NOTICE: This opinion is subject to modification resulting from motions for reconsideration under Supreme Court Rule 27, the Court’s reconsideration, and editorial revisions by the Reporter of Decisions. The version of the opinion published in the Advance Sheets for the Georgia Reports, designated as the “Final Copy,” will replace any prior version on the Court’s website and docket. A bound volume of the Georgia Reports will contain the final and official text of the opinion. In the Supreme Court of Georgia Decided: October 22, 2024 S24A0560. BAKER v. THE STATE. S24A0582. LAGUERRE v. THE STATE. PINSON, Justice. Prentice Baker and Verlaine Laguerre were convicted of malice murder and possession of a firearm during the commission of a fel- ony for the shooting death of Matthew Hardeman. The evidence at trial showed that on the day of the shooting, Hardeman and La- guerre got into a fistfight in front of the home of Andrew “Cali” Ellis, a mutual friend and Hardeman’s neighbor. Everyone agreed Harde- man “won” the fight, and Laguerre left after it ended. Later that day, Laguerre returned with Baker and others to Hardeman’s neighbor- hood, where Baker and Laguerre encountered Hardeman in front of Ellis’s home and shot Hardeman multiple times, killing him. 1 On appeal, Baker and Laguerre each raise claims of error. They both contend that the trial court plainly erred by not instructing the jury on accomplice corroboration. Baker contends that there was not sufficient evidence to support his convictions for malice murder and firearm possession. And Laguerre contends that his counsel pro- 1 In March 2012, a Fulton County grand jury jointly indicted Baker and Laguerre for malice murder (Count 1), felony murder (Count 2), aggravated assault (Count 3), and possession of a firearm during the commission of a fel- ony (Count 4). After their first joint trial ended in a mistrial, a joint jury trial from April 10 to 20, 2018, ended with the jury returning guilty verdicts on all counts. On April 20, 2018, the trial court sentenced each defendant to life in prison for malice murder (Count 1) and a consecutive five years in prison for possession of a firearm during the commission of a felony (Count 4); the re- maining counts were merged or vacated by operation of law. Each defendant timely filed a motion for new trial and later amended his motion through new counsel. After a joint hearing on September 22, 2021, the trial court denied each defendant’s amended motion for new trial in separate March 21, 2022, orders. Each defendant timely filed his notice of appeal, and both appeals were docketed to the April 2024 term of this Court and submitted for a decision on the briefs. We have consolidated those appeals for purposes of this opinion. 2 vided ineffective assistance and that the cumulative effect of the in- structional error and his counsel’s deficient performance require a new trial. Each claim fails. Baker and Laguerre have not established plain error because the evidence they marshal to show that Ellis was an accomplice — mostly evidence that Ellis had a relationship with the defendants and victim and was at the crime scene, and Harde- man’s friend’s speculation that Ellis “probably[ ] got something to do with” the shooting — did not obviously call for a sua sponte instruc- tion on accomplice corroboration. Baker’s sufficiency claim fails be- cause the evidence was sufficient to support his convictions. La- guerre has failed to establish that trial counsel performed deficiently by not objecting to a witness’s testimony about the witness’s fear of the defendants because counsel’s decision to use that testimony later in support of the overall defense strategy was objectively reasonable. And without any errors to assess cumulatively, Laguerre’s claim of cumulative error fails. So we affirm Baker’s and Laguerre’s convic- tions and sentences. 