Updated Sep 7, 2026 · Week 1 board.
The same subtraction, plus one more variable: which book
Our published pick on the spread here reads Arizona (away), market line 10.5, our number 6.18, edge −4.32. Do it yourself: 6.18 − 10.5 = −4.32. That 10.5 belongs to Pinnacle. Our tracker, the same model re-checked against a live line, still carries 6.18, but it's looking at Bovada's number, 10 flat.
Same subtraction: 6.18 − 10 = −3.82. One model, one projected number, two edges. The line just means whichever book you happened to open.
Arizona at Los Angeles, three books, one game
| Book | Spread (home, Chargers) | Favorite / dog price | Total | Over / under price | Moneyline (home / away) |
|---|---|---|---|---|---|
| Pinnacle | −10.5 | +106 / −119 | 47 | −114 / −102 | −509 / +395 |
| Draftkings | −10 | −105 / −115 | 47.5 | −108 / −112 | −500 / +380 |
| Bovada | −10 | −105 / −115 | 47.5 | −105 / −115 | −585 / +410 |
The favorite side: a better number and a better price at the same book
The Chargers side: Pinnacle beats both rivals on both counts at once. Half a point cheaper to cover, −10.5 instead of −10. And it pays you plus money to do it, +106, where DraftKings and Bovada make you lay −105 for the worse number.
No trade-off on this side, which is unusual enough that I checked the prices twice.
The underdog side: no book wins on both axes
The Arizona side flips it. Pinnacle hands you the extra half point, +10.5 instead of +10, then charges −119 for the privilege. DraftKings and Bovada give you fewer points, +10, at a friendlier −115. Which one wins depends entirely on what you think that half point is worth, and that's a question about this number, not about which book is nicer in general.
Price shopping exists for exactly this trade-off, and Pinnacle isn't automatically the answer.
The moneyline gap, priced in real dollars
The moneyline is where the gap gets loud. Backing the Chargers outright costs −500 at DraftKings, which is $500 risked to win $100. Pinnacle wants −509 for the same win, $9 more out of your pocket for the same $100. Bovada wants −585, so that's $85 more risk than DraftKings for the same $100.
Now the other side. Arizona outright pays +395 at Pinnacle, +380 at DraftKings and +410 at Bovada. A $100 ticket at Bovada comes back with $30 more profit than the identical $100 ticket at DraftKings. Same team, same game, same afternoon, and real money left on the table by whoever doesn't look.
Why the published number and the live number can both be right
Both edges are right. They're measured against two different real lines, that's the whole story. The published pick, made Aug 21, 2026, is a snapshot: it wrote down Pinnacle's line at that moment and subtracted. The tracker follows whichever book it's watching, Bovada today, and it doesn't redo the subtraction.
It carries the same projected number forward. Neither one is wrong. They answer the same question, how far we sit from the market, against two market numbers that both exist. The mistake is reading one of them without knowing which book it's pointed at.
Shopping changes the edge number itself
Look at the total on this board. Pinnacle posts 47, half a point under the 47.5 at DraftKings and Bovada. Its over and under prices, −114 / −102, aren't shaped like DraftKings' −108 / −112 or Bovada's −105 / −115 either. Every pair of these books disagrees somewhere: the spread number, the price on it, the total, the price on that, or the moneyline.
So an edge computed against one book, with no book named, is a different number than the same model's edge against another. That's why our published −4.32 against Pinnacle's 10.5 and our tracked 6.18 against Bovada's 10, worth −3.82, don't match. Both are correct subtraction, done against two numbers that were both sitting on a screen today.
The no-vig conversion that turns any one of these prices into a probability is worked out on a flat New England-Seattle board in our companion piece. Book-by-book prices for every Week 1 game are posted on our live NFL odds desk, alongside the odds board. The bankroll question that follows an edge estimate is the Kelly criterion explainer (how much of your bankroll to bet).
The Receipts Drawer
One model, one game, two honest edges: −4.32 at Pinnacle and −3.82 at Bovada. Nothing broke. Three sportsbooks are pricing the same game three different ways and the arithmetic follows them. Pinnacle takes the favorite side outright, better number and better price together.
The Arizona side has no clean winner, just a trade-off you have to price for yourself. On the moneyline that spread is worth $85 on a $500 bet, which is not a thing I'd eyeball. Shop the book, not only the model. If any of the three moves its number before kickoff, run this comparison again against the live odds desk.
