In 1996, the defendant was convicted of murder in the first degree and unlawful possession of a firearm in connection with the shooting death of Carlos Cruz. Based on a showing of ineffective assistance of counsel, the convictions were reversed and the case remanded to Superior Court for a new trial. Commonwealth v. Sena, 429 Mass. 590 (1999). The defendant’s second trial again resulted in a conviction on both indictments. The defendant filed a motion for a new trial, which was assigned to and denied by another judge (as the trial judge no longer served in the Superior Court). On appeal from the convictions and the denial of his motion for a new trial, the defendant claims that (a) counsel was ineffective in that he failed to take any action with respect to alleged underrepresentation of minorities in the jury venire; (b) counsel was ineffective in that he failed to retain an expert witness; (c) “newly discovered” evidence impugning the credibility of the medical examiner warrants a new trial; and (d) the judge erred in admitting the prior recorded testimony of a witness from the first trial. We affirm the convictions, affirm the order denying the defendant’s motion for a new trial, and decline to grant relief under G. L. c. 278, § 33E.
1. Background. We summarize the evidence in the light most favorable to the Commonwealth, reserving certain details for discussion in conjunction with the issues raised.
In the fall of 1991, the defendant was dating Maria Diaz, the mother of the seventeen year old victim, Carlos Cruz. Diaz terminated the relationship in January, 1992, when her son informed her that the defendant was seeing other women. At that time, the defendant threatened to kill Cruz.
On the night of March 21, 1992, Diaz, her new boy friend, Angel Baretto, and her son Cruz were attending a friend’s baby shower at 94 Lewis Street in Lowell. Diaz and Baretto left the party briefly and then returned. As they approached the house, they observed the defendant and Maureen Otero driving down Lewis Street. Diaz flagged down the defendant and began to argue with him. Baretto joined in, and a fistfight ensued, with Baretto the apparent victor. The fight was observed by various *824persons attending the baby shower. When the fight ended, the defendant and Otero drove to the defendant’s apartment on nearby Middlesex Street. The defendant was “mad,” and told Otero that “he was gonna get them back.” The defendant went into his apartment briefly, leaving Otero in the car. When he returned, they drove back toward Lewis Street and parked on another street nearby. Otero described the defendant as still “mad.” The defendant left Otero behind in the parked car, and proceeded to 94 Lewis Street.
Max Maldonado, the father of the baby for whom the party was being held, observed the defendant looking into the front window of the house at 94 Lewis Street. Maldonado saw the defendant motion for Cruz to come outside.1 According to Maldonado, Cruz came outside and stood on the front steps with the defendant. Maldonado then saw the defendant “raise[] his hand and the shot went off.” The defendant’s hand “was pointed at the kid’s face.” Cruz fell to the ground, and the defendant ran off.
Angel Luis Nieves, then eleven years old, was also in attendance at the baby shower. Nieves testified that it was he who summoned Cruz to go to the front door when the defendant signaled for Cruz. Nieves then followed Cruz to the front door and stood in the doorway with Maldonado. According to Nieves, the defendant “pulled out his hand and shot Carlos Cruz.” Nieves then saw the defendant “ scatter[]” toward the back alley near 94 Lewis Street.
Otero was waiting for the defendant where he had left her. The defendant got in the car and “took off.” The defendant told Otero that he had “hit somebody and the gun went off.” As the defendant made this statement to Otero, he gestured with his right arm, extending the arm in front of him and moving it across from right to left. The import of this statement and gesture, and the theory of the defense, was that the gun had gone off by accident when the defendant had used the gun to *825strike Cruz. The defendant told Otero that he was leaving for New York (where he was later found).
Cruz died of a single gunshot wound to the head. Dr. John Krolikowski, the medical examiner who performed the autopsy on Cruz, testified concerning the wound, which was immediately to the left of the victim’s nose. Based on his observations, Dr. Krolikowski opined that the “weapon [was] fired very close to that surface, the facial surface.” Dr. Krolikowski did not observe any evidence of scratches, abrasions or contusions that would reflect the gun’s striking Cruz on the side of the face. On cross-examination, he testified that the wound was “consistent with a near contact wound,” but not a “total contact wound.” He also acknowledged that there was an abrasion on the victim’s face, which he attributed to “blunt trauma.”
