The state appeals a judgment granting petitioner post-conviction relief and setting aside his convictions for four counts of first-degree sexual abuse, ORS 163.427, and two counts of endangering the welfare of a minor, ORS 163.575. The state argues that the post-conviction court erroneously concluded that trial counsel’s failure to object to testimony regarding “treatment recommendations” rendered his performance constitutionally inadequate. Further, the state argues that, in any event, petitioner failed to establish that counsel’s failure to object to the testimony prejudiced petitioner. We agree with the state that the post-conviction court erred and, accordingly, reverse and remand.1
We state the facts consistently with the findings of the post-conviction court, which are binding on us if there is evidence to support them. Derschon v. Belleque, 252 Or App 465, 466, 287 P3d 1189 (2012), rev den, 353 Or 208 (2013). At the time that the abuse underlying petitioner’s convictions came to light, petitioner shared physical custody of his two children, seven-year-old C and her older sibling, N, with their mother, petitioner’s ex-wife. The dissolution of petitioner and his ex-wife’s marriage had been contentious, and petitioner’s relationship with his ex-wife, who had primary custody of both children, was strained. While driving the children to his ex-wife’s house at the conclusion of a visit, petitioner became upset when one of the children expressed a desire not to live with petitioner. During the outburst that followed, petitioner threatened to kill his ex-wife and punched the rearview mirror of his car, breaking it and cutting his hand. Petitioner’s conduct upset the children, and, when they arrived at their mother’s house, the children told her and their stepfather what petitioner had said and done.
Thereafter, petitioner’s ex-wife contacted Haines, an officer in the Portland Police Bureau’s Domestic Violence Reduction Unit. Haines conducted a series of individual interviews with petitioner’s ex-wife and the children, and, although initially focused on petitioner’s threatening behavior, the interviews revealed that petitioner had engaged in *322unconventional conduct around the children. Specifically, petitioner’s ex-wife said that petitioner chose to be nude in his home in front of the children and that he sometimes “cuddled” with C while nude. C confirmed petitioner’s behavior and also indicated that petitioner had touched her vaginal area. That prompted Haines to stop the interview, consult with a detective from the bureau’s Child Abuse Team, and schedule a sexual-abuse evaluation for C with CARES Northwest, a child-abuse-assessment center.
At CARES, C underwent a medical examination, performed by a staff physician, Dr. Bays, and an interview with Findley, a staff social worker and interviewer. During the interview with Findley, C again disclosed that petitioner had cuddled with her while he was nude and that, while cuddling, petitioner had touched her chest and vaginal area. She also said that she had felt petitioner’s erect penis on her leg. After the evaluation, Bays and Findley prepared a report containing multiple treatment recommendations.
During an interview with a detective regarding his conduct with C, petitioner fully admitted that he was often nude in his home in front of the children. He also acknowledged that he sometimes “cuddled” with and “caressed” C while he was nude. Notwithstanding those admissions, petitioner adamantly denied touching C in any sexual way, although he conceded that he may have inadvertently touched C’s vaginal area.
A grand jury subsequently indicted petitioner on eight counts of first-degree sexual abuse and two counts of endangering the welfare of a minor. Petitioner pleaded not guilty and proceeded to a jury trial. At trial, as pertinent to this appeal, the state called Findley, who offered the following testimony about the CARES treatment recommendations:
“We recommended [C] not to have any direct contact with [petitioner] at that time.
* * * *
“We also recommended that [petitioner] have a full psychological evaluation with attention paid to anger management and a sex offender evaluation.
* ‡ ‡ *
*323“We recommended further investigation by *** law enforcement into the allegations of abuse. We also recommended [for the] need to evaluate [the] safety of [N] who has had ongoing contact with [petitioner].”
Petitioner’s trial counsel did not object to that portion of Findley’s testimony.
The jury found petitioner guilty of four counts of first-degree sexual abuse and both counts of endangering the welfare of a minor. Petitioner appealed, and we affirmed his convictions. State v. Logan, 203 Or App 639,129 P3d 281, rev den, 340 Or 359 (2006). He then sought post-conviction relief, alleging 58 claims of inadequate assistance of trial counsel.
