Using NFL Draft Metrics to Predict Player Success

• Activity Number: 349
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 11:15 AM
• Sponsor: Section on Statistics in Sports
• Author(s): Nicholas Kapur*+ and Justin Post and James Gilman
• Companies: North Carolina State University and North Carolina State University and North Carolina State University
• Keywords: Football ; Combine ; Imputation ; Regression ; Ranking

Abstract

Author: Nick Kapur, James Gilman, and Dr. Justin Post

Keywords: Football, Combine, Imputation, Regression, Ranking

ABSTRACT: The NFL draft has become a 3-day spectacle that can change the fortunes of a franchise. There are many draft gurus that attempt to rank players and project their futures; however, most of the analysis these gurus do is subjective. In an effort to make draft grades more objective and data driven, we attempt to use NFL scouting combine data and college affiliation to predict NFL impact. Many players do not complete every drill during the combine, resulting in a lot of missing data. Therefore, we use multiple imputation methods and linear regression models in order to predict player success. We look at which drills relate to having the most success at the different positions, which teams draft well based off our model, and which collegiate conferences produce the most NFL ready players.