Predicting Strategic Voting Behavior with Poll Information

TL;DR

This paper introduces a new heuristic model for strategic voting under poll information, which outperforms existing models in predicting voter behavior by incorporating candidate success and utility considerations.

Contribution

It proposes a novel, tunable heuristic for modeling strategic voting that accounts for poll scores and voter preferences, validated against large-scale data.

Findings

The new heuristic outperforms existing models in prediction accuracy.

Prediction errors are partly due to voters choosing dominated candidates.

Model tuning improves individual voter behavior prediction.

Abstract

The question of how people vote strategically under uncertainty has attracted much attention in several disciplines. Theoretical decision models have been proposed which vary in their assumptions on the sophistication of the voters and on the information made available to them about others' preferences and their voting behavior. This work focuses on modeling strategic voting behavior under poll information. It proposes a new heuristic for voting behavior that weighs the success of each candidate according to the poll score with the utility of the candidate given the voters' preferences. The model weights can be tuned individually for each voter. We compared this model with other relevant voting models from the literature on data obtained from a recently released large scale study. We show that the new heuristic outperforms all other tested models. The prediction errors of the model can be partly explained due to inconsistent voters that vote for (weakly) dominated candidates.