SPINA, J.

The defendant was convicted of murder in the first degree on theories of deliberate premeditation and extreme atrocity or cruelty. He filed a motion for a new trial alleging ineffective assistance of counsel, and he requested an evidentiary hearing. The trial judge denied the motion without a hearing. Her indorsement in the margin said, "for the reasons stated in [the Commonwealth's] opposition." On appeal the defendant alleges error in the denial of his motion for a new trial, the judge's failure to make findings, and the judge's failure to hold an evidentiary hearing on the motion. We affirm the conviction and the denial of the defendant's motion for a new trial. We decline to exercise our power under G. L. c. 278, § 33E.

1. Background. The defendant moved into his girl friend's third-floor apartment in the Dorchester section of Boston in the middle of February, 2008. His girl friend, the victim, had four children, the oldest of whom was six years old. On March 8, 2008, Kristina Ortiz visited the victim at her apartment. The defendant and the victim's four children were there. As Ortiz was leaving, the defendant made a disparaging remark about the victim's children.

That evening the victim sent her six year old son down to the first-floor apartment of a neighbor three times to ask the neighbor to come up to his mother's apartment. Each time the neighbor said she would be right up, but became distracted by her own children and failed to appear. At 9 P.M. the defendant went down to the first-floor apartment and told the neighbor that his "wife was waiting" for her. The neighbor went up to the victim's apartment at around 9:30 P.M. The victim asked the neighbor if the neighbor knew where she could get some cocaine. The neighbor was surprised because she knew the victim was trying to stop using cocaine. The neighbor said she did not know, and left after a brief conversation.

Sometime between 2 and 3 A.M. on March 9 the first-floor neighbor heard "an unusual thud" from an apartment above hers. The victim's apartment was two floors directly above her apartment, but the neighbor could not tell if the noise had come from the victim's apartment. Shortly thereafter she heard footsteps coming down the stairs. She went back to bed.

At about 11:15 A.M. on March 9 the victim's two eldest children appeared at the first-floor neighbor's apartment. The oldest child said, "My mommy and daddy had a fight and he killed her. She's dead." He added that the defendant had left. The next oldest, who was five years old at the time of the incident, testified at trial to the physical beating he saw the defendant inflict on his mother. He saw the defendant push her under a leg of the kitchen table and then sit on the table. The defendant then locked the children in their bedroom.2 The neighbor went upstairs and found the victim lying lifeless on the kitchen floor in a pool of blood. An electrical cord was pulled tight around her neck. The kitchen was in a state of disarray: furniture was overturned, the kitchen table was broken, and laundry was strewn about the room. The neighbor gathered the children, brought them to her apartment, and telephoned the police.

In the meantime, at about 10 A.M. on March 9, the defendant had gone to the home of Doris Serrano, where the defendant's father lived in the basement. He told his father that the victim had "kicked [him] out." His father asked about scratches on the defendant's face. The defendant explained that the victim had scratched him. The defendant left his duffel bag and knapsack in his father's room and went out to have a beer. Later that afternoon the defendant visited his cousin Iliana Pagan (Serrano's daughter), who was a close friend of the victim. Pagan's fiancé was present. The defendant explained that the victim had scratched his face during an argument over drugs. During the defendant's visit Pagan received a telephone call in which she learned that the victim had been found dead in her home. Pagan burst into tears. When her fiancé asked what was wrong, she broke the news in a voice loud enough for the defendant to hear. The defendant said nothing. He bowed his head and put his face in his hands.

Police tried to locate the defendant. They went to Serrano's apartment and asked if Serrano would get in touch with him. Serrano reached the defendant by cellular telephone and told him that his father was looking for him. The defendant returned to Serrano's apartment within minutes. The police asked him to accompany them to Boston police headquarters for questioning. He agreed.

The defendant made a statement that was audiorecorded by police. He told police that he loved the victim and was supposed to marry her. He described what had happened the night of March 8, saying that the victim went "bi-polar" on him. He tried to hug her, but she scratched his face. She threatened to kill herself and call the police if he did not leave. He gathered all his belongings into a duffle bag (which was "heavy") and a backpack, and then left. He took a bus to his father's home, arriving at about 1 A.M. He denied striking the victim or killing her. He also said he loved her children. The defendant said he could not have hit the victim with the kitchen table because he has arthritis and scoliosis and could not lift heavy objects.

