*290¶1 A jury convicted Revere Burnett of six criminal charges, including rape and aggravated sexual abuse of a child. Burnett contends on appeal that he received ineffective assistance of counsel during his trial, because his attorney failed to object to the testimony of the State's expert psychiatrist. After review, we conclude that much of the expert's testimony was proper, and that objection to most of it would therefore have been futile. However, we agree in part with Burnett's arguments, and conclude that his trial counsel was indeed ineffective for failing to object to that part of the expert's testimony in which the expert inappropriately bolstered the victim's credibility. Accordingly, we reverse Burnett's convictions and remand the case for a new trial.
BACKGROUND1
¶2 In 2010, after sixteen years of marriage, Burnett and his wife (Mother) divorced, and Mother was awarded sole legal custody of their three teenage children: Victim, her sister (Sister), and her brother (Brother). The divorce decree granted Burnett parent-time with all three children. In May 2011, less than a year after the divorce, Burnett raised the possibility of increasing the quantity of his parent-time with the children. Within days of Burnett broaching that subject, then-fifteen-year-old Victim disclosed to Mother that Burnett had sexually abused and raped her for many years.
¶3 After investigating the matter, the State charged Burnett with committing various crimes upon Victim, including five counts of aggravated sexual abuse of a child, first degree felonies, see Utah Code Ann. § 76-5-404.1(4) (LexisNexis 2017); two counts of rape of a child, first degree felonies, see id. § 76-5-402.1; and two counts of sodomy upon a child, first degree felonies, see id. § 76-5-403.1.2
¶4 At trial, Victim testified that Burnett first started abusing her when she was "about three or four years old" and that the abuse continued until she was "[a]bout 11." She did not "remember every single incident," but could recall several incidents with specificity. Victim described Burnett digitally penetrating her vagina and anus. She further described Burnett forcing her to manually and orally stimulate him. And she also testified that Burnett raped her on more than one occasion, both vaginally and anally.
¶5 Victim testified that, during the years Burnett abused her, she often felt physically ill, even when the abuse was not actually happening. She experienced headaches, nausea, upset stomach, occasional vomiting, and memory loss. These symptoms progressed over time, and Victim later experienced fainting, pseudo-seizures, hallucinations, and age regression. Eventually, Victim was diagnosed with conversion disorder, a condition "where a person takes psychological distress and expresses it in physical symptoms or ways." Victim started cutting herself when she was thirteen, and she also experienced suicidal ideations that continued throughout the time of trial.
¶6 When asked if she had told anyone about the abuse while it was ongoing, Victim stated that she had not. She explained that she "didn't know any different," and that she did not initially tell anyone about the abuse because she was "scared." She stated that Burnett told her that bad things would happen to her family if she disclosed the abuse, and that she "wasn't supposed to tell anyone, that it was normal." Eventually, Victim told Mother, as well as Victim's psychologist, that Burnett had sexually abused her.
*291¶7 The State called Dr. David Corwin (Expert) as an expert witness at trial.3 Expert is a psychiatrist, and is the director of forensic services for the pediatrics department at the University of Utah, and is board certified in psychiatry, child psychiatry, and forensic psychiatry. He described his practice as focused on working with individuals who had been sexually abused as children, and described for the jury several publications he had authored on child sexual abuse, as well as several awards he had received for his work with victims of child sexual abuse.
¶8 Expert readily acknowledged that he had not treated, interviewed, or examined Victim, and that he was not offering any opinion about whether Victim had actually been abused. Instead, the State explored two main subjects with Expert. First, the State asked Expert about how he, as a trained expert, endeavors to tell the difference between cases in which the victim is advancing truthful claims of sexual abuse, and cases in which the abuse allegations are fabricated. Second, the State asked Expert whether, in his experience, he was aware of certain signs or symptoms that might alert him that a particular child may have been sexually abused.
