Name
Session 2, Speaker 3: Ron Yurko, Assistant Teaching Professor of Statistics & Data Science, Carnegie Mellon University
Date & Time
Saturday, April 12, 2025, 4:10 PM - 4:30 PM
Description
Player tracking data in American football have made it possible to create new measures of defensive ability at halting the ball-carrier's forward progress, addressing the shortcomings of subjective tackle statistics. In this work, we introduce a Bayesian hierarchical framework for modeling the change in ball-carrier momentum (mass times velocity) during contact windows (i.e., tackle opportunities defined by a distance threshold) within a play. Our modeling framework accounts for ball-carrier and defender random effects, with separate distributions for each position. This approach enables us to quantify differences in tackling ability between defensive positions and allows us to compare positional sources of variation. Furthermore, we model the rate at which the ball-carrier's momentum is being reduced by defenders and demonstrate that linebackers excel at quickly suppressing the ball-carrier's forward progress. Our results from the first nine weeks of the 2022 NFL season reveal the top NFL defenders in terms of tackling ability, as well as the best (Derrick Henry) and worst (Alvin Kamara) ball-carriers at maintaining momentum through contact.
Location Name
1401
Full Address
Kline Tower
219 Prospect St
13th and 14th Floors, Registration Table in Room 1401
New Haven, CT 06511
United States
Title
Momentum Interference: A Hierarchical Framework For Modeling Tackling Ability In American Football
Abstract
Player tracking data in American football have made it possible to create new measures of defensive ability at halting the ball-carrier's forward progress, addressing the shortcomings of subjective tackle statistics. In this work, we introduce a Bayesian hierarchical framework for modeling the change in ball-carrier momentum (mass times velocity) during contact windows (i.e., tackle opportunities defined by a distance threshold) within a play. Our modeling framework accounts for ball-carrier and defender random effects, with separate distributions for each position. This approach enables us to quantify differences in tackling ability between defensive positions and allows us to compare positional sources of variation. Furthermore, we model the rate at which the ball-carrier's momentum is being reduced by defenders and demonstrate that linebackers excel at quickly suppressing the ball-carrier's forward progress. Our results from the first nine weeks of the 2022 NFL season reveal the top NFL defenders in terms of tackling ability, as well as the best (Derrick Henry) and worst (Alvin Kamara) ball-carriers at maintaining momentum through contact.
Speaker Bio
Ron Yurko is an Assistant Teaching Professor in the Department of Statistics & Data Science at Carnegie Mellon University, and is the Director of the Carnegie Mellon Sports Analytics Center. His research focuses on developing methods at the interface of inference and machine learning, oriented towards problems in sports analytics and natural language processing. His work has been featured in popular media outlets such as The Athletic, FiveThirtyEight, The Wall Street Journal, and The Washington Post. He is a three-time degree holder from Carnegie Mellon: with a bachelors, masters, and PhD in Statistics. He also has industry experience in both finance and professional sports.
Speaker Headshot
