*352**314Petitioner in this post-conviction proceeding was tried on charges of first-degree sexual abuse and first-degree sodomy in 2001. At trial, petitioner's counsel did not object to certain testimony, and controlling case law at that time from the Court of Appeals held that such testimony was admissible. In 2009, however, the Supreme Court held that that testimony was not admissible. In his post-conviction complaint, petitioner alleged that his trial counsel had failed to provide constitutionally adequate assistance and that he had been prejudiced as a result.

The Court of Appeals affirmed the post-conviction court's grant of partial summary judgment against petitioner. Jackson v. Franke , 284 Or. App. 1, 392 P.3d 328 (2017). The court assumed that counsel exercising reasonable professional skill and judgment would have objected to the testimony so as to preserve the right to seek Supreme Court review of the existing Court of Appeals case law. Even so, the Court of Appeals concluded, petitioner had not been prejudiced. The Court of Appeals decision turned on its conclusion that, as a factual matter, the chance that the Supreme Court would have allowed review in petitioner's case and ruled in his favor was too small for him to demonstrate prejudice, which requires a petitioner to demonstrate that counsel's deficiency had "a tendency to affect the result of the prosecution," a standard that we explained "demands more than mere possibility, but less than probability." Green v. Franke, 357 Or. 301, 321-22, 350 P.3d 188 (2015).

We allowed petitioner's petition for review. We address only the prejudice aspect of petitioner's claim. Although we agree with the Court of Appeals' statement of the applicable test, as outlined in Green , we disagree with its application of the test, for reasons we describe below. In our view, it is not appropriate, or workable as a matter of judicial decision-making, to speculate as to how individual members of the Supreme Court would have viewed a petition for review in petitioner's case, as the post-conviction court suggested, see Jackson, 284 Or. App. at 9, 392 P.3d 328 (quoting post-conviction court ruling). Nor is it correct to conclude, as the **315Court of Appeals did, that because of this court's "complete discretion" regarding whether or not to allow petitions for review, any assessment of the likelihood that such a petition by petitioner would have been allowed would be "nothing but speculation." Id . at 12-13, 392 P.3d 328. Rather, we hold that the test for prejudice under Green requires, in this context as in others, that petitioner show that his lawyer's deficiency had "a tendency to affect the result of the prosecution." 357 Or. at 321, 350 P.3d 188. Here, that inquiry includes an evaluation of whether a petition for review would have been allowed in petitioner's case, using available, objective criteria. As we discuss below, although it is uncertain whether this court would have allowed a petition for review from petitioner, there is "more than mere possibility." Id. at 322, 350 P.3d 188. Given that conclusion, the alleged constitutional inadequacy of his trial counsel, which blocked his appellate counsel from the opportunity to raise the issue on appeal and subsequently in a petition for review, was prejudicial. We remand for the Court of Appeals to consider other issues relating to whether, in fact, the actions of petitioner's trial counsel here fell below constitutionally required standards.

I. LEGAL BACKGROUND AND FACTS

A. Brief Overview of Legal Standards

The issue before us involves whether petitioner's counsel at his criminal trial provided *353inadequate assistance of counsel under the Oregon Constitution (Article I, section 11 ), or ineffective assistance of counsel under the United States Constitution (the Sixth Amendment, made applicable to the states by the Fourteenth Amendment). While the federal and state constitutional provisions are interpreted independently, this court has explained that they are function-ally equivalent, Montez v. Czerniak , 355 Or. 1, 6-7, 322 P.3d 487, adh'd to as modified on recons. , 355 Or. 598, 330 P.3d 595 (2014) (so explaining),1 and the elements of proving constitutionally inadequate assistance of counsel are essentially the same under both constitutions. **316Under the Oregon Constitution, a claim of constitutionally inadequate counsel requires a post-conviction petitioner to show two elements: (1) that trial counsel had failed to exercise reasonable professional skill and judgment, and (2) that that failure "had a tendency" to affect the result of the trial. See, e.g. , Green , 357 Or. at 312, 350 P.3d 188 ; Montez , 355 Or. at 7, 322 P.3d 487. The burden is on the post-conviction petitioner to show both elements. Green , 357 Or. at 312, 350 P.3d 188 ; Montez , 355 Or. at 7, 322 P.3d 487.