3 1. Background (a) The Crimes On October 15, 2011, Hardeman hung out at his sister’s home with three friends: Willie Wimby, Dentrayquise “Tray” Clark, and Andre “Dre” Anderson. On the same day, Ellis, who was Hardeman’s sister’s neighbor, had “a bunch of people” at his home, including La- guerre. Hardeman and Ellis had met before through Hardeman’s sister. At first, Hardeman and Ellis had hung out together, but that stopped when Hardeman started spending time with Wimby. Ellis had warned Hardeman not to hang out with Wimby because, accord- ing to Ellis, Wimby “was into a lot of stuff.” At some point on October 15, Hardeman, Wimby, Clark, and Anderson walked by Ellis’s home while Ellis and Laguerre were out- side. According to Ellis, Hardeman said to Laguerre “something like, ‘What are you looking at me for’ or whatever.” Later that day, Hardeman and Laguerre got into a fistfight in the road in front of Ellis’s home. Clark, Wimby, Anderson, and 4 Clark’s mother all saw the fight. They did not use weapons, but An- derson saw Hardeman hand a pistol to Ellis before the fight and then saw Ellis return the pistol to Hardeman after the fight. Every- one who saw the fight agreed that Hardeman “won.” According to Ellis, Laguerre said after the fight that he was going to get “his redeye,” meaning his gun, and Ellis told Laguerre, “You lose some; just fight [Hardeman] another day.” Ellis also testi- fied that he blamed Wimby for the fight “because he’s the one who got in [Hardeman’s] head” that he needed to fight Laguerre. Ander- son testified that, after the fight, he heard Laguerre say, “I’m from Miami, the murder capital. I’m going to kill you.” Soon after the fight, Laguerre left the neighborhood. Ellis tes- tified that after Laguerre left, Ellis told Hardeman to leave the neighborhood, too, “before they come back . . . They might shoot the house up.” According to Clark, Ellis said, “When [Laguerre’s] brother get here, it’s going to be some gun play. It’s going to be some pistol play. I know this ain’t over with.” 5 After the fight, Hardeman, Clark, Anderson, and Wimby re- turned to Hardeman’s sister’s home. Clark told Hardeman to stay inside, and Clark and Anderson left and returned to their home. Wimby stayed with Hardeman. Meanwhile, Ellis went inside his own home for 15 to 20 minutes, then came back outside. Soon after, Laguerre’s brother and Baker came to Ellis’s home looking for Hardeman. Ellis knew Baker because Ellis, Baker, and Laguerre once lived near each other in a different neighborhood and had hung out together in the past. Ellis testified that Baker had “a little pistol” with him that day, and Baker asked where Hardeman was. Ellis lied and told Baker that Hardeman left after the fight. But while Baker was still at Ellis’s home, Hardeman and Wimby came outside. According to Wimby, he and Hardeman came outside and heard “the suspects” yell “y’all p***y n****s turn around” and he and Hardeman turned around and saw Laguerre and another per- son walking down the street toward them with two other people fol- lowing behind them in a car. Laguerre had a rifle that Wimby 6 thought was an “AK-47.” The person walking with Laguerre took out a handgun and pulled the trigger, but it did not fire at first. The same person cocked the gun again and shot Hardeman. According to Wimby, “the guy who shot [Hardeman] the first time started shoot- ing him some more. Then the guy who [Hardeman] got in the fight with bring [sic] out the AK-47.” Wimby did not know either of the shooters by name but recognized Laguerre because Laguerre would hang out with Ellis, so Wimby said he thought Ellis “probably, got something to do with the s**t right now.” According to Ellis, after Wimby and Hardeman came outside, Laguerre “and them” came around the corner in a car and Laguerre “jumped out the back with a gun.” Baker then cocked his gun, but when he pulled the trigger, it jammed. Baker pulled the trigger again and shot Hardeman in the head. Laguerre fired his gun too, and Ellis heard 40 or more shots. Laguerre and Baker then got back in the car and drove away. Ellis saw about six people in the car, and three of them — Baker, Laguerre, and someone else — had guns. 7 After the shooting, Ellis went back inside his home. The police arrived and sealed off the area outside Ellis’s home. Ellis did not tell the police what he saw at first. But that night, Hardeman’s sister, who had been away from home during the shooting, came to Ellis and asked him what happened. He told her he could not “get in- volved” because he would have to “go to war with them” and they would kill him, but he gave her Laguerre’s name and address. Har- deman’s sister gave that information to the police, and Ellis later spoke to the police and gave them the same address. (b) Photo Lineups Two days after the shooting, Ellis was shown two photo lineups and identified Baker and Laguerre as the men who shot Hardeman. Rosa Clark, Tray Clark’s mother, was also shown two photo lineups. She did not know Laguerre’s name, but she had seen him hanging out with Ellis, whom she knew only by his nickname, “Cali.” Two days after the