Pages

Friday, February 11, 2011

Statistical Analysis of 2011 Recruiting: By State


Last year I became curious about how successful Michigan was at recruiting each state.  That curiosity has spilled over to the 2011 class, and the chart above shows the percentages of successfully recruited players from each state.  This is all based upon offers recorded on the 2011 Offer Board located at the top of the page.  I can't promise that it's 100% accurate, because sometimes reporting of offers is a little bit iffy.  But it's the best I could do.

It should also be pointed out that Brady Hoke's late hiring as Michigan's coach somewhat skews the numbers.  Players who decommitted (Jake Fisher from Michigan; Matt Goudis from California; Kevin Sousa and Dallas Crawford from Florida) would have changed the numbers slightly.

As might be expected, Michigan was the most successfully recruited state.  Nearly 50% of in-state offers were accepted; the decommitment of Fisher was the only thing standing in the way.

Texas was a bit of a surprise.  Michigan pulled in three commitments from Texas - a heavily talented state - while only giving out eleven offers.  But it's a bit curious that Michigan's coaches didn't pursue more players in the state.  Running backs coach Fred Jackson has some contacts in Texas, and even though several Rich Rodriguez assistants had connections in Florida (where Michigan tossed out 43 offers), the success rate in Florida over the past couple years has been abysmal.  Michigan went 3-for-46 (7%) in Florida last year, and that number looks like an unbridled success when compared to the 1-for-43 showing this year (2%).

After crunching the numbers for two consecutive seasons, it seems that Michigan's coaches probably wasted quite a bit of time recruiting the Sunshine State.  Maybe it was a risk-reward thing with Rodriguez (after all, Denard Robinson is from Florida), but a batting average of .020 is horrible, no matter how you slice it.  When you hand out 190 offers, you probably spend quite a bit of time talking to kids or the coaches of kids who have very little chance of attending your university.  Perhaps Rich Rodriguez and Co. would have benefitted from making more intimate connections with fewer kids rather than, as one commenter said, "carpet bombing" the country with offers.

Ohio was once again recruited with some success (37% in 2010; 24% in 2011), and Hoke will surely continue to work hard in the state.  Even though Rodriguez got a large number of 2010 recruits from Ohio, he didn't seem to hit the state very hard in his last season.  The percentage of accepted offers jumped significantly once Hoke was hired.  He reeled in four recruits in approximately three weeks on the job, including the likes of Tamani Carter, Antonio Poole, Keith Heitzman, and Frank Clark.  Prior to his arrival, Rodriguez had accepted only three commitments from Ohioans - Greg Brown, Jack Miller, and Chris Rock.

Altogether, Michigan offered players from 27 different states.  The Wolverines struck out in twenty of them, including:

Pennsylvania: 12
Louisiana, New Jersey: 8
Arizona, Georgia, Indiana, South Carolina: 6
North Carolina, Virginia: 5
District of Columbia: 2
Arkansas, Colorado, Kentucky, Massachusetts, Mississippi, Nevada, New Mexico, Oklahoma, Tennessee, Wisconsin: 1

14 comments:

  1. This is very interesting analysis. It is mind boggling how much recruiting resources were wasted in Florida the last couple years. Hearing that Hoke's staff will be focusing much more on the midwest makes more sense to me now.

    The tough thing about Florida is that you have three big programs in state to compete with, and obviously in the surrounding states as well. So the top players have plenty of options right at home. Perhaps concentrating on Texas and California makes more sense because they do not have as high a concentration of college power programs.

    Denard was a bit unique in that Michigan was the biggest program to offer him at QB. So UM did not have as much recruiting competition with him. If Florida or one of the other big schools there made the same offer, he probably would not have been at UM.

    Maybe UM will not be able to recruit nationally in an efficient way until the team gets back to Big-10 champs level.

    ReplyDelete
  2. @ Painter Smurf 9:29 a.m.

    Yeah, it is a little bit crazy to know that Michigan recruited about 90 kids in two years and only got four of them. Michigan would literally need to offer 450 Floridians to get its full class of 20 commits!

    I think Michigan will be able to recruit nationally in the upcoming year. I just think Hoke and Co. need to be smarter about it than Rodriguez was. They need to recruit to their strengths (Southern California, Texas, and the midwest) and not spread themselves too thin.