B. Proceedings in the Underlying Criminal Case

The post-conviction court granted the state's motion for partial summary judgment against petitioner on his claim that his trial counsel had been inadequate in failing to raise the evidentiary objection at issue here and denied petitioner's cross motion for summary judgment on that claim. The facts necessary to resolve the narrow issue before this court are undisputed, and we take them from the record and decisions below.

Petitioner was indicted in May of 2001 for first-degree sodomy and two counts of first-degree sexual abuse, based on conduct involving M, one of petitioner's three minor children. The charges arose from M's disclosures of abuse after he and his siblings had been placed in the foster home of Gillette. Gillette's mother discovered M, who was approximately 10 at the time, engaged in sexualized behavior with Gillette's son. When Gillette questioned M about it, M told her that petitioner had made M touch him and had sodomized him. Gillette reported the matter to police.

Petitioner was represented at trial by Jonasson. While Jonasson had been appointed only three weeks before the scheduled trial date, petitioner refused to permit Jonasson to seek an extension of time. Petitioner waived the right to a jury, and the case was tried to the court.

One of the witnesses for the state at the trial was Dr. Steinberg, a pediatrician with CARES Northwest who specializes in (among other things) child abuse, and who had examined M. Steinberg testified that she had found no physical evidence of abuse. However, based on Steinberg's interviews with M and her review of M's medical, social, and behavioral history, Steinberg stated that her "medical **317diagnosis" "was highly concerning for sexual abuse." Jonasson did not object to Steinberg's testimony regarding her medical diagnosis. Jonasson knew that Steinberg's testimony "would carry great weight with the factfinder," but he "could not think of any possible objection to her diagnosis."

At that time, the Court of Appeals had issued a plurality opinion holding that a medical diagnosis of sexual abuse was admissible, without suggesting that physical evidence was a necessary precondition. State v. Trager , 158 Or. App. 399, 974 P.2d 750, rev. den. , 329 Or. 358, 994 P.2d 125 (1999). In Trager , the Court of Appeals had decided the case en banc. While all the judges had agreed the evidence was admissible, they had divided over whether the evidence was "scientific evidence" that required a particular foundation.2 This court later denied review *354in Trager , but one justice would have allowed review. Trager , 329 Or. 358, 994 P.2d 125 (Durham, J., would allow).

Jonasson did not know about Trager . Thus, he also was not aware that the Trager court had split en banc, or that, when the defendant in that case petitioned for review, one justice of the Supreme Court had voted to allow review. Nor was he aware that attorneys at the Office of Public Defense Services believed that Trager was wrongly decided and continued to encourage objections to such medical diagnosis testimony in an effort to obtain Supreme Court review in a future case.

The trial court convicted petitioner of first-degree sodomy, but it acquitted him of the other two counts. Petitioner ultimately was sentenced to 130 months in prison.3

**318Petitioner appealed his conviction. His appellate counsel, Allen, did not raise any issue regarding Steinberg's testimony; she concluded that Jonasson's failure to object meant that the issue had not been preserved. The Court of Appeals affirmed without opinion, and this court denied review. See State v. Jackson , 208 Or. App. 757, 145 P.3d 1145 (2006), rev. den. , 342 Or. 473, 155 P.3d 51 (2007).

Two and a half years after this court denied petitioner's petition for review, the Supreme Court effectively overruled Trager . In State v. Southard , 347 Or. 127, 218 P.3d 104 (2009), this court concluded that a medical diagnosis of sexual abuse such as that at issue in this case was not admissible. As an initial matter, the court held that the medical diagnosis did meet the standards required for admission as scientific evidence. Id . at 138-39, 218 P.3d 104. Nevertheless, the court explained that such a diagnosis should be excluded under OEC 403, because the probative value of the diagnosis, when it is not supported by any physical evidence of abuse, is substantially outweighed by the danger of unfair prejudice. Id . at 142, 218 P.3d 104 (summarizing court's "narrow" holding and noting limitations).

C. Post-Conviction Proceedings

Petitioner sought post-conviction relief on a number of grounds, including that Jonasson had been constitutionally inadequate as trial counsel for failing to object to Steinberg's medical diagnosis of sexual abuse.

