence was insufficient as a matter of law to support a finding of guilt as a principal. We affirm the convictions and the denial of the motion for a new trial. After a complete review of the record, we also conclude that there is no basis to exercise our power under G. L. c. 278, § 33E, to reduce his murder conviction to a lesser degree of guilt or to order a new trial.
Background. We summarize the evidence in detail, considering it in the light most favorable to the Commonwealth, and reserving certain details for our analysis of the issues raised on appeal.
In September, 2000, the defendant and codefendant belonged to a gang from the Franklin Hill area in the Dorchester section of Boston known as the Franklin Hill Giants (Franklin Hill) that was engaged in escalating retaliatory violence with another neighborhood gang from the Esmond Street area (Esmond Street). After someone from Franklin Hill was stabbed while walking on Esmond Street on September 9, Kenie Smith, a senior member of Esmond Street, and Richard Green, a senior member of Franklin Hill, met to discuss the situation, in the presence of other gang members, including the defendant. After Green threatened Smith, and Smith made a move to his jacket, the defendant reached into his waistband and did something that sounded as though he were cocking a revolver. On September 12, an Esmond Street gang member was shot, and later that evening, a Franklin Hill gang member was shot. On September 16, at approximately 1 p.m., Smith was driving with three other Esmond Street gang members in a minivan when they spotted Green. Smith pulled in front of Green’s vehicle, and someone from his minivan fired four to five shots into the windshield of Green’s car, wounding Green.
Green was the “head man” of the Franklin Hill gang, supplied “crack” cocaine for sale by gang members, and determined which gang member was allowed to sell drugs on which neighborhood street. After the shooting of Green, several members of Franklin Hill, including the defendant, gathered at Akia Cheshire’s apartment, where the defendant, among others, lived. Later that afternoon, the defendant suggested to Shared Clark, *593another Franklin Hill gang member, that they should go to the Esmond Street area to “kill anybody over there.” Clark responded that he was “not in on this one.”
That afternoon, the defendant asked Terrance Dotson, who was friendly with Franklin Hill members, whether he would put gasoline in a Toyota Cressida automobile that Dotson had recently stolen and leave it for the defendant. Dotson followed his direction. At about 7 p.m., Dotson saw the defendant get into the Toyota automobile with Kyrone Childers.
Shortly thereafter, the defendant, wearing a black hooded sweatshirt, black gloves, and a black “skullie” cap, was driving the Toyota with the codefendant, Willie Johnson, now in the passenger seat, and stopped when he saw Clark, who was standing on the street with two companions. The defendant once more asked Clark to go “up the street with them.” This time, Clark and his two companions followed in Clark’s vehicle.2
Clark testified at trial that the two cars drove to Glenway Street, with the defendant’s car in the lead. They saw three men talking together near the comer of Glenway and Harlem Streets, in Esmond Street territory.3 The codefendant started firing shots at the men, and Clark’s car pulled in front of the Toyota to block traffic. One of the men ran into a nearby store. Another man (Astacio), after being shot in the leg, rolled under a parked van.4 The third man (Stephens) fell onto the sidewalk. Clark testified that the defendant stopped the car, got out, and walked toward the man lying on the sidewalk, firing downward multiple times with a nine millimeter firearm. Immediately after the shooting, the defendant drove the Toyota to the top of Fowler Street, a block away from where the shootings occurred. The defendant and codefendant abandoned the vehicle with the engine running. Clark told them to put the two firearms used in the shooting in his car, and he later hid them in a hamper at his aunt’s house. *594Clark sold one of the guns but retrieved it when the defendant demanded it back.5
At around 8:15 p.m., police officers and paramedics arrived on the scene and found Stephens lying face down on the sidewalk with multiple gunshot wounds to his head and torso that proved fatal.6 Astacio had been shot in the right thigh, but he recovered from his wounds. Police officers at the scene recovered twenty-six nine millimeter shell casings fired from two different nine millimeter firearms.7
*595Michael Boyd testified that approximately one week after the shooting of Green, while Boyd was in the custody of the Department of Youth Services, he spoke by telephone with the defendant, who was a friend. The defendant asked whether he could vote if he was “on the run” for a “body” or for “one of those things,” phrases that Boyd understood to mean that the defendant had killed someone. After Boyd’s release from the custody of the Department of Youth Services in February, 2001, while Boyd was living with the defendant and the defendant’s girl friend, the defendant told Boyd of his involvement in the shootings and gave a number of details: that the assaults were in revenge for the shooting of Green; that there was a meeting at Cheshire’s apartment after Green was shot; that he “sent a guy named Terrence” to get a car and provided money to put gasoline in it; that the defendant drove the car during the shooting and got out of the car; that nine millimeter guns were used in the shooting; that after the shooting the defendant gave the guns to Clark, but Clark sold one of the guns to pay off a debt, which caused a problem between the defendant and Clark because the defendant wanted the gun returned to him. The defendant also told Boyd that the person he shot lived in the Esmond Street area but was not an Esmond Street gang member. He added that this “didn’t matter because we had [a] beef with them anyway,” and “it was revenge.”
At the crime scene, a man who refused to identify himself told police officers that he witnessed the shooting while he was driving his children home. The individual was later identified by a police officer as Sylvester Harrison. Approximately five months after the shooting, on February 13, 2001, Detectives John Martel and Michael Primm of the Boston police department interviewed Harrison at his home and showed him an array of twelve photographs. Detective Martel testified that Harrison did not make a positive identification but picked out two individuals that “look like the two people that were in the *596vehicle.” Harrison told the detectives that he saw the person depicted in photograph number two (the defendant) operate a black Toyota, and “[h]e observed [this person] as being the shooter coming out of the vehicle and shooting at two individuals that were on the sidewalk.” Harrison said the person depicted in photograph number four (who was not the codefendant) was “the passenger of the vehicle.”