shooting, Rosa viewed a lineup and identified Ellis as the person she knew as “Cali.” She viewed another lineup and identified Laguerre as the person she saw Hardeman fight on 8 October 15. Rosa testified that she had been between 50 and 100 percent sure of her identification of Laguerre at the time she made it. She also identified Laguerre in the courtroom as the person she saw fight Hardeman. Rosa did not remember ever seeing Baker in the neighborhood, including at the time of the fight between Harde- man and Laguerre. She did not see who shot Hardeman. Tray Clark was also shown a photo lineup, and he identified Laguerre as the person who fought Hardeman. He also identified Ellis in a photo lineup as the person who made the comment about “gunplay.” Tray was not present at the shooting and had never seen Baker before the trial. Wimby viewed three photo lineups after the shooting. In the first lineup, he identified Ellis as the person he knew as “Cali.” He viewed a second lineup and did not make an identification. And he viewed a third lineup where he identified Laguerre as the person who fought Hardeman and shot him with an AK-47-style gun. 9 (c) Ballistics and Autopsy Evidence Cara McCarthy, an expert in firearms and ballistics analysis, received and analyzed evidence collected from the crime scene. She identified 14 .223-caliber cartridge cases, which she determined were all fired from the same weapon. McCarthy explained that .223- caliber bullets are typically fired from a rifle or long gun designed to be fired from the shoulder but that some pistols can fire .223-caliber bullets too. McCarthy also identified ten .380-caliber automatic car- tridge cases, all of which she determined were fired from the same weapon and are typically fired from a handgun. McCarthy also iden- tified a .380-caliber metal bullet jacket, but she explained that there was no way to tell if that bullet was connected to any of the ten .380- caliber cartridge cases that were submitted for analysis. Finally, McCarthy identified four .22-caliber cartridge cases, all of which she determined were fired from the same weapon. McCarthy also received bullets or fragments that were col- lected during Hardeman’s autopsy and submitted to the GBI for 10 testing. The identifiable bullets included multiple .380-caliber bul- lets and fragments and one .22-caliber bullet fragment. Associate medical examiner Dr. Michael Heniger, an expert in forensic pathology, reviewed Hardeman’s autopsy, which had been completed by someone else. In reviewing the autopsy, Dr. Heninger estimated that Hardeman had at least 20 or 25 gunshot wounds and determined his cause of death was multiple gunshot wounds to the head, torso, and extremities. (d) The Defendants’ Testimony Each defendant testified at trial. Baker denied being anywhere near the crime scene on October 15, 2011. He testified that he knew Laguerre and Ellis, but said he had never been to Ellis’s home and had not hung out with Laguerre since 2009 or 2010. Laguerre testified that he had never visited Ellis’s home, did not own a gun, and did not know Hardeman, Wimby, or many of the other witnesses who testified at trial. He denied both fighting Har- deman earlier in the day and shooting him. 11 2. Claims of Instructional Error Baker and Laguerre each contend that the trial court plainly erred by not sua sponte instructing the jury on accomplice corrobo- ration because there was at least slight evidence that Ellis was an accomplice to the shooting of Hardeman. Under our Evidence Code, “[t]he testimony of a single witness is generally sufficient to establish a fact.” OCGA § 24-14-8. But in felony cases, the Code makes an exception: the testimony of an ac- complice alone is not sufficient to establish a fact, unless the accom- plice’s testimony is corroborated. Id. See also Henderson v. State, 317 Ga. 66, 81 (5) (b) (891 SE2d 884) (2023). An accomplice is a person who “shared a common criminal intent to commit the crimes in ques- tion with the actual perpetrators.” Reese v. State, 317 Ga. 189, 198 (3) (891 SE2d 835) (2023). Generally, that intent is shown through evidence that the person “actively participated in some way in the crimes taking place,” such that they could be considered a party to the underlying crime. See Horton, 310 Ga. at 323 (3) (c); OCGA § 16- 2-20 (b) (1)-(4) (a person is a party to a crime if he “[d]irectly commits 12 the crime,” “[i]ntentionally causes some other person to commit the crime under such circumstances that