¶9 With regard to the first subject, Expert began by testifying that it is "most common" for abuse victims to delay disclosure of the abuse, and offered many reasons for this phenomenon. For example, a child might not know that he or she should disclose instances of sexual abuse; a child might feel loyal to the abuser or feel affection for the abuser; or the abuser might warn or threaten the child. He testified that disclosure is more common in adolescents than in children because adolescents "are beginning to psychologically separate themselves from their parents, and they can begin to conceive of surviving apart from their parents."4 In contrast, "[v]ery young children are really one with their parents and totally dependent upon their parents." Significantly, he stated that divorce, and its attendant changing family dynamics, is one event that can "release[ ]" children from their previous reluctance to disclose, although he acknowledged that there exists a "small percentage" of cases in which children from divorced families "fabricat[e]" abuse allegations. He stated that those are "the most difficult cases of all," and that it is part of his professional duties to "try to figure out which" instances are the ones that involve fabricated allegations.
¶10 Indeed, during re-direct examination, the State asked Expert what he looks for in "making those evaluations" in cases "where there is a custody dispute and there are allegations of child sexual abuse." In response, the following colloquy occurred between the prosecutor and Expert:
A. I look at a lot of things, because those are the most difficult cases that a forensic expert like myself is asked to look at. And that's because the rate of both disclosure of actual sexual abuse goes up when a family breaks up, because the dynamics that keep it secret are gone, but also the possibility of a misperception, either by a child or more commonly an adult attributing it to a child ....
[I]n a small percentage of cases there are either children who have lied or parents who somehow get their kids to say things that are false. And even though that's a very small percentage in *292most studies of this topic, it's very troubling, because sexual abuse is such a serious thing. ...
Q. Now, in-you said that that's a small percentage of these cases. In those cases are the children typically younger?
A. Yes. Yeah, the hardest cases are the three-, four-[or] five-year-olds.
Q. So ... to clarify, in these cases where there is an allegation like a child sexual abuse that's false, it would involve children that are younger on the-
A. Well, in terms of the divorce custody and the possibility of influence, it's easier to do that with a very young child, because they are more susceptible to being misled, in terms of interviewing or what adults say to them. In older kids, they are more capable of making outright false allegations, to actually make something up. And in the studies that have been done, those are usually kids with an ax to grind. They are trying to influence where they are living. And they usually allege garden-variety sexual intercourse, vaginal intercourse.
Q. So what do you mean by garden variety?
A. Really simple, straightforward, as opposed to more strange, more elaborated, more detailed accounts.
Q. So, for example, just a disclosure that he touched my private part, without any more details, or he put his penis in my vagina, without any additional details?
A. Well, that would be sort of vaginal intercourse, put his penis in my vagina. The touching-you know, when the allegations are less severe, when I am trying to look at these various hypotheses, in my mind I think, well, if this is a false allegation why not go ahead and make a, you know, full-blown, culpable sort of clear description than something that's a little hard[er] and less easy to interpret.
Q. So [Defense counsel] asked you about false allegations of child sexual abuse. Based on your training and your experience and your knowledge of the literature in your field, are those common or are they rare?
A. I have to be a little more specific in terms of false by the parent, false by the child. I think the general experience is that false allegations initiated by the children are less common than actual allegations, ... but they do occur.
¶11 With regard to the second subject, the State asked Expert whether and to what extent warning signs exist that alert a medical professional to the possibility of sexual abuse. Expert testified that there are a "variety" of "symptoms" and "disorders" that are "more common among sexually abused and traumatized children than among other children," including "dissociation" (the ability "to compartmentalize and keep these very difficult experiences apart from the rest of one's awareness"), increased anxiety, sleep problems, nightmares, depression, self-mutilation, hallucinations, panic disorders, and suicidal behavior.
¶12 Expert also testified about "conversion disorder," and explained that there are a "continuum of symptoms" common in patients suffering from conversion disorder, including "pseudo-seizures," and specifically noted that "pseudo-seizures" are "one of the most highly correlated among the conversion symptoms" with sexual abuse. Expert conceded that the symptoms of conversion disorder are "not exclusive" to sexual abuse victims, but when asked whether "a diagnosis of conversion disorder" is "consistent with sexual abuse," Expert responded as follows:
It would-it's consistent with it, and in my view I think consistent with is a pretty low bar, because a lot of things are consistent with things. But it's an indicator, it's not definitive, it doesn't in and of itself prove things like sexual abuse, again because it's correlated with other stressors and traumas as well. Um, but it's-it's-it's an indicator.
