Models, Markets, and the Forecasting of Elections

TL;DR

This paper compares probabilistic election forecasts from a model and market prices, finding that combining them improves overall accuracy and proposing a hybrid approach with a trading bot for better predictions.

Contribution

It introduces a hybrid forecasting method combining model and market data, and proposes a novel market design and evaluation criterion for election prediction accuracy.

Findings

Markets outperform models months before the election.

Models perform better closer to the election.

Averaging forecasts improves overall accuracy.

Abstract

We examine probabilistic forecasts for battleground states in the 2020 US presidential election, using daily data from two sources over seven months: a model published by The Economist, and prices from the PredictIt exchange. We find systematic differences in accuracy over time, with markets performing better several months before the election, and the model performing better as the election approached. A simple average of the two forecasts performs better than either one of them overall, even though no average can outperform both component forecasts for any given state-date pair. This effect arises because the model and the market make different kinds of errors in different states: the model was confidently wrong in some cases, while the market was excessively uncertain in others. We conclude that there is value in using hybrid forecasting methods, and propose a market design that incorporates model forecasts via a trading bot to generate synthetic predictions. We also propose and conduct a profitability test that can be used as a novel criterion for the evaluation of forecasting performance.