Finishing How They Started: Evidence of Mean Reversion in the March Madness Tournament

Summarise with VIRAT
Authors: Joey Smith, Joseph Hubert Wilck IV, Clint Buck
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Every year millions of people fill out NCAA men’s Division I basketball tournament brackets hoping to select as many winners as possible. While consistently choosing the better seed to win may be a safe strategy, many research efforts have tried to improve upon this approach. Some of these efforts use complex data and advanced statistical methods, making the process inaccessible to the common fan. Other research uses simpler but more understandable data and techniques. We build on this latter approach and introduce a new measurement system to assess underperformance or overperformance of a given team. This research provides a data-informed way to identify possible upsets. Our research identifies one very significant variable in predicting team performance in the tournament: the team’s preseason ranking. We also find that teams who win their conference tournament often underperform, perhaps due to recency bias and receiving an inflated seed.

Fields of Study & Subjects

Sports Analytics and Performance Sports Performance and Training Sports Gender and Society

Publication details

Journal / source
Bucknell Digital Commons (Bucknell University)
Publication date
1 January 2026
Language
EN