This chapter summarizes the book’s major findings regarding party location, turnout, voter choice, and voter participation. First, the simple party competition model suggests that in unidimensional policy spaces, satisficing by winners plus search by losers produces a string of governmental policies that converges to the median voter’s ideal point when citizens always vote for the party that will better serve their interests. Second, turnout responds in intuitive ways to changes in parameter values. Average participation is higher the further apart the parties, in both two-component models with fixed platforms and in the simple turnout model. Third, voters learn to support parties that better serve their interests. Fourth, voter coordination is usually successful: the Condorcet loser is almost never selected. The behavioral theory of elections based on bounded rationality is formulated in the languages of dynamic systems and stochastic processes.
Entropy in Dynamic Systems