When the leader of an autocratic regime loses power, one of three things happens. The incumbent leadership group is replaced by democratically elected leaders. Someone from the incumbent leadership group replaces him, and the regime persists. Or the incumbent leadership group loses control to a different group that replaces it with a new autocracy. Much scholarship exists on the first kind of transition, but little on transitions from one autocracy to another, though they make up about half of all regime changes. We introduce a new data set that facilitates the investigation of all three kinds of transition. It provides transition information for the 280 autocratic regimes in existence from 1946 to 2010. The data identify how regimes exit power, how much violence occurs during transitions, and whether the regimes that precede and succeed them are autocratic. We explain the data set and show how it differs from currently available data. The new data identify autocratic regime breakdowns regardless of whether the country democratizes, which makes possible the investigation of why the ouster of dictators sometimes leads to democracy but often does not, and many other questions. We present a number of examples to highlight how the new data can be used to explore questions about why dictators start wars and why autocratic breakdown sometimes results in the establishment of a new autocratic regime rather than democratization. We discuss the implications of these findings for the Arab Spring.