¶13 The State asked Expert a hypothetical question: "If you had a patient that had been diagnosed with conversion disorder, was manifesting symptoms such as depression, anxiety, hallucination, self-mutilation and suicidal ideations, would you be concerned about sexual abuse?" Expert answered, "Yes, very concerned, ... [b]ecause all of the things you have listed are known ... to exist at increased *293frequency among victims of child sexual abuse."
¶14 Burnett testified in his own defense at trial, and generally denied all of the allegations against him, the thrust of the defense strategy being that Victim manufactured the allegations against him in an effort to influence the then-pending custody dispute with Mother. Burnett explained that he and Mother divorced in 2010. He stated that Mother and Mother's sister drafted the divorce decree, to which Burnett stipulated, and that Mother "had full legal custody" of Victim. Burnett described Mother taking Victim to various doctors and medical professionals "at least 30 times" in 2010 and 2011 and between twenty and thirty times in 2012, and that Mother did not inform him what the visits were about. Burnett stated that Mother waited until the end of each year before informing him about the medical bills, which were "in the thousands of dollars." Describing Victim's doctor visits, Burnett testified:
[T]hey would see her, they would do tests. There would be nothing wrong with her, and so they would say, well "If such and such happens bring her back and we'll look at her again." And sure enough, she's back in in a couple of days with "X" symptom that they ... suggested.
Burnett stated that Victim "would follow through with whatever [Mother] wanted" and that he "never saw" any of Victim's symptoms.
¶15 Under the terms of the divorce decree, Burnett was not permitted to be involved in Victim's doctor visits. After consulting with a divorce attorney, Burnett decided that he wanted to amend the decree, and arranged to meet with Mother to discuss amendment on May 11, 2011.5 Burnett received a phone call that same day from a detective explaining that Victim had leveled sexual abuse allegations against him.
¶16 During closing argument, Burnett's counsel highlighted the timing of Victim's disclosures, and argued that the case boiled down to an assessment of Victim's credibility. Counsel specifically argued that the allegations against Burnett were "outright falsehoods," and that Victim's allegations were fabricated in an effort to prevent Burnett from modifying the divorce decree.
¶17 In its rebuttal argument at closing, the State also focused on Victim's veracity. The prosecutor argued that Expert's testimony "corroborates" Victim's testimony, stating that Victim had "no ax to grind" and could therefore be considered truthful. The prosecutor emphasized Victim's conversion disorder diagnosis and pseudo-seizures, and specifically reminded the jury that Expert "told you that [conversion disorder and pseudo-seizures are] consistent with the trauma of being sexually abused." The prosecutor also referenced Expert's testimony about how one can separate a truthful accuser from a lying one, and stated as follows:
Dr. Corwin also said that-[c]ounsel talked about, you know, false allegations in child custody cases. And he also-he said that ... specific and unusual disclosures were more reliable. In other words, disclosures where there is simply, oh, he touched my private part, that in the rare cases when there are false allegations, that's what you have. You don't have that in this case. You have a really unusual and specific disclosure.
¶18 The jury convicted Burnett on three counts of aggravated sexual abuse of Victim, two counts of rape of Victim, and one count of sodomy of Victim, but acquitted Burnett of two counts of aggravated sexual abuse of Victim and one count of sodomy of Victim. Burnett appeals.
ISSUE AND STANDARD OF REVIEW
¶19 On appeal, Burnett argues that his trial counsel was ineffective for failing to properly object to Expert's testimony.6 When *294a criminal defendant raises a claim of ineffective assistance of counsel for the first time on appeal, there is no trial court ruling to examine. State v. Hull , 2017 UT App 233, ¶ 12, 414 P.3d 526. We must therefore decide, as a matter of law, whether Burnett received constitutionally ineffective assistance of counsel. Id.
