Dr. Saturday - NCAAF

If you were to go back and review the projections for last year's All-Americans when they were but mere recruits, you'll probably notice something interesting: The rankings seem virtually random. As many of the nation's top players were rated two or three-star mediocrities out of high school as were rated four-star mustangs, and more lower-rated prospects are on the All-America teams than former five-stars. Of the 79 different players (excluding kickers and punters) were voted in some capacity to one of the five NCAA-recognized All-America teams last year, only 13 came into college as five-stars, the cream of the crop. By contrast, more than three times as many of those All-Americans -- 43, to be exact, more than half of the total -- were rated three stars or lower. The top three or four recruiting teams in the country usuallydo better than thatbring in stronger hauls than that, according to the gurus.

If you didn't know any better, you'd think those recruiting stars didn't correspond to future success at all, a theme you might become familiar with over the next week or so.

Fortunately, we do know better. If you look more closely, there's a very good reason for the distribution, beginning with the distribution of stars at the beginning of the process, according to Rivals' database of signees to I-A schools over the last five years:

I would hope that two and three-star players could acquit themselves well enough to produce a large number of stars, since they account for almost 90 percent of the players nationally. Again, using the rosters of the five NCAA-recognized All-America teams -- as voted by the AFCA, the Associated Press the Football Writers of America, the Sporting News and the Walter Camp Foundation -- the situation changes dramatically when you look at the All-America numbers in light of those ratios:

Maybe a raw ratio of 1 in 14 -- or even 1 in 12, or 1 in 10, or whatever it is after accounting for the early departures, injuries and academics that these numbers make no attempt to reflect -- isn't all that impressive; after all, that means far more elite recruits are falling short of their star-studded birthright than are reaching it. Still, if you look at those odds ...

... it must be very tough to go on portraying them as a crapshoot.

digg delicious
more

13 Comments

Post a Comment
  1. Steve
    1. Posted by Steve Thu Sep 03, 2009 9:25 pm EDT

    Report Abuse

    Nice post. I think it would be even more interesting (and a lot of thankless work) to look at honors besides being named All-American. A lot of All-Conference honors are given to players who are arguably all-americans or would be in other years. It would be nice to see how many recruits of various star-ratings achieve the different levels of conference level honors.
  2. jg
    2. Posted by jg Thu Sep 03, 2009 3:40 pm EDT

    Report Abuse

    Although I see your point (as a stat teacher) I feel the need to remind everyone that just because recruiting rankings and AA go hand in hand doesn't necessarily mean there isn't more at work here. To make a sweeping generalization basically all the 4 and 5 star guys go to the USCs, Floridas, etc of the world. You don.t find a lot of five starers on Miami of Ohio or the like. Once these 4/5 star guys get to USC, Florida etc they have access to better coaches, facilities etc. Couldn't that attribute to them making AA? Additionally just like with rankings the "name" factor plays a large role. The AA teams are less likely to include teams from the smaller conferences because of less exposure etc. Couldn't all that be playing into the situation too? Also things like the Notre Dame factor come into play where as soon as a player commits to ND he rises by at least 1 star where if the same player committed somewhere else he would stay the same star wise or even drop. It seems to me that it's more a vicious cycle that is impossible to separate.
  3. Taco
    3. Posted by Taco Thu Sep 03, 2009 3:59 pm EDT

    Report Abuse

    Top programs get top ranked players and top ranked players go to top rated teams. It always has been that way and never changes. Publicity from media leads to a lot of this situation, always has and always will. Just a continuous cycle.
  4. Flea
    4. Posted by Flea Thu Sep 03, 2009 4:12 pm EDT

    Report Abuse

    Why does everyone think recruits get an additional star just because they commit to a big-name school? For as long as I have been following recruiting (about ten years) neither Scout nor Rivals ranks players based on where they are likely to go to school or who they commit to. Players are ranked well before their senior year of high school and the ranking gets adjusted based on football camps and all-star games. I have yet to read once that players get a boost to their ranking based on the name of the school they commit to, or even the names of school looking at the player. The rankings are based on skill and talent that is derived from play, then they are compared to every other player at their position, and all of that is mostly received from scouts going to and watching these kids play.
    If people are going to continue saying "Also things like the Notre Dame factor come into play where as soon as a player commits to ND he rises by at least 1 star where if the same player committed somewhere else he would stay the same star wise or even drop", then please cite your resource because I can not find any information to validate that comment.
  5. gtne91
    5. Posted by gtne91 Thu Sep 03, 2009 8:43 pm EDT

    Report Abuse

    "I have yet to read once that players get a boost to their ranking based on the name of the school they commit to, or even the names of school looking at the player."
    Lance Richardson went from a zero star to a two star after committing to GT early this month. It wasnt the school name that did it, so much as they figured they needed to include him in the database since he committed to a BCS school.
    One thing I also know that happens - 1 star players are guys that are in the database but that hasnt been officially graded by the site (at least that is how Scout works, I assume Rivals is similar). If that player commits to Miami, OH maybe they get around to grading him, maybe not. If he commits to Miami, FL he will get graded, which will probably take him to at least a 2-star.
    If they want to be taken seriously, post commit grading needs to not happen. (Well, you know, accept for the early guys who commit before their senior year. I expect an update after their senior year).
  6. gtne91
    6. Posted by gtne91 Thu Sep 03, 2009 8:43 pm EDT

