STATE v. TWARDOSKI (2026)

¶1 Alan Peter Twardoski appeals the October 26, 2022 Amended Judgment and Commitment Order of the Twenty-First Judicial District Court, Ravalli County, following his felony convictions for Sexual Intercourse Without Consent and Sexual Assault, in violation of §§ 45-5-503 and -502, MCA. We restate and address the following issues:

Issue 1: Whether Twardoski received ineffective assistance of counsel when his attorney failed to object to statistical expert testimony.

Issue 2: Whether the District Court's imposition of a more severe sentence after retrial on remand violated Twardoski's right to due process.

¶2 We affirm.

FACTUAL AND PROCEDURAL BACKGROUND

¶3 In June 2018, a jury found Twardoski guilty of three counts of sexual intercourse without consent, one count of sexual assault, and one count of sexual abuse of children, all felonies. The District Court sentenced Twardoski to the following concurrent prison terms: 50 years, with none suspended, for each count of sexual intercourse without consent and 50 years, with 25 years suspended, for the counts of sexual assault and sexual abuse of children. The District Court designated Twardoski as a Tier II Sexual Offender and restricted Twardoski's eligibility for parole until he completed Phases 1 and 2 of Sexual Offender Treatment.

¶4 Twardoski appealed, and we reversed and remanded for a new trial. State v. Twardoski, 2021 MT 179, ¶¶ 37-38, 405 Mont. 43, 491 P.3d 711. We held that the District Court erred by interpreting Montana's rape shield statute as prohibiting Twardoski from presenting evidence that his accuser, I.A., had been sexually abused in a “unique and identical manner” by a different person two weeks before Twardoski's alleged abuse. Twardoski, ¶ 36. We held that barring evidence of a different abuser in those circumstances denied Twardoski's constitutional right to confront his accuser and present a complete defense. Twardoski, ¶ 36.

¶5 In May 2022, Twardoski's case was retried before a different judge. The State dropped the sexual abuse of children charge on remand but otherwise prosecuted Twardoski using the same theory of the case and presenting evidence similar to the first trial. I.A. testified that when she was thirteen years old, Twardoski used a game called “Truth or Dare” to sexually abuse her. Twardoski's defense was that I.A. fabricated the allegations by accusing Twardoski of the same abusive conduct perpetrated against her two weeks earlier by a different man, Cody Hill. Twardoski maintained that I.A. disliked him for supplying her mother with drugs and that I.A. hoped that she could protect her mother by removing Twardoski from her life.

¶6 During its case in chief, the State called Mary Pat Hansen, APRN, as an expert witness. Hansen is a pediatric nurse practitioner and the clinical supervisor at First STEP Resource Center at St. Patrick Hospital in Missoula. Hansen testified about the forensic interview and medical evaluation she conducted with I.A. after I.A. disclosed Twardoski's alleged sexual abuse. Based on her expertise, the State asked Hansen to explain, generally, the process of victim selection. Hansen testified that people who sexually abuse children may select a particular target based on the child's characteristics. These characteristics include access, the opportunity to be alone with the child, and the child's vulnerabilities. When asked about her experience with children who are victimized multiple times, Hansen explained that her office “frequently” sees children more than once. Hansen considered it “not actually that surprising,” because “if a kid is vulnerable for some reason and they're abused once, those vulnerabilities still exist, and so they may be victimized a second time.” The State asked Hansen to summarize any research that she had reviewed indicating a “victim's likelihood to be reoffended on by a different offender once they've been a victim of abuse.” Hansen responded:

The short summary is that the research shows that if you have been a victim of abuse you are way more vulnerable to being abused again than the kid who has never experienced abuse.

․

There was one summary study I looked at, and I don't have it with me, and I don't remember the name. But it's around -- I think it was 47 percent more likely to be abused. But that was a study that looked at over a thousand articles and tried to come to a number. Generally speaking, I think the take-away that I want you all to have is that if you've already experienced abuse once you are way more likely to be at risk for experiencing abuse a second time.

The State then questioned Hansen about the concept of “grooming” victims and concluded its direct examination.

¶7 Defense counsel did not object to Hansen's testimony. On cross-examination, counsel asked Hansen to distinguish between the frequency that children are abused multiple times by the same person and the frequency that children are abused by different people:

[DEFENSE COUNSEL:] Is there any -- Again, asking for numbers. Either they are coming back with the same abuser or they're coming back with a different abuser. Can you tell me how frequently or tell the jury how frequently it's the same abuser or a different abuser?

[HANSEN:] I don't know that. I think more often than not it's a different abuser, and I say that because we try not to interview children multiple times about the same incident. That said, occasionally we do get asked to reinterview a kid ․

Defense counsel concluded his cross-examination after this exchange.

