This study describes the invention and early development of two social scientific methods, the Delphi method and political gaming. These methods have been developed in the early 1950s at the RAND Corporation, a think tank based in Santa Monica, California, which entertained close relations to the US Air Force. Historically, Delphi and political gaming were the first social scientific methods to systematically use expert opinions for the delineating scenarios of possible futures. Their development is analyzed with the use of two concepts introduced in this study, namely the concept of the epistemic role of the expert and the concept of epistemic hopes. An epistemic role is understood as the bundle of expectations held by the inventors of the methods toward the expert. Epistemic hopes, on the other hand, are defined as the expectations of the social scientists towards capacity, efficience, and usefulness of their methods when applied to policy problems. The comparison of the processes by which Delphi and political gaming were developed shows that although the two methods share several features, the epistemic role of the expert is conceived differently. Whereas Delphi conceives of the expert as carrier of universal knowledge, political gaming conceives of knowledge as being inherently culture-bound. This is explained by pointing out that the two groups of scientists who invented the method were trained in two different philosophical traditions. Also, the analysis shows ambivalent conceptions of the expert among the early Delphi studies. The authors held several epistemic hopes toward the impact of their methods on society. The hope was that the further scientific methodology extended into the realm of policy advice, the more rational a process of policy making would become. Both within the military and the government, a critical mass of high ranking officials was convinced that the social sciences could provide valuable contributions to political decision processes.