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.
Games
Adj margin = offence minus defence EPA per play after the opponent adjustment. Win exp = how often a team with that adjusted margin wins; a 55% win was a coin flip that landed. Win prob range = lowest and highest in-game chance, which is where blown leads and comebacks show.
| Team | Score | vs line | Off EPA/playraw → adj | Def EPA/playraw → adj | Adj margin | Win exp | Win prob range | Turnoverslost / forced |
|---|---|---|---|---|---|---|---|---|
| DAL W | 31–14 | +14.5fav 2.5 | +0.37 → +0.40 | -0.05 → -0.06 | +0.47 | 99%raw 100% | 34% – 100% | 0 / 1 |
| @ TB L | 14–31 | -14.5dog 2.5 | -0.05 → +0.02 | +0.37 → +0.35 | -0.33 | 2%raw 0% | 0% – 66% | 1 / 0 |
| SEA L | 23–41 | -8.0dog 10.0 | +0.08 → +0.15 | +0.39 → +0.34 | -0.19 | 11%raw 2% | 0% – 52% | 2 / 0 |
| @ SF W | 41–23 | +8.0fav 10.0 | +0.39 → +0.37 | +0.08 → +0.07 | +0.29 | 96%raw 98% | 48% – 100% | 0 / 2 |
| NYG W | 31–24 | +9.5dog 2.5 | +0.37 → +0.36 | +0.17 → +0.17 | +0.19 | 89%raw 94% | 27% – 99% | 0 / 0 |
| @ MIN L | 24–31 | -9.5fav 2.5 | +0.17 → +0.13 | +0.37 → +0.40 | -0.27 | 5%raw 6% | 1% – 73% | 0 / 0 |
| BAL L | 17–24 | +0.5dog 7.5 | -0.09 → -0.07 | +0.04 → -0.02 | -0.05 | 37%raw 15% | 8% – 67% | 2 / 1 |
| @ CIN W | 24–17 | -0.5fav 7.5 | +0.04 → +0.03 | -0.09 → -0.10 | +0.12 | 80%raw 85% | 33% – 92% | 1 / 2 |
| MIA L | 31–34 | +11.0dog 14.0 | -0.23 → -0.19 | -0.15 → -0.25 | +0.07 | 67%raw 26% | 3% – 71% | 2 / 3 |
| @ BUF W | 34–31 | -11.0fav 14.0 | -0.15 → -0.15 | -0.23 → -0.24 | +0.10 | 74%raw 74% | 29% – 97% | 3 / 2 |
| LAC L | 30–31 | -3.0fav 2.0 | +0.07 → +0.06 | -0.03 → -0.03 | +0.08 | 72%raw 80% | 29% – 100% | 0 / 4 |
| @ JAX W | 31–30 | +3.0dog 2.0 | -0.03 → -0.04 | +0.07 → +0.05 | -0.08 | 29%raw 20% | 0% – 71% | 4 / 0 |
Quarterbacks this week
Click a column to sort. Schedule = what the opponent adjustment added (+ = faced a tough pass defence). Turnover and sack EPA are what those plays cost; sacks are shared with the line.
| Quarterback | Team | Opp | Dropbacks | Adj EPA/db | Raw EPA/db | Schedule | Total adj EPA | INT / FL | Turnover EPA | Sacks | Sack EPA | WPA | CPOE |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dak Prescott | DAL | TB | 34 | +0.728 | +0.702 | +0.026 | +24.8 | 0 / 0 | +0.0 | 1 | -2.2 | +0.33 | +13.4 |
| Brock Purdy | SF | SEA | 32 | +0.681 | +0.704 | -0.022 | +21.8 | 0 / 0 | +0.0 | 1 | -1.5 | +0.39 | +5.0 |
| Daniel Jones | NYG | MIN | 38 | +0.359 | +0.358 | +0.001 | +13.7 | 0 / 0 | +0.0 | 3 | -4.4 | +0.49 | -1.3 |
| Kirk Cousins | MIN | NYG | 39 | +0.313 | +0.314 | -0.001 | +12.2 | 0 / 0 | +0.0 | 0 | +0.0 | +0.39 | +13.7 |
| Tyler Huntley | BAL | CIN | 31 | +0.129 | +0.110 | +0.019 | +4.0 | 1 / 0 | -4.7 | 2 | -3.4 | +0.09 | -4.3 |
| Joe Burrow | CIN | BAL | 37 | +0.122 | +0.157 | -0.035 | +4.5 | 0 / 0 | +0.0 | 4 | -5.9 | +0.05 | +1.9 |
| Geno Smith | SEA | SF | 39 | +0.113 | +0.072 | +0.040 | +4.4 | 1 / 1 | -9.3 | 3 | -9.0 | +0.20 | +10.7 |
| Justin Herbert | LAC | JAX | 46 | +0.110 | +0.166 | -0.056 | +5.1 | 0 / 0 | +0.0 | 3 | -1.5 | +0.08 | -7.9 |
| Tom Brady | TB | DAL | 70 | +0.037 | -0.051 | +0.088 | +2.6 | 1 / 0 | -5.7 | 2 | -3.7 | -0.19 | -13.0 |
| Trevor Lawrence | JAX | LAC | 49 | -0.107 | -0.122 | +0.015 | -5.2 | 4 / 0 | -16.5 | 2 | -3.2 | -0.20 | -6.8 |
| Skylar Thompson | MIA | BUF | 50 | -0.131 | -0.193 | +0.062 | -6.5 | 2 / 0 | -7.0 | 4 | -6.1 | -0.31 | -17.0 |
| Josh Allen | BUF | MIA | 46 | -0.253 | -0.257 | +0.004 | -11.6 | 2 / 1 | -18.6 | 7 | -15.9 | -0.24 | +5.5 |
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.