Green v. Franke

*303 BREWER, J. A jury convicted petitioner of eighteen sex offenses involving nine victims, ages twelve through seventeen. After an unsuccessful direct appeal, petitioner sought post-conviction relief, alleging that his trial counsel’s performance had not satisfied the minimum requirements demanded by the Oregon and United States Constitutions. Among other claims, petitioner alleged in his pleading that counsel performed inadequately by failing to request an instruction directing the jury to consider the evidence concerning each alleged victim separately and only as that evidence pertained to a specific charge or charges relating to that victim. The post-conviction court entered a judgment denying post-conviction relief. The Court of Appeals reversed. Green v. Franke, 261 Or App 49, 323 P3d 321 (2014). It reasoned that “there was no evident downside to requesting such an instruction; the upside, however, was plain: The jury would have been prohibited from concluding that petitioner had committed the charged acts based on a belief that he had a propensity to commit such acts.” Id. at 58. The Court of Appeals further concluded that petitioner was prejudiced by counsel’s omission because the jury was encouraged by the prosecutor, and permitted by defense counsel’s failure to obtain a limiting instruction, to rely on impermissible propensity inferences in its consideration of most of the charges. Id. at 67-68. We allowed the state’s petition for review to consider the recurring issues of what a post-conviction petitioner must show to establish inadequate performance of counsel and what a petitioner must prove to establish that counsel’s inadequate performance prejudiced the petitioner’s case. We now reverse the decision of the Court of Appeals, and we reverse and remand the judgment of the post-conviction court denying post-conviction relief on petitioner’s instructional claim while otherwise affirming that judgment. PROCEDURAL HISTORY AND FACTS The state charged petitioner with sex offenses against nine victims committed over the course of approximately *304 five years. 1 The victims were girls ranging in age from 12 to 17 years old. Some of the charges were based on sexual contact that, although “consensual” in a colloquial sense, was without lawful consent due to the ages of the victims (in particular, the charges involving victims KN, MZ, and CO). But the majority of the crimes were based on either a theory of forcible compulsion or on lack of consent (specifically, the charges involving victims SB, DH, CH, JA, BB, and RM). 2 Before petitioner’s criminal trial, counsel did not move to sever the charges against petitioner for purposes of trial. By the time of trial, counsel knew that petitioner had acknowledged to police and to an underage girlfriend that he had, in fact, had sex with some of the victims. Petitioner also admitted to counsel that he had had sex with several of the victims. Petitioner insisted, however, that those encounters were consensual, not forcible. Given that information, counsel did not believe that he could mount a plausible defense to the charges involving “consensual” sex — that is, the charges that were based solely on the victims’ ages. Thus, he decided to concede to the jury that petitioner had committed the charged crimes against KN, MZ, and CO. Counsel decided, instead, to focus his efforts on the charges involving the other victims, all of which depended on proof that petitioner had engaged in sexual contact either using forcible compulsion or without their actual consent. Counsel’s theory on the majority of those charges was that, although petitioner may have had sex with several of the girls, he did so only with their consent. Counsel did not *305 concede, however, that petitioner had sex with JA or BB. His defense to the charges involving those victims was that no sexual contact occurred. At trial, the state did not argue that evidence relating to any one of the joined charges was relevant to prove any other charge. However, all the charges were tried to a jury in a single trial. The state called each of the victims to testify about their interactions with petitioner and about sexual acts that he committed against them. Three of the victims described consensual sex with petitioner, and the other six testified that petitioner against their will touched their sexually intimate parts, caused them to touch his sexually intimate parts, or — in the case of the five first-degree rape charges — forcibly compelled them to have sexual intercourse while they either resisted or told him to stop. The state also called police detectives and other witnesses to testify about statements that the victims had made to them. Petitioner’s counsel cross-examined the victims and other witnesses in an attempt to emphasize evidence showing that any sexual contacts that petitioner had with the victims were, in fact, consensual. Counsel stressed that type of evidence on cross-examination even if the charge otherwise was uncontested. 3 Counsel also attempted to persuade the jury that several of the victims were connected to each other and, to some extent, that the evidence about their allegations should be considered together. Counsel did that in two primary ways. First, he pointed out that some of the victims knew each other, either in general or in how they came to report their allegations to police. As an example, during his opening statement, counsel told the jury: “I think detectives are going to testify that Detective Fryett was making one investigation and Detective Young was making another investigation over here, but the majority of these girls really do know each other. And it’s — it started out that there were two victims, and then [petitioner] was *306 on TV, and all of a sudden there’s now nine victims. [The prosecutor] said he thinks the evidence is going to show that these people didn’t know each other; these girls didn’t know