The power to remove the government via no-confidence motion is a powerful tool afforded to the opposition. By triggering the government’s downfall, opposition parties can substantially influence policy direction in parliamentary democracies. Yet, we know surprisingly little about how government and opposition parties interact to determine the occurrence of no-confidence motions and their chance of success. In this project, I develop a simple formal model that identifies the factors influencing when opposition parties propose no-confidence motions and their outcomes. I find support for these expectations by estimating an empirical model that is explicitly derived from the underlying theoretical model. Unlike previous empirical studies of government stability, this project honors the strategic interactions that occur between government and opposition parties. In addition to the possibility of the motion passing, opposition parties are motivated by electoral considerations, which induce different behaviors at various stages of the electoral cycle. This project offers a number of implications for the study of parliamentary politics, including theories of opposition behavior, democratic accountability, and government duration and termination.