The superintendent moved for partial summary judgment, arguing that petitioner had failed to establish either of the requirements for post-conviction relief. First, he asserted that petitioner's counsel was not inadequate for failing to object to Steinberg's testimony, because at the time of petitioner's trial in 2001, the Court of Appeals already had held that a medical diagnosis of sexual abuse, even in the absence of corroborating physical evidence, was admissible. Southard would not be decided for another eight years. Thus, the superintendent asserted, Jonasson's failure to object was not a failure to exercise reasonable professional skill and judgment. Second, the superintendent also argued that petitioner was unable to establish prejudice. Because **319the case law at the time of trial was contrary to petitioner's position, the superintendent argued, as a matter of law petitioner could not demonstrate that, even if his counsel had raised the issue at trial, it would have had a tendency to affect the result in petitioner's case.

Petitioner opposed the motion for partial summary judgment, and also filed a cross-motion for partial summary judgment, arguing that he was entitled to judgment as a matter of law as to the inadequacy of his counsel and the prejudice that he had suffered. He offered affidavits from criminal defense attorneys to support his contention that an attorney exercising reasonable professional skill and judgment would have objected at the time, despite the Court of Appeals *355decision in Trager , although the post-conviction court struck several of petitioner's affidavits on the superintendent's motion.4 Petitioner also argued that, because this court had allowed review in Southard and reversed the evidentiary ruling exemplified by Trager and other cases, he had, as a matter of law, shown that his counsel's deficiency had "tended to affect the result" in his case.

The post-conviction court granted the superintendent's motion for partial summary judgment and denied petitioner's motion. The court concluded that petitioner could not meet either of the elements necessary to show constitutionally inadequate assistance of counsel. First, the court held that a lawyer's failure to anticipate the results in Southard did not fall below the standard of reasonable professional skill or judgment. Second, the court concluded that Jonasson's failure to make the argument did not prejudice petitioner, because petitioner did not offer evidence about the decision making of the individual justices of this court:

"In the absence of an affidavit from an Oregon Supreme Court Justice confirming that petitioner's case would have been the vehicle for the principle of law announced in Southard , petitioner's claims in this regard are unsubstantiated speculation."

**320Subsequently, the matter went to trial on petitioner's other allegations. The court ultimately denied petitioner post-conviction relief on any of the grounds asserted and entered a general judgment for the superintendent.

D. Appeal to Court of Appeals

Petitioner appealed. Among other things, he con-tended that the post-conviction court had erred in granting the superintendent's motion for partial summary judgment and in denying his cross-motion for partial summary judgment regarding Jonasson's failure to object to the medical diagnosis testimony. Petitioner also appealed the post-conviction court's ruling on the motion in limine that struck four of his affidavits.

The Court of Appeals affirmed. As to the two elements required to show constitutionally inadequate assistance of counsel, the court only considered the second one: whether petitioner had been prejudiced. The court concluded that petitioner had shown only a "mere possibility" that the Supreme Court would have allowed review if Jonasson had made a Southard -style objection at trial. Jackson, 284 Or. App. at 12, 392 P.3d 328.

In so holding, the Court of Appeals focused on the counterfactual probability of whether this court would have allowed review in petitioner's criminal case, if there had been an appropriate objection at trial, but the trial court had overruled the objection and the Court of Appeals, following its decision in Trager , had affirmed. Noting that the Supreme Court "has * * * complete discretion whether to allow review" of a Court of Appeals decision, id . at 12, 392 P.3d 328, the court explained that there was "little to no evidence" that "the Supreme Court might have granted review of his case" if Jonasson had objected, id . at 11, 392 P.3d 328. The affidavit of petitioner's appellate counsel, Allen, had not identified "any attributes of petitioner's case that might have made it a particularly good candidate for Supreme Court review." Id . at 12, 392 P.3d 328. "Absent any indication that the Supreme Court was seeking a case with attributes like petitioner's to use as a vehicle for addressing the Southard issue, nothing but speculation supports petitioner's contention that the court might have granted review in his case." Id . at 13, 392 P.3d 328.

**321Given the court's holding that petitioner had not been prejudiced, the court found it unnecessary to decide whether Jonasson's failure to object fell below the reasonable exercise of professional skill and judgment. See id . at 9-10, 392 P.3d 328.