At issue on appeal is petitioner’s claim that trial counsel performed inadequately by failing to object to Findley’s testimony about the CARES treatment recommendations. Specifically, identifying the excerpts of Findley’s testimony set out above, petitioner alleged that those recommendations “left [the jury] with the impression that [C’s] allegations must be true,” because “a professional was making recommendations which supported that finding.” Apparently due to the number of claims raised by petitioner, Findley’s testimony about the CARES treatment recommendations received little attention from petitioner or the state at the post-conviction hearing, and neither party addressed it in the party’s trial memorandum.
The state did present an affidavit in which petitioner’s trial counsel responded to petitioner’s allegations. Although the affidavit did not directly address the CARES treatment recommendations, it explained trial counsel’s general approach to evidentiary objections at trial:
“As a criminal defense lawyer, although there was always a balance to be maintained so that jurors would not perceive that I was disinterested or uninvested in my clients’ innocence, my general approach and belief about objections at jury trials was conservative. I believed, and continue to believe, that jurors often particularly suspect that a criminal defense lawyer making constant objections, whether upheld or not, is trying to keep information from them, and that promoting this sort of feeling can lead them to link their own distaste for such methods to defendant.
*324“Based on this general belief, I did not object whenever possible, but only when I was confident that there was an evidentiary violation and that it would be harmful and important in the context of my overall trial strategy.”
(Emphasis in original.) Trial counsel explained, both in his affidavit and through testimony at the post-conviction hearing, that his trial strategy had been to portray petitioner as an individual who exercised poor judgment and to suggest that, in light of the contentious nature of petitioner’s relationship with his ex-wife, C may have misperceived or been motivated to distort petitioner’s unconventional, but innocent, behavior toward her.
The only discussion of trial counsel’s conduct regarding the CARES treatment recommendations at the post-conviction hearing occurred during the examination of Cohen, a defense attorney whom petitioner had called as an expert witness. When asked whether he found fault in trial counsel’s failure to object to the admission of the treatment recommendations, Cohen answered, “yes,” but explained that he had “mixed feelings” as to the nature of the fault. After acknowledging that “the rules are different for CARES” and that “the case law in Oregon! is] pretty broad [in] allowing [CARES testimony] to come in,” the expert explained that he thought that “you still have to make your record.”2 Cohen then opined that the testimony could have been impeached through a defense witness; that counsel should have objected to it on the basis that Findley “lack[ed] the specific expertise” to make recommendations regarding petitioner and petitioner’s other child, neither of whom Findley had met; and that the treatment recommendations were ultimately “another way *** of backing up [C’s] allegation [s].”
*325At the close of the post-conviction hearing, petitioner and the state each submitted written closing arguments. Petitioner addressed the CARES treatment recommendations in his written closing arguments, but only to restate the basic allegation in his post-conviction petition.3 The state did not address the treatment recommendations directly. The post-conviction court subsequently issued a written order that, after rejecting each of petitioner’s other claims, granted petitioner a new trial on the ground that trial counsel’s failure to object to the CARES treatment recommendations amounted to constitutionally deficient assistance of counsel under both the state and federal constitutions.
In doing so, the post-conviction court reasoned, among other things, that the testimony about the CARES treatment recommendations went “beyond relaying [C’s] statements or even discussing [C’s] consistency in making the statements and into recommendations for treatment as if [C’s] statements were true.” It also reasoned that, “[w]hile not a direct comment on honesty, [the recommendation that petitioner be evaluated] amounts to an indirect comment on credibility from an expert in the field of child sexual abuse investigation.”4
The post-conviction court then went on to determine whether trial counsel’s failure to object to Findley’s testimony regarding the treatment recommendations had prejudiced petitioner. Focusing solely on the vouching aspect *326of Findley’s testimony — that is, that it constituted an impermissible comment on C’s credibility — the post-conviction court concluded that it had. After examining existing case law, the court noted that Findley’s testimony was “similar to opinion evidence that a victim is diagnosed with [having suffered] sexual abuse” and that, under State v. Wilson, 121 Or App 460, 855 P2d 657 (1993), that type of diagnosis did not amount to impermissible vouching. Nevertheless, the court reasoned, Findley’s testimony went “one step further” by recommending that petitioner receive evaluations. According to the post-conviction court, that extra step “encroache[d] into an area where the jury’s discretion bee [ame] supplanted.” On that basis, the post-conviction court granted petitioner a new trial.5 The state filed written objections to the post-conviction court’s order, which were denied. A judgment granting petitioner post-conviction relief followed.
