It’s important to remember as the season progresses that the numbers I show are not the be-all end-all of prediction. Scheduling, key moments in games, players, dynamic play and improved performance over time also have a way of influencing the end result in a manner that defies the steady progress of my analysis week to week.
For instance, I know my numbers didn’t predict the course of the 2017 season with a late surge by Auburn that resulted in their sudden rocketing to the College Football Playoff discussion before a disappointing game in Atlanta scotched that all too soon.