What July's Early Lines Say About the 2026 Season
Analysis · by Rick's Picks Analytics
It's the first week of July, kickoff is eight weeks away, and the betting market has already put a price on 54 college football games. That's it -- 54, out of the several hundred FBS games on the 2026 schedule. We pulled every one of them from our database this week, and the picture they paint is worth a close look: which openers are parked on key numbers, where summer money has already nudged a line, and why these early markets behave nothing like the ones you'll see in game week.
As always, everything below comes straight from our own data. No invented numbers, no "sources say." When our feed contains something suspicious, we'll tell you that too.
The headliners
The books don't post July lines on Kent State at Buffalo. They post them on the games people are already arguing about at the lake. Here are the openers most worth knowing:
| Week | Matchup | Line | Total |
|---|---|---|---|
| 1 | LSU vs Clemson | LSU -10.5 | 50.5 |
| 1 | TCU vs North Carolina (Dublin) | TCU -6.5 | 49.5 |
| 1 | California vs UCLA | Cal -1.5 | 53.5 |
| 1 | Florida State vs SMU | SMU -2.5 | 52.5 |
| 1 | Ole Miss vs Louisville | Ole Miss -6.5 | 55.5 |
| 1 | Auburn vs Baylor | Auburn -6.5 | 59.5 |
| 1 | Oregon vs Boise State | Oregon -24.5 | 53.5 |
| 2 | Michigan vs Oklahoma | Michigan -2.5 | 45.5 |
| 2 | Texas vs Ohio State | Texas -1.5 | 47.5 |
A few things jump out. LSU laying 10.5 to Clemson in a Week 1 marquee game is a bigger number than that series usually carries. Cal -1.5 against UCLA is the closest thing to a pick'em on the board -- two rebuilt rosters and a market that genuinely does not know which one is better yet. And the two Week 2 blockbusters, Michigan-Oklahoma and Texas-Ohio State, both opened inside a field goal, which is exactly where the most interesting line movement happens.
At the other end of the board sit the confidence statements: Ohio State -49.5 over Ball State, Indiana -40.5 over North Texas, Oklahoma -40.5 over UTEP, Alabama -25.5 over East Carolina. Twenty-three of the 54 posted spreads are 20 points or more. July boards skew toward two kinds of games: the ones everyone cares about, and the ones nobody disputes.
Where the key numbers sit
If you've read our key-numbers work, you know the two margins that matter most in college football: 3 and 7. In our 2015-2022 training sample of 5,649 FBS-vs-FBS games, 9.6% of games landed exactly on 3 and 8.5% landed exactly on 7 -- about 18% of all games on just two numbers. That distribution held up out of sample (90% overlap in the top-10 margins on unseen test data, including the completed 2025 season), which is why we treat it as a structural feature of the sport rather than a market quirk.
So where do July's 54 openers sit relative to those numbers?
| Bucket | Games | Examples |
|---|---|---|
| Exactly on 3 | 3 | Stanford -3 vs Hawai'i, UNLV -3 vs Memphis, Virginia -3 vs NC State |
| Within 0.5 of 3 | 3 | Michigan -2.5 vs Oklahoma, SMU -2.5 at FSU, E. Michigan -3.5 vs SJSU |
| Exactly on 7 | 1 | Georgia Tech -7 vs Colorado |
| Within 0.5 of 7 | 4 | TCU -6.5, Ole Miss -6.5, Auburn -6.5 (and one we don't trust -- see below) |
Eleven of 54 games -- roughly one in five -- are on or within a half point of the two biggest key numbers. Those are the lines to watch between now and September, because a half-point move through 3 or through 7 changes the math on a pick more than a full point of movement anywhere else. Georgia Tech -7 against Colorado is the single stickiest number on the board: the book has planted the flag exactly on the second-most-common margin in the sport, and it will take real money to move it off.
