People v. Meskell CA3

• Court: California Court of Appeal
• Decided: April 25, 2023
• Docket: C095788
• Status: Unpublished

Opinion

Filed 4/25/23 P. v. Meskell CA3 NOT TO BE PUBLISHED California Rules of Court, rule 8.1115(a), prohibits courts and parties from citing or relying on opinions not certified for publication or ordered published, except as specified by rule 8.1115(b). This opinion has not been certified for publication or ordered published for purposes of rule 8.1115.

IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA THIRD APPELLATE DISTRICT (Sacramento) ----

THE PEOPLE, C095788

Plaintiff and Respondent, (Super. Ct. No. 20FE000293)

v.

JOHN FRANCIS MESKELL,

Defendant and Appellant.

Two men forced their way into the home of two brothers. One robber fled shortly after entering, while another entered a bedroom where the victims kept a shotgun. A struggle over the shotgun ensued between the robber and the two victims. The shotgun went off, killing one of the victims. During trial, the prosecutor admitted evidence of another robbery defendant was charged with, but acquitted of, to show a common plan between the two robberies. After a trial for the robbery of the brothers, a jury found defendant John Francis Meskell guilty of first degree felony murder and found true a special circumstance.

1 On appeal, defendant makes several contentions: (1) the felony murder and special circumstance jury instructions improperly permitted the jury to find defendant liable under a proximate cause theory rather than as the actual killer, a requirement after the passage of Senate Bill No. 1437 (2017–2018 Reg. Sess.) (Stats. 2018, ch. 1015) (Senate Bill 1437); (2) admission of the prior robbery was relevant only for identity, and the robberies were not sufficiently similar for that purpose; (3) the felony murder and special circumstance statute is unconstitutionally vague because it punishes the same activity; and (4) a parole revocation restitution fine was erroneously imposed. We conclude the jury instructions did not properly advise the jury it was required to find defendant personally killed the victim. We reverse on this issue because there was evidence the second victim fired the shotgun so we cannot say the error was harmless. We also address defendant’s second argument and find the trial court erred in admitting the evidence of the prior robbery because it was not relevant as a common plan. Instead, it was admitted to infer identity, and the crimes were not sufficiently similar for this purpose. We do not address defendant’s remaining arguments. BACKGROUND Defendant was charged with murder (Pen. Code, § 187, subd. (a)),1 with the special circumstance allegation it was committed during a robbery (§ 190.2, subd. (a)(17)), and the enhancement defendant personally used a firearm (§ 12022.53, subd.(b)). A. Trial Testimony Thinh Tran, who was born in Vietnam, testified under an order of immunity he and defendant agreed to rob Cristian Anton’s house on February 8, 2010, because they heard the house had cash and cocaine. Defendant and Tran were both armed with guns

1 Undesignated statutory references are to the Penal Code.

2 defendant supplied: Tran had a .22-caliber handgun, and defendant had a .38-caliber handgun. They both wore gloves, and they waited outside the house that night for someone to come to the house. About 30 minutes to an hour later, a man came home, and defendant pushed him through the front door. Defendant told Tran to watch the man and ran into another room down a hallway. The man tried to grab Tran’s gun and the two struggled over the gun. The gun fired during the struggle, the man released Tran, and Tran ran out the door and away from the house. Tran went to defendant’s home the next day to return the gun, and defendant was angry at him for leaving the house during the attempted robbery the previous day. Defendant told Tran there was another guy in the other room who had a “ ‘long gun,’ ” they fought, and a gun went off. Marius Anton testified he lived with his older brother Cristian in Sacramento on February 8, 2010. Just past midnight that morning he got out of his truck and walked to his front door when two men approached from behind. He described the men as one being big and looking Hispanic and one was smaller and looked Asian. One of the men put Marius in a chokehold and something struck his face. But he couldn’t remember if he was hit with the gun or if the gun went off in his face, and whether it happened outside or inside the house. Once inside, Cristian came out of his room and the larger man shot at him while Marius wrestled with the smaller man near the front door. Marius testified he chased the smaller man outside and away from the house. While outside, Marius saw through a window the bigger robber and Cristian in a bedroom, and this robber may have been pointing a gun at Cristian. Marius grabbed a shovel, ran inside the house to Cristian’s room, and threw the shovel at the robber. As soon as he threw the shovel, he leaped for his shotgun he kept in the room with the intent to shoot the robber. The robber shot at Marius while he was going for the shotgun and Marius initially backed off of the gun. Then Cristian, the robber, and Marius all struggled for the shotgun when it fired once.

3 When asked if he told detectives in 2010 that the robber got the trigger end of the gun and Cristian got the barrel end, Marius testified, “I told them that they were all on it. We all grabbed for it. I grabbed it. He went for it, too.” The prosecutor had Marius read what he told detectives to refresh his memory, but Marius responded, “It’s still not very clear exactly how it went off, but the bottom line was that, you know, the guy was in my room, and he wasn’t supposed to be there, you know. [¶] I had my shotgun, and my intention was to grab him and blow his head off. And you know, what happened happened. . . . Whose hand was on, how it happened, who got their finger in there first, I don’t know. But the gun went off . . . and it didn’t happen [in] slow motion, that’s for sure.” Marius also later testified he “went for the shotgun to blow his head off, he pushed the barrel towards my brother. [¶] It must have happened just like that.” When Marius realized Cristian had been shot, he and the robber looked at each, then the robber ran away. Marius “must have” picked up and dropped the shotgun but couldn’t remember for certain. Both Anton brothers then called 911. On these calls, Marius said, “somebody shot my brother,” “[h]e shot my brother,” and “[t]he guy shot off the shotgun.” Marius identified defendant in court as the bigger robber because he remembered the fear in defendant’s eyes. But two years before trial, about ten years after the shooting, Marius did not pick defendant out of a lineup. Jason W., a neighbor of the Antons, testified to hearing a gunshot early the morning of February 8, 2010. He saw several men run out of the house and saw Marius, who was saying, “ ‘He shot my brother.’ ” In 2020, Jason W. picked defendant out of a lineup as one of the people he saw fleeing the Antons. Several officers testified about interviews with the witnesses. Officer David Treat arrived on the scene and spoke with Cristian while waiting for the ambulance. When officer Treat asked what happened, Cristian, who had an “off-white” zip-tie around one wrist, said he was in his bedroom on his computer when someone came in, “[a]nd then

