Week 4 breakdown

Tight agreement 6 games

Over 27 seasons (2,607 games) favorites here won 69.8% against a 67.6% forecast; in 2023–25 it ran hotter, 74% against 66%. Lean a little toward the favorite. The blend adds 2 points.

• IND at WSH — Indianapolis Colts over the Commanders: Elo 64%, market 66%, System 68%. Tight agreement: go above both.
• NYJ at CHI — Chicago Bears over the Jets: Elo 62%, market 63%, System 65%. Tight agreement: go above both. Keenum is costing 100 Elo; Smith is worth 22 Elo.
• DAL at HOU — Houston Texans over the Cowboys: Elo 61%, market 58%, System 61%. Tight agreement: go above both. Prescott is worth 17 Elo.
• MIA at MIN — Minnesota Vikings over the Dolphins: Elo 81%, market 82%, System 84%. Tight agreement: go above both. Murray is worth 100 Elo; Willis is worth 90 Elo.
• LAC at SEA — Seattle Seahawks over the Chargers: Elo 79%, market 75%, System 78%. Tight agreement: go above both. Lock is costing 30 Elo; Herbert is worth 18 Elo.
• DEN at SF — San Francisco 49ers over the Broncos: Elo 61%, market 59%, System 60%. Tight agreement: go above both. wind 4 mph, 87°F (heat hurts favorites).

Where Elo and the market disagree 4 games

Over 27 seasons (1,102 games) Elo's side won 50% while Elo claimed 65% and the market 48%. The market was right almost every time. Take the market's number, or split the difference at most. This is where Elo is overconfident.

• PIT at CLE — Pittsburgh Steelers over the Browns: Elo 48%, market 58%, System 55%. Big disagreement: trust the market number. Watson is worth 34 Elo; wind 9 mph, 78°F (heat hurts favorites).
• LAR at PHI — Los Angeles Rams over the Eagles: Elo 46%, market 63%, System 59%. Big disagreement: trust the market number. Hurts is worth 17 Elo.
• KC at LV — Kansas City Chiefs over the Raiders: Elo 51%, market 67%, System 63%. Big disagreement: trust the market number. Cousins is worth 70 Elo; Mahomes is worth 46 Elo.
• DET at CAR — Detroit Lions over the Panthers: Elo 52%, market 66%, System 62%. Big disagreement: trust the market number.

Games that swing the most

Importance is how far the league's playoff odds move between a home win and an away win, from simulated seasons. Quality is the strength of the two teams. Both 0–100.

• DET at CAR: importance 54, quality 32
• ATL at NO: importance 49, quality 24
• LAR at PHI: importance 47, quality 43
• DAL at HOU: importance 47, quality 29
• JAX at CIN: importance 47, quality 51

Line movement since the picks froze

• GB at TB: TB 49% → 40% (money against TB)
• IND at WSH: WSH 33% → 39% (money on WSH)
• DET at CAR: CAR 44% → 39% (money against CAR)
• NYJ at CHI: CHI 64% → 58% (money against CHI)
• ARI at NYG: NYG 56% → 51% (money against NYG)
• KC at LV: LV 32% → 36% (money on LV)

Where the markets disagree with each other

The books, Kalshi and Polymarket are within 4 points of each other on every game.

Every game, with a suggested slider

The suggestion is the blend, which is also The System. The Game plan turns it into a week: which games to press, which to keep near 50, and how hard to lean given your place on the leaderboard. Adjust for what you know that the models don't.

Week 3 in review