Harrison testified at trial that on September 16, 2000, he was driving on Page Street toward the intersection of Glenway and Harlem Streets, when he heard shots coming from Glenway Street.8 After hearing the shots, he drove in reverse on Page Street, away from Glenway Street, but, after taking a few turns, drove on Glenway Street past the intersection of Glenway, Page, and Harlem Streets, the very intersection from where the shots came. Driving past this intersection, Harrison saw someone “[o]n the ground . . . [b]y the store on Glenway.” After taking his first right on Fowler Street and dropping his children off at their home on that street, he spoke to police officers at the scene to tell them that a car, which he thought had crashed against another car, was blocking the middle of Fowler Street.9
Harrison testified that “[t]he bottom line to me is I didn’t see who shot the person. I never seen the people’s face.” Pointing at the two defendants, he said, “I never seen that man or that man before in my life.” Harrison said the police officers came to his house late one evening and showed him an array of photographs that they described as “a gang.” The officers pressured him to identify suspects from the array and urged him to “guess” whom he might have seen, even though police officers knew at the time that he was intoxicated. Much of Harrison’s testimony contradicted itself. He said he had not selected any photographs from the array but admitted that he had told the grand jury that he had, explaining that he selected two photographs from the array when he got “tired” of the police officers’ persistence. He said he “did not see their face” and that his “head was turned backwards,” but he conceded that he saw “a quick image.” He said he did *597not see anybody shoot, then admitted that he did see someone shooting “for a second” and that the man who he saw shooting was the same man who drove the car.10
The key witnesses against the defendant — Clark, Dotson, and Boyd — each arguably had received benefit or expected to benefit from their cooperation with the Commonwealth in the investigation. Clark first told police about his role in the shooting after he had been arrested while in possession of a nine millimeter semiautomatic firearm with a large capacity feeding device and faced prosecution as an armed career criminal. Clark’s armed career criminal charge was dismissed, and he was not prosecuted for his role in the shooting of Stephens and Astacio. Dotson had a pending armed robbery case when he testified before the grand jury, to which he later pleaded guilty and served eighteen months, and at the time of trial he was in jail awaiting trial on another pending criminal case.11 Boyd first spoke with the police about this case after he had been arrested on counterfeiting charges in November, 2003, while he was on probation, and he received financial assistance to move outside the Boston area because of his cooperation in this case.
After three days of deliberation, the jury sent a note to the judge that stated, in pertinent part, “We are deadlocked on our decision. Our decision is not unanimous, two individualfs] on opposite sides claim nothing will ever change their view.” The judge, after consulting with trial counsel, dismissed the jury for the day and on the next day delivered an instruction consistent with our holding in Commonwealth v. Rodriquez, 364 Mass. 87, 101-102 (1973) (Appendix), directing the jury to continue their deliberations. After returning to their deliberations, the jury *598asked if it were possible to see the notes of Detective Martel’s interview with Harrison in February, 2001. The judge told the jury that the report was not in evidence and they must rely on their collective memory of the testimony of Harrison and Detective Martel. The jury later asked for “further clarification” on the issues of “witness identification” and “collective memory and collective consciousness.” The judge responded by repeating his identification instruction to the jury. He added that their collective memory should govern their understanding of the evidence and that the jury’s collective memory need not be unanimous. He also told the jury that the law does not recognize the concept of “collective consciousness.” The jury returned their verdicts on December 9, 2005, after eight days of deliberation.
Discussion. 1. Ineffective assistance of counsel. Many of the defendant’s claims on appeal rest on the contention that he was denied the right to the effective assistance of counsel. Commonwealth v. Martinez, 425 Mass. 382, 387-388 (1997). Where, as here, the defendant has been convicted of murder in the first degree, “we consider his claim of ineffectiveness of counsel to determine whether there exists a substantial likelihood of a miscarriage of justice, as required under G. L. c. 278, § 33E, which is more favorable to a defendant than the constitutional standard for determining ineffectiveness of counsel.” Commonwealth v. Gonzalez, 443 Mass. 799, 808 (2005). Where defense counsel is claimed to be ineffective because he failed timely to object during the course of trial, we determine whether there was error and, if so, whether the error was “likely to have influenced the jury’s conclusion.” Id., quoting Commonwealth v. Wright, 411 Mass. 678, 682 (1992). “Under this more favorable standard of review, we consider a defendant’s claim even if the action by trial counsel does not constitute conduct ‘falling measurably below that ... of an ordinary fallible lawyer.’ ” Commonwealth v. Gonzalez, supra at 808-809, quoting Commonwealth v. MacKenzie, 413 Mass. 498, 517 (1992). Where defense counsel made a strategic or tactical decision that the defendant now challenges, we determine whether the decision was “manifestly unreasonable” when made, recognizing that “[mjany decisions of defense counsel that are characterized in hindsight as errors may have been reasonable tactical or strategic *599decisions when made.” Commonwealth v. Mosher, 455 Mass. 811, 827 (2010). See Commonwealth v. Boateng, 438 Mass. 498, 509 (2003).
a. Defense counsel’s failure to file a motion to suppress Harrison’s out-of-court identification. The defendant argues that Harrison’s out-of-court identification was so impermissibly suggestive and so unreliable that any competent attorney would have moved for its suppression. After an evidentiary hearing on the motion for a new trial, in which defendant’s trial counsel and Detectives Martel and Primm testified,12 the judge issued a 137-page memorandum of decision in which he concluded that a motion to suppress would have had only a “minimal chance of successfully suppressing Harrison’s identification.” In view of this minimal chance, the judge concluded that the defendant failed to prove that defense counsel’s decision not to file the motion was manifestly unreasonable when made.
In reviewing the judge’s denial of the defendant’s motion for a new trial, we first consider, as did the judge, whether a motion to suppress the identification likely would have been granted had it been filed. See Commonwealth v. Comita, 441 Mass. 86, 91 (2004) (where defendant claims ineffective assistance of counsel for failure to file motion to suppress, “defendant has to demonstrate a likelihood that the motion to suppress would have been successful”). Under art. 12 of the Massachusetts Declaration of Rights, an out-of-court eyewitness identification is not admissible where the defendant proves by a preponderance of the evidence, considering the totality of the circumstances, that the identification is so unnecessarily suggestive and conducive to irreparable misidentification that its admission would deprive the defendant of his right to due process. See Commonwealth v. Johnson, 420 Mass. 458, 463-464 (1995); Commonwealth v. Thornley, 406 Mass. 96, 98 (1989). See also Commonwealth v. Silva-Santiago, 453 Mass. 782, 794-795 (2009) (Silva-Santiago).13
Based on the evidence at the hearing on the motion for a new *600trial, as well as his knowledge of what occurred at trial, which included Harrison’s trial testimony, the judge found that the defendant failed to prove that Harrison was intoxicated during the identification procedure, that the police detectives pressured Harrison to make selections from the photographic array, or that the detectives told Harrison that the photographic array depicted members of a gang. The judge also found that there was nothing about the photographs in the array that singled out the defendant and that the detectives did not suggest or otherwise emphasize his photograph. Because we conclude, based on our independent review of the evidence, that the judge’s findings were not clearly erroneous, we accept these findings and agree that the defendant would not likely have been successful in proving that the identification procedure was unnecessarily suggestive.