The state contends on appeal that, at the time of petitioner’s trial, a reasonable trial attorney could have concluded that an objection to Findley’s testimony about the CARES treatment recommendations would have failed under Wilson and, hence, was not an appropriate objection to make. Second, it contends that trial counsel’s decision not to object to that portion of Findley’s testimony was strategic and supported trial counsel’s defense strategy. Finally, it contends that counsel’s failure to object to admission of the treatment recommendations did not prejudice petitioner because the evidence was admissible under Wilson. Petitioner responds that the treatment recommendations were inadmissible notwithstanding Wilson because they amounted to a comment on the credibility of petitioner, not merely C.
*327A claim of inadequate assistance of counsel under Article I, section 11, of the Oregon Constitution requires a petitioner to prove, by a preponderance of the evidence, (1) that trial counsel failed to exercise reasonable professional skill and judgment and (2) that the petitioner suffered prejudice as a result — that is, that trial counsel’s failure had a tendency to affect the result of the prosecution. Hayward v. Belleque, 248 Or App 141, 147-48, 273 P3d 926 (2012), rev den, 353 Or 208 (2013) (citing Trujillo v. Maass, 312 Or 431, 435, 822 P2d 703 (1991)). Under the Sixth Amendment to the United States Constitution, a petitioner must make a functionally equivalent showing. Montez v. Czerniak, 237 Or App 276, 278 n 1, 239 P3d 1023 (2010), rev den, 350 Or 571 (2011). Specifically, the petitioner must show that trial counsel’s performance “fell below an objective standard of reasonableness” and that there is a “reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland v. Washington, 466 US 668, 688, 694, 104 S Ct 2052, 80 L Ed 2d 674 (1984).
Because we conclude it to be dispositive, we turn to the issue of prejudice. Whether a petitioner has demonstrated prejudice is a legal question that may depend on the post-conviction court’s findings. Wyatt v. Czerniak, 223 Or App 307, 311, 195 P3d 912 (2008). We thus review the post-conviction court’s ruling on prejudice for legal error and for whether the facts found by the post-conviction court are supported by the record. Derschon, 252 Or App at 466. We are bound by the post-conviction court’s findings if there is evidence in the record to support them. Id.
To establish prejudice on a claim based on a trial counsel’s failure to object to the admission of evidence, a petitioner must establish that the objection would have been well taken when the criminal case was tried. Hayward, 248 Or App at 150 (citing Peiffer v. Hoyt, 339 Or 649, 660, 125 P3d 734 (2005)). The petitioner must then establish that, given the totality of the circumstances, the admission of the objectionable evidence had a tendency to affect the jury’s verdict. Cunningham v. Thompson, 188 Or App 289, 296, 71 P3d 110 (2003), rev den, 337 Or 327 (2004).
