If you follow my work, I believe preseason rankings are useless and should be eliminated due to the inherent bias that goes into them with voters. The same applies to ESPN’s FPI (Football Power Index) because the formula used is rigged.
The FPI stacks teams from the SEC/B1G in the preseason and keeps them propped up when they lose and rewards them when they win, regardless of the opponent.
Here are some prime examples. FSU dominated Alabama in week one and destroyed East Texas A&M in week two, and the FPI has FSU at No. 26, moving up 14 spots. FSU has been the most dominant team through two weeks and not even inside the top 25.
Alabama bounced back to beat a scrub team and automatically jumped up nine spots to No. 4.
How on earth can Alabama be that far ahead of FSU when they just played and got dominated? Notre Dame lost to Miami, but remains 11 spots ahead of the Hurricanes. Miami dropped six spots for dominating Bethune-Cookman 45-3, and Florida only dropped 11 spots for losing to unranked USF. However, Arkansas jumped 14 spots for beating Arkansas State.
Yes, Clemson struggled to beat Troy, but they won and dropped 11 spots. Georgia Tech is 2-0, rolling Gardner-Webb, and only moved up one spot. Auburn is 2-0, hasn't beaten a ranked opponent yet, and has already jumped to No. 11.
Ohio State beat Texas in week one and is still behind them in the FPI rankings.
Here’s what the FPI measures: “Football Power Index, that measures a team's true strength on net points scale; expected point margin vs average opponent on a neutral field.”
So in theory, it’s not measuring anything because it can’t take things like weather or injuries into consideration. Basically, ESPN can manipulate whatever formula they use because you can’t find it anywhere. There’s no rhyme or reason to it other than it tends to favor SEC/B1G teams. Just look at the favoritism shown each week. It's all propaganda for the TV networks.