I did not make this headline up
No this isn't a "3-4"; well it is a 3-man front with the nose over the center,
but not a 3-4, let me explain.
Various people reporting back from practice have noticed three (or fewer) guys with their hands down, and said "oh they're going 3-4." Soon enough people pick up on this, figure a new DC means a new defense, and whiteboards across the state get sales points erased in favor of X's and O's with arrows diving between X's.
Mattison was asked about the 3-4 look in his breakout Q&A and had this to say:
"We ran that last year. What we're doing on defense is trying to see what scheme fits the players."
I know we've hashed the 4-3 under a bunch on this site but I could never walk by a boardroom with football drawings on it (this is probably why I lost that job) so here we go again.
Everyone Runs Everything. Calling a defense a "3-4" or "4-3" or any one thing at this level is not ever going to be accurate, because defenses change up gaps and show different looks so offensive linemen won't know exactly who's got what gap every time. In the process of showing that you might be running 3-4, sometimes you actually have guys two-gapping. If you have a really special player you might do that even more. But general rule is everyone runs everything, and the best you can do to describe any single defense is what their base is. Everything else will stem from the base.
Alignment vs. Philosophy
Alignment is where you put your players before the snap; philosophy is what they're being tasked with. The two-gapping philosophy has become synonymous with the 3-4 alignment, and the one-gapping philosophy is thus tied to the 4-3. The major difference between these philosophies is understood better by using the "gap" terms.
…means you have one to three defensive linemen responsible for controlling a blocker instead of a gap between those blockers. His job is to get into that guy and be in a position to tackle if the run goes to either side. He's not left out to dry; the two-gap philosophy gives you a free hitter from the second level who watches the play then reacts. Think of it as man defense for run fits, i.e. the defenders all have a certain offensive player they're responsible for beating, the free hitter's being the ballcarrier.
You can do this if you have a super large nose tackle, the classic example being Wilfork on the Patriots. You can look at the above diagram and see the downside of two-gapping: if that left guard releases and the center holds up, you have a blocker eating your free hitter downfield. Two-gapping is a luxury you can have if the guy you have two-gapping is able to do it effectively.
A two-gapping defensive lineman needs to get that control established early, so lining up directly over the guy he's going to control is rather important. So a base 3-4, two-gapping defense will line up almost always with the nose tackle directly over the center, thus threatening the two-gap assignment, and still in position to change it up.
Remember that these are not set-in-stone gap assignments, just common ones. You change these up from play to play.
This is your basic "everyone has a gap" defense and the philosophy behind both the 4-3 even/over and 4-3 under. It is zone defense of run fits. Coaches who use this as their base, however, call it the more "aggressive," because football coaches describe the way they eat an ice cream cone "aggressive," yes, but also because the linebackers and linemen aren't diagnosing anything before taking their first steps of the play.
Instructively, one of the terms used often for the 4-3 over is the "Miami 4-3" because Jimmy Johnson used it to great effect when he had access to a lot of very athletic yet not very coachable defensive players. The concept was fast guys hitting their gaps and penetrating upfield to cause disruption.
You'll note that the 4-3 under picture is nearly identical to the 3-4, one-gapping example image above. That's pretty much what a 4-3 under is: a nose tackle and two large DL who set up in a position to take on double-teams so the LBs can attack their gaps. Philosophically everyone has one gap, but guys start taking on different roles: the NT is shaded to strongside so he'll need to be a bit more of a plugger; the SDE is further inside and has to be able to take on doubles like a DT; the other DT is further out and can be a little more end-ish; the WDE has a one-on-one battle with the OT, and can be a little more linebackerish; and the SAM has to be part-DE to compensate for the fact that he's the edge defender instead of the SDE.
Because it looks so much like the 3-4 defense it can threated to do 3-4 things. That gets us to the point of what Durkin does with his defenses, and what Michigan is expected to do this year: threaten two-gapping as a changeup.