ANALYSIS
¶20 Criminal defendants "enjoy the right ... to have the Assistance of Counsel" for their defense. U.S. Const. amend. VI. This right includes the right to effective counsel, Strickland v. Washington , 466 U.S. 668, 686, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), and applies to privately-retained counsel as well as counsel appointed by the court, Cuyler v. Sullivan , 446 U.S. 335, 344, 100 S.Ct. 1708, 64 L.Ed.2d 333 (1980). To show that his counsel was constitutionally ineffective, Burnett must establish that (1) counsel's performance was objectively deficient, and (2) there is a reasonable probability that, but for counsel's deficient performance, Burnett would have received a more favorable outcome at trial. State v. Garcia , 2017 UT App 200, ¶ 19, 407 P.3d 1061. A criminal defendant must establish both elements to prevail on a claim of ineffective assistance of counsel. Archuleta v. Galetka , 2011 UT 73, ¶ 41, 267 P.3d 232.
¶21 "In evaluating counsel's performance under the first Strickland prong, we recognize the variety of circumstances faced by defense counsel and the range of legitimate decisions regarding how best to represent a criminal defendant." Id. ¶ 39 (citation and internal quotation marks omitted). Thus, "a defendant seeking to establish ineffective assistance of counsel must overcome the strong presumption that trial counsel rendered adequate assistance and exercised reasonable professional judgment." Id. (citation and internal quotation marks omitted). To establish deficient performance, a defendant must show "that counsel's conduct fell below an objective standard of reasonableness under prevailing professional norms." Lafferty v. State , 2007 UT 73, ¶ 12, 175 P.3d 530 (citation and internal quotation marks omitted).
¶22 "To establish prejudice under Strickland 's second prong, a defendant must present sufficient evidence to support 'a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different.' " Archuleta , 2011 UT 73, ¶ 40, 267 P.3d 232 (quoting Strickland , 466 U.S. at 694, 104 S.Ct. 2052 ). A showing of "some conceivable effect" on the outcome is insufficient. Menzies v. State , 2014 UT 40, ¶ 91, 344 P.3d 581 (citation and internal quotation marks omitted). Rather, the probability of a "different result must be substantial, not just conceivable." Id. (citation and internal quotation marks omitted).
¶23 Burnett asserts that his counsel was ineffective for failing to challenge two different aspects of Expert's testimony, both of which, he claims, improperly suggested that Victim's testimony was truthful. First, Burnett contends that Expert's discussion of symptoms that appear more frequently in abuse victims created an improper "profile" of a typical victim of sexual abuse. Second, Burnett asserts that Expert, based on his experience in attempting to discern truthful allegations from false allegations, improperly suggested that Victim's testimony was believable. We disagree that Expert's testimony about symptoms constituted improper profiling, but we agree that Expert's testimony about statistics and methods for discerning whether an abuse victim is truthful had the *295effect of improperly suggesting to the jury that Victim was telling the truth.
A
¶24 Burnett argues that Expert's testimony improperly suggested to the jury "that a collection of symptoms like [Victim's] were likely, if not conclusively, caused by sexual abuse." We disagree.
¶25 "[A] witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue." Utah R. Evid. 702(a). While experts may use their expertise to help the factfinder understand issues at trial, experts cannot testify that a particular witness has or has not told the truth. "A witness's credibility may be attacked or supported by testimony about the witness's reputation for having a character for truthfulness or untruthfulness, or by testimony in the form of an opinion about that character." Id. R. 608(a). But (subject to several exceptions, none of which are applicable here) "extrinsic evidence is not admissible to prove specific instances of a witness's conduct in order to attack or support the witness's character for truthfulness." Id. R. 608(b). This rule "prohibits any testimony as to a witness's truthfulness on a particular occasion." State v. Rimmasch , 775 P.2d 388, 391 (Utah 1989), superseded in part by rule as stated in State v. Maestas , 2012 UT 46, ¶ 121 n.134, 299 P.3d 892. And the rule is often applied in sexual abuse cases. See id. at 392-93 (holding that a doctor's testimony was inadmissible under rule 608 because it opined that a witness was telling the truth on a particular occasion: "Well, specifically, in my opinion, one does not give this kind of information with the amount of details and the amount of clarity unless one has experienced it." (internal quotation marks omitted) ).