We come to this conclusion cognizant that the detectives who conducted the identification procedure did not follow the protocol that we adopted in Silva-Santiago, supra at 798, for police to use “in the future” when conducting an identification procedure. The detectives failed to inform Harrison that the alleged wrongdoers may or may not be in the photographs depicted in the array, that it is just as important to clear a person from suspicion as to identify a person as the wrongdoer, that the individuals depicted in the photographs may not appear exactly as they did on the date of the incident because features such as weight and head and facial hair are subject to change, and that, regardless *601of whether an identification is made, the investigation will continue. See id. at 797-798. Nor did the police ask the witness to state how certain he was of his identification, see id. at 798, although Harrison nonetheless volunteered that he was not positive and that the photographs only “look like the two people that were in the vehicle.” Because the identification procedure with Harrison took place before we provided this guidance in Silva-Santiago, supra, the detectives cannot be faulted for failing to follow it. See id. (eyewitness identification admissible despite absence of protocol or comparable warnings to eyewitness).
The defendant asks that we revisit our conclusion in Silva-Santiago, supra at 798-799, that the choice of a simultaneous rather than a sequential display of photographs goes solely to the weight of the identification, not to its admissibility. We suggested in that case that we would revisit this conclusion if the substantial weight of additional empirical evidence demonstrates that “the sequential showing of photographs leads to greater accuracy” in eyewitness identifications. Id. at 798. The defendant contends that the empirical evidence is already clear that sequential display is the better procedure and that simultaneous display is inherently unnecessarily suggestive and conducive to irreparable mistaken identification.
Our review of the empirical research, as well as a survey of the empirical research conducted by a special master appointed by the Supreme Court of New Jersey,14 suggests that the rate of both accurate and inaccurate (i.e., true and false positive) identification is lower where eyewitnesses are asked to identify individuals through a sequential display rather than a simultaneous display. See Report of Special Master at 40 (June 18, 2010), State vs. Henderson, N.J. Supreme Court No. A-8-08 (Master’s Report). See also Steblay, Seventy-Two Tests of the Sequential Lineup Superiority Effect: A Meta-Analysis and Policy Discussion, 17 Psychol., Pub. Pol’y & L. 99, 123 (2011) (where culprit included in array, sequential lineups yield eight per cent fewer correct *602identifications than simultaneous lineups, but where culprit not included in array sequential lineups produce twenty-two per cent fewer errors). Because an eyewitness tends to assume that a suspect’s photograph is present in a simultaneous display, the eyewitness is more likely to identify one of the photographs in the display that looks most like the suspect, and therefore is more likely either to be correct or in error. See Wells, Eyewitness Identification: Systemic Reforms, 2006 Wis. L. Rev. 615, 618-619. However, “there is dispute as to the rate of reduction of accurate identifications as compared to the well-established rate of reduction in inaccurate identifications.” Master’s Report, supra. There is also dispute as to whether a sequential lineup may be preferable under some conditions and a simultaneous lineup preferable in others. Three experts who have surveyed the empirical studies in eyewitness identification have suggested that sequential lineups are superior to simultaneous lineups only when a perpetrator is absent from the array, and that “[sequential lineups . . . produce fewer correct identifications [than simultaneous lineups] when the level of similarity [between suspect and fillers] is high.” McQuiston-Surrett, Sequential vs. Simultaneous Lineups: A Review of Methods, Data, and Theory, 12 Psychol., Pub. Pol’y, & L. 137, 154 (2006).
In addition, we also recognize that the failure to provide warnings comparable to the protocol we adopted in Silva-Santiago, supra at 797-798, including warning the eyewitness that the alleged wrongdoer may not be among the photographs in the array and that it is as important to clear a person from suspicion as to identify a person as the wrongdoer, “substantially increases risk of misidentification.” Master’s Report, supra at 22. What is not clear from the studies is whether, and in what circumstances, the use of the protocol in a simultaneous photographic lineup diminishes the risk of false positive identification to a rate comparable to or less than that in a sequential lineup. We cannot determine whether a sequential display is superior to a simultaneous display and that the use of the latter is unnecessarily suggestive until we learn, at a minimum, whether the rate of false positive identification with the use of the protocol is significantly higher in simultaneous displays than in sequential displays.
Therefore, we believe it is still too soon to conclude that *603sequential display is so plainly superior that any identification arising from a simultaneous display is unnecessarily suggestive and therefore must be suppressed. See, e.g., State v. Henderson, 208 N.J. 208, 257-258 (2011) [slip op. at 30] (“For now, there is insufficient, authoritative evidence accepted by scientific experts for a court to make a finding in favor of either [simultaneous or sequential lineup] procedure”); McQuiston-Surrett, Sequential vs. Simultaneous Lineups, supra at 162 (“research base for [sequential lineups] may not be sufficiently developed from a methodological or theoretical point of view to currently advocate for its implementation to the exclusion of other procedures”). Until we reach such a conclusion, the choice of a simultaneous rather than a sequential display of photographs shall go solely to the weight of the identification, not to its admissibility. Silva-Santiago, supra at 798-799.
The defendant also contends that the photographic array was “grossly improper” because it was composed entirely of suspects. The defendant argues that scientific and legal studies suggest that an all-suspect array “produces sharply inflated false identification rates compared with the single-suspect model.” See, e.g., Master’s Report, supra at 25 (“ordinary and accepted practice among law enforcement agencies is to present an array embedding the suspect among at least five fillers”); United States Department of Justice, Eyewitness Evidence: A Guide for Law Enforcement 29 (1999) (recommending that police “[i]nclude only one suspect per identification procedure”); Wells, Eyewitness Identification: The Importance of Lineup Models, 99 Psychol. Bull. 320 (1986).