¶8 The jury found Twardoski guilty of all charges in the Amended Information: three counts of sexual intercourse without consent and one count of felony sexual assault. Before sentencing, the District Court ordered a second psychosexual evaluation to be completed by a different evaluator than the one who conducted Twardoski's 2018 psychosexual evaluation. The District Court also received an updated Pre-Sentence Investigation Report (PSI) based on the PSI completed for Twardoski's 2018 sentencing.

¶9 At the 2022 sentencing hearing, the State initially recommended that the District Court sentence Twardoski as he was sentenced in 2018. The State took the position that Twardoski did not deserve the opportunity to return to the community due to Twardoski's “adamant denial of any responsibility,” his continued blame of I.A., and his “pretty clear statements” that he did not intend to accept treatment or engage in programming. The State recognized that Twardoski's previous sentence included suspended time for some of the charges, but it advocated that Twardoski not be released “in this lifetime” per I.A.’s wishes. The State recommended that Twardoski's parole eligibility be restricted until he completed Phase 1 and Phase 2 of Sex Offender Treatment.

¶10 Defense counsel proposed the same sentence as was pronounced after the first trial, including the treatment-based parole restriction. Counsel argued that there were no substantial changes between the first and second trial, as the recent evaluator recommended the same Tier II Sexual Offender designation. Counsel pointed to Twardoski's improved mental health and lack of drug use in prison and hoped that Twardoski would quickly complete the recommended sexual offender treatment. Twardoski addressed the court to express regret about how drugs had influenced his life. He stated that he would not refuse sexual offender treatment and that he would “do whatever it takes if there was suspended time or anything.”

¶11 After complimenting I.A.’s personal growth and poise during the second trial, the District Court remarked,

And I am concerned also. We have two sexual offender evaluations that say exactly the same thing. I understand, Mr. Twardoski, you're standing here today saying you will engage in treatment, but both of the evaluators found you were not motivated for any programming. A specific quote, you expressed no insight, guilt, or shame into your part in the rape of your victim. You continue to portray your victim as the sexual aggressor. Twenty-four individual people who knew nothing about you and nothing about her found you guilty. And I'm not sentencing you on the first trial. I'm sentencing you on the second. But we've got 24 people that heard those facts, and all of them agreed that it happened the way [I.A.] said it happened.

The District Court noted that the 2018 evaluator believed that Twardoski needed “rather aggressive methods of confrontation to address [his] culpability,” in addition to sexual offender treatment while in custody.

¶12 As to each count of sexual intercourse without consent and to the count of felony sexual assault, the District Court sentenced Twardoski to prison for 50 years, with no time suspended, with the sentences to run concurrently with each other. The District Court designated Twardoski as a Tier II Sexual Offender and required successful completion of Phases 1 and 2 of Sexual Offender Treatment for Twardoski to become eligible for parole. The District Court also placed a 25-year parole restriction, explaining:

I think you have the ability to check the box and make it through sexual offender treatment. So I am going to put a 25-year parole restriction. That effectively means you will die in prison. You have placed this victim with a life sentence, and I only think it's appropriate that the same occurs to you.

In its Amended Judgment and Commitment Order, the District Court identified both PSIs, the 2018 and 2022 psychosexual evaluations, the State's sentencing recommendations, Twardoski's behaviors, and the harm caused to I.A. as the reasons for the sentence.

STANDARDS OF REVIEW

¶13 Ineffective assistance of counsel claims are mixed questions of fact and law, which we review de novo. State v. Quiroz, 2022 MT 18, ¶ 21, 407 Mont. 263, 502 P.3d 166. Whether a district court violated a defendant's constitutional rights at sentencing is reviewed de novo. State v. Keefe, 2021 MT 8, ¶ 11, 403 Mont. 1, 478 P.3d 830.

DISCUSSION

¶14 Issue 1: Whether Twardoski received ineffective assistance of counsel when his attorney failed to object to statistical expert testimony.

¶15 Article II, Section 24, of the Montana Constitution and the Sixth and Fourteenth Amendments to the United States Constitution guarantee a criminal defendant's right to effective assistance of counsel. Quiroz, ¶ 22. We review ineffective assistance of counsel (IAC) claims by applying the two-pronged test established in Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L.Ed.2d 674 (1984). Quiroz, ¶ 23. Under Strickland, the defendant must demonstrate that (1) counsel's performance was deficient or fell below an objective standard of reasonableness and (2) the deficiency prejudiced his defense. State v. Secrease, 2021 MT 212, ¶ 13, 405 Mont. 229, 493 P.3d 335.