¶26 An expert's testimony describing a sexual abuse victim's symptoms thus has the potential to improperly suggest that if an individual has a symptom common with sexual abuse victims, then that individual has been sexually abused. Stated differently:
Profile testimony that portrays the characteristics of the typical victim of sexual abuse ... has a tendency to mislead and confuse a finder of fact by suggesting that the issue to be decided is whether the accusing child possesses these characteristics, rather than whether the child experienced the specific instances of abuse described.
Id. at 402 n.13 ; see also Utah R. Evid. 403 ("The court may exclude relevant evidence if its probative value is substantially outweighed by a danger of" among other things, "unfair prejudice, confusing the issues, [or] misleading the jury.").
¶27 However, our supreme court has held that evidence that a victim's symptoms are " 'consistent with' " sexual abuse does not amount to inadmissible "profiling" evidence, because such evidence "does not prove directly the ultimate legal conclusion that" a particular victim has been abused. State v. Kallin , 877 P.2d 138, 141 (Utah 1994). Such "consistent with" testimony is usually admissible under rule 403 because "[t]he probative value of such evidence is usually beyond the ken of a jury." Id. Testimony stating that certain symptoms are "consistent with" sexual abuse "may enable the jury to assess the probative relevance of the evidence in light of all other evidence." Id. For these reasons, "experts may testify that a victim's behavior is consistent with sexual abuse without" improperly invading the province of the jury as factfinder. See State v. Christensen , 2016 UT App 225, ¶ 27, 387 P.3d 588 (citing Kallin , 877 P.2d at 141 ).
¶28 Here, Expert testified that various physical maladies (e.g., anxiety, sleep disorders, depression) are "more common among sexually abused and traumatized children than among other children." He also testified that conversion disorder is not only "consistent with" sexual abuse, but that it is "an indicator" of sexual abuse. Expert further testified that he would be "very concerned" about the possibility of sexual abuse if a patient presented with "conversion disorder, [and] was manifesting symptoms such as *296depression, anxiety, hallucination, self-mutilation and suicidal ideations."
¶29 We are unpersuaded that these portions of Expert's testimony inappropriately suggested to the jury what result to reach. Certainly, Expert was clear that he had not examined Victim, and that he was not offering an opinion on whether Victim had been sexually abused. Many of the prosecutor's questions were phrased in terms of whether various symptoms were "consistent with" sexual abuse, questions that have already been deemed proper by prior precedent. See Kallin , 877 P.2d at 141 ; Christensen , 2016 UT App 225, ¶ 27, 387 P.3d 588. And we perceive no material difference between testimony that a particular symptom is "consistent with" sexual abuse and testimony that a particular symptom is "an indicator of" sexual abuse. Those two phrases ("consistent with" and "an indicator of") seem to us to mean more or less the same thing: that people who have suffered sexual abuse often exhibit that symptom, and perhaps exhibit that symptom with a greater frequency than the general population does, but that the mere presence of that symptom does not necessarily mean that abuse has occurred. Even the prosecutor's hypothetical question does not strike us as falling on the wrong side of the admissibility line. Asking whether Expert would "be concerned" about the possibility of sexual abuse if he had a patient who had been diagnosed with conversion disorder and who was also showing signs of depression, anxiety, hallucinations, and suicide is simply another way of asking if those symptoms are "consistent with" sexual abuse.
¶30 Expert stopped well short of offering an opinion that Victim had been abused or even that Victim was in fact experiencing any of the symptoms in question; indeed, Expert testified that he had not interviewed Victim, he had not reviewed her medical records, and he did not review any investigative materials associated with the prosecution. Expert's testimony simply stated, in various ways, that certain symptoms are more commonly associated with sexual abuse victims than with the population at large. This testimony was permissible.7 See State v. Martin , 2017 UT 63, ¶¶ 10, 30, 423 P.3d 1245 (stating that testimony "regarding common behaviors ... of children who have been abused" was permissible because it did not "improperly invade[ ] the province of the jury" (internal quotation marks omitted) ); see also Christensen , 2016 UT App 225, ¶¶ 28-29, 387 P.3d 588 (testimony that a victim's "symptoms were consistent with" post-traumatic stress disorder was permissible when the witness explained that the victim was not one of his patients, that his testimony was not a diagnosis, and that the testimony did not demonstrate that the victim had been assaulted). We cannot conclude that the probative value of this testimony was substantially outweighed by any danger of unfair prejudice. See Utah R. Evid. 403.
