The FPI has been used by ESPN for quite a few years now to predict the outcome of bowl games, and as far as its track record is concerned, it’s pretty accurate. FPI has accurately called the winner of each bowl game 65% of the time, which means if you’re betting against it, the odds are against you.
This year, FPI is extremely confident in some picks. It has Florida State beating Auburn in the BCS National Championship game at about 80%, the odds of Baylor whipping up on UCF in the Rose Bowl at 89%, and Marshall beating Maryland at a solid 60%.
That’s right, unfortunately. FPI and that accurate past doesn’t really like Maryland’s chances against the 9-4 Marshall Thundering Herd. With a point differential of only 4.3, there clearly isn’t a lot of faith in either team to get the job done, but still. Even though this is going to be played in Maryland’s backyard, going up against Rakeem Cato and the Thundering Herd is not going to be an easy task for the Terps.
Their offense is just as explosive, and their quarterback is widely looked at as a probable NFL player in the mold of Russell Wilson. Their defense may get steamrolled from time to time and their competition hasn’t been against anyone stronger than Maryland, but the team still knows how to put together a good game.
Do I think these odds are accurate? No. FPI doesn’t take into account the venue, and it doesn’t take into account injuries or suspensions. It looks strictly at on-field performance and while (by the numbers) these teams are very close statistically, I think Maryland has some extra incentive and a more rounded team. They’ve also gone up against better teams than Marshall. If the players were looking for some added motivation though, they got it.