the other person is not guilty of any crime,” “[i]ntentionally aids or abets” the crime, or “[i]nten- tionally advises, encourages, hires, counsels, or procures another to commit the crime”). A person’s “mere presence at the scene of the crime” is not enough, but “criminal intent may be inferred from pres- ence, companionship, and conduct before, during[,] and after the of- fense.” Lofton v. State, 309 Ga. 349, 352 (1) (846 SE2d 57) (2020) (citation and punctuation omitted). Here, the trial court instructed the jury about the single-wit- ness rule, but not about the accomplice-corroboration exception. Baker and Laguerre now contend that the trial court erred by not instructing the jury on accomplice corroboration based on evidence that they contend showed Ellis was a potential accomplice. But they concede that neither of them asked the trial court to give this in- struction or objected to its omission, so this claim of instructional error is reviewed for plain error. See OCGA § 17-8-58 (b). 13 To establish plain error, Baker and Laguerre must show that the alleged error was not affirmatively waived; was “clear and obvi- ous beyond reasonable dispute”; and likely affected the outcome of the trial. Johnson v. State, 319 Ga. 562, 567 (1) (-- SE2d --) (2024). If those three requirements are satisfied, the appellate court has the discretion to remedy the error only if it “seriously affected the fair- ness, integrity, or public reputation of judicial proceedings.” Id. Fail- ure to establish any one of these requirements means that a defend- ant has not established plain error. Id. (a) When a defendant asks for a particular jury instruction, a trial court must give the instruction if “slight evidence” supports the theory of the charge. Hamm v. State, 294 Ga. 791, 794 (2) (756 SE2d 507) (2014). This includes when a defendant asks for an instruction on the principle of accomplice corroboration under OCGA § 24-14-8. If slight evidence supports a finding that a witness was an accom- plice, the jury needs to know not only “how to decide whether the witness was in fact an accomplice,” but also “the extent to which it can rely on that witness’ testimony by itself to support a conviction.” 14 Id. at 796 (2). Otherwise, the court “leaves open the possibility of a conviction in violation of” the accomplice-corroboration rule. Id. And that is so even if the State presented independent evidence that cor- roborated the accomplice’s testimony, “because the jury, as the ex- clusive judges of credibility, could have rejected the other evidence and convicted solely on the accomplice’s testimony.” Id. at 796 (quot- ing Fleming v. State, 269 Ga. 245, 250 (497 SE2d 211) (1998) (Ben- ham, C.J., dissenting)). Put simply, if there is even slight evidence that a witness was an accomplice, “the failure to give such an in- struction, when it is requested, is error.” Id. But if a defendant does not ask for an instruction on accomplice corroboration and does not object to its omission at trial, the inquiry on appeal is different. In such cases, the trial court’s omission is re- viewed only for plain error, OCGA § 17-8-58 (b), and the question is not whether the court should have given a requested instruction, but whether the court “made a clear and obvious error by not sua sponte instructing the jury on accomplice corroboration.” Ash v. State, 312 Ga. 771, 796 (5) (b) (865 SE2d 150) (2021) (emphasis added). If no 15 one asked for an instruction on accomplice corroboration at trial and there was “no evidence presented at trial that obviously called for” giving that instruction, then a trial court does not commit a clear and obvious error by omitting it. Id. (concluding that the defendant “failed to show that the trial court made a clear and obvious error by not sua sponte instructing the jury on accomplice corroboration,” where “there might have been a slight evidentiary basis for [the de- fendant] to have requested a charge on accomplice corroboration,” but “no evidence presented at trial . . . obviously called for” it). This is not to say that the evidence that a witness was an accomplice must be overwhelming or even substantial — although that would accu- rately describe the evidence in many of our decisions holding that it was clear and obvious error to not instruct the jury on accomplice corroboration. See, e.g., Palencia v. State, 313 Ga. 625, 627-628 (872 SE2d 681) (2022) (three alleged accomplices each indicted with the defendant and pleaded guilty to offenses he