each other very much. But I think the testimony is going to come out that they really did know each other, a lot of them.” In his closing argument, counsel suggested a possible motive for a particular victim (JA) to have fabricated a charge of rape: “Does [JA’s allegation] make any sense? I submit that it does not, because the rape of her did not happen. I don’t know why she’s lying. I don’t have any idea. But I can tell you whether all these girls knew each other before, they all know each other now. And if you don’t think the detectives and the Victims’ Assistance [office] have talked to these people, that’s nonsense. They’ve talked to these girls, they’ve gotten them ready for trial, and they came and testified.” The second way in which counsel attempted to connect evidence involving different victims was by contrasting certain victims’ allegations with what other victims had reported. For example, counsel elicited differences between the circumstances that CH and SB had reported, and he generally noted how the reports of some victims did not fit the “pattern” of what other victims reported. Counsel repeated that theme in closing argument, telling the jury that RM’s report was “out of the pattern of what some of these other folks have said.” For its part, the state emphasized in closing argument that the resolution of disputed factual issues in the case required a credibility assessment: “Now, in hearing from many of the victims, it’s a matter of assessing credibility. And when you go into the jury room to deliberate, we’re not asking you to leave your common sense outside the door. We ask you to draw upon your experience as human beings in assessing credibility.” The prosecutor then explained why the jury should find each victim credible. Referring to evidence that petitioner had attempted to influence the testimony of certain witnesses, the prosecutor argued that, *307 “when you look at [petitioner’s] statements during the course of this trial, and of the investigation, and his tampering with witnesses, they really give you insight into his sexual assaults. And they parallel his predatory nature.” The prosecutor continued: “He’s assaulting victims while they’re sleeping and vulnerable. [SB] and [DH]. He takes the victims by swift attack. [RM], [JA], and [CH]. He takes the victims through manipulation or subtle forms of coercion. [BB] and [KN]. He offers reassurance to each victim, or tries to make them believe that they want it.” The jury ultimately convicted petitioner of all of the charges, including the charges based on a theory of forcible compulsion and lack of actual consent. After an unsuccessful direct appeal, petitioner filed this action for post-conviction relief, asserting several claims of inadequate assistance of trial counsel. The amended petition for post-conviction relief set out 12 allegations specifying the ways that petitioner claimed his counsel’s representation in his criminal trial was constitutionally inadequate. As noted, in the claim that is pertinent on review, petitioner alleged that trial counsel’s performance was inadequate because he “failed to request that the trial court instruct the jury that it was required to consider the evidence concerning each alleged victim separately and only as that evidence related to a specific charge or charges relating to that specific alleged victim.” In response to petitioner’s pleaded claims, the state obtained two affidavits from petitioner’s trial counsel. In general, the affidavits described the challenges that counsel faced in defending the various charges in the criminal case, explained counsel’s investigative efforts, and set out counsel’s reasons for actions that counsel did and did not take in the course of his investigation before trial and in his representation at trial. On the allegation raising counsel’s failure to request a limiting instruction, counsel explained that “the Judge gave instructions about each and every count that the State must prove” and the judge “read each and every count and told the jury they had to find on every count, so I don’t know what counsel is saying.” (Emphasis *308 added.) Counsel opined (using a rape charge as an example) that the instructions were sufficient. 4 Six months after petitioner filed the amended petition, and ten days before trial, petitioner filed a trial memorandum. Although the bulk of the 53-page memorandum was devoted to the other allegations of his petition, it included an argument that clarified petitioner’s claim based on counsel’s failure to ask for a limiting instruction. The memorandum urged that an instruction similar to the one approved in State v. Kitzman, 129 Or App 520, 529, 879 P2d 1326 (1994), aff’d in part and rev’d in part on other grounds, 323 Or 589, 920 P2d 134(1996), would have been appropriate and could “have helped alleviate the potential for the jury to find petitioner guilty of all the charges simply because, with nine separate victims, it believed that he had a propensity for sexually abusing minor females.” 5 That instruction provided: “[Defendant] has been charged with several unrelated counts in a single indictment. A separate crime is charged in each count of the Indictment. Each charge and the evidence pertaining to it must be considered separately by the jury. ****** “In other words, it is your duty to consider the evidence solely for the charge it pertains to and no other charge.” Kitzman, 129 Or App at 529 (alterations in original). 