Predictions · Week 3
The model columns show each model's probability for the home team. Games are sorted into the spots our replays found an edge in, with the record behind each. Once a game ends, the models are scored the way players are, and the week and season tallies show who's forecasting best. Imp. is importance: how much the league's playoff picture swings on the game, from simulated seasons (hover for the game's quality).
Our lock and bold call of the week, and our pick on every game, are for signed-in players. Sign in, it's free.
Locks went 84% over 24 seasons; this season's record is on the leaderboard.
| Game | Elo | Spread | Books | Kalshi | Polymarket | Blend | System | Crowd | Sharp crowd | Imp. | Spot | Result |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ATL at GB | 65% | 69% | 72% | 73% | 74% | 71% | — | — | — | 49 | Clear favorite, mild disagreement | Fri 12:15 AM |
| NE at JAX | 58% | 57% | — | — | — | 60% | — | — | — | — | Tight agreement | Sun 5:00 PM |
| NYJ at DET | 67% | 70% | — | — | — | 71% | — | — | — | — | Tight agreement | Sun 5:00 PM |
| HOU at IND | 44% | 43% | — | — | — | 41% | — | — | — | — | Tight agreement | Sun 5:00 PM |
| KC at MIA | 52% | 26% | — | — | — | 33% | — | — | — | — | Big disagreement | Sun 5:00 PM |
| TEN at NYG | 69% | 66% | — | — | — | 69% | — | — | — | — | Tight agreement | Sun 5:00 PM |
| CIN at PIT | 57% | 43% | — | — | — | 46% | — | — | — | — | Big disagreement | Sun 5:00 PM |
| SEA at WSH | 39% | 43% | — | — | — | 40% | — | — | — | — | Tight agreement | Sun 5:00 PM |
| LAC at BUF | 73% | 63% | — | — | — | 66% | — | — | — | — | Big disagreement | Sun 5:00 PM |
| CAR at CLE | 60% | 43% | — | — | — | 47% | — | — | — | — | Big disagreement | Sun 5:00 PM |
| MIN at TB | 38% | 54% | — | — | — | 50% | — | — | — | — | Big disagreement | Sun 8:05 PM |
| ARI at SF | 74% | 74% | — | — | — | 76% | — | — | — | — | Tight agreement | Sun 8:05 PM |
| BAL at DAL | 35% | 41% | — | — | — | 40% | — | — | — | — | Everything else | Sun 8:25 PM |
| LV at NO | 52% | 59% | — | — | — | 57% | — | — | — | — | Everything else | Sun 8:25 PM |
| LAR at DEN | 57% | 43% | — | — | — | 46% | — | — | — | — | Big disagreement | Mon 12:20 AM |
| PHI at CHI | 57% | 54% | — | — | — | 55% | — | — | — | — | Everything else | Tue 12:15 AM |
The spots
Each game sorted into the spot the replays found an edge in. Rings are Elo, market and spread; the favorite's logo is in the middle. Tiles turn green or red as games finish.
Tight agreement 6 games
Lean a little toward the favorite. The blend adds 2 points. Over 24 seasons (2,312 games) favorites here won 69.2% against a 67.7% forecast; in 2023–25 it ran hotter, 74% against 66%.
JAX over NE
Elo 58% · Mkt 57% · Blend 60%
NE at JAX · Sun 5 PM
DET over NYJ
Elo 67% · Mkt 70% · Blend 71%
NYJ at DET · Sun 5 PM
HOU over IND
Elo 56% · Mkt 57% · Blend 59%
HOU at IND · Sun 5 PM
NYG over TEN
Elo 69% · Mkt 66% · Blend 69%
TEN at NYG · Sun 5 PM
SEA over WSH
Elo 61% · Mkt 57% · Blend 60%
SEA at WSH · Sun 5 PM
SF over ARI
Elo 74% · Mkt 74% · Blend 76%
ARI at SF · Sun 8 PM
Big disagreement 6 games
Take the market's number, or split the difference at most. This is where Elo is overconfident. Over 24 seasons (943 games) Elo's side won 50% while Elo claimed 63% and the market 49%. The market was right almost every time.
Clear favorite, mild disagreement 1 game
The blend without a bump. Nothing to exploit; take the average. Replays: favorites won about as often as the average of the two forecasts said (no surplus either way).
Everything else 3 games
The blend is as good as anything. No edge found either way in the replays.
Season scoreboard
The models scored as players over every decided game this season, on their pre-kickoff numbers.
| # | Model | Games | Points |
|---|---|---|---|
| 1 | Closing spread | 17 | +78.2 |
| 2 | The blend | 17 | +77.4 |
| 3 | The System | 17 | +77.4 |
| 4 | Elo (the model) | 17 | +45.3 |
| 5 | Sportsbooks | 0 | +0.0 |
| 6 | Kalshi | 0 | +0.0 |
| 7 | Polymarket | 0 | +0.0 |
| 8 | The crowd | 0 | +0.0 |
| 9 | The sharp crowd | 0 | +0.0 |