was also charged with); Jackson v. State, 314 Ga. 751, 751 n.1, 752, 755-757 (1) (879 SE2d 410) (2022) (three alleged accomplices each indicted jointly with the 16 defendant for the victim’s murder and pleaded guilty to aggravated assault as a lesser offense of felony murder); State v. Johnson, 305 Ga. 237, 237-238, 238 n.1, 240-241 (1) (824 SE2d 317) (2019) (alleged accomplice jointly indicted with the defendant for the victim’s mur- der); Stanbury v. State, 299 Ga. 125, 127-129 (1), 129-130 (2) (786 SE2d 672) (2016) (alleged accomplice started trial as a co-defendant, took a plea and testified at defendant’s trial). But it must be obvious enough from the evidence that a witness was potentially an accom- plice that it would have been apparent to the trial court — even with no party having raised the issue — that the jury could convict the defendant based on uncorroborated accomplice testimony in viola- tion of OCGA § 24-14-8. See, e.g., Doyle v. State, 307 Ga. 609, 612- 613 (2) (a) (837 SE2d 833) (2020) (alleged accomplice drove defend- ant and his co-defendant to a business after listening to them dis- cuss hurting someone and getting payback, saw the defendant hold- ing a gun while in the car and the co-defendant “rack[ing]” a gun, drove them from the scene after hearing gunshots and stopping to 17 let the defendant back in the car, did not report the crime to the police, and initially denied he knew anything about the shooting). (b) Applying the proper standard here, the trial court did not make a clear and obvious error by not instructing the jury sua sponte on accomplice corroboration because the evidence did not obviously call for such an instruction. Baker and Laguerre contend that there was evidence that Ellis — the only witness to identify Baker as one of the shooters — was an accomplice. In support, they point to (1) evidence of Ellis’s inter- actions with Laguerre, including that he held Laguerre’s pistol dur- ing the fight with Hardeman, said afterwards that there was “going to be some gun play,” was present at the shooting later that day, and did not call for police or medical assistance or talk to police right after the shooting; and (2) Wimby’s testimony that Ellis “probably” had “something to do with” the murder. This evidence did not “obviously call for” a sua sponte instruc- tion on accomplice corroboration. Had either defendant developed and presented a theory at trial that Ellis was an accomplice, it is 18 perhaps conceivable that they might have picked out these snippets of evidence about Ellis from the substantial record at trial and then asked a jury to connect all of those dots and infer that Ellis shared the criminal intent to shoot Hardeman. Such evidence, if marshaled and presented at trial, might have added up to slight evidence that would have required giving an instruction on accomplice corrobora- tion if either defendant had asked for one. But neither defendant presented that evidence in support of an argument that Ellis was an accomplice. 2 And viewed in the context of the trial record as a whole, 2 Baker’s counsel attacked Ellis’s credibility at length, referred to Wimby’s testimony that Ellis “set [the murder] up,” and characterized Ellis as “a potential suspect” during opening statements and closing arguments. And in closing argument, Laguerre’s counsel told the jury that “nothing that [Ellis] says should make you feel good about making a decision when you” begin de- liberations and “you’ve got to wonder what Andrew Ellis really saw.” So alt- hough both defendants attacked Ellis’s credibility, and Baker pointed to Wimby’s testimony as evidence that Ellis may have been involved in the shoot- ing, neither claimed that Ellis committed the shooting together with Laguerre or Baker (or both), and so an accomplice theory would not have been apparent to the trial court. See Thornton v. State, 307 Ga. 121, 125-126 (2) (c) (834 SE2d 814) (2019) (although there might have been “slight evidence that [the alleged accomplice] committed the shooting,” there was no evidence that the alleged accomplice and the defendant “committed the crimes together,” so an accom- plice-corroboration instruction was not supported by the evidence and the trial court did not err, plainly or otherwise, by not giving that charge); Stripling v. State, 304 Ga. 131, 136 (2) (816 SE2d 663) (2018) (even though there “was at least slight evidence that all three [alleged accomplices] were involved in the 19 we cannot say that such evidence highlighted