Updated Sep 7, 2026 · Week 1 board.
The August problem, stated plainly
One book is a number with a company name attached to it. That is not a market. A cross-check needs two independent prices before it means anything, and a single book showing zero disagreement with itself confirms nothing at all, however many times that same number gets refreshed and re-served through the day. That was where our numbers stood in August. It isn't where they stand now, and I'd rather hand you a fact you can check than tell you things have improved.
Today's board: three books, sixteen games, no exceptions
Our market numbers, checked Sep 7 at 6:07 p.m. ET, list three books on all 16 Week 1 games. Pinnacle, DraftKings, Bovada, every game, with no single-book line left anywhere on it. That's a real change and you can go check it yourself, which is the first thing this page has to establish before it does anything else.
Where the three books actually disagree
If all three agreed on every number I'd be more worried, not less. Three books landing on identical prices reads like one line copied twice. They do not agree.
| Game | Pinnacle spread (home) | DraftKings spread (home) | Bovada spread (home) | Spread range |
|---|---|---|---|---|
| Buffalo at Houston | +1 | −1.5 | −1 | 2.5 |
| Cleveland at Jacksonville | −9.5 | −8.5 | −8.5 | 1.0 |
| Washington at Philadelphia | −4.5 | −5.5 | −4.5 | 1.0 |
| NY Jets at Tennessee | −1 | −1.5 | −1 | 0.5 |
Buffalo-Houston is the standout. Pinnacle has the Texans as home underdogs. DraftKings and Bovada both make them favorites. That's a full 2.5 points between the two ends, the widest of any Week 1 game we carry. The moneyline on the same game does the same thing in miniature. Pinnacle prices it −108 and −104, Houston over Buffalo. DraftKings and Bovada both post a flat −110 and −110. So one shop thinks Houston is a shade the better side, and the other two think the game is a true coin flip.
Totals disagree as well, by less. Three games carry a full 1-point range across the three books.
| Game | Pinnacle total | DraftKings total | Bovada total |
|---|---|---|---|
| New Orleans at Detroit | 50.5 | 49.5 | 50 |
| Miami at Las Vegas | 40.5 | 41.5 | 40.5 |
| Dallas at New York Giants | 47.5 | 48.5 | 48.5 |
And no single book is the outlier every time. Pinnacle is high on the first game, low on the second, low again on the third. One mispriced source would sit on the same end of all three. What is left is three books pricing the same football game a little differently, which is what a market looks like from the outside.
Turning book disagreement into a late-swap check
This is where a late-swap decision actually lives, because a published pick has a date stamped on it and the board keeps walking after that. Our spread pick on Buffalo-Houston went out Aug 21 at 1:31 p.m. ET against a line of −1.5, Houston favored. Pinnacle alone has since moved to +1, which makes Houston an underdog at that shop. Nothing broke. That's three-plus weeks of a market moving. Check a locked roster against the old published number and you are checking it against a line that no longer exists at one of the three books you can see. With three prices per game, that drift is visible before kickoff instead of invisible until after.
The Receipts Drawer
Three books on all 16 games, and 2.5 points of disagreement on Buffalo-Houston, which is how you know the disagreement is real rather than copied. Our own pick on that game is three weeks old and one full swing behind on who is favored. That's the watch item here, not a new bet. If you are holding a roster against that pick, pull the live board before kickoff, because at least one book has already flipped sides on you. And three pregame books is still short of a full market. We hold no in-game or live odds, every number above is a pregame snapshot rather than a closing line, and there is no closing-line comparison to run until after kickoff. None of the picks referenced here carries a graded result yet, and the full tally sits below. This is the NFL board only. Our other sports carry their own book counts, on their own day. The live 16-game board is at /odds. Our published picks, Buffalo-Houston included, are on our own NFL picks page. Sunday's player-prop lines sit over at our props page, where the problem runs the opposite way. One book there, not three. I went through that one in Fairplay Support: Player Props Projections & Expected Value.
