The median number of seats won by Islamist political movements, parties, and associations in the 261 parliamentary elections held in Muslim-majority countries since 1970 is precisely zero. This might be surprising, given the attention scholars pay to Islamists’ electoral participation in prominent cases, such as Algeria, Egypt, Jordan, Turkey, and Indonesia. While Islamists won at least one seat in 101 elections, they did not even participate in 146 others. This paper explores spatial and temporal variation in Islamist electoral participation and performance using an original dataset of all parliamentary elections held in Muslim-majority countries since 1970. I find that Islamist participation in elections varies dramatically by region. When Islamists do participate, however, their vote shares exhibit a remarkably similar trend over time. With a few notable exceptions (e.g. Palestine, Iraq, Egypt after Mubarak), Islamists’ vote shares tend to increase from election to election until they stabilize around 11-20%. In countries with the most parliamentary elections since 1970 and where Islamists participate, such as Malaysia, Indonesia, and Bangladesh, Islamist vote shares oscillate around this stable level. I use these data and results to put electoral Islamism in the Arab world in a wider comparative context. Since few free elections have been held in Arab countries, it is difficult to assess how Islamist performance will change with democratization and if Islamists’ recent electoral successes in Egypt, Tunisia, and Morocco are signs of wider regional or global trends. Using this unique global dataset, I explore related questions, such as: 1) Do Islamists perform better in free elections than in less free ones? 2) Do Islamists perform better after a democratic transition if they had participated in elections under authoritarianism? 3) Does the participation and performance of Islamist movements in one state affect the electoral prospects of similar movements in neighboring states? To better understand processes of change over time and identify factors that my “large-N” analysis may have ignored, I compare two Islamist movements that significantly expanded their vote share over time (Turkish and Egyptian Islamists) to those that did not (Indonesian and Bangladeshi). I conclude by drawing implications for Islamists in a more democratic Middle East.