Revictimization Statistics May Be “Educational,” and Harsher Resentencing by a Different Judge Requires Proof of Actual Vindictiveness

1. Introduction

Case: State v. A. Twardoski, 2026 MT 142 (Mont. July 7, 2026).
Court: Supreme Court of Montana.
Parties: The State of Montana (Appellee) vs. Alan Peter Twardoski (Appellant).
Posture: Appeal from an amended judgment entered after retrial and resentencing on remand.

This decision arises from a retrial following the Court’s earlier reversal in State v. Twardoski, 2021 MT 179. After the second trial, Twardoski was again convicted of three counts of Sexual Intercourse Without Consent and one count of felony Sexual Assault. He raised two principal claims on appeal:

  1. Ineffective assistance of counsel (IAC): trial counsel did not object to the State’s expert testimony referencing a study suggesting previously abused children were “around … 47 percent more likely to be abused” again.
  2. Due process at sentencing: the resentencing judge (different from the first) imposed a more severe sentence by adding a 25-year parole restriction; Twardoski claimed judicial vindictiveness for his successful appeal.

The Court affirmed. It declined to decide the IAC claim on direct appeal (leaving it for postconviction litigation) and held that Twardoski did not prove actual judicial vindictiveness in the increased sentence imposed by a different judge.

2. Summary of the Opinion

Issue 1 — Ineffective assistance: failure to object to statistical expert testimony

  • The Court held the record did not establish “why” counsel failed to object and that plausible tactical reasons existed.
  • Because the statistical statement concerned revictimization risk rather than the frequency of false reporting, the testimony was not “clearly inadmissible.”
  • Accordingly, the Court declined to review IAC on direct appeal and indicated the claim is better suited to postconviction proceedings.

Issue 2 — Due process: harsher sentence after retrial

  • The Court recognized the North Carolina v. Pearce framework but emphasized that, under Alabama v. Smith, the “presumption of vindictiveness” applies only when a reasonable likelihood of actual vindictiveness exists.
  • Because a different judge resentenced Twardoski, the Pearce presumption did not apply (relying on Texas v. McCullough and State v. Forsyth).
  • The burden therefore fell on Twardoski to prove actual vindictiveness; the Court held he did not meet that burden given the sentencing court’s articulated, nonvindictive reasons for the 25-year parole restriction.

Separate writing

Justice Bidegaray concurred on Issue 1 but dissented on Issue 2, arguing the sentencing judge’s explicit reference to “twenty-four individual people” (counting both juries, including the first jury from the constitutionally infirm trial) required vacating the 25-year parole restriction and remanding for resentencing before a different judge.

3. Analysis

A. Precedents Cited

1) Ineffective assistance and when IAC is reviewable on direct appeal

  • Strickland v. Washington — establishes the two-prong IAC framework: deficient performance and resulting prejudice. The Court reiterated Strickland as the governing test, as applied in Montana decisions.
  • State v. Quiroz, 2022 MT 18 — confirms the constitutional sources of the right to effective counsel and applies Strickland. It also appears in the Court’s discussion of impermissible credibility/vouching testimony in sexual-assault contexts.
  • State v. Secrease, 2021 MT 212 — emphasizes the threshold “direct appeal vs. postconviction” sorting question, and the general need for a record that explains counsel’s conduct.
  • State v. Mikesell, 2021 MT 288 — articulates the operative screening rule: direct appeal review is appropriate if the record explains “why” counsel acted, or if there is “no plausible justification.” The Court treated Mikesell as the key gatekeeper to deciding whether the IAC claim could be resolved on the existing appellate record.
  • State v. Sinz, 2021 MT 163 (quoting State v. Hinshaw, 2018 MT 49) — underscores that omissions (like failing to object) are often ill-suited to direct appeal because the record rarely captures counsel’s tactical rationale. Sinz also supplies the admissibility framework for “educational” expert testimony in child sexual abuse cases.

2) Expert testimony, vouching, and the “statistical testimony” line

  • State v. Brodniak (partially overruled on other grounds by State v. Van Kirk) — establishes the Montana rule that expert testimony about the frequency of false accusations in rape cases is inadmissible as improper commentary on a witness’s credibility.
  • State v. Reams, 2020 MT 326 — draws the distinction between admissible educational testimony on general causes of false reports and inadmissible statistical testimony about false reporting.
  • State v. Grimshaw, 2020 MT 201 (Grimshaw I) — explains why false-reporting statistics “vouch” for a victim (inviting the inference that the victim is truthful because false reports are rare), and labels that inference improper.
  • State v. Harris — an example of impermissible credibility bolstering (expert calling a victim “trustworthy”), used to contrast permissible explanatory testimony with improper vouching.
  • M. R. Evid. 702 — anchors the admissibility rationale for expert testimony that helps a jury understand matters outside common experience, including dynamics of child sexual abuse.
  • Clausell v. State — supports the proposition that counsel may refrain from objecting for tactical reasons, such as not highlighting harmful testimony.

