stupid predictions

[Bryan Fuller]

Previously: Podcast 14.0A, 14.0B, 14.0C. The Story. Quarterback. Running Back. Wide Receiver. Tight End. Offensive Tackle. Interior OL. Defensive Interior. Edge. Linebacker. Cornerback. Safety. Special Teams. 5Q5A Offense. 5Q5A Defense.

Heuristicland

Turnover Margin

9674598390_9cc93827ae_o

The theory of turnover margin: it is pretty random. Teams that find themselves at one end or the other at the end of the year are likely to rebound towards the average. So teams towards the top will tend to be overrated and vice versa. Nonrandom factors to evaluate: quarterback experience, quarterback pressure applied and received, and odd running backs like Mike Hart who just don't fumble.

Year Margin Int + Fumb + Sacks + Int - Fumb - Sacks -
2007 0.15 (41st) 14 15 2.46(33rd) 14 13 2.17 (67th)
2008 -.83 (104th) 9 11 2.42(33rd) 12 18 1.83 (57th)
2009 -1.00 (115th) 11 5 1.83(68th) 15 13 2.33 (83rd)
2010 -0.77(109th) 12 7 1.38(98th) 15 14 0.85(10th)
2011 +0.54 (25th) 9 20 2.31 (29th) 16 6 1.38 (33rd)
2012 -0.69 (99th) 7 11 1.69 (69th) 19 8 1.38 (28th)
2013 +0.38(33rd) 17 9 1.9 (64th) 13 8 2.77 (109th)
2014 -1.33 (124th) 5 5 2.4 (49th) 18 8 2.2 (63rd)
2015 -0.31 (92nd) 10 2 2.5 (32nd) 10 6 1.4 (28th)
2016 +0.54 (24th) 13 6 3.54(5th) 7 5 1.69 (39th)
2017 -0.31 (90th) 10 7 3.23(8th) 10 11 2.77 (111th)
2018 +0.38 (35th) 11 6 10.5% (3rd) 9 3 5.4% (43rd)
2019 +0 9 11 9.0% (16th) 9 11 6.1% (61st)
2020 +0.38 (35th) 11 6 10.5% (3rd) 9 3 5.4% (43rd)
2021 +0.14 (56th) 8 8 6.8% (62nd) 5 9 3.3% (6th)

Despite having extremely good turnover avoidance Michigan was almost dead even a year ago. The QBs and most of the OL return so sack avoidance should still stay high. The pressure number for Michigan is distorted by the inevitability of Hutchinson—most teams did a lot of dinking. I mostly include this in case Michigan is extraordinarily lucky or unlucky and can expect some regression. This doesn't look like much of anything, projection-wise.

[After THE JUMP: one sort of alarming switch]
[Bryan Fuller]

Previously: Podcast 11.0A, Podcast 11.0B, Podcast 11.0C. The Story. Quarterback. Running Back. Wide Receiver. Tight End. Offensive Tackle. Interior OL. Defensive End. Defensive Tackle. Linebacker. Cornerback. Safety. Special Teams. 5Q5A Offense. 5Q5A Defense.

Heuristicland

Turnover Margin

9674598390_9cc93827ae_o

The theory of turnover margin: it is pretty random. Teams that find themselves at one end or the other at the end of the year are likely to rebound towards the average. So teams towards the top will tend to be overrated and vice versa. Nonrandom factors to evaluate: quarterback experience, quarterback pressure applied and received, and odd running backs like Mike Hart who just don't fumble.

Year Margin Int + Fumb + Sacks + Int - Fumb - Sacks -
2007 0.15 (41st) 14 15 2.46(33rd) 14 13 2.17 (67th)
2008 -.83 (104th) 9 11 2.42(33rd) 12 18 1.83 (57th)
2009 -1.00 (115th) 11 5 1.83(68th) 15 13 2.33 (83rd)
2010 -0.77(109th) 12 7 1.38(98th) 15 14 0.85(10th)
2011 +0.54 (25th) 9 20 2.31 (29th) 16 6 1.38 (33rd)
2012 -0.69 (99th) 7 11 1.69 (69th) 19 8 1.38 (28th)
2013 +0.38(33rd) 17 9 1.9 (64th) 13 8 2.77 (109th)
2014 -1.33 (124th) 5 5 2.4 (49th) 18 8 2.2 (63rd)
2015 -0.31 (92nd) 10 2 2.5 (32nd) 10 6 1.4 (28th)
2016 +0.54 (24th) 13 6 3.54(5th) 7 5 1.69 (39th)
2017 -0.31 (90th) 10 7 3.23(8th) 10 11 2.77 (111th)
2018 +0.38 (35th) 11 6 10.5% (3rd) 9 3 5.4% (43rd)

Let's move to sack rates instead of raw numbers; probably should have done this years ago. Michigan landed a little above average in turnover margin, which is about what you'd expect since they got after the QB more than their quarterback was gotten after.

