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.

TeamScorevs lineOff EPA/playraw → adj Def EPA/playraw → adjAdj marginWin exp Win prob rangeTurnoverslost / forced
NE L 17–47 -25.5dog 4.5 +0.07 → +0.14 +0.77 → +0.68 -0.54 0%raw 0% 0% – 45% 2 / 0
@ BUF W 47–17 +25.5fav 4.5 +0.77 → +0.80 +0.07 → +0.05 +0.75 100%raw 100% 55% – 100% 0 / 2
ARI L 11–34 -19.0dog 4.0 -0.40 → -0.33 +0.10 → +0.07 -0.40 1%raw 0% 0% – 43% 2 / 0
@ LA W 34–11 +19.0fav 4.0 +0.10 → +0.13 -0.40 → -0.43 +0.57 100%raw 100% 57% – 100% 0 / 2
PIT L 21–42 -8.0dog 13.0 -0.11 → -0.15 +0.21 → +0.10 -0.24 6%raw 1% 0% – 63% 1 / 2
@ KC W 42–21 +8.0fav 13.0 +0.21 → +0.23 -0.11 → -0.07 +0.30 97%raw 99% 37% – 100% 2 / 1
SF W 23–17 +9.0dog 3.0 -0.05 → -0.00 -0.14 → -0.17 +0.17 86%raw 76% 43% – 98% 1 / 1
@ DAL L 17–23 -9.0fav 3.0 -0.14 → -0.10 -0.05 → -0.08 -0.02 45%raw 24% 2% – 57% 1 / 1
PHI L 15–31 -8.5dog 7.5 -0.11 → -0.04 +0.06 → -0.03 -0.01 46%raw 9% 0% – 45% 2 / 0
@ TB W 31–15 +8.5fav 7.5 +0.06 → +0.05 -0.11 → -0.11 +0.16 86%raw 91% 55% – 100% 0 / 2
LV L 19–26 -1.0dog 6.0 +0.05 → +0.04 +0.12 → +0.12 -0.08 30%raw 28% 1% – 55% 2 / 0
@ CIN W 26–19 +1.0fav 6.0 +0.12 → +0.08 +0.05 → +0.06 +0.02 55%raw 72% 45% – 99% 0 / 2

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.

QuarterbackTeam Opp DropbacksAdj EPA/dbRaw EPA/dbScheduleTotal adj EPA INT / FLTurnover EPASacksSack EPAWPACPOE
Josh Allen BUF NE 25 +0.981 +0.919 +0.062 +24.5 0 / 0 +0.0 0 +0.0 +0.13 +18.0
Matthew Stafford LA ARI 18 +0.733 +0.695 +0.038 +13.2 0 / 0 +0.0 1 -2.3 +0.16 +11.1
Patrick Mahomes KC PIT 42 +0.517 +0.453 +0.064 +21.7 1 / 0 -6.0 3 -5.3 +0.20 +10.6
Joe Burrow CIN LV 36 +0.219 +0.278 -0.059 +7.9 0 / 0 +0.0 2 -3.8 +0.33 +11.2
Tom Brady TB PHI 41 +0.183 +0.201 -0.017 +7.5 0 / 0 +0.0 4 -4.2 +0.06 +8.2
Mac Jones NE BUF 40 +0.153 +0.078 +0.075 +6.1 2 / 0 -9.7 3 -4.2 -0.09 +4.8
Jimmy Garoppolo SF DAL 25 +0.089 +0.006 +0.082 +2.2 1 / 0 -6.1 0 +0.0 +0.17 -5.3
Derek Carr LV CIN 56 -0.006 +0.008 -0.014 -0.4 1 / 1 -8.1 3 -8.8 -0.08 -6.1
Jalen Hurts PHI TB 46 -0.058 -0.120 +0.062 -2.7 2 / 0 -7.8 2 -2.2 -0.05 -7.5
Ben Roethlisberger PIT KC 46 -0.108 -0.079 -0.029 -5.0 0 / 0 +0.0 2 -3.6 -0.30 -0.4
Dak Prescott DAL SF 47 -0.249 -0.259 +0.011 -11.7 1 / 0 -4.3 5 -7.3 -0.17 -9.5
Kyler Murray ARI LA 36 -0.522 -0.610 +0.088 -18.8 2 / 0 -10.5 2 -3.3 -0.27 -15.3
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: backtests/nfl_game_quality/FINDINGS.md.