3) Due process and vindictive sentencing after retrial

  • North Carolina v. Pearce — provides the foundational due process principle: vindictiveness for successfully attacking a conviction “must play no part” in resentencing; also announces the original prophylactic presumption when the same judge increases a sentence after retrial.
  • Alabama v. Smith — narrows Pearce: the presumption applies only where there is a “reasonable likelihood” of actual vindictiveness; otherwise, the defendant must prove actual vindictiveness.
  • Texas v. McCullough and State v. Forsyth — stand for the proposition that the Pearce presumption generally does not apply when a different judge resentences the defendant. The majority leaned on these cases to place Twardoski in the “actual vindictiveness” regime.
  • Wasman v. United States — cited in the Smith line for the allocation of burdens when the presumption does not apply.
  • State v. Redfern, 2004 MT 277 — Montana’s articulation of Pearce’s due process rule against penalizing a defendant for successfully attacking the original conviction.
  • State v. Knowles, 2010 MT 186 (quoting United States v. Goodwin) — provides the broader due process principle that punishment for exercising rights is constitutionally intolerable.
  • State v. Roundstone, 2011 MT 227 — an “actual vindictiveness” framework in the prosecutorial context used by analogy to frame what “objective proof” of punitive motivation looks like.
  • State v. Grimshaw, 2025 MT 250 (Grimshaw II) — discussed to distinguish cases where the Pearce presumption applied (same judge resentencing) and where explicit references to retraumatizing the victim supported vindictiveness concerns.
  • United States v. Valdez-Lopez, Macomber v. Hannigan — relied upon to reject the idea that a harsher resentencing by a different judge requires “new evidence” or changed circumstances; rather, the second sentencer must supply a logical, nonvindictive explanation.
  • State v. Brown — a comparative example from another jurisdiction addressing when resentencing can be deemed actually vindictive.
  • State v. Bullplume — used to illustrate Montana’s approach in resolving vindictiveness claims (including via comparisons of sentence severity).
  • State v. Santoro, 2024 MT 136 (quoting State v. Webb, 2005 MT 5) — cited for the general proposition that due process applies at sentencing.
  • State v. Keefe, 2021 MT 8 — supplies the de novo review standard for constitutional sentencing issues.

B. Legal Reasoning

1) The Court’s treatment of the “47% more likely” revictimization statistic

The Court centered admissibility on the distinction Montana law draws between: (a) statistics that invite a jury to infer a complainant is truthful because false reports are rare (impermissible vouching), and (b) expert testimony that educates jurors on dynamics of abuse they may not understand (permissible).

Here, the expert’s statistic was not about false reporting rates; it was about the increased vulnerability of previously abused children to further abuse. On the majority’s view, that statistic helped the jury evaluate a specific factual feature of the defense theory—that I.A. could not plausibly have been abused by two different men in close succession—without telling the jury that I.A. was likely telling the truth.

That conclusion drove two downstream holdings:

  • Not “clearly inadmissible”: because the testimony did not fit Montana’s prohibited category (false-reporting frequency/vouching), a sustained objection was not a given.
  • Plausible tactical reasons exist: the Court identified multiple strategic explanations consistent with State v. Mikesell—including avoiding emphasis (Clausell v. State) and using the testimony to suggest that “more often than not” repeated abuse involves a different abuser (a point defense counsel probed on cross).

With plausible explanations on the table and no record explaining counsel’s true motive, the Court treated the IAC claim as a classic postconviction issue, not a direct appeal issue.

2) Actual vindictiveness when a different judge imposes a harsher sentence

The sentencing holding turns on the Pearce/Smith architecture:

  • No Pearce presumption: Twardoski conceded that resentencing by a different judge prevents the Pearce presumption from applying (State v. Forsyth; Texas v. McCullough).
  • Defendant must prove actual vindictiveness: under Alabama v. Smith (citing Wasman v. United States), Twardoski needed to show the harsher sentence was intended to punish him for exercising his right to appeal and obtain a new trial.

