Hey, wasn’t David Pollack a three-star recruit?

As far as I’m concerned, Matt Hinton does the Lord’s work with his posts skewering what I refer to as the macro versus the micro approach to recruiting success.  What I mean by that are the people who conflate the individual success story of the plucky obscure signee who becomes an All-American into an overall approach to recruiting.  You know who I mean – the people who claim that they’d rather have a team made up of hard-working two and three-star recruits than see their school chase after the high-profile guys, because, after all, look at the all-stars who weren’t highly sought after coming out of high school, yada, yada, yada…

The next time somebody makes that argument within earshot, send him or her here.  In particular, this:

On the other hand, if you consider the initial grade as a kind of investment – a projection of the how likely a player is of becoming an elite contributor compared to rest of the field – well, you’d put your money with the “experts” over the chances of finding the proverbial diamond in the rough every time:

Of course, a large number of players in that sample size haven’t finished their careers, but you can divide up the numbers over any time period you’d like – one year, five years, 10 years: The ratio always looks identical on a per-capita basis, and it is not a crapshoot. Four and five-star players are roughly seven times as likely as two and three-star players to land on an All-America team, and the numbers in the NFL Draft tend to be even even more lopsided toward the hyped recruits. All the more reason to want as many of them as you can get your hands on.

Coaches chase the studs not because they’re a guarantee of success, but because the odds are better that they’ll succeed.  If you could sign an unlimited number of recruits and winnow the chaff, it wouldn’t matter.  But when you’ve only got 25/85 chances to get it right, logic suggests you’re going to gravitate towards the surer option every time you get the chance.  Or does anybody really think Nick Saban does what he does because he wants to finish first in Rivals’ recruiting rankings?


Filed under Recruiting

67 responses to “Hey, wasn’t David Pollack a three-star recruit?

  1. Hogbody Spradlin

    I shouldn’t be surprised, but if the chances of a 5 star recruit making All American are a little under 8%, then a great deal of extra recruiting effort is spent for a small marginal return. On the other hand, the big dropoff between the 5 star and 4 star numbers makes me think that they (whoever ‘they’ are) do a good job of identifying the 5 star prospects.

    • What fresh hell is this?

      Don’t get me wrong, I want UGA to get as many 5 star recruits as possible, however, looking at this graph like it’s a scientific experiment with exact, actionable results is horsehockey.
      These “results” are heavily influenced by self-fulflling prophecy. Do coaches give as many practice snaps to a 3 star recruit as they do to a 5 star recruit….how many with the first team?
      Would many (any) coaches start a 3 star player over a 5 star player even if their performance in practice had been equal? Much like the pros, there is pressure to start your high draft picks. It seems that the status of many 5 star recruits has been preordained by both the media and the coaches.
      Many 3 star recruits only get a chance to shine following the injury of a more highly recruited player.
      This is sort of like doing a drug trial where you tell the participants who is on placebo and who gets the real thing.
      Then there is the fact that many recruits are just late bloomers physically.
      How many 5 star recruits at Boise State over the last 5 years?
      Usefull as a guide, but don’t forget to put in the work coaches, and don’t get too wrapped up in stars fans!

      • What fresh hell is this?


        • BMan

          Sort of applying the Malcolm Gladwell explanation in “Outliers,” where the kids with the 5-star labels get more opportunities to practice/play (through All-Star games, increased reps, etc.)?

          • What fresh hell is this?

            Moreso than practice time, the tendency to predetermine a players value based on number of stars and media/fan hype.

            If the pros can’t do it well (talent assessment) after 3-4 more years of playing time and physical maturation, how are coaches to make good decisions about 17 yr. olds based on some recruiting services rating.

            Consider The Ryan Leafs, JaMarcus Russels vs. the Tony Romos, Danny Woodheads (not even invited to the combine or recruited by a d-1 school).