Ellis as a potential accomplice so clearly that the trial court should have given an in- struction on accomplice corroboration without a request. Baker and Laguerre emphasize Wimby’s testimony that Ellis “probably, got something to do with the s**t right now,” but that statement, despite being plucked from Wimby’s lengthy testimony, was bare speculation not based on any specifics about Ellis’s alleged involvement in the shooting — or any facts in evidence — that would suggest that Ellis “actively participated in” the shooting and shared either defendant’s criminal intent to kill Hardeman. Instead, Wimby’s only stated basis for his testimony was Ellis’s association with Laguerre, which is not itself enough to show common criminal intent. See Lofton, 309 Ga. at 352 (1). And it is even more of a stretch to conclude that this statement (or any of this evidence) showed that Ellis shared a criminal intent to commit the crime with Baker, whom Wimby had never seen before and did not identify as one of murder,” there was no evidence that these men committed the shooting with the defendant and his co-defendants, so it was not obvious error to omit an instruction on accomplice corroboration). 20 the shooters. In short, given the evidence presented at trial, the trial court did not commit a clear and obvious error by not giving an in- struction on accomplice corroboration without a request. See Ash, 312 Ga. at 796 (5) (b). Absent a clear and obvious error, neither Baker nor Laguerre have established plain error, so their claims fail. 3. Baker’s Sufficiency Claim Baker contends that the evidence was not sufficient as a matter of constitutional due process to convict him of malice murder and firearm possession. He also contends that Ellis was an accomplice and Ellis’s uncorroborated testimony was the only evidence that identified Baker as one of the shooters, so Baker’s identity was not established as a matter of fact under OCGA § 24-14-8 and, without Ellis’s identification, there was not sufficient evidence to convict Baker. When we review the sufficiency of the evidence as a matter of constitutional due process, we review the evidence as a whole in the light most favorable to the verdict. Jackson v. Virginia, 443 U.S. 307, 319 (III) (B) (99 SCt 2781, 61 LE2d 560) (1979). Our evaluation of 21 the sufficiency of the evidence is independent of the alleged instruc- tional error. See, e.g., Doyle, 307 Ga. at 611 (1) (evidence was suffi- cient as a matter of constitutional due process independent of whether the alleged accomplice’s testimony was corroborated and regardless of the trial court’s plain error by failing to give an accom- plice corroboration instruction). Even assuming Baker’s argument that the evidence was not constitutionally sufficient because of the operation of the statutory accomplice-corroboration rule,3 we still view the evidence at trial in the light most favorable to the verdict. See Jackson, 443 U.S. at 319 (III) (B). And viewed in that light, the evidence at trial showed that Ellis was merely present for the fistfight and shooting, not an ac- complice whose testimony required corroboration. See OCGA § 24- 3 We have indicated before that although OCGA § 24-14-8 “prohibits con- viction on the uncorroborated testimony of an accomplice, … the question of the sufficiency of the evidence to satisfy federal constitutional due process un- der Jackson v. Virginia has no such requirement.” Johnson v. State, 311 Ga. 221, 223 n.2 (857 SE2d 463) (2021). Some of us doubt whether federal consti- tutional sufficiency review properly ignores state-law evidentiary rules in this way, but we need not resolve those doubts here because, as we explain, Baker’s claim fails even if we consider the accomplice-corroboration rule in this analy- sis. 22 14-8. See also Reese, 317 Ga. at 198 (3); Horton, 310 Ga. at 323 (3) (c); Lofton, 309 Ga. at 352 (1). And so Ellis’s identification of Baker as the shooter with the pistol was sufficient on its own to establish that fact. See OCGA § 24-14-8. Moreover, because Ellis’s testimony authorized the jury to find that Baker was one of the two shooters, the jury was also authorized to disbelieve Baker’s testimony that he was not at the crime scene or involved in the shooting in any way, and to consider his disbelieved testimony as substantive evidence against him. See Daughtie v. State, 297 Ga. 261, 263-264 (2) (773 SE2d 263) (2015). That evidence, considered together with the evi- dence as a whole, including Ellis’s identification of Baker, Wimby’s testimony that someone Wimby did not recognize shot Hardeman first and then Laguerre shot Hardeman with an AK-47-type gun, and forensic evidence showing that Hardeman was shot by more than one gun (and some of the bullets may have been fired from a pistol rather than a rifle), authorized Baker’s convictions for malice murder and the related firearm possession offense. See Jackson v. 23 State, 307 Ga. 770, 771-773 (838 SE2d 246) (2020) (sufficient evi- dence to sustain the conviction for malice murder and other crimes based on eyewitness identification of the defendant as the shooter). 