Here's a play from Florida-FSU last year and you'll recognize the alignment as 3-4 (or would be 3-4 except weakside OLB is pulled for a nickel):
Nose tackle is right over the center. Ends are right over the tackles. You are thinking all sorts of Wilforky things. But this is still a one-gapping defense and you'll see why off the snap:
Right there is a good shot of the roles of 4-3 under defenders. The NT and 3-tech have double-teams; as long as they don't give up ground and stay engaged with those dudes they're keeping the LBs clean and keeping their holes closed. The five-tech has a guard (he'd be doubled if the run was going the other way), and the two MLB-types (the WLB and MLB) can flow into gaps (the MLB is blitzing his). The SAM has his edge. The difference here is the WDE has been pulled for a nickel, a dramatization of the fact that a 4-3 under WDE is often in coverage.
Here's how it's drawn up:
(orange arrow means player's in zone coverage and watching his gap; black=rushing).
On that play the nickelback (5'10"/206 junior Brian Poole, Florida's hybrid space player) came down inside and got blocked by the slot receiver, giving up the edge and leading to a big run since the free safety was deep in coverage. If not for that, Poole should have been in perfect position to hold this down.
You Sure Durkin is a 4-3 Guy?
Durkin likes to accomplish the base thing from a gazillion different looks; I predict defensive UFRs with regular opening shots of Brian trying to name things. Here are the formations for every play leading up to that one:
What defense is this? It's a philosophically 4-3 (one-gapping) multiple-front thing that likes to have speed on the edges, either from standup WDE/SAMs or putting hybrid space players in there with edge responsibility. IE what Michigan's been since 2011.
What About 3-3-5?
The 3-3-5 stack also uses two-gapping as a base threat, though neither Rocky Long nor Jeff Casteel have often had access to a nose tackle capable of doing it consistently. The point of the 3-3-5 is it gives up starting position for the threat of attacking from anywhere.
Remember how GERG was terrible at this? It's because he was a 4-3 coach teaching 4-3 philosophy, and that took away the unpredictability of this defense.
I realize Strobel got one. Find a better photo then, pickers of nits.
This has to be talked about. Hoke left a roster that was in relatively good shape considering all the highly rated players who had to stick through some awful program degradation. He signed good classes, and those classes have by and large stuck around and fulfilled their academic duties. But an inordinate amount of them inexplicably didn't redshirt, and because of this there are some holes on the horizon.
I'm sure there are explanations in many of these cases that we are not party to. It's only the sheer volume of head-scratching non-redshirts under Hoke that gives us reason to call all of them into question. Like how I'm sure there are legit medical hardship waivers that occur at Alabama but [graph].
Some guys the coaches were forced to play early, and there's no need to discuss them beyond a mention as such, e.g. Jabrill Peppers. Mason Cole outcompeted a pile of guys to start at left tackle last season. That sort of thing gets a full pass. Beyond that, I've broken each Hoke class into categories of increasing argh:
- WTF. Wasting redshirts on special teams and dime back when last year's dime back is on the bench.
- Pick ONE. Needed bodies at this position, but not all the bodies. Battles for 2nd on the depth chart should be resolved in time for the ultimate loser to have a 5th year as consolation.
- Need the dudes (and other things I don't blame on the coaches). Immediate starters or guys who played because Michigan sorely needed his body and his pulse at that position.
Names that should have redshirted are in red.
Class of 2011
Did you really need both, 2011? [Upchurch]
Hoke arrived to an offensive machine with two years of eligibility remaining, and a nightmare defense of guys who couldn't displace recent departures like Jonas Mouton, Ray Vinopal, Adam Patterson, Greg Banks, and James Rogers. The immediate need was obvious and Hoke rightfully set about recruiting freshmen who could fill those roles. So I'll give him a pass for some of it.
|Hollowell's 2011 contribution was more than scooping up a fumbled kickoff against VT, but it was also more than Ray Taylor's. [Melanie Maxwell|AnnArbor.com]|
Raymon Taylor and Delonte Hollowell. The year following the Never Forget defensive backfield, Hoke recruited five likely cornerbacks: Blake Countess, Raymon Taylor, Delonte Hollowell, Tamani Carter (redshirted, transferred before 2012), and Greg Brown (early enrollee, transferred before 2011 season). The roster still had J.T. Floyd, Courtney Avery and Terrence Talbott (left program summer before 2012 season), available. In a pinch, Troy Woolfolk could have converted back when Thomas Gordon won the free safety job. At least one, and probably two true freshmen would have to play.