4 words to the effect that he tried to rob him and tie him up, and then shot him with a shotgun.” Cristian was taken to the hospital but died shortly after. Officer Elaine Stoops testified she interviewed Marius in March 2010, and he said the Hispanic robber had the trigger end of the shotgun, Cristian had the barrel end, and defendant shot him not on accident. Cristian died from a shotgun wound that was “close, possibly, contact range” to his lower abdomen and groin area; he was 31 years old. Gunshot residue was found on Marius’s hands consistent with holding the barrel of a shotgun and Marius’s blood was found on the barrel of the shotgun. There were also white zip-ties found with blood on them, bullet holes in the wall behind where Marius kept his shotgun, and spent .22 and .38 rounds found in the house; it was determined the .22-caliber rounds were fired from an Astra firearm. B. Jury Instructions The trial court gave standard jury instruction CALCRIM No. 200, which instructed the jury, in part, that if it “believe[d] that the attorneys’ comments on the law conflict with [the court’s] instructions, you must follow [the court’s] instructions.” The trial court gave the jury two theories for first degree murder: felony murder and provocative act murder. For felony murder, the trial court provided instruction CALCRIM No. 540A, which required the jury to find defendant intended to commit robbery, and during the robbery “defendant caused the death of another person.” The instruction explained a “person who was the actual killer may be guilty of felony murder even if the killing was unintentional, accidental, or negligent.” For provocative act murder, the trial court gave CALCRIM No. 560. This instruction stated a “person can be guilty of murder under the provocative act doctrine even if someone else did the actual killing.” The jury had to find: (1) defendant, in the course of the robbery “intentionally did a provocative act,” (2) “defendant knew that the natural and probable consequences of the provocative act were dangerous to human life

5 and then acted with conscious disregard for life,” (3) “[i]n response to the defendant’s provocative act, Marius Anton killed Cristian Anton,” and (4) “Cristian Anton’s death was the natural and probable consequence of the defendant’s provocative act.” The instruction explained a provocative act is an act “[w]hose natural and probable consequences are dangerous to human life,” which means a “reasonable person” would have foreseen there was a “high probability” the act began “a chain of events resulting in someone’s death,” and the act “was a direct and substantial factor in causing Cristian Anton’s death.” The prosecutor alleged two provocative acts: Defendant “[f]ired the .38 caliber revolver toward Cristian Anton when he came out of his bedroom upon entry into the house and fired the .38 caliber revolver at Marius Anton as he attempted to grab the shotgun.” The jury had to agree the prosecution proved at least one of these acts. If the jury found defendant guilty of first degree murder, it had to also decide whether the people proved the special circumstance. The special circumstance instruction required the jury to find defendant intended to commit robbery and defendant “did an act that caused the death of another person.” C. Closing Arguments In closing, the prosecutor argued felony murder is “[d]irect liability, meaning the defendant pulled the trigger.” The difference with the provocative act theory is “defendant pulled the trigger. He caused the death of another person. . . . So if you find the defendant pulled the trigger, he is guilty of felony murder in the first degree. . . . [I]f you cause the death of another person during the course of a robbery or an attempt, you are on the hook whether you intentionally pulled the trigger or not.” For the provocative act instruction, the prosecutor emphasized she did not think the evidence supported the finding because defendant pulled the trigger, stating the only evidence supporting the provocative act instruction was Marius’s testimony he was “not sure whether he did it or the defendant . . . [w]e don’t have a single other piece of evidence to support the provocative act theory that Marius pulled the trigger.”

6 For the special circumstance finding, the prosecutor argued it “mirrors felony murder,” including defendant “has to cause the death. Again, he’s got to be the trigger puller. Doesn’t have to be intentional.” Defense counsel’s closing argument focused on defendant not being one of the perpetrators who robbed the Antons and did not mention the circumstances of the shooting. The prosecutor noted this in rebuttal, stating defense counsel did not discuss “whether the defendant pulled the trigger or not. . . . So if you find the defendant was the person with Tran–and I agree with this based on the evidence–the defendant pulled the trigger.” The prosecutor consequently focused on the evidence establishing defendant’s identity as the robber. D. Verdicts and Sentencing The jury found defendant guilty of first degree murder. The verdict form had three options stating, “[c]heck each applicable box:” felony murder, provocative act murder, and that the jury unanimously agreed on first degree murder but were unable to unanimously decide which theory. The jury marked the box for felony murder. The jury also found true the special circumstance allegation.2 The trial court sentenced defendant to life in prison without the possibility of parole. DISCUSSION I Actual Killer Instructions Defendant first challenges the felony murder and special circumstance jury instructions. He contends they permitted the jury to find him guilty without having to find he was the actual killer, a requirement after Senate Bill 1437. Instead, he needed to

2 After both sides rested, the prosecutor dismissed the firearm use allegation (§ 12022.53, subd. (b)) “based on insufficient evidence.”

7 have only “caused the death,” which, he contends, “confer[s] liability for actions that only qualify as a proximate cause, which is insufficient to establish that a defendant is an actual killer.” Defendant filed a supplemental brief contending People v. Vang (2022) 82 Cal.App.5th 64 (Vang), a recent opinion from a different panel of this court, establishes instructions were erroneously given. The People counter arguing Senate Bill 1437 did not modify the meaning of “actual killer” for felony murder. Instead, actual killer is used only to distinguish from mere participants in the felony. And because defendant was the only participant of the felony in the room when Cristian was shot, he was necessarily the “actual killer” for purposes of the felony-murder statute and special circumstance finding. We find defendant’s position more persuasive and reverse. A. Legal Standards Murder in California generally requires malice aforethought. (§ 187, subd. (a).) “[T]he felony-murder rule provides an exception to the malice requirement for murder.” (Vang, supra, 82 Cal.App.5th at p. 81.) Given the modified intent requirement, the felony-murder rule has a long history of disfavor. “Almost since its inception, the felony- murder rule has been criticized by some courts and commentators as an artificial and unnecessary doctrine that erodes the relation between criminal liability and moral culpability.” (Id. at p. 81-82.) Effective January 1, 2019, the Legislature narrowed murder culpability and the felony-murder rule with Senate Bill 1437. This bill modified section 188 to require murder to have malice aforethought and which “shall not be imputed to a person based solely on his or her participation in a crime.” (§ 188, subd. (a)(3).) Similarly, it modified section 189 to limit murder liable for felony murder to persons who were: (1) the actual killer, (2) were not the actual killer but had the intent to kill, or (3) major participants in the felony who acted with reckless indifference to human life. (People v. Lewis (2021) 11 Cal.5th 952, 959.) This mirrors the requirements necessary to prove a special

