Mailbag: Fumble Rates (Again), Leave Penn State Alone, Can't Happen Here(?)

On fumble recovery rates (again).

Hi Brian,

A sentence in Blue Seoul's recent Nebraska recap led me to re-visit
your claims that fumble recoveries are random. I got data from
teamrankings.com and drew a graph that you may (or may not) find
useful (see attached for 2012 opponents).

In addition, I estimated a simple linear model using fumble recovery
rate in year t to predict fumble recovery rate in year t+1. The model
also controls for the baseline recovery rate of each team by allowing
intercept shifts. Two things stand out. First, past recovery rates
explain relatively little variation in current recovery rates
(R-squared=0.13 in a model with lagged DV and team fixed effects).

Second, the relationship between recovery rates last year and recovery
rates this year is *negative* and statistically significant. If this
general pattern holds true in the case of Michigan this year, we
should expect the team to recover fumbles at a slightly lower rate
than its baseline for the 2002-2012 period (i.e. 61.57%). Obviously,
this is a ridiculously simple model, but it reinforces whatever
evidence you were basing your previous comments on.

The oft-repeated claim here that Michigan's fumble recovery rate in 2011 was a gift from the fickle gods of chance is not one that has gone down smoothly with many readers. The above lays out the numbers case. The previous year's fumble recover rate has very little effect on next year's and what relationship there is tends to take the rate back towards average.

But numbers don't seem like enough for whatever reason. I think I may have given doubters ammunition when I came back from the Glazier clinic reporting that Mattison told a room full of dudes that Michigan emphasizes turnover recovery:

Bonus: For those looking for a reason other than blind luck that Michigan recovered 80% of opponent fumbles this year, in practice all incompletions are live balls. Mattison credited this practice for getting players moving towards the ball at all times and being in position to scoop up live balls in actual play.

I thought this was not that convincing; several commenters seized upon it as a workable rationale why Michigan would be able to continue their fumble-procuring ways, Michigan's coaching does help them recover loose balls, but every other program in the country does something similar. Just because you work at it doesn't mean the results are predictable.

Blue Seoul points out RVB hustle that lead to a fumble recovery against Nebraska…

Ryan slaps the ball out. Check out how far away from the ball Van Bergen is. But he's got his head up, he's disengaged from his blocker, and he's pursuing the ball.

One funny bounce later and it's in RVB's hands. Brian keeps saying that fumble recoveries are just luck and 50-50 propositions. I would disagree and say the fumble recovery percentage is more of a function of the number of each team's players near the ball when the fumble happens. In this case, we were a little lucky because Nebraska had more guys near the ball. But if RVB isn't hustling and getting off his blocker, our chances of getting that ball go from slim to none. So yes, luck plays a part, but I don't believe it's JUST luck or that it will always regress to the mean..

…and Mattison did spend a slice of time describing his deep-seated hatred of "loafs." But so did the tweaked-out LB coach from Illinois who spoke in ALL CAPS. The tweaked-out LB coach from Illinois bellowed his distaste of loafs to three rooms, only one of which he was occupying, for about twenty minutes. The differences there are not great. They're certainly overwhelmed by the fact that these fumble recovery rates are across a sample of 15 or 20.

Here's a last stab. Look at those Alabama numbers, which stretch back to a time when they were run by incompetents. Now they're run by Nick Saban, collector of all first round draft picks and ruthless destroyer of offense. Saban's arrival has coincided with Alabama's fumble recovery rate going… nowhere, maybe getting a tiny bit worse. If coaching had a significant impact on fumble recovery rates, wouldn't Alabama be a perfect case study?*

I think that coaches think they have an impact here, but the things they do are all the same things: run to the ball and be alert. There's no technique and no strategy, so the impact they can have in this department is limited. Therefore Michigan's recovery rate last year was an anomaly and we should not expect it to repeat, or even expect it to be significantly above the 50.3% national rate.

Of course, I said the same stuff about this during the Rodriguez era and the thing stubbornly failed to turn around. If the universe is determined to make me look stupid at least this time I'm hedged properly.

*[I know one team's rate is a hopelessly small sample size but this argument is addressed to people who aren't going to take the numbers to heart.]

I'm not answering this guy's PSU question.

Brian

We're all tired of the Penn State tragedy that has been occupying the headlines since the fall. Some are talking about the conference removing the school as a member. While we've heard such an idea echo throughout the sports world, no one has actually talked about the criteria for the Big Ten actually removing a member. We've also been inundated with schools switching conferences over the past few years and the procedures for making such departures. So, by now, most of us can recite the hoops a school needs to jump through in order to swap conferences. My question is, what is the Big Ten's procedure for removing a member from its ranks?

Thanks,
Chris

I don't know if they even have one but I assume it's something along the lines of getting everyone together to vote on stuff, and then giving them a bunch of money for breach or just to prevent the inevitable lawsuit.