            • Hackerdog

              You’re missing the entire point. You’re confusing anecdotes with data. For every “can’t miss” 5 star kid who doesn’t amount to anything, there are four 3 star kids who don’t. The issue is that nobody pays any attention to a 3 star kid who doesn’t contribute. It doesn’t make the news.

              Conversely, for every 3 star kid who becomes an All American, there are a bunch of 5 star kids welcoming him to the club.

              The data states that the stars of recruiting services are valuable predictors. Nobody said they were foolproof. They’re just valuable.

              • What fresh hell is this?

                The “data” is flawed and full of bias.

                Like I said previously, I want as many 5 star prospects as possible, but the perception affects the results.

                UGA’s 2006 (#4 in country) is case in point.

                Also, all other things being equal, give me a pissed-off, jilted 3-4 star recruit with a chip on his shoulder and something to prove over a 5 star “I don’t have to put in the work because I’m already a star” guy any day of the year.

                • What fresh hell is this?


                  • Mayor of Dawgtown

                    Freshy, not sayin’ any names but the phenomenon described in your final paragraph above had a lot to do with the results on the field for the Dawgs the last 2 years.

                • This is another recruiting line we hear a lot. That’s the way the choice is usually presented: give me a scrappy 3-star over some prima donna blue-chipper.

                  But most of the top prospects are where they are precisely because of good habits, work ethic, and drive to go along with natural ability – and it all looks effortless to us. There is no diminishing level of desire, drive, or heart the higher you go in the ratings. It might be the opposite.

                • Hackerdog

                  So you think that recruiting rankings are useless, but you want as many 5 stars as possible? That’s logically inconsistent. If stars are meaningless, then a 2 star kid has just as much chance of contributing to your team as a 5 star kid. That’s obviously not true. The data proves that it’s not true.

                  So you can hang on to your anecdotes and believe that we can bring in a team of nobodies (how many 2 star kids live within 50 miles of Athens?) and still compete against the best recruiting classes in the country, but you’re not convincing me, especially in the face of actual data to the contrary.

      • Hogbody Spradlin

        What you say, and what most of ther other replies to the original comment say, makes sense. Another couple of ways of looking at the graph are: (i) a lot of recruiting money and effort is panning for small nuggets, and (ii) the graph is an exercise in statistical grouping, not scientific proof.

  2. Scorpio Jones, III

    Excellent piece by Hinton, and I am not surprised at the results of the number crunching. But does this mean if Georgia’s class is ranked lower than say, Bama’s or Florida’s classes the seamy underside of the traditional media (look at me when I am talking to you AJC) will view it as yet another reason Mark Richt is doomed?

    The problem with recruiting “services” (and I have never been sure who they are a service to, other than themselves) is that they, by their own admission, don’t have the staff or time to thoroughly investigate anything about a recruit beyond game stats.

    Which produces Willie Williams’ five stars and David Pollack’s three.

    Wouldn’t you love to know whether SEC coaches are even aware of the number of stars a player has? Or if they care?

  3. X-Dawg

    Some people have bigger fish to fry than worrying about a player’s ranking coming out of HS.

  4. ZDawg

    I love this. I can’t tell you how many times I’ve read on boards the equivalent of “give me a hard working 3 star recruit any day over a 4-5 flashy flashpants…”

    Ok most of the time they didn’t use the word ‘flashpants’

  5. D.N. Nation

    Yeah, but Paul Johnson let Washaun Ealey score, so.

  6. AlphaDawg

    The recruit rankings are simply a Human Hedgefund. Teams hedge thier bet that if they get a handful of 4 and 5 star players who turn out great, they can fill the holes remaining with good 2 and 3 star players to field a competitive team(Auburn). And the Pollacks of a class are just an added bonuses.

  7. Skeeter

    I loved that movie, Flashpants.

  8. Normaltown Mike

    Of course Georgia should want the best. But not all “4 and 5 stars” are created equal. For example, Tennessee prep football has long been known as a weak sister as far as southern football goes. Thus fans (and hopefully coaches) ought to take stats and stars with a grain of salt.

