Started By
Message

GREAT Scientific Analysis on CFB Star Rating and NFL (LSU does pretty good)

Posted on 1/9/14 at 6:58 pm
Posted by MetTiger
Member since Oct 2007
1213 posts
Posted on 1/9/14 at 6:58 pm
Why College Recruiting Services Don't Work

After 1/2/14, I vowed to find a silver lining about the day's event. I came across this article that statistically proves/disproves some myths about how accurate the ratings services (247/Rivals,etc) reflect the real talent of HS players.

Here are some excerpts that pertain to a lot of comments on the TD board for LSU Recruiting (my comments in CAPS):

quote:

Since most players require one or two years of physical development before they are able to contribute on the field, they are better off identifying programs where they are most likely to start or contribute significantly by their sophomore season. Joining a talent-rich team is almost the surest way to never see playing time.
SINCE MANY ARE BUTT HURT ABOUT PLAYERS AT CERTAIN POSITIONS PICKING BAMA OVER LSU, THIS IS A VALID POINT


quote:

Even recruiting service defenders wonder if the recruiting services assign grades based on the services’ proprietary analysis, or if the services merely base grades on the quality of teams pursuing the recruits (i.e. players recruited by large schools such as Alabama are normally graded higher than players pursued by smaller schools).
CASE IN POINT - HOOTIE JONES FALLING FROM 5 TO 4 STARS UPON BAMA DECOMMITTMENT. OR THE FACT THAT ITS A WELL KNOWN FACT THAT ONCE A RECRUIT COMMITS TO BAMA, HIS STAR POWER IMMEDIATELY INCREASES. ON THE OTHER HAND, RECRUITS KNOW THIS AS WELL AND WANT THAT BAMA OFFER TO BOOST THEIR STAR POWER. AND IT DOESN'T HURT THAT BAMA BOOSTERS OWN SOME RECRUITING SERVICES.


quote:

How did these five-star prospects - the players whom the recruiting services considered the best of the very best players in the country - fare in the NFL draft? In a word, poorly. Over a seven year period, among the 200 prospects who received a five-star rating, just 30 were selected in the NFL draft’s first round. Draft Outcomes for Five-Star College Football Prospects (2002-2008) A more generous assessment shows that overall, from 2002-2008, the NFL drafted 53.5 percent of former five-star prospects over all seven rounds.
WHAT WE GET OUT OF THIS IS OF GW3/SN/TB/??, STATISTICALLY ONLY 1 OR NONE WILL GO 1ST ROUND. AND 1 WON'T GET DRAFTED AT ALL. OF COURSE ALL 3 COULD OR ALL 3 COULD BLOW OUT. STATISTICS Y'ALL.


Finally:
quote:

Getting into the NFL is the ultimate meritocracy. It does not matter which school a player attends. If a player has enough talent and determination, NFL scouts will identify him. Not all players may be drafted. After all, there are only 253 or so draft spots. But the best players, barring injury, will be able to forge careers in the NFL regardless of which schools they choose to attend
THIS GOES FOR EVERYONE, INCLUDING LSU. FROM SHEER NUMBERS STANDPOINT, LSU IS ONE OF THE TOP 3 SCHOOL FOR PUTTING PLAYERS IN THE NFL. THAT'S VERY IMPRESSIVE. FOR ANY PLAYER TO SAY THAT GOING SOMEWHERE ELSE GIVES THEM A BETTER SHOT TO THE NFL IS BS. IF ANYTHING LSU DOES, BUT STATISTICALLY IT IS A WASH.



*** My point is, of course we're butthurt when top local players choose to go out of state. On the other hand, the statistics prove that there is a very good chance that the player we replace them with (maybe at another position) may turn out to be better in the long run - even if it's a lesser star player.

