They obviously started with the premise flawed or not that BYU will beat Utah on Sept. 21 in Provo, which would explain part of the discrepancy for these teams. Beyond that, Utah's prospects are more vague than BYU's, with a bunch of potential toss-up games.
You'd have to say Weber State is the only guaranteed win on the Utes' schedule. But the only sure losses I see are at USC and Oregon. So that creates a group of eight games that theoretically could go either way: Utah State, Oregon State, UCLA, Stanford, Arizona State and Colorado at home and Arizona and Washington State on the road.
Victories in half of those games would make the Utes 5-7, right on the oddsmakers' number. I'd take the over, just because a successful under bet would require the Utes going 4-8 or worse. That seems unlikely, because of Utah's atmosphere at Rice-Eccles Stadium and the fact that such a record would be so damaging to the program.
Again, giving BYU a win over Utah obviously led to that 8.5 over-under total.
The Cougars will be trying to break a three-game losing streak in the series, although two of those games were decided by missed field goals at the end. If they do overcome the Utes, the Cougars would have a shot at nine wins, which basically would require beating either Texas or Boise State at home. They could absorb one loss in that pair, along with likely defeats at Wisconsin and Notre Dame.
BYU should beat Middle Tennessee, Georgia Tech and Idaho State at home. So to reach 9-3, the Cougars may have to win four road games: at Virginia, Utah State, Houston and Nevada.
I'd take the under in relation to 8.5 wins, just because there's an opportunity to slip up at any of those venues, and 8-4 would still be a very good record against such a demanding schedule.