Bueno de Mesquita and Lalman's version of an expected utility theory of war is one of the most widely cited theories of international conflict. However, the testing of the theory has lagged its theoretical development. To date, the theory has been tested on only 707 dyad-years, all drawn from Europe between 1816 and 1970. We present a broader test of War and Reason's expected utility theory of war using the same methods as Bueno de Mesquita and Lalman. Our tests include the full international system from 1816 to 1984. We find that the theory receives empirical support using the set of politically relevant dyads for testing, but the relationship is less clear among the population of all interstate dyads.