¶16 When a defendant claims IAC on direct appeal, we must first determine whether the claims are more appropriately addressed in a postconviction proceeding. Secrease, ¶ 14. We review IAC claims on direct appeal if the record explains “why” counsel acted or if there is no “plausible justification” for counsel's action or inaction. State v. Mikesell, 2021 MT 288, ¶¶ 20, 22, 406 Mont. 205, 498 P.3d 192. “Claims involving alleged omissions of trial counsel are often ill-suited for consideration on direct appeal.” State v. Sinz, 2021 MT 163, ¶ 22, 404 Mont. 498, 490 P.3d 97 (quoting State v. Hinshaw, 2018 MT 49, ¶ 21, 390 Mont. 372, 414 P.3d 271).

¶17 Twardoski argues that his trial counsel's performance was deficient by failing to object to Hansen's testimony that children who were previously abused are 47% more likely to be abused again than a child who has never experienced abuse, and that this inadmissible testimony prejudicially bolstered I.A.’s credibility. Twardoski contends that there is no plausible justification for counsel's failure to object to this testimony, making his IAC claim reviewable on direct appeal. The State responds that counsel's lack of objection can be plausibly explained as a tactical decision, as Hansen's testimony was not clearly inadmissible.

¶18 It is well settled that expert testimony about the “frequency of false accusations in sexual assault or rape cases” is inadmissible as an improper comment on the credibility of a complaining witness. Quiroz, ¶ 25; accord State v. Brodniak, 221 Mont. 212, 222, 718 P.2d 322, 329 (1986) (establishing rule), overruled in part on other grounds by State v. Van Kirk, 2001 MT 184, 306 Mont. 215, 32 P.3d 735. But we have also consistently recognized that expert testimony can help explain the complexities of child sexual abuse, which are often outside of the jury's common experience, and aid the jury in understanding and evaluating a victim's testimony. Sinz, ¶ 30 (citing M. R. Evid. 702). Educational testimony regarding “general causes of false reports” in child sexual abuse cases is admissible, but “statistical testimony regarding false reporting” is not. Sinz, ¶ 30 (quoting State v. Reams, 2020 MT 326, ¶ 16 n.1, 402 Mont. 366, 477 P.3d 1118). This distinction rests on the “clear inference” that can be made from the statistical rarity of false sexual assault reports to a high probability that the victim's accusations are true. See, e.g., State v. Grimshaw, 2020 MT 201, ¶ 24, 401 Mont. 27, 469 P.3d 702 (Grimshaw I). Unlike expert witness testimony that assists the jury in understanding a victim's behaviors and symptoms, statistical testimony about false reporting essentially “vouche[s]” for the credibility of the victim's accusations of sexual assault. Grimshaw I, ¶¶ 25, 27.

¶19 Twardoski asserts that the clear inference from Hansen's 47% statistic was that there was a statistical probability that I.A. was abused by both Twardoski and Hill, stamping her allegations with “scientific legitimacy” that improperly bolstered her credibility. See State v. Harris, 247 Mont. 405, 409-10, 808 P.2d 453, 455-56 (1991). But Hansen's statistical testimony did not describe the rarity of false reports of sexual assault or rape, unlike in Brodniak and its progeny. E.g., Brodniak, 221 Mont. at 222, 718 P.2d at 329; Quiroz, ¶ 25. Hansen did not impermissibly comment on whether I.A. was telling the truth, either directly or indirectly. See, e.g., Harris, 247 Mont. at 409-10, 808 P.2d at 455-56 (expert impermissibly testified that victim was an “honest, open country boy” and a “trustworthy child”); Grimshaw I, ¶ 24 (“The clear inference of the statistical testimony in this case was that there is a 92-98% probability that [Defendant] was guilty of the charged offense such that [the victim] was telling the truth as compared to a 2-8% probability [the victim] had made a false accusation.”). Hansen described the process of victimization and explained that children who experience sexual abuse are 47% more likely to be abused again compared to a child who has never been abused. This statistic did not vouch for I.A.’s credibility by suggesting that it was statistically likely that she was making a truthful accusation. See Grimshaw I, ¶ 24. Rather, the statistic permissibly aided the jury in understanding and evaluating I.A.’s testimony that she was sexually abused by two different men. See Sinz, ¶ 31 (holding that educational expert testimony explaining process of victimization was not improper or inadmissible).

¶20 As Hansen's statistical testimony was not clearly inadmissible, several tactical reasons could explain defense counsel's lack of objection. The State points out that Twardoski's counsel may have reasonably wanted to avoid calling “undue attention to the prosecution's case.” Clausell v. State, 2005 MT 33, ¶ 20, 326 Mont. 63, 106 P.3d 1175. The prosecution immediately moved on from this line of questioning and did not emphasize the 47% statistic. Defense counsel may have believed that the statistic would benefit Twardoski's defense, as it could be interpreted that for more than half of children with previous abuse histories, the likelihood of being abused again is virtually the same as the likelihood of a child being abused for the first time. In fact, defense counsel pursued this line of inquiry on cross-examination when he asked Hansen “for numbers” distinguishing how frequently children are abused multiple times by the same person or by different people. The jury could have reasonably inferred from this expert testimony that I.A. was more likely than not abused by a single person, which may have undermined her credibility and bolstered Twardoski's contention that I.A. fabricated her allegations against him by accusing him of the same abusive conduct first perpetrated against her by Hill. These are plausible justifications for defense counsel's lack of objection to Hansen's statistical testimony. See Mikesell, ¶ 22.

