Pipelines / Walls and other geographical myths
The map you see below is a compilation of all the players that have been on the U of M roster starting with Bo and ending with RichRod.
There is a lot of information here...but lets start with the Michigan/Ohio region and work our way out to the rest of the country.
Michigan vs Ohio
There has been a lot of discussion on this board about the Great Wall of Tressel, and how he shut down the border. Lets take a look at how our roster was compiled with Michigan and Ohio kids over the past few decades.
A couple of things stick out here...
- During Bo's era 1 out of 4 kids on the team were from Ohio. From Lloyd's tenure and to the present it was 1 out of 10.
- Our presence in Ohio started to decline long before Tressel got there.
- RichRod always caught a lot of flack for not recruiting Michigan players, but he averaged more Michigan players then previous coaches (Yes, I know we are dealing with sample size, and most of his roster was still Lloyd's players.)
Here is a look at some of the raw data I was working with...
Bo's Midwest:
| Rank | State | # Players | Percentage |
|---|---|---|---|
| 1 | MI | 326 | 41.48% |
| 2 | OH | 201 | 25.57% |
| 3 | IL | 80 | 10.18% |
| 4 | IN | 21 | 2.67% |
| 5 | WI | 12 | 1.53% |
| 6 | MN | 1 | 0.13% |
| Totals | 641 | 81.55% |
Moeller's Midwest:
| Rank | State | # Players | Percentage |
|---|---|---|---|
| 1 | MI | 55 | 32.35% |
| 2 | OH | 26 | 15.29% |
| 3 | IL | 17 | 10.00% |
| 4 | IN | 9 | 5.29% |
| 5 | WI | 3 | 1.76% |
| Totals | 110 | 64.71% |
Lloyd's Midwest:
| Rank | State | # Players | Percentage |
|---|---|---|---|
| 1 | MI | 183 | 41.22% |
| 2 | OH | 44 | 9.91% |
| 3 | IL | 24 | 5.41% |
| 4 | IN | 19 | 4.28% |
| 5 | WI | 5 | 1.13% |
| 6 | MN | 4 | 0.90% |
| 7 | IA | 1 | 0.23% |
| 8 | NE | 1 | 0.23% |
| Totals | 281 | 63.29% |
Rich Rod's Midwest:
| Rank | State | # Players | Percentage |
|---|---|---|---|
| 1 | MI | 59 | 46.83% |
| 2 | OH | 14 | 11.11% |
| 3 | IL | 5 | 3.97% |
| 4 | IN | 1 | 0.79% |
| 79 | 62.70% |
Regional and National Footprint
The graphs you see below break down our roster by Region....
I was pretty amazed to see that 82% of Bo's roster was compiled of MidWest kids. I'll update this or do a Part 2......but it will be interesting to see how Hoke's first class will break down once it's complete. Right now its leaning torwards Bo's percentages.
** note: All data was pulled from http://bentley.umich.edu/athdept/football/football.htm
[Ed-M: I used basically the same database for this year's HTTV article on recruiting Ohio, and ran into a similar problem: walk-ons. Bentley doesn't say who's a scholarship player and who isn't. Also the nature of a 4- to 5-year turnover cycle makes it hard to see what effect a coach is having except by tracking trends, not total % of players. For example, Bo's Ohio recruiting had been trending down for a long time in his later years, so that when Moeller took over it looks like the story is Mo recruited nationally and Bo was a Midwest guy. In truth, Bo took over a team from Bump that had been capping at 80% people from a mitten- or trash can-shaped state. Bo started by scouring Ohio but once his program was established he went all around the country. To wit:
If you break that into trend-lines...
...the conclusions are radically different than the percentages above suggest. Also RR and Mo weren't really there long enough to make any judgments, except the first chart shows a pretty radical growth in Ohio prospecting by RR replacing in-state success of Carr.]
to see that RR's percentages were roughly the same as Lloyd's. Well done.
Is this data recruiting classes or just the roster? If it's the roster, then there is a 5-year lag until all of a coach's kids get in the system.
Its just the roster.....I know that skews things especially for Moeller and RichRod based on sample size......but if I wanted to do it based on recruiting classes I dont know where to find the recruiting data from 20 - 30 years ago.
But in the short run it loses meaning.
Certainly for Bo, and I think that given the length of Lloyd's tenure, you can use roster data for those 2. I'd argue the roster through 2008 and maybe even 2009 is primarily attributable to Lloyd also.
Moeller didn't last as long, but to some degree he was a continuation of the coaches around him.
For Rodriguez (and eventually Hoke), you should be using the recruiting data. His first recruiting class (2009) includes players like Lewan and Gordon who are (under Hoke) entering their Soph year of eligibility.
