frank beamer #1
Meta: New interim column name is interim. Rhymes with "talkin' points" if you have a heavy Midwest accent. Hakn means to nag in Yiddish, literally to bang on […a pot or teakettle]. The reference.
Every touch is a little bit of magic. [Fuller]
Early last month Brian forwarded me a reader question about the relative experience of Michigan's players, and asked for a lot of research:
What has been the average age and game experience of each of the teams’ skill groups over the course of the season for each of Hoke’s years coaching here?
I’d love to see a table or graph that showed age/game experience by skill group by year of tenure for all the skill groups. Just data.
Everyone says – players aren’t developing. I’m not sure whether it’s true or a function of getting better but younger less experienced guys on the field.
My impression Defense is improving – and that’s where Hoke started recruiting (if memory serves) – those are some of his third year guys now (still juniors and RS Sophs) – getting better all the time. Offense – a year behind defense from age/experience. Mostly Sophs and RS Fresh. If that pattern is right and holds, a defense of 4th and 3rd year guys next year and an offense of 3rd and second year guys should continue to improve the product. No?
Off the cuff, we were plotting out age progression of Hoke's recruiting classes back in 2012 (when most of the 2013 class was signed) and concluding that 2015 was the probable germination point. I think a big part of why Hoke was let go was Michigan doesn't at all seem on track for that to happen. As Hackett mentioned in his press conference, the 2015 team should be one of the most experienced we've fielded in memory across the board (provided there's no mass exodus, which is hardly a guarantee).
Yay for Good News! How Good's Our GNews?
To get a real answer I really think we'd need other teams to compare it with, and that's way too much work. Also not all positions are created equal and relative experience does not say how quality the experienced players are: the 2003 and 2005 teams were nearly identical, but the 2003 was one of the best under Lloyd while the latter we thought of at the time as painful. Deciding which positions mature at what rate and have which effect of outcome is beyond the scope of this study. But I found two ways to approximate an answer:
1) Long ago I started keeping a spreadsheet of players, going back to the mid-'90s, with what years they were on the roster, when they left, and why. With some updating that was able to produce a list of how many scholarship players Michigan had available each year back to '97, broken up by year-in-program and eligibility and whatnot. By that count Michigan has the oldest team in 2015 in the post-championship era, with 85 accumulated years (average at UM for 1997-2014 is 68) since high school on offense and 83 (average is 61) on defense.
2) I scoured the Bentley team history pages (the links at the right on that page), for how many starts each player had. This turned out to be quite the rabbit hole, hence why it took me so long to produce a response. After fixing a bazillion duplicates and spelling errors and whatnots (like for example they have the Gordons mixed up), I had a list of starts by season of every Michigan player going back to 1994, which I've put on Google Docs for your perusal.
There's some other good tabs at that link if you like exploration.
[Money chart and more after the jump]
"Whatever you need to make you feel, like you've been the one behind the wheel, the sunrise is just over that hill."
—Cursive, The Gentleman Caller, The Ugly Organ
How about some good news regarding Michigan's football team? One runaway success you can attribute to this coaching staff is they've managed to hold onto their players, especially the ones they recruited. Better news: the thing about a lot of the teams that finished in the Top 10 in 2013 were they had lots of upperclassmen starters. Experience is still a big deal, and the only way to get that is to go a lot of years in a row without losing half your roster. Better better news: Michigan is (likely) going to be one of those teams in the not-too-distant future. Let's go right to the table:
% of PLAYERS REMAINING AFTER X YEARS FROM RECRUITING CLASSES 1993-'13
|Class||Recruited by||Class Size||1 year||2 years||3 years||P.O.E.*||Usage**|
|Average for 1993-2010||21||92%||85%||74%||58%||77%|
*(Played out eligibility, i.e. nonredshirted Sr's who played 4 years + guys who played 5)
**(Eligible seasons the class netted divided by 4 x class size)
There will be attrition from Hoke's classes as the position battles shape out, but for awhile there Michigan was regularly coming into a recruiting class's redshirt sophomore season with a third of that class already departed. As of now the only guy from that awesome 2012 haul not on the roster is Kaleb Ringer. You have to go back to the class of 2000*, which didn't qualify Reggie Benton, to find a class to make it this far as intact. It was so long ago that a guy from that class is now one of Michigan's coaches.