8 circumstance enhancement under section 190.2. (§ 190.2, subds. (b)-(d).) Thus, “Senate Bill 1437 made the crime of felony murder subject to the same elements of proof required for a felony-murder special-circumstance finding under section 190.2.” (Vang, supra, 82 Cal.App.5th at p. 83.) “A person convicted of murder, attempted murder, or manslaughter whose conviction is not final may challenge on direct appeal the validity of that conviction based on the changes made to Sections 188 and 189 by Senate Bill 1437.” (§ 1172.6, subd. (g).) “In criminal cases, a trial court has a sua sponte duty to instruct on the general principles of law relevant to the issues raised by the evidence and necessary for the jury’s understanding of the case. [Citation] We apply a de novo standard when determining whether jury instructions were complete and correctly stated the law. [Citation.] We consider the challenged instruction in the context of the instructions and record as a whole to determine whether it was reasonably likely the jury misapplied the law. [Citation.]” (Vang, supra, 82 Cal.App.5th at pp. 83-84.) B. Analysis 1. Instructional Error The prosecutor tried defendant on an actual killer theory. The trial court consequently instructed the jury on both felony murder and special circumstances under the actual killer theory. We must therefore decide whether the jury instructions properly advised the jury it had to find defendant was the actual killer. Defendant contends they did not as the instructions permitted liability if defendant “caused” Cristian’s death. At least four recent cases have dealt with this exact issue after Senate Bill 1437, and these cases support defendant’s position. In People v. Garcia (2020) 46 Cal.App.5th 123 (Garcia I), the defendant either handed his co-perpetrator a roll of duct tape that was used to cover the victim’s mouth or taped the victim’s mouth himself during a robbery; the victim eventually died from

9 asphyxiation. (Id. at pp. 145-146.) The jury was instructed on robbery-murder special circumstance using CALCRIM No. 730, including the element the defendant “did an act that caused the death of another person.” (Id. at pp. 149-150.) On appeal, the appellate court found “the California Supreme Court has used the term ‘personally killed’ when describing liability of an ‘actual killer’ for the felony-murder special circumstance,” and “ ‘[p]roximately causing and personally inflicting harm are two different things.’ ” (Id. at p. 151.) Thus, it concluded “[t]he clear import of these cases and the statutory scheme set out in section 190.2 is that the meaning of “actual killer” under section 190.2(b) is literal. The actual killer is the person who personally kills the victim, whether by shooting, stabbing, or — in this case — taping his mouth closed, resulting in death by asphyxiation.” (Id. at p. 152.) And “ ‘did an act that caused the death’ ” instruction did not require the jury to find defendant personally killed the victim; this was legal error. (Id. at p. 155.) In People v. Lopez (2022) 78 Cal.App.5th 1, the defendant’s DNA was found at the scene of a murder. The defendant admitted to being at the scene but said he went there with another person and never went into the room where the victim was killed. (Id. at pp. 7-9.) The jury had found the defendant guilty of first degree murder and found true a special circumstance finding. (Id. at p. 10.) The defendant filed a petition for resentencing based on the changes from Senate Bill 1437. Relying on the record of conviction, the trial court denied the petition, finding his conviction was based on him being the actual killer. (Id. at pp. 10-11.) The appellate court reversed based on the jury instructions, because “the jury necessarily found that defendant ‘caused the death of another person’ and ‘did an act that caused the death of another person.’ ” (Id. at p. 16.) Relying on Supreme Court case law similar to Garcia I, the Lopez court found “actual killer” means “personally killed.” (Id. at pp. 16-17.) And the “the Legislative history of Senate Bill No. 1437 suggest the Legislature, in revising the law of felony murder and

10 accomplice liability for murder, intended actual killer to have that same meaning; that is, someone who personally killed the victim.” (Id. at p. 17.) The Lopez court also found support in Garcia I: even though “[t]he [Garcia I] court interpreted the term actual killer in section 190.2 [subdivision] (b) as applied to section 190.2 [subdivision] (a)(17) . . . we find no reason to believe the Legislature intended for the term actual killer to have a different meaning in section 189 [subdivision] (e)(1), particularly given the Legislative history of Senate Bill No. 1437.” (Lopez, supra, 78 Cal.App.5th at p. 19.) Because there was no direct evidence the defendant committed the killing, the court concluded the record of conviction did not establish the defendant was ineligible as a matter of law for resentencing after Senate Bill 1437. (Lopez, at p. 20.) In Vang, the defendant forced the victim into his car and, while he was driving away, she jumped out of the moving car and died. (Vang, supra, 82 Cal.App.5th at p. 64.) The jury found the defendant guilty of first degree felony murder and found true the special circumstance based on instructions requiring “defendant caused the death of another person.” (Id. at p. 80 & fn. 3.) On appeal, another panel of this court determined, after reviewing the legislative history of Senate Bill 1437, that the intent of the bill was “to impose punishment commensurate with the person’s culpability.” (Vang, at p. 88.) We concluded: “the term ‘actual killer’ was intended to limit liability for felony murder — in cases where section 189, subdivision (e)(2) or (3) do not apply — to the actual perpetrator of the killing, i.e., the person (or persons) who personally committed the homicidal act. In other words, the intent was to conform California law to the ‘agency theory’ of felony-murder liability, under which criminal culpability is restricted to deaths directly caused by the defendant or an accomplice, as distinguished from the ‘proximate cause’ theory of felony murder, under which a defendant is responsible for any death that proximately results from the unlawful activity.” (Ibid.) We determined this interpretation was also “consistent with precedent discussing what it

11 means to be an ‘actual killer’ for purposes of section 190.2, the felony-murder special- circumstance statute.” (Ibid.) And we found support for our analysis in Garcia I and Lopez. (Id. at pp. 90-91.) Thus, we concluded “[t]he jury was permitted to find defendant guilty of felony murder, with a special circumstance, based on an invalid legal theory.” (Id. at p. 91.) Finally, in People v. Garcia (2022) 82 Cal.App.5th 956 (Garcia II), the defendant assaulted and robbed the victim who died shortly after the attack from cardiac arrhythmia. (Id. at pp. 961-962.) A jury found the defendant guilty of first degree murder prior to Senate Bill 1437 and the trial court denied the defendant’s petition for resentencing because the defendant was ineligible as a matter of law as the actual killer. (Garcia II, at pp. 959-960.) On appeal, another panel of this court stated: “[w]hen there is more than one contributing cause of death in a murder case, the law is clear: ‘[A]s long as the jury finds that without the criminal act the death would not have occurred when it did, it need not determine which of the concurrent causes was the principal or primary cause of death. Rather, it is required [only] that the cause was a substantial factor contributing to the result.’ ” (Id. at p. 964.) Under this understanding, “the record of conviction unequivocally establishes that defendant was the ‘actual killer.’ He was the sole perpetrator of a robbery where a death occurred as a direct consequence of his actions.” (Id. at p. 969.) Thus, we concluded, “the trial court properly evaluated defendant’s petition for resentencing under the ‘actual killer’ provision of section 189, subdivision (e)(1) to determine whether he was guilty of murder under a theory that remains valid after Senate Bill No. 1437.” (Ibid.) We did not find Garcia I, Lopez, or Vang to be inconsistent with this conclusion because, unlike those cases, the defendant in Garcia II personally did the act that was a substantial factor in bringing about the victim’s death. (Id. at pp. 969-971.) These cases are consistent on the legal issue central to the resolution of defendant’s appeal. Defendant’s jury was instructed solely on an actual killer theory, and

