Lowy, David A., J.
In January 1996, an Essex County jury convicted the defendant, Angel Echavar-ria, of murder in the first degree in the shooting death of Daniel Rodriguez. He was sentenced to life imprisonment in State Prison, and the Supreme Judicial [643]*643Court upheld his conviction on December 22, 1998. Commonwealth v. Echavarria, 428 Mass. 593 (1998).
This matter is before the court on the defendant’s motion for new trial, pursuant to Mass.R.Crim.P. 30(b). The defendant claims that the interests of justice require that he be granted a new trial, due to ineffective assistance of counsel, the prosecution’s failure to disclose exculpatory evidence, and newly discovered DNA evidence and new research concerning the fallibility of certain eyewitness identification procedures. An evidentiary hearing on the matter was held on December 18 and 19, 2014, and March 6, 2015. For the following reasons, the defendant’s motion is allowed.
BACKGROUND and FINDINGS OF FACT
The following facts are gleaned from the trial record, as well as the evidentiary hearing on the defendant’s motion for new trial, with further facts reserved for discussion below. The principal witness for the prosecution, Isidoro Rodriguez (“Isidoro”), testified that on the evening of January 7, 1994, he and his brother Daniel Rodriguez sought to enter the Lynn apartment where Isidoro lived at the time. They were confronted by two armed men who had gained access to the apartment, one of whom Isidoro later described as being taller and lighter skinned, and the other as shorter and darker skinned. The two men dragged Isidoro and Daniel into the apartment. The taller, lighter skinned man bound Isidoro’s hands, brought him into a bedroom, ordered him to lie on the floor, and threw a shirt over his face. Shortly thereafter, Isidoro untied himself, and found his brother, who had been bound hand and foot and shot twice in the head, dead in the bathroom. The two armed men had already fled the apartment.
Police arrived on the scene, and interviewed Isidoro. John Garvin, who at the time of the murder was assigned to investigate homicides as a supervising sergeant with the Essex County District Attorney’s office, testified at trial that Isidoro provided the following description of the lighter-skinned perpetrator on the night of the murder: “Hispanic, maybe being Puerto Rican descent, approximately twenty years of age, five foot ten inches tall, with a stocky, chunky build. He had brown eyes and a medium complexion ... no facial hair, short curly hair on the top of his head and shaved sides.” Trial trans. 1/24/96 at 122. A flyer that was disseminated to local departments described the lighter-skinned individual as stocky or chunky, with a shaved-sides hair style.
Hours after the murder, at the police station, Isidoro took part in a photographic identification process. According to Mr. Garvin’s testimony, he reviewed six to seven mugbooks with over two hundred photographs per book. He singled out one of the photographs as depicting a person whom he said looked like the lighter skinned perpetrator. Officer Raymond Guillermo, a Spanish-speaking officer assigned to the anti-crime unit, served as an interpreter for Isidoro during the photo identification process.
At the probable cause hearing, regarding Isidoro’s identification of the photograph, Isidoro’s testimony was as follows:
[Isidoro]: I told Ramon [Guillermo] that one of them looked like maybe and he said “Are you sure”? And I said Ramon I am too nervous and Ramon said, “But you have to know, what kind of people, what kind of person” and I said “No I don’t know.”
[Defense]: Now with regard to the one photograph that you thought looked like somebody, did you think that he looked like the person that had been in your apartment?
[Isidoro]: Yes that’s right.
[Isidoro]: I kept looking at him, I talked to Ramon, I explained to Ramon “Can’t tell, Ramon whether it is this or not” but he said, “You know this is not a joke you have to know how these things are.”
[Isidoro]: ... I told Ramon, I said I’m scared, I’m nervous (inaudible) a few hours of the day so I did not affirm in any way that it was that person, you know that I was certain that it was that person.
PCH trans. at 63-66.
At trial, the jury heard the following testimony regarding Isidoro’s identification of the photograph:
[Prosecutor]: What photo did you see?
[Isidoro]: A person that looked — looked alike.
[Prosecutor]: Did you say anything to the police?
[Isidoro]: Yes.
[Prosecutor]: What did you tell them of that picture?
[Isidoro]: I told them that the person looked alike.
[Prosecutor]: Did you pick out that picture?
[Isidoro]: They took it out. I picked it. I told them that yes, that it looked alike.
[Prosecutor]: Did you tell them that was one of the men?
[Isidoro]: No. No, I never said that. I said that it looked like that person.
Trial trans. 1/22/96 at 86-87.
Mr. Garvin and Officer Guillermo gave the following trial testimony:
[Mr. Garvin]: [Isidoro] stated through Officer Guillermo — that Mr. Bonafacio — the photograph of Mr. Bonafacio looked like one of the subjects in his house that night. Then he stated that it wasn’t the subject, but it was—
[Prosecutor]: Sony. I didn’t hear.
[644]*644[Mr. Garvin]: He stated it wasn’t the subject that was there, but he was similar to the person. He had some of the same features.