It immediately became apparent that one would be Countess. So to fill out the two deep they would need to burn Taylor or Hollowell's shirt. Hollowell arrived as the quintessential Cass Tech mite corner. The guy was 164 pounds, but saw some action at dime back vs. Nebraska, and recovered the fumble at the end of the first half. Taylor had two tackles and a personal foul.
Brennen Beyer and Frank Clark. Going into the season Beyer was a SAM and Clark a WDE. The difference between those positions in Michigan's 4-3 under was not very great, particularly because when Beyer was inserted it was for a 5-2 look. The WDE's depth chart was Craig Roh and Jibreel Black; SAM was Jake Ryan and Cam Gordon. The reason I say one would have played anyway is the rush end position has a lot rotation, and Black was already the starter in the nickel formation.
There wasn't much to differentiate the two in aggregate play; Beyer was the more consistent, Clark the more explosive. The coaches chose to have them compete through the year instead of preserving one. Had they done so Beyer was the obvious choice despite Clark's higher ceiling. Beyer was smaller and Michigan had Roh to be a more solid edge defender, but only Clark to be a merchant of chaos (remember the Sugar Bowl interception). On the other hand Frank had a rough history before Glenville, and could have used an adjustment season. Either way he would have been dismissed after last year's incident.
Needed dudes etc.
Blake Countess and Desmond Morgan won starting jobs on the 2011 defensive reclamation project. They also both would lose a season to injury so we have them back yay. Thomas Rawls I'm not broken up about, though he will be a pretty good MAC back this year. RBs usually have most of the "it" they ever will as freshmen, and if they do become long-term starters the toll it takes on their bodies means they're often better off moving through their careers early. A redshirt year can make a guy a better blocker, or put some distance between a good back and his heir, or let a smaller guy fill in. Matt Wile is a special pass even though they wasted his redshirt on kickoff duties (and punting during Hagerup's first suspension). I learned recently that Wile made it clear from the start he intended to graduate in four years and do engineering things.
[Save your anger for after the jump.]
This is the second installment of a comprehensive look at quarterbacks whom Jim Harbaugh recruited and coached. Part 1 looked at his WKU recruits, his work with the Oakland Raiders, and his first head coaching job at San Diego. A few trends that came out:
- He recruits at least two QBs per class
- They tend to look like shooting guards: tall, athletic, gangly, on the border of dual-threat/pro-style. He scouts them at multiple sports.
- Their teams usually perform above or far above the usual for that program.
- He likes them smart.
We are now entering the Stanford phase, so it’s a good thing we could notice item #4 above before the sample was ruined.
We also got an idea of how Harbaugh coaches them. He likes his heady guys to memorize a million things they can think about pre-snap. When he has one of those guys, they go to the line with three plays called, and the quarterback decides which by defensive alignment. Conversely, post-snap reads are super-simplified and drilled mercilessly so that his QB barely has to think about his progressions during a play.
This week we get into his last two stops before Michigan.
Head coach and quarterbacks coach, 2007-'10
2007: Harbaugh took over at Stanford in December 2006 with Kellen Kiilsguaard, a high three-star dual-threat, and L.D. Crow, an early-recruited academic from the South, already committed. Crow was on a lot of 2007 early watch lists (I know because I was reading those religiously for Mallett news) and Stanford's first commit of that class, but he was passed by a lot of guys by Signing Day (not Nick Foles, Kellen Moore, or Ryan Lindley). Kiilsguaard would eventually switch to safety. Harbaugh couldn't lure another QB but did get a transfer from Michigan. MGoBlog's Brian Cook:
Redshirt sophomore quarterback Jason Forcier can read the writing on the wall -- it says "Jesus Christ, that kid can throw eighty yards" -- and is transferring to Stanford effective at the end of the semester. Lloyd is not happy about it.
Forcier, like all Forciers, was an accurate Marinovich project with enough legs to be classified as a dual-threat but not enough to overshadow his passing.
