Our UCLA vs Maryland prediction is simple to state and awkward to bet: the model has UCLA winning straight up on Saturday, September 26, 2026, with a projected margin of 3.8 points. The market agrees on the side, with UCLA at -125 on the moneyline.
The tier is PASS. It is an official-card PASS, which means no official action. We are not backing this game. Below is why the model likes the Bruins and why that still does not turn into a bet.
Why the model is on UCLA
The baseline starts with UCLA by 4.5. The model then checked whether the quarterback matchup should move it. Both Iamaleava and Washington are three-start quarterbacks, so there is a readable sample on each, but not enough to justify a precise adjustment. The overlay stays at 0.0 points.
The real case for UCLA is scoring against a depleted Maryland defense. Mathis is out for the season and Joyner remains barred from games. Those are two defensive absences that matter when the Bruins want to sustain drives.
UCLA also owns double-digit wins over California, San Diego State and Purdue. That is a wider slice of FBS opposition than Maryland's résumé offers, since Maryland's aggregate numbers are inflated by a Hampton game.
The case against UCLA
Maryland has a real path. It went on the road and beat FBS UConn 38-14, and the model's concern is that the same kind of efficient offensive game shows up against a UCLA defense that gave up 38 to Purdue. Add home field for Maryland and a cross-country trip for UCLA, and the Terrapins are not a soft landing spot.
The bigger issue is on the UCLA offense. Matthews is out for the season. Mizell and Phifer carry questionable designations, and Indestad's availability is uncertain. That thins both the receiving options and the protection. If Washington can reproduce Maryland's UConn-level efficiency, a restricted UCLA passing game may struggle to keep pace.
Maryland has its own holes. Teasley is unavailable and Williams is likely absent. Practice participation and historical recovery timelines do not establish clearance for either, so we treat them as out for planning purposes.
The number that matters
The model's own number is UCLA -3.8, and straight up it has the Bruins winning. At -125 on the moneyline, the market is also leaning UCLA. So this is not a case of the model fighting the market on direction.
What we are saying is narrower. The model likes the side, the injury sheets on both teams are unsettled, and the result is a game we are not betting. We will not dress that up as a soft play or a lean. It is simply not a bet we are making.
Off-field noise
UCLA's athletics department is in flux. Harris is the permanent athletics director after Jarmond's dismissal. That is context, but there is no evidence it has changed football preparation or play-calling, so the model gives it no weight in the number.
The verdict
Verdict: PASS. The model has UCLA winning at -125, but we are not backing this game. Its case rests on Maryland's depleted defense and UCLA's wins over California, San Diego State and Purdue. The risk is a Maryland offense that hit UConn for 38 facing a Bruins defense that allowed 38 to Purdue. Enjoy the game, and we will grade the result in public either way.
- Event · College Football Week 4 · Saturday, September 26, 2026
- Market · UCLA -125 — recorded when the pick was posted
- Model · Heismann, Botcappers' AI College Football model — how it is built and what it does not know
- Record · every pick is timestamped before the event and graded in public, win or lose
- Statistics · the model's own match/game database, current to the posting time above. Figures a reader cannot open here are internal to that database, not published third-party data.