12 both the felony murder and special circumstance instructions failed to define an actual killer. Instead, the instructions permitted liability if the jury found defendant “caused the death of another person” and “did an act that caused the death of another person.” The courts in Garcia I, Lopez, and Vang concluded actual killer means someone who personally killed the victim and the same “caused” language employed in defendant’s trial is insufficient for these purposes. This reasoning applies to both felony murder and the special circumstance statute. We find no reason to conclude the legal analysis shared by these cases was erroneous or misguided, and we adopt it here. Consequently, we find defendant’s jury was improperly instructed on both felony murder and the special circumstance finding. The core reasoning of the People’s argument is that defendant must be liable as an actual killer because he was the sole participant of the felony in the room during the killing. The People contend, “[t]hese differentiations between participants are unnecessary where a sole perpetrator commits a qualifying felony that results in a death because under these circumstances there are no ‘participants’ with whom to share culpability. All criminal conduct, as well as the resulting death, is attributable to the sole perpetrator of the qualifying offense, the only possible killer.” The People contend felony murder effectively imposes “strict liability” for sole perpetrators. (See People v. Huynh (2012) 212 Cal.App.4th 285, 309-310 [“the felony-murder rule imposes a type of strict liability on the perpetrator acting on his or her own”].) Underlying this argument seems to be an assumption that murder liability should be imposed automatically for participation in a felony, a reasoning invalidated by Senate Bill 1437. But we need not decide whether the People’s argument is still true in every case after Senate Bill 1437 because it is not true in this specific case. The People may be correct that Senate Bill 1437 did not change the definition of an actual killer. But it is undisputed that after this bill, felony murder, just like the special circumstance finding, is restricted to three classes of defendants. And defendant here was

13 tried only under one theory: actual killer. Under the actual killer theory, defendant must have personally killed Cristian. But the instructions permitted the jury to find defendant guilty of felony murder and find true the special circumstance under a proximate cause theory, an invalid theory for actual killer liability. In practical terms, this meant the jury could find defendant guilty of felony murder and find true the special circumstance even if Marius shot Cristian if it found defendant “caused” Marius to shoot Cristian. In this circumstance, defendant would not have personally killed Cristian. So, defendant being the sole participant of the felony at the time of the shooting did not automatically make him the actual killer. The People also rely on Garcia II, but that case is not inconsistent with our conclusion. We dealt with a unique fact pattern in Garcia II, where the victim died later from a heart attack caused by the defendant’s assault. The issue was whether the defendant could be an actual killer “[w]hen there is more than one contributing cause of death in a murder case.” (Garcia II, supra, 82 Cal.App.5th at p. 964.) We found that under these circumstances, a person is an actual killer if they were a “substantial factor” in the death and concluded, “because defendant’s acts in robbing [the victim] (including assault) were a substantial factor in bringing about [the victim’s] death, defendant ‘personally killed’ [the victim] and was the ‘actual killer’ within the meaning of section 189, subdivision (e)(1).” (Id. at pp. 964, 970-971.) We, therefore, rejected the defendant’s argument the Legislature “intended section 189, subdivision (e) to mean that there is no ‘actual killer’ for purposes of felony-murder liability where death resulted from a preexisting medical condition aggravated by the stress of the underlying felony.” (Id. at p. 968.) The victim in Garcia II died both because of defendant’s attack and because of the victim’s preexisting medical condition. That is not the situation we have here, where the shotgun blast certainly caused Cristian’s death and only one person could have pulled the trigger, but it is not beyond doubt who that was. This case is more like Garcia I, Lopez,

14 and Vang, where the issues were whether there was evidence the defendants were the sole cause of the victim’s death. In Garcia I, it was either the defendant or a co-perpetrator that taped the victim’s mouth. In Lopez, it was either defendant or another unknown person who killed the victim. In Vang, there was no dispute it was the victim who jumped from the car. Consequently, in Garcia II, we found those cases inapplicable because “evidence in those cases did not show (as the evidence did in this case) that the defendant personally committed the act(s) that directly resulted in the victim’s death.” (Garcia II, supra, 82 Cal.App.5th at p. 971.) That does not conflict with our conclusion here. Thus, we conclude the instructions given for felony murder and the special instructions constituted legal error.3 2. Prejudice We must next determine whether this error was prejudicial. “Where, as here, the verdict was based on a legally invalid theory, reversal is required unless we can say, beyond a reasonable doubt, that the error did not contribute to the verdict.” (Vang, supra, 82 Cal.App.5th at p. 91.) Under this standard, “the prosecution bears the burden of proving that the error was harmless. [Citation.] That is, the presumption is that we must reverse, unless [we find] the prosecution proves the error was harmless beyond a reasonable doubt.” (People v. Avalos (2022) 85 Cal.App.5th 926, 953.) “The test is not whether a hypothetical jury, no matter how reasonable or rational, would render the same verdict in the absence of the error, but whether there is any reasonable possibility that the error might have contributed to the conviction in this case. If such a possibility exists, reversal is required.” (People v. Lewis (2006) 139 Cal.App.4th 874, 887.)

3 We make an explicit call for the language in these CALCRIM instructions to be considered for modification in light of our opinion here and the opinions in the other cases we have cited.

