2025 UT App 106

THE UTAH COURT OF APPEALS STATE OF UTAH, Appellee, v. MAP TOUCH NGOY, Appellant.

Opinion No. 20221005-CA Filed July 10, 2025 Third District Court, Salt Lake Department The Honorable Amber M. Mettler No. 181910498 Erick B. Grange, Attorney for Appellant Derek E. Brown, John J. Nielsen, and Andrew F.

Peterson, Attorneys for Appellee JUDGE MICHELE M. CHRISTIANSEN FORSTER authored this Opinion, in which JUDGES GREGORY K. ORME and RYAN M. HARRIS concurred.

CHRISTIANSEN FORSTER, Judge: ¶1 Map Touch Ngoy appeals his conviction of two counts of aggravated sexual abuse of a child. He argues that the trial court erred in denying the directed verdict motion he made at trial based on insufficient evidence. He additionally argues that the court plainly erred in admitting certain expert testimony. We disagree and affirm.

State v. Ngoy

BACKGROUND ¶2 In October 2018, the police were called after twelve-year- old Mary 1 disclosed to her parents that Ngoy, her grandmother’s husband, had sexually abused her the previous year. As a result of the allegations, the State charged Ngoy with two counts of aggravated sexual abuse of a child for touching Mary’s breasts and genitals. 2 See Utah Code § 76-5-404.1.

¶3 At trial, the State called Mary as its first witness, and she testified as to the abuse by Ngoy. She explained that when her family had first come to live with her grandmother and Ngoy (and several other extended family members), she and her three siblings would sleep on the floor of the “upstairs living room.”

Mary stated that once, in the middle of the night, she was in that room, asleep on her back, when Ngoy came in and began touching her. She recalled that Ngoy first “rubbed up and down [her] legs” and then “proceeded to put his hands under [her] shirt.” She stated that Ngoy unclipped her bra, moved her shirt and bra out of the way, and began “to kiss and rub” her breasts with “[h]is hands and his mouth.” According to Mary, Ngoy also removed the upper part of her leggings and “touched [her] vaginal area.”

She additionally testified that “throughout” the episode, Ngoy “kissed [her] mouth.” Mary stated that the episode ultimately ended by Ngoy pulling her pants back up and putting her shirt down and then retreating to the kitchen.

1. A pseudonym.

2. Ngoy was charged with a third count of aggravated sexual abuse of a child based on conduct alleged by Mary’s younger sister. However, the jury acquitted Ngoy on that charge, and thus, only the charges corresponding to Mary’s allegations are at issue in this appeal.

20221005-CA 2 2025 UT App 106 State v. Ngoy

¶4 Although Mary acknowledged that in the darkness of the room she could not see Ngoy but “was only able to feel him,” and that he never said anything, she explained how she nonetheless knew it was Ngoy: “Because he has no teeth and when he kissed me on my lips I didn’t feel [any] teeth.” She related that during the episode, she “tried to pretend that [she] wasn’t in that situation and that it was all a dream.” And she said that she did not report the abuse to anyone right away because she “was afraid that no one would believe [her].”

¶5 The State thereafter presented its remaining witnesses, including a forensic interviewer (Expert) who had experience testifying “as an expert about child disclosures and about Children’s Justice Center interviews.” During questioning regarding how a forensic interview generally proceeds, the following exchange about the likelihood of delayed disclosure occurred: Q. Okay. Do children always report abuse right away?

A. No. Q. Have you done interviews with children who have disclosed abuse but delayed a substantial period of time before disclosing?

A. Yes.

Q. How common is that?

A. It’s very common. Delayed disclosure is something that is very common, particularly with sexual abuse. There’s research that indicates that as much as up to 60 to 80 percent of abuse is either disclosed after a significant delay or never disclosed. But certainly we see and I’ve seen in my

20221005-CA 3 2025 UT App 106 State v. Ngoy

experience, children who delay the disclosure anywhere from, you know, it could be several weeks to several years.

The questioning then turned to the discussion of possible reasons for a delayed disclosure of abuse.