6 *309 Ten days later, the state filed and served its own trial memorandum. The state’s response to the claim at issue consisted of the following: “The State incorporates by reference the two affidavits of [trial counsel] that respond to petitioner’s claims. As trial counsel’s affidavits confirm, petitioner’s claims lack merit. The requests for jury instructions by the State and trial counsel covered the key elements of petitioner’s case. Petitioner will fail to prove that prejudice resulted from counsel not requesting one additional instruction. To be successful at this post-conviction level, petitioner would have to prove that: (1) petitioner had a legal basis to request an additional instruction; (2) petitioner actually wanted the instruction; (3) the trial court would have allowed the request and instructed the jury accordingly; (4) the one additional instruction would have altered any juror’s vote, and; (5) the one additional instruction would have caused enough jurors to change their votes so that the jury verdicts would have been different. Petitioner cannot satisfy his burden of proof on these issues, and therefore cannot show any trial counsel deficiency or resulting prejudice.” (Internal citations to record omitted.) 7 Like many post-conviction cases, this case was tried in the main on a documentary record. 8 Neither party called trial counsel as a witness, nor did they present any further sworn statement from him. Instead, trial counsel’s “testimony” consisted solely of his pretrial affidavits. The hearing itself was short. At the hearing, post-conviction counsel focused on other claims in the petition and did not mention the failure to request a limiting instruction. At the conclusion of the hearing, the post-conviction court rejected each of petitioner’s claims, commenting on each one only briefly. With respect to counsel’s failure to request a limiting instruction, the court said: “I don’t think it *310 would have changed anything at all. Clearly, I think that was a proper instruction. Again, I don’t think the — the result would have been any different.” In short, the post-conviction court did not decide whether or not counsel’s performance was inadequate with respect to the instructional claim; instead, the court concluded that petitioner could not prevail on that claim because he had failed to establish prejudice. Petitioner appealed the ensuing judgment denying post-conviction relief. The Court of Appeals affirmed with respect to all but one of petitioner’s claims; as noted, it reversed on the claim involving trial counsel’s failure to request a limiting instruction. This court’s decision in State v. Leistiko, 352 Or 172, 282 P3d 857, adh’d to as modified on recons, 352 Or 622, 292 P3d 522 (2012), was central to the Court of Appeals’ holding. In Leistiko, this court held, among other things, that evidence of other sexual crimes is not relevant under OEC 404(3) to prove a defendant’s propensity to commit a charged offense. Leistiko, 352 Or at 180. Based on that holding, the Court of Appeals concluded in this case that petitioner’s trial counsel should have identified the “great and obvious danger” that the jury would rely on impermissible propensity inferences in petitioner’s case. Green, 261 Or App at 58. 9 The court ultimately concluded that counsel performed inadequately because “there was no evident downside to requesting such an instruction.” Id. (citing Pereida-Alba v. Coursey, 252 Or App 66, 71, 284 P3d 1280 (2012), rev’d, 356 Or 654, 342 P3d 70 (2015)). The court arrived at that conclusion without expressly considering how the limiting instruction would have meshed, or potentially conflicted, *311 with counsel’s defense strategy at trial. That comes as no surprise, because (as discussed in greater detail below) neither party addressed either before the post-conviction court or the Court of Appeals the effect, if any, of counsel’s trial strategy on the adequacy of his performance in failing to request a limiting instruction. Turning to the issue of prejudice, the Court of Appeals — again, using Leistiko as an example — noted the importance of instructions limiting a jury’s use of evidence in “joined proceedings.” Id. at 59. The court concluded that the absence of a limiting instruction prejudiced petitioner with respect to all of his convictions, except the ones as to which he admitted guilt. Without a limiting instruction, the court reasoned, the jury “may well have,” was “permitted” to, or was “invited to” make improper inferences about petitioner’s propensity to commit sex offenses against one underage victim based on evidence that was admissible only with respect to other charges involving different victims. Id. at 63-67. On that ground, the Court of Appeals held that trial counsel’s deficient performance had a tendency to affect the outcome of the case and, therefore, petitioner had suffered prejudice. See id. at 59. LEGAL PRINCIPLES AND STANDARD OF REVIEW Post-conviction relief is warranted when there has been a “substantial denial” of a petitioner’s “rights under the Constitution of the United States, or under the Constitution of the State of Oregon, or both, and which denial rendered the conviction void.” ORS 138.530(l)(a). The state-based right to adequate assistance of counsel derives from Article I, section 11, of the Oregon Constitution, which provides, in part, that “ [i] n all criminal prosecutions, the accused shall have the right * * * to be heard by himself and counsel.” The federal right to effective assistance of counsel derives from the Sixth Amendment to the United States Constitution’s guarantee that “[i]n all criminal prosecutions, the accused shall enjoy the right *** to have the Assistance of Counsel for his defence.” This court has said that, although those provisions are “worded differently,” they “embody similar objectives.” Krummacher v. Gierloff, 290 Or 867, 871, 627 P2d 458 (1981). *312 In evaluating whether a criminal defense lawyer has rendered inadequate assistance under the Oregon Constitution, our analysis ordinarily proceeds in two steps: “‘First, we must determine whether petitioner demonstrated by a preponderance of the evidence that [his or her counsel] failed to exercise reasonable professional skill and judgment. Second, if we conclude that petitioner met that burden, we further must determine whether he proved that counsel’s failure had a tendency to affect the result of his