On the roster were a pair of fliers in Alex Loukas, a 6'4/193 Purdue-al-threat (see: every other Purdue quarterback of the period), and Marcus Rance, a barely three-star guy from Washington whose next best offer was Idaho. Harbaugh also inherited a 5th year senior and on-and-off starter in T.C. Ostrander, an Elite 11 prospect who signed as a 4-star and two spots below Brady Quinn in a deep year for pro-style QBs. Ostrander split time with 2006 3rd round draft pick (and former 5-star) Trent Edwards for three seasons.
Last among inherited bullets was Tavita Pritchard, a 2005 3-star pro-style guy ranked just behind Colt McCoy. Pritchard had thrown one pass—that incomplete—and was sacked on three other career snaps before Harbaugh arrived.
Ostrander suffered a seizure the week Stanford would go into #1 USC as 41-point underdogs. Against a brutal defense, Pritchard wasn't doing too hot—he'd go 11/30 for 149 yards, 1 TD and 1 INT in that game. But then 20 of those yards were a laser to Richard Sherman, and another 10 were the fade to Mark Bradford to win it. Pete Carroll wanted to know what Harbaugh's deal was.
Harbaugh's deal was he was recruiting quarterbacks. Andrew Luck committed at the end of June 2007, before Stanford had played a game under Harbaugh. The interest in Stanford was already there for the academic Texan, and meeting Jim sealed it.
Jim continued to recruit a second QB for the class. Targets included Dayne Crist (Notre Dame), Jerome Tiller (ISU), Sean Renfree (Duke), Ted Stachitas (Wake Forest), Wayne Warren (Rutgers), B.R. Holbrook (New Mexico), and another Texas prospect, Robert Griffin III. RGIII turned down Harbaugh's offer because of Stanford's admissions policy:
“I was graduating early, and Stanford wasn’t allowing early graduates to enroll and that was the biggest issue,” Griffin said.
So Stanford wound up with just one quarterback for the class. Luck was the epitome of the Harbaugh quarterback recruit: valedictorian smart, extremely productive in high school, cool demeanor, and some wiggle. Under Harabaugh he would develop into the best pro prospect since Peyton Manning, whom Luck displaced.
|Year||W-L||Player||Att||Cmp%||Yds||Rtg||Att||Yds||YPP||TD Rate||Int Rt|
TD rate and INT rate on the right are percentages for all attempts (passing and rushing
The sophomore Luck won the job over the incumbent senior Pritchard in 2009, but it was his junior season, 2010, when he really became Andrew Luck.
[Jump for 2009-2010 targets and the San Francisco story]
Brian buzzed me last week with a recruiting question on early enrollees:
1) Are EEs less likely to redshirt?
2) Are EEs more likely to start as freshmen? Underclassmen?
3) Are EEs more likely to be all-conference? Drafted?
4) Are EEs more likely to stick around as 4th and/or 5th year Seniors?
I hear a lot about the benefits of being an EE; you get on campus early, you get to start working out with the team trainers and players, start taking classes, etc. I think Clemson has something crazy like 12 EEs this year and I can't imagine that hurts their team development. I'm just curious if it actually gives any empirical advantage to those who do so.
Thanks and Go Blue!
I don't have national data, but I've got the early enrollees in my giant spreadsheet (see "EE" column). I'm pretty sure of things since 2008, but before that I had to rely on Michigan's press releases from signing days and spring games.
2015: Malzone, Cole
2014: Speight, Canteen, Harris, Cole, Mone, Ferns, Watson
2013: Douglas, Butt, Bosch, LTT, Taco, Dymonte
2012: Ringer, Bolden, Wilson
2011: Greg Brown
2010: Gardner, White, Hopkins, J.Rob, J.Jackson, Ricardo, Pace
2009: Forcier, V.Smith, Campbell, LaLota, M.Jones, Hawthorne, Vlad
2007: Mallett, Helmuth, Chambers
2006: C.Brown, Boren
2005: Kevin Grady
Just from reading that list you'll notice transition years have relatively few of them; a healthy Michigan probably has six or seven guys enrolling in January each year. You'll also note a lot of guys who left for one reason or another.