¶6 When the State rested its case, Ngoy made a motion for a directed verdict. He argued, among other things, that the State had not submitted sufficient evidence to support Mary’s identification of him as her abuser where Mary never saw him but only relied on her lack of feeling teeth for her identification. The trial court, recognizing that it must “view the evidence in a light most favorable to the State,” denied the motion.

¶7 The defense then presented its witnesses, and closing arguments were given. Thereafter, the jury deliberated and returned a guilty verdict on both charges concerning Mary. Ngoy now appeals.

ISSUES AND STANDARDS OF REVIEW ¶8 Ngoy argues that the identity evidence produced by the State at trial was insufficient and, thus, the court erred in denying his motion for a directed verdict. “A trial court’s ruling on a motion for a directed verdict is a question of law, which we review for correctness, giving no particular deference to the trial court’s legal conclusions.” State v. Hirschi, 2007 UT App 255, ¶ 15, 167 P.3d 503 (quotation simplified); see also State v. Stricklan, 2020 UT 65, ¶ 30, 477 P.3d 1251. Nonetheless, “a defendant has a substantial burden on appeal to show that the trial court erred in denying a motion for directed verdict.” Stricklan, 2020 UT 65, ¶ 30 (quotation simplified). The defendant “must show that, when viewed in the light most favorable to the State, no evidence existed from which a reasonable jury could find beyond a reasonable

20221005-CA 4 2025 UT App 106 State v. Ngoy

doubt that the defendant committed the crime.” Id. (quotation simplified).

¶9 Additionally, and in the alternative, Ngoy argues that Expert should not have been allowed to give statistical probability testimony about the likelihood of delayed disclosure of sexual abuse. Ngoy concedes that this issue was not preserved for appeal, but he relies on the plain error exception to the appellate preservation rule. “To prevail under plain error review, a defendant must demonstrate three elements: (1) that an error occurred, (2) that the error should have been obvious to the trial court, and (3) that there is a reasonable likelihood that the outcome would have been more favorable to the defendant in the absence of the error.” State v. Milligan, 2012 UT App 47, ¶ 6, 287 P.3d 1 (quotation simplified).

ANALYSIS I. Sufficiency of the Evidence ¶10 Ngoy argues that the trial court erred in denying his motion for a directed verdict based on an asserted insufficiency of the evidence, specifically, that there was insufficient evidence that Ngoy abused Mary. See generally Utah R. Crim. P. 17(o) (“At the conclusion of the evidence by the prosecution, or at the conclusion of all the evidence, the court may issue an order dismissing any information or indictment, or any count thereof, upon the ground that the evidence is not legally sufficient to establish the offense charged therein or any lesser included offense.”). In reviewing the denial of a motion for a directed verdict based on a claim of insufficiency of the evidence, “we will uphold the trial court’s decision if, upon reviewing the evidence and all inferences that can be reasonably drawn from it, we conclude that some evidence exists from which a reasonable jury could find that the elements of the crime had been proven beyond a reasonable doubt.” State v. Montoya, 2004 UT 5, ¶ 29, 84 P.3d 1183 (quotation simplified).

20221005-CA 5 2025 UT App 106 State v. Ngoy

We ultimately conclude that this condition was satisfied in the instant case.

¶11 We acknowledge that “identification of the defendant as the person who perpetrated the crime charged is an essential element that must be proven beyond a reasonable doubt.” State v. Granados, 2019 UT App 158, ¶ 25, 451 P.3d 289 (quotation simplified), cert. denied, 462 P.3d 800 (Utah 2020). It is also well- settled that a jury’s determinations may be based on reasonable inferences but not on speculation. Salt Lake City v. Carrera, 2015 UT 73, ¶ 11, 358 P.3d 1067. “[T]he difference between an inference and speculation depends on whether the underlying facts support the conclusion. A jury draws a reasonable inference if there is an evidentiary foundation to draw and support the conclusion. In the case of speculation, however, there is no underlying evidence to support the conclusion.” Id. ¶ 12.