trial.’” Montez v. Czerniak, 355 Or 1, 7, 322 P3d 487, adh’d to as modified on recons, 355 Or 598, 330 P3d 595 (2014) (quoting Lichau v. Baldwin, 333 Or 350, 359, 39 P3d 851 (2002) (internal citations omitted)). Only general statements can be made about what constitutes the exercise of professional skill and judgment. Krummacher, 290 Or at 873. Generally speaking, “counsel must * * * prepare himself on the law to the extent appropriate to the nature and complexity of the case.” Id. at 875. Counsel need not, however, “expend time and energy uselessly or for negligible potential benefit under the circumstances of the case.” Id. at 874. This court’s review of a post-conviction court’s determinations is not open-ended. We review such proceedings for errors of law. Peiffer v. Hoyt, 339 Or 649, 660, 125 P3d 734 (2005). A post-conviction court’s findings of historical fact are binding on this court if there is evidence in the record to support them. Lichau, 333 Or at 359. If the post-conviction court failed to make findings of fact on all the issues — and there is evidence from which such facts could be decided more than one way — we will presume that the facts were decided consistently with the post-conviction court’s conclusions of law. Id. ANALYSIS As discussed, in this case, petitioner contends that his trial counsel’s performance was inadequate either because counsel did not make a conscious choice to forego asking for a limiting jury instruction or because, if he made that choice, no reasonable counsel would have done so. This court recently addressed a similar problem in Pereida-Alba v. Coursey, 356 Or 654, 342 P3d 70 (2015). Because the *313 analysis in that case is central to our analysis here, we first discuss that case in some detail. The petitioner in Pereida-Alba, having been convicted of first-degree robbery, asserted in a post-conviction action that his trial counsel “either did not decide or reasonably could not have decided to forego giving the jury the option of convicting him of the lesser-included offense of third-degree robbery.” Id. at 656. In Pereida-Alba, neither party presented testimonial evidence from the petitioner’s trial counsel. The state did, however, assert before the post-conviction court that counsel’s failure to request the posited instruction would have been a reasonable tactical decision. In that regard, the state argued that, “given the store employee’s favorable testimony, [the] petitioner’s trial counsel reasonably could have made a tactical choice to limit the jury’s options to convicting petitioner of first-degree robbery or acquitting him.” Id. at 660. The post-conviction court ruled that no reasonable counsel would have failed to ask for an instruction on the lesser-included offense and entered judgment in the petitioner’s favor. The Court of Appeals affirmed. It observed that the mandatory minimum sentence for first-degree robbery is 90 months’ imprisonment, while the guidelines sentence for third-degree robbery ranges from probation to 16 months. Pereida-Alba, 252 Or App at 71. Explaining that the evidence permitted the petitioner to argue that he had committed third-degree robbery but not first-degree robbery, the court reasoned: “The upshot is that there was no evident downside to petitioner from requesting an instruction on third-degree robbery and a significant potential benefit to him from doing so.” Id. The Court of Appeals concluded, based on its assessment of the potential risks and benefits, that “the post-conviction court reasonably could [and implicitly did] infer that the defense attorney’s failure to request an instruction on third-degree robbery was attributable to the attorney’s failure to consider whether to make such a request.” Id. The court also concluded that counsel’s failure to consider whether to request an instruction on third-degree robbery amounted *314 to inadequate assistance. Id. Finally, the Court of Appeals concluded that counsel’s omission prejudiced the petitioner because “the jury did not have a complete statement of the law.” Id. at 72. After an extensive examination of trial counsel’s defense of the petitioner in the underlying criminal case, this court concluded on review that counsel reasonably could have determined that the absence of an instruction on third-degree robbery would not have impermissibly skewed the jury’s verdict towards a conviction for first-degree robbery. Pereida-Alba, 356 Or at 668. This court further concluded that counsel reasonably could have believed that, when faced with the choice of convicting the petitioner of the doubtful crime of first-degree robbery or acquitting him altogether, the jury would choose the latter course. Id. The remaining question was whether counsel had made a conscious choice to forego asking for an instruction on third-degree robbery and, if she had failed to consider that option, what the legal consequences of that failure were. This court explained that “the failure to consider every possible tactical choice does not automatically establish inadequate assistance.” Id. at 670. In so concluding, this court implicitly rejected as incomplete the Court of Appeals’ framing of the inadequate assistance inquiry in terms of whether there “was no evident downside to petitioner from requesting an instruction” and “a significant potential benefit to him from doing so.” See id. at 661. That is, this court in effect held that, even if such a cost-benefit inquiry suggests that counsel’s failure to seek a limiting instruction may have been deficient, the question remains whether the omission constituted inadequate assistance, especially when viewed in light of the strategy that the petitioner’s counsel did pursue. See id. at 674. This court in Pereida-Alba further concluded that it could not assume that the post-conviction court implicitly found that counsel failed to consider asking for an instruction on third-degree robbery. The court noted that the