Using Folded Seats-Votes Curves to Compare Partisan Bias in the 2020 Presidential Election with Partisan Bias in the Five Other Presidential Elections in the 21st Century

Abstract

Using a folded seats-votes curve, we examine partisan bias in the 2020 presidential election and compare it to partisan bias in the five other presidential elections in the 21st century. 2020 and 2016 are extreme outliers with respect to the absolute magnitude of partisan bias in the Electoral College. In 2016, 2020, and 2000 bias runs in a pro-Republican direction; in the other three elections in this century, the opposite is true. But partisan bias can vary with where on the seats-vote curve we look to find bias, and we identify the lowest vote share at which there is no partisan bias (if such exists).