The core factual dispute was how to interpret the resentencing judge’s remarks referencing both juries (“twenty-four individual people”). The majority acknowledged those comments were “improvidently offered” and could be read, in isolation, as problematic. But it held that the comments did not carry Twardoski’s burden because the sentencing court supplied multiple legitimate reasons for the parole restriction (lack of accountability, evaluator findings of low motivation for treatment, severity and harm, and a concern that treatment completion could be merely “check[ing] the box”).

Critically, the majority rejected the argument that the harsher sentence was suspect because there was no “new evidence” after remand: relying on Texas v. McCullough, United States v. Valdez-Lopez, and Macomber v. Hannigan, it held that due process does not require the second sentencer to justify a harsher sentence only with information unavailable to the first sentencer. The touchstone is whether the record shows a logical, nonvindictive rationale and whether the defendant can prove punitive motive.

C. Impact

1) Expert testimony: revictimization statistics occupy a safer lane than false-reporting statistics

This opinion clarifies an important boundary in Montana’s “statistics in sexual-abuse trials” doctrine. The Court reaffirmed the strict prohibition on statistics that function as credibility vouching (the Brodniak/Grimshaw I line), while signaling that at least some quantitative statements about abuse dynamics (like revictimization vulnerability) may be treated as “educational” under State v. Sinz and M. R. Evid. 702.

Practically, litigants can expect future admissibility fights to focus on:

  • whether a statistic invites a “probability the victim is telling the truth” inference (impermissible), or
  • whether it explains behavior or context without directing a credibility conclusion (potentially permissible).

2) IAC litigation: failures to object remain difficult to win on direct appeal

The decision reinforces that “failure to object” IAC claims commonly belong in postconviction proceedings, where counsel’s reasons can be developed. Even when the challenged testimony contains a numerical claim, the Court will look for “plausible justification” and will not presume deficiency if admissibility is debatable.

3) Resentencing after retrial: the “actual vindictiveness” burden is heavy without the Pearce presumption

For Montana resentencing practice, the opinion highlights three consequences:

  • When a different judge resentences, defendants generally will not get the Pearce presumption and must prove actual vindictiveness.
  • A harsher sentence may be upheld even without new evidence, so long as the resentencing judge articulates logical, nonvindictive reasons on the record.
  • The dissent flags a litigation roadmap: explicit reliance on a prior reversed conviction (or constitutionally infirm proceedings) may be argued as objective proof of impermissible motive—an argument the majority rejected on this record but did not foreclose categorically.

4. Complex Concepts Simplified

Ineffective assistance of counsel (IAC) and the “Strickland test”

  • Deficient performance: Was the lawyer’s conduct objectively unreasonable under professional norms?
  • Prejudice: Is there a reasonable probability the outcome would have been different without the error?

On direct appeal, courts often cannot answer the first question because the trial record rarely reveals the attorney’s strategy. Postconviction proceedings can add evidence (e.g., attorney testimony) about why objections were or were not made.

“Vouching” vs. “educational” expert testimony

  • Vouching (not allowed): An expert effectively tells the jury the complainant is likely telling the truth—often by using statistics about how rarely people lie about sexual assault.
  • Educational testimony (often allowed): An expert explains dynamics of abuse (grooming, vulnerability, delayed disclosure) to help jurors understand evidence without instructing them how to decide credibility.

Pearce “presumption of vindictiveness” vs. “actual vindictiveness”

  • Pearce presumption: In certain settings (classically, the same judge increases a sentence after retrial), the law may presume vindictiveness unless rebutted.
  • Actual vindictiveness: When the presumption does not apply (as with a different resentencing judge), the defendant must prove the harsher sentence was imposed to punish the defendant for exercising appeal/trial rights.

5. Conclusion

State v. Twardoski (2026 MT 142) makes two notable contributions to Montana criminal procedure:

  1. It treats a revictimization-vulnerability statistic (“47 percent more likely to be abused”) as not “clearly inadmissible” and thus insufficient, on this record, to support direct-appeal IAC review—reinforcing that many failure-to-object claims belong in postconviction proceedings.
  2. It clarifies resentencing doctrine when a different judge imposes a harsher sentence after retrial: the Pearce presumption does not apply, no “new evidence” requirement exists, and the defendant must prove actual vindictiveness—an onerous burden when the resentencing court articulates multiple nonvindictive, record-based reasons.

The dissent underscores an unresolved fault line: whether, and when, explicit reliance on a prior reversed (and here, constitutionally infirm) verdict should be deemed objective proof of vindictiveness. The majority’s holding indicates that, at minimum, such remarks will be evaluated in context and may be outweighed by robust, nonvindictive sentencing explanations.