¶12 Ngoy argues that there are numerous reasons why Mary’s identification was not reliable evidence to support the jury’s conclusion that Ngoy was Mary’s abuser, thus making the jury’s conclusion pure speculation. He first analogizes to the dangers of eyewitness and earwitness identification and argues that an identification based on touch is “an even more unreliable method to identify a person.” Relatedly, he argues that Mary’s perceptual abilities would have been “significantly decreased from the stress of the alleged incident,” further increasing the unreliability of her identification. But while these points may have been valid factors for the jury to consider in weighing Mary’s identification testimony, their mere existence does not preclude the jury from ultimately crediting the identification and relying on that identification in reaching a guilty verdict.

¶13 Indeed, despite the dangers generally inherent in eyewitness identification, criminal convictions are frequently based on admissible identifications by the crime victim or other eyewitnesses, see Utah R. Evid. 617(b) (providing that the

20221005-CA 6 2025 UT App 106 State v. Ngoy

admissibility of eyewitness identification turns on whether, considering certain enumerated factors, it is shown that the jury could “reasonably rely on the eyewitness identification”), and the very cases cited by Ngoy establish that even earwitness identification may, under certain circumstances, be adequate to sustain a conviction, see State v. Booker, 709 P.2d 342, 345 (Utah 1985) (“[A] voice identification alone is considered insufficient to support a conviction unless shown to be especially reliable.” (emphasis added)); State v. Silva, 2000 UT App 292, ¶ 15, 13 P.3d 604 (“For identification by voice of a person to be sufficient to sustain a conviction, the identifier must either be familiar with the voice of the person identified or that person’s voice possesses some peculiar characteristic which could not be easily mistaken.” (emphasis added) (quotation simplified)). Thus, the relevant question is not whether a type of identification is, by definition, reliable, but, instead, whether the circumstances surrounding a specific identification make it such that a jury could reasonably rely upon it.

¶14 Regarding the specific circumstances present here, Ngoy argues that a lack of peculiarity of the touch and Mary’s lack of familiarity with Ngoy render Mary’s identification unreliable.

Specifically, Ngoy argues that (1) “not having teeth is common” and “is not a peculiar characteristic sufficient to identify a person” and (2) no particular familiarity with the touch involved was established at trial because the State presented “no evidence that [Mary] was familiar with Ngoy kissing her on the lips or anywhere” and no evidence that she generally “could distinguish between a kiss on the lips from a person with teeth and a kiss on the lips from a person without teeth.” 3

3. Ngoy also argues that Mary’s identification is unreliable under the circumstances here because Mary “failed to initially identify Ngoy even by sight at trial.” But this characterization strikes us as (continued…)

20221005-CA 7 2025 UT App 106 State v. Ngoy

¶15 As to the peculiarity of the touch, we disagree. We think it reasonable for a jury to infer—absent any suggestion of a break- in or an intruder—that the universe of potential abusers was limited to those individuals residing in the household. Likewise, it would be reasonable for a jury to infer that toothlessness was a relatively unique characteristic among the individuals living in the house, the vast majority of whom were children or young adults. Thus, we are convinced that toothlessness would, under the circumstances present here, amount to a unique and highly reliable identifying feature and that the jury could therefore reasonably rely on an identification based on that feature.

¶16 As to familiarity, we agree with Ngoy that the State did not present any evidence that Mary, who had recently come to live

somewhat misleading. The referenced exchange during Mary’s direct examination is as follows: Q. . . . So when you met your [grandmother], was Map Ngoy already her husband when you met her?

A. Yeah.

Q. Do you see Map Ngoy today?

A. No. Q. So asking about [your grandmother’s] husband, do you see him today?

A. Yes.

Q. Okay. Could you tell us where he’s sitting and something that he’s wearing?

A. He’s sitting to my left with a black blazer and black tie.

It seems quite likely that the initial “No” response was due to confusion as to what was being asked, not from any actual struggle Mary had in recognizing Ngoy in the courtroom. And at any rate, it is unclear how a delay in visual identification at trial— following several years of no contact—would impact the reliability of Mary’s identification via touch at the time of the abuse.