Plot the retention of the 2012 class to this point with the state of the classes before it coming into their 3rd season. It's stunning:
Years after coaching changes seem to witness an exodus spike, followed by a return to normal, which is to be expected. The last few years though…
*Even better was 1998. Henson (Yankees), Terrell (early NFL), Fargas (transfer to USC) and fullback Dave Armstrong (unrenewed 5th) were the only losses, and that was just a year of eligibility from each of them. Considering they were recruited after the championship year that's astounding.
[Jump for lots more charty charts.]
Selfie: Go save Christmas again.
Kids! Gandalf the Maize's follow-up on which factors seem to be the most predictive of offensive line play compared the r-squareds of 19 things, from o-line experience as a whole, to their recruiting stars, to the left tackle individually, depth, QB talent, how much the team goes long, etc., and found the ONLY factor with relatively high significance is…
Interior. Offensive. Line. Experience!
Let's get a DotW to the wizard, and tell Funk to put some years on his charges, RIGHT NOW.
Ron Utah followed up by showing the relative age of Michigan's whole roster versus teams of significance.
Part II by dnak438 on Michigan's offensive regression this week went back and added 2011-12 to the study of YPP versus opponents, tracking it by dividing Michigan's yards per play each game versus the average that team gave up. The results are charts that really show the history of Al Borges's various offensive strategies:
Here's the progression:
[after the jump]
Hypothetical activities by a 67-year-old Jimi Hendrix if he were still alive | Lies, Deceit, & Stuff
I spent most of today trying to play with this diary by airvipermb, which spent some time yesterday on the front page before I knocked it back. For those who don't remember the Jimi-headed versions from last year, the OP did a tremendous job of going through Big Ten rosters and putting down how many upperclassmen each team was projected to play as starters and on the two-deep. What this doesn't do is provide any predictive information.
For that reason this isn't front page material. Not yet at least. But I'd like to help it get there.
First, upperclassman starters in 2011, in table format. I changed it to percent; starter % of upperclassmen is out of 11, two-deep is % of upperclassmen out 22. Because I'm pretty sure this is how airvipermb did it (likely reasoning: too hard to find data otherwise) a redshirt sophomore is an underclassman while a true junior is an upperclassman.* Your most experienced Big Ten two-deeps next year (UPDATE: added deltas):
|Team||'11 Starters||Delta||'11 Two-Deep||Delta|
The author was optimistic but this says Michigan's defense is still the youngest in the conference excepting younger siblings. Does that matter? The O.P. suggests it does; the data say NSFMF.
Here's how this all looked last year:
|Team||2010 DFEI %ile||2010 Yds/G||2010 Starter%||2010 Two-Deep%|
I showed the Yards per game because that's the sorting metric the author used. I'm gonna talk DFEI from here on since I'm an advanced stats fan but if you're not such, mentally upgrade Penn State and Wisconsin for hard-nosed grind-it-out game planning, and downgrade Illinois, Indiana and Purdue for "not getting it." It doesn't matter because these numbers are all over the place:
HOWEVA, if you do the same thing on pure recruiting the results are also bouncy. Here's 4- and 5-stars recruited by Big Ten teams from '06 to '08, out of 22 spots available:
Actually this is pretty un-bouncy except Nebraska and Iowa outperformed by a lot and Penn State and Michigan (infinite ARGH!!!) underperformed. Iowa is our super-duper experienced two-deep so maybe that explains them and Michigan's young roster explains Michigan. But then what's Penn State's problem? This study doesn't say. Future study: I would love it if we could get a spreadsheet of all of the Big Ten players on the 1- and 2-deep.
* Let's use Nebraska 2010 as an example for how this can throw us off. You appropriately call them a great defense and rank them second to Ohio State, which I think we can concur on. Here's Nebraska's defense as of this date last year. I count six upperclassman starters there. However Cam Merideth, Baker Steinkuhler, Sean Fisher, Will Compton and P.J. Smith are all from the Class of 2008, i.e. they're redshirt sophomores. Also from that class: Alfonzo Dennard, counted as a junior. See the problem? You've got returning starters three years removed from high school counted the same as Courtney Avery (a true freshman who was a QB in H.S.) last year, despite there being vast difference between their respective expectations of experience-based contribution.
