Our reproduction of David Sasser's model (davidsasser.com/nfl). Each team's points come from its points per game, yards per play and yards per game, and the opponent's allowed versions. All prior games count (2025 incl. playoffs), weight halving every 4 games, no opponent adjustment. Refitted each run on his published projections (220 rows, fit RMSE 1.09 pts). Updated 15 min ago.
| Repro pick | Edge | Pts | Kickoff | Game | Repro score | Repro line | Repro total | Market | His line | His total | Δ line |
|---|---|---|---|---|---|---|---|---|---|---|---|
| MIA +3.5 | +5.6% | 2.7 | 2026-11-15 13:00 ET | MIA @ IND | 20.9–21.7 | IND −0.8 | 42.7 | IND −3.5 · 38.5 | IND −1.7 | 41.6 | +1.0 |
| DAL +4.5 | +4.0% | 2.2 | 2026-11-15 16:25 ET | SF @ DAL | 27.2–24.9 | SF −2.3 | 52.1 | SF −4.5 · 55.5 | SF −6.0 | 51.0 | -3.7 |
| ARI +9.5 | +3.2% | 1.9 | 2026-11-15 16:05 ET | LA @ ARI | 26.3–18.7 | LA −7.6 | 45.1 | LA −9.5 · 39.5 | LA −8.3 | 44.9 | -0.7 |
| NE +2.5 | +3.2% | 1.9 | 2026-11-15 09:30 ET | NE @ DET (N) | 23.4–24.0 | DET −0.6 | 47.4 | DET −2.5 · 51.5 | DET −1.1 | 48.0 | +0.4 |
| KC −3.5 | +2.2% | 1.5 | 2026-11-15 13:00 ET | KC @ ATL | 24.8–19.7 | KC −5.0 | 44.5 | KC −3.5 · 45.5 | KC −4.3 | 45.4 | +0.7 |
| GB −3.5 | +1.1% | 1.2 | 2026-11-15 13:00 ET | MIN @ GB | 17.1–21.8 | GB −4.7 | 39.0 | GB −3.5 · 42.5 | GB −5.8 | 40.2 | +1.1 |
| BAL −4.5 | +1.1% | 1.2 | 2026-11-16 20:15 ET | LAC @ BAL | 18.3–24.0 | BAL −5.7 | 42.3 | BAL −4.5 · 38.5 | BAL −6.1 | 39.4 | +0.4 |
| CAR +2.5 | +0.4% | 0.9 | 2026-11-15 13:00 ET | CAR @ NO | 21.5–23.0 | NO −1.6 | 44.5 | NO −2.5 · 48.5 | NO −1.0 | 45.5 | -0.6 |
| NYG −2.5 | −0.7% | 0.6 | 2026-11-12 20:15 ET | WAS @ NYG | 20.0–23.1 | NYG −3.1 | 43.1 | NYG −2.5 · 41.5 | NYG −1.9 | 43.6 | -1.2 |
| NYJ +8.5 | −0.7% | 0.6 | 2026-11-15 13:00 ET | BUF @ NYJ | 27.3–19.3 | BUF −7.9 | 46.6 | BUF −8.5 · 50.5 | BUF −10.5 | 46.7 | -2.6 |
| LV +6.5 | −0.8% | 0.5 | 2026-11-15 16:05 ET | SEA @ LV | 24.9–18.9 | SEA −6.0 | 43.8 | SEA −6.5 · 40.5 | SEA −6.7 | 40.7 | -0.8 |
| HOU −4.5 | −1.7% | 0.2 | 2026-11-15 13:00 ET | HOU @ CLE | 24.1–19.4 | HOU −4.7 | 43.5 | HOU −4.5 · 40.5 | HOU −5.3 | 42.5 | -0.6 |
| TEN +10.5 | −1.9% | 0.2 | 2026-11-15 13:00 ET | JAX @ TEN | 28.7–18.4 | JAX −10.3 | 47.1 | JAX −10.5 · 46.5 | JAX −10.5 | 44.1 | -0.2 |
| CIN −3.5 | −2.1% | 0.1 | 2026-11-15 20:20 ET | PIT @ CIN | 23.2–26.8 | CIN −3.6 | 50.0 | CIN −3.5 · 49.5 | CIN −3.4 | 48.0 | -0.2 |
Sorted by edge. Edge = chance the pick covers, treating the final margin as normal around the repro line (sd 13.5 pts), minus the 52.4% needed at −110. Ignores key numbers (3, 7), so edges near those are overstated or understated. Pts = repro line vs market line. Lines are written favorite first (DAL −3.5 = Dallas by 3.5). Hover a column header for its definition.
Held-out = each week predicted by a fit that never saw that week. ATS uses his posted market line (sheet line for Week 3), final scores, and the held-out repro line.
| Week | Source | Games | Held-out line RMSE | Max miss | His ATS | Repro ATS |
|---|---|---|---|---|---|---|
| 10 | page | 14 | 1.45 | 3.8 | 0–0 | 0–0 |
| 9 | page | 15 | 1.58 | 3.6 | 7–8 | 4–11 |
| 8 | page | 14 | 1.40 | 2.6 | 7–7 | 5–9 |
| 7 | page | 14 | 1.24 | 2.3 | 4–9–1 | 4–9–1 |
| 6 | page | 14 | 1.71 | 3.1 | 4–10 | 3–11 |
| 5 | page | 15 | 1.46 | 2.4 | 7–7–1 | 5–9–1 |
| 4 | page | 16 | 1.41 | 3.2 | 12–3–1 | 14–1–1 |
| 3 | sheet | 16 | 1.22 | 2.8 | 9–7 | 9–7 |
| PF/g | opp PA/g | YPP | opp YPP allowed | Yds/g | opp Yds allowed/g | home | intercept |
|---|---|---|---|---|---|---|---|
| 0.246 | 0.269 | 2.460 | 1.699 | 0.009 | -0.005 | 1.631 | -12.849 |
The yards inputs overlap heavily, so read the weights together, not one at a time.