20221005-CA 8 2025 UT App 106 State v. Ngoy

with Ngoy, was familiar with his touch or specifically how, considering her youth and likely inexperience, she could tell that the person who kissed her was toothless. And certainly, more specific testimony as to what exactly Mary felt during the kissing and the details of how she was able to know that the person kissing her was toothless would have been helpful to the jury.

Indeed, as Ngoy points out, Mary’s testimony that she did not feel teeth could also lead to an alternative reasonable inference that the person had teeth but that Mary just did not feel the teeth during the kissing. Nonetheless, “the fact that we can identify an equally plausible alternative inference is not nearly enough to set [a] verdict aside.” State v. Ashcraft, 2015 UT 5, ¶ 25, 349 P.3d 664 (quotation simplified).

¶17 We recognize that our task on appeal is not to weigh competing inferences that flow from the evidence presented but to, instead, simply assess “whether the jury’s verdict is reasonable in light of all of the evidence taken cumulatively, under a standard of review that yields deference to all reasonable inferences supporting the jury’s verdict.” Id. ¶ 24. Because we determine that, under these circumstances, toothlessness is a unique characteristic that would render an identification based thereon sufficiently reliable for the jury to consider, and because Mary’s identification cleared the bar—albeit maybe not by much—of “some evidence” from which a reasonable jury could infer that Ngoy was the perpetrator, see State v. Montoya, 2004 UT 5, ¶ 29, 84 P.3d 1183 (quotation simplified), we must affirm. 4

4. Although the State argues that Mary’s identification was sufficient evidence to support Ngoy’s conviction, it also points to other circumstantial evidence that it argues also supports the conviction, namely, Ngoy’s “silence in the face of an accusation” and Mary’s younger sister’s separate allegations of inappropriate touching by Ngoy. But we do not consider these factors in our (continued…)

20221005-CA 9 2025 UT App 106 State v. Ngoy

II. Expert Testimony ¶18 Ngoy argues that the trial court plainly erred in allowing Expert to testify that “[t]here’s research that indicates that as much as up to 60 to 80 percent of abuse is either disclosed after a significant delay or never disclosed.” Ngoy asserts that such testimony “invited a numerical conclusion that [Mary] was telling the truth rather than allowing the jury to decide the truth based on the evidence.” In doing so, he relies on holdings in both State v. Rammel, 721 P.2d 498 (Utah 1986), and State v. Iorg, 801 P.2d 938 (Utah Ct. App. 1990).

¶19 Nearly identical expert testimony and appellate argument were recently addressed by this court in State v. Garcia-Cardiel, 2024 UT App 174, 561 P.3d 692, cert. denied, 564 P.3d 959 (Utah 2025). There, the State’s expert had testified “about some research that suggests that 60 to 80 percent of all abuse is not reported at all or not reported until adulthood.” Id. ¶ 20 (quotation simplified). On appeal, the defendant challenged this testimony, arguing that “it was plain error for the trial court not to exclude it” because it “resemble[d] the inadmissible probability evidence” given in Rammel and Iorg. Id. ¶20 In considering this argument, the Garcia-Cardiel court determined that the expert testimony at issue did not run afoul of

sufficiency assessment. The accusation of abuse was vague at best (Mary’s father testified that he—in a language in which he was not fluent—told Ngoy “to keep his hands off [his] kids” and that Ngoy seemed “pretty confused about what [he] was talking about”), and it is unclear whether Ngoy was given any real opportunity to respond (Mary’s father testified that he “just walked away” after confronting Ngoy). And where the jury acquitted Ngoy of the charge based on alleged abuse of Mary’s sister, it is unlikely that the jury credited those allegations in arriving at its guilty verdict.