petitioner had the burdens of production and persuasion on his claim and that there was no direct evidence that his counsel inadvertently had failed to seek a lesser-included offense *315 instruction; the question then became “whether the circumstantial evidence in the record [was] sufficient to meet petitioner’s burden of production.” Id. at 672. Because such circumstantial evidence did exist, this court remanded the case to the post-conviction court with the following instructions: “If the post-conviction court finds on remand that petitioner has not met his burden of persuasion on that issue and that petitioner’s counsel made a tactical choice, then that choice was a reasonable one for the reasons explained above. Conversely, if the post-conviction [court] finds that petitioner has met his burden of persuasion and that his counsel failed to consider asking for an instruction on the lesser-included offense of third-degree robbery, then the question whether petitioner’s counsel provided constitutionally adequate assistance becomes slightly more complex. As this court recently explained, the failure to consider an issue or undertake a particular investigation does not automatically constitute inadequate assistance. However, the absence of strategic thought or direction on the part of a defense team can constitute inadequate assistance. “As Montez explains, whether the failure to consider an issue constitutes inadequate assistance will turn on, among other things, whether the strategy that defense counsel did employ was reasonable, the relationship between the evidence or theory that defense counsel failed to consider and the strategy that counsel did pursue, and the extent to which counsel should have been aware of the strategy that petitioner now identifies.” Id. at 673-74 (internal quotations and citations omitted). This court in Pereida-Alba ultimately held that, if, on remand, the post-conviction court found that counsel failed to consider asking for an instruction on third-degree robbery, then the post-conviction court would have to decide whether that failure constituted inadequate assistance based on the reason for that failure, considered in light of the strategy that the court found that counsel did pursue. Id. at 674. As further background, we briefly review certain principles pertaining to the joinder of criminal charges involving offenses of the same or similar character. ORS 132.560 provides, in pertinent part: *316 “(1) A charging instrument must charge but one offense, and in one form only, except that: ut 111 * “(b) Two or more offenses may be charged in the same charging instrument in a separate count for each offense if the offenses charged are alleged to have been committed by the same person or persons and are: “(A) Of the same or similar character. * * * * “(3) If it appears, upon motion, that the state or defendant is substantially prejudiced by a joinder of offenses under subsection (1) or (2) of this section, the court may order an election or separate trials of counts or provide whatever other relief justice requires.” In analyzing questions of joinder and severance of criminal charges under ORS 132.560(3), a trial court must focus on “any circumstance” that impairs a defendant’s right to a fair trial. State v. Miller, 327 Or 622, 633, 969 P2d 1006 (1998). 10 Of particular relevance is the “probable effectiveness of limiting instructions given to the jury by the court.” State v. Staley, 142 Or App 583, 589, 923 P2d 650 (1996), rev den, 324 Or 560 (1997). Where charges of the same or similar character are jointly tried, this court has recognized that “trial courts have the authority to give limiting instructions to juries that require them to consider evidence only for a particular purpose or in regard to a particular element.” State v. Moore / Coen, 349 Or 371, 391, 245 P3d 101 (2010), cert den, 563 US 996, 131 S Ct 2461, 179 L Ed 2d 1225 (2011). In particular, “[w]hen a trial court declines to sever joined offenses, and evidence relating to one offense is not admissible to prove another joined offense, a trial court ordinarily will instruct the jury to consider the evidence on each offense separately *317 to prevent the jury from using the evidence offered to prove one offense to decide another joined offense.” Leistiko, 352 Or at 178 (“recognizing the risk that a jury may use evidence admitted to prove one count in deciding whether the state has proved a joined count”). In fact, “[w]hen evidence which is admissible as to one party or for one purpose but not admissible as to another party or for another purpose is admitted, the court, upon request, shall restrict the evidence to its proper scope and instruct the jury accordingly.” OEC 105. With that background, we return to this case. As noted, the state has not asserted that evidence that defendant committed a crime against'a particular victim was admissible to prove that petitioner committed a charged offense alleged against a different victim. 11 Nor does the state dispute that competent defense counsel ordinarily would seek an appropriate limiting instruction in a case where multiple charges involving different victims and separate criminal episodes have been joined for trial. Rather, the state asserts that, although neither party offered evidence from petitioner’s trial counsel explaining why counsel did not request a limiting instruction, counsel’s strategy in the criminal trial was to acknowledge petitioner’s guilt as to the charges involving consensual sexual contact, but to defeat the charges involving forcible sexual contact, and a *318 non-propensity instruction could have conflicted with that strategy. The state explains: “[T]rial counsel believed that petitioner had no plausible defense to the charges that were based on consensual sexual contact with the victims. Thus, he focused his overarching trial strategy on creating doubt about the more serious charges that depended on forcible