Trial trans. 1/24/96 at 129-30.
[Prosecutor]: What did he say about that photo?
[Officer Guillermo]: That he pointed to a photo.
[Prosecutor]: What did he say?
[Officer Guillermo]: “That looks like him, but it’s not him.”
[Prosecutor]: Soriy. What did he say?
[Officer Guillermo]: He said, “That looks like him, but it’s not him.”
Trial trans. 1/25/96 at 30-31.
Several days after Isidoro identified the photograph, the officers showed him another photo array including a more recent photograph of the same man, which Isidoro again singled out as looking like the lighter-skinned perpetrator. The photograph actually depicted a man named Mariano Bonafacio. According to a police report authored by then-Lieutenant J. Michael Roach, which was entered in evidence at the eviden-tiaiy hearing, victim Victor Batista previously identified Mr. Bonafacio from a photo array as the person who had shot Batista in the chest on November 11, 1993, less than two months before the murder in this case and approximately half a mile away from where Daniel Rodriguez was murdered. Lynn police obtained a search warrant for a car Mr. Bonafacio was seen driving days after that shooting, and a search of the car revealed a bag of cocaine and a pager.
The court credits the testimony of Mr. Echavarria’s attorney, Charles Robson, that he never saw the Bonafacio report. The Court also credits the testimony of Mr. Garvin at the evidentiary hearing that he was unclear as to the scope of any investigation that was done concerning Mr. Bonafacio, or if a report was generated and turned over to the District Attorney’s office concerning Mr. Bonafacio’s alleged involvement in the shooting of Mr. Batista.
Several days after singling out the photograph of Mr. Bonafacio at the police station, Isidoro saw Mr. Echavarria and another man — Mr. Echavarria’s co-defendant at trial — in a Lynn barbershop. Isidoro spoke to the defendant, telling him he looked like someone Isidoro knew from Puerto Rico, and asking him his name. Mr. Echavarria gave his real name. Later that day, Isidoro saw the same two men in a nearby restaurant. Believing them to be the two armed men from the night of the murder, Isidoro went to the Lynn police station. Ultimately, Mr. Echavarria and his friend were arrested.
Mr. Echavarria, at the time of his arrest, was five feet ten inches tall and weighed 135 pounds. His booking sheet, which was admitted in evidence at the evidentiaiy hearing on the motion for new trial, indicates that Mr. Echavarria had a goatee. His booking photograph and the photograph of Mr. Bonafacio that Isidoro identified the night of the murder clearly depict two different people. Mr. Echavarria was bom in the Dominican Republic and came to the United States from the Dominican Republic when he was twenty-six years old. Isidoro testified at trial that the lighter-skinned man spoke “Puerto Rican Spanish,” based on his pronunciation of the word “cabrón.” Trial trans. 1/22/96 at 72.
The Court credits the testimony of Dr. Michael O’Laughlin that based upon Isidoro’s testimony at trial, the perpetrators spoke enough words such that somebody familiar with the Puerto Rican accent would accurately have been able to distinguish whether their accents were Puerto Rican. The Puerto Rican accent is different from the Dominican accent, and is one of the most distinctive accents in the Spanish-speaking world. It is distinctive in intonation and it has been influenced more by English than any other Spanish accent. There are significant differences in sounds and in meanings of words between Dominican and Puerto Rican accents.
About one year after Daniel Rodriguez was murdered, police interviewed a second eyewitness, Gary Sevinor, who was at the time serving a state prison sentence. According to Sevinor’s trial testimony, he was present at the apartment the night of the murder for the purpose of buying cocaine; he provided money to an individual who left the apartment to get the drugs, when shortly thereafter there was a knock on the door. Mr. Sevinor opened the door, and the individual to whom he had given the money returned with two other people, one of whom was holding a gun to the individual’s neck. Mr. Sevinor stated that he looked at the two intruders for “like two seconds” before they ordered him to the floor. Trial trans. 1/18/1996 at 154. He described one man as lighter than the other, and said he focused more on the lighter-skinned man.
During the police’s initial interview with Mr. Sevinor, Mr. Garvin presented him with a photo array which included a photograph of Mr. Echavarria. According to Mr. Garvin’s trial testimony, Mr. Sevinor studied the array for “quite a while, a minute or two,” before he identified the photograph of Mr. Echavarria. Trial trans. 1/24/96 at 163. On a photocopy of the array, beneath Mr. Echavarria’s photograph, Sevinor wrote, “I couldn’t swear that this was the man that said on the floor face down.” Trial trans. 1/18/96 at 184-85. Mr. Sevinor testified, “I was pretiy sure that was him, but I wouldn’t want to swear on it. I only saw him for two seconds.” Id. at 183.
Mr. Sevinor also testified that he had been tied up by the lighter-skinned man, who had “ripped the telephone cord that [Mr. Sevinor] was talking on out of the wall,” trial trans. 1/18/1996 at 159, and used that to bind him. Three telephone wires were recovered by police from the crime scene, and Orchid Cellmark [645]*645Laboratories completed DNA testing of the ends of the wires in 2011. The DNA profile revealed by the testing was a mixture consistent with two individuals. Mr. Echavarria was excluded as a contributor to the profile.