*328With those standards in mind, we turn to the law in effect at the time of petitioner’s criminal trial in 2001. At that time, the Oregon Supreme Court had repeatedly made it clear that, ‘“in Oregon, a witness, expert or otherwise, may not give an opinion on whether he believes [another] witness is telling the truth.’” State v. Keller, 315 Or 273, 284, 844 P2d 195 (1993) (quoting State v. Middleton, 294 Or 427, 438, 657 P2d 1215 (1983)); see also, e.g., State v. Milbradt, 305 Or 621, 629-30, 756 P2d 620 (1988) (same); State v. Brown, 297 Or 404, 444, 687 P2d 751 (1984) (same). However, those cases did not stand for the proposition that all evidence tending to bolster the credibility of a witness is inadmissible.6
The Oregon Supreme Court articulated in Middleton the distinction between inadmissible vouching as to the credibility of a witness and admissible evidence that merely tends to bolster the credibility of a witness. There, the defendant was convicted of the first-degree rape of his 14-year-old daughter. After describing the rape consistently on a number of occasions, the child recanted her allegations in a written statement. At trial, the child testified that the defendant had raped her, and the defendant impeached her testimony with her written recantation. Later in the trial, over the defendant’s objections, two social workers who worked with sexually abused children testified that the child’s behavior was consistent with typical reactions of child victims of intrafamilial sexual abuse.
On review, the Supreme Court rejected the defendant’s argument that the evidence amounted to an impermissible comment on the child’s credibility. The Supreme Court reasoned:
“Defendant contends that the evidence given by the experts here was a direct effort to support the credibility of the complaining witness. It is true that[,] if the jurors believed the experts’ testimony, they would be more likely to believe the victim’s account. Neither of the experts *329directly expressed an opinion on the truth of the victim’s testimony. Much expert testimony will tend to show that another witness either is or is not telling the truth. This, by itself, will not render evidence inadmissible.”
Middleton, 294 Or at 435 (internal citation omitted).
Thus, as conceived in Middleton, the issue reduced to whether the testimony at issue “directly expressed an opinion on the truth” of another witness’s statement or merely “tend[ed] to show that another witness either is or is not telling the truth.” Id. However, a witness’s opinion on the truth of another witness’s statement need not be explicit in order to be direct. Hence, whether a statement falls into the former category or the latter is not always obvious. But an examination of the court’s treatment of the issue in two subsequent cases, Milbradt and Keller, and of our application of that line of cases in Wilson, helps clarify the distinction.
In Milbradt, the defendant was convicted of a number of sexual offenses against two young women who had cognitive impairments. On review, the Supreme Court concluded that a psychologist’s testimony — specifically that he saw no evidence of deception on the part of the victims and that what the victims had reported represented their experiences — constituted improper vouching. Quoting Middleton, the court explained:
“‘We expressly hold that in Oregon a witness, expert or otherwise, may not give an opinion on whether he believes a witness is telling the truth.’ An opinion that a person is not deceptive, could not lie without being tripped up, and would not betray a friend (to wit: the defendant) is tantamount to the same thing.”
305 Or at 629-30 (quoting Middleton, 294 Or at 438).
Similarly, in Keller, a case involving first-degree sexual abuse of a five-year-old child, the court dealt with expert testimony from a CARES doctor about the child’s CARES evaluation. After citing both Middleton and Milbradt, the court concluded:
“Dr. Bays testified on direct examination during the state’s case-in-chief that ‘[tjhere was no evidence of leading or coaching or fantasizing’ during the child’s interview at *330C.A.R.E.S. and that, in that interview, the child was ‘obviously telling you about what happened to her body.’ Each of those statements amounts to testimony that the child was credible.”
Keller, 315 Or at 285.
In Wilson, we applied that line of cases to a CARES diagnosis of sexual abuse. There, a child reported that the defendant had repeatedly sexually abused her. The child was taken to CARES for a sexual-abuse evaluation. After reviewing an interview with the child and conducting a medical examination, which revealed no physical evidence of abuse, Dr. Bays diagnosed the child with having been sexually abused. The defendant objected to the admission of Bays’s diagnosis, on the grounds that “any comment on an alleged diagnosis * * * would be a direct comment with regards to the credibility of the [child.]” 121 Or App at 462. The trial court overruled the defendant’s objection.