sexual contact. He wanted the jury to believe that, although petitioner had sexual contact with several of those victims, it was — like the offenses to which he had admitted — consensual.” (Emphases in original.) The state urges that, given that strategy to portray all of the sexual encounters in which petitioner engaged as consensual, a reasonable attorney might “have wanted the jury to infer that the fact that petitioner had consensual sex with some victims made it more likely that sex with other victims was also consensual,” which would have conflicted with a limiting instruction of the kind that petitioner claims should have been requested. The state also points out that counsel emphasized in argument that many of the victims knew each other (suggesting that they had perhaps collaborated on their versions of what occurred) and counsel in argument compared the accounts of some of the victims to those of other victims. In the state’s view, counsel could not have made those arguments if the limiting instruction had been given. Petitioner responds that, in the event that the jury found that petitioner committed a sex offense against a particular victim, petitioner needed to mitigate the risk of being convicted of a charge involving a different victim based on a perceived propensity to sexually abuse underage girls. According to petitioner, that need was not necessarily incompatible with the trial strategy that counsel pursued. As petitioner sees it, his counsel’s defense strategy reduced to trying to impeach the victim’s credibility, and impeaching a victim’s testimony with evidence that she knew another victim or that she described a different “pattern” of abuse from other victims would not have been inconsistent with admonishing the jury that — except for the limited purpose of impeachment — it was required to consider the evidence *319 concerning each victim separately and only as that evidence related to a specific charge involving that victim. See OEC 105 (requiring limiting instruction upon request when “evidence which is admissible as to one party or for one purpose but not admissible as to another party or for another purpose is admitted” (emphasis added)). The difficulty with the arguments that the parties make about the effect, if any, of counsel’s trial strategy on petitioner’s instructional claim is that those arguments were not made either before the post-conviction court or the Court of Appeals. For that reason, the post-conviction court’s ruling in this case did not include an express or implied finding about whether counsel made a strategic choice to forego requesting a limiting instruction. That circumstance invites the question as to which party bore the burden of raising the issue of whether counsel made a reasonable tactical choice. To date, we have not expressly decided whether a post-conviction petitioner’s procedural and evidentiary burdens also include an obligation to identify, and then rebut, any possible tactical reasons that counsel may have had for a litigation action or omission. 12 Nor have the parties addressed that issue on appeal or review. As a further complication on review, the Court of Appeals’ framing of the inadequate assistance inquiry was incomplete in asking whether there “was no evident downside to petitioner from requesting an instruction *** and a significant potential benefit to him from doing so.” See Green, 261 Or App at 58 (quoting Pereida-Alba, 252 Or App at 71). Consistent with this court’s holding in Pereida-Alba, even where a cost-benefit inquiry suggests that counsel’s failure to seek a limiting instruction was deficient, the question remains whether the omission constituted inadequate *320 assistance of counsel, particularly when viewed in light of the strategy that counsel did pursue. Pereida-Alba, 356 Or at 674. Neither the post-conviction court nor the Court of Appeals had the benefit of this court’s decision in Pereida-Alba in considering petitioner’s claim. Moreover, as was the circumstance in Pereida-Alba, the post-conviction court in this case could have found from the evidence before it either that counsel did or did not consider the need for a limiting instruction. To summarize: (1) the parties did not expressly litigate before the post-conviction court the issue of the effect, if any, of counsel’s trial strategy on the adequacy of counsel’s performance in failing to request a limiting instruction; (2) this court has not expressly decided whether a post-conviction petitioner’s procedural and evidentiary burdens include anticipating the argument that counsel made a tactical decision that was inconsistent with the act or omission complained of, and the parties have not addressed that issue on appeal or review; (3) neither the post-conviction court nor the Court of Appeals had the benefit of this court’s decision in Pereida-Alba in considering the adequacy of counsel’s performance in this case; and (4) the record before us would permit the post-conviction court to draw opposing inferences as to whether trial counsel did or did not consider requesting a limiting instruction and how such an instruction would have fit with counsel’s overall trial strategy. In light of those circumstances, we conclude that the proper course of action is to reverse the decision of the Court of Appeals, reverse the post-conviction court’s judgment denying post-conviction relief with respect to petitioner’s instructional claim, and remand that claim to the post-conviction court for further consideration in light of this court’s decision in Pereida-Alba. In reaching that conclusion, we recognize that the post-conviction court did not actually decide whether counsel’s failure to request a limiting instruction constituted inadequate assistance; instead, the court concluded that petitioner failed to demonstrate prejudice. As noted, the Court of Appeals rejected the post-conviction court’s conclusion with respect to prejudice. On