Week 1 has sixteen games, and four of them touch the key number three on at least one of the two fully tracked books. Only one sits there at both books. Another game is parked on seven at both. The strangest market is Bills at Texans: Pinnacle and Bovada disagree about which team is favored.
The board below is the shared Pinnacle and Bovada snapshot captured at 4:00 PM ET on August 30 from the odds feed. It is a market report, not a projection. The point is to show the price that existed, the places books split, and the rows the feed did not have.
| Game | Pinnacle | Bovada | Read |
|---|---|---|---|
| Patriots at Seahawks | SEA -3.5 | SEA -3.5 | agree |
| 49ers at Rams | LA -3 | LA -3.5 | Pinnacle holds three |
| Browns at Jaguars | JAX -7.5 | JAX -8.5 | widest same-favorite gap |
| Bears at Panthers | CHI -2.5 | CHI -2.5 | agree; road favorite |
| Jets at Titans | TEN -2.5 | TEN -2.5 | agree |
| Buccaneers at Bengals | CIN -3.5 | CIN -4 | half-point split |
| Saints at Lions | DET -7 | DET -7 | seven at both books |
| Falcons at Steelers | PIT -3 | PIT -3 | three at both books |
| Bills at Texans | BUF -1 | HOU -1 | opposite favorites |
| Ravens at Colts | BAL -3.5 | BAL -3.5 | agree; road favorite |
| Packers at Vikings | MIN -1 | MIN -1.5 | half-point split |
| Cardinals at Chargers | LAC -10.5 | LAC -10 | half-point split |
| Dolphins at Raiders | LV -3.5 | LV -3.5 | agree |
| Commanders at Eagles | PHI -4 | PHI -4.5 | half-point split |
| Cowboys at Giants | DAL -2.5 | DAL -3 | Bovada holds three |
| Broncos at Chiefs | KC -3 | KC -2.5 | Pinnacle holds three |
The key numbers are doing real work
Falcons at Steelers is Pittsburgh -3 at both books, the cleanest key-number market on the slate. Broncos at Chiefs is Kansas City -3 at Pinnacle and -2.5 at Bovada. The 49ers-Rams split points the other way: Los Angeles -3 at Pinnacle and -3.5 at Bovada. Cowboys at Giants puts Dallas -3 at Bovada and -2.5 at Pinnacle. Three of the four games touching three are book-dependent. Saints at Lions is the lone unanimous seven, with Detroit -7 on both feeds.
That is why “the line” is often the wrong phrase. There are lines, plural, and a half point around three or seven can separate a push from a decision. The useful read is not that one book must be right. It is that a bettor who ignores the split is volunteering to take the worse version of the same opinion.
Bills at Texans is the board's argument
Seven games match exactly between Pinnacle and Bovada. Seven are separated by a half point. Browns at Jaguars is a full point apart, Jacksonville -7.5 versus -8.5. Bills at Texans is stranger: Pinnacle makes Buffalo a one-point road favorite, while Bovada makes Houston a one-point home favorite.
The moneyline rows confirm that the side reversal is present in the captured feeds rather than a spread-sign typo. Bovada shows Houston -105 and Buffalo -115; Pinnacle shows Houston -107 and Buffalo -105. DraftKings appears on two of the sixteen games in the same snapshot, so it cannot settle the disagreement. The FanDuel feed has no current Week 1 rows after early June. That absence belongs in the report because coverage is part of provenance.
Road favorites are lighter than the recent baseline
Pinnacle makes the road team the favorite in four of sixteen games: Ravens at Colts, Cowboys at Giants, Bears at Panthers, and Bills at Texans. That is 25 percent of the current board. Bovada flips Bills-Texans, leaving three road favorites.
In NFL game results, true road favorites appeared in twenty-five of sixty-four Week 1 games from 2022 through 2025, or 39.1 percent. Across all regular-season games in that completed window, the count was four hundred thirty-four of one thousand eighty-seven, or 39.9 percent. The current board is lighter than that comparison. It is not proof of a betting angle; it is a clean description of the market shape.
The standing gaps have not moved during shared coverage
Pinnacle and Bovada first overlap on all sixteen games at 7:15 PM ET on August 27. Through the August 30 snapshot, roughly sixty-nine hours later, neither book changes a spread on any game. The differences in the table are standing disagreements, not a race where one book has already reacted and the other is lagging.
Bovada has an earlier window beginning August 22. Inside that feed it changes four games before the shared coverage begins, including Browns-Jaguars from -7.5 to -8.5 and Ravens-Colts from Baltimore -3 to -3.5 on August 24. Because Pinnacle has no matching rows in that earlier window, the data cannot name who moved first. It can only show where the number started, where Bovada carried it, and where simultaneous comparison becomes possible.
The May snapshots are clues, not a movie
Six matchups also have FanDuel rows from May 30: Falcons-Steelers, Ravens-Colts, Bills-Texans, Bears-Panthers, Patriots-Seahawks, and Commanders-Eagles. Those early lines were within a half point of the available Pinnacle rows. Each matchup has a different favorite in the August snapshot.
The tempting move is to write a dramatic story about how the market crossed the summer. The database does not support that story. Pinnacle has no rows between June 3 and August 27 for those games, while Bovada begins on August 22. The honest sentence is narrower: the early and current snapshots disagree, and the path between them is missing.
What to watch next
Bills at Texans is first because two books cannot agree on the direction of the spread. Ravens at Colts is next because both books sit a half point from three. The next report will have a longer simultaneous window, which means it can distinguish a standing split from an actual move without guessing between missing timestamps.
For the mechanics behind the numbers, read How NFL Spreads Work, Closing Line Value Explained, and the vig and hold guide. The watchpoint is simple: record the next changed row, the book, and the capture time. Until that row exists, the board is still the board above.
Sources
the odds feed supplies spread, moneyline, book, and capture time for the live pregame sections. NFL game results supplies the completed 2022–2025 road-favorite comparison. Both were queried read-only on August 30, 2026. No result in this article is treated as a graded betting record.
Line movement versus recorded public ticket share
This chart remains empty until a verified source table binds public ticket share and line movement for the same market observations.
Model calibration from graded predictions
Calibration points render only when a verified source binds prediction probabilities to settled outcomes for the same observations.