The Court credits the testimony of Dr. Carl Ladd that evidence stored in a dry environment is suitable for testing twenty to thirty years later, but notes that at the time these cords were seized and stored, anti-contamination measures taken during the collection of evidence were less stringent than they are today, and that the manner in which evidence was treated in the courtroom could expose it to aerosal contamination as well as contamination from skin contact.
At trial, both Isidoro and Mr. Sevinor identified the defendant as the lighter-skinned of the two perpetrators.
Mr. Echavarria’s attorney, Charles Robson, told the jury in his opening statement, “You’re going to hear Mr. Echavarria say that he wasn’t even there. He wasn’t at this apartment. He’s never been at that building. Didn’t even know where it was.” Trial trans. 1/17/96 at 161-62. The court does not credit the testimony of Mr. Echavarria that he told Mr. Robson he wanted to testify, and it credits the testimony of Mr. Robson that Mr. Echavarria was not adamant that he wanted to testify. Over the course of the trial, Mr. Robson never called Mr. Echavarria as a witness.
Mr. Robson was retained by Mr. Echavarria for $2,500.00. He did not retain an investigator, and at the time of Mr. Echavairia’s trial had complaints pending against him from the Board of Bar Overseers. He stipulated to a Petition for Discipline and was suspended from the practice of law for three years, from 1997 through 2000.
On January 26, 1996, the jury convicted Mr. Echavarria of first degree murder, armed assault in a dwelling, armed assault and battery, and armed robbery. The judge ordered entry of findings of not guilfy for Mr. Echavarria’s co-defendant notwithstanding the jury’s verdicts. The Commonwealth did not appeal the judge’s order as to the co-defendant.
RULINGS OF LAW
A judge may grant a new trial “at any time if it appears that justice may not have been done.” Mass.R.Crim.P. 30(b). “Judges are to apply the standard set forth in rule 30(b) rigorously and should only grant such a motion if the defendant comes forward with a credible reason which outweighs the risk of prejudice to the Commonwealth.” Commonwealth v. Wheeler, 52 Mass.App.Ct. 631, 635-36 (2001). Anew trial will be ordered only where the court, after “review[ing] the evidence and the case as a whole, considering the strength of the Commonwealth’s case, as well as the nature and significance” of the defendant’s asserted grounds for a new trial, is “left with uncertainly that the defendant’s guilt has been fairly adjudicated.” Commonwealth v. Chase, 433 Mass. 293, 299 (2001), citing Commonwealth v. Amirault, 424 Mass. 618, 647 (1997).
Mr. Echavarria argues that he is entitled to a new trial on several grounds: ineffective assistance of counsel, because Mr. Robson told the jury in his opening statement that they would hear the defendant’s testimony, but then never called him as a witness, and because he failed to present Isidoro’s testimony from the probable cause hearing relating to his identification of the photograph of Mr. Bonafacio immediately following the murder; the prosecution’s alleged failure to provide the defense with evidence that Mr. Bonafacio had shot someone less than two months before the murder, and that Mr. Bonafacio was a drug dealer; and newly discovered DNA evidence excluding the defendant as a contributor to the DNA profile identified from the telephone cords that were used to tie up one of the witnesses at the time of the murder. The defense also argues that scientific knowledge regarding the unreliability of eyewitness identifications, yet to be developed at the time of trial, constitutes newly discovered evidence that demonstrates the flawed nature of the identification procedures that resulted in Gary Sevinor’s identification of the defendant. The Court bases its ruling today on the grounds of ineffective assistance of counsel.
“Claims of ineffective assistance of counsel raised after plenary review under G.L.c. 278, §33E, are reviewed under the applicable constitutional standard.” Commonwealth v. Drew, 447 Mass. 635, 639 n.2 (2006), citing Commonwealth v. Wright, 411 Mass. 678, 682 n. 1 (1992). Here, the defendant must show that there has been serious incompetency, inefficiency or inattention of his counsel which falls measurably below that which might be expected from an ordinary fallible lawyer, and that the failure deprived the defendant of an otherwise available, substantial ground of defense. Commonwealth v. Medeiros, 73 Mass.App.Ct. 571, 577 (2009); Commonwealth v. Saferian, 366 Mass. 89, 96 (1974). Put another way, to satisfy the second prong of the ineffectiveness standard, the defendant must make a showing of “ ‘reasonable probability’ that ‘but for counsel’s unprofessional errors, the result of the proceeding would have been different,’ ” Commonwealth v. Mahar, 442 Mass. 11, 15 (2004), quoting Strickland v. Washington, 466 U.S. 668, 694 (1984), or that “better work might have accomplished something for the defense,” Commonwealth v. Satterfield, 373 Mass. 109, 115 (1977). “A strategic decision of counsel amounts to ineffective assistance ‘only if it was manifestly unreasonable when made.’ ” Commonwealth v. Montez, 450 Mass. 736, 754 (2008), citing Commonwealth v. Martin, 427 Mass. 816, 822 (1998).
