2026 NFL Week 7 Digest
All 14 games of Week 7 in the 2026 NFL season are now graded. Before kickoff our model picked 10 of the 14 winners, the same number as the market favorite. THE PLAY, our one headline call for each game, won 8 and lost 6, a gain of $7.24 on $70 staked. During play the live engine's win probability scored a Brier of 0.141 over 11,540 moments, behind the incumbent model at 0.136 and behind the market at 0.126, where lower is better.
Pregame calls
Before kickoff the model picked a winner in 14 games, and those picks won 10 and lost 4. The market favorite, the side the market priced above 50%, won 10 and lost 4. Against the spread our side won 9 and lost 5, and on totals it won 8 and lost 6. Scored as probabilities, our pregame win chances had a Brier score of 0.194 against 0.228 for the market, where lower is better, so ours were the closer forecasts this week.
THE PLAY and the $5 ledger
THE PLAY is the one headline call we publish for each game, and the $5 ledger stakes every pregame pick at $5 at the price it showed. THE PLAY won 8 and lost 6 across 14 games, winning 57.1% of the time at an average price of 51.0 cents, a gain of $7.24 on $70 staked. The largest single return was a gain of $5.20 on DAL by 5.0+ in DAL at PHI, priced at 49 cents.
Every pick in the ledger together, 3,928 of them, won 42.0% of the time at an average price of 43.2 cents and finished with a loss of $1,580.07 on $19,640 staked. By market, player props finished with a loss of $1,759.82, spreads a gain of $312.34, totals a loss of $105.21 and moneylines a loss of $27.37.
The live engine
During play the live engine reprices every game about every ten seconds. Across 11,540 in-game moments in 14 games its win probability scored a Brier of 0.141. The incumbent, a simpler baseline model, scored 0.136 and the market scored 0.126 on the 11,401 moments that had a market price. Lower is better.
| Quarter | Moments | Engine | Incumbent | Market |
|---|---|---|---|---|
| Q1 | 2,136 | 0.218 | 0.222 | 0.210 |
| Q2 | 3,982 | 0.178 | 0.171 | 0.160 |
| Q3 | 2,252 | 0.123 | 0.122 | 0.103 |
| Q4 | 3,170 | 0.054 | 0.045 | 0.046 |
| Game | 11,540 | 0.141 | 0.136 | 0.126 |
Overtime is left out of this grade, so 38 overtime moments were set aside.
In-game edges and hot plays
An in-game edge is a stretch where the engine disagreed with the market price on the same side, counted once at the first price it showed. Across 14 games the engine logged 802 edges, and they won 50.4% of the time at an average price of 51.2 cents. Held to the final they returned a loss of 2.2 cents per $1 contract after fees, and sold 15 minutes after entry they returned a loss of 4.3 cents.
Hot plays are the edges that come right after a score, a turnover or a big play. Across 14 games the engine logged 427 hot plays, and they won 52.9% of the time at an average price of 53.3 cents. Held to the final they returned a loss of 1.7 cents per $1 contract after fees, and sold 15 minutes after entry they returned a loss of 4.6 cents.
Live player props
Live prop projections were graded in 14 games, 429 props in all. At kickoff the average projection missed the final stat by 17.0, measured in each stat's own units, and by the end of the third quarter the average miss was 7.9. The Brier score of the projected chance of going over the line was 0.271 at kickoff and 0.111 at the end of the third quarter. The market's own chance scored 0.241 on the 330 props with a market price.
- Receiving yards, 158 props: an average miss of 22.6 yards at kickoff and 9.8 yards at the end of the third quarter.
- Receptions, 155 props: an average miss of 1.5 catches at kickoff and 0.7 catches at the end of the third quarter.
- Rushing yards, 88 props: an average miss of 21.2 yards at kickoff and 8.1 yards at the end of the third quarter.
- Passing yards, 28 props: an average miss of 58.5 yards at kickoff and 36.6 yards at the end of the third quarter.
The best and worst calls
The best call is the pregame winner pick where we disagreed most with the market and were right, and the worst is the one where we disagreed most and were wrong. The best came in SF at ATL: we gave ATL a 55% chance to win when the market gave them 43%, and ATL won 29 to 20. The worst came in CLE at TEN: we gave TEN a 50% chance to win when the market gave them 43%, and CLE won 27 to 24.
Every game
| Game | Final | Our winner | Spread | Total | THE PLAY | $5 result |
|---|---|---|---|---|---|---|
| NE at CHI | CHI 30, NE 24 | CHI 56%, won | NE +3, lost | Under 50.5, lost | NO: CHI by 3.0+, lost | −$5.00 |
| PIT at NO | PIT 16, NO 10 | PIT 60%, won | PIT −3, won | Over 43.5, lost | Over 43.5, lost | −$5.00 |
| CIN at BAL | CIN 35, BAL 13 | CIN 63%, won | CIN −3, won | Over 46.5, won | CIN -3.0, won | +$4.76 |
| CLE at TEN | CLE 27, TEN 24 | TEN 50%, lost | TEN +2.5, lost | Over 36.5, won | Over 37.0, won | +$4.55 |
| IND at MIN | MIN 16, IND 10 | MIN 74%, won | MIN −6.5, lost | Over 41.5, lost | Over 41.5, lost | −$5.00 |
| MIA at NYJ | MIA 28, NYJ 25 | NYJ 55%, lost | MIA +3.5, won | Over 39.5, won | Over 39.5, won | +$4.55 |
| NYG at HOU | NYG 27, HOU 21 (OT) | HOU 66%, lost | NYG +10, won | Over 42.5, won | Over 42.5, won | +$4.55 |
| SF at ATL | ATL 29, SF 20 | ATL 55%, won | ATL +2.5, won | Over 50.5, lost | Over 50.5, lost | −$5.00 |
| TB at CAR | TB 43, CAR 13 | CAR 64%, lost | CAR −3, lost | Over 44.5, won | Over 44.5, won | +$4.55 |
| DEN at ARI | DEN 30, ARI 16 | DEN 71%, won | DEN −4.5, won | Over 42.5, won | Over 43.5, won | +$4.55 |
| GB at DET | DET 20, GB 17 | DET 63%, won | DET −3.5, lost | Over 55.5, lost | DET -3.5, lost | −$5.00 |
| LA at LV | LA 35, LV 21 | LA 70%, won | LA −3.5, won | Over 48.5, won | Over 48.5, won | +$4.55 |
| KC at SEA | SEA 20, KC 3 | SEA 62%, won | SEA −3, won | Over 42.5, lost | Over 42.5, lost | −$5.00 |
| DAL at PHI | DAL 38, PHI 21 | DAL 69%, won | DAL −4.5, won | Over 43.5, won | DAL by 5.0+, won | +$5.20 |
How to read this
A win rate means little without the price paid, so every rate here sits beside the average price, and a price of 53 cents means the market put the chance at about 53%. A Brier score measures how far a probability sat from what happened, from 0 for a perfect forecast to 1 for a certain one that was wrong, so lower is better. In-game returns are per $1 contract after fees.
Every game, every edge and every prop behind these numbers is on the NFL results page, graded as each game finishes.