A Quantitative Modeling of Rushing Ability in the NFL

Loading...
Thumbnail Image
Authors
Roberts, William
Issue Date
2016
Type
Thesis
Language
en_US
Keywords
Research Projects
Organizational Units
Journal Issue
Alternative Title
Abstract
In professional football, rushing has often been considered one of the most important aspects related to moving the ball on offense. In the past, the ability of rushing players has been measured strictly by the number of total yards they produce. Other statistics such as touchdowns, first downs, or even fumbles have not been formally considered in the ability of a rushing player. In this paper, models will be created to attempt to measure the ability of modern day rushing players that include several, or all of the previously mentioned statistics. Multiple linear regression models to predict rusher rankings of all-time great rushers were estimated using a set of quantitative performance variables. The models were then applied to the 180 rushing players who played during the 2005-2014 NFL regular seasons. This set includes all of the qualified rushers in the NFL during this time frame. The most statistically significant model was the model that included Yards per Carry, Touchdown Rate, Fumble Rate and Rushing Attempts. These were seen as measures of ability to consistently move the ball, ultimate success and failure respectively, and the durability of each rushing player. The results of the players were compared to the results of the all-time best rushers from Bleacher Report who played during the same time period in order to determine if the all-time best rushers were truly the bets rushers of the era. The models were created to be used with data from the entire career of a player and may not be accurate for shorter lengths of time.
Description
v, 35 p.
Citation
Publisher
Kalamazoo College
License
U.S. copyright laws protect this material. Commercial use or distribution of this material is not permitted without prior written permission of the copyright holder.
Journal
Volume
Issue
PubMed ID
DOI
ISSN
EISSN