Proven forecasting model predicted Harris loss to Trump in US presidential race
Source citation: Enns, P. K., Colner, J., Kumar, A., & Lagodny, J. (in press 2024). Understanding Biden’s exit and the 2024 election: The State Presidential Approval/State Economy Model. PS: Political Science & Politics, 1–14.

In this article, the creators of an election forecasting model showed how it predicted the winner of the 2024 presidential race. With data from 100 days prior to the election, the model found that Kamala Harris had only a 25 percent chance of taking the Electoral College. Authors Enn et al. also outlined their model’s historic accuracy. It predicted the 2020 presidential winner in 49 of 50 states, and would have correctly forecasted the winner in 95 percent of the states in presidential elections since 2000. The model relies on state-level economic data and presidential approval ratings, and is now supplemented by individual-level data from 89 historic surveys from several sources, including CBS News/New York Times polls from ICPSR. Enn et al. also described this year’s updates to the model, including the addition of indicators for Hispanic respondents.