State police Sergeant John Busa had retrieved a .38 caliber semiautomatic weapon from underneath a parked car in front of 94 Lewis Street. Testing confirmed that it was the gun used in the shooting of Cruz.2 Sergeant Busa pointed out the sharp front edges on the barrel end of the gun’s slide, opining that if the gun had been used to strike someone, the victim would have been cut by those sharp edges. Sergeant Busa also testified concerning his observations of the wound on Cruz’s face.3 In his opinion, the stippling on one side of the wound indicated that the gun had been “almost pressed to the skin at the time of the discharge.”
2. Discussion, a. Ineffective assistance of counsel. “In evaluating a claim of ineffective assistance of counsel in a case of murder in the first degree, we begin by determining whether there was a serious failure by trial counsel. If so, then we determine whether the failure resulted in a substantial likelihood of a miscarriage of justice.” Commonwealth v. Harbin, 435 Mass. 654, 656 (2002). We give deference to trial counsel’s tactical decisions, see Breese v. Commonwealth, 415 Mass. 249, 251 (1993), and unless such a decision was “manifestly *826unreasonable when made,” we will not find ineffectiveness. Commonwealth v. LaCava, 438 Mass. 708, 713 (2003), quoting Commonwealth v. Martin, All Mass. 816, 822 (1998). The defendant argues that his trial counsel was ineffective in failing, to request that certain jurors be selected “out of turn” to compensate for underrepresentation of minority jurors in the venire, and in failing to call an expert witness on “the critical issue . . . whether or not the gun had made contact with the victim’s face.” Both claims lack merit.
i. Jury venire. During empanelment, defense counsel observed that there was only “one black person” and “one person of possible Hispanic extraction” in the venire. The defendant now argues that counsel was ineffective when he failed to request any remedy for this alleged underrepresentation of racial and ethnic minorities, specifically, a request that the judge seat the two minority members observed in the venire out of turn instead of proceeding with a random draw of jurors. He also asks that we “promulgate a rule encouraging a trial judge to [ejmpanel jurors out of order in instances such as this, where the minority underrepresentátion is blatantly obvious.”
The record before us does not support the defendant’s claim of unconstitutional underrepresentation of minorities in the venire. To establish such a claim, the defendant must demonstrate that (1) the group allegedly discriminated against is a distinctive group in the community, (2) that the group is not fairly and reasonably represented in jury venires in relation to its proportion of the community’s population, and (3) that the underrepresentation is due to systematic exclusion of the group in the jury selection process. Commonwealth v. Arriaga, 438 Mass. 556, 562-563 (2003); Commonwealth v. Tolentino, 422 Mass. 515, 518-519 (1996). Here, the defendant has failed to make an adequate showing on the second and third elements of his claim, and there can be no ineffective assistance in counsel’s failure to request a remedy for alleged underrepresentation that has not been established.
Based on the comment made by trial counsel identifying one African-American and one possible Hispanic juror in the venire, the defendant calculates the percentage of those minorities in the venire at approximately one per cent, a figure below the *827proportion of those minorities in the Middlesex County population at large.4 A mere visual assessment of the jurors, however, is not a sufficient basis on which to establish the racial and ethnic composition of the venire. See Commonwealth v. Tolentino, supra at 520, and cases cited. Furthermore, the single venire brought into the court room for the defendant’s trial “is an unacceptably small sample for the purpose of any statistical showing of underrepresentation.” Commonwealth v. Arriaga, supra at 564. Rather, “[a] defendant must present evidence of a statistically significant sample, usually requiring analysis of the composition of past venires.” Id. In this case, no such evidence has been presented, and the defendant has failed to demonstrate any underrepresentation of minorities in the Middlesex County jury pool.5 Defense counsel’s failure to request a remedy for alleged underrepresentation therefore did not constitute ineffective assistance of counsel.6
ii. Failure to call expert witness. The defense theory was that the victim’s death resulted from an accidental firing of the gun when the defendant used it to strike the victim in the face. To advance this theory, defense counsel effectively cross-examined the Commonwealth’s witnesses, eliciting their opinions on aspects of the forensic evidence that might be consistent with the defense version. There was testimony from several expert witnesses as to how close the gun was to the skin surface of the victim’s face at the time it went off (the defense view being that actual contact at the time of firing would be consistent with the gun’s striking the victim’s face and discharging on the force of that blow). Dr. Krolikowski had testified on direct examination that the gun was “very close” at the time it discharged, characterizing the wound as a “near contact wound” (meaning “near the surface” of the skin). On cross-examination, defense *828counsel obtained clarification that by the term “near,” Dr. Krolikowski meant that the distance could be as little as one inch. Then, Sergeant Busa’s opinion placed the gun even closer, testifying that the gun was “almost pressed to the skin at the time of the discharge, but. . . at an angle,” which he characterized as an “incomplete contact” wound. Finally, the Commonwealth’s forensic chemist testified that she found occult blood inside the barrel of the gun (but no evidence of blood on the outside surface), consistent with “a close contact shot.”
