Error Bars

Something I find irritating about all college football polls, even the vastly superior Blogpoll, is they don't give voters room to express uncertainty. A search for the phrase "I don't know, man" on MGoBlog's main page yields 70 hits; surprisingly, only six of them are from Blogpoll posts describing the hebdomadal^* struggle to figure out who should be ranked 20^th and who should be ranked 25^th.

You could argue with a straight face that, based on results, Michigan could be the third best team in the country. After all, its only losses are to an undefeated Alabama and and undefeated Notre Dame. Michigan could be better than undefeateds like Kansas State, Oregon State, Louisville, or Rutgers, and you'd probably bet on Michigan to beat Rutgers right now.

Strict resumé balloting errs in the opposite direction and ranks each team according to how bad they could possibly be. Who has Michigan beaten? Purdue? Air Force? Illinois!? UMass!?? With a set of wins like that, you can reasonably argue that U-M is the 50^th or 60^th best team of the country. Especially if you argue that Notre Dame is a mirage that didn't actually beat Stanford. Resumé balloting is more honest than voting with your gut, but it still hides the fact that, six weeks into the season, where many teams should rank is still a mystery. Is Michigan third? Unlikely. Is Michigan fiftieth? Also unlikely. But we just don't know, man.

GIF of Stanford's overtime touchdown credit SBNation, obviously.

One thing we do know is the defense put in an amazing performance against Illinois.  They were held to 3.3 yards per carry (with a standard deviation of 5.1 yards). These two stats indicate that not only did the D hold the Illini in check, but that they kept them from pulling off many big runs; in fact, Illinois only had one run of over ten yards all day, the Nathan Scheelhaase dash that knocked him out of the game. If you calculate the standard error about the mean, it's 0.14 yards, suggested that if U-M and Illinois face of again and again, Michigan would hold them to under 3.5 YPC again and again and again. That's consistency. That's dominance.

On the one hand, we have a vast amount of uncertainty about how good this team is overall. On the other hand, we have a lot of evidence, especially over the last three weeks, to suggest that Greg Mattison has worked his magic once again and produced another unparalleled defense. There are lots of reasons to be confident about the second half of the season, but there are also lots of reasons to be unsure. If we can be leave this game 100% sure of anything, it's that Danielle is the luckiest girl on earth, even luckier than Tommy Rees.

Photo by Ace.

With Michigan State coming up, it's not the time for smack talk yet. Let's let Jake Ryan continue his hebdomadal smacking next Saturday and then we'll talk.

^*Hebdomadal means "occuring every seven days." Similarly, "nychthemeral" mean "occurring every twenty-four hours." These words have subtly different meanings than "weekly" and "daily." English is weird. Thanks to @sgtwolverine for retweeting the word of the day.

Everyone else has a solution for the BCS, so why can't we?

NCAA institutions are a varied lot: both private and public, both large and small, both teaching-oriented and research-oriented, both colleges and universities. However, without doing the research, there is at least one thing that I am certain every single one of them has: a professor of statistics.

The Michigan Statistics Department has an awesome logo. That is all.

In this post we will discuss two of the many problems with the BCS. The first is that there are humans involved in the formula who are swayed by things such as "brand name," "reputation," and "whatever bullshit Gary Danielson spews during the fourth quarter of the SEC Championship game." Vishnu Parasuraman at Grantland proposes replacing the polls with a committee. The only effect that would have is that the biases of "brand name," "reputation," and "Gary Danielson" would be hidden in a board room in Indianapolis instead of being out in the open for us to mock. Lloyd Carr is the sort of person who'd be on the selection committee, and he's already a Harris Poll voter. You don't see the NCAA putting Ken Pomeroy on the March Madness committee.

This is not to say that computers are unbiased; the second problem with the BCS is that the computer rankings are hobbled and slanted by the silly biases of Jeff Sagarin, Peter Wolfe, Wesley Colley, Kenneth Massey, Richard Billingsley, and some dudes at the Seattle Times (a.k.a. Seattle's fourth best newspaper, behind the Stranger, the Weekly, and the Sinner). Only six. That's ridiculous.

My solution to these problems: no more committees, no more polls. Only computers. But not just six computers, that's not enough! In order to come up with a reasonable BCS computer ranking formula, we need hundreds of rankings. Every school in the NCAA should be invited to submit a team ranking system before the start of the season, and the official football rankings should be a combination of hundreds of rankings created by the smartest statistics students at all of the NCAA's member institutions.

Here's the plan. After the jump.

Damn Lies: Connecticut

[Author's note: I'm going to try this year, in the vein of statistical analysis, to try and find things that struck my from the Michigan box score after each game. However, because I do not have as much time as I used to to write meaningful things, this analysis will be limited to 120 seconds of looking at the box score and attempting to find something. Sometimes it will build off something that struck me while watching the game, other times it will just be something that stands out. My impressions could be wholly and completely incorrect, my conclusions could be completely wrong, but this is why I am calling this feature "Damn Lies", because well, we're looking at statistics.]

1). Michigan was 15 for 19 on third down, and 1 for 2 on fourth down yesterday.

Michigan needed an average of six yards yesterday on those 19 third downs. Their average game on those third downs was 7.3 yards. Of the 14 conversions, Denard Robinson ran for 7 of them himself, and passed for six of them. Michael Shaw's two yard run in the first quarter was the only third down conversion for which Denard was not partially or wholly responsible. More amazingly, of the four failed third down conversions, none were the result of an incomplete pass. Which is why Michigan likely only needed to punt once.

2). Michigan had drives of 96 yards (5:57), 77 yards (2:35), 75 yards (8:05), 96 yards (4:22) and 51 yards (7:59).

If you're convinced that Rich Rodriguez's up-tempo style does not protect leads (or by extension, the defense), look at those drive totals. While Michigan did have some short drives, Michigan ate up the clock with a long early TD drive, a long TD drive to start the third quarter, and a long drive late that came up short, but left Connecticut without much time to do anything.

3). Denard Robinson targeted 8 different players yesterday, but a running back scored the only receiving touchdown of the day.

I thought that Stonum and Roundtree came up on the short end of the stick yesterday, and I was partially right, Roundtree was targeted once, the first play of the second half, and lost one yard on his catch. Stonum actually caught five balls, so my bad on that one.

[Author's addition (Monday): So, missed this initially, but Roundtree was hurt during the game, which would further explain this. Sorry Shaft!]

4). The officials called a total of three penalties in the game. Two false starts on the Connecticut offense, one personal foul on the Michigan offense.

I am not saying this is the most penalty free game I have ever seen, but I would be hard pressed to remember a game where there was a gap between penalties from the second drive of the game to the final drive of the game. Whether this was an effect of coaching, officiating, or a combination thereof remains to be seen.

5). Denard Robinson was 19 of 22 passing yesterday, after being 14 of 31 passing...for all of 2009.

Robinson's Quarterback Rating was 188% higher than his average QB rating for all of last season. More impressively, his 197 yards are 56.12% of his rushing total from last season. I have no idea what this means, it's just what it is.