*356II. DISCUSSION

We begin by noting that the only issue before us is whether the post-conviction court erred in holding, as a matter of law, that trial counsel, in failing to seek exclusion of the medical diagnosis testimony, caused petitioner prejudice. The trial court had ruled against defendant on both elements of the claim that counsel was constitutionally inadequate. See, e.g. , Green , 357 Or. at 312, 350 P.3d 188 (discussing the two elements of inadequate assistance of counsel claims). The Court of Appeals, however, reached its decision based entirely on the conclusion that the superintendent was entitled to summary judgment on the prejudice element. By letter to the parties, we asked them to focus their briefing on that element. Accordingly, for purposes of this opinion we will assume-without deciding-that an attorney exercising reasonable professional skill and judgment would have objected to the medical diagnosis, despite the existing Court of Appeals decision in Trager .5

We begin with a brief review of the standard for determining prejudice in the context of constitutionally inadequate assistance of counsel. This court generally has phrased the prejudice standard as whether counsel's error had a "tendency to affect the result of the prosecution." See Green , 357 Or. at 321, 350 P.3d 188 (tracing phrase to Krummacher v. Gierloff , 290 Or. 867, 883, 627 P.2d 458 (1981) ). We have explained that, at least where counsel's error occurs in a jury **322trial, "the tendency to affect the outcome standard demands more than mere possibility, but less than probability." Id . at 322, 350 P.3d 188.6 But courts ordinarily apply the same standard to bench trials. See Kimmelman v. Morrison, 477 U.S. 365, 106 S.Ct. 2574, 91 L.Ed.2d 305 (1986) ; Maxfield v. Nooth, 278 Or. App. 684, 688, 377 P.3d 650 (2016).7

With those observations, we return to the facts and procedural posture of this case. The question is whether petitioner was prejudiced-that is, whether the failure of petitioner's trial counsel to object to that testimony had a tendency to affect the result of the prosecution.

We conclude that it did. This case involved allegations of sexual abuse unsupported by physical evidence, and it was tried in 2001. The only witnesses to the alleged crime were M and petitioner. Under those circumstances, although some parts of Steinberg's testimony were admissible, her "medical diagnosis" of M as being "highly concerning for sexual abuse" offered little that was of probative value, while carrying with it a "substantial *357risk" that the factfinder would be "overly impressed or prejudiced" by "a credentialed expert, surrounded with the hallmarks of the scientific method." Southard , 347 Or. at 140-41, 218 P.3d 104. In our view, the admission of that evidence created more than a mere possibility that such evidence influenced the determination of guilt. See **323Stevens v. State of Oregon , 322 Or. 101, 110, 902 P.2d 1137 (1995) (finding prejudice because "trial counsel's inadequate performance denied petitioner highly valuable impeaching evidence from disinterested witnesses that would have called into question pivotal testimony of the complaining witness").

But that conclusion does not end our inquiry, even as to the prejudice prong, as petitioner recognizes. Rather than argue for a general rule that a 2009 decision from this court should be applied retroactively to his 2001 trial, petitioner instead asserts that he can establish prejudice by demonstrating that his counsel's deficiency had a "tendency to affect the result" in his case. Petitioner acknowledges that if his trial counsel had objected to Steinberg's testimony when the case was tried, the objection likely would have been overruled by the trial court, based on Trager. And that ruling likely would have been affirmed by the Court of Appeals when it decided his case in 2006. Thus, petitioner cannot demonstrate that, had the proper objection been made, the outcome after trial or before the Court of Appeals would have been different. To prevail, then, petitioner must show that there is "more than mere possibility," Green , 357 Or. at 322, 350 P.3d 188, that if the objection had been properly made his appellate counsel would have taken up the issue and ultimately presented it to this court in a petition for review, and that this court would have allowed review and reversed petitioner's conviction.

The superintendent asserts that this theory of prejudice is not cognizable at all, either under Article I, section 11, or the Sixth Amendment. The superintendent argues that determining prejudice based on present case law would lead to incongruous results. He notes that petitioner, whose counsel did not make an objection, would obtain a benefit that was denied to other criminal defendants whose counsel had made an unsuccessful objection prior to Southard . He identifies several petitions for review that he asserts had raised a Southard -style issue, but which this court had denied.

