Why Pass Rush Win Rate Lies to You

· 3 min read
nfl statistics

Pass rush win rate (PRWR) has become the go-to metric for evaluating edge rushers and interior defensive linemen. ESPN popularized it. Twitter repeats it. Front offices reference it. But most people citing PRWR don’t understand what it actually captures, and more importantly, what it misses.

What PRWR actually measures

ESPN’s pass rush win rate tracks whether a pass rusher “wins” his individual matchup within 2.5 seconds of the snap. A “win” means beating the blocker, not necessarily getting to the quarterback. The data comes from NFL Next Gen Stats player tracking.

The 2.5-second threshold is the key detail everyone ignores. If a rusher wins at 2.6 seconds, it doesn’t count. If he wins at 2.4 seconds but the QB already threw the ball, the win still counts even though it produced zero pressure.

Where it breaks down

Consider two edge rushers:

import pandas as pd

rushers = pd.DataFrame({
    "player": ["Rusher A", "Rusher B"],
    "prwr": [25.1, 18.3],
    "pressures": [42, 58],
    "sacks": [8, 12],
    "hurries": [51, 67],
})

print(rushers.to_string(index=False))
Output
player prwr pressures sacks hurries Rusher A 25.1 42 8 51 Rusher B 18.3 58 12 67

Rusher A has a significantly higher PRWR, but Rusher B generates more pressures, sacks, and hurries. How? A few reasons:

  1. Snap count context. Rusher B might rush the passer on 150 more snaps. Volume matters.
  2. Win conversion. Not all wins produce pressure. Rusher A might win his rep but get redirected before reaching the pocket.
  3. Scheme effects. If Rusher A faces constant double teams, his PRWR drops, but that might be creating one-on-one opportunities for teammates.

The 2.5-second problem

The arbitrary cutoff creates systematic bias. Quick-twitch speed rushers look better. Power rushers who collapse the pocket at 3+ seconds look worse. Bull rushes that move the entire pocket backward but take 3.1 seconds? Not a win.

This isn’t hypothetical. Some of the best interior rushers in the league, guys who physically move 300-pound guards into the quarterback’s lap, grade out worse in PRWR than smaller, faster players who technically “beat” their blocker but never actually disrupt the play.

A better framework

Instead of treating PRWR as the definitive pass rush metric, think of it as one input in a broader evaluation:

  • PRWR tells you about initial snap-to-engagement speed
  • Pressure rate tells you about actual disruption
  • ESPN’s pass rush win rate + PFF pressure grades together give a more complete picture
  • Team-level adjusted sack rate captures scheme and supporting cast effects

The best analysts use all of these together. The worst analysts tweet a PRWR ranking and call it a day.

The Box quote applies here

“All models are wrong; some are useful.” PRWR is useful. It measures a real thing, and at the extremes, it’s highly predictive. A guy with a 30%+ PRWR is almost certainly elite. But treating it as ground truth, especially in the 15-22% range where most starters live, is exactly the kind of false precision that gives analytics a bad name.

Context always beats a single number. That’s true in statistics, and it’s true in football.

← all posts