The pathologist who performed the autopsy determined that death was caused by a combination of ligature strangulation (probably by the electrical extension cord found around the victim's neck) and a sharp incision to the front of the victim's neck that severed her right carotid artery and jugular vein and completely divided her trachea (windpipe). The strangulation occurred before the incision wound. The victim had suffered blunt trauma to her head. She also had been exposed to a caustic chemical, such as bleach, after death. The pathologist could not determine if the incision wound was caused by drawing a sharp blade from right to left or from left to right.

Police recovered the duffel bag and backpack the defendant had left in his father's room. Inside the duffel bag was a "CharlieCard," a fare card used for Massachusetts Bay Transportation Authority (MBTA) services, that had been used at 11:33 P.M. on March 8 on an MBTA bus that passed within a few blocks of the victim's apartment. Also inside the duffel bag was a receipt from a 7-Eleven store that evidenced a cash purchase at 12:02 A.M. on March 9, 2008. The backpack contained personal items, including a notebook, a pair of sandals, and some clothing.

The notebook had served as a journal. The defendant had made an entry on January 11, 2008, in which he wrote:

The tread on the defendant's left sandal was similar in size and pattern to a footwear impression made in blood within a few feet of the victim's body. The impression left at the crime scene lacked sufficient detail to support a definitive comparison.

The victim was found to be a potential source of deoxyribonucleic acid (DNA) evidence recovered from reddish-brown stains on the heel of the defendant's right sandal, three areas on the defendant's duffel bag, and the handle and blade of a knife found in the victim's kitchen sink, as well as a brown stain on the defendant's shirt, where 1 in 39 quintillion Caucasians, 1 in 1.7 sextillion African-Americans, and 1 in 260 quadrillion Southeastern Hispanics would have the same genetic profile. The victim was also determined to be a possible source of DNA recovered from reddish-brown stains on the defendant's denim pants containing a mixture of DNA from two individuals, where 1 in 44 trillion Caucasians, 1 in 2.5 quadrillion African-Americans, and 1 in 1.8 trillion Southeastern Hispanics would have the same genetic profile. Both the victim and the defendant were determined to be potential contributors to a mixture of DNA from three or more individuals found on the upper half of the sole of the defendant's right sandal.

The defense theory was that the defendant did not kill the victim. He had no motive to kill the victim, whom he loved, and he left her apartment after they argued. He contended there was not enough time between the visit by the first-floor neighbor at 9:30 P.M. and the CharlieCard activity at 11:33 P.M. for him to kill the victim, pack his belongings, and attempt to cover his tracks at the scene with bleach or other caustic substance.

2. Standard of review. The defendant asserted multiple claims of ineffective assistance of counsel in his motion for a new trial. Because he has been convicted of murder in the first degree and his appeal from the denial of his motion for a new trial has been consolidated with his direct appeal, we consider his claims of ineffective assistance of counsel to determine if any error has created a substantial likelihood of a miscarriage of justice, as required by G. L. c. 278, § 33E. This standard of review is more favorable to the defendant than the constitutional standard for determining ineffective assistance of counsel. See Commonwealth v. Wright, 411 Mass. 678, 682 (1992). Under this more favorable standard, we consider whether there was error by trial counsel, regardless of whether trial counsel's performance fell measurably below that of an ordinary fallible lawyer and, if there was, whether the error was likely to have influenced the jury's verdict. Id. However, a strategic decision by an attorney constitutes error only if it was manifestly unreasonable when made. See Commonwealth v. Smith, 456 Mass. 476, 482 (2010).

A judge is required to conduct an evidentiary hearing on a motion for a new trial only if a substantial issue is raised by the motion or affidavits. See Commonwealth v. Wallis, 440 Mass. 589, 596 (2003); Mass. R. Crim. P. 30 (c) (3), as appearing in 435 Mass. 1501 (2001). In that regard a judge considers the seriousness of the issues raised and the adequacy of the defendant's showing on those issues. See Commonwealth v. DeVincent, 421 Mass. 64, 67 (1995). A judge is not required to credit assertions in affidavits submitted in support of a motion for a new trial and may evaluate them in light of factors pertinent to credibility, including bias, self-interest, and delay. See Commonwealth v. Grant, 426 Mass. 667, 673 (1998). A judge may rely on his or her knowledge of the trial and evaluation of the witnesses and evidence at the trial when reaching a decision on a motion for a new trial, including whether to decide the motion without an evidentiary hearing. See Commonwealth v. DeVincent, supra at 69.

