Backtest History
285 graded games
Every graded prediction from the frozen walk-forward backtest, next to the closing spread and the final margin. Lines are quoted for the home team.
| Season | Wk | Matchup | Score | Model | Close | Result | Model err |
|---|---|---|---|---|---|---|---|
| 2021 | 16 | DET @ ATL | 16–20 | -6.1 | -7.5 | -4.0 | 2.1 (model closer than the closing line) |
| 2021 | 16 | IND @ ARI | 22–16 | +0.4 | -3.0 | +6.0 | 5.6 (model closer than the closing line) |
| 2021 | 16 | JAX @ NYJ | 21–26 | -1.6 | -2.5 | -5.0 | 3.4 |
| 2021 | 16 | LA @ MIN | 30–23 | +2.1 | +3.5 | +7.0 | 4.9 |
| 2021 | 16 | LAC @ HOU | 29–41 | +12.3 | +13.5 | -12.0 | 24.3 (model closer than the closing line) |
| 2021 | 16 | MIA @ NO | 20–3 | +0.7 | +3.0 | +17.0 | 16.3 |
| 2021 | 16 | NYG @ PHI | 10–34 | -11.7 | -11.0 | -24.0 | 12.3 (model closer than the closing line) |
| 2021 | 16 | PIT @ KC | 10–36 | -13.2 | -10.5 | -26.0 | 12.8 (model closer than the closing line) |
| 2021 | 16 | SF @ TEN | 17–20 | +5.1 | +3.5 | -3.0 | 8.1 |
| 2021 | 16 | TB @ CAR | 32–6 | +10.3 | +11.5 | +26.0 | 15.7 |
| 2021 | 16 | WAS @ DAL | 14–56 | -9.8 | -10.5 | -42.0 | 32.2 |
| 2021 | 15 | ARI @ DET | 12–30 | +14.4 | +13.0 | -18.0 | 32.4 |
| 2021 | 15 | ATL @ SF | 13–31 | -10.1 | -9.5 | -18.0 | 7.9 (model closer than the closing line) |
| 2021 | 15 | CAR @ BUF | 14–31 | -10.6 | -14.0 | -17.0 | 6.4 |
| 2021 | 15 | CIN @ DEN | 15–10 | -3.1 | -3.0 | +5.0 | 8.1 |
| 2021 | 15 | DAL @ NYG | 21–6 | +10.8 | +11.5 | +15.0 | 4.2 |
| 2021 | 15 | GB @ BAL | 31–30 | +11.5 | +9.5 | +1.0 | 10.5 |
| 2021 | 15 | HOU @ JAX | 30–16 | -4.2 | -6.0 | +14.0 | 18.2 (model closer than the closing line) |
| 2021 | 15 | KC @ LAC | 34–28 | +2.6 | +3.0 | +6.0 | 3.4 |
| 2021 | 15 | LV @ CLE | 16–14 | -2.4 | +3.0 | +2.0 | 4.4 |
| 2021 | 15 | MIN @ CHI | 17–9 | +7.6 | +7.0 | +8.0 | 0.4 (model closer than the closing line) |
| 2021 | 15 | NE @ IND | 17–27 | +1.1 | +1.0 | -10.0 | 11.1 |
| 2021 | 15 | NO @ TB | 9–0 | -10.8 | -11.5 | +9.0 | 19.8 (model closer than the closing line) |
| 2021 | 15 | NYJ @ MIA | 24–31 | -13.3 | -10.5 | -7.0 | 6.3 |
| 2021 | 15 | SEA @ LA | 10–20 | -5.4 | -7.5 | -10.0 | 4.6 |
| 2021 | 15 | TEN @ PIT | 13–19 | +3.8 | -1.0 | -6.0 | 9.8 |
| 2021 | 15 | WAS @ PHI | 17–27 | -8.1 | -10.0 | -10.0 | 1.9 |
| 2021 | 14 | ATL @ CAR | 29–21 | -5.4 | -2.5 | +8.0 | 13.4 |
| 2021 | 14 | BAL @ CLE | 22–24 | -5.2 | -3.0 | -2.0 | 3.2 |
| 2021 | 14 | BUF @ TB | 27–33 | -3.1 | -3.5 | -6.0 | 2.9 |
| 2021 | 14 | CHI @ GB | 30–45 | -13.7 | -11.5 | -15.0 | 1.3 (model closer than the closing line) |
| 2021 | 14 | DAL @ WAS | 27–20 | +3.7 | +6.5 | +7.0 | 3.3 |
| 2021 | 14 | DET @ DEN | 10–38 | -11.6 | -13.0 | -28.0 | 16.4 |
| 2021 | 14 | JAX @ TEN | 0–20 | -10.2 | -8.5 | -20.0 | 9.8 (model closer than the closing line) |
| 2021 | 14 | LA @ ARI | 30–23 | -6.0 | -3.0 | +7.0 | 13.0 |
| 2021 | 14 | LV @ KC | 9–48 | -8.1 | -10.0 | -39.0 | 30.9 |
| 2021 | 14 | NO @ NYJ | 30–9 | +5.5 | +4.5 | +21.0 | 15.5 (model closer than the closing line) |
| 2021 | 14 | NYG @ LAC | 21–37 | -9.1 | -9.5 | -16.0 | 6.9 |
| 2021 | 14 | PIT @ MIN | 28–36 | -5.7 | -3.5 | -8.0 | 2.3 (model closer than the closing line) |
| 2021 | 14 | SEA @ HOU | 33–13 | +10.7 | +10.0 | +20.0 | 9.3 (model closer than the closing line) |
| 2021 | 14 | SF @ CIN | 26–23 | +4.4 | +2.5 | +3.0 | 1.4 |
| 2021 | 13 | ARI @ CHI | 33–22 | +6.5 | +7.5 | +11.0 | 4.5 |
| 2021 | 13 | BAL @ PIT | 19–20 | +4.6 | +4.0 | -1.0 | 5.6 |
| 2021 | 13 | DAL @ NO | 27–17 | +4.5 | +6.0 | +10.0 | 5.5 |
| 2021 | 13 | DEN @ KC | 9–22 | -4.7 | -8.5 | -13.0 | 8.3 |
| 2021 | 13 | IND @ HOU | 31–0 | +12.4 | +10.0 | +31.0 | 18.6 (model closer than the closing line) |
| 2021 | 13 | JAX @ LA | 7–37 | -11.7 | -14.0 | -30.0 | 18.3 |
| 2021 | 13 | LAC @ CIN | 41–22 | -2.7 | -2.5 | +19.0 | 21.7 |
| 2021 | 13 | MIN @ DET | 27–29 | +9.0 | +7.0 | -2.0 | 11.0 |
| 2021 | 13 | NE @ BUF | 14–10 | -1.5 | -3.0 | +4.0 | 5.5 (model closer than the closing line) |
Model and Close are home lines; Result is the final margin on the same axis. Model err is the absolute miss of the model's margin, shown green when the model was closer than the closing line.