The nature of regional bias in Heisman voting

This study explores regional bias in Heisman voting from 1990–2016 using a negative binomial regression model with player-year fixed effects. Analysis confirms finalists receive higher vote tallies in home regions, on average. Additionally, results show regional vote tallies are decreasing in the fraction of other finalists in-region. Furthermore, evidence reveals finalists receive higher vote tallies for each game played against in-region teams and lower vote tallies for each game played by other finalists against in-region teams. Analysis is augmented by showing the recent increase in national television coverage of college football has been accompanied by a decline in regional bias. 5