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, 2026
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 |
|---|---|---|---|---|---|---|---|---|---|
| JAX | 6–2 | +0.168 | -0.133 | +0.300 | +0.287 | 6.4 | -0.4 | +9.1 | +8 |
| SF | 6–1 | +0.197 | -0.055 | +0.253 | +0.247 | 5.8 | +0.2 | +4.5 | -2 |
| BUF | 6–1 | +0.204 | +0.035 | +0.169 | +0.146 | 5.2 | +0.8 | +3.7 | -2 |
| KC | 5–2 | +0.134 | -0.023 | +0.157 | +0.169 | 4.9 | +0.1 | +1.4 | +5 |
| LA | 4–4 | +0.032 | -0.115 | +0.147 | +0.153 | 5.4 | -1.4 | -1.5 | -4 |
| CIN | 5–2 | +0.134 | +0.008 | +0.127 | +0.129 | 4.9 | +0.1 | +2.6 | +2 |
| CHI | 6–2 | +0.082 | -0.038 | +0.121 | +0.113 | 5.8 | +0.2 | +5.4 | -4 |
| MIN | 6–1 | -0.118 | -0.192 | +0.074 | +0.074 | 4.6 | +1.4 | +4.6 | +6 |
| GB | 5–3 | -0.008 | -0.066 | +0.058 | +0.066 | 4.5 | +0.5 | +0.1 | +5 |
| SEA | 4–4 | -0.062 | -0.120 | +0.058 | +0.040 | 4.5 | -0.5 | -0.9 | -3 |
| TB | 3–5 | -0.094 | -0.152 | +0.057 | +0.068 | 4.3 | -1.3 | +4.9 | +2 |
| NE | 5–3 | +0.065 | +0.009 | +0.055 | +0.041 | 4.8 | +0.2 | +1.9 | -7 |
| DEN | 6–2 | -0.008 | -0.048 | +0.040 | -0.014 | 5.0 | +1.0 | +0.2 | +5 |
| PIT | 6–2 | -0.001 | -0.039 | +0.037 | +0.054 | 4.3 | +1.7 | +2.9 | +4 |
| CAR | 3–4 | +0.045 | +0.032 | +0.013 | +0.009 | 3.3 | -0.3 | -2.1 | +0 |
| NYJ | 2–6 | +0.025 | +0.024 | +0.001 | +0.024 | 4.0 | -2.0 | -0.3 | +1 |
| HOU | 2–5 | +0.009 | +0.009 | -0.000 | +0.003 | 3.4 | -1.4 | -3.6 | +3 |
| DET | 5–2 | +0.116 | +0.116 | -0.000 | +0.012 | 3.4 | +1.6 | -0.3 | +3 |
| LV | 4–4 | +0.025 | +0.070 | -0.045 | -0.052 | 3.6 | +0.4 | +2.1 | +7 |
| NYG | 4–2–1 | -0.029 | +0.017 | -0.047 | -0.015 | 3.3 | +1.2 | +3.8 | -2 |
| DAL | 3–5 | +0.094 | +0.141 | -0.048 | -0.033 | 3.0 | -0.0 | -2.1 | +0 |
| WAS | 2–4–1 | -0.028 | +0.029 | -0.056 | -0.048 | 2.7 | -0.2 | -0.4 | +7 |
| BAL | 4–5 | -0.042 | +0.015 | -0.057 | -0.040 | 3.8 | +0.2 | -6.4 | -1 |
| ATL | 3–5 | -0.109 | -0.049 | -0.060 | -0.069 | 3.6 | -0.6 | +0.0 | -4 |
| CLE | 5–3 | -0.088 | +0.007 | -0.096 | -0.076 | 3.2 | +1.8 | -0.6 | +0 |
| ARI | 3–5 | -0.036 | +0.061 | -0.096 | -0.118 | 2.8 | +0.2 | -2.1 | -1 |
| NO | 2–5 | -0.050 | +0.115 | -0.164 | -0.141 | 2.0 | +0.0 | -2.5 | -5 |
| IND | 2–6 | -0.112 | +0.064 | -0.176 | -0.176 | 2.6 | -0.6 | -5.2 | -11 |
| LAC | 0–7 | -0.256 | -0.066 | -0.190 | -0.214 | 1.7 | -1.7 | -8.5 | -6 |
| MIA | 1–6 | -0.060 | +0.150 | -0.209 | -0.205 | 0.9 | +0.1 | -1.6 | -3 |
| TEN | 1–7 | -0.117 | +0.096 | -0.213 | -0.176 | 1.6 | -0.6 | -0.9 | +1 |
| PHI | 2–6 | -0.162 | +0.088 | -0.250 | -0.234 | 0.8 | +1.2 | -7.1 | -4 |
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.