We are not convinced that the rate of false positive identification is greater with all-suspect arrays, but we are persuaded that the danger that a false positive identification will result in a wrongful prosecution is greater with an all-suspect array, because an eyewitness who is mistaken in an identification will nonetheless point to a suspect in the case and the police will not know that the identification is mistaken. See id. at 322 (“there is no response that the eyewitness can make that can be classified as a known error in actual cases that use the all-suspect model”). Where the array consists of only one suspect and fillers, a mistaken identification will be apparent, because the filler has *604no connection to the investigation. Therefore, an all-suspect array significantly and needlessly increases the potentially unjust consequences that may arise from a false positive identification. Unless there are exigent or extraordinary circumstances, the police should not show an eyewitness a photographic array, whether simultaneous or sequential, that contains fewer than five fillers for every suspect photograph. See State v. Henderson, supra at [26] (“lineups should include a minimum number of fillers . . . [and] there appears to be general agreement that a minimum of five fillers should be used”). We expect police to follow our guidance to avoid this needless risk.15 Because this is new guidance, we do not address in this opinion whether the failure to follow it will make the identification inadmissible in evidence. Cf. Silva-Santiago, supra at 798.16
We do not conclude that a substantial likelihood of a miscarriage of justice arose from the use of an all-suspect array in this case. The photograph that Harrison identified as the man in the Toyota automobile was not the codefendant; but the codefend-ant, and not the man identified, was the person charged by the Commonwealth, which suggests that the police did not lock onto a suspect based on Harrison’s identification. Moreover, the heart of the Commonwealth’s case against Walker was the testimony of Clark, Dotson, and Boyd, not Harrison’s equivocal and retracted prior identification.17
Finally, the defendant urges that we revisit our jurisprudence under art. 12 of the Massachusetts Declaration of Rights and *605common-law principles of fairness, and require a trial judge to serve as a gatekeeper to exclude unreliable eyewitness testimony even where a defendant is unable to prove by a preponderance of the evidence that the identification is unnecessarily suggestive and conducive to irreparable misidentification. In Commonwealth v. Jones, 423 Mass. 99 (1996), we recognized that common-law principles of fairness dictate that an unreliable identification arising from “especially suggestive circumstances” should not be admitted in evidence even where the police were not responsible for the suggestive confrontation. Id. at 108-109. See Commonwealth v. Odware, 429 Mass. 231, 236 (1999) (“in some circumstances an identification that has been tainted, but not by the government, may become so unreliable that its introduction into evidence is unfair”). We have not suggested, however, that in the absence of an unnecessarily suggestive police identification procedure or especially suggestive circumstances, a judge must serve as a gatekeeper of eyewitness identifications offered by the Commonwealth and admit only those identifications the judge finds to be reliable. Two years after we decided the Jones case, we rejected the argument that nonsuggestive pretrial identifications should be suppressed on the ground that they were unreliable, declaring, “When the procedures are not suggestive, the pretrial identifications are admissible without a further showing.” Commonwealth v. Payne, 426 Mass. 692, 694 n.3 (1998), quoting Commonwealth v. Warren, 403 Mass. 137, 139 (1988).
We recognize that, in State v. Henderson, supra, the Supreme Court of New Jersey recently concluded that, where a defendant presents some evidence of suggestiveness that could lead to a mistaken eyewitness identification, the defendant is entitled to a pretrial hearing where the judge will consider all of the factors relevant to the reliability of the identification, including both “[s]ystem [vjariables” (defined as “variables within the State’s control,” such as the pretrial identification procedure, id. at 248) and “[e]stimator variables” (defined as “factors beyond the control of the criminal justice system,” such as factors related to the incident, the witness, or the perpetrator, id. at 261). The court declared, “[I]f after weighing the evidence presented a court finds from the totality of the circumstances that defendant has demonstrated a very substantial likelihood of *606irreparable misidentification, the court should suppress the identification evidence.” Id. at 289. The court noted that, under its previous framework, a defendant needed to prove that police procedures were “impermissibly suggestive” before a motion judge could consider estimator variables that bear on reliability.18 Id. at 286. Now, where a defendant has met the initial burden of showing some evidence of suggestiveness, a motion judge may consider all evidence relevant to reliability, including both system and estimator variables, in determining whether the defendant has proved a very substantial likelihood of irreparable mis-identification. Id. at 289. See State v. Hubbard, 48 P.3d 953, 963 (Utah 2002) (“Even if law enforcement procedures are appropriate and do not violate due process, eyewitness identification testimony must still pass the gatekeeping function of the trial court and be subject to a preliminary determination — whether the identification is sufficiently reliable to be presented to the jury”).
Our convening of a study committee on eyewitness identification reflects our willingness to revisit our jurisprudence regarding such evidence, see note 16, supra, and to consider, among other alternatives, the approach established in New Jersey, but we will not do so in a case where the defendant did not move to suppress the identification and where the issue is whether the defendant’s attorney was ineffective in failing to do so. Where, as here, the pretrial identification was not obtained through an unnecessarily suggestive police identification procedure or especially suggestive circumstances, and where the alleged unreliability of the eyewitness identification arose from distance, lighting, the brevity of the observation, and the emotional state of the eyewitness at the time of the observation, there was no substantial likelihood of a miscarriage of justice from the admission of the eyewitness testimony. See Commonwealth v. Odware, supra at 236; Commonwealth v. Payne, supra. The jury were capable of making an informed assessment of the accuracy of *607such an identification and assessing its weight, aided by vigorous cross-examination and the judge’s instructions. See Commonwealth v. Odware, supra; Commonwealth v. Jones, supra at 110. See also Commonwealth v. Vardinski, 438 Mass. 444, 450-451 (2003) (“It is crucial to the fact finder’s assessment of the truth to allow the defendant to probe fully on cross-examination the infirmities of the [eyewitness] identification”).19
For these reasons, we conclude that defense counsel was not ineffective for failing to move to suppress Harrison’s pretrial identification, and its admission did not result in a substantial likelihood of a miscarriage of justice.20
b. Defense counsel’s failure to object to hearsay in Harrison’s out-of-court identification. The defendant also claims that defense counsel rendered ineffective assistance because she failed to object to Detective Martel’s testimony that Harrison said he saw the person depicted in photograph number two (the defendant) operate a black Toyota and “[h]e observed [this person] as being the shooter coming out of the vehicle and shooting at two individuals that were on the sidewalk.” The defendant contends this was hearsay that went beyond the permissible scope of an identification. We conclude there was no error in the admission of this testimony.