¶21 Without a complete record explaining defense counsel's actions, we can only speculate as to whether the lack of objection to Hansen's testimony was a tactical decision. See Mikesell, ¶¶ 21-22. Twardoski's IAC claim is better suited for consideration in postconviction proceedings, and we decline to review it on direct appeal. See Sinz, ¶ 26.

¶22 Issue 2: Whether the District Court's imposition of a more severe sentence after retrial on remand violated Twardoski's right to due process.

¶23 The Montana and U.S. constitutions guarantee the right to due process at sentencing. State v. Santoro, 2024 MT 136, ¶ 30, 417 Mont. 92, 551 P.3d 822 (quoting State v. Webb, 2005 MT 5, ¶ 18, 325 Mont. 317, 106 P.3d 521). Punishing a defendant for exercising a statutory or constitutional right is a due process violation “of the most basic sort.” State v. Knowles, 2010 MT 186, ¶ 28, 357 Mont. 272, 239 P.3d 129 (quoting United States v. Goodwin, 457 U.S. 368, 372, 102 S. Ct. 2485, 2488, 73 L.Ed.2d 74 (1982)). A court denies a defendant due process by imposing a heavier sentence as punishment for successfully attacking his original conviction. State v. Redfern, 2004 MT 277, ¶ 12, 323 Mont. 225, 99 P.3d 223 (citing North Carolina v. Pearce, 395 U.S. 711, 722-24, 89 S. Ct. 2072, 2079-80, 23 L.Ed.2d 656 (1969), overruled in part on other grounds by Alabama v. Smith, 490 U.S. 794, 109 S. Ct. 2201, 104 L.Ed.2d 865 (1989)). Such judicial vindictiveness “must play no part” in resentencing after a new trial. Pearce, 395 U.S. at 725, 89 S. Ct. at 2080.

¶24 To protect against judicial vindictiveness, the Pearce Court established a prophylactic requirement that “whenever a judge imposes a more severe sentence upon a defendant after a new trial, the reasonings for ․ doing so must affirmatively appear.” Pearce, 395 U.S. at 726, 89 S. Ct. at 2081. The U.S. Supreme Court has since restricted Pearce’s “presumption of vindictiveness” to apply only in circumstances where there is a reasonable likelihood that the sentencing authority's actual vindictiveness produced the increase in sentence. Smith, 490 U.S. at 799, 109 S. Ct. at 2204-05. Where the presumption does not apply, it is the defendant's burden to prove actual vindictiveness. Smith, 490 U.S. at 799-800, 109 S. Ct. at 2205 (citing Wasman v. United States, 468 U.S. 559, 569, 104 S. Ct. 3217, 3223, 82 L.Ed.2d 424 (1984)).

¶25 Twardoski concedes that because he was resentenced by a different judge upon retrial, the Pearce presumption of vindictiveness does not apply. State v. Forsyth, 233 Mont. 389, 422, 761 P.2d 363, 384 (1988), abrogated in part on other grounds by State v. Curtis, 241 Mont. 288, 787 P.2d 306 (1990); Texas v. McCullough, 475 U.S. 134, 140, 106 S. Ct. 976, 979-80, 89 L.Ed.2d 104 (1986). Twardoski argues that the second judge's comments at resentencing and imposition of an additional 25-year parole restriction without new evidence demonstrate actual vindictiveness. The State responds that the second judge did not punish Twardoski for exercising his constitutional rights because she provided an “on-the-record, wholly logical, nonvindictive reason” for the second sentence.¹ McCullough, 475 U.S. at 140, 106 S. Ct. at 980.