------------------------------------------
Good work putting this together in a pretty format.
I think this has been looked at to some degree before. Rodriguez recruited Michigan slightly less and increased focus on Ohio and Pennsylvania relative to Lloyd/Mo. [not a real surprise given he knew those areas from WVU-time.] Overall he took more midwest recruits and the Florida-focus was vastly overstated. However, the origin of the gripe comes from Rodriguez missing out on high profile players and taking fewer OSU oferees away from them in Ohio (something that Lloyd didn't do super often either, but he was better at keeping Michigan guys home, at least until the latter years e.g. Perry, RoJo)
Keep up the great work, I enjoy anything that exposes a platitude.
Your main point that there is a downward trend does not require recruiting data, the roster data clearly is enough, the cause of the downward trend is the next question.
---------------------------------------------------------------------------------
New
Though I see your point, I still don’t think you need recruiting data. You could mine the same roster data, then make a year by year comparison chart showing the % of incoming freshman from Michigan and the % of incoming freshmen from Ohio.
I believe the main complaint during RR's tenure and at the end of Lloyd's was that we weren't getting the top talent from Ohio anymore. When RR was here we only grabbed a few players from Ohio that had OSU offers. Pretty much every verbal from a recruit we have gotten this year either has an OSU offer, would have gotten one if they didn't have just 15 or so schollies, or would have one if the NCAA stuff wasn't changing their strategy.
Agreed. At the end of the day, it matters little where players came from. The entire recruiting class can come from Delaware for all I care as long as they're great players.
I don't want to seem like I'm bashing the poster though - the diary is still interesting.
Delaware is fine, but God please not Oregon, Idaho, Montana, North Dakota, South Dakota, or Maine, and don’t even get me started on Alaska, and Hawaii
RR hates the U.P.!
I'm tired of haters ripping on RR for his supposed discrimination against the UP! He gave Yoopers a fair chance!
Commence board meltdown.
We would expect things to change in terms of roster composition over time to be less midwest dominated because of the population shifts from Midwest to Sunbelt states in general for us, so to see that change from Bo to Lloyd is expected. We'll need a few more years of Hoke to see how his staff will recruit and do before we can make any conclusions on where we will get our talent.
If we can assume 1 player out of x thousand is D-1 caliber, then it stands to reason that more populous areas would have more players recruited from them.
From just 2000 to 2010, the U.S. population increased by about 10% nationally. Here's the regional breakdown.
is the flow of information. If Bo was finding out about a recruit in Florida or Arizona it was via telephone and maybe VHS tape sent by mail. Hard to do an in-home visit before cellphones, internet, and chartered private planes. It was simply a different era where EVERYONE recruited more local kids.
I knew that there have been a lot of players over the years from New Jersey but this really drives that point home.
Yeah, and we haven't even been a major player there, just getting the odd kid now and then. Rutgers is sitting on a gold mine, but has never taken full advantage.
i like all the info on the team break down. but part of me wonders that nationally the michigan brand spread west to CA, TX and other states after Desmond won the heisman. I could be wrong but i think that with all the success in the early 90's it help pull more kids from those states. Like now with a "slight" backslide in recent years we are not neccessarlly so prevelant out in CA. who knows football changes, kids change, recruiting changes.....
Don't forget the Pete Carroll/Hollywood/La-La-Land factor that USC boosted into the stratosphere after tanking in the early/mid-90's.
I'll have to start a rant-ey board post about "Michigan brand." (Not an assault on you, Protoman) but geez, the "Michigan brand" is about the U, not ONLY the football team. Maybe use "Michigan Football brand" and carry-on?
Good stuff. I'm equally surprised that Bo had 82% of his kids from the south, especially given the provincial nature of college recruiting back during his era - kids just didn't move around as much.
I always thought RR received a raw deal from a recruiting standpoint - he lost some high-profile recruits in-state because of their coach's strong ties to MSU, not because he didn't try to get them.
The Tressel wall definitely intensified as the years went on, though I agree it started before he showed up.
Bo had 82% of his kids from the midwest (note that the OP says 1 in 4 were from Ohio).
He only had 2% from the south.
The shades of blue are a bit close, so it's understandable.
And to further reassure the poster above you, I checked the financials for Housing & Food Services during the Bo years and there was not a separate line item on the budget for "Grits."
I like grits, but I nearly blew coffee out my nose. Thanks.
Rich Rod hates Grand Rapids! FIRE RICHROD.
and vice versa.
Wonder if RR went out to Vegas to recruit...whoever that is from Nevada. Thanks for posting the trends, it'd be interesting to see one that tracks geographically where the offers are going. I know it would be impossible to do for more than a couple years back, but it would give a bit more data to work with. All I am aware of is the 2012 offer list in the useful stuff tab, does mgoblog have records from any further back?
