might understand better. Let's use the bowl game last year as an example: Washington was favored by 4-5 over BYU and won by 15. So BYU was not favored in the game, and yet they still lost ATS. Had Washington only won by a FG then underdog BYU would have won ATS, or outperformed expectations. Looking at a different example, take the last BYU/Utah game in 2013 (since not beating Utah is a common criticism of Bronco by many BYU fans). Here BYU was favored by ~7. Not only did they lose the game outright, but even if they had won the game by 6 or fewer points, Utah would have still won ATS versus a favored BYU team meaning a BYU win by less than 7 would have been an underperformance despite it going in the W column of their win/loss %. That is why just looking at winning percentage over time is a terrible way to substantiate an argument for or against a coach or team.

ATS doesn't care who is "favored"; but rather how a team performs against odds-maker expectations for them. Which is why it is a more objective measure for how a team actually performs in any game relative to how they are expected to perform than simply win/loss percentage. Of course one games doesn't dictate a trend that dictates a team's performance; but over the course of a season or several season, trends of performance ATS become very accurate of how a team is over or under performing. BYU under Bronco has under-performed ATS. No way to argue that fact.