Do election outcomes lead to affective polarization? A growing set of studies pose this question, fearing that the fundamental mechanism of democracy—competitive elections—may also pose a risk to its stability. However, many political stimuli occur around election time. How can researchers isolate the effect of winning or losing itself? In this study, we use a pre-post regression continuity design (RDD) to study the effect of 648 close US House, Senate, and state-level presidential outcomes between 1996 and 2020 on the attitudes of survey respondents who experienced them. When controlling for pre-election polarization, people whose party saw a “close win” emerge more polarized than their opposition. However, this is primarily because of movement among the election’s losers, who depolarize by feeling less warmly toward their own party. If election losses weaken people’s commitment to democracy, anger at the opposition may not be the reason.