15 The People do not carry their burden here. True, there was strong evidence defendant pulled the trigger. Multiple witnesses, including police officers, testified that shortly after the shooting Marius and Cristian implied defendant shot Cristian. Marius stated on the 911 calls “he shot my brother” and the “guy shot off the shotgun.” There was also physical evidence indicating Marius did not fire the gun. He had gunshot residue consistent with holding the barrel of the gun and his blood was found on the barrel of the gun. The prosecutor also emphasized for felony murder and the special circumstance allegation defendant had to be the “trigger puller.” However, there was some evidence Marius did fire the shotgun, or at least created a reasonable doubt as to who fired the gun. The most significant evidence is his testimony at trial. Marius, the only living eyewitness to the events, was asked several times about his previous statements, and he refused to adopt them. Instead, he maintained he did not know who shot his brother, saying it is “still not very clear exactly how it went off” and “[w]hose hand was on [it], how it happened, who got their finger in there first, I don’t know.” And the physical evidence never conclusively established who pulled the trigger. Marius testified he might have picked up the gun after the shooting. It is possible this could have resulted in Marius’s blood being on the barrel. The gunshot residue was also consistent with Marius holding the barrel of the shotgun but there was no testimony this was the only consistent scenario. Thus, there was some evidence creating doubt defendant personally killed Cristian. And though the prosecutor argued defendant must be the actual “trigger puller” for felony murder, the jury was instructed under CALCRIM No. 200 that the instructions were to be followed over the arguments of counsel. The prosecutor’s arguments lessen the likelihood of prejudice but do not eliminate it. The jury not checking the box for the provocative act theory does not eliminate any prejudice. The prosecutor argued there were two options for how Cristian was shot: either defendant pulled the trigger, and felony murder applied, or Marius pulled the

16 trigger, and the provocative act theory applied. But the instructions did not present these options as the prosecutor argued. First, because the jury was given a deficient instruction on what constituted being the “actual killer,” there was an overlap between the two instructions. The jury could have found defendant caused the death with a provocative act, creating a scenario where both instructions applied, and the jury just checked the box for felony murder as it was not required to find either theory not true. Second, there are several scenarios where the provocative act theory could have been found not true even if Marius shot Cristian. For example, the jury had to find one of the specific provocative acts provided in the instruction and both involved defendant shooting the .38-caliber gun, either at Cristian when he came out of the bedroom or at Marius when he attempted to grab the shotgun. The jury’s findings would be consistent if it found defendant did not fire the shots but did find defendant caused Cristian’s death by trying to grab the shotgun out of Marius’s hand, which was not a charged provocative act. Or the jury could have believed a reasonable person would not have foreseen there was a high probability of a death of another person resulting from the alleged provocative acts. Or the jury could have concluded the provocative act was not a substantial factor in Cristian’s death. We need not catalogue all possible differences between finding defendant caused Cristian’s death and the provocative act theory based on Marius shooting Cristian. It is enough that, based on the evidence, the jury could have found defendant did an act that caused Marius to shoot Cristian but also not find true the provocative act theory based on the given instructions. This leaves the legal possibility the jury would have found defendant not guilty of first degree murder if it had been properly instructed. The People argue “there is sufficient evidence that [defendant] caused Cristian’s death.” The prejudice standard we must apply is not whether the evidence was sufficient to support a finding defendant personally shot Cristian, which we do find. Nor are we tasked with considering whether it was likely the jury did find defendant personally shot Cristian, which is possible. Instead, it is whether we can say beyond a reasonable doubt

17 the instructional error did not contribute to the verdict. (People v. Lewis, supra, 139 Cal.App.4th at p. 887.) From the evidence, we cannot conclude this is the case. We, therefore, conclude the error was prejudicial and must reverse defendant’s first degree murder conviction and the special circumstance finding. However, because we find there is substantial evidence indicating defendant did personally kill Cristian, defendant may be retried on the felony murder and the special circumstance. (People v. Franco (2009) 180 Cal.App.4th 713, 726.) II Uncharged Robbery Defendant also argues the trial court committed a reversible error in permitting evidence defendant committed another robbery. Defendant contends the trial court admitted the robbery on the basis of a common scheme or plan, but in reality it was used to prove identity. He asserts this is erroneous because identity requires a higher degree of similarity between the two crimes, and the trial court properly found the robberies were not similar enough to establish identity. The People counter the court properly found the two robberies sufficiently similar to be used for common design or plan and the evidence was admitted only for that purpose. In any case, the People contend any error was harmless. Although we reverse defendant’s conviction based on instructional error, we address this issue for the guidance of the trial court in the event of a retrial. (Pina v. County of Los Angeles (2019) 38 Cal.App.5th 531, 551 [“To assist the parties and the court, we address other alleged errors regarding issues likely to arise on retrial.”].)4 We find the admission of this uncharged offense was an abuse of discretion.

4 We do not, however, address defendant’s two other arguments as they relate to sentencing.

18 A. Additional Facts Defendant was acquitted of robbery after a jury trial in 2011. Over the course of a day, a transcript from that trial was read into the record of defendant’s 2021 trial. This transcript covered the testimony of Lee and May Kwong who described two men entering their home in Sacramento around noon on February 5, 2009. Lee, who was 88 years old at the time of the robbery, described sitting in his family room when he heard a loud noise from the back of his house. He went into the hallway and saw two men coming in from the back door, a bigger man with darker skin and a skinny smaller man who was white. One or both of the men hit Lee in the head, making him bleed, but he couldn’t remember what they hit him with. They then tied his hands and legs together with black zip-ties and one of the men put his foot on Lee’s back to keep him down. Lee testified the men were in his house for two to three hours. During this time, the men never made any verbal demands but one of them started going through the rooms and turning things over. They eventually took $150 cash, a credit card out of Lee’s wallet, a bag with about $50 cash, some collectible currency, and a 1942 Luger handgun with its ammunition. The robbers also had completely broken the back door and caused damage throughout the house. Lee testified he was in the hospital for two or three days recovering from the attack. Lee had a hard time identifying the assailants, thinking the bigger one had slightly darker skin that made him look Hispanic. Lee told law enforcement neither of the men were of Asian descent. At the 2011 trial, Lee said defendant had a similar skin color and build as the larger assailant but could not be certain it was him. May, who turned 84 years old the day after the robbery, testified to seeing Lee come back into the family room bleeding from his head. Someone then tied up her hands and put a jacket over her face. May said the men were in their house for about two hours and took some jewelry and an expensive dress. After the men left, she was able to free herself and Lee, and call 911. At this point Lee “looked pretty bad,” he had been

19 pleading with the assailants “You gonna kill me,” but was still able to walk to the ambulance. May did not identify defendant at a lineup in August 2010, but said defendant could “pass” as the bigger assailant because he had the same size. She described the smaller assailant as having blond hair and had told officers both men were wearing gloves. In 2011, defendant was tried and acquitted of the Kwong robbery. At the trial for the current offenses, law enforcement officers testified to investigating defendant for the Kwong robbery because one of the zip-ties used on the Kwongs had defendant’s DNA. Officer’s searched defendant’s garage and found two firearms, none of which was Lee’s 1942 Luger handgun. Defendant said the weapons must have been Tran’s as he let Tran use his garage; Tran denied ownership. Defendant also said people often thought he was Hispanic because he was part Korean and part Caucasian. One of the guns recovered from defendant’s garage was a .22-caliber Astra with unidentified DNA, which was uploaded into a DNA database on December 9, 2010. This DNA was later found to be Marius’s, which led to law enforcement investigating defendant for the Anton robbery. A zip-tie found at the Anton residence had DNA that was retested and was determined to be a thousand times more likely to be defendant’s than a random individual. Whereas the DNA on the zip-tie in the Kwong robbery was more than sextillion times likelier to be defendant’s than a random individual. Before the current trial, the prosecutor filed a motion to admit evidence of the Kwong robbery. The trial court found this was not relevant to prove intent because “it’s not really contested that there is an intent to steal here.” Instead, it would be relevant to identity, because “[m]ostly the argument is that it was not defendant who did it.” But the court noted that “identity does require the highest level of similarity” between the current offense and the prior conduct. The court analyzed both robberies and found that they had some similarities, including that in both cases: there were two male assailants who wore