Defense counsel also, on cross-examination of Dr. Krolikowski, introduced evidence that, in addition to the bullet wound itself, the victim’s face had an abrasion and contusion caused by “blunt trauma.” He also elicited testimony concerning Sergeant Busa’s “shock testing” on the gun, with Sergeant Busa conceding that the gun had fired once after being struck from behind several times with a hammer (thereby illustrating that the gun could discharge accidentally under the force of a blow). Defense counsel then used this forensic evidence in closing, arguing to the jury that it was consistent with the version of events that the defendant had relayed to Otero in the immediate aftermath of the incident.
The defendant now contends that his counsel was ineffective in failing to retain his own expert witness to address these forensic issues. At the outset, we note an anomaly in the claim. On appeal following the first trial, the defendant’s then appellate counsel made the same argument, i.e., that trial counsel at the first trial had been ineffective in that he had failed to consult any forensic expert who could help analyze this evidence to support the defendant’s theory of an accidental firing.7 Having obtained a reversal and remand for a new trial, the attorney who represented the defendant on that successful appeal proceeded to represent him at the second trial. In other words, the same attorney who had contended that failure to present expert testimony, or “even consult a forensic pathologist,” was ineffective assistance of counsel mandating a new trial, chose not to *829retain an expert or present testimony from a forensic expert at the second trial. Against this backdrop, it would be difficult to characterize counsel’s decision with respect to the forensic evidence as anything other than deliberate and tactical.8
Whatever the explanation for counsel’s failure to retain an expert to assist the defense on retrial, the expert affidavit submitted with the motion for a new trial fails to support the claim that that failure resulted in a substantial likelihood of a miscarriage of justice. To begin with, the defendant’s claim attributes undue significance to whether the gun was in physical contact with the victim’s skin at the moment of discharge. As the motion judge noted, “Contact between the gun and the victim’s face neither proves nor disproves an accident. Shooting someone in the face could be a premeditated murder whether the gun barrel was touching the skin, pushed into the skin or a short distance from the skin at the time the shot was fired.”'
Moreover, to the extent that establishing the gun’s proximity to or contact with the victim’s face was helpful to the defense as at least being consistent with his accident theory, the expert affidavit now before us does not offer anything on that issue beyond what was already acknowledged by the Commonwealth’s witnesses and conceded in the prosecutor’s closing argument. Indeed, Carl Majeskey, the expert now proffered by the defendant, expresses the opinion that Sergeant Busa “evaluate[d] this wound correctly.” Majeskey characterizes the wound as a “partial contact” wound, just as Sergeant Busa called it an “incomplete contact” wound. Based on Sergeant Busa’s testimony, the prosecutor’s closing argument made the very point that the defendant now seeks to make by way of Majeskey’s affidavit: “[Ojne portion of that gun was touching one portion of [the victim’s] nose.” While Majeskey goes on to criticize certain details of Dr. Krolikowski’s testimony, the criticism focuses on points that were never tied to the allegedly critical issue of the degree of contact between the end of the *830gun and the victim’s skin surface (e.g., quibbling that Dr. Krolikowski described the wound as “round” at one time, while referring to it as “oval” or “ovoid” at other times).9 Majeskey’s affidavit, with its stated concurrence in Sergeant Busa’s analysis of the issue, does not suggest that additional expert testimony for the defense would have offered anything of substance beyond that in Sergeant Busa’s testimony (as emphasized in the prosecutor’s closing).
b. Newly discovered evidence. The defendant argues that his motion for a new trial should have been allowed on the ground of newly discovered evidence that allegedly casts doubt on Dr. Krolikowski’s credibility. A defendant seeking a new trial on the ground of newly discovered evidence must first establish that the evidence was not discoverable at the time of trial despite the due diligence of the defendant or defense counsel. Commonwealth v. Jones, 432 Mass. 623, 633 n.6 (2000). Commonwealth v. Salvati, 420 Mass. 499, 507 (1995). The defendant must then show that the newly discovered evidence “casts real doubt on the justice of the conviction.” Id. at 506. In order to obtain a new trial on the ground of newly discovered evidence, there must be “a substantial risk that the jury would have reached a different conclusion had the evidence been admitted at trial.” Commonwealth v. Moore, 408 Mass. 117, 126 (1990), quoting Commonwealth v. Grace, 397 Mass. 303, 305-306 (1986).
