North Dakota State vs New Mexico sits on Saturday, October 24, 2026, and the board has New Mexico -4.5 at home, with the moneyline on that side at -200. For a College Football Week 8 game, that's a healthy number for New Mexico, and the model agrees with the direction. It just doesn't agree with the size.
Straight up, the model likes New Mexico to win. Against the spread, it's a different conversation, and that gap is the whole story here.
Why the model still likes New Mexico
The model's starting point for this game is New Mexico by 4.8, built off New Mexico's broader body of work. The adjustment that actually moves the number is on the North Dakota State side: Laubstein's loss of the quarterback role and the rushing option that came with it, with no replacement announced. That's a real subtraction, worth 2.0 points in the model's view, because it strips out a dimension North Dakota State had been using to extend possessions.
From there, the model treats Fa'avae as a known quantity after four starts rather than an unknown, and it folds two verified findings on Layne into one continuing season-long absence that prior results already account for, not two fresh deductions stacked on top. Net it out and New Mexico projects to win by 2.8.
The case for New Mexico, and its limits
New Mexico's path at home runs through defensive containment paired with enough offense to hold a lead. The clearest data point in its favor is holding Oklahoma to 14 points on the road, which is a stiffer test than the games against lighter G5 competition on its schedule. That's a meaningful distinction, but it's not proof of trench superiority by itself. Final scores alone don't establish who wins up front, and there's no verified film breakdown on either side to settle that question independently.
New Mexico also isn't walking in clean. Its other season-long losses are still restrictions on the offense, and neither a concussion question nor an eligibility question gets presumed resolved just because time has passed. Returning home after a trip to Hawai'i is worth noting, but it doesn't earn an extra bump on top of the home number already baked in, and nothing here supports tacking on a weather or motivation adjustment either.
Where North Dakota State can flip it
North Dakota State's counter is to lean on what it already showed this season: a shutout of Wyoming and a win over Nevada, both against competition it could dominate up front. The question is whether that translates against a New Mexico defense that's missing Jaxton Eck and Wheeler, which is the kind of absence that should, in theory, open rushing lanes.
But two things cap how far that logic can be pushed. Hayes is working through a foot issue, and Punni is expected to miss multiple weeks, and both constrain how much the model can assume about quarterback mobility and depth in the backfield. Hayes making his next start confirms he's playing, not that he's back to full speed, and that distinction matters when the whole case for North Dakota State depends on rushing volume.
The number that matters
Here's the gap: the market is asking New Mexico to win by 4.5, and the model's own projection has that margin at 2.8. The model and the market point the same direction, toward a New Mexico win, but they don't agree on the size of it, and that daylight between the two numbers is exactly why this one doesn't clear as a bet.
The verdict
This is a PASS. There's no official action on North Dakota State vs New Mexico today, and that's not a knock on the matchup, it's just not a bet we're making. Watch it as a game where a banged-up North Dakota State offense tries to lean on a defensive identity it's already proven, against a New Mexico team that's favored for real reasons but is carrying more injury and eligibility questions than the final score will show.
- Event · College Football Week 8 · Saturday, October 24, 2026
- Market · New Mexico -4.5 -200 — 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.