20 gloves, the victims were present, and the robberies were initiated by hitting a victim in the head. The “most distinctive” similarity to the trial court was the use of the zip-ties. The trial court also found both incidents had similar “cross-admissib[le]” evidence: DNA that could have been defendant’s was found on zip-ties in both offenses and guns found in defendant’s garage “related to both cases.”5 Thus, “there are a lot of aspects of similarity. There are a couple of areas that are signature. There is cross-admissibility.” The trial court also did not find evidence of the prior robbery was prejudicial because: it was not remote; presenting the evidence would take “maybe one and a half days out of a ten-day trial. So it wouldn’t be an undue consumption of time”; the other case was not much stronger than the present case; there was less violence in the other case; and the evidence was cross-admissible. Still, the court recognized “there clearly is prejudice just having information from another case where there is an allegation that defendant committed an additional or another type of crime, but then you do have to balance that out with all the probative value.” The court, going back to the crimes’ similarity, was “reluctant to call [it] a signature crime because robberies do occur. They are not infrequent. [¶] Going into someone’s house might be less frequent than a burglary, for example, having the victims present, but still I am not convinced that that would be an identity that could come in for identity.” But it did find the evidence “fairly compelling under a common plan or scheme theory” because “[i]t could show that the defendant had a common method, methodology, or common plan that he used in general to commit a home invasion type robbery, and I laid those out already when I went

5 This appears to be a misstatement of the evidence. The Kwongs never testified guns were used during the robbery of their home. The only relation the guns found in defendant’s garage had between the two cases was that the search was initiated because defendant’s DNA was found at the Kwong residence and one of the guns recovered was eventually determined to have been used in the Anton robbery. But again, there was no evidence either of these guns were used during the Kwong robbery so it is not readily apparent how these guns would have been admissible at the Kwong robbery trial.

21 through the various similarities between the crimes.” Thus, the court permitted the admission of evidence relating to the Kwong robbery. The trial court gave a limiting instruction for use of the Kwong robbery under CALCRIM No. 375. This instruction stated if the jury found defendant committed the uncharged robbery by a preponderance of the evidence “you may, but are not required to, consider that evidence for the limited purpose of deciding whether the defendant had a plan or scheme to commit the offenses alleged in this case.” In closing arguments, the prosecutor argued that, though the jury in 2011 did acquit defendant of robbing the Kwongs, the standard here was lower. Here, the jury needed to find only that defendant committed the robbery by a preponderance of the evidence, which was satisfied by the “overwhelming” DNA evidence from the zip-ties in the Kwong robbery. The prosecutor then went through the similarities between the two robberies, mirroring the trial court’s reasoning, and concluded, “Is this the defendant’s common scheme or plan when he commits robberies? That’s all you can consider it for. [¶] So that’s how we know it’s the defendant.” B. Legal Standards Evidence “a person committed a crime, civil wrong, or other act” is inadmissible to establish one’s propensity for criminal behavior, but is admissible when relevant to prove “some fact,” such as “motive, opportunity, intent, preparation, plan, knowledge, [or] identity.” (Evid. Code, § 1101, subds. (a), (b).) To be relevant, the uncharged misconduct must have a degree of similarity with the charged offense, and the degree of similarity depends on the fact to prove. (People v. Ewoldt (1994) 7 Cal.4th 380, 402 (Ewoldt).) On one end of the scale is intent, which requires the “least degree of similarity (between the uncharged act and the charged offense).” (Ibid.) To be admissible to prove intent, “the uncharged misconduct must be sufficiently similar to support the inference that the defendant ‘ “probably harbor[ed] the same intent in each instance.” ’ ” (Ibid.)

22 More similarity is required when the prior bad act is introduced to prove a common plan or scheme with the charged act. Here, the “evidence of uncharged misconduct must demonstrate ‘not merely a similarity in the results, but such a concurrence of common features that the various acts are naturally to be explained as caused by a general plan of which they are the individual manifestations.’ [Citation.] ‘[T]he difference between requiring similarity, for acts negativing innocent intent, and requiring common features indicating common design, for acts showing design, is a difference of degree rather than of kind; for to be similar involves having common features, and to have common features is merely to have a high degree of similarity.’ ” (Ewoldt, supra, 7 Cal.4th at pp. 402-403.) On the end of the similarity spectrum is identity. To prove identity, the uncharged and charged conduct “ ‘must share common features that are sufficiently distinctive so as to support the inference that the same person committed both acts. [Citation.] “The pattern and characteristics of the crimes must be so unusual and distinctive as to be like a signature.” ’ [Citation.] Requiring a ‘highly unusual and distinctive nature [for] both the charged and uncharged offenses virtually eliminates the possibility that anyone other than the defendant committed the charged offense.’ ” (People v. Roldan (2005) 35 Cal.4th 646, 706, disapproved on another ground in People v. Doolin (2009) 45 Cal.4th 390, 422, fn. 22.) If sufficiently similar to be relevant to prove the asserted fact, “the trial court then must consider whether the probative value of the evidence ‘is “substantially outweighed by the probability that its admission [would] . . . create substantial danger of undue prejudice, of confusing the issues, or of misleading the jury.” ’ ” (People v. Foster (2010) 50 Cal.4th 1301, 1328.) “ ‘ “ ‘Because evidence of other crimes may be highly inflammatory, its admissibility should be scrutinized with great care.’ ” ’ ” (People v. Jones (2013) 57 Cal.4th 899, 930.)

