Background
Tyler Speth, an insurance office employer, was charged with seven counts of sexual assault-related conduct against a 16-year-old employee (A.B.) who worked for him between June and November 2022. A.B. alleged that Speth offered to pay her for sexual favors and sexually assaulted her on two occasions: once involving penis-to-vagina intercourse in the office basement on August 9, 2022, and again involving oral sex in the sales office between August 11–14, 2022. A jury trial in July–August 2023 heard testimony from 23 witnesses, including A.B., her family, coworkers, investigating officers, and forensic analysts.
The jury convicted Speth on all seven counts. The circuit court sentenced him to five years of initial confinement and six years of extended supervision, to be served concurrently. Speth then filed post-conviction motions seeking judicial recusal and a new trial based on claims of ineffective assistance of trial counsel and newly discovered evidence.
The Court’s Holding
The Court of Appeals affirmed the conviction and rejected all of Speth’s post-conviction arguments. On the recusal claim, the court found that Speth failed to rebut the presumption of judicial impartiality. Although the trial judge disclosed knowing Sherry Lange—who observed trial and was connected to a witness—through decades of civic organization membership, Speth presented no evidence that this relationship created objective bias or a “serious risk of actual bias” under Wisconsin law.
The court also rejected Speth’s fourteen ineffective assistance of counsel claims. The court credited trial counsel’s testimony that a key alibi witness (Puttkamer) never disclosed specific alibi information before trial, making counsel’s failure to call her testimony non-deficient. For another potential witness (Gerstner), counsel reasonably declined to present his testimony because contradictory photographs undercut the alibi defense. The decision not to call employee Connor Smith—whose testimony about being overpaid by Speth was offered in opening but withheld at trial—was a reasonable strategic choice because four other employees had already testified to similar facts, making Smith’s testimony cumulative and potentially “laborious” late in a week-long trial. Counsel had also presented and highlighted unsubstantiated allegations against a restaurant coworker to attack A.B.’s credibility, so no deficiency arose from failing to develop that evidence further.
Key Takeaways
- Trial counsel’s strategic decisions receive substantial deference on appeal; courts will not second-guess reasonable trial strategy, even if hindsight suggests a different path.
- Counsel is not ineffective for failing to present evidence that was not disclosed during pretrial preparation, nor for declining to call witnesses whose testimony would be cumulative or contradictory.
- A judge’s civic or social association with trial participants does not create a “serious risk of actual bias” unless the party seeking recusal presents clear evidence of a substantial relationship affecting impartiality.
- Post-conviction relief for ineffective assistance requires the defendant to prove both deficient performance and prejudice; courts may resolve claims on the deficiency prong alone.
Why It Matters
This decision reinforces the high bar for obtaining post-conviction relief in Wisconsin based on claims of ineffective assistance of counsel. By crediting counsel’s trial testimony about witness preparation and strategic rationale—and declining to second-guess cumulative evidence decisions—the court illustrates how broadly appellate courts defer to trial counsel’s judgment. For criminal defendants, the ruling underscores that simply identifying evidence counsel did not present, or witnesses counsel did not call, does not establish ineffectiveness; a defendant must show counsel’s performance fell below an objective standard of reasonableness and caused prejudice to the defense.
The decision also clarifies Wisconsin’s standard for judicial recusal: absent concrete evidence of substantial personal ties or actual bias, a judge’s disclosure of a civic association with someone present at trial does not warrant removal. This may affect how trial judges handle disclosures in smaller communities where professional and civic circles overlap.