Dale Zimmerman, PhD

Previously published analyses of NCAA men’s basketball tournament game outcomes since its 64-team format began in 1985 have uncovered some apparent anomalies, such as 12-seeds defeating 5-seeds more often than might be expected, and 10-, 11-, and 12-seeds making it to the Sweet Sixteen more often than 8- and 9-seeds.  However, this literature does not offer or provide supporting evidence for any plausible explanations for these anomalies.  I seek to provide an explanation based largely on the distribution of team strengths, as estimated by a linear model for the difference in score of all games between NCAA teams in a given season.  The impact that automatic qualifiers (e.g., conference tournament champions) have on the anomalies is examined, as is the extent to which the Sweet Sixteen anomaly would disappear if teams were re-seeded between the first and second rounds.  We also investigate anomalies in the NCAA women’s basketball tournament, which differ substantially from those of the men’s tournament.

Dale Zimmerman, PhD