4. Laguerre’s Ineffectiveness Claims Laguerre contends that his trial counsel rendered constitution- ally ineffective assistance because she did not object, move for mis- trial, or seek a curative instruction in response to several statements by Wimby about his fear of the defendants. Laguerre contends that counsel should have objected to Wimby’s statements as irrelevant under OCGA § 24-4-401 and § 24-4-402, as more prejudicial than probative under OCGA § 24-4-403, or as impermissible character ev- idence under OCGA § 24-4-404 (a) and that Wimby’s testimony im- properly suggested that Laguerre posed a future danger.4 See Ster- ling v. State, 267 Ga. 209, 210 (477 SE2d 807) (1996) (improper for 4 In his reply brief, Laguerre also contends that counsel was ineffective for not objecting to and moving to strike Wimby’s statements as non-respon- sive. But that contention is not properly before us because it was raised for the first time in his reply brief. See Bradley v. State, 318 Ga. 142, 145 n.4 (2) (897 SE2d 428) (2024). 24 the State to argue a defendant poses a threat of future danger in the guilt-innocence phase of trial). To succeed on a claim of ineffective assistance of counsel, a de- fendant must show that his counsel’s performance was deficient and caused him prejudice. See Strickland v. Washington, 466 U.S. 668, 687 (III) (104 SCt 2052, 80 LE2d 674) (1984). See also Harmon v. State, 319 Ga. 259, 265 (3) (903 SE2d 28) (2024). To be deficient, an attorney’s conduct must be “objectively unreasonable under the cir- cumstances.” See Harmon, 319 Ga. at 265 (3). See also Strickland, 466 U.S. at 689-690 (III) (A). To show prejudice, the defendant must establish “a reasonable probability that the result of the trial would have been different absent counsel’s deficient performance.” See Harmon, 319 Ga. at 265 (3). See also Strickland, 466 U.S. at 694- 695 (III) (B). (a) Wimby made the following statements at trial, none of which counsel objected to, which Laguerre now contends was consti- tutionally ineffective assistance of trial counsel: On direct examination, the prosecutor asked Wimby: 25 Q. Did you know who the guy with the AK-looking gun hung around with? A. Yeah; [Ellis, aka] Cali. Cali is the one who gave them all the information. I don’t even know how the f**k Cali is on the witness stand like he’s just my f**king friend and telling the truth. He, probably, got something to do with the s**t right now; they getting locked up right now. He’s, probably, go- ing to be the reason I get killed. This they first time ever seeing me ever since this happened. They didn’t know how I looked. I had to change my whole f***ing face, dreads and everything, just so they don’t know who the f**k I am. But, now, [Baker and Laguerre] know who I am. They done seen me; everybody. Nobody had seen me since the day that s**t happened. Now, I done showed myself, who the f**k is going to protect me? After the prosecuting attorney passed the witness but before defense counsel began cross-examination, Wimby spontaneously said: “I re- ally feel threatened for my life because, without me, there is no case. Then they out on bond. It’s so easy to come and try to find me right now. Who the f**k is going to protect me right now?” On cross-examination, Laguerre’s attorney asked: Q. So do you realize that you were on the train all this time with Mr. Laguerre? A. On the train? That’s what you said? Q. Yeah. A. With that being said, he’s been knowing how I look. 26 Q. He sees you on the MARTA train all the time. A. Check this, that’s a whole got [sic] damn lie. The way this man feeling right now, I am the only person he really want to find in order to beat his case. If he had a gun right now, this man will shoot the s**t out of me right now in order for me not to be up here telling this story, ma’am. Those boys you’re talking about your clients see me on the