¶31 Because Expert's testimony as it related to symptoms of sexual abuse victims was admissible, any objection from Burnett's trial counsel would have been futile. "Failure to raise futile objections does not constitute ineffective assistance of counsel." State v. Edgar , 2017 UT App 54, ¶ 13, 397 P.3d 656 (internal quotation marks omitted) (quoting State v. Kelley , 2000 UT 41, ¶ 26, 1 P.3d 546 ). Therefore, Burnett's trial counsel was not ineffective for failing to object to these portions of Expert's testimony.
B
¶32 Burnett next contends that Expert's testimony improperly bolstered Victim's credibility by suggesting that scientific factors and probabilities indicated that her testimony was truthful. We agree that this portion of Expert's testimony crossed the line, and conclude that Burnett's trial counsel provided ineffective assistance by failing to challenge that portion of Expert's testimony.
*297¶33 Our supreme court weighed in on this subject in State v. Rammel , 721 P.2d 498 (Utah 1986). In that case, the credibility of the prosecution's star witness was at issue. See generally id. At trial, that witness testified, under a grant of immunity, that the defendant had committed the crime in question, and stated that he knew this because he had driven the defendant away from the crime scene in a getaway car. Id. at 499-500. This story, however, was not the same story the witness had told to police when he was first questioned; at that point, the witness denied any involvement whatsoever in the criminal enterprise. Id. at 499.
¶34 In an effort to shore up the witness's credibility, the prosecution called a police detective to testify that "he did not consider it unusual" for the witness to "lie to him when [he] was first interrogated." Id. at 500. The detective presented himself as an expert in police interrogations, and as someone who had "experience interviewing several hundred criminal suspects." Id. Based on this expertise, the detective testified that "no criminal suspect ever admitted 'right off the bat' to committing a crime," and that "because most suspects lie when initially questioned by police, it would not have been 'unusual' for [the witness] to lie during the first police interrogation." Id. The trial court admitted the evidence, but the supreme court reversed, for three reasons.
¶35 First, the court noted that any evidence that calls into question a witness's truthfulness "must go to that individual's character for veracity," and because the detective's testimony "did not relate to [the witness's] character for veracity, but instead invited the jury to draw inferences about [his] character based upon [the detective's] past experience with other suspects," "the testimony was inadmissible." Id. Second, the court noted that "foundation" for the detective's testimony "was utterly lacking," because "[t]here was no showing that the anecdotal data from which the detective drew his conclusions had any statistical validity." Id. at 501. Finally-and most significantly for present purposes-the court held that, even if a proper statistical foundation had been laid, the evidence should still have been excluded "because its potential for prejudice substantially outweighed its probative value." Id. The court stated as follows:
In this case, the prosecution attempted to establish, in effect, that there was a high statistical probability that [the witness] lied [in his initial interview]. Even where statistically valid probability evidence has been presented-and [the detective's] testimony hardly qualifies as such-courts have routinely excluded it when the evidence invites the jury to focus upon a seemingly scientific, numerical conclusion rather than to analyze the evidence before it and decide where truth lies. Probabilities cannot conclusively establish that a single event did or did not occur and are particularly inappropriate when used to establish facts "not susceptible to quantitative analysis," such as whether a particular individual is telling the truth at any given time.
Id . (citations omitted); see also id. (stating that, even "assuming statistics were available, [the] statistical likelihood of [an] eyewitness to err would not assist the trier of fact to determine whether a particular eyewitness was telling the truth" (citation omitted) ); State v. Iorg , 801 P.2d 938, 941 (Utah Ct. App. 1990) (stating that our supreme court has "continued to condemn anecdotal 'statistical' evidence concerning matters not susceptible to quantitative analysis such as witness veracity, as one of the categories of evidence leading to undue prejudice").
