Week 5 Model Picks

Week 4, if you had bet the posted card: 4-0-1 on spreads, 2-1 on parlays, +9.01 units. Jacksonville +2.5 won outright. Chicago, Miami, and San Francisco all covered, and Seattle -7 pushed on a seven-point win. Two of the three moneyline parlays hit, at +388 and +233. The one that missed had all three legs lose. Season to date: -2.40 units, up from -11.41 a week ago. Here is Week 5, on the same rules. A spread has to clear a raw edge of 4.0. Totals stay on the page and off the card. Three 1-unit moneyline parlays. The budget is eight units. This week spends eight.

Our Bets

Spreads (5 units)

• 2 units on Giants +4.5 (-115). The Model has New York winning by 1.7 against Washington -4.5. That is a 6.2-point edge, the biggest on the board, and an outright side split. The odd part: the Giants row is Jameis Winston at 34.5 and Jayden Daniels is at 59.2. The Model likes New York anyway, so the quarterback is not what is driving this. If New York starts someone else, the number moves.
• 1 unit on Patriots -3.5 (-110). The Model has New England by 9.2 against 3.5. Drake Maye is at 61.1, Kirk Cousins at 64.9. The edge comes from the team numbers, not the quarterbacks.
• 1 unit on Bills +3.0 (-110). Monday night. The Model has Buffalo by 1.1 against Los Angeles -3. That is a 4.1-point edge and the other side split on the card. Josh Allen 67.7, Matthew Stafford 57.9.
• 1 unit on Jaguars -7.5 (-102). The Model has Jacksonville by 11.6 against Philadelphia. The edge is 4.08, so it just clears. This is the London game, 9:30 AM Eastern. Trevor Lawrence is at 70.8, Jalen Hurts at 58.5. The Jaguars covered for us last week. The Model does not care about that, and you shouldn’t either.

Parlays (1 unit each)

• Giants ML (+160) / Patriots ML (-198) / Lions ML (-258). Pays +443.
• Bills ML (+140) / Seahawks ML (-162) / Bengals ML (-325). Pays +408.
• Bears ML (-135) / Steelers ML (-142) / Jaguars ML (-345). Pays +283.

No Thursday spread. Tampa Bay at Dallas is DAL -9.5 and The Model calls it Dallas by 5.7. That is a 3.8-point edge, and it stays off. Same rule as Tampa Bay at 3.99 last week. Dallas at -500 and Houston at -425 are too short to be parlay legs.

The Model

Same ensemble: Ridge, LightGBM, XGBoost, and a small neural net, refit on games through Week 4, using team stats, EPA, and starter QBR. Week 4 was its best week: 10-6 straight up and 11-4-1 against the spread. Through four weeks it is 36-28 straight up and 32-31-1 against the spread. It is still behind the market on margin error, 10.1 to 9.8. One good week does not change that.

The table is the full 15-game card: per-model margin, pick, DraftKings spread, raw edge, straight-up win probability, and moneyline. Four teams are on bye.

Over/Under Picks

Totals went 6-10 in Week 4 and are 24-40 on the season. We are showing the numbers and not betting them. The biggest total edge is Detroit at Arizona Under 54.5. The Model says 46.9. Still not a bet.

Where we disagree with Vegas

• NYG at WAS: WAS -4.5; model MOV +1.7 Giants. On the card for 2 units. Side split. The Winston caveat is above.
• LV at NE: NE -3.5; model MOV +9.2. On the card.
• BUF at LA: LA -3.0; model MOV +1.1 Bills. On the card. Side split.
• PHI at JAX: JAX -7.5; model MOV +11.6. On the card at 4.08.
• TB at DAL: DAL -9.5; model MOV +5.7. Off the card at 3.8, and it plays tonight. Tampa Bay’s row is Jalon Daniels at 49.7.
• CIN at MIA: CIN -6.5; model MOV +2.8 Cincinnati. Off the card at 3.7. Miami is Malik Willis at 39.2 again.
• DEN at LAC: DEN -3.5; model MOV +0.2 Chargers. Off the card at 3.7. Side split. Justin Herbert’s row is still at 29.5.
• HOU at TEN: HOU -7.5; model MOV +3.9 Houston. Off the card at 3.6.
• MIN at NO: MIN -2.5; model MOV +0.3 Saints. Off the card at 2.8. Side split. Kyler Murray is at 33.9 for Minnesota.

Starter notes. Chicago’s row is Tyson Bagent at 75.1, a change from Case Keenum last week. Herbert at 29.5 is the same flag as last week. If that is an injury split and not the real Herbert, the Chargers number is stale. The Giants are Winston at 34.5. Check who starts before betting the 2 units.

Remember, the best way to use this is as one input into your own thinking. The Model does not know who is holding out, and it only knows the quarterback nflverse listed.

These are our personal bets and opinions, not betting advice. Wager at your own risk and within your means.

Good luck everyone.