We allow the admission in evidence of a pretrial identification even where the identifying witness denies or does not remember having made such an identification, provided the identifying witness testifies at trial and is subject to cross-examination concerning the pretrial identification, and we do not limit the use of *608such evidence to impeachment. Commonwealth v. Cong Due Le, 444 Mass. 431, 441-442 (2005). The witness in the Cong Due Le case, during a pretrial identification procedure, had identified the photograph of one codefendant as the person who had attacked his friend, and the photograph of the other codefendant as the person who had struck him and knocked him to the ground. Id. at 433-434. At trial, however, the witness admitted having identified the photographs of the two codefendants but claimed that he identified them as persons he had seen outside the nightclub on the evening of the assault, not as the assailants. Id. at 435. We concluded that the admission of the detective’s testimony regarding “the contents of [the witness’s] statement of identification” was proper under our common-law rules of evidence and did not violate the defendants’ right of confrontation. Id. at 442.
The defendant contends that Harrison’s out-of-court statements went beyond a statement of identification and into the realm of inadmissible hearsay. We have recognized that “[a]bsent context, an act or statement of identification is meaningless. . . . [Identification evidence must be accompanied either by some form of accusation relevant to the issue before the court, or some form of exclusionary statement, in order to be relevant to the case.” Commonwealth v. Adams, 458 Mass. 766, 772 (2011).21 We have concluded that an eyewitness’s out-of-court statement identifying a defendant as the person shooting at the eyewitness’s friend is part of the context of the identification, see id. at 770-771, but a statement regarding the number of shots fired, the color of the firearm, and the defendant’s behavior after the shooting goes beyond the context of the identification of the shooter. See Commonwealth v. Martinez, 431 Mass. 168, 175-176 (2000).
The judge, in denying the defendant’s motion for a new trial, *609concluded that Harrison’s statement that the person in photograph number two looked like the person who operated the car and then fired shots at two individuals on the sidewalk was “limited context integral to the identification” and therefore properly admitted. The judge also concluded that Harrison’s description of the car as a “black Toyota” was “extraneous to the identification,” and was inadmissible hearsay, but was harmless in the circumstances in view of abundant other evidence that the car involved in the shooting was a black Toyota. We agree. Because an objection by defense counsel to this testimony would have been overruled except to exclude the statement regarding the black Toyota, and because the admission of that statement was harmless, we conclude that defense counsel was not ineffective for failing to object.
*608“The extent of the statement needed to provide context will vary from case to case, depending on what the witness remembers or denies at trial. We emphasize that the rule was not intended to render a witness’s entire statement admissible, but only so much as comprises relevant evidence on the issue of identification. Judges have broad discretion in this area, and parties who intend to offer pretrial statements of identification are well advised to bring the matter to the attention of the trial judge at the earliest practicable time, preferably in a motion in limine.”
*609c. Defense counsel’s failure to object to the prosecutor’s closing argument. In his closing argument, the prosecutor told the jury that Harrison had “a good look” at “the driver, the person doing the shooting,” and that Harrison “tells us” that the face of the driver looked like the face of the defendant. The defendant maintains the prosecutor’s “glaring misstatements” went to the “heart of the case,” and that “defense counsel’s failure to object to the statements was constitutionally ineffective.”
In denying the motion for a new trial, the judge determined that the statement about a “good look” was “proper argument based on the evidence before the jury and permissible inferences drawn therefrom.” The judge recognized that Harrison did not testify at trial that the face of the shooter looked like the face of the defendant, but instead said that to Detective Martel, who reported it to the jury in his testimony. The judge concluded that the prosecutor’s statement was not improper because, despite “the prosecutor’s inartful phraseology,” a reasonable jury would have understood that the prosecutor was referring to the pretrial identification of the photographs made by Harrison, not to his trial testimony.
Because the judge would have overruled defense counsel’s objections to these closing argument statements and was within his discretion in doing so, we conclude that defense counsel’s failure to object did not constitute ineffective assistance of counsel. See Commonwealth v. Raymond, 424 Mass. 382, 391 *610(1997) (prosecutor’s statements must be considered in context in which they were made to determine what jury likely would have understood them to mean).
d. Defense counsel’s failure to introduce evidence of third-party confession. In her cross-examination of Dotson, defense counsel elicited from him that he had stated to Detective Michael Devane in October, 2000, that one of Green’s brothers had shot the victim in the chest and another brother had “finished him in the head.” He testified “that was the word” he had heard. On redirect examination, Dotson testified that this was the word “that was going around the hood,” and that he had passed on the rumor to the police, having no firsthand information. In her cross-examination of Detective Devane, defense counsel elicited that Dotson had told him that two of Green’s brothers shot the victim. However, on redirect examination, when the prosecutor asked the detective if he recalled “all of what Mr. Dotson had to say” regarding the Green brothers’ involvement in the shooting, the detective testified that Dotson had recanted that statement and admitted it was not true. He said Dotson “changed it to something he heard and then backed off even hearing that.” The defendant asserts that he was denied effective assistance of counsel because his trial attorney did not attempt to elicit from Dotson or Detective Devane the information contained in the detective’s report of this interview that Dotson had heard this information from the Green brothers.
We agree with the judge that it was not manifestly unreasonable for defense counsel to fail to attempt to elicit this information. In the police report, Detective Devane wrote that Dotson initially stated that he heard this information from the Green brothers during a conversation with them, then said he overheard a conversation between the Greens, then denied hearing any conversation, and then admitted that his statement that the Green brothers committed the shooting was not true. If the defendant’s attorney had elicited from Dotson or Detective Devane that Dotson had initially told the detective that he heard the Greens admit to this crime, the prosecutor would have elicited from Dotson or Detective Devane that Dotson immediately recanted this information and admitted it was not true. By not pressing this point, defense counsel was able to leave the jury with the *611impression that Dotson had said the word on the street in the neighborhood was that the Green brothers had committed the shooting. If she had pressed the point, she risked eliciting testimony that Dotson had admitted even that was not true. See Commonwealth v. Pilial, 445 Mass. 175, 185-187 (2005) (not manifestly unreasonable for defense counsel to fail to impeach sexual assault victims with information suggesting they fabricated allegations against friend’s father for petty reasons); Commonwealth v. White, 409 Mass. 266, 273 (1991) (reasonable tactical decision not to call witness who would have testified that complainant changed story where witness would also have testified that defendant changed story).22
Having addressed each of the claims of ineffective assistance of counsel raised by the defendant in his motion for a new trial, we affirm the denial of the motion and move to the other claims raised by the defendant.