1) Are EEs less likely to redshirt?
Yes. 32% of EEs redshirted as freshmen versus 65% of those who enrolled in fall (those who never enrolled not counted). They obviously came to play.
2) Are EEs more likely to play early? Yes, but they're less likely to play overall. Here's the average number of starts per their season in the program for players who would be eligible*. Notice the difference?
* "Would be eligible" means I've removed redshirt (including medical), and transfer years, and 5th years of guys who never redshirted. Those lost to attrition otherwise are counted.
That is wow. It is extremely weird for there to be as many starts for true freshmen as for third- or fourth-year players. This shows that early enrollees are more likely to play as freshmen, but were progressively less likely to be starters each season thereafter.
You also can see the average start numbers per eligible player are rather low.
It's more accurate to say you find out what they are much earlier. Notably, NONE of the early enrollees to earn starts at Michigan redshirted initially (the 11 starts by a 5th year are all Gardner's).
It's also worth nothing that it wasn't the same guys contributing to those columns. Your true freshman EEs with more than 3 starts were Mason Cole (12), Jake Butt (8), Tate Forcier (12), and Darryl Stonum (10). Those guys—for reasons of injury, Denard, or temporal existence—contributed just 8 starts to the sophomore column, which is filled instead by Boren, Vincent Smith, and Jarrod Wilson.
3) Are EEs more likely to be all-conference? Drafted?
That seems to be much more relative to their talent, but we'd need national data to make that assumption. One day I'll add NFL draft information to the big spreadsheet; maybe we'll discover something then.
4) Are EEs more likely to stick around as 4th and/or 5th year Seniors?
As you probably guessed from the above chart, they are way LESS likely, and from the data it appears that's mostly because they're flight risks. Even if you figure all of the current players make it to graduation, early enrollees at Michigan have an average of 1.92 (!) lost seasons of eligibility out of the four they get, compared to 0.82 for fall entries.
This remains true even if you remove all the guys currently on the team. Here's a breakdown of the % of former players (from 1993 class to 2014) who stuck around X amount of years by when they enrolled:
|Seasons at M—>||5||4||3||2||1||DNQ|
That is stark. A good third of early enrollees left the program after just a year, and the hits kept on coming. When you total up all the eligible seasons of enrollees lost to various types of attrition, the EEs were particularly likely to be giving those seasons to other schools:
|% of Season Lost To:||Fall||Early|
|Gave up football||10%||2%|
Of the 37 early enrollees, six played out their eligibility and 13 are currently on the team. Three losses were natural attrition (Mike Jones was an unrenewed 5th, Hopkins gave up football, and Pace was a medical loss), and three were dismissals (Forcier, Stonum, and Austin White). That leaves 12 transfers: Boren, Mallett, Helmuth, Chambers, Emilien, LaLota, Ricardo, J-Rob, G.Brown, Ringer, Bosch, and Ferns.
Only the first two of those transfers wound up helping Power Five programs, though Bosch and Ferns still have the opportunity to do so. Mallett and Boren would have been guaranteed starters on the 2008-forward teams. The rest seem to be guys who were buried on the depth chart and realized it early.
What have we learned?
An early enrollee is more likely to care extremely about early playing time. They chose Michigan in part for an immediate opportunity to start, thus raising the likelihood of early playing time. However they are way more volatile in attrition.
Your expectations of an early enrollee from Michigan's smallish sample is that you'll find out right away if he's going to be either a long-term starter or a non-major contributor. A lot of these guys come to compete for an open spot, and either win it or move on.
Michigan's offense this year is facing the mother of all X factors in its quarterback situation. Brady Hoke left the rest of the team in relatively great shape, but its most important position in a Shane-or-die position.
Hoke and his staff recruited just Russell Bellomy (a last-minute flier stolen from Purdue) in the hybrid 2011 class, and skipped a quarterback altogether in 2012 because they already had a commitment from Morris in 2013. This was a bad idea then, and worked out awfully for Michigan. Bellomy's injury ruined any chance of a badly needed redshirt for Shane, so even if Morris worked out he'd be gone after 2016. And if he didn't work out: Michigan was going into this year hoping to catch lightning in a freshman from either lone 2014 recruit Wilton Speight, or early enrollee Alex Malzone.