The judge must make findings of fact necessary to resolve the defendant's allegations of error of law in a motion for a new trial. See Mass. R. Crim. P. 30 (b), as appearing in 435 Mass. 1501 (2001). A judge's failure to make findings required by rule 30 (b) is "not fatal ... where the ultimate conclusion is clearly evident from the record," Commonwealth v. Lanoue, 392 Mass. 583, 586 n.2 (1984), or where we are satisfied that "on review of the whole case manifest injustice would [not] result," Commonwealth v. Preston, 393 Mass. 318, 322 n.4 (1984). See Commonwealth v. Dunnington, 390 Mass. 472, 478 (1983).

3. Discussion. We turn to the defendant's claims of ineffective assistance of counsel.

a. The defendant faults trial counsel for failing to exploit the time difference between the defendant's use of his CharlieCard at 11:33 P.M. on March 8 and the "unusual thud" heard by the first-floor neighbor on March 9 between 2 and 3 A.M. followed by the sound of footsteps she heard shortly thereafter going down the stairs, both of which the Commonwealth relied on to link the defendant to the killing. He also faults counsel for failing to request an alibi defense and pursue a third-party culprit defense (which he claims would explain the thud and footsteps heard by the first-floor neighbor).

In his closing argument trial counsel did in fact highlight inconsistencies in the timeline. However, the time of death had not been established by the autopsy. The pathologist could only opine that the victim had been dead more than twenty-four hours by the time of the autopsy on March 11, 2008. Because it was not clear precisely when death occurred, an alibi defense would not likely have succeeded where the defendant did not have an alibi for the time between 9:30 and 11:33 P.M. on March 8, when the jury could have determined that the killing occurred.

Had trial counsel pursued a third-party culprit defense, about which the defendant offers no details, the Commonwealth was ready to offer evidence of the defendant's jailhouse confession to another inmate that included many details not publicly known, together with a detailed diagram of the crime scene. After trial counsel announced that the defendant would not testify, the prosecutor decided not to call as a witness at trial the inmate to whom the defendant had confessed. Counsel's decision to make do with what he had rather than pursue a highly risky strategy not likely to produce favorable results (but likely instead to yield highly damaging admissions from the defendant) was a reasonable tactical decision we infer from the record. "Neither ineffectiveness nor a likelihood of a miscarriage of justice arise from counsel making the best he can out of the circumstances of the crime." Commonwealth v. Hung Tan Vo, 427 Mass. 464, 471 (1998).

This theory of ineffectiveness was not supported by any affidavit filed on behalf of the defendant.3 We are satisfied that no substantial question was presented that required an evidentiary hearing, and that on review of the whole case there was no substantial likelihood of a miscarriage of justice in the absence of written findings by the judge. The existence of a reasonable tactical decision by trial counsel in proceeding as he did is clearly evident from the record.

b. The defendant contends that trial counsel was ineffective for failing to object to testimony by Serrano to the effect that she lied to the defendant by telling him his father was looking for him. She said she knew that if she had said the police were looking for him he would not come. The defendant argues that this was impermissible comment on his credibility. See Commonwealth v. Triplett, 398 Mass. 561, 567 (1986).

Although probably objectionable, Serrano's remark did not create a substantial likelihood of a miscarriage of justice. Contrary to the defendant's assertion, the prosecutor did not refer to, or even allude to, Serrano's statement in his closing argument. Trial counsel elicited from Serrano on cross-examination that she had been with the victim and the defendant for about fifteen minutes on March 6, two days before the victim's death, and that they appeared to be getting along. Serrano detected no tension between them at the time. Moreover, the defendant went with police voluntarily to give a statement, offering no resistance. We are satisfied that Serrano's remark was fleeting and isolated. It was hardly the kind of prejudicial comment that permeates the testimony of a key Commonwealth witness on a critical issue in the case such that a new trial is required. Compare Commonwealth v. Triplett, supra. Counsel was not ineffective.

c. The defendant asserts that trial counsel was ineffective for failing to recognize that the incision on the victim's neck was caused by a left-handed person, for failing to recognize from available information that the defendant was right-handed, and for failing to consult with an appropriate expert to show that the defendant could not have caused the incision wound. This is the only claim of ineffective assistance of counsel supported by affidavit.

The defendant presented the affidavit of a physician who is a recognized expert on knife wounds and has testified as an expert both for the Commonwealth and for defendants. Based on the autopsy report and autopsy photographs, it was the physician's opinion that "[t]he pattern of this incision [wound to the neck] is most consistent with an assailant delivering the incision using his left hand while positioned behind the victim." The defendant filed an affidavit stating that he is and always has been right-handed. Affidavits from his mother and his older sister similarly attested to his right-handedness. The defendant also submitted medical records indicating two injuries purportedly consonant with right-handedness.

