The Priest/Klein model predicts both trial rates and plaintiff win rates as functions of three structural parameters: the decision standard, parties' uncertainty in estimating case quality, and the degree of stake asymmetry across parties. Previous tests of the model are unsatisfactory because most have concentrated on its prediction of a 50 percent win rate, which only obtains as a limiting case. We gather independent evidence that describes the model's three parameters and compare it with estimates from a structural model that simultaneously estimates both trial and win rates. The model fits the data for four of our six case types. A four-parameter model, in which plaintiffs face greater uncertainty than defendants, can explain the two anomalous case types.