Multiple Testing Procedures for Sports Data: An Analysis of the MLB Hall of Fame

• Activity Number: 43
• Type: Contributed
• Date/Time: Sunday, July 29, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Sports
• Author(s): Leah Jager*+
• Companies: U.S. Naval Academy
• Keywords: multiple testing ; high dimensional data ; baseball ; sports

Abstract

A hot topic in statistics involves developing methodology for high-dimensional data problems. Major applications of these methodologies lie in genomics -- for each person in a gene expression microarray study, investigators collect gene expression data on thousands of individual genes. This is not unlike the scenarios that exist in the world of sports statistics, although on a much smaller scale. In today's world, a large number of individual statistics are collected on a single athlete or team. Because of this, analysts of sports data may need to consider some of the same issues that face analysts of high-dimensional data.

In this paper, we consider the similarities of certain sports analyses to the analysis of microarray data in genomics. We review the issue of multiple testing, which often arises in high-dimensional analyses. Then we apply these multiple-testing procedures to analyze player attributes that are associated with successful entry in the Major League Baseball Hall of Fame. Finally, we discuss other possible applications of this methodology to the world of sports.