Still Raining, Still Dreaming
Jimi eat galaxy.
Until a few hours ago, it looked like the SEC was ready to swallow up one of the last quasi-worthwhile bits of the Big XII that doesn't have its own channel. This of course sparked another round of speculating how many galaxies the Big Ten must own to keep up, from oakapple and justingoblue. This being the Big Ten, it's not who would come (except Notre Dame) so much as who can fit the academic criteria without being a.) Ivy League, or b.) Athletically challenged.
Justin took it from an academic expenditure perspective, which is an interesting way to get around having to use academic rankings like U.S. News & World Report's, and avoids the weirdness from endowments-based ranking. The candidates are Duke, UNC, Virginia, V-Tech, Pitt, Rutgers, Cuse, Mizzou and Notre Dame, though all but Duke, UNC and Pitt would be near-bottom in the conference at research spending.
For the wargames route, consult oakapple, who took a more pragmatic, dominos (NTD's) effect approach. After six previous dominos it's the Big Ten's turn and…
VII. What Does the Big Ten Do?
The short answer, at least for now, is: probably nothing. The Big Ten is already in a position of strength. It has no particular need to expand. There are only two institutions that could improve the Big Ten’s current product: Texas and Notre Dame. The Irish have chosen repeatedly to remain independent, and for reasons noted above, the Longhorns are more likely to choose the Pac-12.
Galactic plans are on hold until such time as the Pac 487 annexes China and the Big East is trying to teach the French to play football, or Zoltan demands such, whichever happens first.
Scoreboard! Thanks to M-Wolverine at Fan Day, and M-Wolverine's camera.
EGD had some thoughts for a Top 10 (which means 11 not 12) other ideas for Jerel Worthy tattoos. I'm surprised nobody suggested just getting a chip drawn on his shoulder. I'm also surprised it wasn't the 2008 scoreboard, because that's exactly what my MSU friends/family still troll me with (the reminder of 2008 is what stings). Anyway these are good but I want pics! Those of you with Photoshop/GiMP, or who are Samara Pearlstein, get on this!
And THE_KNOWLEDGE is apparently using up the last miles on his current time machine lease to predict the standout of the 2012 class will be… well I can't ruin it.
Let's take some small sample sizes and extrapolate wildly. It will be fun. Here's Bill Connolly breaking down expected improvement from teams that return varying numbers of defensive starters:
So Cincinnati returns 11 defensive starters. That's probably a good thing, right? But how good? And how much can a bad defense improve in one offseason just because of experience? Let's take a look, shall we?
Average Change In Def. F/+, Last Three Years Starters
N Avg Chg in
1 1 -12.4% 2 4 -10.9% 3 10 -8.4% 4 32 -2.1% 5 53 -1.1% 6 69 -0.5% 7 85 1.1% 8 56 1.5% 9 37 4.2% 10 9 6.0% 11 3 5.4%
So basically, if you return between five and eight starters, you are likely not going to change much, but three or fewer is a problem, and nine or more is a good thing.
F/+ is Connolly's advanced metric; it's play-based instead of drive-based like FEI. Don't be fooled by the % symbol—the metric is percentage based and from context it's clear the difference is meant to be added to the score, not multiplied. Since the best defenses are around +17% and the worst around –13%, 6% is about a fifth of the entire scale.
Michigan is, unsurprisingly, right at the bottom of that scale at 115th. They were 12% worse than an average defense down-to-down. The good news is they return 9-ish starters, losing Greg Banks, James Rogers and Jonas Mouton while reacquiring Troy Woolfolk. (They also lose Ray Vinopal and Obi Ezeh, but Ezeh had been replaced and Michigan should get JT Floyd back so let's call it a wash.)
The numbers are thin at both ends of the spectrum but, hey, extrapolating wildly from small sample sizes. Doing so says Michigan's defense will storm forward from 115th nationally to…
I have no source for this, unfortunately.
