Do NFL Teams Run Too Much?

Short answer: yes. Long answer: yes, and the data has been screaming this for a decade.

The Numbers

Since the NFL began tracking detailed play-by-play data, one finding has remained remarkably consistent: passing plays generate more Expected Points Added (EPA) than rushing plays at nearly every down and distance.

Play TypeAvg EPA/Play (2020-2025)
Pass+0.07
Rush-0.05
Difference0.12 EPA/play

That gap compounds. Over a 16-game season, a team that shifted just 5% of their rush attempts to passes could expect roughly 0.5 additional wins — purely from play-calling optimization.

"But You Need to Run to Set Up the Pass"

This is the most common counterargument, and it's largely unsupported:

  1. Play-action doesn't require rushing — Play-action passing is highly effective regardless of a team's rushing volume. Multiple studies have shown no statistical correlation between rushing frequency and play-action success.
  2. Rushing doesn't "open up" the pass — Defenses adjust to tendencies in real time, not based on season-long run/pass ratios.
  3. The causation is backwards — Teams that pass well build leads, and then run to burn clock. The rushing stats are a result of winning, not the cause.

Why Coaches Still Run

Several factors keep the run game alive beyond what analytics would suggest:

  • Clock management — Running does kill clock, which has genuine tactical value with a lead
  • Quarterback preservation — Fewer dropbacks means fewer hits
  • Cognitive bias — Coaches remember the big rushing game, not the 200 games where the run was stuffed for 3.2 YPC
  • Accountability — A failed run feels like bad luck; a failed pass feels like a bad decision

The Exception

There is a legitimate case for rushing: short-yardage and goal-line situations, where the EPA gap narrows significantly. Rushing also has value in end-of-game clock scenarios. Nobody is arguing for zero rushes.

The argument is about degree. Most NFL teams still run the ball 40-45% of the time in neutral game scripts. The optimal rate, based on EPA and win probability modeling, is likely closer to 30-35%.

The Skeptic's Take

The NFL is a passing league that coaches still treat as a rushing league. The teams that figure this out first — and most thoroughly — gain a real competitive advantage. The data is clear. The question is whether coaches will listen.


Related: Quarterback Wins Are Not a Stat