Our James Madison vs Old Dominion prediction is simple to state: the model has James Madison winning this Saturday, September 26, 2026, and its own number is James Madison -7.7. The listed side is James Madison -6.5, with the moneyline on the pick at -225. The conviction tier is PASS. This is an official-card PASS, so we are not backing it and there is no official action.
That is worth saying up front, because a model that likes a side and a card that says pass can look like a contradiction. Here is what the model sees, and what keeps us from treating it as a bet.
Why the model likes James Madison
The model projects James Madison to win by 8.2 before any adjustment, and it lands at -7.7 as its final margin. The main support is defense. James Madison allowed 13 points against FBS Liberty and 13 against San Diego State. The San Diego State game was a road win, and the model treats it as more informative than the blowout of FCS Wagner. A shutdown of a lower-division opponent tells you little. Holding two FBS teams to 13 each tells you more.
The model is also careful about what those scores prove. It says they do not independently establish an edge in the trenches or in scheme. They show a defense that has been hard to score on, and that is the repeatable part of the case.
The quarterback question
Coleman has two starts, and the verified, ongoing absence of Evans means James Madison has to win with Coleman. It cannot count on Evans returning or providing relief. The model does not add a fresh numerical penalty for that, since the situation is already the established baseline. It still leaves this side leaning on a quarterback with a thin track record.
The other injury note is Henderson, who remains unavailable. Replacement linebackers are handling tackling and run fits. The model does not add a new penalty for that either, since the absence is already established and priced into its read.
The case against
Old Dominion has a plausible path. It gave up 20 points to East Carolina in a competitive defensive performance, and the model's own risk note says a repeat of that, combined with enough offense from Henicle at home, could turn a close game into an Old Dominion win. The path runs through Coleman failing to sustain scoring drives.
The Old Dominion loss at Power-conference Virginia Tech also deserves an allowance. It came against a tougher schedule slot than James Madison has faced, so the raw result should not be read as a weakness.
The number that matters
The model's -7.7 sits above the -6.5 on the pick, and the moneyline of -225 shows how strongly the market leans the same way. Even so, this is not a bet we are making. We are not calling it a play, and we are not going to dress it up as one. The model agrees with the market on the winner. Whether that agreement becomes a wager is a separate question, and on this game the answer is no.
If you follow the game, watch two things: whether Coleman can sustain drives, and whether Old Dominion can reproduce its East Carolina defensive showing.
The verdict
Verdict: PASS. The model has James Madison winning, with a projected margin of 7.7 points against a listed -6.5. We are not backing it, and there is no official action on this game.
- Event · College Football Week 4 · Saturday, September 26, 2026
- Market · James Madison -6.5 -225 — 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.