    In CMR’s case, there appears to be some diminishing returns of chasing hot shots from far off non football hotbeds like Maryland, Oakland and San Diego and glossing over players from Georgia.

    By my recollection, CMR’s only hotshot from a non football factory state is Knowshon. Am I missing one? I’m sure we can all name several big busts.

  9. And Pollack is ALWAYS the example. You’d think these people would notice that the phenomenon they’re advocating happens rarely, so that they have to use the same example over and over again to support the argument.

    • Scorpio Jones, III

      So, then, you are saying the Newbergians are always right and that Pollack is the only example of a 3-star who turned out to hit above his average?

      I don’t understand.

    • King Jericho

      Don’t forget that Bruce Miller dude from UCF this year. He wanted to be a bulldog super bad and we blew it! How could we have been so dumb!?

      • Scorpio Jones, III

        Or Tim Jennings.

      • Normaltown Mike

        Thanks for proving my point Jericho!

        in 2006 Georgia was clearly too busy focusing on a “the #2 Tight End nationwide” from Oakland, a “Parade All America” from Maryland and a 3 star 6′ 1″ DT from Cerro Gordo (really) to worry about the likes of lowly Jasper Brinkly, Casper Brinkley and Bruce Miller.

        Why recruit Thompson and Woodstock GA when you can head to Oakland, Maryland and Cerro Gordo!

        • Me

          The best I remember the Brinkley twins were recruited by UGA in high school,but wound up at MGC I believe. Afterwards one of them was again recruited by UGA, not sure why both were not. Maybe grades? They wanted to play together, so SOS and USCe apparently had no qualms about taking them both.

          • Sanford222View


            Georgia coaches are such idiots to miss on Miller. Thankfully for us he chose UCF over Florida, Bama, LSU, USC, and Texas.

            • Normaltown Mike

              Luckily Georgia “hit” on Akeem Hebron, Ricardo Crawford, Darius Dewberry, Michael Lemon and Brandon Wood.

              By my limited Georgia math skills, Garner and CMR are so amazing at talent evaluation that they offered 5 guys on defense in one class that have distinguished themselves with (in order): minor possession, nothing, nothing, assault and nothing.

              I’m not holding it against Garner and CMR for missing Miller. I’m holding it against them for missing on Akeem Hebron, Ricardo Crawford, Darius Dewberry, Michael Lemon, Brandon Wood AND Miller.

              • Mayor of Dawgtown

                I am holding it against them for the minor possession, nothing, nothing, assault and nothing.

  10. sUGArdaddy

    Well, I’d argue that New Jersey has pretty good high school football. They send a lot of players to Penn St and Ohio St.

    If a player is a stud from Florida, Georgia, Louisiana, Texas, California, New Jersey, Pennsylvania, and Ohio, your bets are pretty safe.

    If they’re from South Carolina, North Carolina, Alabama, Mississippi, Arizona, Virginia, Oklahoma there’s a good chance they played pretty good competition.

    If they’re from anywhere else…buyer beward. They just don’t don’t take it as seriously in those other parts of the world.

    There are always exceptions. But I like the fact that we’re turning more focus to Georgia. The truth is, I think Richt recruited Georgia hard when he got here and thought, “Hhmmm, I’ve figured out that I can win an SEC title w/ Georgia boys, but I’m not sure about a National Title. I think I’ll try to get some national guys to win the big one.” That backfired. I really think it was a conscious decision to try to win a national title and it didn’t work, and now he realizes that winning an SEC title will get you in the hunt most years, and we can do that with Georgia guys.

    I also think he realizes now that it did something to our program. Many of our stars might not have been ready to bleed for UGA because it wasn’t in their blood. 5 star guys are good. 5 star guys from Georgia are better. Just think about that ’08 team that was wildly talented and where those guys were from.

    Our 4 best offensive players were from Texas, Jersey, North Carolina & South Carolina.

