The Billion Dollar Statistical Concept: How a Lack of Understanding of a "Simple" Statistical Technique Causes Huge Losses and Poor Decisions in Sports (and Drug Development!)
March 19, 2012
4:00 PM - 5:00 PM
George R. Brown Hall
CAPITAL ONE LECTURE SERIES
This talk will discuss hierarchical modeling in sports and present numerous examples when this simple idea is ignored -- and the mistakes that are made because of this. Examples from baseball, golf, and many other sports will be explored to show the lack of understanding of a relatively simple statistical idea. This lack of undertsanding has enormous negative effects in decisions that are made. Interestingly, these decisions are the same as poor decisions that are made in drug development -- again resulting in poor decisions -- and billions in losses. As more and more "information" is being collected in every field of sport, business, and science, this simple idea is becoming more critical to understand and utilize.
BIO: Scott Berry is President and Senior Statistical Scientist at Berry Consultants. Since 2000 he has been involved in the design of innovative and adaptive clinical trials in drugs, devices, and biologics. His research interests are in Bayesian methods, computations, simulations, and hierarchical models. In addition to his innovative work in biostatistics he is a renowned sports statistician, with more than 40 articles, ranging from the Journal of the American Statistical Association to ESPN the Magazine. He received his PhD from Carnegie Mellon University (1994) and his BS from the University of Minnesota (1990). He spent 5 years at Texas A&M University (1995-2000) before founding Berry Consultants with Don Berry.
*Reception to follow talk.