Postmortem: How Much Did They Actually Improve?
After Michigan's 4-0 start the offense skidded downwards into mediocrity and inconsistency, gradually erasing the early giddiness around these parts. Early in the year, I said "the question is not whether this will be Michigan's best rushing offense of the decade, but by how much," which… well… Alan Greenspan has something to say about that.
Still, for anyone who actually watched the team this year and last, the idea that things didn't improve is laughable. The favorite tool of folks who like to claim this is Michigan's scoring offense in the Big Ten. Check this pre-OSU Michigan Monday for an example. Okay, actual scoring—despite being the point of everything—is is kind of a wildly unreliable metric, especially when your quarterbacks are freshmen and very, very turnover-happy. But when a tempo-free aerial says basically the same thing when it comes to yards per attempt…
…the offense has not exactly reached juggernaut status.
Still, it did improve significantly. All of this focus on conference play ignores that last year Michigan was 1-3 outside the conference and lost to a 3-9 MAC team because it scored 10 points. Against Notre Dame the net offensive output was probably negative because of five horrendous turnovers. This year the MAC snacks were swatted away and Michigan was one of many teams to scorch TAH-NOO-TAH's blitz-mad ways. They were in the ballpark of crappy, and this represents a step forward from last year, when Michigan football was the Indiana basketball of the Big Ten: too terrible to even fit on the scatterplot. To repeat a theme of late, this is progress of a not-very-fun variety.
There is noise yet in the scatterplot, though, as it makes Northwestern's dink-mad offense look worse than Michigan's when it wasn't. Let's take away all the noise caused by varying numbers of drives, varying average drive start, and opposition offenses, and just look at how efficient Michigan was on a drive-by-drive basis.
Brian Fremeau maintains a rating that does this called the Fremeau Efficiency Index. It's similar in concept to the numbers the Mathlete has posted here over the last year or so: find the average success rate in Situation X and measure teams by how far above or below that break-even line they are. Fremeau has many numbers; we'll look at a few. OE is "offensive efficiency" and is just a measure of how much you score relative to the D-I average. It's a tempo-free stat roughly equivalent to Points Per Possession in basketball. The second, FEI, is this in the words of its author:
College football rating system based on drive-based Game Efficiency data that rewards playing well against good teams, win or lose, and punishes losing to poor teams more harshly than it rewards defeating poor teams.
I couldn't find a more specific definition of what exactly this means; IMO, this is less interesting as a performance measure than it is an attempt to make the ratings more plausible to human eyes looking for a ranking system. So the focus should be on the raw numbers.
100% Awesome Mediocrity
Raw and adjusted, your results:
Year | OE | OE Rk | Off FEI | OFEI Rk |
---|---|---|---|---|
2008 | -0.29 | 98 | -0.15 | 80 |
2009 | -0.05 | 65 | -0.01 | 61 |
(This rating excludes I-AA games, as all serious attempts to quantify college football do.)
So there you go: from one of the worst teams in D-1 to totally mediocre. Totally mediocre seems acceptable, or better, when you're graduating four starters, have a freshman quarterback, play most of the year without your best offensive lineman, don't get to use your senior tailbacks all that much, and suffer from another epic turnover plague.
The Unfortunate Flipside
The other side of the ball is sort of horrifying:
Year | DE | DE Rk | Def FEI | DFEI Rk |
---|---|---|---|---|
2008 | -0.14 | 44 | -0.13 | 45 |
2009 | 0.12 | 75 | 0.04 | 63 |
…but a little less horrifying than i expected. Michigan's defense under Scott Shafer was actually sort of good-ish, but submarined by terrible field position and a ton of drives faced because of the offense. FWIW, I don't think this reflects poorly on Robinson yet; one year doesn't tell you much of anything. Also, Michigan's two best defensive performances of the year (relative to the opposition) were against Minnesota and Northwestern, after Shafer was basically fired. (And, yes, after Michigan's disastrous attempt to switch to the 3-3-5 cost them the Purdue game.)
The Net
Improvement + implosion = ?
Your net numbers in efficiency: Michigan went from 85th last year to 62nd this year. In FEI terms, Michigan went from 71st to 56th. That is almost exactly in line with what I believe is a reasonable take on the team: it was terrible last year and significantly better this year, but the amount of progress was disappointing not just relative to expectations after 4-0 but relative to those in the preseason.
Attempting to downplay the real improvement that was obvious to anyone watching the offense in something other than the fetal position is silly, and a sign they're about to bring up the Braylon Edwards #1 jersey "controversy" as further evidence that Rodriguez should be fired.
December 2nd, 2009 at 6:25 PM ^
December 3rd, 2009 at 2:54 PM ^
December 2nd, 2009 at 8:54 PM ^
December 3rd, 2009 at 11:47 AM ^
December 2nd, 2009 at 10:45 PM ^
December 2nd, 2009 at 11:32 PM ^
December 2nd, 2009 at 11:27 PM ^
December 3rd, 2009 at 1:10 AM ^
December 3rd, 2009 at 10:46 AM ^
December 3rd, 2009 at 1:48 PM ^
December 4th, 2009 at 12:39 AM ^
Comments