Game quality
How well each team played, beyond the score: EPA (expected points added) per play, adjusted for the opponent. A big day against a bad defence counts for less. Season views include playoff games.
Teams, 2025
Per-game averages. Expected wins sums each game's win expectancy, so wins minus expected is mostly luck. Click a column to sort.
| Team | Record | Adj off EPA | Adj def EPA | Adj margin | Raw margin | Expected wins | Wins − expected | Avg vs line | Turnover diff |
|---|---|---|---|---|---|---|---|---|---|
| SEA | 17–3 | +0.049 | -0.155 | +0.204 | +0.199 | 15.4 | +1.6 | +8.0 | +6 |
| LA | 14–6 | +0.143 | -0.055 | +0.198 | +0.184 | 15.6 | -1.6 | +3.0 | +14 |
| NE | 17–4 | +0.099 | -0.071 | +0.170 | +0.192 | 15.9 | +1.1 | +5.3 | +2 |
| HOU | 13–6 | -0.024 | -0.177 | +0.152 | +0.139 | 13.0 | +0.0 | +4.7 | +13 |
| BUF | 13–6 | +0.150 | -0.002 | +0.152 | +0.159 | 13.8 | -0.8 | +0.7 | +0 |
| JAX | 13–5 | +0.037 | -0.084 | +0.121 | +0.129 | 11.6 | +1.4 | +6.1 | +9 |
| PHI | 11–7 | +0.013 | -0.095 | +0.108 | +0.104 | 12.1 | -1.1 | -1.4 | +9 |
| DEN | 15–4 | +0.021 | -0.086 | +0.107 | +0.117 | 13.0 | +2.0 | +0.6 | +0 |
| DET | 9–8 | +0.081 | -0.016 | +0.096 | +0.085 | 10.2 | -1.2 | +0.2 | +4 |
| GB | 9–8–1 | +0.115 | +0.022 | +0.093 | +0.094 | 11.0 | -1.5 | -2.3 | +2 |
| IND | 8–9 | +0.095 | +0.016 | +0.079 | +0.081 | 10.1 | -2.1 | +2.5 | +0 |
| KC | 6–11 | +0.032 | -0.036 | +0.068 | +0.062 | 9.4 | -3.4 | -1.0 | -2 |
| CHI | 12–7 | +0.074 | +0.012 | +0.062 | +0.053 | 11.6 | +0.4 | +2.3 | +17 |
| LAC | 11–7 | -0.037 | -0.066 | +0.029 | +0.040 | 10.1 | +0.9 | -0.4 | +4 |
| BAL | 8–9 | +0.044 | +0.031 | +0.013 | +0.011 | 9.2 | -1.2 | -2.5 | -3 |
| MIN | 9–8 | -0.142 | -0.137 | -0.005 | -0.021 | 7.4 | +1.6 | +1.9 | -8 |
| SF | 13–6 | +0.064 | +0.076 | -0.012 | -0.019 | 9.3 | +3.7 | +1.0 | -13 |
| TB | 8–9 | +0.004 | +0.023 | -0.019 | -0.023 | 7.7 | +0.3 | -3.0 | +6 |
| PIT | 10–8 | -0.000 | +0.025 | -0.025 | -0.038 | 9.0 | +1.0 | -0.4 | +12 |
| ATL | 8–9 | -0.041 | -0.012 | -0.028 | -0.018 | 8.2 | -0.2 | -2.0 | +5 |
| NO | 6–11 | -0.109 | -0.065 | -0.044 | -0.021 | 7.1 | -1.1 | -0.2 | -3 |
| NYG | 4–13 | +0.006 | +0.072 | -0.066 | -0.076 | 7.0 | -3.0 | +1.4 | -1 |
| DAL | 7–9–1 | +0.090 | +0.178 | -0.088 | -0.076 | 6.3 | +1.2 | -2.7 | -10 |
| ARI | 3–14 | -0.016 | +0.083 | -0.099 | -0.104 | 5.7 | -2.7 | -4.1 | -1 |
| MIA | 7–10 | -0.030 | +0.088 | -0.118 | -0.104 | 5.7 | +1.3 | -1.4 | -5 |
| CLE | 5–12 | -0.217 | -0.097 | -0.120 | -0.105 | 5.1 | -0.1 | -0.5 | -8 |
| CAR | 8–10 | -0.059 | +0.065 | -0.124 | -0.131 | 6.5 | +1.5 | +0.7 | -3 |
| WAS | 5–12 | +0.015 | +0.146 | -0.132 | -0.135 | 5.5 | -0.5 | -2.9 | -12 |
| CIN | 6–11 | -0.015 | +0.124 | -0.139 | -0.143 | 5.4 | +0.6 | -2.4 | -3 |
| LV | 3–14 | -0.217 | +0.026 | -0.243 | -0.245 | 4.3 | -1.3 | -6.0 | -6 |
| TEN | 3–14 | -0.140 | +0.121 | -0.261 | -0.274 | 3.0 | +0.0 | -3.8 | -7 |
| NYJ | 3–14 | -0.155 | +0.148 | -0.303 | -0.296 | 2.3 | +0.7 | -5.7 | -18 |
How the ratings are built
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 2026-11-07 11:05 UTC, from nflverse. Write-up:
backtests/nfl_game_quality/FINDINGS.md.