2. Limitation on use of exculpatory evidence. Through co-defendant counsel’s cross-examination of Boston police Sergeant Detective James J. Wyse, the jury learned that an “unknown black male” had approached a police officer who had responded to the crime scene and told him that two Hispanic males had left the suspects’ car, “possibly” dropped something near a Cadillac up the street, and fled in a Honda or Toyota automobile.28 The judge, without objection, instructed the jury that they may not consider this evidence for the truth of the matters asserted in the statement but may consider it only “for purposes of your assessment of the police investigation.” The defendant argues that the judge erred in providing this limiting instruction and should have permitted the jury to consider this statement for the troth of the matters asserted. Because the defendant did not object to the limiting instruction, we determine whether this limitation was error that resulted in a substantial likelihood of a miscarriage of justice.24
Where evidence is offered for its truth to prove that a third *612party is the true culprit, “we have permitted hearsay evidence that does not fall within a hearsay exception only if, in the judge’s discretion, ‘the evidence is otherwise relevant, will not tend to prejudice or confuse the jury, and there are other “substantial connecting links” to the crime.’ ” Silva-Santiago, supra at 801, quoting Commonwealth v. Rice, 441 Mass. 291, 305 (2004). Here, it is not apparent from the statement whether the “unknown black male” personally observed the crime or obtained the information from a third party and, if he did observe it, whether he saw the perpetrators well enough to identify them as Hispanics. Nor was there evidence of the emotional state of the declarant that may have permitted the statement’s admission as an excited utterance. The judge could reasonably conclude that such unreliable testimony would tend to prejudice or confuse the jury if admitted for its truth. In these circumstances, we conclude that the judge acted within his discretion in limiting the use of this evidence, and that the limitation did not create a substantial likelihood of a miscarriage of justice.
3. Evidence of drug dealing. The defendant claims that the judge erred by permitting the Commonwealth to introduce “entirely tangential” and highly inflammatory evidence that the defendant sold crack cocaine. Over the defendant’s objection, Clark testified that the defendant was involved in the sale of crack cocaine from Akia Cheshire’s apartment, but he never saw the defendant make any drug transactions or handle any narcotics. In addition, apart from evidence about Green’s drug dealing, several police officers and civilian witnesses testified to the prevalence of drug dealing among members of the Franklin Hill gang. We review for prejudicial error.
While evidence of prior bad acts, such as drug dealing, may not be introduced for the purpose of showing the accused’s *613propensity to commit the crime charged, evidence of prior bad acts may be admitted where it is not unduly prejudicial and is relevant “to establish motive, opportunity, intent, preparation, plan, knowledge, identity, or pattern of operation.” Commonwealth v. Horton, 434 Mass. 823, 827 (2001). We leave to the judge’s sound discretion whether the probative value of the evidence outweighs the risk of unfair prejudice. Id. at 828. We conclude that the judge did not abuse his discretion by admitting this testimony.
The Commonwealth’s theory, as the prosecutor stated in his closing argument, was that the shooting of the victim and Asta-do appears senseless unless one understands its context. The jury heard evidence from which they could infer that the viciousness of the shooting was motivated by rage that Green, a key source of drugs and therefore of revenue for the Franklin Hill gang, had been shot. The defendant’s participation in the gang’s drug dealing provided motive for his participation in the shooting. The judge’s forceful admonitions to the jury reasonably ensured that the jury considered the drug testimony for the limited purpose of establishing motive and not for the forbidden purpose of inferring the defendant’s bad character. See Commonwealth v. O’Laughlin, 446 Mass. 188, 208-209 (2006) (no unfair prejudice to defendant where judge gave limiting instructions on evidence of prior drug abuse relevant to motive). Because testimony about drug dealing was relevant to the defendant’s motive and the judge gave forceful hmiting instructions, we conclude that the judge did not abuse his discretion in finding that the probative value of the testimony outweighed the risk of unfair prejudice.25
4. Denial of request for an alibi instruction. The defendant *614maintains the judge’s denial of defense counsel’s request for an alibi instruction violated his constitutional right to a fair trial. Because the issue was preserved for appeal, we review for prejudicial error, and we find none.
While a judge may choose to give an alibi instruction, it is well settled that an “alibi instruction is not required where the charge as a whole makes clear that the Commonwealth must prove beyond a reasonable doubt that the defendant committed the crime for which he was indicted.” Commonwealth v. Thomas, 439 Mass. 362, 371 (2003). See Commonwealth v. Medina, 380 Mass. 565, 579 (1980) (not error to omit alibi instruction where judge “otherwise made clear that the burden of showing that the defendant was present at the time and place, and thus capable of committing the crime, remains on the Commonwealth”). Here, the judge repeatedly emphasized throughout trial, and forcefully instructed in his charge to the jury, that the Commonwealth alone bore the burden of proving every element of each crime charged beyond a reasonable doubt.26
5. Sufficiency of the evidence as to the charge of armed assault with intent to murder. On the charge of armed assault with intent to murder, the Commonwealth proceeded solely on an attempted battery theory of assault,27 with the defendant as principal, and the judge instructed the jury accordingly.28 The *615defendant argues that the evidence was insufficient to convict him as a principal of armed assault with intent to murder Astacio because “[tjhere was no evidence presented that Astacio was ever even close to [the defendant’s] line of fire.” In reviewing the judge’s denial of a motion for a required finding, we determine “whether the evidence offered by the Commonwealth, together with reasonable inferences therefrom, when viewed in its light most favorable to the Commonwealth, was sufficient to persuade a rational jury beyond a reasonable doubt of the existence of every element of the crime charged.” Commonwealth v. Lao, 443 Mass. 770, 779 (2005), S.C., 450 Mass. 215 (2007), and ante 12 (2011), quoting Commonwealth v. Campbell, 378 Mass. 680, 686 (1979).