We are not persuaded. The superintendent focuses on what petitioner seeks to gain-the application of Southard to his case, even though Southard was decided only after petitioner's appeal from his conviction had concluded. But **324the issue before us is prejudice, and, in that context, the more relevant consideration is what petitioner lost . And what he lost, because of his trial counsel's failure, was the opportunity to petition this court for review of his criminal case and to have this court apply to his case in the first instance the rule later announced in Southard -the correct interpretation of the evidentiary rules that applied at the time of petitioner's trial.8 If petitioner can show that that lost opportunity had a tendency to affect the outcome of his case, then petitioner will be able to show prejudice.

The superintendent also argues that, when considering the adequacy of trial counsel, we should only look to potential effects on the result of the trial, not any potential appeal. Constitutional inadequacy is not limited to trial counsel, however: We have recognized that appellate counsel also can be constitutionally inadequate. See Guinn v. Cupp , 304 Or. 488, 496, 747 P.2d 984 (1987). The superintendent's argument thus asks us to hold that, while trial counsel can be constitutionally inadequate at trial and appellate counsel can be constitutionally inadequate on appeal, there is no remedy for a trial counsel whose deficiencies prevent appellate counsel from presenting a meritorious issue on appeal.

*358We decline to engage in such analytical hair-splitting. Because of our rules governing preservation, and for other reasons besides, the tasks of trial and appellate counsel are interdependent. Here, trial counsel's alleged error in failing to object to the disputed testimony meant that the error was not preserved, and therefore competent appellate counsel was unlikely to be able to raise it successfully on appeal or in a petition for review. Petitioner's argument is that his trial counsel's error denied him the opportunity to have the merits of his evidentiary objection even considered by those courts and that he can establish that, had he been given the opportunity, there was a tendency that it would **325have affected the result in his case. He acknowledges that, given controlling case law at the time, had his objection been raised, in all likelihood it would have been rejected by the trial court and the Court of Appeals. He argues, however, that tensions in the existing case law-and the fact that this court decided the issue in his favor in Southard -indicate that this court might well have allowed review in his case. Thus, he asserts, he has demonstrated, in the words of Strickland v. Washington , 466 U.S. 668, 694, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different."

To prevail, therefore, petitioner must show that there was a "tendency"-"more than mere possibility, but less than probability," Green , 357 Or. at 322, 350 P.3d 188 -that three events would have occurred in conjunction. First, that reasonable appellate counsel would have raised the issue that petitioner's trial counsel had preserved. Second, that this court would have allowed review in petitioner's case. And, third, that this court would have ruled in petitioner's favor and reversed his conviction. In this case, the first and third parts of that chain are not in serious dispute. Petitioner has presented uncontradicted affidavits that his appellate counsel would have raised the issue and properly presented it to this court in a petition for review. And, given this court's ultimate holding in Southard , there is little doubt that this court would have found the challenged evidence inadmissible had we decided petitioner's case. Moreover, for the reasons discussed above, the evidence that petitioner argues that his trial counsel should have objected to was significant in his case, such that its admission was not harmless error, and reversal would have been appropriate. The key dispute in this case, however, is whether this court would have allowed review in petitioner's case at all.

The Court of Appeals accurately observed that this court has complete discretion as to whether to allow review of any particular case, and parties can never be certain whether or not review will be allowed in a particular case. After all, this court's decisions on whether to allow or to deny a petition are discretionary, generally unexplained, and rest on a multiplicity of factors. See ORAP 9.07 (identifying 16 **326nonexclusive criteria that Supreme Court "considers * * * to be relevant to the decision whether to grant discretionary review") but certainty is not required. The test, as explained in Green , is whether a petitioner can show that his counsel's error had a "tendency" to affect the outcome of his or her case, which is quite different. And the absence of certainty does not render any evaluation of a tendency to affect the outcome purely speculative.

