Three Bowl Projections Agree About Michigan Until They Have To Name The Bowl

Michigan's season has been assigned a destination twice this summer, by three men, and they do not agree on the destination.

Kyle Bonagura and Mark Schlabach published ESPN's preseason playoff and bowl projections on August 22. Both of them put Michigan in the ReliaQuest Bowl, at Raymond James Stadium in Tampa, at noon on December 31. They disagree only about the opponent, Bonagura taking LSU and Schlabach taking Oklahoma.

Steven Lassan published Athlon Sports' version on July 7. He has Michigan in the Citrus Bowl, at Camping World Stadium in Orlando, on January 2, against Ole Miss. (July 7 is six weeks before the AP poll existed, which is worth carrying rather than hiding, because it means Lassan was not reading the same room the other two were.)

Same team, same conference, same December, and two different stadiums three hours' drive apart. That is what the preseason agreement about Michigan looks like when you make it be specific about something.

The part they do agree on, and it is the same three teams every time#

Michigan is not in the twelve-team playoff field in any of the three brackets. It is also not the fourth-best team in its own conference in any product that ranks the conference, and the three teams above it are the same three in all of them.

Lassan's Big Ten preseason power rankings, published August 5, run Ohio State, Indiana, Oregon, then Michigan.

Daire Carragher's dark-horse piece at Pro Football Focus, published August 20, calls Michigan one of the outsiders with a shot rather than one of the contenders. His words: "Beyond the obvious contending trio of Ohio State, Indiana and Oregon, it's probably one of USC, Michigan or Iowa that represents the best shot at an outsider run."

Bonagura's twelve and Schlabach's twelve contain exactly three Big Ten teams each, and it is the same three: Ohio State, Indiana, Oregon. Neither of them wrote the PFF piece and the PFF piece was published two days earlier at a different company.

Bill Connelly's final preseason SP+ numbers, published at ESPN on August 13, project the Big Ten's top four as Ohio State, Oregon, Indiana and Penn State. Michigan is not in that four either. Connelly's own summary of the conference: "There's definitely a 'top three and everyone else' vibe here."

Four organisations. Seven products between July 7 and August 22. The same three teams sitting above Michigan in every one of them that puts teams in order.

Which is mostly the poll, doing arithmetic#

The Associated Press released its preseason Top 25 on Monday, August 17. Michigan is 16th with 718 points. Ohio State is first with 1,672 points and 40 of the first-place votes.

A twelve-team bracket takes twelve teams. Sixteenth is four places outside it.

So before anybody decides that a projection is a judgment about Michigan, the check worth running is whether the projections depart from the poll at all up there. They do not. AP's top seven is Ohio State, Oregon, Georgia, Notre Dame, Texas, Indiana and Miami. All seven appear in Bonagura's field. All seven appear in Schlabach's field.

Nothing has been said about Michigan by either projector that the poll had not already said five days earlier. A bracket that takes twelve was applied to a team ranked sixteenth, and it produced the answer that operation produces. The ReliaQuest Bowl is not an opinion about the Wolverines. It is a subtraction.

The one instrument that shows its work#

Six of the seven products are somebody's judgment, written down. Nobody is hiding that. Lassan's rankings are Lassan's rankings, Carragher's dark horses are prose, the AP poll is 69 people voting, and ESPN's bowl projections are two named men guessing in August about January.

SP+ is the exception, and it is worth separating out. Connelly describes it as "a tempo- and opponent-adjusted measure of college football efficiency," a predictive rating rather than a résumé, and he says in terms that the projections "aren't intended to be a guess at what the AP Top 25 will look like at the end of the year."

That is a different kind of machine from a magazine's list, and it produces the same neighbourhood. Michigan is outside the Big Ten's projected top four. It also carries the fifth-toughest projected schedule in the country, which is the highest of any Big Ten team except Ohio State.

The instruments do not share a method. They share an answer.

The schedule is the mechanism, and it checks out#

Carragher says why, and it is the least mysterious sentence in any of this. "USC and Michigan both draw all three of the favorites, while Iowa avoids both Oregon and Indiana."

Michigan's own published schedule says he is right. Indiana comes to Ann Arbor on October 24, which is game seven. Michigan goes to Eugene on November 14, game ten. Michigan goes to Columbus on November 28, game twelve.

Eight home games, four away, and the two the AP ranks highest are the road ones. The season opens with Western Michigan, Oklahoma, UTEP and Iowa at home in September, and then hands Michigan Penn State, Indiana, Oregon and Ohio State across a stretch that runs from October 17 to the end of November.

That is the shape of a schedule that produces a good record or a playoff résumé and does not often produce both.

Bryce Underwood is the part of this that is a person#

Lassan ranked all eighteen Big Ten starting quarterbacks on August 21 and put Underwood sixth. Michigan is fourth on his team list. (Sixteen days separate the two lists, and nothing obliges a magazine's team rankings and its position rankings to agree.)

The quarterback rates two places below his own team, on the same magazine's two products, and the gap has names in it. The three teams above Michigan in Lassan's power rankings supply the top three quarterbacks on his quarterback list: Julian Sayin at Ohio State first, Dante Moore at Oregon second, Josh Hoover at Indiana third.

That is where four places outside a bracket actually lives. Not in a magazine's opinion of Michigan's roster, which Lassan calls a strong one with a defense and an offensive line among the conference's best, but at one position, in a conference where the three teams ranked ahead of Michigan on his list are the three with the quarterbacks ranked ahead of Underwood.

Carragher gets to the same place from a different direction and calls Underwood "Michigan's biggest variable." His supporting number is about movement rather than accuracy: Underwood's 42 scrambles are the fifth-most of any returning Power Four quarterback, and he had the quickest release of anyone in that top fifteen. Underwood threw for 2,428 yards last season with 11 touchdowns and nine interceptions.

What the agreement is not#

Two of the seven products carry the same masthead. Every one of them was published before a snap. A sixth projector, Brett McMurphy's Week 0 bowl projections at On3, sits behind a subscriber wall and was not read. Seven products is a count, not a consensus.

And the sharpest argument against reading too much into the convergence comes from inside one of the products. Lassan's August 5 power rankings say Michigan and USC "should be squarely in the mix to make the College Football Playoff," and his own July bracket does not have Michigan in it. That is one editor, one magazine, one summer, holding both.

Which is the honest size of the finding. Seven documents put Michigan in the same band because they are all reading the same offseason and mostly the same poll. Ask them anything narrower and one of them sends Michigan to Tampa and another sends it to Orlando, and neither of them has watched this team play.

Michigan plays Western Michigan in Ann Arbor at 7:30 on Saturday night, and the game is sold out. On Sunday morning the sixteenth-ranked team in the country will have done something a projection can be checked against, for the first time since any of this was written down.