Mr. Echavarria argues that trial counsel rendered ineffective assistance in several instances, including for promising the jury, in his opening statement, that [646]*646they would hear the defendant testify that “he wasn’t even there. He wasn’t at this apartment. He’s never been at that building. Didn’t even know where it was,” and then failing to call the defendant to the stand.
“A promise by defense counsel in his opening statement to produce key testimony, followed by a failure to deliver it may, without more, constitute ineffective assistance of counsel.” Commonwealth v. Duran, 435 Mass. 97, 109 (2001), citing Anderson v. Butler, 858 F.2d 16, 19 (1st Cir. 1988), aff'd sub nom. Commonwealth v. Anderson, 408 Mass. 803 (1990). “If counsel fails to deliver on a promise that the defendant will testify, a danger arises that the jury may presume that the defendant is unwilling to testify under the pressure of cross-examination under oath, or that the defense is otherwise flawed.” Duran, 435 Mass. 97 at 109. See also, e.g., Ouber v. Guarino, 293 F.3d 19, 28 (1st Cir. 2002) (“When a jury is promised that it will hear the defendant’s story from the defendant’s own lips, and the defendant then reneges, common sense suggests that the course of trial may be profoundly altered. A broken promise of this magnitude taints both the lawyer who vouchsafed it and the client on whose behalf it was made”). However, trial counsel’s failure “to produce evidence to which counsel alludes in an opening statement” will not, in all cases, in and of itself constitute ineffective assistance of counsel. Duran, 435 Mass. 97 at 109. The court should consider “whether the opening statement reflected inadequate preparation, incompetency, or inattention, and whether the subsequent failure to produce the evidence was a decision forced upon counsel by events over which he had no control, or otherwise was supported by strategic justifications.” Commonwealth v. McMahon, 443 Mass. 409, 425 (2005) (internal quotes and citations omitted).
The Commonwealth argues that “the only reasonable conclusions that can be drawn from the combined [trial] and appellate record are that: 1) Mr. Robson referenced the defendant testifying in opening statement because he intended to put him on the stand to deny the allegations; 2) the case developed in such a way that Mr. Robson concluded that it was no longer in the defendant’s best interest to testify; 3) after discussion, he and the defendant agreed to the same; and 4) counsel’s actions were reasonable at all junctures.” The Court accepts the first of these propositions. There is insufficient evidence on this record to determine whether the second and third propositions are accurate; however, even assuming that they are, the Court does not agree with the Commonwealth’s proposition that Mr. Robson’s actions were reasonable.
Certainly, an attorney’s strategy may alter over the course of a trial; what appeared to be sound strategy on day one may no longer be prudent or possible to pursue several days later, if the trial has developed in unexpected ways. But given the “drastic ramifications” that can result from the failure to present critical evidence that has been announced in an opening statement, McMahon, 443 Mass. at 425, an attorney crafting her opening statement must exercise caution and a reasonable degree of foresight, not least when deciding whether to announce that the defendant will testify. See, e.g., Ouber, 293 F.3d at 28. (“[OJn the eve of trial, a thoughtful lawyer may remain unsure as to whether to call the defendant as a witness. If such uncertainty exists ... it is an abecedarian principle that the lawyer must exercise some degree of circumspection”).
Courts “cannot fault counsel for not guarding against the unforeseeable.” Id. at 29. See, e.g., McMahon, 443 Mass. at 426 (where the defendant’s lawyers had spent many hours preparing the defendant to testify, and were “shocked” by his declaration, mid-trial, that he would not take the stand, “(t]he record amply supported] the judge’s determination that the defense opening was a strategic decision that had failed through no fault of counsel”); Commonwealth v. Nardone, 406 Mass. 123, 127 (1989) (“counsel’s failure to call a ballistics expert, as promised, was clearly a decision forced upon him by events over which he had no control,” when retained expert refused to testify mid-trial and substitute testimony “would not appreciably advance defendant’s theory of the case”). The record here, however, does not indicate that circumstances of the trial changed to a degree which should not reasonably have been foreseeable to Mr. Robson.
According to the Commonwealth’s assessment of the trial record,
Mr. Robson may well have determined that the defendants’ more than day long cross-examination of Isidoro had been fruitful in undermining his credibility — e.g., his admission that he had lied at the probable cause hearing when he said he did not smoke marijuana the day of the murder — and when the defense was also permitted to expose the fact that Isidoro had voluntarily absented himself from trial. In addition, in connection with the defendants’ motions for required finding of not guilty at the close of the Commonwealth’s case, Judge Grabau referenced that he “spoke candidly last night when we had a conference,” in a manner suggesting that he had conveyed to counsel reservations he had concerning the strength of the Commonwealth’s case. This also may have colored Mr. Robson’s perception as to whether it was in his client’s best interests to testify, or to rest on the case that had already been presented.
