Sunday, Sep 13, 8:20 PM ET
Where the best number is
Across the 2 books we archive — which is a real limit on this, and we would rather say so than imply a survey of the market. Lines only: two books on the same number with different juice read as equal here.
Both prices with the bookmaker's margin removed. This is the market's number, not ours — we don't forecast NFL games, and we publish why.
Teams the market priced at 55–60% won 493 of 879 games since 1999.Slightly worse than the price claimed, by 1.4 points.
The two prices add up to 104.2%. Everything over 100% is the book's edge, taken whichever side you pick.
DAL · 17 games, 600 attempts · 2025 figures
No games played yet this season — every figure here is from 2025.
35.3 attempts per game · team runs 64 plays at 57% pass · sacked on 4.9% of dropbacks
7.6 yards per attempt · 67.3% completions · 0.180 EPA per dropback
NYG has allowed 7.1 yards per attempt across 17 games
Following the FanDuel total · one of 2 books archived · 419 prices on record
Opened 47.5 on Wed, Jul 29, 47 days before kickoff.
418 reprices since, 809 points of movement in all. It sits 78.5 — 31 points higher than the open. It ranged from 22.5 to 78.5 along the way, retracing 778 points of that movement.
Half of the movement was over by Wed, Sep 2, 8 AM ET.
56 points at once — 78.5 to 22.5 on Wed, Sep 2, 8 AM ET.
Every sentence above is arithmetic on prices we recorded and the times we recorded them. Nothing here says where this number goes next.
Counts from 6,967 regular-season games since 1999. No model, no forecast — this is the history of the number itself, not a claim about this game.
Favourites of 2.5 covered 244 of 504 games.
The range covers a coin flip.
They won outright in 269 of 504.
No history held for a total of 78.5 — not rounded to a nearby number.
Home handicap. Dashed rules are the key numbers.
Over line.
CeeDee Lamb
Receiving Yards
over
-114
FanDuel
under
-114
FanDuel
CeeDee Lamb
Receptions
over
+116
FanDuel
under
-154
FanDuel
Dak Prescott
Passing Yards
over
-114
FanDuel
under
-114
FanDuel
George Pickens
Receiving Yards
over
-114
FanDuel
under
-114
FanDuel
George Pickens
Receptions
over
+124
FanDuel
under
-166
FanDuel
Javonte Williams
Rushing Yards
over
-114
FanDuel
under
-114
FanDuel
Jaxson Dart
Passing Yards
over
-114
FanDuel
under
-114
FanDuel
Jaxson Dart
Rushing Yards
over
-114
FanDuel
under
-114
FanDuel
Malik Nabers
Receiving Yards
over
-114
FanDuel
under
-114
FanDuel
Malik Nabers
Receptions
over
+116
FanDuel
under
-154
FanDuel
Every number here is a price we observed or arithmetic on prices we observed. “Market %” is the book's own two-sided price with the vig removed — not a forecast. We publish no prop projection: the model failed its pre-registered calibration gate on 2026-07-31, and a number that failed its own test doesn't go on the site.
Every point above is a price we observed and stored, with the time we saw it. Nothing on this page is a prediction, a rating, or a recommended bet.
Every row is a rate measured over games already played, never a probability — the same components behind each quarterback's Index below. No games played yet this season for at least one side, so those figures are last season’s. Our passing model failed its calibration test; we publish that on the accuracy page.
Expected volume from the inputs above. Unvalidated — it is arithmetic on what has already happened, not a forecast that has been scored against outcomes.
The market has his team at 40.5 points implied.
The index ranks this quarterback and this matchup against every qualifying passer, over games already played. It is not a probability and not a bet — our passing-yards model failed its calibration test, and we publish that on the accuracy page.
NYG · 12 games, 339 attempts · 2025 figures
No games played yet this season — every figure here is from 2025.
28.3 attempts per game · team runs 61 plays at 51% pass · sacked on 9.4% of dropbacks
6.7 yards per attempt · 63.7% completions · 0.009 EPA per dropback
DAL has allowed 8.1 yards per attempt across 17 games
Expected volume from the inputs above. Unvalidated — it is arithmetic on what has already happened, not a forecast that has been scored against outcomes.
The market has his team at 38.0 points implied.
The index ranks this quarterback and this matchup against every qualifying passer, over games already played. It is not a probability and not a bet — our passing-yards model failed its calibration test, and we publish that on the accuracy page.
Across 6,967 regular-season games since 1999, 24 of 28 spread rungs with a usable sample had a 95% interval containing 50%. In other words the number itself has never been an edge, at any rung, in twenty-seven seasons. What it is good for is knowing which numbers are expensive to cross — and a rung holding fewer than 30 games is shown greyed rather than dropped.