There are a couple of things about Judge Bybee's opinion in this case that I marginally dislike, including the fact that he seems to stretch out to resolve an intracircuit conflict. (He claims that he's required to resolve the relevant standard because this is a uniquely close case, but from the tone and content of the opinion, I'm just not sure that I really think he's of that view.)
On the merits, my biggest disagreement -- though (again) I didn't feel massively strongly in this regard -- was with his discussion of the admissiblity of statistical evidence. Judge Bybee holds that the district court erroneously excluded plaintiff's statistical evidence (in this "pattern and practice" discrimination suit) because flaws in this evidence generally go only to weight, not admissibility. But he recognizes that errors may indeed be so troubling to justify exclusion of the evidence entirely -- at a minimum, they often may make the evidence more prejudicial than probative. And my sense was that the district court may well have made the right call here, for precisely that reason. The evidence that was excluded didn't at all attempt to control for the qualifications of the applicants; plus, the expert's conclusion was that there was "no statistical evidence . . . that the selection process [] were unbiased" -- in other words, that the statistics did not prove a lack of bias -- rather than that the statistics actually proved bias. That's the kind of conclusion that might well confuse a jury, and particularly given the alleged methodological flaws, I'm not sure Judge Bybee gets it right.
Remember, after all, that we're talking about whether the exclusion of this evidence was an abuse of discretion. Judge Bybee articulates this standard, but you don't hear much about it in the opinion, nor is there a lot of reflection on this point evident in Judge Bybee's analysis. I might (or might not) have made a different call as the district court judge. But it's hard to find that the evidentiary call made here was an abuse of discretion, and the deference given to district courts under Rules 402, 403, and 702 doesn't find much expression in this opinion. Which may make for both bad law and bad discrit court adjudications.
