Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We use data from the Associated Press college American football poll to analyze two types of ex-post optimality of social learning in a non-lab setting. The poll is a weekly subjective ranking of the top 25 teams, voted on by over 60 sports journalists. Voters potentially can learn from their peers by observing the aggregate ranks before updating their individual ranks. Our results indicate that, while voters do learn from their peers to some extent, the informativeness of peer ranks appears to be under-valued. Copyright Springer Science+Business Media New York 2014