But wait! Our sample sizes are not small enough and our extrapolation is not making out with other nubile young extrapolations in front of a television camera. Bill added a second factor, the previous year's defense, and finds that a defense with an F/+ under –10% that returns nine starters should expect (for a given confidence level that is not high at all) to improve by 8.6%, which would see them get to…
You might be able to argue that Mike Martin wasn't right and the team was even younger than the average team that returns nine starters and GERG is rubbing stuffed animals on the faces of other stuffed animals at a tearful tea party and for the first time in a long time they'll just run one damn defense per year and that they should expect to improve even more. You're probably setting yourself up for disappointment. Like installing the spread 'n' shred, digging out of a hole this big is a multi-year project.
Ridiculous wallpaper part 7. Via user monuMental and his ridiculous talents:
How could this possibly happen? So when people say things along the lines of "could we really have the worst defensive talent in the Big Ten" and I say "yes," no one believes me. This is usually because one sophomore four star in the starting lineup at a particular position looks like talent and two fifth-year-senior three stars do not. Here's the Iowa two-deep on defense:
There are 22 players. Five of them are underclassmen, only one of those a starter. Nine are seniors, and this is minus a senior starting linebacker who would shove a freshman out the door. Michigan's starting lineup has as many sophomores (Floyd, Kovacs, Roh) and freshmen (Gordon, Gordon) as the entire Iowa two-deep, and where Iowa has seniors backed by seniors or sophomores backed by juniors in many places Michigan has freshmen, freshmen, and more freshmen. This is why it's impossible to tell anything about Greg Robinson yet. You could take an established genius and give him this roster and the results would be, oh, I don't know… somewhat depressing:
|Pass Efficiency Defense||92||139.01|
|Tackles For Loss||66||5.83|
That's not Michigan. It's USC. USC's secondary:
- Senior CB Shareece Wright, a top 50 recruit
- Freshman CB Nickell Robey, a top 250 recruit
- Sophomore S Jawanza Starling, a top 250 recruit
- Sophomore S TJ McDonald, a top 50 recruit
This is a "talented" secondary. It is also awful because it has one upperclassman; they're trying to bolster things by moving freshman and starting WR Robert Woods to nickelback. That sounds familiar except in Michigan's case it's a guy who should be a linebacker moving from wide receiver and being forced to start instead of being Courtney Avery.
Michigan does not have near that amount of recruiting mojo, nor does it have the veteran consistency of Iowa. Yes, if Michigan is not more experienced and less awful next year it's time to focus the firey finger of blame entirely on Rich Rodriguez. Not quite yet, though.
Parachute in. While everyone was looking at that guy in the parachute he was looking at us:
AIM FOR THE YELLOW.
Alabama-rama. Some final thoughts on the Alabama game, but first Dave Brandon:
"I just thought it was a terrific opportunity for our team, our coaches and our fans," Brandon said. "But we got the numbers right, we worked really hard to make sure there was plentiful availability of tickets, because we wanted to get that right for our fans. We hope to think of this as a preseason bowl trip where we can bring lots of people and really make it a special Labor Day weekend."
Booting the Notre Dame game is not an option, so Michigan will go on the road to face Ohio State, Notre Dame, Nebraska, and Alabama (sort of) in the same season. Have fun, senior Denard. Anyway, thoughts:
- I am almost certain the reason introducing a middleman is preferable to a home-and-home are the messed up TV contracts. If Michigan plays a home and home with Alabama they split the extra TV money with the rest of the conference. It sounds like by doing this neutral site thing they are getting the financial windfall all to themselves. If you dislike this trend—and as a guy who would rather travel to Tuscaloosa than Dallas, I do—the only solution is to let teams keep all the profits from their nonconference games to themselves.
- As to why it's in an irrelevant place like Dallas: when ND started its "barnstorming" games it quickly discovered it couldn't play anyone in a relevant location because TV contracts prohibit anyone from playing a neutral site game in their conference's geographical footprint unless that game is going to be on the appropriate network. The result was ND-Washington State in Texas.
- This will be the biggest nonconference game played against anyone other than Notre Dame in…a very long time. Maybe the 1996 Colorado game? Michigan accidentally played a 13-0 Utah team in 2008 but in terms of pregame hype that pales in comparison. Washington was #9, IIRC, so Alabama will outstrip that.
Etc.: Hockey has a critical road test against UNH this weekend. People are still projecting us for New Year's Day. The Daily takes a look at college amateurism and whether it can or should go away. BWS picture pages the Webb TD.