Backtest History
2761 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 |
|---|---|---|---|---|---|---|---|
| 2017 | 17 | NYJ @ NE | 6–26 | -18.4 | -17.0 | -20.0 | 1.6 (model closer than the closing line) |
| 2017 | 17 | LV @ LAC | 10–30 | -11.0 | -7.0 | -20.0 | 9.0 (model closer than the closing line) |
| 2017 | 17 | SF @ LA | 34–13 | +5.2 | +6.5 | +21.0 | 15.8 |
| 2017 | 17 | WAS @ NYG | 10–18 | +3.5 | +5.0 | -8.0 | 11.5 (model closer than the closing line) |
| 2017 | 16 | ATL @ NO | 13–23 | -4.5 | -5.5 | -10.0 | 5.5 |
| 2017 | 16 | BUF @ NE | 16–37 | -14.0 | -11.0 | -21.0 | 7.0 (model closer than the closing line) |
| 2017 | 16 | CLE @ CHI | 3–20 | -7.5 | -5.5 | -17.0 | 9.5 (model closer than the closing line) |
| 2017 | 16 | DEN @ WAS | 11–27 | -1.0 | -3.0 | -16.0 | 15.0 |
| 2017 | 16 | DET @ CIN | 17–26 | +5.7 | +3.0 | -9.0 | 14.7 |
| 2017 | 16 | IND @ BAL | 16–23 | -14.0 | -13.5 | -7.0 | 7.0 |
| 2017 | 16 | JAX @ SF | 33–44 | +5.5 | +4.0 | -11.0 | 16.5 |
| 2017 | 16 | LA @ TEN | 27–23 | +7.0 | +5.5 | +4.0 | 3.0 |
| 2017 | 16 | LAC @ NYJ | 14–7 | +11.2 | +7.5 | +7.0 | 4.2 |
| 2017 | 16 | MIA @ KC | 13–29 | -11.5 | -11.0 | -16.0 | 4.5 (model closer than the closing line) |
| 2017 | 16 | MIN @ GB | 16–0 | +9.5 | +8.5 | +16.0 | 6.5 (model closer than the closing line) |
| 2017 | 16 | NYG @ ARI | 0–23 | -2.7 | -3.0 | -23.0 | 20.3 |
| 2017 | 16 | LV @ PHI | 10–19 | -8.8 | -10.0 | -9.0 | 0.2 (model closer than the closing line) |
| 2017 | 16 | PIT @ HOU | 34–6 | +7.3 | +8.0 | +28.0 | 20.7 |
| 2017 | 16 | SEA @ DAL | 21–12 | -3.0 | -4.5 | +9.0 | 12.0 (model closer than the closing line) |
| 2017 | 16 | TB @ CAR | 19–22 | -8.8 | -10.0 | -3.0 | 5.8 (model closer than the closing line) |
| 2017 | 15 | ARI @ WAS | 15–20 | -2.4 | -4.0 | -5.0 | 2.6 |
| 2017 | 15 | ATL @ TB | 24–21 | +5.6 | +7.0 | +3.0 | 2.6 (model closer than the closing line) |
| 2017 | 15 | BAL @ CLE | 27–10 | +7.7 | +6.5 | +17.0 | 9.3 (model closer than the closing line) |
| 2017 | 15 | CHI @ DET | 10–20 | -4.8 | -5.0 | -10.0 | 5.2 |
| 2017 | 15 | CIN @ MIN | 7–34 | -12.2 | -12.5 | -27.0 | 14.8 |
| 2017 | 15 | DAL @ LV | 20–17 | +2.5 | +3.0 | +3.0 | 0.5 |
| 2017 | 15 | DEN @ IND | 25–13 | +2.6 | +3.0 | +12.0 | 9.4 |
| 2017 | 15 | GB @ CAR | 24–31 | +0.0 | -3.0 | -7.0 | 7.0 |
| 2017 | 15 | HOU @ JAX | 7–45 | -10.4 | -10.5 | -38.0 | 27.6 |
| 2017 | 15 | LA @ SEA | 42–7 | -0.6 | +1.0 | +35.0 | 35.6 |
| 2017 | 15 | LAC @ KC | 13–30 | +4.0 | -1.0 | -17.0 | 21.0 |
| 2017 | 15 | MIA @ BUF | 16–24 | -1.1 | -3.5 | -8.0 | 6.9 |
| 2017 | 15 | NE @ PIT | 27–24 | +3.1 | +2.5 | +3.0 | 0.1 (model closer than the closing line) |
| 2017 | 15 | NYJ @ NO | 19–31 | -17.3 | -16.5 | -12.0 | 5.3 |
| 2017 | 15 | PHI @ NYG | 34–29 | +5.5 | +7.5 | +5.0 | 0.5 (model closer than the closing line) |
| 2017 | 15 | TEN @ SF | 23–25 | -5.0 | -2.5 | -2.0 | 3.0 |
| 2017 | 14 | BAL @ PIT | 38–39 | -5.3 | -6.0 | -1.0 | 4.3 (model closer than the closing line) |
| 2017 | 14 | CHI @ CIN | 33–7 | -7.9 | -6.0 | +26.0 | 33.9 |
| 2017 | 14 | DAL @ NYG | 30–10 | +0.5 | +4.0 | +20.0 | 19.5 |
| 2017 | 14 | DET @ TB | 24–21 | +0.8 | 0.0 | +3.0 | 2.2 (model closer than the closing line) |
| 2017 | 14 | GB @ CLE | 27–21 | +2.3 | +2.5 | +6.0 | 3.7 |
| 2017 | 14 | IND @ BUF | 7–13 | +2.5 | -3.5 | -6.0 | 8.5 |
| 2017 | 14 | MIN @ CAR | 24–31 | +4.0 | +2.5 | -7.0 | 11.0 |
| 2017 | 14 | NE @ MIA | 20–27 | +15.4 | +10.0 | -7.0 | 22.4 |
| 2017 | 14 | NO @ ATL | 17–20 | -0.9 | -2.5 | -3.0 | 2.1 |
| 2017 | 14 | NYJ @ DEN | 0–23 | +3.6 | +1.0 | -23.0 | 26.6 |
| 2017 | 14 | LV @ KC | 15–26 | -4.7 | -4.5 | -11.0 | 6.3 (model closer than the closing line) |
| 2017 | 14 | PHI @ LA | 43–35 | +1.0 | +1.0 | +8.0 | 7.0 |
| 2017 | 14 | SEA @ JAX | 24–30 | -3.9 | -3.0 | -6.0 | 2.1 (model closer than the closing line) |
| 2017 | 14 | SF @ HOU | 26–16 | -1.6 | -1.0 | +10.0 | 11.6 |
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.