The development of causal models for forecasting voting in elections has become an important sub-discipline of political science. The models’ predictive validity is of interest to researchers whose theories of voting behavior are represented by the models, and to decision-makers whose plans vary depending on their expectations of who will win an election.

The PollyVote divides available models into three categories:

1. Retrospective voting models
2. Prospective voting models
3. Mixed (retrospective & prospective) voting models

The combined models component forecast is the simple average across the three components.