But… seriously? We're talking about removing a university with 40k people involved with it because of one very, very tragic event? I get that the Penn State athletic department had warped itself into a place where common human decency took a back seat to covering your ass, but the people responsible will be held accountable and the culture of the place will change radically. The unique situation PSU found itself in will not repeat. As long as they take the proper steps to assure that, let's just grimly recalibrate how we feel when we look at the Paterno statue and move on. A start: apply the same FOIA laws to Penn State that apply to public universities everywhere.

Annihilating the PSU program is pointless. Maybe I can see hobbling it for a bit, but when we start talking about booting universities from conferences or cancelling their nonconference games and forcing them to play on the road for two years or the death penalty or the NCAA fate worse than death, you lose me. While I've been pretty uncompromising in re: JoePa himself, at some point you have to think of the town, the university, and the program as things that aren't just JoePa. There's a massively damaged community out there whose only crime was to believe in Joe Paterno. They've been getting the 2009 Illinois treatment for going on a year now. At least give them the escape of a crisp fall day against Wisconsin, conflicted though it must be.

I'm all for the symbolic eviscerating of Paterno's record, and if you want to slap some other sanctions that erase the "one of two schools never to have been hit with a major violation" claim, fine. Burning down the program doesn't affect any of the perpetrators more than jail and death does.

It can't happen here?

Brian:

What nags at me about Penn State is this: How far is the "Grand Experiment" away from "This is Michigan"? Obviously, I don't mean, "Do you suppose we've ever aided and abetted a predatory pedophile?" But let's face it: We've seen "the good of the program" used to defend some behavior we wouldn't countenance in another context.

I think in this case what allowed the culture to develop was the cloistered nature of Penn State, culturally and geographically. The main reason this most likely couldn't happen in Ann Arbor is there are a lot of people in and around Ann Arbor who couldn't care less about football. "This is Michigan" on an optic-yellow T-shirt might mean Bo or Desmond or Yost or Man-Ball. But "Michigan" means JFK and the Peace Corps on the steps of the Union, the polio vaccine, and Space, Bitches, Space. I'm not saying everyone in State College loved football; I'm not saying everyone affiliated with the Michigan football program "just lugs the damn refrigerator" as quietly as they should. I am saying enough people of non-athletic consequence pass through Ann Arbor for one reason or another that a sports coach who spends too much time saying "look at me" is going to get his comeuppance before a jury is hearing 48 counts of criminal sexual conduct against one of his assistants.

No one who puts any stock into the importance of "the program" is immune from having his perspective skewed a bit, of course; the Florida A&M band was big enough that they could initiate newcomers however it wanted -- until one of them got killed. But I feel like a cosmopolitan locale -- be it a state capital or a cultural center -- provides a deterrent on any one entity getting TOO big. In State College, there was no adequate deterrent. The Sandusky scandal was likely a perfect storm for which the ingredients will never again exist: Among other things, we've likely seen the last of the 40-plus year football coaching tenures. But I think the next time we're walking toward Stadium and Main on a Saturday afternoon and we pass someone walking in the opposite direction -- toward some library or lab or city hall -- hauling some figurative fridge we'd never care to bother with, it's worth thinking gratefully, "This is Michigan."

Best,
Matt

Mostly I just posted this so people could read it. Comment follows out of obligation.

Michael Weinreb described the strangeness of this event after growing up in State College beautifully for Grantland*. His piece tends to confirm your hypothesis. I believe that if we were to transport Bo into an anachronistic Whole Foods at the height of his popularity, at least half the people in there would wonder who the jerk making fun of the meat was. Any sports fan who spends a lot of time in Ann Arbor has had the sort of mutually jaw-dropping conversation with someone from the opposite pole of sports/art.

Meanwhile, the university itself is sometimes kind of sort of at war with the athletic department in a way common across the country. Rodriguez fought with the admissions department, sure, but so did Lloyd Carr. Lloyd Carr! And he lost from time to time. The local media is all up in Michigan's business. Carr (Lloyd Carr!) had to deal with the local paper bombing Michigan's practice of giving athletes easy classes, soon followed by the Free Press stretching the most minor of major violations into a leathery program condemnation. Unlike Penn State, Michigan is subject to open records requests. So, yeah, It can't happen here.

But it is an extreme outlier. It's easy to feel good about not having that be possible. Maybe we should take this as an opportunity to learn from what went on at Penn State and adapt to prevent things in the 1-5%-as-horrific range from happening here. Michigan complies with FOIA requests but charges outrageous prices to do so, blocking certain investigations. When the Daily asked for 400k employee purchasing card records they were quoted a price of "unspecified thousands of dollars"—the same records cost $200 from MSU, $181 from Iowa, and were free from Illinois and Ohio State. That is a pale shadow of the PSU secrecy that should be dispelled.

Other than that concrete suggestion, I'm not sure what actions Michigan could take, or even what the problem might be given that Michigan's coach has been around one year and probably won't make it past 15 before retiring.

*[Yes, he was sour about Michigan's team at some point over the last year. Let's get over it.]