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 | 13 | NYG @ MIA | 9–20 | -8.6 | -6.5 | -11.0 | 2.4 (model closer than the closing line) |
| 2021 | 13 | PHI @ NYJ | 33–18 | +8.2 | +5.0 | +15.0 | 6.8 (model closer than the closing line) |
| 2021 | 13 | SF @ SEA | 23–30 | +4.6 | +3.0 | -7.0 | 11.6 |
| 2021 | 13 | TB @ ATL | 30–17 | +10.9 | +11.0 | +13.0 | 2.1 |
| 2021 | 13 | WAS @ LV | 17–15 | +0.9 | -1.0 | +2.0 | 1.1 (model closer than the closing line) |
| 2021 | 12 | ATL @ JAX | 21–14 | +1.1 | +1.5 | +7.0 | 5.9 |
| 2021 | 12 | BUF @ NO | 31–6 | +6.1 | +7.0 | +25.0 | 18.9 |
| 2021 | 12 | CAR @ MIA | 10–33 | +1.2 | +1.0 | -23.0 | 24.2 |
| 2021 | 12 | CHI @ DET | 16–14 | +7.2 | +2.0 | +2.0 | 5.2 |
| 2021 | 12 | CLE @ BAL | 10–16 | -2.6 | -3.0 | -6.0 | 3.4 |
| 2021 | 12 | LA @ GB | 28–36 | -3.3 | +2.0 | -8.0 | 4.7 (model closer than the closing line) |
| 2021 | 12 | LAC @ DEN | 13–28 | +1.5 | +2.5 | -15.0 | 16.5 (model closer than the closing line) |
| 2021 | 12 | LV @ DAL | 36–33 | -9.0 | -7.0 | +3.0 | 12.0 |
| 2021 | 12 | MIN @ SF | 26–34 | -4.4 | -3.5 | -8.0 | 3.6 (model closer than the closing line) |
| 2021 | 12 | NYJ @ HOU | 21–14 | -6.4 | -3.0 | +7.0 | 13.4 |
| 2021 | 12 | PHI @ NYG | 7–13 | +6.0 | +4.0 | -6.0 | 12.0 |
| 2021 | 12 | PIT @ CIN | 10–41 | -1.9 | -3.5 | -31.0 | 29.1 |
| 2021 | 12 | SEA @ WAS | 15–17 | -0.3 | +1.0 | -2.0 | 1.7 (model closer than the closing line) |
| 2021 | 12 | TB @ IND | 38–31 | +1.0 | +3.0 | +7.0 | 6.0 |
| 2021 | 12 | TEN @ NE | 13–36 | -7.1 | -7.5 | -23.0 | 15.9 |
| 2021 | 11 | ARI @ SEA | 23–13 | -3.2 | -2.5 | +10.0 | 13.2 |
| 2021 | 11 | BAL @ CHI | 16–13 | -3.2 | +5.0 | +3.0 | 6.2 |
| 2021 | 11 | CIN @ LV | 32–13 | -0.6 | +1.0 | +19.0 | 19.6 |
| 2021 | 11 | DAL @ KC | 9–19 | +2.3 | -2.5 | -10.0 | 12.3 |
| 2021 | 11 | DET @ CLE | 10–13 | -13.3 | -13.0 | -3.0 | 10.3 |
| 2021 | 11 | GB @ MIN | 31–34 | +2.2 | +1.0 | -3.0 | 5.2 |
| 2021 | 11 | HOU @ TEN | 22–13 | -12.5 | -10.5 | +9.0 | 21.5 |
| 2021 | 11 | IND @ BUF | 41–15 | -6.2 | -7.5 | +26.0 | 32.2 (model closer than the closing line) |
| 2021 | 11 | MIA @ NYJ | 24–17 | +5.8 | +3.5 | +7.0 | 1.2 (model closer than the closing line) |
| 2021 | 11 | NE @ ATL | 25–0 | +9.9 | +7.0 | +25.0 | 15.1 (model closer than the closing line) |
| 2021 | 11 | NO @ PHI | 29–40 | -1.8 | -2.5 | -11.0 | 9.2 |
| 2021 | 11 | NYG @ TB | 10–30 | -11.2 | -11.5 | -20.0 | 8.8 |
| 2021 | 11 | PIT @ LAC | 37–41 | -3.9 | -5.5 | -4.0 | 0.1 (model closer than the closing line) |
| 2021 | 11 | SF @ JAX | 30–10 | +7.7 | +6.5 | +20.0 | 12.3 (model closer than the closing line) |
| 2021 | 11 | WAS @ CAR | 27–21 | -5.4 | -3.5 | +6.0 | 11.4 |
| 2021 | 10 | ATL @ DAL | 3–43 | -6.7 | -8.0 | -40.0 | 33.3 |
| 2021 | 10 | BAL @ MIA | 10–22 | +5.9 | +8.5 | -12.0 | 17.9 (model closer than the closing line) |
| 2021 | 10 | BUF @ NYJ | 45–17 | +12.0 | +13.0 | +28.0 | 16.0 |
| 2021 | 10 | CAR @ ARI | 34–10 | -10.4 | -7.0 | +24.0 | 34.4 |
| 2021 | 10 | CLE @ NE | 7–45 | -0.5 | -2.5 | -38.0 | 37.5 |
| 2021 | 10 | DET @ PIT | 16–16 | -10.4 | -9.0 | 0.0 | 10.4 |
| 2021 | 10 | JAX @ IND | 17–23 | -12.7 | -10.5 | -6.0 | 6.7 |
| 2021 | 10 | KC @ LV | 41–14 | -1.8 | +2.5 | +27.0 | 28.8 |
| 2021 | 10 | LA @ SF | 10–31 | +3.3 | +3.5 | -21.0 | 24.3 (model closer than the closing line) |
| 2021 | 10 | MIN @ LAC | 27–20 | -2.5 | -3.0 | +7.0 | 9.5 (model closer than the closing line) |
| 2021 | 10 | NO @ TEN | 21–23 | -7.4 | -3.0 | -2.0 | 5.4 |
| 2021 | 10 | PHI @ DEN | 30–13 | -3.4 | -1.0 | +17.0 | 20.4 |
| 2021 | 10 | SEA @ GB | 0–17 | -1.8 | -3.0 | -17.0 | 15.2 |
| 2021 | 10 | TB @ WAS | 19–29 | +10.9 | +10.0 | -10.0 | 20.9 |
| 2021 | 9 | ARI @ SF | 31–17 | -4.8 | -5.5 | +14.0 | 18.8 (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.