¶26 We have not previously evaluated a defendant's claim of actual vindictiveness by a second sentencing judge. See, e.g., Forsyth, 233 Mont. at 422, 761 P.2d at 384 (summarily holding that defendant made no showing of actual vindictiveness); State v. Bullplume, 2011 MT 40, ¶¶ 18-21, 359 Mont. 289, 251 P.3d 114 (resolving vindictiveness allegation on the basis that second sentence was not more burdensome than first); State v. Grimshaw, 2025 MT 250, ¶ 35, 424 Mont. 445, 578 P.3d 890 (Grimshaw II) (applying presumption of vindictiveness to resentencing by same judge). When evaluating actual vindictiveness, other jurisdictions assess whether the second sentence was intended to punish the defendant for exercising a right. See, e.g., United States v. Valdez-Lopez, 4 F.4th 886, 893-94 (9th Cir. 2021) (holding that the judge's comments in the second sentencing did not indicate punishment for defendant's decision to go to trial); State v. Brown, 309 Kan. 369, 435 P.3d 546, 552 (2019) (holding that actual vindictiveness motivated resentencing because 12-month increase was based solely on victims’ additional pain and suffering caused by appeal). This standard echoes our inquiry into actual prosecutorial vindictiveness, which requires “objective proof ‘that the prosecutor's charging decision was motivated by a desire to punish [the defendant] for doing something that the law plainly allowed him to do,’ or evidence of the prosecutor's bad faith or malicious actions.” State v. Roundstone, 2011 MT 227, ¶ 39, 362 Mont. 74, 261 P.3d 1009 (internal citations omitted).

¶27 Twardoski analogizes the District Court's comments to those determined to be vindictive in Grimshaw II, ¶¶ 36-37, and State v. Robledo, 282 Or.App. 96, 386 P.3d 136, 140 (Or. Ct. App. 2016). But unlike the present case, the Pearce presumption of vindictiveness applied in both Grimshaw II and Robledo because the defendant was resentenced by the same judge. Grimshaw II, ¶ 35; Robledo, 386 P.3d at 138, 140-41. The Grimshaw II and Robledo judges also stated a clearly vindictive reason for imposing an increased sentence: that the defendants’ appeals and retrials on remand required the victims to testify multiple times. Grimshaw II, ¶¶ 36-37; Robledo, 386 P.3d at 138, 140-41. To be sure, the District Court's comments that “[t]wenty-four individual people” found Twardoski guilty and agreed “that it happened the way [I.A.] said it happened” were, at best, improvidently offered. Out of context, they could at least suggest that the District Court resentenced Twardoski in part by relying on the first jury's verdict, which resulted from erroneous rulings that violated Twardoski's constitutional right to confront his accuser and present a complete defense. But unlike the clearly vindictive rationale for imposing increased sentences in Grimshaw II and Robledo, the second judge in this case did not tie the 25-year parole restriction to any additional burden put on I.A. or other witnesses by Twardoski's appeal or retrial.

¶28 The District Court's references during resentencing to the first jury do not demonstrate that it impermissibly relied on the unconstitutional first verdict or Twardoski's successful appeal, as the District Court provided ample legitimate justification for the 25-year parole restriction. See Valdez-Lopez, 4 F.4th at 893-94 (holding that second sentencing judge's comments about defendant's decision to go to trial did not indicate punishment for defendant's exercise of Sixth Amendment right to trial, as comments were not entire basis for sentence). The District Court's “[t]wenty-four ․ people” comments were only one part of its discussion of Twardoski's continued unwillingness to take responsibility for causing lifelong harm to I.A. The District Court observed that despite Twardoski's statement during the resentencing hearing that he would engage in treatment, the 2018 and 2022 psychosexual evaluations both found that Twardoski was not motivated to do so. The District Court elaborated that Twardoski “expressed no insight, guilt, or shame into [his] part in the rape of [his] victim,” that he “continue[d] to portray [his] victim as the sexual aggressor,” and that the 2018 psychosexual evaluation indicated Twardoski needed “rather aggressive methods of confrontation to address [his] culpability.” Before pronouncing the 25-year parole restriction, the District Court explained that it believed that Twardoski could “check the box and make it through sexual offender treatment.” The District Court explained that the 25-year parole restriction was appropriate to ensure that Twardoski received the same “life sentence” that he placed on I.A. by sexually abusing her. Considered in their entirety, the District Court's comments indicate that the 25-year parole restriction was motivated by a desire to punish Twardoski for the severity of his offenses and refusal to accept accountability, and to mitigate Twardoski's ability to circumvent the treatment-based parole restrictions. Given the ample explanations for the 25-year parole restriction—and absent a presumption of vindictiveness—Twardoski has failed to carry his burden of proving that the District Court's references to the first unconstitutional verdict indicate an intent to punish Twardoski for successfully appealing his first conviction.

¶29 Twardoski relies on McCullough to assert that the additional parole restriction was intended to punish Twardoski for appealing his original conviction, as the appeal was the only changed circumstance between his first and second sentencing hearings. As the State observes, McCullough interpreted Pearce as requiring a second sentencing judge to provide “no more” than an “on-the-record, wholly logical, nonvindictive reason for the sentence.” McCullough, 475 U.S. at 140, 106 S. Ct. at 980. In McCullough, the second sentencing judge was not determined to be vindictive because she relied on new evidence presented at the second trial to impose a harsher sentence following retrial. McCullough, 475 U.S. at 143-44, 106 S. Ct. at 981-82. But the Court did not require that a second sentencing judge's rationale be based on new evidence or changed circumstances between the first and second sentencings. See McCullough, 475 U.S. at 138-40, 106 S. Ct. at 979-80; Valdez-Lopez, 4 F.4th at 892 (“[T]he law does not require the second sentencer to offer reasons that were unavailable to the first sentencer.”); accord Macomber v. Hannigan, 15 F.3d 155, 157 (10th Cir. 1994) (“[I]t is not necessary that the second sentencing judge rely on and provide facts not available at the time of the first sentence to support the more severe sentence.”).

