Gaming techniques have long been used to help acquire vicarious experience in the management of crises and the conduct of military operations. The more familiar kinds of political-military games have been used primarily to train decisionmakers, and to exercise various kinds of procedures and operations. This Note reports on an effort to determine the potential utility of gaming as a research tool, i.e., as a way to formulate and test a variety of research hypotheses related to the management of superpower crises. This goal entailed several requirements not typical of a traditional political-military game. These requirements included the need for (1) a system for recording the game play for later analysis, and (2) a game design that permitted the value of key variables to be changed and the game to be "rerun." The authors describe the design of the games and the scenario employed, outline some tentative research hypotheses, and present their conclusions about a methodology for using gaming as a research tool.