    Our stud D-linemen were from Florida and Jersey. 2 of our starting DBs were from South Carolina.

    Does that make a difference? I don’t know. But maybe that stuff adds up after a while. Contrast that with the ’02 team which had Fred, Reggie, Stinchcomb, Gilbert, Boss, Greenie, and Pollack as leaders and all from Georgia.

    What happened between ’07 and ’08? Maybe it was because guys like Thomas Brown graduated? None of this stuff is science and every team has it’s own personality. But it getting the best of the best in Georgia is a move in the right direction and fit in a peice like Jenkins and LeMay when you need to. Oh, and Beat Boise!

    • Normaltown Mike

      I think the problem began in 2005 when CMR watched Georgia win the SEC and ink the best QB he’d ever seen. The combination of success on the field and in recruiting nationwide lulled him into thinking he had the program on the autopilot that Bowden had for over a decade.

      Sadly, he didn’t pay attention to two monstrous harbingers:
      (1) the fact that a first year coach in Gainesville was able to beat him despite not having “his” players (Georgia hid behind the injury of ONE player to explain this way)
      (2) the incredibly disgusting Sugar Bowl performance by the Willie Dawg Defense (CMR explained this away as a result of “playing in Atlanta for the third time in six weeks”.

      2006 was largely wasted on CMR’s square peg round hole playcalling and at any rate Bammer canned Shula and paved the way for a severe upgrade in Saban.

      2005 was his highwater mark*. In a little over a year from the SEC CG, CMR dropped Georgia from top of the conference to middle of the pack. I hope he turns it around, but I have strong suspicions he won’t.

      * a home loss to South Carolina and a 21 point loss to UT does not a great season make.

      • sUGArdaddy

        There’s probably a lot of truth to what you’re saying. I also always thought that things would have been different if Shock was one year younger or Matthew was one year older. Having Matthew redshirt behind DJ in ’05 or ’06 would have been priceless and we might have never missed a beat.

        Sometimes, it’s just not that easy.

        • Normaltown Mike

          I’d have to think Stafford going pro early was the biggest kick to the groin of CMR’s QB relationships.

          Yes, I know he was staring at big money. But (1) how often do QB’s from private schools jump and (2) how many pro style QB’s suffer career ending injuries in college?

          What’s more, he almost certainly knew he would be drafted by Detroit in 09. By spending his first contract years there, he might not have the opportunity to a second mega lucrative contract in his first bite of the free agency apple (assuming he is continually battered as he has been so far). Had he played 09 at UGA, there’s a good chance he’d be in St Louis right now (he would’ve projected better than Bradford). The chance for success b/w those 2 franchises is night and day.

          • ruteger

            I’m not sure how Stafford’s decision to leave can be questioned. He pretty much knew he’d be #1, but he could have a sub-par senior season that could result in him dropping like Leinart did. He knew he’d go to the sad-sack Lions, but he might go to them the following year too (they ended up picking #2 that year). He knew that if he stayed he could possibly go #1 the next year and make just as much or more in his initial contract, but he’d also be turning down one work year worth millions of dollars.

            • Scorpio Jones, III

              Stafford also had some sort of mystical connection to the Lions.

              He lived on the same street as Bobby Layne or something.

              • Dog in Fla

                And the Lions, in turn, had a mystical connection with Matt, the result of which was two season ending shoulder injuries.

  11. simpl_matter

    I’m shocked the % chance is not substantially greater for 5-star recruits. The two sample pools are so dramatically diffent in size, 171 vs 11,675! That means there’s one 5-star recruit for every sixty-eight (68) 2/3-star recruits.

    I GTP but, my take-away for those numbers is the importance of being able to evaluate the talent of sub 5-star recruits (like David Pollack). There were 50 5-star All-Americans in those numbers and 201 All-Americans that were less than 5-star. Sure, go after any 5-star you have a legit shot at but, if you are wasting your limited recruiting time & budget going after a guy you have a 25% to land, you are probably going to miss the boat on a few diamonds in the rough that you just didn’t evaluate thoroughly.

