William L. Meadow & Cass R. Sunstein, Statistics, Not Experts, 51 Duke L.J. 629 (2001).
Abstract: The legal system should rely much more than it now does on statistical evidence. It should be cautious about the judgments of experts, who make predictable cognitive errors. Like everyone else, experts have a tendency to blunder about risk, a point that has been shown to hold for doctors, whose predictions significantly err in the direction of optimism. We present new evidence that individual doctors' judgments about the ordinary standard of care are incorrect and excessively optimistic. We also show how this evidence bears on legal determinations of negligence, by doctors and others.