Debate is a "mind sport" that requires fair and impartial judging. This study examines debate rounds at the Tournament of Champions from 2004 to 2009. We use a binomial choice model to estimate the marginal effects of regional bias, sex bias, and side bias, using transitive predictions to control for idiosyncratic quality. We find evidence of region and side bias but not sex bias. These factors may explain the significant number of nontransitive outcomes in the data. Finally, we suggest some policy remedies to mitigate the impact of biases and further applications of our methodology. (JEL C25, J16, J52, J71, L83)