Approximately five months prior to the defendant’s second trial, Dr. Krolikowski signed a “Temporary Agreement To Restrict The Practice of Medicine” (temporary agreement), which he then submitted to the Board of Registration in Medicine (board). In that temporary agreement, Dr. Krolikowski agreed not to read prostate biopsy slides for a period of four months pending the outcome of an ongoing board investigation. The board instituted disciplinary proceedings against Dr. Krolikowski in May, 2002, which resulted in a consent order *831prohibiting his practice of surgical pathology. As recited in the consent order, a hospital’s review of approximately 3,000 surgical pathology cases read by Dr. Krolikowski found that twenty-nine of his readings were “sub-optimal” and that three prostate slides had been “misread.” The consent order specified that there had been no complaints alleging substandard practice by Dr. Krolikowski “in the areas of clinical pathology, autopsy pathology and forensic pathology,” the only restriction therefore being in the practice area of “surgical pathology.” At trial, Dr. Krolikowski described his then current employment as a “forensic pathologist” in a “Department of Forensic Sciences,” his training and fellowships in “forensic pathology,” and his board certifications in “anatomic pathology, clinical pathology and forensic pathology.”
The defendant does not — and cannot — contend that information concerning the temporary agreement was not discoverable at the time of the first trial. As such, the temporary agreement, and its referenced disciplinary investigation, cannot qualify as “newly discovered” evidence. Instead, the defendant points to the 2002 consent order, contending that the results of that investigation qualify as newly discovered evidence that casts doubt on the justice of the conviction. We agree with the motion judge that this evidence has no such effect on the defendant’s conviction. First, nothing in the consent order suggests that Dr. Krolikowski was not competent in forensic pathology, the principal area of his training and practice and the sole subject of his testimony at trial. To the contrary, the consent order expressly states that the board’s concerns focused exclusively on Dr. Krolikowski’s reading of surgical pathology slides and specimens, and that there was no indication of substandard practice in other branches of pathology, specifically including his practice of forensic pathology. As such, the proposed use of the consent order would be solely for purposes of a very general attack on Dr. Krolikowski’s credibility. “It is well established that ‘[njewly discovered evidence that tends merely to impeach the credibility of a witness will not ordinarily be the basis of a new trial.’ ” Commonwealth v. Lo, 428 Mass. 45, 53 (1998), quoting Commonwealth v. Ramirez, 416 Mass. 41, 47 (1993). And, as discussed above, the points of disagree*832ment with Dr. Krolikowski’s testimony were themselves minor and inconsequential, and already addressed by Sergeant Busa’s opinion (the opinion the Commonwealth endorsed in its closing). The absence of this sole item of potential impeachment of Dr. Krolikowski does not “cast[] real doubt on the justice of the conviction.” Commonwealth v. Grace, 397 Mass. 303, 305 (1986).
c. Admission of prior recorded testimony. The defendant claims that the judge erred when he allowed the prosecution to introduce the prior recorded testimony given by Fernando Santana at the first trial. Specifically, the defendant contends that the Commonwealth failed to establish that Santana was “unavailable” to testify at the second trial, and that the opportunity to cross-examine Santana at the first trial was inadequate to address alleged new issues that emerged at the second trial. See Crawford v. Washington, 124 S. Ct. 1354, 1367-1369 (2004), and cases cited (testimonial statement of absent witness admissible only if witness unavailable and defendant had prior opportunity to cross-examine).10
Before allowing the Commonwealth to introduce prior recorded testimony, the judge must be satisfied that the Commonwealth has made a good faith effort to locate and produce the witness at trial. Commonwealth v. Roberio, 440 Mass. 245, 248 (2003), quoting Commonwealth v. Siegfriedt, 402 Mass. 424, 427 (1988). “Whether the Commonwealth carries its burden on the question of sufficient diligence in attempting to obtain the attendance of the desired witness depends upon what is a reasonable effort in light of the peculiar facts of the case.” Commonwealth v. Lopera, 42 Mass. App. Ct. 133, 136 (1997), *833quoting Commonwealth v. Hunt, 38 Mass. App. Ct. 291, 295 (1995).