Although estimating probabilities as to how courts will decide legal issues is difficult, in some cases, including this one, our "tendency to affect the result" standard for determining prejudice requires us to do so. The superintendent argues that this approach is inappropriate and that it invites the post-conviction court to inquire into the identities, past and present, of individual members of the Court of Appeals and this court, as well as the perceived views of those members on the legal issues in dispute or factual aspects of a case to which individual judges might be sympathetic. That approach could lead the post-conviction court to take evidence on the confidential, internal actions within the appellate courts. We see precisely those potential pitfalls in the post-conviction court's suggestion that a petitioner should offer affidavits of Supreme Court justices regarding whether the court might have allowed review. By doing so, the post-conviction court expressly invited the parties to *359put into issue the individual membership of this court (and, potentially, of the Court of Appeals) and the individual thought processes of those judges.

Strickland , however, makes it clear that the prejudice inquiry should not depend on a decisionmaker's factual identity or personal characteristics. As the Supreme Court stated:

"[The assessment of prejudice] should not depend on the idiosyncrasies of the particular decisionmaker, such as unusual propensities toward harshness or leniency. Although these factors may actually have entered into counsel's selection of strategies and, to that limited extent, may thus affect the performance inquiry, they are irrelevant to the prejudice inquiry. Thus, evidence about the actual process of decision, if not part of the record of the proceeding under review, and evidence about, for example, **327a particular judge's sentencing practices, should not be considered in the prejudice determination."

466 U.S. at 695, 104 S.Ct. 2052. That is no less true under Green . As a matter of law, then, the prejudice analysis should not involve any consideration of the individual justices, or the factors that might or might not have motivated specific judges in deciding whether to allow review. For that reason, the post-conviction court's suggestion that a post-conviction petitioner should obtain affidavits from justices of this court was misplaced. Not only is such evidence unnecessary, but it is also irrelevant. The private views of specific members of this court are of no weight in assessing prejudice.

Whether review would have been allowed, for purposes of the prejudice inquiry, must be determined based on objective indicators that allow meaningful assessments of the likelihood of a petition for review being allowed. Assessing how this court would have decided the issue in dispute, however, will typically be easier than determining in retrospect whether review would have been allowed. That is because, in evaluating the prejudice prong, we have one tremendous advantage: hindsight.9 Perhaps the most salient indicator-and one present in this case, as we discuss below-is whether, at or around the time that the petition at issue was before this court, this court allowed review of a petition raising the same issue and ultimately issued an opinion favorable to the arguments of the person now seeking post-conviction relief. But other indicators that could be considered in assessing whether a petition for review might have been allowed include whether the issue **328is a significant one, whether there is conflict on the legal issue in question-e.g. , inconsistent Court of Appeals decisions or dissenting or concurring opinions in that court, or conflict between decisions of the Court of Appeals and of this court-and whether this court has denied review of petitions presenting the same issue in the past. Similarly, conflicts or inconsistencies between recent United States Supreme Court decisions and Oregon cases might indicate that a petition for review focused on such a decision would be a likely candidate for review. Also relevant, when taking into account past denials of review, is whether the denied petitions presented the issue clearly and cleanly, and whether any justices would have allowed review. A justice's public vote to allow a petition, as distinct from that justice's private views or inclinations, is an indication that the issue may be one that this court would be interested in deciding in a future case. And, as our rules make clear, this court also considers, in ruling on petitions *360for review, whether a Court of Appeals decision appears wrong and, if so, whether the error "results in a serious or irreversible injustice." ORAP 9.07(14)(a).10

Of course, in assessing whether this court likely would have allowed review of a particular hypothetical petition, it would be difficult to imagine stronger evidence than that the court did, in fact, allow a petition raising the identical issue at or around the same time the hypothetical petition would have been considered. Conversely, if this court never allowed review of such a case, or only did so long after a post-conviction petitioner's petition for review would have been filed or in response to unforeseen subsequent developments in the law, that would indicate that review likely would not have been allowed in the petitioner's case.

**329In this case, petitioner argues that, had the evidentiary issue been preserved, there is enough evidence to satisfy his burden of showing, to the degree required by the Green "tendency" standard, that there is more than a mere possibility that this court would have allowed the petition for review in his criminal case. Although assessing counterfactual possibilities always involves some degree of uncertainty, several objective indicia support petitioner's contention.