From left: Morris, Speight, Malzone, Gentry, O'Korn. O'Korn won't be eligible in 2015 due to transfer rules but gives Michigan a guy they didn't have between Morris Speight.
This won't happen under Harbaugh. The former Michigan and NFL star likes lots of bullets and lots of competition at his old position, which he personally coaches. Harbaugh has already added the high-ceilinged Zach Gentry, a perfect complement to the high-floor Malzone. By this time next year (unless there's attrition), Michigan should have the above plus two years of eligibility remaining on Houston transfer John O'Korn, and likely one or two of the nation's best freshmen.
What I'd like to do, then, is go back through Harbaugh's quarterbacks—the starters and the recruits—to see if we can find any common threads in the type of guy he adds to the pile, and the type of guy who emerges from it.
|Jim used his Orlando offseason home as a base from which to recruit the talent-rich region for WKU. [USA Today]|
Recruiting assistant, 1994-2001
Bo's former defensive backs coach Jack Harbaugh was coaching at Western Kentucky, and struggling through his first few years, when the school decided it would cut two assistant coaching positions and a handful of scholarships (they already put very little toward equipment). His sons offered to do some scouting and recruiting for him—John from Cincinnati and Jim from his house in Orlando—and the harvest from those recruits was an WKU's rise to a I-AA national championship in 2002 and eventual reclassification into Division I-A.
The Jim-John co-op (John was doing much of the scouting, passing on guys Indiana couldn't recruit) was personally credited with 17 players on the national championship team. Nick Baumgardner got the story of Jim's first quarterback recruit, Willie Taggart:
Harbaugh explained he was trying to round up some talent for his dad's program. He told Taggart that he and his father were watching tape of Manatee and asked, "Who is that little skinny guy?" Jim said he thought he should play quarterback in college and he'd come by the school on Tuesday at lunch to discuss it further.
Taggart hung up and assumed it was a prank or something. "I called my high school coach and he checked on it and said, 'Yeah, Jim is Jack Harbaugh's son.'
Manatee was a Tampa area powerhouse back then, so Taggart's a guy who absolutely would have shown up on recruiting radars today, and had FBS programs looking at him then. He became the best QB in school history, improving a 2-8 team in in 1995 to 7-4 in '96 and 10-2 in '97. Taggart is now USF's head coach, and was an assistant for Jim at Stanford.
Taggart returned to WKU in 2000 and ran an option offense that rotated between three candidates. The original winner was Jason Johnson. They got the 6'3/200 Johnson out of Palmetto, but with the limited scholarships they couldn't offer him one out of high school. Johnson went to a military college for a few years before being re-recruited:
It was during that second season that Johnson had to renter the recruiting game. He was in contact with a number of Division programs, including Clemson, South Carolina, Kansas State and Indiana, but in the end Western won out.
Donte Pimpleton was the second, a local-ish dual-threat kid who wound up playing receiver—there isn't anything on the internet connecting his recruitment to the brothers. The third candidate, and the starter of the 2002 championship team, was Jason Michael, another local recruit, onetime Jim Harbaugh assistant in SF and now the OC of the Tennessee Titans,
Jim did recruit Alan Ogletree, an overlooked athlete from Atlanta who ended up starting at every position in the defensive and offensive backfield for the Hilltoppers (QB, RB, FB, WR, CB, SS, FS, K, P).
[After the jump: Raiders and San Diego]
Taco-ranked starters are far more likely than Glasgows [Fuller]
Every year, as college football recruiting becomes the only football thing left to pay attention to until spring, we are suddenly struck by an army of pundits so arrogantly attached to their "recruiting stars don't matter" narratives that they don't bother to care that math is against them.
