Bill Connelly ran one million preseason SP+ simulations for the 2026 college football season, and one of them ended with BYU playing for a national title after a five-loss run. The same set of projections also gave the Cougars only 1.4% odds to win it all, which is why the most extreme result jumps out now.
The numbers are being searched because preseason college football projections are landing just as the 2026 season approaches, and Connelly's simulation work turns that anticipation into something measurable. In his setup, BYU was not projected as a favorite by any stretch. It was given the 18th-best title odds at 1.4%, while Texas Tech sat at 4.6%, and BYU's path to the CFP from the Big 12 came in at 28.4%.
Connelly's first simulation showed how the model can go sideways fast. BYU opened as an 11-seed against Oregon and lost by 15 points. That result was ordinary enough. The oddity came when the same preseason math was run out to its fullest, because one simulation produced a national champion with five losses, and the Cougars were the team that lived through it.
In that scenario, BYU finished 7-5 before reaching the Big 12 title game, then upset Texas Tech to win the conference and enter the CFP as the 12-seed. From there, the Cougars beat No. 5 LSU, No. 1 Wisconsin and No. 4 Syracuse before reaching the title game against No. 5 Northwestern. The route was so unlikely that it barely resembled the preseason ranking that put BYU in long-shot territory in the first place.
The same simulation run also produced another curiosity at No. 999,123, which Connelly labeled simply “huh?” That version featured a semifinal with LSU, BYU and Wyoming, a reminder that when a model is pushed through a million turns, even modest probabilities can surface in strange combinations. Wyoming had only a 3.5% chance to win the Mountain West, and Southern Miss had only a 1.7% chance to win the Sun Belt, yet the larger exercise still found room for them in the noise.
That is the point of the exercise. Preseason SP+ projections are not a forecast of certainty; they are a map of possibility, and the map gets stranger the farther it is run. BYU's 1.4% title chance makes the five-loss championship path feel impossible until the simulation reaches enough repetitions to find it. Connelly did not say which team won the national title in the highlighted run, but the larger message is already clear: in college football, a tiny preseason number can still hide a very improbable door.