On appeal, the defendant argued that “Bays’ diagnosis of the child should have been excluded, because the jury could have inferred from that diagnosis that Bays believed the child’s statements.” Id. at 465. After discussing Middleton and Keller, we rejected the defendant’s argument. We first distinguished Keller on the basis that Bays had not testified that the child “had not been coached or led and had not fantasized the incident” — that is, she had not directly commented on the credibility of the child. Id. We then concluded that the diagnosis was admissible, even though it tended to indicate that the child was telling the truth:
“Bays testified that, on the basis of her evaluation of the child’s interview, physical examination andhistory, she diagnosed the child as having been sexually abused. Although, if believed,, Bays’ testimony supported the child’s testimony, that does not render it a direct comment on the child’s credibility. It was an opinion as to the proper medical diagnosis. A medical doctor is not precluded from testifying as to her medical diagnosis simply because the jury may infer from that testimony that another witness is or is not telling the truth.”
Id. (emphasis added).
*331From those cases, we glean that, under Wilson, the dispositive question at the time of petitioner’s criminal trial was whether Findley’s testimony about the treatment recommendations was directed, either expressly or implicitly, at the credibility of C — or of any other witness — or whether it simply allowed an inference in support of C’s testimony. With that in mind, we return to the merits, under Wilson, of an objection to Findley’s testimony about the CARES treatment recommendations as an impermissible comment on the credibility of C. As noted, the post-conviction court acknowledged the similarity between the treatment recommendations at issue here and the diagnosis of sexual abuse in Wilson. It also acknowledged that, given Wilson, “the fact that [Findley’s] testimony, if believed, supports the credibility of [C’s] statements does not amount to a direct comment on credibility.” We agree with that conclusion.
As noted, petitioner challenged Findley’s testimony about three treatment recommendations: that C have no contact with petitioner, that petitioner receive a psychological and a sexual-offender evaluation, and that N’s safety be evaluated in light of her ongoing contact with petitioner. Unlike the testimony in Milbradt and Keller, Findley’s testimony did not implicate C’s capacity for lying; it did not identify any evidence of leading or coaching on the part of C — or the lack thereof; and it offered no opinion on whether C was obviously testifying about what had happened to her body. Thus, under Wilson, it was admissible.
Of course, like the testimony in Wilson, Findley’s testimony would allow the jury to draw inferences in support of C’s credibility. Regarding the first and second recommendations — that C have no contact with petitioner and that petitioner undergo evaluations — the jury could infer that, at the time of Findley’s initial interview with C, he found her allegation that petitioner had abused her to be credible enough to warrant that C have no further contact with petitioner pending further investigation of petitioner. And the post-conviction court relied on that aspect of Findley’s testimony in concluding that petitioner had suffered prejudice.
*332In doing so, the post-conviction court characterized Findley’s testimony as going “one step further” by recommending that petitioner be evaluated. Petitioner makes a similar argument on appeal, contending that Findley’s testimony about the treatment recommendations would have been impermissible under Wilson because “it was not simply a ‘diagnosis’ of abuse suffered by a victim, but was a ‘diagnosis’ of the defendant being an abuser,” and as such was a comment on the credibility of both C and petitioner.7 The state argues in response that the treatment recommendations were “more innocuous” and, therefore, fell short of the “full-blown medical diagnosis” of sexual abuse approved in Wilson. However, whether or not Findley’s testimony went further — or was more innocuous — than the diagnosis in Wilson, the pertinent inquiry is whether the difference between the two rendered the treatment recommendations “a direct comment on *** credibility” as that phrase was used in Wilson. It did not.
Although the post-conviction court did not rely on the third treatment recommendation- — viz., that N’s safety be evaluated — in determining prejudice, we conclude that it, too, was admissible under Wilson. We acknowledge that that part of Findley’s testimony could inferentially support C’s testimony in at least two ways. First, it would allow the jury to infer that Findley believed that petitioner’s temper posed a threat to N’s safety. On that point, the jury had heard evidence of the contentious nature of petitioner’s divorce and petitioner’s violent outburst while driving the children to his ex-wife’s house. Second, it would allow the jury to infer that Findley believed C when she said that petitioner had engaged in abusive conduct toward N. However, under Wilson, the fact that Findley’s testimony allowed for those inferences did not render it an impermissible comment on C’s credibility, and, accordingly, any objection to it would have failed.
