The Michigan Difference
We spend an awful lot of time griping about the defense here and what it is doing for opposing offense's stats. I thought I would instead look at what our offense is doing to our opponents' defensive stats. So brace yourself for lots of charts.
Here is how our opponents' defenses have looked thus far:
Opponent | Games | Yards | Yds/gm |
NCAA Rank |
Connecticut | 8 | 2914 | 364.25 | 60 |
Notre Dame | 9 | 3538 | 393.11 | 77 |
Bowling Green | 9 | 3828 | 425.33 | 98 |
Indiana | 9 | 3518 | 390.89 | 76 |
MSU | 10 | 3279 | 327.90 | 28 |
Iowa | 9 | 2651 | 294.56 | 8 |
Penn State | 9 | 3114 | 346.00 | 45 |
Illinois | 9 | 3087 | 343.00 | 43 |
Well, how did Michigan's offense do compared to these teams' season averages?
Opponent | Avg Yds/gm | Michigan |
M % of Avg |
M % w/o M |
Connecticut | 364.25 | 473 | 130% | 136% |
Notre Dame | 393.11 | 532 | 135% | 142% |
Bowling Green | 425.33 | 721 | 170% | 186% |
Indiana | 390.89 | 574 | 147% | 156% |
MSU | 327.90 | 377 | 115% | 117% |
Iowa | 294.56 | 522 | 177% | 196% |
Penn State | 346.00 | 423 | 122% | 126% |
Illinois | 343.00 | 676 | 197% | 224% |
So our offense has gained more yards than what every one of our opponents' defenses yield per game.
What would their statistics look like if they hadn't played us? I went and calculated what each team's Total Defense season average would be and their resulting ranking with the FBS statistics:
Opponent |
Average Yds/game |
Rank |
Without M Yds/game |
Rank |
Difference |
Connecticut | 364.25 | 60 | 348.71 | 50 | -10 |
Notre Dame | 393.11 | 77 | 375.75 | 70 | -7 |
Bowling Green | 425.33 | 98 | 388.38 | 75 | -23 |
Indiana | 390.89 | 76 | 368.00 | 62 | -14 |
MSU | 327.90 | 28 | 322.44 | 27 | -1 |
Iowa | 294.56 | 8 | 266.13 | 5 | -3 |
Penn State | 346.00 | 45 | 336.38 | 34 | -11 |
Illinois | 343.00 | 43 | 301.38 | 14 | -29 |
Average difference: -12.25 places.
So there we have the Michigan Difference. Playing Michigan so far this year has cost our opponents on average of 12 places in their NCAA Total Defense statistic. I suspect an analysis of rushing offense, passing offense, or scoring offense would yield silmilar results.
The Leaders, and Best. Let's Go Blue!
Edit: As per suggestions in comments, added % over average w/o Michigan, above.
Added another chart for the defense:
Opponent |
Gms |
TO |
TO/gm |
vs. M |
TO/gm - M |
Rank w/ M |
Rank w/o M |
Difference |
Connecticut | 8 | 2708 | 338.50 | 343 | 337.86 | 86 | 88 | +2 |
Notre Dame | 9 | 3618 | 402.00 | 535 | 385.38 | 49 | 59 | +10 |
Bowling Green | 9 | 2663 | 295.89 | 283 | 297.50 | 111 | 110 | -1 |
Indiana | 9 | 3599 | 338.50 | 568 | 378.88 | 50 | 64 | +14 |
MSU | 10 | 4168 | 416.80 | 536 | 403.56 | 36 | 48 | +12 |
Iowa | 9 | 3688 | 409.78 | 383 | 413.13 | 42 | 41 | -1 |
Penn State | 9 | 3325 | 369.44 | 435 | 361.25 | 68 | 72 | +4 |
Illinois | 9 | 3261 | 362.33 | 561 | 337.5 | 71 | 88 | +17 |
Average difference: +7.13 places
Overall conclusion: Our offense is doing more damage to our opponents' defensive stats than our defense is helping our opponents' offensive stats. And WTF was up with the MSU game? We whiffed on both sides of the ball on that one.
November 8th, 2010 at 1:06 PM ^
...how our defensive stats would improve if you took away all of our opponents yards. I think we'd look a little better than our current stats.
November 8th, 2010 at 1:10 PM ^
One thing I picked up from that is the spread between NCAA defenses.
For example, there's only a handful of elite defenses. Look at Iowa and Illinois lines. By shaving 30 YPG off Illinois, they improve 29 spots. Do the same to Iowa, and they improve only 3 spots. It's much more crowded in the middle of the pack. The difference between being top 5 in anything in the NCAA (like, I don't know, say Michigan in the top 5 in offensive yardage) is a huge advantage.
By the way, I would expect the opposite end of the spectrum to also hold true (cough Michigan defense). I would be interested to see how much opposing offenses have improved by virtue of having played us.
Very interesting, worthwhile analysis.
November 8th, 2010 at 2:57 PM ^
You really need to take the games versus Michigan out of the opponents average. Then compare the defensive numbers vs other opponents (as an average) vs what they did vs Blue. Your numbers would be even more convincing and more skewed toward the Michigan offensive dominance.
November 8th, 2010 at 3:09 PM ^
Using Iowa as the example:
8 games (not including Michigan): 2129 yds total
266 yds/gm avg
522 yds vs Blue
196% M % of Avg (using your formula)
This would follow for all of the games since Michigan piled up more yards in every game than the opponents average, generally significantly more.
November 8th, 2010 at 3:24 PM ^
stat is that Illinois would normall be 14th in the nation defensively, but now they're 49th. WOAH.
November 8th, 2010 at 5:33 PM ^
I guess I am just glad we don't play ourselves. Although it will probably help with the "Average drop" numbers purely because there aren't that many teams below us to begin with!
November 8th, 2010 at 6:01 PM ^
PSU is better defensively than I thought. I didn't know their YPG was right around Illinois's. UConn looks suprisingly competent defensively as well.
BTW, this should probably be a diary.
November 8th, 2010 at 6:23 PM ^
I initially kind of threw this together on the fly without a lot of planning, then it just kind of grew. I'll do an updated version after Purdue and put it there next week.