Comm. PF at 39-40 (citations to trial transcript omitted). These may well have been reasonable justifications for not calling Mr. Echavarria to the stand. Bombshell revelations, however, they were not. That the credibility of the Commonwealth’s key identification witness would be undermined over the course of trial was certainly well within the realm of foreseeabil[647]*647ity at the outset of proceedings; Mr. Robson predicted as much when he told the jury in his opening, “(y]ou’re going to hear the person who was in that apartment give different stories and different identifications of these people. You’re going to hear he gives one identification one time and another identification another time”; and when he asked the jury “to pay strict attention to the discrepancies and the opportunities of the so-called eyewitnesses.” Trial trans. 1/17/96 at 162-63.
If Mr. Robson’s strategy was indeed to call Mr. Echavarria to testify depending on how Isidore’s credibility presented over the course of the trial and the apparent overall strength of the Commonwealth’s case, and if he was anticipating at the outset that Isidore’s credibility would be undermined with evidence of inconsistent “identifications” and accounts, and that the Commonwealth would present no evidence aside from eyewitness testimony linking the defendant to the murder, it should have been apparent to him at the time he delivered his opening statement that there was a significant chance that he would not, in the end, find it necessary or prudent to call Mr. Echavarria to testify.
“There is no principle, or requirement, that one must name all one’s witnesses ... in the opening; indeed, some defendants’ lawyers choose not to open at all. One keeps options open by keeping silent.” Anderson, 858 F.2d at 18 (1st Cir. 1988). While “an opening statement can be critical in preventing a jury from forming a one-sided view at the trial’s outset,” Commonwealth v. Ramos, 66 Mass.App.Ct. 548, 552 (2006), in the context of this case Mr. Robson surely could have delivered an effective opening statement without telling the jury they would hear directly from Mr. Echavarria. Under the circumstances, the strategic benefit of announcing at the beginning of his opening statement that Mr. Echavarria would testify was non-existent. Cf. McMahon, 443 Mass. at 425 (“the strategic benefit of announcing specific anticipated testimony in the opening statement may outweigh the risk that the testimony will not be available”). The Court agrees with the defendant that “it strains credulity to view counsel’s promise in his opening as a strategic or tactical choice made with appropriate consideration of its ramifications.” Def. PF at 23. See Ouber, 293 F.3d at 28-29, quoting Anderson, 858 F.2d at 18 (“Even ‘if it was . . . wise [not to have the witness testify] ... it was inexcusable to have given the matter so little thought at the outset as to have made the opening promise’ ’’). The fact that Mr. Robson was paid only $2500.00 to represent Mr. Echavarria, did not hire an investigator, failed to impeach Isidore’s testimony, as discussed below, with powerful evidence from the probable cause hearing, and was later suspended from the practice of law for reasons including neglect of clients’ cases, only tends to strengthen the defendant’s argument. So does his testimony at the evidenfiaiy hearing that he never saw ihe Bonafacio report; this indicates that he did not investigate with rigor a man whom Isidore, on two separate occasions and from two different photographs, identified as looking like the lighter skinned perpetrator.1 Compare Commonwealth v. Ramos, 77 Mass.App.Ct. 1120 (2010), 2010 WL3463352 at *3 (Mass.App.Ct. 2010) (“There [was] no serious allegation . . . that trial counsel was not thoroughly prepared, nor . . . any suggestion that he made the challenged decisions without careful thought. As the motion judge found, counsel was an experienced criminal defense attorney ... He clearly gave the matter his careful attention after conducting thorough preparation”). Furthermore, Mr. Robson did not request an instruction explaining to the jury that they were not permitted to draw negative inferences from Mr. Echavarria’s failure to testify, and this may plausibly be viewed as yet another indication of Mr. Robson’s inattentiveness. The Commonwealth contends that this was a prudent move to avoid drawing attention to the fact that the defendant did not testify, but the Court finds this argument unavailing; the juiy could not have forgotten they did not hear Mr. Echavarria speak after being told that he would do so. The Court concludes that under the circumstances it was manifestly unreasonable for Mr. Robson to promise the defendant’s testimony in his opening statement.
It is certainly conceivable that the jury’s verdict may have been different, were it not for Mr. Robson’s error. Mr. Robson, who delivered his opening statement immediately after the prosecutor delivered his, began by telling the jury that “now [they were] going to hear the other side of the story,” and that “the rest of the story was totally different” from what they had just heard. They were going to hear Mr. Echavarria tell that them he was not even there, Mr. Robson explained, that he did not even know where that apartment building was. Coming as it did at the beginning of Attorney Robson’s statement, this assertion likely would have impressed upon the jurors that one of the key ways through which they could expect Isidore’s testimony to be discredited would be through the contrary testimony of Mr. Echavarria. This case is distinguishable from Commonwealth v. DiCicco, in which the Court of Appeals concluded that “defense counsel's statement, that the defendant might testify, was far from a dramatic promise.” 44 Mass.App.Ct. 111, 123 (1998). Again, though the Commonwealth argues otherwise, Attorney Robson’s promise was not one that the jury was likely to forget over the course of the trial.