“Under the attempted battery theory, the Commonwealth must prove that the defendant intended to commit a battery, took some overt step toward accomplishing that intended battery, and came reasonably close to doing so.” Commonwealth v. Melton, 436 Mass. 291, 295 (2002). Here, the evidence, when viewed in the light most favorable to the Commonwealth, established that on the evening of September 16, 2000, the defendant armed himself with a nine millimeter semiautomatic weapon and drove in a stolen automobile to territory controlled by Es-mond Street with the intent to “kill anybody over there” in revenge for the shooting of Green. At the intersection of Glenway, Harlem, and Page Streets, the passenger in the defendant’s car fired the first shots, a wounded Astacio rolled under a van, and the defendant then left the car and began shooting multiple times in a downward direction on Glenway Street. Based on the ballistics evidence, a reasonable jury could infer that at least one shot from each of the two firearms used by the shooters was fired near the van under which Astacio rolled, which suggests that each of the shooters attempted to shoot Astacio after he was wounded.29 This evidence was sufficient to permit a reasonable jury to conclude *616beyond a reasonable doubt that the defendant intended to commit a battery on Astacio, took overt steps toward accomplishing that aim, and came “reasonably close” to doing so. See Commonwealth v. Melton, supra.
6. Review under G. L. c. 278, § 33E. Having thoroughly examined the record, we see no reason to exercise our authority under G. L. c. 278, § 33E, to reduce the murder in the first degree conviction to murder in the second degree, or to order a new trial.
7. Conclusion. For the reasons stated above, we affirm the defendant’s judgments of conviction, and the denial of his motion for a new trial.
So ordered.
Willie Johnson was charged with murder in the first degree, assault and battery by means of a dangerous weapon, and possession of an unlicensed firearm.
At the time of trial, one of Shared Clark’s companions was deceased. The other, Christopher Robinson, testified that he did not recall ever being in a vehicle with Clark on the day of the murder.
The transcript refers to Harvard Street, but according to the map in evidence, the street that intersects Glenway Street is Harlem Street. We infer that the reference to Harvard Street is a stenographic error.
Jose Astacio asserted his right against self-incrimination pursuant to the Fifth Amendment to the United States Constitution and did not testify at trial.
The trial testimony of two nonparticipant eyewitnesses corroborated Clark’s description of the shooting. Jonathan Baskin testified that at approximately 8 p.m. he heard a loud sound and looked out of the third-story window of his home near the intersection of Glenway and Harlem Streets. He “saw a gentleman standing and firing a gun . . . in a downward direction” on the sidewalk on Glenway Street near Harlem Street. Baskin described the man as a dark-skinned African-American male of “average height,” which he set variously as between five feet, six inches, to five feet, ten inches tall. After the shooting, he saw the man jump into the passenger side of a car that “took off up Glen-way Street.” Baskin did not see the face of the man. At trial, he identified a photograph of a dark-colored Toyota Cressida as the car used by the shooter.
Glennis Pena testified that, while she was on her second-story porch on Fowler Street, which intersects Glenway Street, a block away from the intersection where the shooting occurred, she witnessed a car going “very fast” up the street. The car stopped further up Fowler Street, and two people got out and entered a car that was “coming from behind.” She did not recognize the occupants of the cars.
Dr. Mark Flomenbaum, the chief medical examiner, testified at trial about the wounds to Stephens. Dr. Flomenbaum did not perform the autopsy or write the autopsy report but testified because the medical examiner who had done so was no longer with the medical examiner’s office. Without objection, Dr. Flomenbaum testified that, according to the autopsy report, the victim had been shot fifteen times; the injuries caused by eight of the bullets that struck him would independently have been fatal. Three shots were fired to the back of Stephens’s head, and another was consistent with having struck Stephens when he was “belly down.” Three pages of the autopsy report were admitted in evidence, also without objection. While this testimony and the autopsy report were inadmissible hearsay whose admission violated the defendant’s right of confrontation under the Sixth Amendment to the United States Constitution, see Commonwealth v. Barbosa, 457 Mass. 773, 784 (2010), cert. denied, 131 S. Ct. 2441 (2011), the defendant does not contend on appeal that defense counsel was ineffective for failing to object to their admission or that their admission produced a substantial likelihood of a miscarriage of justice. We independently consider this error under G. L. c. 278, § 33E, and conclude that no substantial likelihood of a miscarriage of justice resulted from its admission, because there was abundant other reliable evidence that the shooter repeatedly fired at Stephens while he lay helplessly wounded on the sidewalk.
The police also recovered a light-colored T-shirt from the abandoned *595Toyota and a Red Sox baseball cap found under a Cadillac automobile that was parked near the Toyota. Deoxyribonucleic acid samples from the baseball cap did not match that of either defendant, nor did fingerprints found in the interior of the Toyota and on its registration plate. The T-shirt was sent to the crime laboratory but there was no testimony about any results of forensic testing or whether such testing was done.
Page Street becomes Harlem Street at the intersection with Glenway Street.
Sylvester Harrison could not identify the photograph of the black Toyota as the car he saw blocking Fowler Street that night. Clark testified that the Toyota had front end damage before it was used in the shooting.
When asked to estimate the length of Page Street, Harrison replied that it was probably a distance of about “three quarters of a football field,” the equivalent of 225 feet, and that he was at the beginning of the street when he heard the shots. Page Street begins at McLellan Street, from which Harrison entered Page Street, and ends at Glenway Street. The defendant contends from this testimony that Harrison was 225 feet away from Glenway Street when he saw the shooter, but this ignores Harrison’s testimony that he drove in reverse on Page Street when he realized the shots were coming from Glenway Street, which suggests that he had driven some distance toward Glenway Street before he decided to go in reverse.
Terrence Dotson denied that he had been promised anything regarding his armed robbery case or his pending case.
Harrison died before the evidentiary hearing on the motion.
Where the defendant satisfies this burden, the out-of-court identification is per se excluded as a violation of the defendant’s right to due process under art. 12 of the Massachusetts Declaration of Rights. Commonwealth v. Johnson, 420 Mass. 458, 462-463 (1995). In contrast, under the Fourteenth Amendment *600to the United States Constitution, a motion judge must apply a two-step analysis to the question of admissibility. The judge asks first whether the eyewitness identification was obtained by a police procedure that was unnecessarily suggestive. See Manson v. Brathwaite, 432 U.S. 98, 110 (1977). If it was, the judge then asks whether, notwithstanding the unnecessarily suggestive procedure, the eyewitness identification was reliable under “the totality of the circumstances.” Id. at 113. Because “reliability is the linchpin,” the identification, if found reliable, is admissible even where obtained through an unnecessarily suggestive procedure. Id. at 114. The United States Supreme Court expressly rejected the “per se” rule of exclusion as “go[ing] too far since its application automatically and peremptorily, and without consideration of alleviating factors, keeps evidence from the jury that is reliable and relevant.” Id. at 112. Because the standard for the admissibility of an identification under the Massachusetts Constitution is more favorable to a defendant than the standard under the United States Constitution, we need not consider the defendant’s Federal constitutional claims concerning Harrison’s out-of-court identification.
