This dissertation examines the tie between campaign spending and voter calculus where candidates’ platforms are perceptually fixed. It also presents a spatial voting model addressing the equilibrium platform implications of a shared, catastrophic utility loss to political parties, incurred when voter preferences are inadequately represented. In the first two chapters, I adapt models of economic growth and consumer behavior to address optimal campaign resource allocation across both time periods and the focus of spending. These decisions are assumed to play a critical role in defining the non-policy “valence” elements of voter calculus and to also affect the salience of various issue dimensions. This approach yields simple rules by which candidates can maximize their expected appeal on Election Day. I test the implications of the formal models using data on television advertising by national political campaigns in the 2000 and 2004 election cycles. In the third chapter, office-seeking candidates choose platforms given a period-specific probability distribution regarding the position of the median voter. Platform divergence results from an intratemporal trade-off between short-term electoral gain and long-run system stability. In equilibrium, the level of divergence between parties is shown to be increasing in the level of uncertainty about the preferences of the median voter, which is a function of prior realizations of this policy position.