I think the Nevada guy is George Morales, the long snapper that Carr mysteriously recruited his last year.
Awesome work! I'd love to see a similar breakdown for weight under those 4 coaches as well. Only if you have enough time. Again, great job!
Without the prevalence of national recruiting sites, easy travel, workout/practice videos, highlight reels, etc. it would make sense that the maority of Bo's kids came from within a region that is much more easily traveled and scouted. I mean, this was before cell phones as well.
To me, this isn't as much the fact that we're getting away from recruiting the midwest. To me, it's more of the fact that there are many more resources available for kids to get exposure on a national scale to big-time national programs.
back in the 70's, there wasn't cable TV. There was one, maybe 2, games on each saturday. About the only time a kid in Texas, Florida or California saw Michigan on TV was against OSU or in a bowl game (and the Big Ten only let the conference champion go to a bowl game up until the '75 season). Even for most of the 80's there wasn't the all you can buffet of football games on TV like there is now.
I don't think that the South, namely Florida and other pipeline states were quite what they are now as compared to Bo's time. Florida started being seen as a hotbed of recruiting in the early 80s, with the rise of the Miami Hurricane teams.
I appreciate the time that you put in, but the results are not useful. Why?
1. You used roster make-up, not recruits. While this would work over an extended period of time, it does not work for a coach like RR who was here for 3 seasons. During RR's first season, the team was almost entirely Lloyd's players. Counting these as RR's players makes no sense. Even during year 3, after massive attrition, about 1/2 of the team was made up of Lloyd players.
2. You looked purely at raw numbers, not recruiting profiles. Let me explain. When Moe LC were coaching Michigan, we routinely owned Michigan. It was fairly rare for us not to secure most o the top 10 in-state players. Sure, MSU or some other school might come in and pick off a few each, but we dominated the instate recruting battles. During the past few years, the top in-state talent did not come to Michigan to the same degree. This is not a criticism of RR, just a fact that if you look at the star rankings of the instate talent that we secured the last 2-3 years, it does not compare (overall) to the prior years.
3. OSU and Tress' fence. I see your point that the percentage of Ohio kids on the roster started to fall off before Tressel took over at OSU. But, the point with Tressel in my mind has always been that he came in and locked down TOP Ohio talent. So, while the numbers were falling pre-tressel, before he got there, the Ohio kids that we signed were still top players. Post-Tressel, OSU locked down the majority of top Ohio talent.
Speak for yourself.
The charts and graphs provide me with an informative and interesting picture of the geographical trends in the origin of Michigan players across four coaching regimes. The more granular details you refer to are also interesting, but the fact that they're not present doesn't mean the article is without value to other people besides you.
Maybe you should follow up with your own analysis using the three points you mention.
I agree with parts of your long-winded critique (and made some of the same above) but some of what you're saying isn't true.
It seems (or at least feels) true that Michigan won more than it's fair share of elite in-state recruits in the past, but the slide can't be blamed on Rodriguez. The exodus, if there was any, began under Carr. In 2005 Michigan landed 5 of the top 10 (typical I'd say), in '06 is was 3, in '07 it was....zero.
Rodriguez got 5, 2, 2 while he was head coach. That 9 in 3 years to Carr's total of 3 in the previous 3 years.
This is something thats repeated over and over. Rodriguez is villified, rightfully, for many things but the quality of his recruiting shouldn't be one of them.
When you actually look at recruiting data you see that the quality of Rodriguez's recruiting was about the same as Carr's which was the same as Moellers which was the same as Bo's which will be the same as Hoke's. There's some noise from year to year but Michigan largely recruits itself.
Great work, thanks for the analysis.
One red dot is obviously missing though, and it's a big one. New Orleans! (OK, technically Metairie, LA.) The home of Carvin Johnson, one of our better performing freshmen last year. I'll let you off easy though, as even Les Miles couldn't find this kid on the radar.
However, I will not rest easy until credit is given to Rich Rod and Fred Jackson for finding such a gem. And with that, commence "Rich Rod could only find gems in the Deep South because he hated Michigan" meme...
to rosters with many more kids from other parts of the country can also be partially explained by the massive population growth in the Sunbelt states that accelerated in the late '60s, coupled with the population declines in the states of the so-called "Rust Belt" in the late '90s and 2000s.
cellphones, chartered flights, and the internet
I must be getting old. During Bo's reign, we all bitched about the over-relaince on Midwest kids. We would have killed for a roster made up of 15% Southeast kids, ~10% West kids, and ~10% Northeast kids. Killed. It would have been a prayer answered.
