A recent article by Susan Sample purports to resolve a debate that has generated a great deal of scholarly attention over the past two decades, whether arms races are associated with the escalation of militarized disputes to war. In response, we outline a research agenda designed to reconfigure the arms race-war debate rather than to perpetuate the controversy. We argue for better theoretical specification of the arms race-war relationship, and for empirical tests to sort out whether this relationship is direct, indirect, or spurious. We also advocate some methodological changes including more longitudinal studies, a broader spatial domain to include minor powers, less reliance on military expenditure data, and multivariate (rather than bivariate) tests of the arms race-war relationship.