The defendant's assertions that trial counsel failed to recognize critical details is purely speculative. In addition, even if this defense had been presented to the jury, it likely would not have influenced the jury's conclusion. See Commonwealth v. Wright, 411 Mass. at 682. The defendant told the detectives who interviewed him, "I cannot lift heavy objects ... I cannot really grasp, like grasp, certain things .... Any time I try to grasp something hard, all I feel is a pain and it goes straight numb cuz you can feel the bone right here. I don't do much lifting. I can't exercise." Although the defendant may be right-handed, he told police he is unable to use his major hand for rigorous projects. Moreover, he has not claimed that he could not grasp something, such as a knife, with his left hand and use it to cut the victim's throat. Nor has he claimed that he would not have been able to strangle the victim with an electrical cord. The defendant admitted carrying his heavy duffel bag and his backpack when he left the victim's apartment to go to his father's home. The jury could well have believed that the defendant could not have cut the victim's throat with his right hand, but instead used his left hand. Moreover, the evidence strongly suggested that the victim's throat was cut after she was strangled. The jury could have concluded that the neck wound was inflicted after the victim collapsed to the floor, not while the defendant was standing behind her, as the defendant's expert implied. Cutting the victim's throat while she was in that condition would not have been difficult, even for someone with disabilities the defendant claims to have.

The defendant has not raised a substantial issue about his right-handedness that would have required the judge to hold an evidentiary hearing. The record strongly refutes a conclusion that only a left-handed person could have cut the victim's throat. Written findings were not required to resolve any issues. We conclude that the defendant has not shown that counsel was ineffective in failing to pursue this issue.

d. Contrary to the defendant's argument, trial counsel was not ineffective for failing to request a Daubert-Lanigan hearing, or otherwise failing to object to or moving to strike the expert testimony concerning the comparison of the treads on the defendant's footwear with a footwear impression made in the blood at the crime scene. See Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993); Commonwealth v. Lanigan, 419 Mass. 15 (1994). The defendant emphasizes that of the four factors considered by the Commonwealth's expert, namely, (1) pattern, (2) size, and two individualizing factors — (3) wear and tear, and (4) distinctive (random) marks — the witness acknowledged that only factors (1) and (2) were similar, and because there were insufficient details as to factors (3) and (4) to enable the witness to conclude that they, too, were similar, the witness should not have been permitted to give an opinion that essentially was speculative. The expert opined that a bloody footprint impression at the crime scene could have been made by the defendant's left sandal, but he could not give a definitive opinion.

Judges have broad discretion in deciding whether to admit expert testimony. Commonwealth v. Fitzpatrick, 463 Mass. 581, 603 (2012). The test is whether the testimony "will assist the trier of fact to understand the evidence or to determine a fact in issue." Mass. G. Evid. § 702 (2014). See Commonwealth v. Dockham, 405 Mass. 618, 628 (1989). An expert opinion that is not definitive but expressed in terms of observations being "consistent with" a particular cause, or words of similar effect, does not render the opinion inadmissible on the ground that it is "speculative." See Commonwealth v. Azar, 32 Mass.App.Ct. 290, 302-303 (1992), S.C., 444 Mass. 72 (2005).

Here, the expert was asked if he had an opinion "to a reasonable degree of scientific certainty" whether the defendant's left sandal was "consistent with" the bloody footprint observed at the crime scene. He said he had such an opinion, and that the defendant's left sandal "could have" made the bloody impression, but the impression "was not detailed enough for a more definitive conclusion." Having explained to the jury how he applied the four factors, he further explained how they shaped his opinion, which essentially neither excluded the defendant's sandal nor led him to opine regarding the existence of a definitive match. Instead he was led to an inconclusive result. Trial counsel made this quite clear in his cross-examination, that is, the expert's opinion did not express the existence of a definitive match. The expert's opinion was not improper. See Commonwealth v. Azar, supra.