Now we're all pointing to 15% Southeast kids, ~10% West kids, and ~10% Northeast kids as a sign of some kind of recruiting failure and are anxiously awaiting the day when Brady Hoke brings us back to the glory days of Bo.
Be careful what you ask for. Don't think you won't start bitching about a roster with 80% of the kids from the Midwest.
Let's face it, we don't know what we really want, we just want to win. If we're not winning enough, then we want whatever is different than what we have now.
During Bo's day when we were winning the Big 10 but losing Bowl Games, this was our wish list:
- A wide open passing oriented attack, less reliance on winning games 10-7,
- A quick-strike offenese that can come from behind, less empahsis on ball-control and running out the clock,
- Play-calling that would forego field goals and go for it on fourth downs,
- A defense that could defend the pass, not just the run,
- A shot at an MNC. No Rose Bowl tie-in,
- A grass field, so that we were comfortable with the surface when it came to Bowl Games,
- A national recruiting class that de-emphasized "slow" Midwesterners,
- Warmer weather,
We wound up getting all of these things during the Moller and early/mid Lloyd years, even a little Global Warming thrown in. Were we happy? Only as long as we were winning. When Lloyd and then RR starting losing, we were desperate to get back to our Bo/Midwest roots.
Except the problem is that when we were actually living back in those Bo/Midwest roots days, we were desperate for something else.
From my years in school in the early 90's, the rap often heard against Mo was he recruited too many top recruits from outside of the Midwest who weren't "excited to play for Michigan." So when these blue-chip recruits couldn't crack the starting lineup, they became cancerous and the team chemistry went to hell in '93 and '94. No idea if this was really true, and as noted above, Mo's sample size is too small based on rosters. I'd be interested in seeing his recruiting classes again though.
When Lloyd came in, he said he was going to focus on guys who "wanted to play for Michigan." Obviously, every kid wants to start, but there certainly seemed to be great team chemistry in the mid-to-late 90's under Lloyd. Whether that had anything to do with recruiting these "Michigan fans" (as opposed to recruiting really talented kids who lived up to their potential), I (again) have no idea.
Well, aint this just a goegraphical oddity......two weeks from everywhere!
Sorry. Couldn't resist.
Bo, Gary, Lloyd, and RR all were good recruiters. RR was the only coach of the four that was 15-22 in three years. I could go on, but I'll save my rant for a better day.
I agree. 15-22 after 3 years makes a coach horrible. A losing record after 3 years can only mean that RR sucks. No other factors are involved. Only wins and losses matter.
Now a quick question...what would you say of a coach that only had 10 wins in his first 3 seasons at a new job and when he left said job, his winning percentage wasn't as high as the schools overall winning percentage? Or do you want to rant about Brady Hoke being a shitty coach also?
My sense is has always been that the issue is not that we're getting fewer players from the midwest, but fewer of the elite players out of ohio. That would be really hard to measure though.
But it doesn't "prove" anything about pipelines unless you just look at recruits. RR's rosters were, roughly:
2008: 75+% Lloyd recruits
2009: 50+% Lloyd recruits
2010: 25+% Lloyd recruits
So of course it's going to look more similar than it might have in a hypothetical RR year 5.
Second problem is that rosters reflect post-attrition numbers. Isn't it plausible that transfers are more likely for recruits from outside the local area, whereas walkons will be local? I'm talking McGuffie, Mallet, etc. I don't have any stats on this to back up this idea, but it seems plausible. In a small sample size situation like this, it could also skew the roster figures. RR had a lot of attrition, and if that attrition was primarily among out of Mich/Ohio recruits, then you have something consequential on your hands.
Keep in mind that recruiting was much different in Bo's early days compared to what it is now. The disimination of information was not as fluid as it is now with the internet and a plethora of recruiting agencies. As a result, it is much easier to recruit nationally now than it was then. That might be another reason why Bo's midwest percentage was so much higher than his successors.
I really appreciate the graphs especially where we've gotten every player from since the Bo years.
If you want to prove though that Tressel didn't put up a fence to Ohio for Michigan then you need to look at lists of the top recruits in Ohio in the Tressel years and before and how many of them ended up coming to Ohio State rather than Michigan and compare to the years before.
The problem with what you have is that if Michigan recruited 25 players in Ohio that were ranked the top 75 through 100 players in the state of Ohio against Tressel who had recruited the top 15 players in Ohio and got the rest of his roster elsewhere it would look like the Ohio border is open when really OSU would have been recruiting better.
Exaggerated example yeah but you can get the point. I'm not sure I want to see what the numbers look like though I would hate to see anything prOSU especially when we know they only got it by offering free stuff.
...organized by year and coach.
Comments