Defendant seeks reversal of a judgment convicting him of two counts of third-degree sexual abuse, ORS 163.415; one count of furnishing alcohol to a minor, ORS 471.410(2); and one count of contributing to the sexual delinquency of a minor, ORS 163.431. He argues that the trial court should have excluded, under OEC 403, the victim’s testimony that defendant refused to use a condom during four of their sexual encounters. In a previous appeal, we reversed defendant’s convictions and remanded to the trial court after accepting the state’s concession that the trial court had failed to analyze the admissibility of that evidence under the four-step analysis required by State v. Mayfield, 302 Or 631, 645, 733 P2d 438 (1987). State v. Sewell, 222 Or App 423 (2008), adh’d to on recons, 225 Or App 296 (2009) (Sewell I). On remand, the trial court analyzed the disputed testimony under Mayfield and concluded that the evidence was properly admitted because its probative value was not substantially outweighed by unfair prejudice. The court again entered a judgment of conviction; defendant appeals, assigning error to the trial court’s admission of the disputed testimony. We affirm, concluding that the court did not abuse its discretion in determining that the probative value of the disputed evidence to bolster the victim’s credibility was not substantially outweighed by the danger of unfair prejudice.
This case turns on the trial court’s admission of disputed evidence under OEC 403. That rule states that,
“[although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay or needless presentation of cumulative evidence.”
When a trial court is presented with a request to exclude evidence as unfairly prejudicial under OEC 403, the court must:
“[(1)] analyze the quantum of probative value of the evidence and consider the weight or strength of the evidence [;] * * * [(2)] determine how prejudicial the evidence is, to what extent the evidence may distract the jury from the central *465question whether the defendant committed the charged crime[;] *** [(3)] balance the prosecution’s need for the evidence against the countervailing prejudicial danger of unfair prejudice[;] *** [and (4)] admit all the proponent’s evidence or * * * admit only part of the evidence.”
Mayfield, 302 Or at 645. The court commits legal error if it “fails to exercise discretion, refused to exercise discretion or fails to make a record which reflects an exercise of discretion.” Id.
We take the facts and some of the procedural history of this case from Sewell I:
“According to the state’s evidence, the conduct giving rise to defendant’s convictions occurred during the weekend of August 20 through August 22, 2004. At that time, defendant was 35 years old, and the victim was 17. The victim’s eighteenth birthday was August 29, 2004. On August 20, defendant, the victim, and two other people had dinner together. After dinner, defendant bought alcoholic beverages, and defendant, the victim, and the victim’s friend, G, spent the evening in the victim’s residence drinking them. The victim testified that she and defendant later went into her bedroom and had sexual intercourse. G, who was sleeping in the living room, testified that she overheard them.
“The victim testified that she and defendant had sexual intercourse again on the morning of August 21. Later that day, defendant bought more alcoholic beverages and, that evening, defendant, the victim, and G once again drank the beverages in the victim’s residence. G testified that she overheard defendant and the victim having sex in the victim’s room later that evening. G left the residence on the morning of August 22. The victim testified that she and defendant stayed at her residence and that they had sexual intercourse twice that day. Defendant left the victim’s residence that night. The prosecutor separately asked the victim, with respect to four of the five sexual encounters, whether defendant refused her request that he wear a condom. In each instance, the victim answered affirmatively that defendant refused to wear a condom.
“* * * At trial, defendant’s theory of the case was that he and the victim had a consensual sexual relationship and that they did not have sexual relations until she turned *46618. Defendant also argued that, if he had sexual relations with the victim during the alleged weekend, it was because defendant did not know that the victim was only 17 years old at the time.”
222 Or App at 425-26.
As for the victim’s testimony that defendant refused to wear a condom,
“[defendant objected to that testimony on the ground that it was irrelevant under OEC 401 and, alternatively, that its prejudicial effect substantially outweighed its probative value under OEC 403. The trial court overruled defendant’s objections, explaining as follows:
“‘Frankly, that — she has a right to explain the circumstances surrounding the matter, and I don’t believe a weighing was necessary regarding the prejudice vis-a-vis the probative value. It explains * * * her memory. It makes it more clear to the jury what she remembers what occurred there.
“‘All very, I think, to the State’s view, important stuff. I don’t think that — the issue is whether she had intercourse or not, but the — whether the condom was used or not really doesn’t add a whole lot to that. It just adds to the circumstances surrounding, so I’m going to — that’s the reason I overruled your objections.’”
“The jury convicted defendant of two of the five counts charging him with third-degree sexual abuse, specifically, of the counts involving the acts that G testified she overheard on the evenings of August 20 and 21, and it found him not guilty of the three remaining third-degree sexual abuse charges. The jury also convicted defendant on the two remaining counts, furnishing alcohol to a minor and contributing to the sexual delinquency of a minor.”
Id. at 426-27.