First, at the time petitioner was tried, the controlling Court of Appeals case on the admissibility in a sexual abuse prosecution of a medical diagnosis that the alleged victim had been sexually abused (without discussing any need for physical evidence of abuse) was Trager. Although the Supreme Court had denied the petition for review of Trager , as we noted previously, one justice had voted to allow review. See Trager, 329 Or. 358, 994 P.2d 125 (Durham, J., would allow).

Second, although Trager may have been controlling, there was tension between that decision and several decisions from this court holding that medical experts were not permitted to vouch for a person who asserted that the defendant had sexually abused them. In State v. Milbradt , 305 Or. 621, 756 P.2d 620 (1988), this court had strongly rejected the testimony of a psychologist who asserted that two alleged victims of sexual abuse were not "deceptive" and that their statements "represented their experience." In this very similar context, the court stated:

"We have said before, and we will say it again, but this time with emphasis-we really mean it-no psychotherapist may render an opinion on whether a witness is credible in any trial conducted in this state."

305 Or. at 629, 756 P.2d 620 (emphasis in original). See also State v. Middleton , 294 Or. 427, 438, 657 P.2d 1215 (1983) ("We expressly hold that in Oregon a witness, expert or other-wise, may not give an opinion on whether he believes a witness is telling the truth."). In light of the existing cases, which highlight both the importance of the issue and the tension between this court's decisions and Trager , this court's decision in 2008 to allow review in Southard was not unexpected.

**330Yet the most important piece of evidence is Southard itself, in which review was allowed on the same issue that petitioner argues that his trial counsel should have preserved. Petitioner's petition for review in his appeal from his conviction was filed on November 22, 2006, and was denied on March 7, 2007. State v. Jackson , 342 Or. 473, 155 P.3d 51 (2007) (denying review). The petition for review in Southard was filed on November 14, 2007, and review was allowed on April 16, 2008. 344 Or. 401, 182 P.3d 200 *361(allowing review). This court allowed the petition in Southard just over a year after petitioner's petition for review was denied.

On the other side of the scale, as the state notes, there are several orders from this court between 2002 and 2008 denying review in cases that raised a similar issue. While one could debate how closely some of those cases resemble this one, the court clearly had at least one or two other opportunities to take the "medical diagnosis of sexual abuse" issue before Southard and declined to do so.11 That evidence weighs against the likelihood of this court allowing review in petitioner's case. However, two similar petitions in which review was denied do not scupper petitioner's case, given that he need not prove to a certainty that review would have been allowed, or even that review was more likely than not. Petitioner needs only to show a "tendency" that the result in his case would have been affected. Similar petitions for review that were denied are relevant pieces of evidence but are not dispositive-adding up the similar-but-denied petitions cannot substitute for a full consideration of the objective indicators bearing on the question of whether review would have been allowed. Here, the ongoing tension between Trager , Milbradt , and Middleton , and, more importantly, the court's eventual decision to allow review and then to reverse the Court of Appeals in Southard , indicate **331that it is not correct to say, as the Court of Appeals did, that "nothing but speculation supports petitioner's contention that the court might have granted review in his case." Jackson, 284 Or. App. at 13, 392 P.3d 328. No one could have predicted with any certainty whether this court would have allowed review in petitioner's case, but we agree with petitioner that there was more than a mere possibility that, if the issue had been preserved and adequately presented, this court would have allowed review and reversed his conviction.

For the foregoing reasons, we conclude that the Court of Appeals erred in affirming the post-conviction court's entry of partial summary judgment in favor of the superintendent on the prejudice prong of petitioner's claim that his trial counsel was constitutionally deficient in not objecting to the medical diagnosis of sexual abuse. That issue turned on facts that are essentially undisputed, and we conclude that petitioner's motion for partial summary judgment on that claim should have been granted as to the prejudice prong.

Having resolved the narrow issue presented in this case, we turn to the proper disposition on review. Although we have concluded that the Court of Appeals erred on the issue that it decided, there are other relevant assignments of error that the Court of Appeals did not reach, including whether petitioner's trial counsel's failure to object fell below constitutionally required standards and whether the post-conviction court erred in excluding affidavits potentially relevant to that question. Accordingly, we reverse and remand for the Court of Appeals to consider those additional questions.

The decision of the Court of Appeals is reversed, and the matter is remanded to the Court of Appeals for further proceedings consistent with this opinion.