The judge in this case did not instruct the jury at any point that they were not permitted to draw negative inferences from the defendant’s failure to testify. Compare, e.g., DiCicco, 44 Mass.App.Ct. at 123 (finding no prejudice where “the trial judge forcefully instructed the juxy that no such [negative] inference could be drawn” from the defendant’s failure to testify); Duran, 435 Mass. at 110 (counsel’s performance not ineffective where “the judge gave clear instructions on the point to the jury, admonishing them that ‘no infer[648]*648ence unfavorable to the defendant [couldl be drawn by [them] by reason of his not testifying’ ”); Yancey v. Hall, 237 F.Sup.2d 128, 135 (D.Mass. 2002) (no prejudice where “the trial judge repeatedly instructed the jury— before opening statements, before closing statements, and in the jury charge — that [defendant] was not required to testify and that no adverse inferences could be drawn from his failure to do so”). Furthermore, Mr. Robson did not establish through other evidence that Mr. Echavarria was not at the scene of the crime the night of the murder, or that he had made statements to that effect. Compare Duran, 435 Mass. at 110 (“[d]efense counsel did not abandon the defendant’s alibi defense; he called . . . the defendant’s father and aunt, both of whom testified that the defendant was home at the time of the murder”); DiCicco, 44 Mass.App.Ct. at 123 (“defense counsel informed the jury that any anticipated testimony by the defendant would substantially track his statements to the police which they had already heard from Trooper Cox and a copy of which was in evidence”).2
In the context of this trial, Mr. Robson’s promise that the jury would hear from Mr. Echavarria was a serious mistake that could have had serious consequences for the defendant. His performance, at it related to the unfulfilled promise, was perhaps not as egregiously prejudicial as defense counsel’s in Anderson; there, counsel “continued to base his defense on the defendant’s mental state even after failing to present promised crucial psychiatric testimony in relation to that argument,” see Yancey, 237 F.Sup.2d at 133-34, and here Attorney Robson, as the trial progressed, seemed to focus his attempts on poking at Isidoro’s credibility through various impeachment methods, rather than continuing to emphasize that Mr. Echavarria was not at the scene of the crime. Yet by giving the jury reason to doubt Mr. Echavarria’s story, he may quite possibly have undermined the entire premise of the defense, which was of course that Mr. Echavarria had been falsely identified.
Even assuming that the prejudice arising from Mr. Robson’s unfulfilled promise may not quite have risen to constitutionally unacceptable levels, when the Court considers this error together with Mr. Robson’s failure to impeach Isidoro’s trial testimony with his probable cause hearing testimony regarding his identification of the photograph of Mr. Bonafacio and the infirmities of the identification evidence, the conclusion becomes inevitable “that better work might have accomplished something material for the defense.” Commonwealth v. Satterfield, 373 Mass. 109, 115 (1977).
At the probable cause hearing in this case, Isidoro testified that following the murder, when presented with several mug books as part of a photo identification process, he singled out a photograph of Mr. Bonafacio and told Officer Guillermo that it “looked like maybe,” and that he said “can’t tell. . . whether it is this or not.” At trial, however, when the prosecutor asked Isidoro whether he told the officers that the photograph was one of the men, Isidoro simply answered, “No. No, I never said that. I said that it looked like that person.” As prior inconsistent statements,3 Isidoro’s assertions at the probable cause hearing would have been admissible for impeachment purposes on cross examination,4 yet Attorney Robson did not attempt to introduce them.
The Commonwealth argues that “[i]t is far from evident that the . . . somewhat unclear accounts by Isidoro [at the probable cause hearing] differed enough from his trial testimony to even serve any impeachment purpose. ” Comm. PFat 9. The Court disagrees. Isidoro’s testimony at the probable cause hearing highlights that at the time of the initial photo identification process, a few hours after having the opportunity to observe the lighter-skinned perpetrator’s face for several minutes, he believed the photograph of Mr. Bonafacio — whose appearance in the photograph differed noticeably from Mr. Echavarria’s appearance at the time of his arrest— may have depicted the lighter-skinned perpetrator, though he was unwilling to “confirm” that he was “certain” that it was that person. Isidoro’s trial testimony, on the other hand, does not allude to any confusion or hesitation on his part, and can be understood to suggest that he did not believe the photograph might in fact depict his attacker. What Isidoro actually said when he singled out the photograph of Mr. Bonafacio out of over one thousand photographs bears directly on his reliability as an identifying witness.
