A paradox in the comparative literature on electoral systems is that one of the most common systems in Europe – flexible-list proportional representation systems – may be the least understood. Any study of flexible-list systems must start by acknowledging a puzzle: why candidates spend time and effort striving to win preference votes when typically these votes make no difference between election and defeat. Offering the first comprehensive multi-country test of this key puzzle, we provide evidence from Belgium, the Czech Republic, and Slovakia that parties will promote to better list ranks in the next election those candidates who are successful at winning preference votes, thereby improving their prospects of election in the longer term and incentivizing them to cultivate personal reputations. Our findings have important implications for party scholars and practitioners when designing, or reforming, political institutions.