review, the state challenges the Court of Appeals’ analysis of prejudice, raising several reasons why, in the state’s view, *321 the Court of Appeals was wrong and the post-conviction court was right. If we were to agree with the state’s arguments in that regard, it would obviate the need for a remand. We therefore briefly address one argument that the state makes and that is appropriate for us to resolve at this juncture. The state argues that the Court of Appeals applied the wrong legal standard for prejudice. The state asserts that, “[r]ather than evaluating what likelihood a limiting instruction would have had on the jury’s verdict, the Court of Appeals evaluated what likelihood it could have had.” (Emphases in original.) The post-conviction court used the probability-based formulation of the prejudice standard for which the state advocates when it concluded that, if counsel had requested a limiting instruction, the result “would” not have been different. We conclude, however, that the post-conviction court was mistaken in doing so. As the Court of Appeals explained, to prove prejudice, petitioner had to show that his trial counsel’s deficient exercise of skill and judgment had a “tendency to affect the result of the prosecution.” Green, 261 Or App at 59. This court first used the phrase “tendency to affect the result of the prosecution” more than 30 years ago in Krummacher, wherein the court concluded that “only those acts or omissions by counsel which have a tendency to affect the result of the prosecution can be regarded as of constitutional magnitude.” 290 Or at 883. Although the court in Krummacher did not elucidate the meaning of that term, the court applied it by concluding that the petitioner’s trial counsel’s failure to object to a leading question did not have a tendency to affect the outcome of the prosecution because the omission was “inconsequential.” Id. In Stevens v. State of Oregon, 322 Or 101, 902 P2d 1137 (1995), the petitioner established that his trial counsel in a rape prosecution had failed to adequately investigate and discover witnesses who would call into question the complaining witness’s credibility. The Court of Appeals had affirmed the post-conviction court’s denial of post-conviction relief on the ground that the petitioner had not proved “substantial [] prejudice[].” Stevens v. State of Oregon, 129 Or App 533, 537, 879 P2d 893 (1994). The Court of Appeals had *322 described substantial prejudice as omissions by trial counsel that “would have affected the outcome of the case” and held that the omissions likely had not made any difference in the outcome of the prosecution. Id. On review, as it had in Krummacher, this court framed the prejudice inquiry in terms of whether the inadequate performance of the petitioner’s counsel had a tendency to affect the outcome of the trial. Stevens, 322 Or at 110. This court concluded that counsel’s errors “denied [the] petitioner highly valuable impeaching evidence from disinterested witnesses that would have called into question pivotal testimony of the complaining witness.” Id. The omitted evidence had a tendency to affect the outcome, this court concluded, because there were “no other witnesses to the alleged rape and no physical evidence of abuse or trauma,” and the prosecution “necessarily turned on the credibility of the complaining witness and of [the] petitioner.” Id. at 108-09. In Lichau, this court concluded that the petitioner was prejudiced by his trial counsel’s unreasonable failure to investigate a possible alibi defense to charges of rape, sodomy, and sexual abuse of the petitioner’s underage niece. Lichau, 333 Or at 363-64. The court noted that, if counsel had presented an alibi defense, the jury would have heard additional evidence that the petitioner was not in the state when the sexual abuse allegedly had occurred, and the jury would have been instructed that the prosecution had the burden of proving, beyond a reasonable doubt, that the petitioner was present in the state during the narrow period of time at issue. Id. at 364-65. Because the case turned primarily on the credibility of the petitioner and the victim, the court concluded that “the presentation of [the alibi] evidence could have tended to affect the jury’s consideration of the niece’s and petitioner’s version of events.” Id. at 365 (emphasis added). As the described cases indicate, where the effect of inadequate assistance of counsel on the outcome of a jury trial is at issue, it is inappropriate to use a “probability” standard for assessing prejudice. Instead, because many different factors can affect the outcome of a jury trial, in that setting, the tendency to affect the outcome standard demands more than mere possibility, but less than probability. As the *323 court stated in Lichau, the issue is whether trial counsel’s acts or omissions “could have tended to affect” the outcome of the case. Id. (emphasis added). 13 Because the post-conviction court appears to have applied the wrong legal standard for prejudice, it is inappropriate to affirm the post-conviction court’s judgment on prejudice grounds on the record before us. The decision of the Court of Appeals is reversed. The judgment of the circuit court denying post-conviction relief is reversed and remanded with respect to petitioner’s instructional claim, and is otherwise affirmed. Footnotes [1] In particular, petitioner faced trial on five counts of first-degree rape, two counts of third-degree rape, six counts of second-degree sexual abuse, three counts of third-degree sexual abuse, one count of third-degree sodomy, and one count of contributing to the sexual delinquency of a minor. [2] Five of the charges {i.e., the first-degree rape charges) involved a theory of forcible compulsion. The remaining charges all involved sexual conduct that the state alleged was non-consensual. Although for three charges (those involving KN and CO and one of the charges involving