Laboratory experiments have established the existence of cognitive biases, but their explanatory power in real-world economic settings has been difficult to quantify. We evaluate the extent to which a cognitive bias, confirmatory bias, affects the opinions of experts in a real-world environment. In the Associated Press Top 25 College Football Poll, expert pollsters are tasked with assessing team quality, and their beliefs are treated week to week with game results that serve as signals about an individual team’s quality. We exploit the variation provided by actual game results relative to market expectations to develop a novel regression-discontinuity approach to identify confirmatory bias. We construct a data set that matches more than 20 years of individual game characteristics to poll results and betting market information and show that teams that slightly exceed and barely miss market expectations are exchangeable. The likelihood of winning the game, the average number of points scored by teams and their opponents, and even the average week of the season are no different between teams that slightly exceed and barely miss market expectations. Pollsters, however, significantly upgrade their beliefs about a team’s quality when a team slightly exceeds market expectations. The effects are sizable—one fifth of the standard deviation in poll points in a given week can be attributed to confirmatory bias, which is equivalent to nearly half of the voters in the poll ranking a team one slot higher when they slightly exceed market expectations. This type of updating suggests that even when informed agents make repeated decisions, they may act in a manner consistent with confirmatory bias.