    ReplyDelete
  3. Its an interesting but not surprising funding. I think theres danger in reading too much into this. Not all scholarship offers come with the same level of effort. And I'd guess the out of state oferees were generally higher regarded than the local kids. We know that RR's staff took the mentality that an offer was just used to spur interest. Its almost like it was a starting point for them. Whether this was right or wrong can't be shown by counting and dividing.

    If it takes 450 offers to get 20 4-stars out of florida and only 50 to get them out of the midwest -- so what. This is a results oriented business. Recruiting was one area (perhaps the only area) where he didn't fail.

    I find plenty of flaws in his recruiting approach but I don't see "recruiting batting average" as a particularly enlightening or meaningful one.

    -Lank

    ReplyDelete
  4. To flip this a little bit, you could look at the total 5 stars, total 4 stars, and total 3 stars and notice that Michigan has a much higher success rate with 3 stars and very low success rate with 5 stars. But no one would reasonably conclude that means Michigan shouldn't bother recruiting elite prospects.

    -Lank

    ReplyDelete
  5. @ Lankownia 4:30 p.m.

    "Recruiting was one area (perhaps the only area) where he didn't fail."

    I'm not sure if I agree with you here. In fact, I don't think so. The point of recruiting is to get good football players and put them on the football field in Michigan Stadium. The performance of the team over the last three years suggests that he didn't quite succeed. There are certainly some exceptions (Denard, Roundtree, etc.), but I don't think you can point to Michigan and say, "Now THERE'S a team that knew how to recruit."

    And we'll have to agree to disagree on the "recruiting batting average" thing. I don't think it's the only factor to be considered (i.e. if you offer 450 players and gets 20 future All-Americans, obviously there's no fault). But when you're not getting impact players, perhaps it's time to change your strategy.

    Again, it remains to be seen whether this recruiting tactic will work. All 380 kids offered over the past two seasons are either soon-to-be sophomores or incoming freshmen, so we can't pass judgment. But if the coaches are taking the time to watch film, talk to coaches, do background checks, mail players, call players, e-mail players, etc., and 23% of your time is spent talking to kids that will add only 1 or 2 kids to your recruiting class...well...that seems like a bit of a waste.

    ReplyDelete
  6. True. I just meant by recruiting rank which, as much as we want to argue matters, is still an inexact science. Player development and coaching matter can mean more than talent.

    RR's classes were ranked on par with Lloyd and they were just as focused on local players (Michigan and Ohio) as Lloyd. RR just offered way more out of state kids. We'll never know if this was wise or not.

    I suspect in many cases the staff didn't put enormous effort into scouting kids. Your logic about effort is just a risk/reward issue and can be applied to argue that Michigan shouldn't recruit 5-star players because they have less chance of getting them than 3-stars. This is obviously silly.

    ....Also, consider this: in 1 month on the job, how many offers has Hoke put out or agreed to honor? 50? Multiply this number by 12 (or 8 if you only want to consider off-season). Now consider that Hoke was working much of the last month without a full staff.

    This tells you it doesn't necessarily take in-depth knowledge to offer recruits.

    I realize you didn't raise this issue as a Hoke vs Rodriguez thing but the tone seems to be critical of Rodriguez in favor of a more conventional old-school Michigan approach. And don't get me wrong - I actually agree with the conclusion. I just don't see the batting average by state issue as being real meaningful.

    The only advantage I see to recruiting local kids is that they're less likely to transfer after they arrive.

    ReplyDelete
  7. @ Lankownia 8:26 p.m.

    You and I both know that the following two statements aren't equal:

    a) Michigan coaches might be wasting their time by offering Floridians because they're unlikely to come to Ann Arbor.
    b) Michigan coaches might be wasting their time by offering 5-stars because they're unlikely to come to Ann Arbor.

    So far I count 33 new offers given out by Hoke, including both 2011 and 2012 recruits. The common thread with most of those recruits is that the vast majority have come from the midwest or their old stomping grounds in California. And surely he won't continue to give out 33 offers a month.

    Am I being critical of Rodriguez when he offered 90 kids in Florida and only got a handful? Yeah, I am. I didn't write the post strictly to be critical - it's an interesting topic to me, regardless of the coach - but if it comes across as being skeptical of handing out nearly 200 scholarship offers a year, well, I guess I got my point across.