This Court recognizes that “[ilmpeachment of a witness is, by its very nature, fraught with a host of strategic considerations,” Commonwealth v. Knight, 437 Mass. 487, 502 (2002) (internal quotes and citation omitted), and that “[gjenerally, failure to impeach a witness does not amount to ineffective assistance of counsel.” Commonwealth v. Fisher, 433 Mass. 340, 357 (2001). See also, e.g., Commonwealth v. Montez, 450 Mass. 736, 757-58 (2008) (in first-degree murder case, where physical evidence linked defendant to crime, trial counsel's decision not to impeach witness who testified regarding her identification of the defendant as the man who had previously assaulted her not manifestly unreasonable; the court credited counsel’s testimony “that he asked [the witness] no questions ... because it onlywould have served to reinforce her testimony . . . [and] his strategy was to treat [this] matter, like the other prior bad act evidence, as having nothing to do with the . . . murder” in the instant case); Commonwealth v. Rice, 441 Mass. 291, 306 (2004) (defense counsel not ineffective for failing to cross examine witness with line of questions that “likely would have undermined” a defense; “[a]voiding this risky subject was not manifestly unreasonable”). Even where a reviewing court “see[s] no tactical advantage to the defendant in counsel’s decision not to impeach,” it will still defer to counsel’s judgment if “nothing further of any significance would have been gained if counsel” had pursued a particular avenue of impeachment. Commonwealth v. Mitchell, 428 Mass. 852, 855 (1999) (citing its decision in Commonwealth v. Daigle, 371 Mass. 541, 543 (1980), in which the court concluded [649]*649that “where defense counsel did not produce criminal records of the defendant’s accomplices in order to impeach them, but did establish that the accomplice-witnesses were ‘seasoned criminals’ through their own admissions and the circumstances of the robbery, defense counsel’s approach was not manifestly unreasonable”). Here, the Commonwealth points to the fact that trial counsel did question Isidoro regarding other portions of his probable cause hearing testimony, including whether he drank a beer the night of the offense or smoked marijuana that night, as proof that counsel was familiar with Isidore's testimony at the hearing and made a conscious decision not to cross examine Isidore regarding his identification of Mr. Bonafacio’s photograph. However, unlike in the cases cited above, the Court can neither discern from the trial record any reasonable strategic justification for Mr. Robson’s decision, nor conclude that impeaching Isidore’s testimony would have added little value for the defense.
At the evidentiary hearing on Mr. Echavama’s motion for new trial, trial counsel agreed with the Commonwealth that he believed the “quality of [Isidore and Sevinor’s] identification and their befievability in front of the jury was really almost the sole issue at issue in the case.” Evid. H. trans. vol. 1 at 61. Impeaching Isidore with his probable cause hearing testimony regarding the photograph would have gone to the heart of this issue, presented no apparent risk of undermining other areas of the defense, and arguably would have more compellingly called into question Isidore’s ability to make an accurate identification than the other lines of questioning Mr. Robson pursued in his cross examination of Isidoro — for example: “[H]aving had beer, marijuana, a gun to your head, and being tied up, you don’t feel that your ability to recognize anybody was affected in any way, is that correct?” Trial trans. 1/24/96 at 50-51. Mr. Echavarria would clearly have stood to benefit from the introduction of this portion of Isidore’s probable cause hearing testimony, and in failing to introduce it, Mr. Robson performed below the level which might be expected from an ordinary fallible lawyer.
At various points during the trial, the Commonwealth emphasized Isidore’s conduct during the initial photo identification process. The prosecutor elicited testimony from Lieutenant Garvin and Officer Guillermo that Isidore was explicit that the photograph looked like the perpetrator, but was not him; the prosecutor also stated in his closing argument that
... [Isidore] says, “I saw him and I saw him well.” And it was tested. It was tested that night by almost, what, a thousand photographs? . . . One, it looks like. But remember, they said it wasn’t. That’s what they told you he said, and that’s what he said, it wasn’t
Transcript 1/25/96 at 101 (emphasis added).
Isidore’s probable cause hearing testimony could have served as a meaningful counter to these assertions that Isidore made a confident negative identification of the photograph of Mr. Bonafacio. As the Supreme Judicial Court acknowledged in its decision in Commonwealth v. Gomes, “[s]tudy after study demonstrates . . . that jurors place the greatest weight on eyewitness confidence in assessing identifications even though confidence is a poor gauge of accuracy.” 470 Mass. 352, 366 (2015) (quoting Justice Sotomayor’s dissent in Perry v. New Hampshire, 132 S.Ct. 716, 739 (2012)). Left unim-peached, Isidore’s trial testimony gave the impression that he never questioned whether Mr. Bonafacio might have been the fighter-skinned perpetrator; Isidore’s probable cause hearing testimony, if introduced, could plausibly have altered the jury’s perception of Isidore’s confidence and its assessment of his subsequent identification of Mr. Echavarria. Furthermore, it could have provided the defense with fodder for an argument that the jury should not credit the testimony of Officer Guillermo and Mr. Garvin. In short the jurors would have had evidence to rebut the Commonwealth’s argument that Isidoro never picked out Mr. Bonafacio as a potential perpetrator out of approximately one thousand photographs.
