Electoral systems are uniquely distributive political institutions that shape political outcomes, yet are themselves endogenously shaped outcomes of political choices. In Poland, party system development has involved not only parties adapting to electoral institutions in each election, but also parties modifying these institutions prior to every election. We model electoral system change as driven by partisan self-interest in maximizing seat share and test it in five episodes of electoral system change in Poland from 1989 to 2001, comparing parties' support for electoral law alternatives to their expectations of seat shares from those alternatives. Data consists of opinion polls, roll-call votes, Sejm records, constitutional committee transcripts, and interviews with political actors who designed and chose the Polish electoral institutions. The findings clearly show that party support for each electoral law was closely linked to the perceived effect on that party's seat share, with this linkage growing more consistent over time.The relationship between electoral and party systems, nearly all scholars recognize, is mutual: electoral institutions shape party systems, but themselves are formed in an environment of partisan electoral competition. Despite this recognition, however, electoral studies has been overwhelmingly concerned with the political consequences of electoral laws, rather than examining precisely how electoral laws themselves arise as consequences of political processes. Political experience, however, demonstrates repeatedly that while actors do maximize their self-interest by adapting their strategies in response to institutions, they also struggle to modify the institutional settings that transform their strategies into outcomes (Tsebelis 1990).Despite its importance, this cycle of strategic adaptation and institutional modification remains understudied and incompletely understood. With the exception of a handful of case studies, the politically endogenous origins of electoral institutions has yet to receive systematic comparative treatment. Remington and Smith (1996) have examined the choice of electoral systems in post-communist Russia, Brady and Mo (1992) in Korea, and Bawn (1993) in West Germany in the postwar period. Other approaches to explaining electoral systems tend to focus on explain-