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.
Matt Ryan — 2021
Adj EPA / dropback
+0.021
Raw EPA / dropback
+0.011
Dropbacks
601
Turnover EPA
-67.2
WPA
+1.22
| Week | Opp | Dropbacks | Adj EPA/db | Raw EPA/db | Schedule | Total adj EPA | INT / FL | Turnover EPA | Sacks | Sack EPA | WPA | CPOE |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | PHI | 38 | -0.452 | -0.420 | -0.032 | -17.2 | 0 / 0 | +0.0 | 3 | -8.4 | -0.06 | -9.3 |
| 2 | TB | 47 | -0.074 | -0.134 | +0.060 | -3.5 | 3 / 0 | -21.6 | 1 | -1.7 | +0.08 | +7.2 |
| 3 | NYG | 39 | +0.129 | +0.132 | -0.003 | +5.0 | 0 / 1 | -1.3 | 3 | -4.0 | +0.21 | +4.2 |
| 4 | WAS | 43 | +0.289 | +0.289 | +0.000 | +12.4 | 0 / 0 | +0.0 | 1 | -2.0 | +0.24 | -1.8 |
| 5 | NYJ | 45 | +0.370 | +0.474 | -0.104 | +16.6 | 0 / 0 | +0.0 | 0 | +0.0 | +0.34 | +6.4 |
| 7 | MIA | 41 | +0.335 | +0.273 | +0.063 | +13.7 | 1 / 0 | -3.3 | 1 | -1.7 | +0.45 | -3.3 |
| 8 | CAR | 30 | -0.285 | -0.277 | -0.008 | -8.5 | 2 / 0 | -7.7 | 3 | -5.1 | -0.21 | +5.2 |
| 9 | NO | 32 | +0.657 | +0.541 | +0.115 | +21.0 | 0 / 0 | +0.0 | 2 | -2.2 | +0.71 | +14.8 |
| 10 | DAL | 23 | -0.585 | -0.648 | +0.064 | -13.4 | 2 / 0 | -8.8 | 2 | -2.7 | -0.07 | -23.5 |
| 11 | NE | 32 | -0.360 | -0.382 | +0.023 | -11.5 | 2 / 0 | -8.9 | 4 | -6.9 | -0.24 | -2.9 |
| 12 | JAX | 30 | +0.029 | +0.162 | -0.134 | +0.9 | 1 / 0 | -1.5 | 1 | -1.6 | +0.08 | -2.1 |
| 13 | TB | 46 | +0.050 | -0.015 | +0.065 | +2.3 | 0 / 0 | +0.0 | 5 | -10.3 | -0.06 | +4.8 |
| 14 | CAR | 29 | +0.183 | +0.179 | +0.003 | +5.3 | 0 / 0 | +0.0 | 0 | +0.0 | +0.13 | +8.4 |
| 15 | SF | 35 | -0.188 | -0.198 | +0.010 | -6.6 | 0 / 1 | -4.4 | 3 | -7.6 | -0.10 | +3.5 |
| 16 | DET | 27 | +0.174 | +0.314 | -0.139 | +4.7 | 0 / 0 | +0.0 | 3 | -3.5 | +0.23 | +9.0 |
| 17 | BUF | 28 | -0.285 | -0.347 | +0.062 | -8.0 | 0 / 1 | -5.3 | 5 | -10.4 | -0.23 | -0.4 |
| 18 | NO | 36 | -0.025 | -0.123 | +0.098 | -0.9 | 1 / 0 | -4.4 | 3 | -4.6 | -0.27 | +1.1 |
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.