Not a lot to say as a result. Michigan wasn't an outlier and shouldn't expect to be better or worse this year than you might expect. Michigan gets their QB and four OL back, which should help keep Michigan TOs low. If they can keep up the pressure you'd expect them to be modestly positive again at the end of the year. But turnovers are a fickle mistress, as always.

[After the JUMP: position switches are minimal]

Previously: The story, the secondary, the linebackers, the defensive line, the offensive line, the receivers, the running backs, the quarterbacks, special teams, defensive questions, offensive questions.

Heuristicland

Turnover Margin

denard-fumble

The theory of turnover margin: it is nearly random. Teams that find themselves at one end or the other at the end of the year are highly likely to rebound towards the average. So teams towards the top will tend to be overrated and vice versa. Nonrandom factors to evaluate: quarterback experience, quarterback pressure applied and received, and odd running backs like Mike Hart who just don't fumble.

Year Margin Int + Fumb + Sacks + Int - Fumb - Sacks -
2007 0.15 (41st) 14 15 2.46(33rd) 14 13 2.17 (67th)
2008 -.83 (104th) 9 11 2.42(33rd) 12 18 1.83 (57th)
2009 -1.00 (115th) 11 5 1.83(68th) 15 13 2.33 (83rd)
2010 -0.77(109th) 12 7 1.38(98th) 15 14 0.85(10th)

WELCOME TO YET ANOTHER YEAR where I predict Michigan's turnover rate plunges towards zero. I'm seriously this time though.

For the first time on this chart Michigan should have a non-insane person running things. In 2007, it was either injured Henne or Mallet; 2008 was death, 2009 was freshmen QBs, and last year was essentially a redshirt freshman. With Denard's return this is the first time since 2006 Michigan can expect their QB to be less turnover prone than the year before. (This obviously goes out the window in the event of a major injury to Denard. Also out the window: life, hope, puppies.)

But… I'm seriously this time. Even if Rodriguez had some weird evil turnover juju when he was around he's gone. Turnovers regress like a mofo. People have argued with me about this plenty and I do believe them somewhat:

  • NFL turnover margins regress like a mofo and always will.
  • College TO margins might have extra regression because low turnover teams tend to have senior quarterbacks and then break in new ones, and high turnover teams tend to have young quarterbacks who return. What looks like randomness is potentially roster turnover.
  • Sucky defenses case fewer turnovers because things are easy.

So Rodriguez-era stuff was negative because the defenses were turrible and the quarterbacks were young. The defense does trace a largely negative track as it declines from 29 turnovers in the last Carr year to 20 in RR year 1, 16 in RR year 2, and 19 in RR year 3. Turnovers from the offense are about constant in the era of lots of freshmen, but in 2006 Michigan coughed it up just 12 times.

If Robinson remains healthy Michigan should improve significantly. The defense has to suck less and Robinson's responsibility should improve rapidly relative to players more than a year removed from being novelty freak shows. I'm afraid that Robinson is just a fumble-prone guy—Mike Hart didn't need experience to hold on to the damn ball—but the interception rate should dip considerably.

On the other side of the ball, a defense that rushes more than three players and has Martin, RVB, and Roh should get back to at least average in sacks. The center of the Gaussian distribution here is probably –3 turnovers on the year; even that would be massive improvement.

Position Switch Starters

Theory of position switches: if you are starting or considering starting a guy who was playing somewhere else a year ago, that position is in trouble. There are degrees of this. When Notre Dame moved Travis Thomas, a useful backup at tailback, to linebacker and then declared him a starter, there was no way that could end well. Wisconsin's flip of LB Travis Beckum to tight end was less ominous because Wisconsin had a solid linebacking corps and Beckum hadn't established himself on that side of the ball. Michigan flipping Prescott Burgess from SLB to WLB or PSU moving Dan Connor inside don't register here: we're talking major moves that indicate a serious lack somewhere.

mark-moundros-nw

Last year there were a half-dozen of varying severity. This year I'm not sure there are any, except insofar as people on the defense are all switching positions because of the scheme change. I'm not sure how much those count.