20221005-CA 10 2025 UT App 106 State v. Ngoy

Rammel and Iorg because the testimony “was neither anecdotal nor used to bolster [an individual’s] truthfulness.” Id. ¶ 22. The court reasoned that the expert’s statistical evidence “came not from his experience interviewing victims but from research he was familiar with as an expert in the field.” Id. The court also reasoned that the expert “never testified specifically about” the victim disclosures from that case or “weigh[ed] in on another witness’s credibility”; instead, the expert “spoke only in general terms about delayed reporting of child abuse.” Id. Thus, the court concluded, the expert testimony “was admissible under Utah law, and there was no error when the trial court did not exclude it.” Id. ¶21 Although Ngoy acknowledges that Garcia-Cardiel “appears to be on point” and would suggest the trial court did not plainly err in allowing Expert’s probability testimony, Ngoy asserts that that case “is wrong and should be overturned because it contradicts established Utah precedent.” Specifically, Ngoy suggests that although this established precedent may allow “testimony that an event is ‘common,’” it does not allow “testimony that something is ‘statistically more likely’ to occur.” (Quoting State v. Nunez-Vasquez, 2020 UT App 98, ¶¶ 63, 66, 468 P.3d 585, cert. denied, 474 P.3d 945 (Utah 2020).)

¶22 We do not agree with Ngoy’s reading of Utah appellate precedent. While some precedent to which Ngoy refers has indeed treated as inadmissible certain testimony setting forth statistical probabilities, it has not established that statistical testimony is per se inadmissible. See Rammel, 721 P.2d at 501 (“Even where statistically valid probability evidence has been presented . . . 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

20221005-CA 11 2025 UT App 106 State v. Ngoy

telling the truth at any given time.” (emphasis added) (quotation simplified)); State v. Burnett, 2018 UT App 80, ¶ 43, 427 P.3d 288 (“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. . . . The part of [the expert witness’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.” (emphasis added) (quotation simplified)), cert. denied, 432 P.3d 1232 (Utah 2018); State v. Wright, 2013 UT App 142, ¶¶ 36–37, 304 P.3d 887 (determining that the admission of certain “quasi- statistical information” in the expert’s testimony was harmless “even assuming for purposes of appeal that [the testimony] . . . was inadmissible” (emphasis added)), cert. denied, 312 P.3d 619 (Utah 2013); Iorg, 801 P.2d at 941 (determining that a police officer’s testimony was not more probative than prejudicial when it was based on “her anecdotal statistical experience with late reporting in sexual abuse cases” (emphasis added) (quotation simplified)); see also State v. Francis, 2025 UT App 104, ¶ 82 (“[W]hile Utah appellate decisions have expressed hesitancy about allowing experts to rely on concrete statistics, we don’t understand there to be a rule categorically prohibiting every reference to any statistic.”).

¶23 And in State v. Nunez-Vasquez, 2020 UT App 98, 468 P.3d 585, cert. denied, 474 P.3d 945 (Utah 2020), this court affirmed the admission of the challenged expert testimony, distinguishing from certain prior cases—some of which concerned statistical expert testimony—by explaining, “In those cases, the problem with the testimony was that it focused directly on the veracity of another witness’s testimony.” Id. ¶ 66. The Nunez-Vasquez court determined that the testimony challenged in that case, however, was permissible “because [the expert] did not opine as to the

20221005-CA 12 2025 UT App 106 State v. Ngoy

veracity of [the victim’s] testimony.” Id. Thus, Nunez-Vasquez also fails to support any assertion that statistical evidence is per se not allowed.

¶24 In sum, we determine that Garcia-Cardiel is directly on point, that it does not contradict prior Utah precedent, and that it controls our decision here. We accordingly decline Ngoy’s invitation to overturn it. Based on Garcia-Cardiel and as explained above, we conclude that Expert’s probability testimony was admissible and that, therefore, Ngoy’s plain error argument fails.

CONCLUSION ¶25 Because the State presented some evidence from which a reasonable jury could determine that Ngoy committed the charged crimes, we must uphold the trial court’s decision denying the directed verdict motion. Additionally, the court did not plainly err in allowing the State’s expert witness to share probability testimony. Affirmed.

20221005-CA 13 2025 UT App 106