The Court reviews this case under the standard of “substantial risk of miscarriage of justice.” See Commonwealth v. Valentin, 470 Mass. 186, 188 (2014). A substantial risk of a miscarriage of justice exists if there is a “serious doubt whether the result of the trial might have been different” in the absence of a defense attorney’s errors at trial. Id. at 189, quoting Commonwealth v. Azar, 435 Mass. 675, 687 (2002). Under this standard, a court asks whether the errors are “sufficiently significant in the context of the trial to make plausible an inference that the jury’s result might have been otherwise but for” the errors, Commonwealth v. Alphas, 430 Mass. 8, 13 (1999) (internal quotes and brackets omitted), and in making such determination, considers the strength of the Commonwealth’s case against the defendant. See Commonwealth v. Azar, 435 Mass. 675, 687 (2002). See also, e.g., Commonwealth v. Thayer, 39 Mass.App.Ct. 396, 399 (1995) (finding prejudice partly in light of the fact that “[t]his was not a case in which evidence of the defendant’s guilt was overwhelming”); Commonwealth v. DeJesus, 71 Mass.App.Ct. 799, 806 (2008) (concluding that trial counsel’s failure to object to certain evidence did not prejudice defendant given the strength of other evidence). The record in this case shows that the evidence against Mr. Echavarria was weak. The Commonwealth presented no physical evidence linking Mr. Echavarria to the murder of Daniel Rodriguez,5 and Isidore’s testimony, upon which the Commonwealth’s case depended entirely, was problematic in more ways than one. Isidore demonstrated a marked lack of mental acuity over the course of trial — particularly, as the trial judge noted, “lack of education, sophistication, [and] very little conception as to spatial relationships and . . . distance and time.” Trial trans. 1/26/96 at 17. His description the night of the murder of the lighter-skinned perpetrator as a “stocky” or “chunky” was inconsistent with Mr. [650]*650Echavarria’s physique; at the time of his arrest, Mr. Echavarria was five feet ten inches and weighed 135 pounds. Indeed, at the evidentiary hearing, Mr. Garvin testified that on the day of his arrest, “compared to what [Mr. Garvin] would call chunky [Mr. Echavarria] wasn’t.” Evid. H. trans. at 77. Isidore’s description of the perpetrator as a Puerto Rican male certainly did not fit Mr. Echavarria, who like Isidore was from the Dominican Republic; the testimony of Dr. O’Laughlin at Mr. Echavarria’s evidentiary hearing underscores the significance of this distinction. Furthermore, though Mr. Sevinor’s identification provided some corroboration for Isidore’s identification of Mr. Echavar-ria, as the judge acknowledged at trial, “it wasn’t one hundred percent corroboration.” Trial trans. 1/26/96 at 17. Though today’s scientific research, if it had been available at the time of trial, would have provided for more opportunity to attack the integrity of Mr. Sevinor’s identification,6 there were significant weaknesses in the identification which should have been apparent to the jury even in 1996. For example, the jury heard that Mr. Sevinor had only seen the lighter-skinned perpetrator for several seconds, that he had to study the array for some time before picking out Mr. Echavarria’s photograph, and that when he did so he wrote beneath the photograph that “he couldn’t swear to it” that this was the culprit. They also heard that Mr. Sevinor’s identification occurred more than a year after the night of the murder.
Had Mr. Robson not promised Mr. Echavarria’s testimony and then failed to deliver on that promise, the jury may have found Mr. Echavarria’s defense more credible; and had Mr. Robson introduced Isidore’s testimony from the probable cause hearing, the jury would have had more reason to doubt the reliability of Isidore’s identification. The jury’s perception of the relative integrity of each side thus altered, it is possible that the jury would have considered with more wariness the overall weakness of the Commonwealth’s case as a whole, and that they would have concluded that the evidence “fell short of eliminating a reasonable doubt” that the defendant had not been falsely identified. See Crayton, 470 Mass at 254.
We put our faith and our fate in the jury, and almost twenty years ago, the jury found Mr. Echavarria guilty after two men identified him as the perpetrator. It is not appropriate for this Court to grant a new trial simply because it has some “lingering doubt” about the outcome of the first trial. See Commonwealth v. Amirault 424 Mass 618, 637 (1997). The granting of a motion for a new trial in a case in which the Supreme Judicial Court has already conducted plenary review is seldom granted and for good reason; such motions should only be granted with great reticence considering the prejudice to the Commonwealth and society’s interest in resolving yesterday’s problems with finality. Here, however, the weakness of the Commonwealth’s case, along with the performance of Mr. Echavarria’s counsel which fell measurably below that which might be expected from an ordinary fallible lawyer, leaves the Court with a compelling belief that justice may not have been done in this case.
Mr. Echavarria may well have murdered Daniel Rodriguez or he may well be factually innocent. Either way, the Court has scrutinized the record in this case and has been cautious not to let deep personal concerns regarding factual innocence color its determination that Mr. Echavarria received ineffective assistance of counsel. This Court is left with uncertainty that the Defendant’s guilt has been fairly adjudicated.
ORDER
For the foregoing reasons, the defendant’s motion for a new trial is ALLOWED.