    • Hackerdog

      If you have a 25% chance of landing a 5 star recruit, and that kid is 700% more likely to be an All American than a 2 star kid you have a 95% chance of landing, you should go for the 5 star kid. That’s almost like turning down a free lottery ticket.

  12. JBJ

    Sure everyone would love a team full of 5 stars. The problem is that when you get to the 4, 3, and 2 star athletes, the picture becomes much less clear and the hooting and hollering over this 3 star guy or this 4 star guy is just ridiculous.

    I would love to see a personality on the 2, 3, and 4 group to determine if there is some aspect which separates the 1-2% from the rest.

  13. This is one of those arguments that has more annoyed me than anything else as a Dawg fan the last several years. The old “I’d rather have guys that love Georgia (i.e. 2/3 star recruits from in-state) than the guys that have their eyes on the NFL (i.e. the superstud 4/5 stars out of staters such as AJ, Knowshon, Stafford)” argument. While having guys that wanted to bleed red and black their whole lives and will bust their tails to get there is nice, you win championships with the AJs, Knowshons, and Staffords of the world. Granted, I think this argument is more born out of those that felt that Stafford owed something to them and expect loyalty from guys like him while showing none of their own. As the old saying goes “It ain’t about X’s and O’s, it’s about the Jimmy’s and Joe’s”.

  14. Senator,
    David Hale did some nice work on this topic last year. Interesting read and along the lines of your take.
    One of the many reasons I miss DH… http://ugadogsblog.blogspot.com/2010/02/fun-with-numbers-from-promise-to.html

  15. ChemDawg


    The “Pollackers” will say “all Americans are great. How about wins? Everyone knows 4,5 star players are “talented”, but are they team players, do they practice hard to implement the game plan. All Americans that lose games are irrelevant.” Fair point.

    Compare stars to wins and championships. Here are the last 4 champs with stars (5,4,and 3 respectively) in the four seasons leading up to and including the championship year from Rivals (includes oversigning):
    Au – 5, 39, 60 *
    Al – 9, 53, 40
    FU – 12, 53, 25
    LSU – 8, 50, 31

    Georgia has 5 five star and 73 four 73 three over the last 6 years combined. Depressing right, in terms of five stars, but I don’t see nearly as many 3 star guys as we seem to have except in the case of Au (which is why they have an asterisk…among other reasons)

    On the other hand…USC (real USC) has 18 five star recruits over the same six years. Hasn’t helped them.

    Given Auburn’s championship among this small sampling, there may be some validity to a team of 3 stars that are “All in” may indeed give odds toward a championship season if not as well as stocking up on rediculous talent.

    Expanding this methodology beyond SEC NC teams is tricky because the SEC champ was not in the title game and therefore a true national champ was not crowned.

  16. Chrisfrmatl

    Star rating has ALOT to do with offer list and attention from big schools…

  17. Derek

    Newflash: Kids who are really good in high school tend to be good in college. Guys that are really good in college tend to be pretty good pros.

    I don’t think the NFL guys are going around saying: Hey lets not take the hotshot QB out of college because we may get a Ryan Leaf, lets go get a guy in the sixth round as he’ll no doubt become the next Tom Brady.

    It is stupid to suggest that any talent evaluator can find enough “diamonds in the rough” to remain competitive with those who always take those who, on the surface, appear to be the best players. I’ll take my chances with as many 5-stars as I can get. Yes we all were disappointed in Miller, but AJ turned out decent didn’t he. And for every #47 who turned out great I can name you a dozen other 3 stars who never contributed.

  18. Scorpio Jones, III

    So what we do is get the best kids we can get, hope they qualify with the NCAA, and in our specific case, the “faculty-administration approval committee” (Bet they ain’t got one of them at Bammer.), then hope we can coach them up.
    “faculty-administration approval committee” is not the real name of this group….can’t remember what we call it. I have a tweet in to Jamar Cheney, but he has not answered yet….proly out looking at new Porsches or something.

