The killing of a head of state is among the most severe and consequential forms of political violence. But to date, there have been no systematic studies of the incidence of such assassinations, with the few existing case studies tending to emphasize the uniqueness of those events. Drawing on existing theories of social protest and contentious politics, we argue instead that institutional and sociopolitical factors should be important correlates of assassination. We examine empirically the implications of this theory, using data on the incidence of assassinations of heads of state between 1946 and 2000. Our findings suggest that institutional factors related to leadership succession, institutionalized power, and levels of repression interact to influence the occurrence of such killings. Notable in this respect is our conclusion that, while repressive leaders are at greater risk for assassination, the effect of repression is moderated by executive power, such that weak, repressive leaders in nondemocratic systems face the highest risk of assassination. Our findings dovetail neatly with the broader literature on other forms of insurgency, suggesting that assassinations are but one manifestation of the larger phenomenon of political violence.