Depression and Accuracy: Evidence from the 2010 FIFA World Cup

Before and during the 2010 Soccer World Cup, participants made probabilistic forecasts for the outcomes of the tournament. We examine the relationship between depression levels and performance in this real-world forecasting task. Across two different waves of predictions, for both continuous and categorical classifications of depression, and with multiple measures of prediction accuracy, we find that depressed forecasters were less accurate. The poorer accuracy amongst the more depressed forecasters was mediated by neglect for base-rate probabilities: The depressed participants assigned probabilities that departed more substantially from the baserates, particularly for low base-rate events. Given the high incidence of depression in the workforce, the importance of probabilistic forecasting as a judgment task, and that we may be the first to look at forecasting accuracy on an ecologically valid task with objective outcomes over which the forecaster has no control, these finding may have important implications for both theory and practice.