From a numbers standpoint, though, the most noticeable shift might have come up top. Ohio State handled Rutgers easily, 49-27, in Columbus on Saturday evening, but the bar for dominance is awfully high among the 99th-percentile teams. The Buckeyes' overall SP+ rating slipped just enough for idle Alabama to ease into the top spot for the first time all year. Does this really matter? No -- the Tide and Buckeyes still hold a solid advantage over anybody remaining on their respective schedules. But a new No. 1 is always noteworthy, even if it's a team that has spent large portions of the last 12 seasons in the top spot.
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.
Here are the full numbers.
This week's movers
With yet another batch of teams making their season debuts, we once again had a solid number of market adjustments this week.
Twelve teams moved up more than 10 spots from last week. Six of them have played one game.
- Colorado (1-0): up 21 spots from 90th to 69th
- Maryland (2-1): up 20 spots from 88th to 68th
- Washington State (1-0): up 19 spots from 59th to 40th
- Buffalo (1-0): up 19 spots from 78th to 59th
- Toledo (1-0): up 18 spots from 96th to 78th
- Western Michigan (1-0): up 16 spots from 86th to 70th
- Georgia Southern (5-2): up 15 spots from 82nd to 67th
- Iowa (1-2): up 12 spots from 28th to 16th
- Nevada (3-0): up 12 spots from 85th to 73rd
- Coastal Carolina (7-0): up 11 spots from 49th to 38th
- Kansas State (4-3): up 11 spots from 67th to 56th
- Miami (Ohio) (1-0): up 11 spots from 113th to 102nd
The most interesting team to me here is Iowa. The Hawkeyes are the prototype "numbers like them more than humans" team in that they have now lost two games by an eyelash -- to Purdue after they fumbled while driving for the game-clinching points and lost via late field goal, and to Northwestern after blowing a 20-7 lead and coming up short on two late drives into Wildcat territory -- and won one via utter destruction. They obliterated Michigan State, going up 35-0 at halftime and cruising, 49-7.
This odd résumé combination has now moved them up to 16th overall and second in the Big Ten West ... more than six adjusted points per game ahead of unbeaten Northwestern. That is jarring to the eye but somewhat justifiable on paper.
Similarly, 11 teams moved down more than 10 spots this week. Five of them are coming off of their season debuts.
- Rutgers (1-2): down 20 spots from 87th to 107th
- Ohio (0-1): down 19 spots from 69th to 88th
- Stanford (0-1): down 18 spots from 45th to 63rd
- Boise State (1-2): down 15 spots from 26th to 41st
- Duke (2-6): down 15 spots from 82nd to 87th
- Ball State (0-1): down 15 spots from 84th to 99th
- South Carolina (2-4): down 14 spots from 65th to 79th
- Air Force (1-2): down 13 spots from 81st to 94th
- UCLA (0-1): down 12 spots from 62nd to 74th
- Oregon State (0-1): down 11 spots from 75th to 86th
- Tennessee (2-4): down 11 spots from 54th to 65th
Georgia's season hasn't gone quite as planned, but the Dawgs can at least take pleasure in having broken Tennessee. The Vols were 2-0 and a solid 22nd in SP+ and led UGA before getting overrun, 27-0, in the second half. They got blown out by both Kentucky and Alabama, and followed that up with another second-half collapse at Arkansas: they led 13-0 at half and lost 24-13. The SEC East now has two top-10 teams ... and only two top-55 teams.
Air Force's drop, by the way, was a funky result of conference-level adjustments now that MWC teams have played a decent number of games. SP+ wasn't programmed for an "only two teams in your conference play non-conference games" scenario like what the MWC is facing, and it's reacting to it oddly, such that Boise State's blowout loss to BYU dragged everyone down a bit. Weird math for a weird season.
Why does SP+ hate Notre Dame?
After a good week against the spread (66% overall, moving it back up to 54% for the season), SP+ is apparently particularly cocky and anti-social. It responded to Notre Dame's biggest win in forever by ... dropping the Fighting Irish one spot to eighth.
What gives? Two things, mainly. (No, one of them isn't a "don't let your fans rush the damn field and surround your players during a damn pandemic" point deduction.)
1. Statistically, the game was more of a tie than a win. Good, predictive ratings don't live in a world of 1s and 0s; a dominant win is worth more than a slight win. My postgame win expectancy measure -- which takes the key, predictive stats from a given game, tosses them into the air and says "you could have expected to win this game X% of the time" -- placed the Irish at 42% for this one. Clemson held a success rate advantage of six percentage points (45%-39%) and operated better in the red zone. That's usually enough for an advantage.
That the Irish overcame that and won anyway made the win even more stirring, but SP+ didn't respond quite as well to the win as you might initially think. (This also explains why Clemson eased back ahead of idle Wisconsin for third overall. The Tigers played a top-10 team to a statistical draw on the road.)
2. BYU and Oregon looked incredible. Notre Dame didn't sink so much as the Cougars and Ducks rose more. BYU utterly dominated what had been a top-30 Boise State team on the road, winning by a 51-17 margin that could have been even worse. Oregon got perhaps an artificial boost from playing a diminished Stanford that was without its starting quarterback, but while the Cardinal pulled off a few big plays and earned some turnovers luck, Oregon's success rate advantage (56%-37%) suggests that Stanford was fortunate to stay within 21.
Does any of this particularly matter? Not really. Notre Dame will now be favored in every remaining game until the ACC Championship, as will Clemson, and the Irish obviously have an excellent CFP case at the moment regardless of what SP+ thinks. But never underestimate the numbers' ability to tell fans, "Well actually, that good thing wasn't as good as you think."
Brian Kelly: 'We proved something to ourselves' by beating Clemson
Notre Dame coach Brian Kelly opens up about Notre Dame's upset win over No. 1 Clemson.