In Week 2, both of the Big Ten's favorites lost. Michigan State fell on the road to Oregon, while Ohio State lost at home to Virginia Tech. At that point, we all decided that our suspicions were correct; the Buckeyes were in a full-fledged rebuilding year, and despite the loss in Eugene, Michigan State would likely cruise to the Big Ten title.
The following six weeks have muddied the water. A young Ohio State team has found its legs, playing nearly elite offense and special teams and establishing a higher level of play on defense. Meanwhile, Michigan State has cleared every hurdle, mostly with ease, but has yet to play a full, 60-minute game and has shown just enough cracks to make you wonder.
Actually, the numbers aren't wondering at all. They think the Buckeyes are good enough to win in East Lansing and take the Big Ten title. Ohio State is up to eighth in the F/+ rankings, while Michigan State is laboring at 18th, 10 spots below its AP ranking. Why is that, exactly?
Push 'em out, push 'em out, waaaaay out
The F/+ ratings are the combination of two sets of ratings: my S&P+ and Brian Fremeau's FEI. They look at different stats in different ways. About two-thirds of my ratings are based on play-by-play data, one-third is based on drive data, and all of it is weighted equally. Fremeau's measure is based solely on drive data and applies extra weight to games against good teams. They travel two paths with two different philosophies, but both ratings end up correlating about the same to wins and losses, and the combined F/+ correlates even better.
That State ranks fifth in S&P+ but 25th in FEI tells you where to start looking. If you're much lower in FEI than S&P+, either you struggle to close out drives on one side of the ball (or both) or you've performed worse against good competition than others. For State, it's both.
Michigan State's offense is closing out drives quite well. The Spartans are averaging 5.1 points per scoring opportunity (drives with a first down inside the opponent's 40), 20th in the country. But while the defense isn't allowing tons of opportunities, they're giving you touchdowns on the chances you get. Opponents are averaging 4.8 points per scoring opportunity, 108th in the country.
A lot of this stems from big plays. The Spartans play a high-risk, high-reward defense, one that suffocates opponents and prevents them from playing efficient football. But if an opponent finds a leak in the defense, it turns into a burst dam.
More Michigan-Michigan State
Last year, State ranked first in the country in success rate, but 74th in IsoPPP, which measures the magnitude of the successful plays you're allowing. This year, following the loss of quite a few defensive difference-makers, the efficiency is still there -- State currently has the second-best success rate in the country. But the big plays are even bigger: State is 124th in IsoPPP.
Big plays only matter a little if you're not allowing many of them. Efficiency über alles. But Oregon managed to average 7.2 yards per play against the Spartans, and while State was able to swallow up Indiana's passing attack last Saturday, receiver Shane Wynn ripped off a 75-yard touchdown run, and star running back Tevin Coleman had rushes of 65 and 30 yards. And despite an overwhelmed freshman quarterback, IU was still within a 28-17 margin midway through the third quarter before Michigan State pulled away.
Vs. good competition
That brings us to the other factor for disagreement between FEI and S&P+: performance against good teams.
Against Jacksonville State, Eastern Michigan, Wyoming, Purdue, and Indiana, Michigan State is winning games by an average score of 55-17, averaging 7.2 yards per play and allowing 4.6 yards per play. Complete domination.
But against Oregon and Nebraska, the only two top-60 teams yet on the schedule, State was outscored, 68-54, averaging 5.7 yards per play and allowing 5.5. The Spartans were handling Nebraska just fine until the fourth quarter, when the Huskers made a late charge, aided by a punt return touchdown. Still, if we're supposed to be looking at Michigan State as an elite national title contender, we'd like to see a combined performance better than this against the No. 7 (Oregon) and No. 17 (Nebraska) teams.
This could just be a sample size issue. It's hard to base strong conclusions off of two performances, after all. But it's a warning sign. Michigan State looked like an elite team for two quarters against Oregon and three against Nebraska, but that won't get you very far against other elite teams. And one of those will be coming to town in a couple of weeks.
How does that apply to this weekend?
It doesn't. Michigan is technically the third-best opponent State has faced in 2014, ranking 55th in F/+ (72nd on offense, 42nd on defense, 82nd on special teams). The Wolverines have had a bye week to regroup from a series of results that put head coach Brady Hoke on either the hottest or second-hottest seat in the country. They are a complete disaster when they fall behind schedule (113th in Passing Downs S&P+), and State's defense treats you like every down is a passing down.
Michigan does bring some strengths to the table -- for one thing, the Wolverines are better at running the ball this year with sophomores Derrick Green and De'Veon Smith combining to average 5.5 yards per carry (a lot of which came against bad defenses, yes); for another, they're quite stout against the run themselves -- but State brings more. Anything is possible in college football, but it will take a significant letdown for the Spartans to lose.
If one of the two Big Ten favorites lose this week, then, it will probably be Ohio State. The Buckeyes are smoking hot and hold most of the advantages against Penn State, but the Nittany Lions do still have a top-10 defense (they're No. 9 in Def. F/+, actually ahead of No. 13 Michigan State), and if Ohio State's redshirt freshman quarterback J.T. Barrett is due a letdown, it's likely to come against an excellent defense in a hostile locale. Granted, Penn State will still have to score to win, and that's not a given.
As it pertains to the Big Ten race, each team is allowed a mulligan. The winner of the November 8 game in East Lansing will likely determine the
West East division winner even if a favorite falls once.