Here's a dossier:

  • RVB is now a full time three-tech instead of a 5-tech on a three-man line. He's already started as a three tech in his career.
  • Roh is now a WDE full time instead of a 3-3-5 OLB/DE.
  • Kenny Demens is now a MLB instead of a snack for a guard.
  • Thomas Gordon is a starting safety instead of a SLB/safety-type-object.
  • Some wide receivers are flipping outside from the slot.
  • Third string TE Steve Watson was on defense last year.

None of this comes anywhere close to Mark Moundros maybe starting at LB, Cam Gordon starting at FS, and Roh moving to LB. Anyone who's starting is moving to a spot they've played before or goddamn well should have (Roh).

The lone exception is Thomas Gordon, who is going to be playing at a new position after being a random DB his freshman year, then a spur. That's still not flipping sides of the ball.  It is a concern. At least this year there are no obvious panic moves. Sliding Gordon from a nickelback to safety is not starting John Ferrara or pushing Mark Moundros as the solution at MLB.

An Embarrassing Prediction, No Doubt

Worst Case

There's no bottom if Denard and a couple of other key defensive players are hurt. Leaving the worst-worst case out, a relatively healthy Michigan has no business losing to WMU, EMU, Minnesota, or Purdue at home.

San Diego State, Northwestern, Illinois are all losable but Denard should be able to snake at least one of those. 5-7 is the floor.

Best Case

The schedule is fairly soft, with no true road games until Michigan State (the game at Northwestern will be at least half M fans) and both Penn State and Wisconsin rotating off. If the offense maintains its current level of productivity and Mattison mediocres the defense real good, the only game that still seems entirely out of reach is Nebraska.

That's not to say Michigan can reasonably expect to win all games in reach. Taking more than two from Notre Dame, Michigan State, Iowa, and the Akron State Golden Bobcats seems to be irrational optimism. 9-3 is about all you can reasonably hope for.

Final Verdict

There are a lot of ugly predictions like 5-7, 6-6, and 7-5 from the newspaper folk after their fifty words on the running backs* and it's easy to see why if you're looking at the surface. If you look at the final scores of last year's games it's easy to find extra losses but not extra wins.

If you look at the yardage margins and turnovers it's an entirely different picture. Michigan is poised for a big bounce. Robinson should cut down on his enormous mistakes considerably and a defense that bothers to rush will increase those of opponents. Nineteen starters return; Brendan Gibbons will either be much better or quickly replaced. GERG is gone. The offense will change and that's a drag but the things that made Robinson so insane are not that hard to exploit and he is still rapidly developing. This looks like a team that had a combination of bad luck and youth last year that should improve by leaps and bounds.

The catch: depth. It is a huge issue on both sides of the ball, with a half-dozen players essentially irreplaceable. Injuries happen; with Michigan which injuries will be huge. Huyge or Heininger or Cam Gordon going down is no big deal. Losing Denard or Martin or Demens is massive. A fully healthy Michigan looks like a (fringe) contender for a division crown, but football teams are not fully healthy.

OOC
9/3 WMU Must win
9/10 Notre Dame Tossup
9/17 EMU Must win
9/24 SDSU Lean to win
Conference
10/1 Minnesota Must win
10/8 @ NU Lean to win
10/15 @ MSU Lean to loss
10/29 Purdue Must win
11/05 @ Iowa Tossup
11/12 @ Illinois Lean to win
11/19 Nebraska Probable loss
11/26 Akron State Lean to loss
Absent:

Wisconsin, Penn State, Indiana

I add it up and I come up with eight wins and change. Assume one irreplaceable player is annihilated and that comes back down to an even 8-4. Unlike last year, when I predicted 7-5 but thought 6-6 was more likely than 8-4, I think Michigan is more likely to surprise to the positive until such time as we have another Woolfolk ankle explosion pity party.

Some commenters have suggested that the exactingly specific predictions in the previous posts today suggest I'd be predicting something better than 8-4, but I think turnovers, while getting much better, will still be in the red. Though the special teams issues can't be as bad they will still be a problem that could kill Michigan in a close game.

Also, 50th in advanced metrics is still bleh territory since they correct for schedule strength. For example, that's worse than Purdue and Penn State last year; the Nittany Lions gave up at least 21 points in every Big Ten game and Purdue got bombed for at least 34 five times in conference.

--------------

*["Michael Shaw is expected to start but power back Fitzgerald Toussaint will also see time. If he had any newshole anymore we would tell you about Vincent Smith, but oh well."

/end running back "scouting".]