The KSC disagreed, noting that the COA panel appeared to have independently assessed witness credibility and noting that "[i]nterestingly, the panel's credibility calculus did not include any reference to the jury's acquittal of Brooks on five of the seven counts upon which S.C. testified nor any rumination upon what inference might be drawn therefrom about the jury's view of S.C.'s credibility."
Ultimately, the KSC determined that the panel had gone too far in evaluating the strength of the state's case rather than apply the correct test, which asks if there is a reasonable probability of a different outcome:
Given the panel's stated reasons for finding no prejudice, we discern that it ran afoul of the oft-stated rule that an appellate court will not determine the credibility of witnesses or weigh conflicting evidence. "The jury is charged with the responsibility of weighing the evidence and determining witness credibility. Appellate courts do not reweigh the evidence or decide which witnesses are credible." State v. Corbett, 281 Kan. 294, 310, 130 P.3d 1179 (2006).
The relative importance of testimony about the presence or absence of a penile scar was something the jury was supposed to decide. The panel's queries about the scar evidence, e.g., whether such a scar is permanent, went to the weight to be attached to that evidence and invaded the province of the jury. Likewise, the panel was figuratively on jury duty when it assessed the credibility of the victim. Appellate courts should "let the jury decide whether to believe a witness." State v. Leaper, 291 Kan. 89, 108, 238 P.3d 266 (2010) (Johnson, J., concurring). Certainly, appellate courts should not excuse constitutional violations based upon a cold record assessment of a witness' believability.Applying the proper test, the KSC held that, but for defense counsel's deficient performance, Mr. Brooks had shown a reasonable probability that the outcome of the case would have been different. As a result, the KSC reversed and remanded for a new trial.