Hits and misses: Preseason college football win totals revisited

Michigan State's preseason win total was 4.5 by Caesars and 5.5 by SP+, but the Spartans sit at 7-0. AP Photo/Carlos Osorio

After 60 or more FBS games in each of the first five weeks of the season, there were only 51 in Week 6 and 52 in Week 7. There are 54 in Week 8, and four of the AP's top 15 teams are off; there are no matchups between two ranked teams, and the six highest-ranked teams in action are all favored by at least 20 points.

We can never rule anything out, but after all the action and chaos of the first seven weeks, it's likely this will be a pretty tame weekend. A bye week for us, if you will.

A bye week is a prime opportunity for coaches to turn their gaze inward and do some good, hard self-scouting. So let's do the same.

All things considered, my SP+ ratings have performed pretty well this season. Despite all the uncertainty of whether we could trust anything we saw in the strange, disjointed 2020 season, SP+ preseason projections held up, and it is currently at 51.8% against the midweek Caesars spread (FBS vs. FBS games only) and, thanks to some dynamite FCS numbers, 56.5% against the closing spread (FBS vs. FCS games included).

There have been a few more big whiffs than normal: the absolute error (the absolute-value average of the actual scoring margin minus the projected margin) is at 13.1 points per game, 0.7 higher than it was in a brilliant 2019. But really, that was to be expected. The sportsbooks themselves have had a higher absolute error this year as well.

In the name of self-scouting, let's check our work. I listed SP+ win projections for every FBS team before the season -- let's see where the hits and misses have come from and what lessons we might learn.


Largest preseason disagreements:

LSU Tigers (4-3)

Preseason expectations: 8.5 wins per Caesars, 7.1 per SP+ (under)
Current odds of hitting the preseason under, per SP+: 99%