With so few top teams playing in Week 11, it was easy to assume we wouldn't see much movement atop either the polls or computer rankings. The SP+ ratings still managed to do some shuffling, though.
Wisconsin's second dominant performance of the season again allowed the Badgers to jump idle Clemson into third, while Notre Dame's comprehensive win over Boston College gave the Fighting Irish just enough juice to hop Florida into fifth. Throw in a slight drop for Oregon -- which beat Washington State but had to labor to do it -- and a slight, opponent-adjustment-driven rise for idle Ohio State, and each of the top 18 teams in SP+ either rose or fell at least one spot. None moved much, but they all danced at least a bit.
(That's right, idle Ohio State jumped idle Alabama this week. The Buckeyes' previous opponents looked a little better than Bama's, apparently. The numbers are always churning behind the scenes whether you play or not.)
What is SP+? In a single sentence, it's a tempo- and opponent-adjusted measure of college football efficiency. I created the system at Football Outsiders in 2008, and as my experience with both college football and its stats has grown, I have made quite a few tweaks to the system.
More than ever, it's important to note that SP+ is intended to be predictive and forward-facing. It is not a résumé ranking that gives credit for big wins or particularly brave scheduling -- no good predictive system is. It is simply a measure of the most sustainable and predictable aspects of football. If you're lucky or unimpressive in a win, your rating will probably fall. If you're strong and unlucky in a loss, it will probably rise.
This week's movers
With fewer teams making their season debuts and/or playing far different than projected, we didn't see quite as much extreme movement in the overall rankings -- only 15 teams moved up or down by double digits, as opposed to 27 last week.
Moving up
Eight teams moved up at least 11 spots this week.
Nevada: up 16 spots from 73rd to 57th
San Diego State: up 15 spots from 61st to 46th
Arizona: up 15 spots from 100th to 85th
Houston: up 13 spots from 64th to 51st
Louisiana: up 12 spots from 46th to 34th
San Jose State: up 12 spots from 81st to 69th
Oregon State: up 12 spots from 86th to 74th
Kent State: up 11 spots from 111th to 100th
Arizona couldn't quite pull off an upset against USC, but Kevin Sumlin's Wildcats certainly fared better than projected overall and rose accordingly. (It helps to have a really low bar to clear -- they were projected 122nd in defensive SP+ and rose to 115th after holding the Trojans to only 6.6 yards per play and 34 points.)
Moving down
Seven teams moved down at least 10 spots:
Michigan State: down 19 spots from 62nd to 81st
Minnesota: down 17 spots from 31st to 48th
Miami (Ohio): down 13 spots from 102nd to 115th
Boston College: down 12 spots from 52nd to 64th
Baylor: down 12 spots from 50th to 62nd
Virginia: down 11 spots from 54th to 65th
Southern Miss: down 10 spots from 91st to 101st
Virginia moved to 3-4 after a 31-17 win over Louisville, but the Cavaliers still fell by a decent amount. How? By winning in a pretty unsustainable way.
Louisville averaged 8.0 yards per play to UVA's 5.7 and finished eight drives in opposition territory to UVA's six. But Virginia scored from the most high-leverage play in football -- a long touchdown off a turnover (Noah Taylor's 85-yard pick six, which produced a nearly 14-point swing) -- and recovered a fumble at its 35 as well.
My postgame win expectancy number, which looks at the game's key, predictive stats, tosses them into the air and says, "With these stats, you should have expected to win this game X% of the time," projected UVA to have only a 22% chance with the stats at hand. Paired with their win over UNC, that's two straight unlikely (and very entertaining) wins for the Hoos.
Why doesn't SP+ hate Penn State?
Each week in this space, I've looked at a team that doesn't seem to get as much credit from the numbers as its record demands. This week, I'll look at the opposite: an outlier ranking much higher than its record suggests it should.
As you would expect, Penn State fell in the ratings following a 30-23 loss to Nebraska. But the Nittany Lions didn't fall much. In fact, despite an 0-4 record, they still rank 16th overall, ahead of teams with infinitely better records -- 7-1 Miami, 4-0 Indiana and 4-0 Northwestern, for instance. How is this possible?
There are a couple of reasons for why PSU hasn't completely plummeted:
1. Preseason projections are still pretty heavy. The way SP+ works is, you start with your projected preseason ratings, and they slowly get phased out during the season. They are never completely phased out, as priors make predictive systems more successful whether it feels like they should or not. Still, through four games, more than half of Penn State's rating is derived from the fact that the Nittany Lions were projected sixth in the preseason. It takes a while to fall.
2. SP+ basically sees wins in the Indiana and Nebraska games. In the season debut against Indiana, the Nittany Lions outgained the Hoosiers by 277 yards, generated 11 more first downs and finished four more drives inside the 30 than Indiana did. On Saturday in Lincoln, PSU outgained Nebraska by 203 yards, generated 13 more first downs and finished seven of 11 drives inside the Huskers' 25. (Nebraska finished five drives inside PSU's 25.) But the Nittany Lions failed to score on either of two late red zone trips and handed NU 10 points from turnovers -- a 26-yard fumble return score and an interception returned to the PSU 15, which set up a field goal.
If you think of a sport as a combination of (A) setting up chances to win the game and (B) converting them, PSU did (A) just fine. Postgame win expectancy was 95% against IU and 83% against NU, which means that there was a 79% chance of winning both games and only a 1% chance of losing both.
Events with a 1% chance happen 1% of time, however, and PSU is 0-4 because of it.
In Penn State, SP+ basically sees a team that was projected sixth, likely "won" two games, lost to the top team in the country by an acceptable margin and laid one egg, last week's dreadful game against Maryland. It has therefore dropped the Nittany Lions only a fair amount instead of a lot. And since their next game is against an Iowa team ranked higher than them, it's possible that PSU could play pretty well, lose to a superior team ... and remain in the top 20 at 0-5. I'm sure that wouldn't get me yelled at on Twitter at all.