Can Michigan bounce back from a shocking loss and beat Minnesota this weekend? The markets have Michigan as a two-touchdown favorite, but is this too many points?
Predicting this game got more complicated with the sudden retirement of Minnesota head coach Jerry Kill. How will Minnesota play without its leader?
Let's break down the matchups and see if analytics can shed any light on the game.
Michigan's defense against Minnesota's offense
Minnesota's offense will struggle against Michigan's elite defense.
Minnesota no longer has RB David Cobb, who rushed for 183 yards on 5.7 yards per carry last season against Michigan. Their offense has changed without Cobb, as they run the ball on 51 percent of plays compared with 68 percent last season.
This increased percentage of passes makes QB Mitch Leidner the focus of the offense. He has responded by completing a higher percentage of his passes but for fewer yards per attempt than last season. Minnesota has resorted to a short passing game.
No matter what Minnesota tries, they'll find it difficult against Michigan's defense. The Wolverines defense has allowed 3.79 yards per carry, best in the nation, and 3.04 yards per pass attempt, second best in the nation. (These numbers count sacks as pass plays.)
Michigan's offense against Minnesota's defense
Michigan sets up its offense with the run, as the Wolverines have run on 57 percent of plays this year. This matches up well against a Minnesota defense that plays the pass better than the run.
In my numbers that take yards per play and adjusts for strength of schedule, Minnesota ranks 16th in pass defense. The Gophers' two cornerbacks, Eric Murray and Briean Boddy-Calhoun, show up on 2016 NFL draft lists.
When Michigan throws, it should tend to target TE Jake Butt or any other receiver not covered by these two players.
However, Minnesota ranks 61st in rush defense, providing an opportunity for Michigan to control the game on the ground. In addition, Minnesota defensive end Theiren Cockran is listed questionable to play in this game. Expect Jim Harbaugh to give Minnesota a heavy dose of De'Veon Smith and Drake Johnson.
How does Jerry Kill's absence affect this game?
Minnesota coach Jerry Kill retired from football this week due to health concerns. However, I'm not sure this matters much.
Kill has kept his staff together through jobs at Southern Illinois, Northern Illinois and now Minnesota. Tracy Claeys, the defensive coordinator, takes over as head coach and has been with Kill since 1995. Coordinators Matt Limegrover (offense) and Jay Sawvel (special teams) have also been on the staff for at least 14 years.
We can also get some insight in Kill's absence from the 2013 season, when he missed some games due to the same health issues. At The Power Rank, I assign teams a rating by taking margin of victory in games and adjusting for schedule strength. Minnesota made a 7-point improvement from 2012 to 2013. Kill's absence didn't prevent his team from making this large improvement.
Last, the markets haven't moved much because of Kill's retirement. Michigan has moved from a 13.5- to 14-point favorite after the news broke.
My preseason numbers had Michigan as a 1-point favorite over Minnesota.
The Wolverines have improved dramatically, as their rating has increased 4 points since the beginning of the season. That might not seem like much. However, if two teams of equal strength play on a neutral field, they each have a 50-percent win probability. If one team raises its rating by 4 points, that team now has a 62-percent win probability.
Michigan could still improve in the remainder of the season, but there's a ceiling without a stronger downfield passing game.
It might seem like Minnesota has had a disappointing season so far, as the Gophers lost by large margins to Northwestern and Nebraska. However, they beat Purdue by 28, and a 6-point loss to playoff contender TCU counts as a strong performance.
Overall, Minnesota has barely budged from my preseason projection. This implies Michigan should be a 5-point favorite, which leaves a 35 percent chance that Minnesota pulls the upset. This predicted point spread is probably low, as I don't think my model gives Michigan enough credit for its special teams play. Michigan ranks first in ESPN's special teams numbers.
Michigan should win this game, especially if it can get an early lead and rely on the defense and running game. However, the Wolverines will find it difficult to beat Minnesota if they fall behind and have to throw the ball against a strong secondary.
If Michigan beats Minnesota by more than two touchdowns predicted by the markets, it will signify continued improvement for this team.