Here, approximately one week prior to the scheduled retrial, the Commonwealth attempted to contact Santana, who had been a cooperative witness at the first trial. See Commonwealth v. Hunt, supra at 294 n.3 (Commonwealth not required to “commence its efforts to locate a witness until shortly before trial”). Unable to find Santana at the expected address, the police pursued other avenues, ultimately learning, by way of information from Santana’s roommate, that Santana was in Vega Alta, Puerto Rico. However, no one had an address or telephone number for him there. The Puerto Rican authorities were enlisted to help search for Santana, and they checked possible addresses under that name (which they described as a very common name in Puerto Rico). By the fourth day of trial (the day on which the Commonwealth rested its case), Santana had still not been found. The Commonwealth met its burden of demonstrating a good faith diligent effort to locate Santana.
It is undisputed that the defendant had an opportunity to cross-examine Santana at the first trial, and in fact cross-examined him extensively. The defendant claims, however, that that cross-examination was inadequate in that it did not probe all of the inconsistencies between Santana’s multiple statements to the police and his trial testimony. That the earlier cross-examination did not cover every detail and every possible avenue of impeachment that counsel would now like to pursue does not change the fact that the defendant had the requisite opportunity for cross-examination. See Commonwealth v. Roberio, supra at 250-252; Commonwealth v. Childs, 413 Mass. 252, 262-264 (1992); Commonwealth v. Siegfriedt, supra at 428-429.
The defendant also contends that the prior opportunity for cross-examination was inadequate because there was no occasion to question Santana about whether Nieves had witnessed the shooting, as Nieves did not testify at the first trial. That a subsequent trial involves additional evidence introduced against the defendant does not mean that the opportunity for cross-examination at an earlier trial is inadequate to satisfy the confrontation clause, and nothing in Nieves’s testimony at the second trial introduced new, substantive issues into the case. *834Nieves testified that he and Maldonado were at the doorway at the time of the shooting and that he, as did Maldonado, saw the defendant raise his hand and shoot the victim. The purpose of Nieves’s testimony was to bolster that of Maldonado with respect to this point.11 Santana, however, had observed the earlier fight and had seen the defendant come up to the doorway and signal the victim to come out, but he had not seen the actual shooting. He had only heard the sound of the gun firing, and thus had nothing to add with respect to whether the defendant had shot the victim on purpose or had struck him in the face resulting in the accidental discharge of the weapon. In short, Santana’s testimony could neither add nor detract from Nieves’s testimony with respect to what the defendant has consistently characterized as the crucial issue in the case.
The defendant suggests that further cross-examination of Santana was necessary to address whether Nieves was in fact present at the doorway at the time of the shooting, and that he thus had no opportunity to use Santana’s testimony to undercut the entirety of Nieves’s version of events. This argument ignores the fact that, as given, Santana’s testimony at the first trial made no mention of seeing Nieves in the vicinity of the shooting, but only made reference to Maldonado. There was nothing to “cross-examine” Santana about on this issue, as the version he gave was already favorable to the defense on that point. (Indeed, the defense was better off with a version of Santana’s testimony that made no reference to Nieves whatsoever than it would have been with an opportunity for Santana to add any further information as to Nieves’s location at the time of the shooting.) At the first trial, the defendant had every opportunity and motive to question Santana about what he saw, including which party guests were or were not in a position to see the confrontation between the defendant and Cruz. Nothing about Nieves’s later testimony so changed the prosecution’s evidence as to undermine the defendant’s prior opportunity and motive to cross-examine Santana on all relevant issues. As such, we find no error in the judge’s decision to admit that prior recorded testimony.
*835d. Review under G. L. c. 278, § 33E. We have reviewed the entire record as required by G. L. c. 278, § 33E, and conclude that there is no basis for ordering a new trial or directing the entry of a verdict of a lesser degree of guilt.
Judgments affirmed.
Order denying amended motion for new trial affirmed.