¶36 Much of Expert's trial testimony in this case is inadmissible under the standards set forth in Rammel , because "its potential for prejudice substantially outweigh[s] its probative value." See Rammel , 721 P.2d at 501 ; see also Utah R. Evid. 403. In this case, as in Rammel , the State portrayed Expert as an expert in being able to discern truthful sexual abuse allegations from false ones. The jury heard testimony that Expert had published papers and articles on the subject, and that he had presented at conferences all over the country. Expert told the jury, at least four times, that it is only in a "small percentage" of cases that children lie about sexual abuse, and he explained to the jury some of the things he looks for in judging an accuser's credibility:
*298In older kids, they are more capable of making outright false allegations, to actually make something up. And in the studies that have been done, those are usually kids with an ax to grind. They are trying to influence where they are living. And they usually allege garden-variety sexual intercourse, vaginal intercourse.
The clear import of Expert's testimony is that only in a "small percentage" of sexual abuse cases-especially cases in which the allegations emerge in tandem with a divorce in the family-are the allegations fabricated, and that most cases of fabrication involve a child "with an ax to grind" and/or a child who is making only "garden-variety" allegations.8 Even though Expert stopped short of offering an opinion that Victim herself was telling the truth, there is no question that, by presenting (and later, in closing argument, emphasizing) this testimony, the prosecution was clearly "invit[ing] the jury to draw inferences about" Victim's credibility "based upon [Expert's] past experience with other" cases and studies. See Rammel , 721 P.2d at 500. Accordingly, we see no way to meaningfully distinguish this case from Rammel .9 This testimony was improper, and had counsel made a timely objection, there is a "reasonable probability" that the testimony would have been excluded. See Edgar , 2017 UT App 54, ¶ 13, 397 P.3d 656.
¶37 "[B]efore we will reverse a conviction based on ineffective assistance of counsel," however, "we must be persuaded that there was a lack of any conceivable tactical basis for counsel's actions." State v. King , 2010 UT App 396, ¶ 31, 248 P.3d 984 (citation and internal quotation marks omitted). The State suggests that Burnett's trial counsel could have had tactical reasons for not objecting to Expert's testimony: in the State's view, counsel could have decided not to object because Expert's testimony supported, at least in part, defense's theory that Victim fabricated her allegations against Burnett. The State asserts that Expert
repeatedly testified that the types of physical and psychological symptoms [Victim] happened to exhibit were common among people who experience severe childhood trauma, not just sexual abuse. And trial counsel drew out testimony that divorce could be one such trauma. Even though [Expert] testified that child sexual abuse was a more likely cause than divorce, [Expert's] testimony supported the defense at least in part.
We are not persuaded, chiefly because this tactical reason could only possibly apply to the first portion of Expert's testimony, discussed above, in which Expert testified that certain symptoms were more common in sexual abuse victims than in the general population. Any tactical advantage in allowing Expert to so testify would have been realized by Expert's admissible testimony; the State's proffered tactical reason therefore offers no possible reason for counsel not to have objected to the second (inadmissible) portion of Expert's testimony. We simply cannot imagine any convincing tactical reason for counsel to have consciously decided not to raise a *299Rammel objection to Expert's testimony regarding credibility.
¶38 And we think it rather evident that Expert's testimony regarding credibility was prejudicial to Burnett. Indeed, we have already determined that the potential for unfair prejudice substantially outweighed any probative value the testimony might have had. In order "[t]o prevail on a claim of ineffective assistance of counsel, [Burnett] must demonstrate that counsel's performance was deficient and that the deficient performance was prejudicial-i.e., that it affected the outcome of the case." State v. Crespo , 2017 UT App 219, ¶ 22, 409 P.3d 99 (citation and internal quotation marks omitted). We conclude that counsel's deficient performance did, in fact, prejudice Burnett. In other words, had counsel properly challenged Expert's testimony, we are convinced that there is a "reasonable probability" that Burnett would have received a more favorable outcome. See Garcia , 2017 UT App 200, ¶ 19, 407 P.3d 1061.