23 “ ‘On appeal, a trial court’s resolution of these issues is reviewed for abuse of discretion. [Citation.] A court abuses it’s discretion when its ruling “falls outside the bounds of reason.” ’ ” (People v. Catlin (2001) 26 Cal.4th 81, 122.) C. Analysis 1. Relevance: Common Plan vs. Identity We first examine the relevance of this evidence. The trial court admitted evidence of the Kwong robbery to show a common plan with the Anton robbery. But we conclude the Kwong robbery was instead used to prove identity. As the California Supreme Court has noted, the difference “is subtle but significant.” (Ewoldt, supra, 7 Cal.4th at p. 394, fn. 2.) This subtle difference is determinative in this case so we find a full analysis of their differences will prove helpful. The meaning of “common plan” has evolved. In People v. Tassell (1984) 36 Cal.3d 77, the California Supreme Court determined “common plan” required evidence the two crimes were part of one singular plan or “ ‘grand design.’ ” (Id. at p. 84.) Ten years later, the court invalidated this rule in Ewoldt. (Ewoldt, supra, 7 Cal.4th at p. 398 [“In [Tassell], this court departed from its previous approach to the question of the admissibility of uncharged criminal acts to demonstrate a common design or plan.”].) The court explained the correct rule: “ ‘The presence of a design or plan to do or not to do a given act has probative value to show that the act was in fact done or not done.’ [Citation.] For example, a letter written by the defendant stating he planned to commit a certain offense would be relevant evidence in a subsequent prosecution of the defendant for committing that offense. [Citation.] The existence of such a design or plan also may be proved circumstantially by evidence that the defendant has performed acts having ‘such a concurrence of common features that the various acts are naturally to be explained as caused by a general plan of which they are the individual manifestations.’ [Citation.] Evidence of a common design or plan, therefore, is not used to prove the defendant’s intent or identity but rather to prove that the defendant engaged in the

24 conduct alleged to constitute the charged offense.” (Id. at pp. 393-394.) Thus, “[t]o establish the existence of a common design or plan, the common features must indicate the existence of a plan rather than a series of similar spontaneous acts, but the plan thus revealed need not be distinctive or unusual.” (Id. at p. 403.) The court explained the distinction between admitting evidence to prove a common plan as opposed “to prove intent or identity, is subtle but significant.” (Ewoldt, supra, 7 Cal.4th at p. 394, fn. 2.) “Evidence of a common design or plan is admissible to establish that the defendant committed the act alleged” because “ ‘the act is still undetermined.’ ” (Ibid.) Conversely, “[e]vidence of identity is admissible where it is conceded or assumed that the charged offense was committed by someone, in order to prove that the defendant was the perpetrator. For example, in a prosecution for shoplifting in which it was conceded or assumed that a theft was committed by an unidentified person, evidence that the defendant had committed uncharged acts of shoplifting in the same unusual and distinctive manner as the charged offense might be admitted to establish that the defendant was the perpetrator of the charged offense.” (Ibid.) Commonality is sufficient to prove identity is sometimes called a modus operandi. “[A]n inference of identity can be drawn from ‘a distinctive modus operandi,’ ” but “to be admissible on this basis, ‘the evidence must disclose common marks or identifiers, that, considered singly or in combination, support a strong inference that the defendant committed both crimes.’ [Citation.] ‘These common marks must be distinctive rather than ordinary aspects of any such category of crime. They must be sufficiently distinctive that they bear defendant’s unique “signature.” Reaching a conclusion that offenses are signature crimes requires a comparison of the degree of distinctiveness of shared marks with the common or minimally distinctive aspects of each crime.’ ” (People v. Earle (2009) 172 Cal.App.4th 372, 394.)

25 Our Supreme Court in People v. Bradford (1997) 15 Cal.4th 1229 provided a helpful analysis of what kind of evidence is required to satisfy the similarity requirements for modus operandi. In that case, the court found the crimes similar enough where the “defendant was acquainted with each victim, had utilized the career and monetary ambitions of each to induce her to accompany him, and had taken each victim to the same remote desert location during a period at or preceding the commission of each murder.” (Id. at p. 1317.) It contrasted this with another case that found insufficient similarity, People v. Bean (1998) 46 Cal.3d 919, and explained that in Bean, “both victims were females of the same age who were killed within three days of each other, in the same general vicinity and consistently with the same general scheme to burglarize their residences, were insufficiently distinctive to permit a finding of a common modus operandi. We noted that in one offense two assailants entered a couple’s residence through a window, attacked the victim with a ball peen hammer, and took substantial property, whereas in the other, a single assailant gained entry through the door and took a purse, causing the victim to die of a heart attack.” (Bradford, at p. 1317.) Common plan and modus operandi are related, but distinct. Ewoldt and its disapproval of Tassell provides additional insight on this point. In Tassell, the court found that “[t]here being no issue of identity, it is immaterial whether the modus operandi of the charged crime was similar to that of the uncharged offenses. While the People rely on the ‘common plan or scheme’ rationale for admissibility, under the circumstances that is merely a euphemism for ‘disposition.’ The evidence was not admissible.” (People v. Tassell, supra, 36 Cal.3d at p. 89.) In Ewoldt, the court rejected this rationale, explaining “it is relevant that the modus operandi of uncharged offenses committed by a defendant is markedly similar to the charged offense, even when the evidence is admitted not to prove identity or intent, but to establish that the uncharged offenses and the charged offense are manifestations of a common design or plan. [Citation.] The decision in Tassell failed to recognize that evidence of similar, uncharged misconduct may be used not only to prove

26 intent or identity, but also to show a common design or plan.” (Ewoldt, supra, 7 Cal.4th at p. 399.) This tells us common plan and modus operandi are opposite sides of the same coin. The type of similarities analyzed for “common plan” evidence and “modus operandi” evidence is the same. They both require “common features” or “common marks” between the charged and uncharged conduct. The critical difference is the purpose of the evidence. A “common plan” proves an act, and whether this act exists as part of the same plan. “Modus operandi” proves identity, and whether it was the defendant who committed the act. The name assigned to the collection of commonalities is insignificant, which may be why some cases use common plan and modus operandi interchangeably. (See People v. Johnson (2013) 221 Cal.App.4th 623, 635 [“Appellant maintains that ‘the other-crimes evidence was insufficiently similar to the charged incident to justify its admission’ to prove intent and common design or plan. We disagree. . . . The modus operandi used to gain admission into the residences was the same” (italics added)]; People v. Chhoun (2021) 11 Cal.5th 1, 27-28 [citing the passage from Johnson].) What is significant is the fact to prove and the level of similarity required to prove it. This distinction between the fact to prove is decisive here. The trial court began its analysis by noting the evidence was relevant to prove identity: “[m]ostly the argument is that it was not the defendant who did it.” And the parties acted as though the only issue was identity. Defense counsel’s entire closing argument was devoted to arguing it was not defendant who robbed the Antons, and the prosecutor commented on this in rebuttal before focusing on the identity evidence. The trial court’s analysis of the Kwong robbery evidence consequently focused on whether there was a sufficient similarity to meet the “highest level” needed to be used for identity evidence. It ultimately could not conclude the robberies were sufficiently similar to establish a signature necessary to prove identity, so it then shifted and found the crimes similar enough to be a common

