Game quality
What each result was worth beyond the score. Every play is graded by EPA (expected points added), then adjusted for who it came against: a defence's rating is what it allowed to everyone else that season, so a big day against a bad defence counts for less. Season views include playoff games.
Davis Mills — 2021
Adj EPA / dropback
-0.100
Raw EPA / dropback
-0.104
Dropbacks
424
Turnover EPA
-48.9
WPA
-0.54
| Week | Opp | Dropbacks | Adj EPA/db | Raw EPA/db | Schedule | Total adj EPA | INT / FL | Turnover EPA | Sacks | Sack EPA | WPA | CPOE |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | CLE | 19 | +0.042 | +0.024 | +0.019 | +0.8 | 1 / 0 | -3.4 | 1 | -1.3 | -0.08 | -16.9 |
| 3 | CAR | 31 | -0.240 | -0.233 | -0.007 | -7.4 | 0 / 0 | +0.0 | 4 | -7.0 | -0.10 | +6.1 |
| 4 | BUF | 24 | -0.905 | -0.955 | +0.050 | -21.7 | 4 / 0 | -16.3 | 3 | -4.8 | -0.43 | -15.9 |
| 5 | NE | 32 | +0.613 | +0.560 | +0.053 | +19.6 | 0 / 0 | +0.0 | 3 | -3.8 | +0.43 | +9.3 |
| 6 | IND | 45 | -0.196 | -0.200 | +0.004 | -8.8 | 2 / 0 | -9.9 | 2 | -3.2 | -0.19 | -2.0 |
| 7 | ARI | 34 | -0.472 | -0.482 | +0.010 | -16.1 | 0 / 1 | -5.9 | 2 | -8.0 | -0.16 | +0.1 |
| 8 | LA | 43 | -0.064 | -0.160 | +0.096 | -2.7 | 1 / 0 | -5.2 | 5 | -9.2 | -0.29 | +8.4 |
| 13 | IND | 16 | -0.774 | -0.767 | -0.007 | -12.4 | 0 / 0 | +0.0 | 2 | -3.8 | -0.01 | -25.4 |
| 14 | SEA | 50 | -0.067 | -0.040 | -0.027 | -3.3 | 0 / 0 | +0.0 | 2 | -2.8 | -0.03 | -3.5 |
| 15 | JAX | 31 | -0.022 | +0.113 | -0.135 | -0.7 | 1 / 0 | -3.6 | 1 | -0.8 | +0.01 | -2.2 |
| 16 | LAC | 28 | +0.578 | +0.572 | +0.006 | +16.2 | 0 / 0 | +0.0 | 1 | -1.8 | +0.46 | +7.6 |
| 17 | SF | 35 | -0.455 | -0.458 | +0.003 | -15.9 | 1 / 0 | -4.5 | 3 | -5.7 | -0.38 | +1.2 |
| 18 | TEN | 36 | +0.284 | +0.296 | -0.013 | +10.2 | 0 / 0 | +0.0 | 2 | -3.0 | +0.24 | +6.8 |
Opponent adjustment. A unit's rating for a game is its average over its other games that
season, blended with 16 games' worth of prior (league average plus half of last season's gap from it;
24 games for pass defence, which is noisier). Early in a season that leans on last year; by
midseason it is mostly this year. Chosen by how well it predicts a held-out game: it cut that error by about
a quarter against no shrinkage, and adjusted numbers agree between odd and even weeks more than raw ones do
(team EPA margin r 0.52 → 0.55, defence 0.32 → 0.36, QB EPA/dropback 0.47 → 0.49, 2016–2025).
Win expectancy is a logistic fit of wins on EPA margin over 6,036 completed games (it names the winner 86.5% of the time). These describe what happened; one game is not a forecast. Play-by-play credits the quarterback with every sack and dropback, so a bad line shows up in his numbers. Seasons 2016–2026, built 2026-09-14 01:25 UTC, from nflverse. Write-up:
Win expectancy is a logistic fit of wins on EPA margin over 6,036 completed games (it names the winner 86.5% of the time). These describe what happened; one game is not a forecast. Play-by-play credits the quarterback with every sack and dropback, so a bad line shows up in his numbers. Seasons 2016–2026, built 2026-09-14 01:25 UTC, from nflverse. Write-up:
backtests/nfl_game_quality/FINDINGS.md.