    Report Abuse

    I was doing a mini-study of 5-star guys in 2004 last week. I was using Scout's rankings and excluding juco 5-stars. IIRC, there were 31 5-star guys and I divided them into 3 categories - star, okay, bust. Star and bust are obvious, okay was guys who contributed, maybe even started a good bit, but was by no means what you would expect from a 5-star recruit.
    Anyway, while about 50% of them were stars, 30% (10, I think) were busts.
    While there is a clear correlation between stars and success (5-stars will on average due much better than 3-stars), the fact that you can have a 30% bust rate for 5-star guys suggests the filtering isnt that good. Yeah, there will never be a perfect system, Rhett Bomars is hard to predict, but Willie Wiliams wasnt.
  7. Flea
    7. Posted by Flea Thu Sep 03, 2009 4:12 pm EDT

    Report Abuse

    Yeah, that was my point. 1 star players have no rating, they just exist. I would agree with your statement that it wasn't the fact the school that got him the additional star directly, but more that once it was known he was being looked at by a major school, they probably went out and contacted the schools scouting reports to add to his profile. Once that is completed, according to what I have read, they then do a comparison of said player to all other players at the position, and adjust the ranking from there, which would lead to him getting somewhere between 2-5 stars.
    Like I said, the additional ranting is not a direct result of, as it is being stated on here recently, "just because is recruiting him."
    I think Scout and Rivals have it down pretty solid as far as how they rate players. I think commen sense should prevail when looking at these rating though. They should never be looked at as a definitive way of determining future sucess, but more of a loose barometer of overall player talent and potential compared to that players peers, within the same position. Obviously there are way more factors involved in determining the success of one player over another, i.e. coaching changes, offensive/deffensive scheme changes, injuries, player character off the field, etc.
  8. Flea
    8. Posted by Flea Thu Sep 03, 2009 4:12 pm EDT

    Report Abuse

    Also, gtne91, THWG!
  9. Steve
    9. Posted by Steve Thu Sep 03, 2009 9:25 pm EDT

    Report Abuse

    @gtne91
    I'm just wondering, what criteria did you use for "bust". Did injuries come into the discussion, because I can see how something like that, or some kind of off field problems, could contribute to the "bust" ratio. Would a guy like Bomar, who was good enough to start at Oklahoma before his off field problem, count as a "bust"? Especially given that he still had a decent enough career after leaving Oklahoma and still might be NFL bound?
    Remember also that with 5 stars only excluding JCs its a pretty small sample size, so it could just be randomness to a certain extent.
  10. Michael
    10. Posted by Michael Thu Sep 03, 2009 8:27 pm EDT

    Report Abuse

    Although the recruiting services deny it, I have to think that knowing USC, Texas, OU, LSU, and Florida have interest in a kid BEFORE they do their rankings becomes a factor. I don't mean to discredit the process entirely because I think it's about as good as we can hope for, but I do think there is a bias towards ranking players higher by default when their name is followed by USC. With that said, it may be a fairly safe assumption that if LSU likes a player, he's probably pretty good.
  11. gtne91
    11. Posted by gtne91 Thu Sep 03, 2009 8:43 pm EDT

    Report Abuse

    Steve,
    I counted Bomar as a bust. If you end up below FBS, you busted. Ditto injuries and arrests and etc. A bust is a bust. The borderline bust guy for me was Xavier Lee. I had him as a bust even though he started some and contributed some more. Just not enough to make it to okay. Many of the guys (especially outside the skill positions) I had never heard of and could only go from numbers. If you were a 5-star guy and played in 3 games (0 starts) in 4 years, you were a bust. Maybe it was injury? I dont know. You still busted.
    I also agree about randomness, I will probably end up doing some more years to build up a better supply of data. It was just the first I looked at, with the exception of one guy who was just granted a 6th year(and he went in the star category anyway - his 2 years as a starter [with all conference accolades] after his injuries was good enough for me), the 2004 class is done.
  12. B
    12. Posted by B Thu Sep 03, 2009 3:01 pm EDT

    Report Abuse

    For 5 star players, you would need to consider also that the Floridas and USCs of the world are sometimes recruiting multiple 5 star players at the same position, thereby making it less likely that they would all be named AA. USC signed to 5 star RBs in 2007, so assuming that one of them start enough to make AA, the other one is left out. There are some exceptions, but I not many.
  13. B
    13. Posted by B Thu Sep 03, 2009 3:01 pm EDT

    Report Abuse

    Sorry, that is supposed to be "two 5 star players".

Dr. Saturday

Add to My Yahoo! RSS

Matt Hinton

Dr. Saturday is a college football blog edited by Matt Hinton. Email him tips and feedback.

Related Photo Gallery

Y! Sports Blogs

Dr. Saturday Recent Readers