The Analytics Behind Winning: ML Models of NFL Performance

Authors

• Derek Bernardino, St. John Fisher University
• Dylan Devine, St. John Fisher University
• Noah Geiger, St. John Fisher University
• Ricky Hill, St. John Fisher University

Document Type

Poster Presentation

Publication Date

4-17-2026

Keywords

fsc2026

Abstract

Our project applies machine learning techniques to predict NFL team wins using historical data spanning the 2020–2025 seasons. Drawing on a comprehensive set of team-level statistics including offensive and rushing performance, passing yardage, points scored and allowed, point differential, and win-loss records, we trained regression models to identify the key drivers of team success. Our analysis incorporated features across offensive, defensive, and divisional dimensions to capture the multifaceted nature of NFL performance. The resulting model demonstrated strong predictive accuracy, achieving an R² of 0.92, indicating that over 90% of the variance in team wins can be explained by the selected features. These findings suggest that a data-driven approach can reliably forecast team performance.

Comments

Poster presented at the 2026 Fisher Showcase, St. John Fisher University, April 17, 2026.

Publication Information

Bernardino, Derek; Devine, Dylan; Geiger, Noah; and Hill, Ricky, "The Analytics Behind Winning: ML Models of NFL Performance" (2026). Fisher Showcase 2026. Paper 119.
https://fisherpub.sjf.edu/fsc2026/119

Please note that the Publication Information provides general citation information and may not be appropriate for your discipline. To receive help in creating a citation based on your discipline, please visit https://libguides.sjf.edu/citations.