    By the way, counting from the 2008-2010 classes, 8 deserters were from local areas (MI, OH, IN) and 6 were national recruits (CA, TX, NJ, etc.). Obviously the percentage of those kids would take a little light research that I don't feel like doing at nearly midnight, but I would guess that local recruits aren't that much more likely to stay than national recruits. For every Justin Feagin, there's a Boubacar Cissoko. For every Anthony Lalota, there's a Dann O'Neill.

    ReplyDelete
  8. I entirely agree with Thunder. If you’ve offered a kid, then you’ve probably expended resources on him, and RR’s Florida strategy had an exceedingly meager return. On top of that, the ones you get are mostly those the Southern schools didn’t want. Even Denard Robinson probably wouldn’t be at Michigan if the major Florida schools had offered him at QB.

    I’m not saying (and Thunder isn’t saying) that Michigan shouldn’t recruit nationally, but you have to be realistic. Spraying offer letters in directions where they’re unlikely to be accepted is not a good use of time.

    The guy who said RR’s classes were on par with Lloyd’s is just plain wrong; they were not. It is even worse if you consider attrition. Recruiting class rankings are basically frozen after signing day, and don’t count whether the kid actually made it to Michigan and stayed with the program. Although there is no national ranking of attrition, my sense is that Michigan in recent years has fared worse than would be considered normal.

    Another metric is: when was the last time a Michigan class was rated more highly than an Ohio State class? (Not lately.) When was the last time a kid who had an Ohio State offer chose Michigan instead? (Not lately.)

    How many five-star kids has Michigan signed since Lloyd retired? Just one (per Rivals): William Campbell. Although recruiting rankings are inexact, there is overwhelming evidence that the “stars” are highly correlated with collegiate success. Michigan is highly unlikely to make it back to elite status without landing and retaining more of the four/five-star kids.

    ReplyDelete
  9. @ Marc Shepherd 2:16 p.m.

    I think you would be correct in assuming that Michigan's attrition problems have been worse than normal. One might argue that the SEC suffers from a great deal of attrition, but that's largely because they oversign and then force kids out. So I don't think that really counts. Attrition has been a huge problem for Michigan the past few years. I don't really give Rodriguez much credit for getting Demar Dorsey, Taylor Hill, Austin White, etc. when they never played a down for Michigan.

    According to Rivals, Michigan had a better recruiting class than OSU in 2010, 2007, 2005, 2004, and 2003.

    Also, I have to say...

    Michigan got several kids in 2010 with offers from Ohio State, including Cullen Christian, Will Hagerup, Marvin Robinson, and Devin Gardner, if I recall correctly.

    Zero kids picked Michigan over OSU in 2011, but there were a handful in 2010.

    ReplyDelete
  10. RR's recruiting classes were ranked on par with Carr's. There wasn't a substantial difference on signing day. This fact indicates to me that there was nothing inherently wrong with RR's approach. Particularly if you want to argue that the W-L record was a substantial drag on recruiting.

    After signing day, RR ran into problems with his recruiting classes. I'm not sure what the reason is but his attrition seemed to be higher than 'normal'. I first assumed that it was because RR chose to recruit more distant recruits, who had a greater tendency to get home-sick. But when I looked at the numbers, RR was getting just as many local (MI and OH) kids as Carr, so that wasn't it.

    Regardless of the recent anecdotal evidence, being a CA or FL or TX kid just gives you one more difficulty/reason/excuse to leave the program.


    Thunder,

    I'd be willing to bet a higher percentage of the Florida kids recruited were 4 or 5 stars. I wonder if there is any significant difference when you account for star ranks. For example, is Michigan's success rate with 4 star offers in Florida much different than their success rate for 4 stars in OH? Or is the FL 'batting average' just a function of degree of difficulty (more competition for higher stars).

    We can agree that 3,4, and 5 stars have different cost benefit calculation than 3 stars right? If the Florida guys are higher caliber than the OH/MI guys, then you expect and accept the lower batting average. Right?