On appeal in Sewell I, defendant asserted that the trial court committed legal error because it had failed to engage in the four-step analysis required by Mayfield. The state conceded that point, but argued that, nonetheless, the admission of the disputed evidence was harmless. We accepted the state’s concession, concluded that the error *467was not harmless, and reversed defendant’s convictions and remanded to the trial court. In the context of analyzing whether the conceded error was harmless, we discussed the probative value and potential prejudice of the disputed testimony. As discussed more fully below, we noted that the disputed testimony had “little to do with either party’s theory of the case” and “was, at best, marginally relevant to the state’s proffered purpose of showing that the sexual encounters occurred before the victim’s eighteenth birthday.” We also stated that “the evidence *** could have persuaded the jury that defendant was an irresponsible adult male who deserved to be punished for his conduct with a much younger female.” Id. at 430. Thus, we concluded that we could not say that there was little likelihood that the admission of the disputed testimony affected the verdict. See State v. Poitra, 206 Or App 207, 212, 136 P3d 87, rev den, 341 Or 245 (2006) (“Under Article VII (Amended), section 3, of the Oregon Constitution, an error is harmless if there is little likelihood that it affected the verdict.”).
At the hearing on remand, the state urged the trial court to adhere to its original ruling and allow the evidence; defendant asserted that the harmless error discussion in Sewell I essentially required the court to exclude the evidence under OEC 403 because we determined that the evidence was marginally relevant and potentially prejudicial. The court analyzed the evidence under the four-step process of Mayfield and decided that the prejudicial effect of the evidence did not substantially outweigh its probative value. The trial court concluded that the credibility of the victim was “extraordinarily critical” and that the disputed testimony was persuasive to establish the victim’s credibility. As for prejudice to defendant, the court concluded that any prejudice from the disputed testimony was “marginally slight” because “whether he wore a condom or not seems to me to be really — when you look at the more serious charge he’s facing, did he have sex with an underage child? Whether he wore a condom or not to the — as far as the jury being prejudiced, I find the prejudice to be extraordinarily mild.”
The court went on to balance the state’s need for the evidence against the danger of unfair prejudice:
*468“The balancing now against the need as I — for the prejudicial evidence against the need *** for it against the probative value, again, I — the victim in this case, credibility was really a determinative factor in the outcome of the case, explaining the circumstances surrounding the event. Exploring the memory this victim had of — and memory was a very critical part of this, as a matter of fact.
“My recollection is part of the defense that the defense presented was, well, we had the wrong date entirely. That this occurred sometime after she turned 18 years of age. For her to be able to explain, with particular emphasis on what had happened on that date, and circumstances surrounding it, including her request that the defendant wear a condom, I think fleshes out her memory and her credibility.
“And so I find that on the balance that the State — that the State’s evidence, that they needed that evidence, and the State had a right to the evidence, and the State — that the prejudice was so mild, if not existing at all. I don’t find there is any possibility of admitting part of the evidence. It’s all or nothing, and I did.
«* * * * *
“But the bottom line is I think * * * the case really comes down to: Did the State need it? And I find an extraordinary need for this in this case, because of the victim. Frankly, she did not present well in this case. And I’m satisfied that the jury found accordingly based upon their verdict.
“The State needed the evidence in this matter and I— the weighing here, the act of intercourse, sexual intercourse with a child under the age of 18 strikes me as being so much more significant, and that’s what we’re trying, than whether or not the defendant chose not to wear a condom that I find in the weighing process here in this matter, that it does not strike me as being very significant.”
We review a trial court’s decision to admit or exclude evidence under OEC 403 for an abuse of discretion. State v. Titus, 328 Or 475, 481, 982 P2d 1133 (1999). We generally defer to the trial court’s decision regarding “whether the probative value of the evidence is substantially outweighed by the potential for prejudice.” State v. Williams, 313 Or 19, *46929-30, 828 P2d 1006, cert den, 506 US 858 (1992). “An abuse of discretion occurs when a court exercises its discretion ‘to an end not justified by, and clearly against, evidence and reason.’” State v. Mason, 100 Or App 240, 243, 785 P2d 378 (1990)). As a general rule, OEC 403 favors admissibility and places the burden on the party seeking exclusion of the evidence, but it also allows a means of excluding distracting evidence from a trial. State v. O’Key, 321 Or 285, 320, 899 P2d 663 (1995).
The probative value of the evidence is “essentially a measure of persuasiveness that attaches to a piece of evidence.” State v. Lawson, 352 Or 724, 757, 291 P3d 673 (2012). Put another way, it is the strength of the relationship between the proffered evidence and the proposition sought to be proved. Id. The critical inquiry in determining whether evidence is unfairly prejudicial is whether the evidence improperly appeals to the preferences of the trier of fact for reasons that are unrelated to the power of the evidence to establish a material fact. State v. Supanchick, 245 Or App 651, 666, 263 P3d 378 (2011), rev allowed, 352 Or 107 (2012). For example, evidence is unfairly prejudicial in a situation “in which the preferences of the trier of fact are affected by reasons essentially unrelated to the persuasive power of the evidence to establish the fact of consequence.” O’Key, 321 Or at 300.