    Meanwhile the Newbergians are telling everybody who wants to know our class sucks compared to Bama, or that Bama got the guys we really wanted, and the AJC is wailing Richt is doomed.

    You know, coaching college football team is not an easy job, in spite of what wisdom you may read here and there.

  19. Dog in Fla

    “Or does anybody really think Nick Saban does what he does because he wants to finish first in Rivals’ recruiting rankings?”

    I think it’s because Nick wants a first place finish on Rivals. After all, since running Urban out of the league, spitting the bit against Auburn, thumping Michigan State to make up for the Utah Sugar Bowl loss and the consecutive championships on oversigning.com, what other competition is left?

  20. shane#1

    The problem with recruiting is that no one can look at an 18 year old kid and tell how he will turn out. Greg Blue lived out of trash cans and slept on the street untill his HS coach took him in. Greg went on to become a great Dog and to graduate. Mett was raised in a good middle class home and was kicked out of UGA. Even so, I would think the four and five star kids from good programs who happen to be the biggest and fastest would surely help a coach hedge his bets. However, there are many great athletes playing for small schools down here in South Georgia that never get noticed. Does Thomas Davis ring a bell?

  21. Mayor of Dawgtown

    The real problem is misclassifying. Obviously someone f#cked up when classifying Pollack–he was way better than the rating services thought. On the other end of the scale you have a Bryan Evans. Some schools and their respective coaching staffs place too much emphasis on what some pencil-neck who never wore a jockstrap thinks about how good a player a kid is, IMO. The hard work is identifying for yourself the players that are the right ones to recruit. It is just too easy to rely on Rivals and go along with the herd chasing a wideout labeled a 5 star recruit just because he can run a 4.3 forty when in fact he gets alligator arms as soon as he is hit the first time. I almost wish these services didn’t exist.

  22. 69Dawg

    To me the biggest problem is the level of competition that the kids play against. In the world of midgets the 5 foot guy is king. I’ll take 3 stars from 6A any time not so much from 2A. Look at Mr Brown, he was all world in the Christian league in Memphis but he has struggled in college cause he is up against the best. We have seem to go after some guys from small schools and don’t say anything about Walker because we all knew he was a beast.

  23. Vinings Dog

    I think we should add another dimension to this debate, and that is how many starts do you get out of a five star player. Florida got to use Tebow for four years, and, although he did not start in 2006, he definitely contributed to that team’s sucess. KM played a little less than two years for us, and Stafford less than a full three. Green, with injuries in ’09, suspension in ’10 and the NFL in ’11, played just over two years.

    We talk about all the talent we have had, but we have not had it on campus long enough. Knowshon could have been a fifth year senior THIS YEAR.

    I wish someone would do some analysis on this – adding up the amount of playing time a great player had at his alma mater.

    • Mayor of Dawgtown

      Correct VD. In the new era of college football you better play ’em while you can because the good ones will leave. Hence, redshirting should not be used for skill position players. It’s OK when it gives an O-lineman or D-lineman a chance to physically mature plus they do not tend to leave early as much. Otherwise, the kiss of death. I hope CMR has learned this lesson. He sure should have by now.

      • Normaltown Mike

        I agree. OL and QB are obvious redshirts. Any other position is a mistake. CMR’s decision to RS Knowshon is now looking more and more like Goff’s treatment of Robert Edwards as a DB for 2 years – and that’s not company you want to keep.

        There is also the added advantage of washing out the disappointments faster. We’ve got several guys that never made the field or only in mop up duty but b/c CMR redshirted them, they are stuck here for 5 years.

        I’m not suggesting we give em the Saban treatment, but I’m guessing it would be 2 or 3 new spots a year that would open up by hurrying the process along with the deadweight.