¶30 The District Court provided several on-the-record, wholly logical, nonvindictive reasons for Twardoski's 25-year parole restriction based on the second sentencing hearing, the 2018 and 2022 PSIs, and the 2018 and 2022 psychosexual evaluations. See McCullough, 475 U.S. at 140, 106 S. Ct. at 980. During the second sentencing hearing, the State took the position that Twardoski did not deserve the opportunity to return to the community because of his continued denial of responsibility, blaming the victim, and his stated intent to decline treatment. The State also communicated I.A.’s wishes that Twardoski not be released “in this lifetime.” Both PSIs reported that Twardoski's actions would have a lifelong impact on the victim. The District Court reasoned that “[v]ery little has changed” with Twardoski since the first sentencing judge wrote, “The [c]ourt believes Defendant is a danger to society and likely can never be rehabilitated.” The District Court noted that the second psychosexual evaluation by a provider of Twardoski's choosing indicated that he was “not motivated to engage in any programming or rehabilitative training.” The District Court concluded that “Defendant's utter lack of accountability and willingness to victimize a vulnerable girl 48 years his junior requires that the only appropriate sentence is one which results in the Defendant spending the rest of his life in prison.” These nonvindictive reasons, plus the District Court's concern that Twardoski could circumvent the treatment-based parole restrictions by “check[ing] the box,” adequately explain the second sentencing judge's imposition of a 25-year parole restriction.

¶31 Twardoski has failed to carry his burden of proving that actual vindictiveness motivated the District Court's increased sentence. Smith, 490 U.S. at 799-800, 109 S. Ct. at 2205.

CONCLUSION

¶32 We decline to review Twardoski's IAC claim on direct appeal. Twardoski has not proved that his second sentence was the product of actual judicial vindictiveness. We affirm his conviction and sentence.

¶33 I concur in the Court's resolution of Issue One because the Court declines to decide Twardoski's ineffective-assistance claim on direct appeal. I would make explicit that our disposition is without prejudice to Twardoski's ability to raise the claim, if otherwise timely and properly pleaded, in postconviction proceedings. I respectfully dissent from Issue Two. The District Court imposed a concededly more severe sentence after Twardoski successfully appealed a constitutionally infirm conviction, after both parties recommended the original sentence, after the State dismissed one charge, and after the court acknowledged that the relevant sentencing information had not materially changed. Because the court expressly relied on the fact that “twenty-four individual people” had found Twardoski guilty—including the first jury whose verdict followed a trial this Court held unconstitutional—I would vacate the 25-year parole restriction and remand for resentencing before a different judge.

¶34 I agree with the Court that Twardoski's ineffective-assistance claim is better suited for postconviction proceedings. The record shows counsel did not object to Hansen's testimony, but it does not explain why counsel remained silent. Claims alleging trial counsel's failure to object often require a developed record explaining counsel's reason for action or inaction. State v. Mikesell, 2021 MT 288, ¶¶ 20-22, 406 Mont. 205, 498 P.3d 192; State v. Sinz, 2021 MT 163, ¶ 22, 404 Mont. 498, 490 P.3d 97. The distinction between impermissible statistical credibility testimony and permissible educational testimony also complicates the direct-appeal inquiry. See State v. Quiroz, 2022 MT 18, ¶¶ 24-31, 407 Mont. 263, 502 P.3d 166. I therefore would not decide on this record whether counsel performed deficiently or whether prejudice resulted.

¶35 I part company with the Court on sentencing. Due process requires that “vindictiveness against a defendant for having successfully attacked his first conviction must play no part” in the sentence imposed after retrial. North Carolina v. Pearce, 395 U.S. 711, 725, 89 S. Ct. 2072, 2080, 23 L.Ed.2d 656 (1969), overruled in part on other grounds by Alabama v. Smith, 490 U.S. 794, 109 S. Ct. 2201, 104 L.Ed.2d 865 (1989). Montana law recognizes the same constitutional limit. A court denies due process when it imposes a heavier sentence to punish a defendant for setting aside the original conviction. State v. Jackson, 2007 MT 186, ¶ 14, 338 Mont. 344, 165 P.3d 321.