Michigan typically gets taken to the woodshed in these articles for recurrently not matching recruiting expectations with on-field results. This discrepancy does exist beyond the normal J.T. Turners that everybody gets, and for various interrelated reasons: attrition spikes, spottily shoddy coaching, program instability, recruiting shortfalls. Anecdotally, there are examples I can point to, especially in the early aughts, when an otherwise two-star athlete was bumped to a three-star because Michigan offered. That explains less about how Wisconsin and Michigan State thrive on 2- and 3-stars, and more about how Michigan has recruited very few guys under a consensus 3-star.
However every time we find a new way to compare recruiting data to performance data, we consistently discover that recruiting stars handed out by the services correlate to better players. No, a 5-star isn't an instant superstar, but the 25-30 five-stars each season are consistently found to be about twice as likely to meet some performance metric (NFL draft, All-conference, team success, etc.) as the pool of 200-odd four-stars, who are consistently more likely to meet performance thresholds of the 400-odd three-stars, etc.
Today I present a new metric for proving it: starts.
|Example of raw data, via UM Bentley Library.|
ALL the Starts
My project over Christmas was to take the data from Bentley's team pages (example at right), scrub the hell out of it, and produce a database of who started what years, at what positions, at what age, with what recruiting hype, etc.
A few weeks back I released the initial results of my starts data. We noticed there were a lot of problems in that. I went back and did a lot of fixing, mostly just finding more weird errors in the Bentley pages I'd culled the data from, sometimes emailing the guys themselves to ask things like "Was there a game in 2001 that either you or B.J. didn't start?"
I think I've got it cleaned up now; at least the total number of starts for each season matches 22 players per game.
Recruiting By Starts
Starting in 1996 we start getting relatively uniform star rankings for recruits, though I had to translate Lemming rankings and such into stars (he had position rankings and national lists that line up with what we call recruits today). So I took the average of available star ratings of all players to appear on Michigan's Bentley rosters from the Class of 1996 through the Class of 2010, and put 'em against the number of starts generated. Guess what: recruiting actually matters.
|2- or 2.5-stars||29||271||9.3|
Even with Michigan's notorious luck, the 5-stars were expected to give you about two seasons of starts, compared to the 8 or 9 games you'll get out of a 2- or 3-star. That is significant, and offers a bit more evidence toward the general statement about recruiting stars: the higher the star rating, the more likely he is to be a good college football player, though at best you're at 50-50.
As for walk-ons, I've linked to the list of the 217 guys in that time period who made the Bentley rosters and weren't special teamers, in case you doubt me. The Order of St. Kovacs have accomplished great things for Michigan, but turning up one of those guys anywhere other than fullback has been rare indeed.
I'm going to try to use the starts data above to get predictive. The scatter plot of the 1996-2010 group was pretty linear so I'm just going to plug in a linear equation:
Expected Starts on Avg M Team = Stars x 5.30 - 6.35
And that gives us a reasonable expectation of Michigan starts to expect from a class based on their rankings:
click big makes
For the Class of 2011-2014 projections, I just guessed by hand, so those projections are going to be increasingly inaccurate once I'm predicting 2017 starters and whatnot.
The chart above has two stories to tell: 1) The strength of a recruiting class is strongly correlated to the value that class will produce in starters, and 2) the damage done by attrition to the 2005 and 2010 classes created ripple effects for several classes afterwards.
An Average Michigan Team:
By some quick averages I was able to get an average makeup of a starting 22. I took the average number of starts by experience (i.e. year in the program) for the classes of 1995-2010, adjusted those numbers for a 13-game schedule, then divided by 13 games to get an idea of what the starters ought to be against years of interest.
|Senior / RS Jr||7||5||4||8||8||6||8||9|
|Junior / RS So||6||10||5||4||7||7||5||6|
|Soph / RS Fr||3||3||6||2||1||2||3||4|
|AVG starter age||3.55||3.27||3.18||3.82||3.50||3.27||3.77||3.50|
By this the last two teams look extraordinarily young—about as young as the 2008 team or younger. The 2012 team by contrast seems like a wasted opportunity. FWIW I counted Devin, not Denard, as the quarterback, or it would have been even older. That fits the narrative: 2012 was a wasted opportunity, as a line with three 5th year seniors (two of whom were long-term productive starters) plus Lewan and Schofield was coached into one of the worst offensive lines in memory.