One more structural detail: 46 of the 54 spreads end in a half point. Only 8 are integers. That's deliberate. A half-point line can't push, and in a thin summer market with wide disagreement, books prefer prices that resolve cleanly rather than numbers that invite two-way action around a push.
Summer money is already moving lines
Our feed carries an opening spread for 47 of the 54 games, and six of them have already moved:
| Matchup | Open | Now | Direction |
|---|---|---|---|
| Michigan vs Oklahoma | -1.5 | -2.5 | Toward Michigan |
| Texas vs Ohio State | -2.5 | -1.5 | Toward Ohio State |
| Florida State vs NMSU | -28.5 | -29.5 | Toward FSU |
| South Carolina vs Kent State | -33.5 | -34.5 | Toward South Carolina |
| Troy vs Sam Houston | -17.5 | -16.5 | Toward Sam Houston |
| Nevada vs W. Kentucky | +4.5 | +4.0 | Toward Nevada |
The two that matter are at the top. Michigan-Oklahoma moving from -1.5 to -2.5 and Texas-Ohio State moving from -2.5 to -1.5 are both one-point moves inside the 1-to-3 range, where every half point carries real win-probability weight. Neither has crossed 3 yet. If either one does before kickoff, whoever took the early number on the right side got meaningfully better value than the game-week price -- and that, not the July board itself, is the actual argument for paying attention this early. Early markets are where price discovery happens slowly and visibly.
An honest note about July data
Our brand is publishing what the data says, including when the data is wrong. Two lines in this pull fail the smell test:
- Georgia -6.5 vs Tennessee State. Tennessee State is an FCS program. No functioning market makes Georgia a one-score favorite over an FCS opponent; this is almost certainly a feed artifact, and we've excluded it from any analysis above beyond counting it.
- Notre Dame +20.5 vs Wisconsin. Notre Dame as a three-touchdown home underdog to Wisconsin is not a real price. This looks like a sign or home/away designation error in the source data -- the kind of thing neutral-site and rescheduled games produce constantly.
If we quietly dropped those rows and said nothing, you'd never know. But this is exactly why we route every spread in our pipeline through one canonical utility with a fixed sign convention (negative = home favored), and why "the database says so" is never the end of the conversation. July feeds are messier than November feeds. Anyone doing analysis on early lines who doesn't say that out loud is hiding something.
How July markets differ from game week
Three practical differences, none of them subtle:
- Coverage is sparse. 54 games have spreads; only 12 of those even have a total posted. Most of the board simply doesn't exist yet.
- Limits are low. Books post these numbers to generate engagement and gather information, not to take size. A modest amount of sharp interest can move a July line a full point -- as Michigan-Oklahoma just showed -- in a way that would take far more money in game week.
- Information is thin. No August camp reports, no depth charts, no transfer-portal stragglers resolved. The line is a projection of a projection, which is precisely why openers that survive the summer untouched (like that Georgia Tech -7) tell you the market's prior is strong.
The bottom line
Fifty-four games, eleven parked on or near the key numbers 3 and 7, six already moved by summer money, and two lines we flat-out don't believe. That's the July 2026 board.
We're not telling you any of these openers is a value play -- our standard for that is the same as always: a hypothesis tested on 2015-2022 data, validated out of sample on 2023-2025, clearing the 52.4% break-even rate at -110 pricing, with Bonferroni correction when we test multiple angles at once. No July number has earned that treatment yet, and pretending otherwise would be tout-speak.
What the early board is good for is context. It tells you where the market's priors are strong (Georgia Tech -7, Ohio State -49.5), where they're genuinely uncertain (Cal -1.5, Texas -1.5), and which lines are one sharp opinion away from crossing a key number. When our 2026 Analyst Picks start rolling out, this is the baseline they'll be measured against -- and as always, we'll show you the results either way.
Rick's Picks publishes statistical analysis of college football for informational and entertainment purposes. Nothing here is betting advice. 21+. If gambling is affecting you or someone you know, call 1-800-GAMBLER.