The defendant contends that where only two out of the four factors produced positive results, the expert's opinion did not even rise to the level of a preponderance of the evidence and thus was speculative. We disagree. There is no suggestion in the record that a proper analysis requires a particular "score" among the four factors. Rather, as with many areas of forensic science, prescribed factors that must be applied when considering a particular matter ultimately involve a matter of judgment, and are intended to guide and shape the expert's reasoning. How the expert proceeds with the application of those factors is usually fertile ground for cross-examination.4 What is often crucial is how the expert presents his or her opinion and analysis to the jury. Of particular concern is the danger that the jury is misled into an understanding that the "science" at hand is "hard" science, when in fact it is "soft" science. There was no danger of that in this case. The jury were given the visual images of the defendant's left sandal and the bloody impression made at the crime scene, and the expert led the jury through the factors he applied in his analysis. Although the witness should not have been asked if he had an opinion to a reasonable degree of scientific certainty but instead should have been asked if he had an opinion to a reasonable degree of certainty within the study of footwear impression comparison, or comparable words, it was readily apparent to the jury that the opinion given was a matter of judgment, and not a scientific result driven by precise mathematical calculations. See Commonwealth v. Pytou Heang, 458 Mass. 827, 848-850 (2011). It was made clear to the jury that this was a matter they could weigh for themselves, and the judge instructed them that they could accept or reject an expert's opinion.

The evidence had probative value that was enhanced when juxtaposed with the expert testimony about the DNA evidence from the defendant's right sandal. The prosecutor did not argue unfairly from the testimony of the expert on footwear impression, as the defendant contends. The prosecutor spent a significant amount of time discussing the DNA evidence in his closing argument. At the end of that discussion he spoke briefly about the footwear impression testimony, arguing essentially that, when viewed together, the DNA testimony and the footwear impression testimony provided strong circumstantial evidence that the defendant was the person who killed the victim because the victim's blood made its way on to the defendant's right sandal at about the same time the defendant stepped in her blood and left a footwear impression with his left sandal. The powerful synergistic effect of the expert testimony was an entirely reasonable and proper inference to draw.

Finally, the defendant has not shown that had trial counsel moved for a hearing under Commonwealth v. Lanigan, supra, the Commonwealth's expert probably would not have been allowed to testify. Cf. Commonwealth v. Comita, 441 Mass. 86, 91 (2004) ("in order to prevail on an ineffective assistance of counsel claim on the ground of failing to file a motion to suppress, the defendant has to demonstrate a likelihood that the motion to suppress would have been successful").

For the foregoing reasons, we conclude that counsel has not been shown to be ineffective as to this claim.

e. There is no merit to the defendant's claim that counsel was ineffective for arguing on manslaughter in his closing argument, a theory that was inconsistent with the primary theory at trial, which was that the defendant was not the killer. The Commonwealth's case was very strong, and trial counsel carefully avoided the admission of the defendant's jailhouse confession. Trial counsel requested a manslaughter instruction based on theories of heat of passion and reasonable provocation (the victim first scratched his face and then "kicked [him] out"). His argument to the jury was more in passing than it was inconsistent with the primary trial strategy. He argued at the very end of a closing argument that spanned approximately twenty-three pages of the transcript, "You must return — must return a verdict of not guilty. And at the most, at most, the government has proven manslaughter." The argument was hardly the "abrupt switch" in strategy about which the defendant complains. Rather, in the context of the entire closing argument and the entire trial, it was the gentle planting of a small seed. It served primarily as a quiet introduction to the judge's instructions, and not a shift in strategy. The requested instruction also gave the jury, and the defendant, an additional option between guilty of murder and not guilty of murder. Without an affidavit from the defendant or counsel as to what, if anything, was discussed in this regard, we cannot say that counsel's strategy was manifestly unreasonable. This is especially true in light of the understated manner in which counsel proceeded on this issue.

Finally, the judge's instruction on manslaughter was the model instruction. Counsel's failure to object to the instruction was not ineffective assistance of counsel. See Commonwealth v. Tassinari, 466 Mass. 340, 356-357 (2013) (manslaughter charge nearly verbatim to model instruction — no error). Taken as a whole, we think the jury understood that a verdict of guilty of murder in the first degree required proof beyond a reasonable doubt of the absence of reasonable provocation and the heat of passion, and that there was no error as in Commonwealth v. Acevedo, 427 Mass. 714, 717 (1998).

For the foregoing reasons, we conclude that trial counsel has not been shown to have been ineffective. We also discern no error in the denial of an evidentiary hearing, and we conclude that there was no substantial likelihood of a miscarriage of justice in the judge's failure to make written findings.

4. Review under G. L. c. 278, § 33E. We have reviewed the briefs and the entire record and conclude that there is no reason for us to reduce the degree of guilt or order a new trial.

Judgment affirmed.

Order denying motion for a new trial affirmed.