¶36 The Court correctly concludes that the Pearce presumption does not apply because a different judge resentenced Twardoski after retrial. Texas v. McCullough, 475 U.S. 134, 140, 106 S. Ct. 976, 979-80, 89 L.Ed.2d 104 (1986); State v. Forsyth, 233 Mont. 389, 422, 761 P.2d 363, 384 (1988). But that conclusion does not end the inquiry. When the presumption does not apply, a defendant still may obtain relief by showing actual vindictiveness. Smith, 490 U.S. at 799-800, 109 S. Ct. at 2204-05; McCullough, 475 U.S. at 138, 106 S. Ct. at 978-79. I do not dispute that a different sentencing judge may independently evaluate the record and impose a harsher lawful sentence. The due process violation here is narrower: the court's own explanation made the constitutionally invalid first verdict part of the rationale for making the sentence more severe by adding the 25-year parole restriction. In my view, the objective record makes that showing here.

¶37 The State does not dispute that the added 25-year parole restriction made Twardoski's second sentence more severe. That concession is correct. A sentence may become more burdensome even when the term of years remains the same. Jackson, ¶ 15. Before remand, Twardoski's parole eligibility depended on successful completion of Phases 1 and 2 of sex-offender treatment. After remand, the District Court added a 25-year parole restriction and told Twardoski that the restriction “effectively means you will die in prison.”

¶38 Although a different sentencer is not categorically required to identify new information, the absence of any materially changed sentencing information matters here because the sentencing court itself supplied an impermissible explanation for the increase. The State asked the District Court to impose the same sentence Twardoski received after the first trial, stating that the new psychosexual evaluation “really mirrors in most ways” the 2018 evaluation. Defense counsel likewise asked the court to impose “the same sentence that was imposed the first time around,” with “the only parole restriction being that he needs to complete SOP 1 and 2” before parole eligibility. The District Court acknowledged that “we have two sexual offender evaluations that say exactly the same thing.” The State also dismissed the sexual abuse of children charge before retrial. Thus, Twardoski stood before the second sentencing judge convicted of fewer charges, facing materially unchanged sentencing information, and supported by a joint recommendation for the original parole restriction.

¶39 Against that record, the District Court stated:

Twenty-four individual people who knew nothing about you and nothing about her found you guilty. And I'm not sentencing you on the first trial. I'm sentencing you on the second. But we've got 24 people that heard those facts, and all of them agreed that it happened the way I.A. said it happened.

The sequence matters. The court did not merely note that the second jury convicted Twardoski. It counted both juries. It invoked the combined force of “twenty-four individual people.” It did so even though this Court reversed the first conviction because the first trial violated Twardoski's constitutional right to confront his accuser and present a complete defense. State v. Twardoski, 2021 MT 179, ¶ 36, 405 Mont. 43, 491 P.3d 711.

¶40 The majority acknowledges that the “24 people” comment was, “at best, improvidently offered,” and that it could suggest reliance on the first jury's verdict. I would not minimize the constitutional significance of the remark, view it in isolation, or take it out of context. In context, the remark was part of the court's explanation for why Twardoski's denial of responsibility justified a parole restriction. The court first stated that both psychosexual evaluators found him unmotivated for treatment and that he expressed “no insight, guilt, or shame.” It then counted “twenty-four individual people” who found him guilty. The sequence shows that the court used both verdicts as proof that Twardoski's continued denial was aggravating. But one of those verdicts—the first one—resulted from a trial this Court held unconstitutional because that jury reached its verdict without evidence this Court later held Twardoski had a constitutional right to present. Twardoski, ¶ 36. A constitutionally invalid verdict may not be used as part of the premise for increasing punishment after a successful appeal. Using a constitutionally infirm conviction to establish the premise for a harsher sentence is incompatible with due process. Due process does not permit a sentencing court to rely on a reversed, unconstitutional conviction to justify a harsher penalty; it protects the right to appeal only if a defendant can exercise that right without facing a sentencing penalty for having obtained reversal.

¶41 The majority gives dispositive weight to the District Court's other stated reasons: the seriousness of the offenses, the harm to I.A., Twardoski's lack of accountability, his treatment needs, and the court's concern that he could “check the box” by completing treatment. Those are legitimate sentencing considerations. But legitimate considerations do not cleanse an impermissible one. Pearce requires that vindictiveness for a successful appeal “must play no part” in the sentence imposed after retrial. The question is not whether the court could have imposed a severe sentence in the first instance. The question is whether the increased parole restriction after remand rested, in part, on Twardoski's successful appeal and retrial. The District Court's own words show that it did not. A sentence may be supported by many lawful considerations, but it cannot stand if the sentencing court also relied on a constitutionally impermissible one that increased the punishment.