train, ma’am, it’s a whole lie. Finally, on further cross-examination, Laguerre’s attorney asked: Q. You are not under the impression that anybody knows where you live right now? A. I am all the way under the impression that y’all know exactly where I’m at. Yes, I do feel scared for my life. Now that I know that he’s out on bond and all that and I go home, guess what, he’s going home, too. But guess what make it even better, now, [Laguerre] know how I look. He’s out on bond. I’m scared for my life. Do I get any type of protective custody or anything? I’m so scared. Is anybody going to help me? At the hearing on Laguerre’s motion for new trial, his trial counsel testified about the defense strategy, saying the focus was “identity; that [Laguerre] did not commit the offense” or participate in the fight with Hardeman earlier that day. Counsel recalled that Wimby “was acting very bizarre” at trial and “was not answering questions.” She explained that Wimby “had gone into sort of a 27 stream of consciousness and just started saying a number of dispar- aging things, and was concerned about his safety,” and that “[h]e was acting irrationally, in [counsel’s] opinion.” Counsel testified that she did not object to these comments, ask for a curative instruction, or move for mistrial, and when asked if she had a strategic reason not to do so, said “no.” She did, however, ask Wimby on cross-exam- ination about riding MARTA with Laguerre (and later elicited testi- mony from Laguerre that he regularly encountered Wimby on MARTA without incident while Laguerre was out on bond after the shooting). And she testified that she did so to undermine Wimby’s statement that Laguerre was someone he should fear and that La- guerre might kill him. (b) Notwithstanding counsel’s subjective evaluation of her per- formance, deficiency under Strickland is evaluated on an objective basis. See Richardson v. State, 318 Ga. 690, 697 n.7 (2) (b) (899 SE2d 685) (2024). And objectively, counsel’s decision not to object but in- stead to question Wimby and Laguerre about encounters on MARTA 28 to “undermine” his supposed fear of Laguerre was not an unreason- able strategy regardless of the merits of any possible objection that could have been raised under Rules 401 through 404 or as an im- proper comment on future dangerousness. See Gaston v. State, 307 Ga. 634, 642 (2) (d) (837 SE2d 808) (2020) (“Decisions about cross- examination do not amount to deficient performance unless they are so unreasonable that no competent attorney would have made them under similar circumstances.”) (citation and punctuation omitted). Indeed, that decision fit the overall defense strategy of showing that Laguerre was not the person who fought or shot Hardeman since Laguerre denied being present for both the shooting and the earlier fistfight. And this strategy would have benefited from casting doubt on Wimby’s identification of Laguerre, with whom Wimby had ap- parently been riding MARTA without ever noticing. So Laguerre has not shown that trial counsel rendered deficient performance by questioning Wimby about these encounters with Laguerre on MARTA to undermine his testimony, rather than objecting to that testimony or seeking other remedial measures. See id. Cf. Rashad v. 29 State, 318 Ga. 199, 212-213 (3) (e) (897 SE2d 760) (2024) (concluding counsel did not perform deficiently “by failing to object or to move for a mistrial or for curative instructions to testimony by [a witness] that [the defendant] characterizes as bad character evidence” be- cause, “[i]t would not be objectively unreasonable for trial counsel, as a matter of trial strategy, to refrain from objecting to this testi- mony so as not to draw attention to it”). And without a showing of deficient performance, Laguerre’s claim of ineffective assistance of counsel fails. See Strickland, 466 U.S. at 689-690 (III) (A); Harmon, 319 Ga. at 265 (3). 5. Laguerre’s Claim of Cumulative Error Laguerre contends that he was prejudiced by the cumulative effect of the trial court’s failure to give an accomplice-corroboration jury instruction and his counsel’s deficient performance in failing to object to Wimby’s testimony about his fear of Laguerre. As discussed above, Laguerre has not established that an error occurred, nor have we assumed any error in resolving his claims. Without any errors to 30 assess cumulatively, this claim fails. See Blocker v. State, 316 Ga. 568, 583 (5) (889 SE2d 824) (2023). Judgments affirmed. All the Justices concur. 31