The Supreme Court of New Jersey appointed a special master to conduct a “plenary hearing” to examine the scientific evidence regarding eyewitness identification. See Report of Special Master at 3 (June 18, 2010), State vs. Henderson, N.J. Supreme Court No. A-8-08. At the hearing, more than 200 published scientific studies, articles, and books were made part of the record, and seven expert witnesses on eyewitness identification testified. Id.
We urge district attorneys to provide training to police in proper identification procedures to ensure that the risk of a wrongful conviction arising from a mistaken identification is diminished.
Because eyewitness identification is the greatest source of wrongful convictions but also an invaluable law enforcement tool in obtaining accurate convictions, and because the research regarding eyewitness identification procedures is complex and evolving, we shall convene a study committee to consider how we can best deter unnecessarily suggestive procedures and whether existing model jury instructions provide adequate guidance to juries in evaluating eyewitness testimony.
The defendant’s additional alleged constitutional bases for suppression are groundless and merit only brief comment. That a police detective who was investigating the case composed and presented the array (that is, that the identification procedure was not “double-blind”) and that the interview was not recorded do not without more render the identification evidence unduly suggestive. See Commonwealth v. Silva-Santiago, 453 Mass. 782, 797, 799 (2009).
The Supreme Court of New Jersey, in contrast with this court, had earlier adopted the two-step analysis in Manson v. Braithwaite, 432 U.S. 98, 110 (1977), in which a judge first determines whether an identification procedure was impermissibly suggestive and, if so, then considers in the totality of the circumstances whether the identification was nonetheless reliable. State v. Henderson, 208 N.J. 208, 237 (2011).
The defendant claims that Harrison’s identification is unreliable because he was too distant from the men he observed, he saw only a “quick image,” he was focused more on the safety of his children than the identity of the shooter, and his pretrial identification occurred five months after the incident. The defendant’s cross-examination and the judge’s instructions to the jury were adequate to address the danger of unreliability arising from these sources.
Because we conclude that there was no reasonable likelihood that Harrison’s pretrial identification would have been suppressed had defense counsel so moved, we need not address whether it was manifestly unreasonable for defense counsel to make the strategic decision not to challenge its admissibility and use Harrison’s testimony to argue that the police were overzealous in their investigation of the murder by pressuring him to make a false identification. See Commonwealth v. Acevedo, 446 Mass. 435, 446-447 (2006) (defense counsel ineffective for strategic decision only where decision was manifestly unreasonable when made).
We added in Commonwealth v. Adams, 458 Mass. 766, 772 (2011):
In view of the conclusion that defense counsel’s actions were not manifestly unreasonable, we need not determine whether the Green brothers’ purported admission to the crime would have been admissible in evidence.
Neither the defendant nor the codefendant is Hispanic. Both are African-American.
The defendant initially sought to admit this out-of-court statement through *612the testimony of Officer Brian Mahoney, who spoke to the unknown black male. The judge denied its admission on the ground that it was not an excited utterance. The defendant at that time did not seek its admission for the limited purpose of evaluating the adequacy of the police investigation. When the defendant proposed its admission for this limited purpose, the judge allowed the admission of the evidence through the testimony of Boston police Sergeant Detective James J. Wyse, who testified to the contents of Officer Mahoney’s report of the interview. Because the defendant does not contend on appeal that the statement was admissible as an excited utterance, we do not consider the objection preserved, but the standard of review is irrelevant here because we find no error by the judge.
Our confidence that the defendant did not suffer unfair prejudice from the admission in evidence of the defendant’s participation in drug dealing is strengthened by the fact that there was far more extensive evidence regarding his codefendant’s drug dealing, and the jury found the codefendant not guilty. Clark testified that he had seen the codefendant selling illegal drugs approximately twice a week through the summer of 2000. Trooper Patricia Riley testified that she personally observed the codefendant “multiple times” in the summer of 2000 while conducting physical surveillance of the sale of crack cocaine in the Franklin Hill housing complex. Not surprisingly, it was codefendant’s counsel, not the defendant’s, who took the lead in objecting to testimony about illegal drug dealing. If any unfair prejudice attached to the evidence of *614drug dealing, it was the codefendant, not the defendant, who would have borne its brunt.
We add that the alibi defense in this case was weak. The evidence established that the shooting occurred between 8 and 8:15 p.m. The only evidence of an alibi came from the testimony of Jose DeJesus, an acquaintance of the defendant, who said that he saw the defendant and others when he arrived at Alda Cheshire’s apartment on the evening of the shooting. He estimated that he arrived at Cheshire’s apartment some time after 8 p.m., based on “the math” he had done with defense counsel concerning the timing of his earlier activities that day.
The Commonwealth did not elect to proceed on the assault theory of “immediately threatened battery,” under which it would have been required to “prove that the defendant intentionally engaged in menacing conduct that reasonably caused the victim to fear an imminent battery.” Commonwealth v. Melton, 436 Mass. 291, 295 n.4 (2002). See Commonwealth v. Porro, 458 Mass. 526, 530-531 (2010) (examining difference between imminent threatened battery and attempted battery).
While the prosecution sought and obtained a joint venture instruction as to the charge of murder, it did not seek a joint venture instruction as to the charge of armed assault with intent to murder.
Officer Mark Lydon, a ballistics expert called by the Commonwealth, testified that twenty-six shell casings were recovered at the scene. Subsequent forensic testing indicated that twelve of those casings were from bullets fired from one nine millimeter semiautomatic weapon and fourteen were from bullets fired from another nine millimeter semiautomatic weapon. A map introduced in evidence and used by Detective Lydon during his testimony showed that a shell casing from one weapon was located under the right rear tire of the van, and another casing from that same weapon was recovered on the pavement *616near the right rear of the van. A shell casing from the second weapon was recovered near the right rear of the van on the sidewalk, slightly apart from a group of casings from the second weapon that were recovered close to where the victim had been shot. Detective Lydon testified that the shooter most likely was standing to the left and front of where the shell casings were recovered, and although some of the casings may have rolled or bounced, they would not have traveled “too far” from where they were ejected.