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.
Ryan Tannehill — 2021
Adj EPA / dropback
+0.018
Raw EPA / dropback
+0.035
Dropbacks
603
Turnover EPA
-89.0
WPA
+1.00
| Week | Opp | Dropbacks | Adj EPA/db | Raw EPA/db | Schedule | Total adj EPA | INT / FL | Turnover EPA | Sacks | Sack EPA | WPA | CPOE |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | ARI | 41 | -0.467 | -0.480 | +0.014 | -19.1 | 1 / 2 | -14.6 | 6 | -20.8 | -0.25 | -6.6 |
| 2 | SEA | 43 | +0.154 | +0.173 | -0.019 | +6.6 | 0 / 1 | -5.5 | 3 | -8.7 | -0.04 | +5.4 |
| 3 | IND | 28 | +0.133 | +0.118 | +0.015 | +3.7 | 2 / 0 | -9.1 | 1 | -2.2 | -0.08 | +1.2 |
| 4 | NYJ | 57 | -0.110 | +0.005 | -0.115 | -6.3 | 0 / 0 | +0.0 | 7 | -10.1 | +0.13 | -2.4 |
| 5 | JAX | 25 | +0.135 | +0.263 | -0.128 | +3.4 | 0 / 0 | +0.0 | 3 | -4.1 | +0.28 | +0.2 |
| 6 | BUF | 29 | +0.320 | +0.242 | +0.078 | +9.3 | 1 / 0 | -4.4 | 0 | +0.0 | +0.43 | +2.6 |
| 7 | KC | 27 | +0.504 | +0.516 | -0.013 | +13.6 | 1 / 0 | -4.8 | 1 | -1.3 | +0.25 | +14.9 |
| 8 | IND | 36 | +0.231 | +0.218 | +0.014 | +8.3 | 2 / 0 | -4.9 | 3 | -2.9 | +0.43 | +2.5 |
| 9 | LA | 30 | -0.045 | -0.141 | +0.097 | -1.3 | 1 / 0 | -3.5 | 3 | -2.9 | -0.04 | +0.4 |
| 10 | NO | 29 | +0.187 | +0.084 | +0.103 | +5.4 | 0 / 0 | +0.0 | 2 | -3.8 | +0.09 | +2.6 |
| 11 | HOU | 54 | -0.287 | -0.185 | -0.102 | -15.5 | 4 / 0 | -20.4 | 2 | -2.1 | -0.36 | +3.8 |
| 12 | NE | 23 | -0.378 | -0.409 | +0.032 | -8.7 | 1 / 0 | -3.9 | 2 | -3.4 | -0.09 | -13.2 |
| 14 | JAX | 35 | -0.111 | +0.022 | -0.133 | -3.9 | 0 / 0 | +0.0 | 4 | -5.8 | +0.06 | -3.1 |
| 15 | PIT | 36 | -0.193 | -0.240 | +0.047 | -7.0 | 1 / 0 | -5.9 | 4 | -6.1 | -0.59 | +5.2 |
| 16 | SF | 33 | +0.255 | +0.235 | +0.020 | +8.4 | 0 / 0 | +0.0 | 4 | -5.3 | +0.28 | +11.9 |
| 17 | MIA | 19 | +0.139 | +0.079 | +0.060 | +2.6 | 0 / 0 | +0.0 | 1 | -2.6 | +0.03 | +6.8 |
| 18 | HOU | 33 | +0.485 | +0.568 | -0.083 | +16.0 | 0 / 0 | +0.0 | 1 | -2.4 | +0.59 | +7.9 |
| 20 | CIN | 25 | -0.199 | -0.178 | -0.021 | -5.0 | 3 / 0 | -12.1 | 1 | -2.1 | -0.12 | -0.3 |
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.