What are Rutgers’ bowl eligibility chances? A dive into SP+ and FPI rankings

After a convincing win versus Temple and a strong effort to win at Syracuse, Rutgers football stands at 2-0 in advance of a (probably sneakily competitive) game against FCS stalwart Delaware next Saturday. In advance of the next game, a post about Rutgers’ bowl eligibility chances is in order.

A reminder, with a full season in 2021, we are back to the pre-2020 rules with respect to bowl eligibility. A team needs six wins this season (and only one of them can be versus an FCS opponent). Last year was a strange year in college football for lots of reasons not worth enumerating—but one important difference was teams did not have to be close to/at .500 records to be bowl eligible. This is why you may have watched a 3-7 Mississippi State team play in the Armed Forces Bowl last season (for instance).

So, if Rutgers hopes to play in a bowl game after this season, they’ll need to cobble together six wins. For the sake of argument, I’m assuming a six-win Rutgers team gets a bowl bid (i.e., there will be a place to put Rutgers if bowl-eligible). This may be arguable, especially with the No. 7 Big Ten team supposed to play in the now-canceled Redbox Bowl, but for purposes of this article, let’s assume there will be enough bowl slots for all the Big Ten teams that hit the magic number of six wins.

OK, so on to the advanced analytics. (You thought you were getting off easy, but with me writing this article, think again!) There are several analytical models in college football, for my money my favorite is Bill Connelly’s SP+ metric (paywall link here) but in my humble opinion it has a flaw that is relevant to 2021 Rutgers football. Specifically, it seems to weight previous season performance more heavily than current season performance until 4-5 weeks into the current season, when conference play starts.

Why is this problematic for predicting Rutgers’ bowl eligibility chances? I think there are a couple of reasons. First, rebuilds are tricky things to time. As I’ve mentioned in this space before, well-coached college teams have a tendency to outperform projections quicker than expected. Second, last year was really wonky, and even if SP+ decided to solve for it by using 2019 full season data instead (not sure if this is case), Rutgers’ numbers would be really wonky either way because 2019 was also a weird year for Rutgers football.

(For whatever it’s worth, SP+ is, like many, thinking Rutgers will fall back to earth later this season. RU is ranked 72nd out of 130 FBS schools at the moment overall, with defensive efficiency ranked in the mid-40s and ranked 90^th in offensive efficiency. If you’re familiar with my KenPom analysis during basketball season, you can take efficiency to mean the same thing here – it’s a measure of the quality of a unit adjusted for speed of play.)

This is a long-winded way of saying, in this specific instance, I actually prefer ESPN’s Football Power Index (FPI) to analyze Rutgers’ chances of hitting six wins. You can check out the data here. In short,

• Rutgers right now has a 76.3% (!!) chance of making it to six wins this season,
• They are projected to have a season record of 6.5-5.5 (the decimal points are because of the uncertainty in the projection),
• Plus a 0.8% chance of winning the division (which, I mean, LOL, but also this is higher than Maryland or Indiana’s chances right now which is a cool nugget)

When I look at this data, I put my skeptical analyst hat on. There’s an old saying by the ethicist Stephen Jay Gould, “the median is not the message” which certainly applies here.

FPI is built off hundreds of historical simulations and it’s telling us that historically, teams like Rutgers, who start the season like Rutgers, land at six or more wins 76.3 percent of the time. But (a) very few teams who start 2-0 have a conference slate as difficult as Rutgers, (b) FPI may have a tendency to overweight turnovers, which Rutgers has been good at forcing this season (+8 through two games is ::checks notes:: really good), and (c) I’m not sure FPI is great at predicting the quality of FCS opponents, because it’s an FBS model, and Delaware is a stronger than average FCS team with a real chance to scare Rutgers on Saturday. FPI may be assuming a 3-0 start, and we know better than that as Rutgers fans.

If I had to guess, Rutgers is more than a toss-up to make it to six wins, but if Vegas offered me 3:1 against Rutgers hitting six wins, I’d take that bet. While the Big Ten may end up be weaker than usual this season, and Rutgers may be able to take advantage of this later in the season in a few games, I wouldn’t book tickets to the Beef O’Brady’s Bowl just yet.

For now, in the early going, I find these numbers encouraging, and a bit surprising in a positive direction. At the very least, it’s cool Rutgers is 2-0 allowing us to have this discussion.