*333For those reasons, we conclude that petitioner failed to demonstrate by a preponderance of the evidence that trial counsel’s failure to object to Findley’s testimony about the CARES treatment recommendations prejudiced petitioner. It follows that the post-conviction court erred in granting relief to petitioner on the ground that it did.
Reversed on appeal; affirmed on cross-appeal.
Petitioner filed a cross-appeal, raising four assignments of error. We reject those assignments without discussion.
In explaining that “the rules are different for CARES,” Cohen referred back to his answer to an earlier question. Asked to explain his use of the term “vouching statements,” Cohen had testified:
“You’re talking about two categories. You’re talking about case law authorized vouching, which is not what we call it, * * * which is what happens when CARES comes in because they are a medical facility and says we diagnosed that this child was sexually abused by the defendant. That is not considered vouching, * * * even though I respectfully disagree * * *. However, any other form of vouching is verboten.”
Notably, the post-conviction hearing — and, thus, Cohen’s testimony — occurred before the Oregon Supreme Court decided State v. Southard, 347 Or 127, 218 P3d 104 (2009), which dealt with the admissibility of expert evidence in sexual-abuse cases.
The entirety of petitioner’s closing argument about the treatment recommendations was:
“Trial [counsel also] failed to object to the CARES social worker recommending that Petitioner’s other child be evaluated for safety concerns [as well] as recommending that Petitioner have an anger control assessment. By making those recommendations, the jury was left with the impression that the allegations must be true since a professional was making recommendations which supported that finding.”
The post-conviction court also reasoned that trial counsel should have objected under OEC 401, OEC 403, and OEC 404 to Findley’s testimony about the treatment recommendation regarding N’s safety, on the basis that that part of Findley’s testimony was irrelevant and unfairly prejudicial. Despite that conclusion, both the state and petitioner frame the issue on appeal to be whether Findley’s testimony about N was inadmissible as an improper comment on C’s or petitioner’s credibility — which was the post-conviction court’s focus when it concluded that petitioner had suffered prejudice as a result of the admission of Findley’s testimony. As it is, notwithstanding petitioner’s failure to defend the trial court’s ruling, we conclude that any objection under OEC 401, OEC 403, or OEC 404 to Findley’s testimony about the treatment recommendation regarding N’s safety would have failed.
In reaching its conclusion, the post-conviction court also considered State v. Southard, 347 Or 127, 218 P3d 104 (2009), which had been decided roughly one month before the entry of the post-conviction court’s order. Under Southard, the post-conviction court noted, “the diagnosis and treatment would be inadmissible.” However, although Southard was decided before the post-conviction court entered its order, it postdated plaintiff’s trial by nearly eight years. As the state points out, because trial counsel’s performance is to be viewed as of the time of trial, we have held that a failure to anticipate Southard’s holding does not render representation constitutionally inadequate. Umberger v. Czerniah, 232 Or App 563, 565, 222 P3d 751 (2009), rev den, 348 Or 13 (2010) (“To render assistance that is constitutionally adequate, trial counsel need not be clairvoyant.”). To the extent that the post-conviction court relied on Southard in concluding that trial counsel’s performance was constitutionally deficient, it erred.
The Supreme Court further examined the issue of vouching evidence in the context of CARES testimony in 2010 in State v. Lupoli, 348 Or 346, 234 P3d 117 (2010). However, because our analysis is limited to the law in effect at the time of petitioner’s criminal trial, we do not consider the effect of Lupoli on the admissibility of the evidence in this case.
By their terms, Findley’s treatment recommendations were a “diagnosis” of neither C nor of petitioner. Petitioner’s characterization of Findley’s testimony as a ‘“diagnosis’ of the defendant being an abuser” calls to mind the Supreme Court’s decision in State v. Sanchez-Alfonso, 352 Or 790, 801-02, 293 P3d 1011 (2012) (no basis under OEC 702 for admission of CARES report that “[the victim] was physically abused by [the defendant]”). Here, however, petitioner raised no claim alleging that trial counsel was ineffective for failing to object to the treatment recommendations on the ground that they violated OEC 702.