Partisan Lean of States: Electoral College and Popular Vote

Abstract

We compare federal election results for each state versus the USA in every second year from 1992 to 2018, to model partisan lean of each state and its dependence on the nationwide popular vote. For each state, we model both its current partisan lean and its rate of change, as well as sensitivity of state results with respect to the nationwide popular vote, using Bayesian linear regression. We apply this to simulate the Electoral College outcome in 2020, given even (equal) nationwide popular vote, as well as 2016, 2008, and 2004 nationwide popular vote. We backtest 2012 and 2016 elections given actual popular vote. Taking equal popular vote for two major parties, we prove that the Electoral College is biased towards Republicans.