Don't miss the comments, especially an insightful one from Roger Clegg regarding the study's timing, on my earlier post about the Daye diversity study. Paul Caron was kind enough to cite to my rebuttal, but blogs like JDJournal blithely repeat the press release with a headline that the study "proves" the benefits of diversity.
One commenter writes: "You assume that LSAT scores or college grades are a predictor of success as a law student, and that success as a law student is a predictor of competence in the practice of law." I don't just assume it. I know it to be a fact. The race is not always to the swift nor the battle always to the strong, but that's where the smart betting man puts his money. As Richard Sander has demonstrated, LSATs do predict law school success; law school success does have predictive value for career success. Not everyone with a perfect LSAT will become a millionaire (some of us will foolishly toil in the think-tank mines and fritter away their skills with other non-profit work), and not every smart attorney was Order of the Coif in a top-six school, but ceteris paribus, law firm hiring attorneys have Bayesian reasons to want graduates with good grades from top law schools.
Now, concededly, there's more than one way to be a success as an attorney. Someone with extraordinary social skills can squeak past the bar exam, and use his social skills to recruit many clients with profitable cases, and then hire other attorneys to do the intellectual heavy lifting; someone else with exceptional charisma and oral communications skills can persuade juries (or threaten to persuade juries) of facts that just aren't so, and use those results to attract clients and negotiate favorable settlements before appeals courts throw out the cases. If you have these rare skills, you have the capacity for far nicer Christmas parties than the average Harvard Law Review editor, and your LSAT is irrelevant—but so is the name on your law school diploma, and thus your relevance to the conversation about opportunity and the need for racial preferences.