¶42 The District Court's “check the box” concern does not cure the constitutional problem. The first sentence already addressed treatment by conditioning parole eligibility on successful completion of Phases 1 and 2 of sex-offender treatment. If the court believed a treatment-completion condition was inadequate because Twardoski could complete treatment without genuine accountability, that concern may have supported a parole restriction in the first instance. But the court did not explain why the same evaluations and same treatment concern justified adding 25 years only after Twardoski successfully appealed and was retried. The court's own explanation supplies the missing link: it counted both verdicts.

¶43 I do not question the sentencing court's authority to consider Twardoski's denial of responsibility, treatment needs, danger to the public, or the harm caused to I.A. Those considerations may support a severe lawful sentence. They do not, however, explain why the court added a 25-year parole restriction after a successful appeal when the State recommended the original sentence, the defense recommended the original sentence, the State had dismissed one charge, and the court recognized that the two evaluations “say exactly the same thing.” Those objective facts reinforce why the District Court's express reliance on both verdicts matters.

¶44 The majority reads McCullough too broadly. McCullough holds that the Pearce presumption does not apply when a different judge imposes the second sentence and the record provides an on-the-record, logical, non-vindictive reason for the increase. The majority does not give adequate weight to the facts that made the increased sentence permissible in McCullough. The second sentencing judge in McCullough relied on new objective evidence presented at the second trial that had not been presented at the first and that showed the defendant's greater role in the killing. McCullough, 475 U.S. at 143-44, 106 S. Ct. at 981-82. Here, the record contains no comparable new objective aggravating information. The psychosexual evaluations “say exactly the same thing.” McCullough removes the presumption here; it does not insulate an increased sentence from actual-vindictiveness review when the sentencing court expressly relied on a constitutionally invalid first verdict. On this record, that combination is sufficient to establish actual vindictiveness. I do not suggest the sentencing judge acted from personal animus. Actual vindictiveness in this context is an objective due process inquiry: whether the increased sentence was based, in whole or in part, on Twardoski's successful exercise of appellate and trial rights.

¶45 The majority's reliance on United States v. Valdez-Lopez, 4 F.4th 886 (9th Cir. 2021), and Macomber v. Hannigan, 15 F.3d 155 (10th Cir. 1994), does not answer the due process problem here. Those cases stand for the proposition that, absent a presumption of vindictiveness, a second sentencer need not identify facts unavailable to the first sentencer before imposing a harsher sentence. I agree with that proposition. But neither case permits a sentencing judge to count a constitutionally invalid verdict as part of the reason for increasing punishment. Indeed, Valdez-Lopez recognizes that, if the stated reason is vindictive, “there is no need for a presumption”; the defendant may show actual vindictiveness. Valdez-Lopez, 4 F.4th at 892 (citing Smith, 490 U.S. at 799-800, 109 S. Ct. at 2204-05). Valdez-Lopez therefore confirms the narrower point relevant here: the absence of a presumption does not foreclose relief when the sentencing court's own stated reason shows actual vindictiveness. That is the due process problem I see in this record.

¶46 The Court also distinguishes State v. Grimshaw, 2025 MT 250, 424 Mont. 445, 578 P.3d 890 (Grimshaw II), and State v. Robledo, 282 Or.App. 96, 386 P.3d 136 (2016), too narrowly. I do not rely on Grimshaw II or Robledo to import a same-judge presumption into this different-judge case. I rely on them for the more basic proposition that a sentencing increase after a successful appeal cannot rest on the defendant's having obtained the appeal and forced the State to prove its case again. Grimshaw II, ¶¶ 31-36; Robledo, 386 P.3d at 140-41. The District Court did not say that it punished Twardoski because I.A. had to testify again. But the court did rely on the fact that two juries had convicted him. By explicitly invoking the combined weight of two juries immediately before imposing a 25-year parole restriction, the District Court made Twardoski's successful appeal and retrial part of the rationale for increasing his sentence.

¶47 This is a narrow dissent. I would not disturb the jury's verdicts. I would not decide the ineffective-assistance claim on direct appeal. I would not hold that every harsher sentence imposed by a different judge after retrial requires new evidence or changed circumstances. I would hold only that, when a sentencing court adds a 25-year parole restriction after a successful appeal and expressly counts a constitutionally invalid first verdict as part of its explanation for doing so, due process requires resentencing.

¶48 I would vacate the 25-year parole restriction and remand for resentencing before a different judge on that restriction. That remedy leaves intact the convictions, the 50-year concurrent prison terms, the Tier II designation, and the treatment-based parole condition. I therefore concur in part and dissent in part.

FOOTNOTES

1.   The State does not dispute that the additional 25-year parole restriction made Twardoski's second sentence more severe than his first.

Justice James Jeremiah Shea delivered the Opinion of the Court.

We Concur: CORY J. SWANSON, C.J. BETH BAKER, J. JIM RICE, J.