MZ) the lack of consent was based on the victims’ ages, for the remaining ten charges, the lack of consent was based on defendant’s use of “force” only in the sense of exercising physical control over the victims so that the sexual contact occurred despite their lack of actual consent to it. See generally State v. Ofodrinwa, 353 Or 507, 532, 300 P3d 154 (2013) (phrase “does not consent” in sexual abuse statute includes lack of capacity to consent due to age as well as lack of actual consent). [3] For example, counsel cross-examined victims KN, MZ, and CO to ensure that the jury heard that the encounters were consensual, even though counsel conceded — given those victims’ lack of capacity to consent due to their ages— petitioner’s guilt on those charges. [4] The other affidavit from counsel did not refer to the claim at issue. [5] OEC 404(3) provides: “Evidence of other crimes, wrongs or acts is not admissible to prove the character of a person in order to show that the person acted in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.” [6] In Kitzman, as in this case, the defendant was charged with multiple sex offenses committed against several underage victims. The Court of Appeals rejected the defendant’s argument that the post-conviction court had erred in denying his motion to sever the charges for trial, in part, because that court had mitigated the prejudicial effect that is inherent in admitting evidence of crimes against different victims “by giving defendant’s instruction that he was charged with unrelated incidents and that each charge, and the evidence pertaining to it, must be considered separately.” Kitzman, 129 Or App at 530. [7] The state did not argue before the post-conviction court or the Court of Appeals that petitioner’s propensity argument was beyond the scope of his pleaded claim or that the state was unfairly surprised by the elaboration of that argument in petitioner’s trial memorandum. [8] OES 138.620(2) provides, in part: “If the petition states a ground for relief, the court shall decide the issues raised and may receive proof by affidavits, depositions, oral testimony or other competent evidence.” [9] That danger, the court concluded, was greater due to comments that the prosecutor made during closing argument that the court believed suggested propensity purposes for the evidence. Id. at 57-58. In particular, the court emphasized the prosecutor’s statement: “What are the chances that you’re going to have nine people coming before you, six disclosing nonconsensual sexual touching, with the varied backgrounds, with the same theme of manipulation, deceit, and coercion? What are the chances?” Id. In another part of the opinion, the court referred to the prosecutor’s statements about petitioner’s “predatory nature” and about petitioner’s “patterns of sexual assault.” Id. at 54. [10] In State v. Barone, 329 Or 210, 986 P2d 5 (1999), cert den, 528 US 1086, 120 S Ct 813, 145 L Ed 2d 685 (2000), this court upheld the trial court’s decision denying the defendant’s motion to sever three homicides for which the defendant was indicted. However, the trial court in that case required that the prosecution “build a ‘fire wall’ between the three cases and to ‘present the cases totally separately.’” Id. at 216. Prosecutors in Barone made three separate opening and closing arguments; each case was presented separately. [11] The Court of Appeals similarly observed that the state had not contended before that court “that the evidence of unlawful sexual conduct pertaining to any particular victim was nonetheless admissible as to a different victim for a noncharacter purpose ‘such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.’” Green, 261 Or App at 60 (quoting OEC 404(3)). We note that this court recently held, in State v. Williams, 357 Or 1, 346 P3d 455 (2015), that, irrespective of OEC 404(3), “the admission of ‘other acts’ evidence to prove character and propensity under OEC 404(4) depends on whether the risk of unfair prejudice outweighs the probative value of the evidence under OEC 403.” Williams, 357 Or at 20. If, on a new trial in this case, the state were to offer other crimes evidence with respect to one or more victims to prove a charge involving a different victim, the trial court would be required to address any objection that defendant might make under OEC 403. We have no occasion here to predict what ruling the trial court might make if that set of circumstances were to unfold. [12] We note that, in Pereida-Alba, this court said: “When, as in this case, the state argues that the petitioner’s trial counsel made a reasonable tactical choice, that argument does not shift the burden of production and proof; rather, as petitioner acknowledges, the burden of production and proof remains with him.” Pereida-Alba, 356 Or at 662 (emphasis added). Perida-Alba thus left the burden of proof and persuasion with the petitioner when the state raises the possibility that trial counsel made a choice for tactical reasons. But Perida-Alba did not speak to what happens if the state does not raise that argument. [13] In contrast, where the prejudice inquiry focuses on a different type of outcome, for example, whether in the absence of inadequate assistance of counsel the petitioner would have entered a guilty plea, this court has described the tendency to affect the outcome test in terms of actual probability. Thus, in Moen v. Peterson, 312 Or 503, 513, 824 P2d 404 (1991), the court held that, “to prevail in this case, [the] petitioner must show by a preponderance of the evidence that, had counsel informed him of the possibility of a minimum sentence, or had he otherwise been aware of it, he would not have pleaded no contest.” (Footnote omitted; emphasis added.) See also Trujillo v. Maass, 312 Or 431, 437, 822 P2d 703 (1991) (framing prejudice inquiry in terms of whether the petitioner “would have withdrawn his plea” if counsel’s advice had been adequate).