27 plan. But, crucially, the trial court never explained how a common plan was relevant to the Anton robbery. Put another way, what act in the Kwong robbery could help prove what occurred in the Anton robbery. The fact the crimes might be similar enough to establish a common plan did not make a common plan relevant. (See People v. Alcala (1984) 36 Cal.3d 604, 634 [cautioning that calling evidence a “ ‘plan or scheme’ . . . adds nothing to a case for the admission of prior offenses. The proffered evidence must still be analyzed to determine whether it proves something material, disputed, and beyond bare disposition.”].) The Kwong robbery was not relevant to prove an act because there was no question a death occurred during the robbery. Though there was conflicting evidence on exactly how Cristian died, a common plan with the Kwong robbery could not shed light on this fact because no death occurred with the Kwong robbery. The shooting was instead the result of a lack of a plan, involving a shotgun the victims possessed, a fact the robbers evidently failed to foresee. Even though the jury instructions stated the evidence could be considered only in deciding “whether the defendant had a plan or scheme to commit the offenses alleged,” it could not make the Kwong robbery relevant for that purpose, and instead created an independent risk of instructional error. (People v. Cross (2008) 45 Cal.4th 58, 67 [“Giving an instruction that is correct as to the law but irrelevant or inapplicable is error.”].) And any clarity the instructions could provide was severely undermined by the prosecutor stating in closing, “That’s all you can consider it for. So that’s how we know it’s the defendant.” This is an explicit call to use the evidence to establish identity. Ultimately, this case aligns with the shoplifting hypothetical posed in Ewoldt, where it was “assumed that the charged offense was committed by someone,” so the Kwong robbery was admitted, “to prove that the defendant was the perpetrator.” (Ewoldt, supra, 7 Cal.4th at p. 394, fn. 2.)

28 2. Similarity The crimes were also not sufficiently similar to establish a common plan, let alone an identity. The similarities, as the court found, were: two male assailants, wearing gloves, who had zip-ties, the victims were present, and the victims were hit. The similarities of the two of these becomes questionable on closer inspection. Marius was not sure when and how he was hit, but Tran admitted he was the one that tussled with Marius. Contrary to the trial court’s finding, the evidence indicated this attack on Marius was not part of a plan but a spontaneous act resulting from Marius fighting back. And the most significant similarity to the trial court, the zip ties, were in a manner dissimilar: The ones used in the Kwong robbery were black whereas the ones used in the Anton robbery were white. Regardless, any similarity between the two robberies pale in comparison to the material dissimilarities: the timing of the crimes — one around noon and the other just after midnight; the location of the crimes — the houses were in different parts of Sacramento; the type of victims — an older couple compared with young men; the manner of entering the home — breaking down the back door compared with waiting outside and forcing someone inside the front door; weapons used — the Kwongs did not testify to seeing the robbers have guns whereas guns were brought to the Anton robbery; the target of the robbery — anything of perceived value compared to cash and cocaine; dates of the crime — they were a year apart; and the accomplices — a blonde man not of Asian descent and Tran, who was Vietnamese. These dissimilarities exhibit a lack of a shared plan. Also, the “cross-admissib[le]” evidence could not establish the crimes had commonalities. This evidence included defendant’s DNA on the zip-ties used to tie up the Kwongs and the gun found in defendant’s garage with Marius’s blood. The trial court implied it thought this evidence indicated a “signature.” However, this is evidence of defendant’s identity as one of the Kwong robbers, not evidence of how similarly the

29 crimes were executed. Evidence of defendant’s identity as one of the Kwong robbers was required for the jury’s consideration of the prior crime evidence because it had to find by a preponderance of the evidence defendant committed that robbery. But what is important for the admissibility of the Kwong robbery, to prove either a common plan or identity, is how defendant committed the crimes, examining the common features between the Kwong robbery and the Anton robbery. This so-called cross-admissible evidence had no tendency to prove a common scheme or modus operandi because there is no indication it was part of defendant’s common plan or modus operandi to leave zip-ties at each crime scene with his DNA. In summary, these crimes had only superficial similarities. The Kwong robbery was used to prove identity, the central issue in the case. However, as the trial court ultimately concluded, the crimes were not similar enough to meet the heightened similarity requirement for identity or a modus operandi. But even if the Kwong robbery was admitted to establish a common plan, the limited similarities between the two crimes exhibit no shared common plan. Thus, the Kwong robbery did not have sufficient similarity with the present robbery, and therefore no admissible relevance to the present case, regardless of whether they were considered for identity or for common plan. 3. Prejudice Uncharged conduct is inherently prejudicial even when relevant. Such evidence is overwhelmingly prejudicial, and consequently inadmissible, when the relevance is slight. (Ewoldt, supra, 7 Cal.4th at p. 404 [“ ‘Since “substantial prejudicial effect [is] inherent in [such] evidence,” uncharged offenses are admissible only if they have substantial probative value’ ”].) That is the case here. Over the course of a day, the jury in defendant’s trial heard Lee’s testimony about being beaten and stepped on, and May’s testimony that Lee had told the intruders they were going to kill him. This was highly prejudicial evidence that had no relevance to the present case. Even assuming the Kwong robbery exhibited a common plan, beating an elderly man had no similarity with the

30 present case; there was no evidence defendant beat anybody during the Anton robbery. And though the Kwong robbery did not result in a death, that does not erase from the jury’s mind the testimony of violence against an elderly couple. Given the lack of relevance of the Kwong robbery, it was not admissible. But even if it had minimal relevance, this relevance was not sufficient to overcome the inflammatory nature of the evidence. Thus, relying on the record before us, the Kwong robbery was prejudicial. 4. Conclusion Ultimately, given the lack of probative value of the Kwong robbery, the trial court abused its discretion in admitting this evidence. Finding error, the next step would be to consider whether this prejudiced defendant and rendered his trial fundamentally unfair. (People v. Winkler (2020) 56 Cal.App.5th 1102, 1164.) We need not make this determination as we are reversing on instructional error. It is enough for us to find it was in error to admit evidence of the Kwong robbery under the circumstances of defendant’s 2021 trial as guidance for any retrial.

31 DISPOSITION Defendant’s conviction for first degree murder and the special circumstance finding is reversed and the matter is remanded for possible retrial and such further proceedings as may be just under the circumstances.

\s\ , McADAM, J.*

We concur:

\s\ , ROBIE, Acting P. J.

\s\ , MAURO, J.

* Judge of the Yolo County Superior Court, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.