Reports about runaway jury awards have become so common that it is widely accepted that the U.S. jury system needs to be "fixed". Proposals to limit the right to a jury trial and increase judicial discretion over awards implicitly assume that judges decide cases differently than juries. We show that there are large differences in mean awards and win rates across juries and judges. But if the types of cases coming before juries are different from those coming before judges, mean award and win rates may differ even if judges and juries would make the same decisions when faced with the same cases. We find that most of the difference in judge and jury mean awards can be explained by differences in the sample of cases coming before judges and juries. On some dimensions, however, there remain robust and suggestive differences between judges and juries. Copyright 2000 by Oxford University Press.