This Note is the companion piece to earlier work, which described a methodology for analyzing and gaming opponent reasoning and reported on its employment during and after the recent conflict with Iraq. It examines four crises in which either British or U.S. interests were threatened, testing the key decisionmakers for the "limited rationality" and for the incrementalist or more goal-driven approach to risk taking. It also identifies the policy implications of these case studies at both the broad strategic and the crisis-specific levels, and evaluates the relative effectiveness of the military and economic tools of political coercion. The workhorse of British and U.S. policies has been the strategy of extended deterrence. The authors explore the disjunction between the theory and practice of extended deterrence in each of the cases they survey. Aside from the decisionmaking problems caused by time constraints and imperfect information, there are powerful general psychological influences — of which frustration and positive or negative feelings about one's current situation are the most important. Also, aggressors routinely underestimate the effects of their opponents' maritime capabilities for blockade, strategic lift, and bombardment, suggesting that a form of "analytic bias" is present. The policy implications are (1) U.S. decisionmakers and their staff organizations must habituate themselves to the practice of carrying along multiple models of opponent reasoning; and (2) in some cases, the United States must be prepared to give unambiguous political and military warning in the face of looming crisis, necessitating bilateral or multilateral treaties with friendly nations in areas of vital interest.