¶39 In this case, there was no confession, no third-party eyewitnesses, and no physical evidence that Victim had been sexually abused; the only direct evidence that abuse occurred was Victim's testimony. Our supreme court has described convictions in such cases as "not strongly supported by the record." See Gregg v. State , 2012 UT 32, ¶ 30, 279 P.3d 396 (stating that "[t]here was no independent physical evidence that supported or contradicted [the victim's] testimony, and therefore, the conviction is not strongly supported by the record"); see also State v. Stefaniak , 900 P.2d 1094, 1096 (Utah Ct. App. 1995) (recognizing prejudicial error in the admission of bolstering evidence in a child sex abuse case where "[t]here was no physical evidence introduced, nor was there testimony from" other witnesses, and where the "case depended on the jury's assessment of the victim's credibility versus the defendant's" (citation and internal quotation marks omitted) ); Iorg , 801 P.2d at 941-42 (recognizing prejudicial error in the admission of bolstering evidence in a child sex abuse case where the "case hinged on credibility"). Moreover, in this case the jury returned a split verdict, acquitting Defendant on all charges related to Sister and Brother, and acquitting Defendant on some of the charges related to Victim. Our supreme court has indicated that a split verdict can be an indication that "the jury was conflicted about the evidence and the competing versions of events offered by the victim and" the defendant, and that therefore the elimination of certain evidence may very well have mattered. State v. Richardson , 2013 UT 50, ¶ 44, 308 P.3d 526.
¶40 The important impact of Expert's testimony was highlighted by the State during closing argument. In discussing the evidence in the record that might corroborate Victim's account, the prosecutor referred the jury to Expert's testimony about credibility, and reminded the jury that Victim had "no ax to grind" and that she had offered "specific and unusual disclosures" that Expert had indicated were "more reliable." This argument was the last thing the jury heard before beginning its deliberations. The prosecution's emphasis in closing argument of Expert's testimony regarding credibility is not only an indicator that the State considered that testimony important corroborative evidence, but also that the testimony was important enough to make a difference. Cf. State v. Martinez , 2015 UT App 193, ¶ 22, 357 P.3d 27 (explaining, in the context of evaluating a trial court's decision on a criminal defendant's mistrial motion, that "[b]revity, vagueness, and absence from closing arguments are among the factors that militate in favor of a finding that improperly admitted evidence did not substantially influence the verdict").
¶41 Accordingly, we conclude that, had counsel properly objected to Expert's testimony regarding credibility, there exists a reasonable probability that Burnett would have obtained a more favorable outcome at trial.
CONCLUSION
¶42 In sexual abuse cases, assuming proper foundation is laid, properly qualified experts are permitted to testify that certain physical symptoms or maladies appear more frequently in victims of sexual abuse than in the general population. In so doing, experts *300may offer an opinion that symptoms are "consistent with," or even "an indicator of," sexual abuse. In this context, it is not necessarily impermissible for hypothetical questions to be asked of such an expert about whether the expert would be "concerned about" sexual abuse if a patient exhibited certain symptoms, since this is simply a different way of asking whether symptoms are "consistent with" abuse. Thus, the part of Expert's testimony that discussed symptoms in this way was proper, and Burnett's trial counsel was not ineffective by not objecting to it.
¶43 However, not even properly qualified experts are permitted to offer statistical evidence, anecdotal or otherwise, that informs the jury, even indirectly, that a witness is more or less likely to be telling the truth on a particular occasion. Such evidence constitutes improper comment on a witness's credibility, and "invites the jury to focus upon a seemingly scientific, numerical conclusion rather than to analyze the evidence before it and decide where truth lies." Rammel , 721 P.2d at 501. The part of Expert's testimony in which he informed the jury that only a "small percentage" of sexual abuse allegations are fabricated, and in which he discussed some of the techniques he uses as an expert to distinguish between cases involving true allegations and cases involving fabricated allegations, was categorically inadmissible. Counsel had no conceivable tactical reason not to object to that testimony, and there is at least a reasonable probability that the outcome of the trial would have been different had the testimony been excluded. Accordingly, Burnett received ineffective assistance of counsel.
¶44 The judgment of the trial court is reversed, and this case is remanded for a new trial.