    The biggest problem here is the assumption that all offers require the same investment in time. This seems bogus. There's no question that coaches emphasize certain recruits over others. The coaches probably spent far more time driving to local games and meeting with local coaches than they did for the FL recruits. I don't think you can just ignore that. Michigan wiffed on Zettel and Jernigan
    and Prince Holloway, but the level of effort placed in recruiting the 3 was almost surely not the same.
    When I said 50, I was counting guys that were reaffirmed by Hoke. His recruiting class was 20 strong plus we heard he had offers out to many other prospects that were not landed. I'd guess it was 50 offers just for the 2010 class, in one month, albeit an atypical month. Its not really the point, since I don't expect Hoke to put out 400 offers, but as you like to wait till things are over and done with, we'll have to wait till Feb 2012 to find out for sure.

    ReplyDelete
  11. @ Lankownia 1:53 p.m.

    I agree with most of what you write here. I guess neither one of us really knows how much time the coaches spend on recruiting Anthony Zettel vs. Prince Holloway. While the coaches probably spend a lot of time driving to places like West Branch to see Zettel play, they also spend 3 hours on a plane every time they want to go to Florida. And I would guess that if they want to "background checks" (i.e. talking to coaches and teachers, checking transcripts, etc.) about certain kids, it's easier to find out background info on a kid from Detroit than it is a kid from Pahokee, FL.

    I think we both can agree that sending out official offers takes time (film review, background checks, paperwork, etc.), and therefore, 190 offers are going to consume more time and resources than 100 offers. The cost-vs.-reward is yet to be determined because a lot of those kids haven't had a chance to "pan out" yet, so maybe those few Florida kids we got will all be superstars like Denard Robinson.

    The question for me seems to go like this:

    Could Michigan have found a Vincent Smith in OH, IL, or MI? Do you need to hand out 46 offers in Florida to get a Richard Ash when you have Jonathan Hankins in your backyard? Etc. I don't want to turn this into a discussion about whether Hankins should have received an offer, because that's tangential. But does it really pay to recruit all those kids in Florida and get very few of them when you can toss out a local offer and get the kid without much effort, thus saving time/resources for the big fish? I don't know for sure, but it's something to weigh when deciding on a recruiting strategy.

    ReplyDelete
  12. Also clouding issue is RR's use of 'non-commitable' offers. You could argue these are not unlike the 'verbal' offers that the '12 recruits have been given, but there seems to be a difference when the formality of a written offer is in place. I get the sense that the background checks you describe weren't always performed (e.g. Dorsey).

    I think you could certainly argue that a local version of Vincent Smith was gettable or at least a player of that caliber (like say Rawls.) The question I ask is - who cares? Is a local version of Smith any better? If you invest the same total time in offering 3 Smiths from a distance as you do 2 local Smiths, who really cares as long as you land the one Smith. The only difference may be fan preference for local players.

    Ultimately, I agree with you about the benefits of local recruiting. With local kids you have the inherent advantage of proximity. You can form relationships a little more easily, you can be plugged in to the rumor grapevine a little easier. You can worry a little less about a player getting homesick after he arrives, and you know he's a little less likely to have something to runaway/back to.

    To me, the Austin White situation indicates a bigger problem with RR's recruiting strategy than geography. White was a kid with character question marks that most had pegged for MSU. RR didn't seem to mind. He was either indifferent or oblivious and I'm not sure which is worse. White and others like him are now gone and that didn't come as a surprise to everyone.

    ReplyDelete
  13. To me, the bottom line is that RR got a lot of his 4 stars out of Florida and therefore the effort was worth it. We can't say much about the level of effort between different recruits, but that doesn't mean we should assume they're all the same.

    RR's strategy was both successful and imperfect, as was Carr's. Maybe the flaws in strategy were significantly different, but my opinion is that geography played little to no role in these flaws, whatever they may be.

    ReplyDelete
  14. While we agree that offers take time in general, and some recruits require a substantial investment/effort, we can also agree that others do not. We need look no further than the last couple weeks when Hoke has issued 20-30 offers to the '12 class. He can't have invested that much time in these offers.

    These offers represent the bare minimum effort required to offer a guy and at this level you could offer hundreds upon hundreds. But, obviously it takes more effort to actually land the kids.

    I think we all agree that results are what matters here, but I think as fans we'd like the Michigan offer to hold a certain prestige and not be handed out so easily. You want an offer to carry cache, like OSU and Texas programs have right now. You want a recruit to feel like 'if I don't jump on this soon, someone will take my spot'.

    At the same time, its a new world. Kids are taking it personally if they aren't offered by fall of their junior year. In that environment, its challenging to find the right approach.

    ReplyDelete