Development and Properties of a Pace-Adjusted Offensive Efficiency Metric in the NFL (2020–2025)

Abstract

We propose and validate a simple pace-adjusted efficiency metric, Eff = PPG / (YPG × SPP), where PPG is points per game, YPG is yards per game, and SPP is seconds per play. Using team-level data for all 32 NFL teams across the 2020–2025 seasons, we demonstrate that this metric captures meaningful variation in offensive performance beyond raw scoring or yardage. The metric shows moderate year-to-year stability and aligns directionally with advanced analytics such as EPA/play and Success Rate.

Introduction

Traditional NFL efficiency measures (e.g., yards per play or points per game) often overlook tempo. Our metric normalizes scoring output by both volume (YPG) and pace (SPP), providing a tempo-adjusted view useful for fantasy football, betting, and roster evaluation.

Methods

Data were compiled from Pro-Football-Reference (YPG, PPG), Sharp Football Analysis and StatRankings (SPP), and nflfastR-derived proxies for EPA/play and Success Rate. The efficiency metric is defined as:

Eff=PPGYPG×SPP\text{Eff} = \frac{\text{PPG}}{\text{YPG} \times \text{SPP}}

\text{Eff} = \frac{\text{PPG}}{\text{YPG} \times \text{SPP}}

Home/Away splits were used to compute HA_EFF_Ratio = Eff_Home / Eff_Away. Teams were tiered annually into High/Mid/Low by Eff_Season.

 Descriptive Statistics and Raw Data Tables

League-wide Eff averaged approximately 0.0020–0.0024 across seasons, with notable variation. The metric correlated modestly with EPA/play proxies (~+0.086) and Success Rate (~+0.063). 

Year-to-year Spearman rank correlation for team Eff ranged from 0.48 to 0.62, indicating moderate stickiness. 

Tables 1–4 demonstrate both league-level stability and meaningful team-level differentiation. 

Discussion

This metric offers an interpretable, computationally lightweight alternative to complex EPA models while incorporating tempo. It behaves consistently with established advanced statistics. The raw data tables confirm that the metric captures real variation in offensive performance and responds logically to known scheduling factors.

Conclusion

The proposed pace-adjusted Eff metric is valid, stable, and practically useful for weekly fantasy and analytical workflows. The inclusion of raw data tables provides transparency and supports reproducibility.