“From Vouching to Variables” – State v. Molde Recognizes Purely Statistical Evidence on False Reporting as Permissible Expert Testimony

1. Introduction

State v. Jobert L. Molde, 2025 WI 21, presented the Wisconsin Supreme Court with a narrow yet recurring evidentiary problem: whether an expert’s statement that approximately one percent of child-sexual-assault (CSA) disclosures are fabricated constitutes impermissible vouching for a specific complainant’s credibility under the Haseltine rule.

The case arose from allegations that the defendant, Jobert Molde, sexually assaulted his daughter “Lauren” when she was eight or nine. At Molde’s jury trial Dr. Alice Swenson—who had supervised Lauren’s forensic interview—answered a juror’s question about false reporting rates by citing the “one-percent” statistic. Defense counsel lodged no objection, Molde was convicted, and on post-conviction review claimed ineffective assistance of counsel (IAC) for failing to object under State v. Haseltine, 120 Wis. 2d 92 (1984). The court of appeals reversed, relying on its own decision in State v. Mader, 2023 WI App 35, holding that near-certain statistics (99% truthful) always cross the vouching line.

In a unanimous opinion authored by Justice Hagedorn—and accompanied by a forceful concurrence from Justice Karofsky—the Supreme Court reversed the court of appeals, created a new evidentiary distinction, and clarified both the contours of the Haseltine rule and the scope of counsel’s duty to object. The Court expressly overruled Mader and kindred cases to the extent they barred statistical testimony per se.

2. Summary of the Judgment

Holding. “Statistical evidence alone on the likelihood of false reports does not violate the Haseltine rule.” Expert testimony is improper only when it makes (or necessarily implies) a direct assessment of the particular witness’s credibility.
Application. Dr. Swenson’s 1%-false-disclosure figure was generalized, did not reference Lauren’s veracity, and was therefore admissible. Counsel’s failure to object was not deficient; Molde’s IAC claim fails.
Precedent Overruled. The Court overrules Mader and any other decisions treating high-confidence statistics as intrinsic vouching.
Case Result. Court of Appeals decision reversed; Molde’s conviction stands.

3. Detailed Analysis

3.1 Precedents Cited & Their Influence

The majority walks through forty years of Wisconsin jurisprudence on “vouching.” Key cases include:

State v. Haseltine (1984) – The foundational rule: no witness may opine that another witness is telling the truth.
State v. Robinson & State v. Jensen (1988) – Upheld expert testimony describing typical victim behaviors, so long as experts stopped short of endorsing a specific complainant.
State v. Romero (1988), State v. Kleser (2010) – Found impermissible explicit or implicit endorsements of credibility.
State v. Maday (2017) – Clarified that testimony about indicia of coaching/honesty observed is permissible if it does not pronounce ultimate veracity.
State v. Mader (Ct. App. 2023) – Had treated 99% truth-rate testimony as vouching; now expressly overruled.

These authorities illustrate a spectrum: from general educational information (permissible) to applying conclusion to the witness (impermissible). The majority situates Dr. Swenson’s remarks on the former side; it was “precisely the kind of generalized evidence that might assist the jury.”

3.2 The Court’s Legal Reasoning

“Assist, not supplant.” Under Wis. Stat. §907.02 an expert must “assist the trier of fact.” Assistance becomes usurpation only when the expert ties her opinion directly to the complainant’s truthfulness.
Statistics ≠ Credibility Verdict. Pure statistics merely add a probabilistic datum; they do not allocate the complainant to the “true” category. The jury remains free to weigh the stat’s reliability, the methodology, and all other evidence.
Symmetry Principle. If defense could introduce statistics showing high false-reporting rates, the State should likewise be able to present statistics showing low rates. Evidentiary doors swing both ways; admissibility cannot depend on which side benefits.
Rule 904.03 & Daubert Gatekeeping. Reliability, potential prejudice, and methodological rigor still govern admission. The Court simply removes the categorical bar formerly erected by lower-court precedent.
IAC Analysis. Because an objection would have lacked merit, counsel’s performance was not deficient under Strickland.

3.3 Impact on Future Litigation

Evidentiary Landscape. Attorneys may now elicit well-founded statistics on false-reporting rates without fearing automatic reversal.
Expert Preparation. Experts must still avoid the “extra step” of applying group statistics to an individual witness (“therefore this child is truthful”).
Trial Strategy. Litigants opposing such testimony must shift from Haseltine objections to Daubert, §904.03, or cross-examination on sampling bias, false-negative rates, or differing study definitions of “false.”
Appellate Guidance. Courts of appeals gain a bright-line rule: numbers alone do not equal vouching. This fosters uniformity and reduces reversals premised on misapplied Haseltine doctrine.
Child-Witness Jurisprudence. The decision dovetails with national trends permitting “CSA accommodation evidence” to educate juries on delayed disclosure, memory fragmentation, and rarity of fabrication.

4. Complex Concepts Simplified

Haseltine Rule. A ban on any witness (expert or lay) telling the jury, “Believe this other witness, they are truthful.” It protects the jury’s exclusive role in judging credibility.
Vouching vs. Education. • Vouching = Saying this child is truthful. • Education = Explaining general patterns (“children often delay disclosure” or “false reports are rare”).
Statistical Testimony. Descriptive, group-level data derived from peer-reviewed studies. It is admissible if methodologically reliable and relevant.
Daubert Standard (Wis. Stat. §907.02). Gatekeeping test requiring that expert opinions rest on reliable methodology applied to the facts.
Ineffective Assistance of Counsel (IAC). To prove IAC, a defendant must show (1) counsel’s performance was deficient and (2) prejudice resulted. If the underlying objection would have failed, counsel is not deficient.

5. Conclusion

State v. Molde fortifies the evidentiary line by clarifying that “numbers are not endorsements.” Generalized statistics regarding false reporting may enlighten jurors without treading on their duty to decide individual credibility. By overruling Mader, the Court resolves intra-state tension and aligns Wisconsin with jurisdictions that admit low-incidence statistics where methodologically sound. Practitioners must now pivot from categorical vouching objections to substantive critiques of reliability and prejudice, while experts must remain vigilant not to conflate population data with person-specific conclusions. The decision preserves the jury’s fact-finding prerogative yet equips it with empirical context—balancing fairness to defendants, educational value to jurors, and respect for victims’ experiences.