On appeal, defendant contends that the probative value of the disputed testimony was slight because the victim’s testimony “did not make it more likely that the conduct occurred.” Pointing to the parties’ theory of the case, defendant contends that the issue in the case was whether defendant had intercourse with the victim before her eighteenth birthday, and whether defendant used a condom did not add much to that issue. Defendant also highlights our harmless error discussion in Sewell I, in which we described the disputed testimony as “marginally relevant to the state’s proffered purpose of showing that the sexual encounters occurred before the victim’s eighteenth birthday” because “evidence that defendant refused to wear a condom had little to do with either party’s theory of the case.” 222 Or App at 430.
*470As for the prejudicial effect of the testimony, defendant argues that it was substantial, given that “[e]vidence that a 35-year-old defendant refused to honor a 17-year-old victim’s request that he wear a condom is emotionally-charged evidence; it evokes societal misgivings, even outrage, against predatory, older * * * men and underage girls.” Again, defendant points to our discussion of harmless error in Sewell I, in which we stated:
“Notwithstanding that the evidence at issue here had minimal relevance to either party’s theory of the case, that evidence nevertheless could have persuaded the jury that defendant was an irresponsible adult male who deserved to be punished for his conduct with a much younger female. That impression could have been reinforced on each of the four occasions that the prosecutor asked the victim whether defendant had refused to wear a condom during their sexual encounters and she answered affirmatively.
«* * * * *
“The jury could have been unsympathetic to the victim and still have believed that defendant was a menace to society who should be incarcerated because of his penchant for having unprotected sex with young females.”
Defendant contends that we recognized the prejudicial impact of the evidence in our harmless error analysis.
The state, on the other hand, maintains that the record of the remand hearing reflects that the trial court “exercised its discretion within its range of permissible options.” That is, there was no abuse of discretion. Regarding the probative value of the disputed testimony, the state asserts that, even though we characterized the evidence as marginally relevant in Sewell I, relevance in the context of a harmless error analysis is fundamentally different than when examining the probative value of evidence under OEC 403. The state contends that, as reflected at the remand hearing, the state proffered the disputed testimony to bolster the victim’s credibility because credibility was a determinative factor in the case. The trial court concluded that the victim’s ability to remember specific details and “to be able to explain, with particular emphasis on what happened on that date, and the circumstances surrounding it, including *471her request that defendant wear a condom * * * fleshes out her memory and her credibility,” thus making her more believable as a witness.
As for whether the disputed testimony was unduly prejudicial, the state argues that the trial court correctly concluded that any prejudice was mitigated by the nature of the proceeding itself. That is, defendant’s concern that the disputed evidence is “emotionally-charged evidence [that] evokes societal misgivings, even outrage, against predatory, older * * * men and underage girls” are the same concerns that are stirred up by the fact that he was charged with having sexual intercourse with an underage girl. In short, whether the intercourse was unprotected or not adds little to those concerns. As the court explained on remand, the fact that the sex was unprotected would have added little to the jury’s outrage because “sexual intercourse with a child under the age of 18 strikes [a person] as being much more significant * * * than whether or not the defendant chose not to wear a condom.”
In sum, the trial court determined that the disputed evidence was critical to the state’s case because the victim’s credibility was of central importance and, in the court’s view, the more detail that the victim could provide of the sexual encounters, the more persuasive was her testimony that the encounters occurred on the dates that she claimed (before she turned 18). The central issue in the case was when defendant and the victim had sexual intercourse, and the disputed testimony bolstered the likelihood that her testimony on that critical fact was accurate. The trial court also concluded that any prejudice was mild because the conduct that defendant was charged with — sex with an underage girl — evoked the same societal misgivings as would the disputed testimony and that there was little chance that the evidence of defendant’s refusal to wear a condom would have added significantly to the “jury’s outrage.” The state needed the evidence to establish the victim’s credibility because the victim “did not present well” and, because any prejudice was slight, the trial court concluded that the probative value was not substantially outweighed by the danger of unfair prejudice.
*472Although this case presents a close discretionary call —and, indeed, our harmless error discussion in Sewell I leaves the impression that this court, if presented with the matter in the first instance, might have struck the discretionary balance differently — we cannot conclude that the trial court’s determination was outside the range of legally permissible outcomes. We remanded in order to allow the trial court to engage in the OEC 403 analysis that it failed to perform initially, and, on remand, the court made an extensive record that reflects its exercise of discretion by analyzing and weighing the probative value of the evidence against its potential for unfair prejudice. In particular, the court determined that because of the victim’s presentation, the disputed evidence was critical to establish the credibility of her account of the events, an observation which we are not equipped to make in reviewing the appellate record of the trial. We cannot say that the court’s determination in that regard was outside the range of legally permissible outcomes.
Further, our harmless error discussion in Sewell I did not bind the trial court’s exercise of discretion under OEC 403 on remand. In Sewell I, we analyzed the disputed testimony for harmless error — that is, whether there was little likelihood that the admission of the evidence affected the verdict. That analysis is fundamentally different than the one with which we are tasked here — that is, whether the trial court’s determination under OEC 403 constituted an abuse of its discretion.
Affirmed.