In this article, the authors propose an extension to Fiorina's balancing model based on voters' electoral expectations and test this extension and several implications of the theory. The authors examine the observed pattern of ticket-splitting and find it less consistent with a balancing perspective than with an alternative approach stressing separation between presidential and congressional voting. They also address the relationship between party polarization and ticket-splitting; their results indicate that the occurrence of split ballots does not increase with polarization. A further test identifies respondents with both the sophistication and the motive to engage in balancing behavior. According to this analysis, balancing considerations influence, at most, the very small group of voters whose sophistication and electoral expectations give them the tools and the incentive to pursue balance with a split ballot. Ticket-splitting appears to result far more from incumbency and cross-pressured voters holding candidate evaluations at odds with their partisan learnings.