I believe that the LSU football program sells itself and we are in generally good stead:

* 3 National Championships (2 BCS)
* Tiger Stadium (voted best stadium)
* Richest Football Talent State (esp for NFL)
* 4 consecutive 10 win seasons

Of course we need to constantly improve facilities and recruiting efforts. But the sky may at times be dark - but it's certainly not falling....





:geauxtigers:
This post was edited on 1/9/14 at 7:01 pm
Posted by MasterofTigerBait
Member since May 2009
7592 posts
Posted on 1/9/14 at 7:02 pm to
WHY SO MUCH YELLING????
Posted by Ironhead985
Louisiana
Member since Jun 2013
8714 posts
Posted on 1/9/14 at 7:04 pm to
Nice analysis, bro
Posted by RollTigers
Member since Dec 2010
3274 posts
Posted on 1/9/14 at 7:04 pm to
Caps lock much?
Posted by Stingray
Shreveport
Member since Sep 2007
12420 posts
Posted on 1/9/14 at 7:05 pm to
quote:

On the other hand, the statistics prove that there is a very good chance that the player we replace them with (maybe at another position) may turn out to be better in the long run - even if it's a lesser star player



:sigh:
Posted by bosoxjo13
Baton Rouge
Member since Jun 2008
3249 posts
Posted on 1/9/14 at 7:07 pm to
Interesting find. Good post. Sky is always falling on TD though
Posted by MetTiger
Member since Oct 2007
1213 posts
Posted on 1/9/14 at 7:18 pm to
As I said in the post, the CAPS are meant to differentiate from the article quotes. I didn't do the science, I just posted it.
Posted by novabill
Crossville, TN
Member since Sep 2005
10430 posts
Posted on 1/9/14 at 9:15 pm to
quote:

Morris Claiborne and Tyrann Mathieu were initially pretty low star ratings and Jacob Hestor also.


Scrubs
Posted by Victrola Wayne
Member since Sep 2013
104 posts
Posted on 1/9/14 at 10:26 pm to
Interesting read. Thanks for the effort. It's nice when a post here has some substance.
Posted by JakeFromStateFarm
*wears khakis
Member since Jun 2012
11888 posts
Posted on 1/9/14 at 10:28 pm to
I LIKE TO TYPE IN ALL CAPS TOO!!!!!!!




J/k. Good stuff
This post was edited on 1/9/14 at 10:29 pm
Posted by Kingpenm3
Xanadu
Member since Aug 2011
8952 posts
Posted on 1/9/14 at 11:06 pm to
Thanks
Posted by FLBooGoTigs1
Nocatee, FL.
Member since Jan 2008
54460 posts
Posted on 1/10/14 at 1:56 am to
Nice read and personally I think those recruiting services are full of shite anyway. It is a business and should be treated as that. I trust our coaching staff in evaluating talent.
Posted by glaucon
New Orleans, LA
Member since Aug 2008
5292 posts
Posted on 1/10/14 at 7:50 am to
Everything about those stats tells me that the Star system is working. 5 stars are drafted more often than 4 starts who are drafted more than 3 stars. Also, his analysis is shite. The football factory thing is perhaps the more ridiculous of all. You can't take how much of a percentage of the entire draft a school produces to see how effective it is at getting players into the NFL. You have to look at the percentage of players that play for the school go out to get drafted and play professionally. Those 8 Alabama players that got drafted (and probably a few others that were un-drafted free agents) represent about a third to a half of the whole graduating class of Alabama. His stats are demonstrating the opposite of what he is arguing.

Edit: Also, if you want to do a real "scientific" analysis why not judge the star system based on what it is actually suppose to judge in terms of contributions on the collegiate level. Compare stars to starts and stats in college not whether or not a player in drafted in the NFL.
This post was edited on 1/10/14 at 8:00 am
Posted by I20goon
about 7mi down a dirt road
Member since Aug 2013
12823 posts
Posted on 1/10/14 at 8:08 am to
quote:

glaucon
for the most part I'm agreeing with you... except for...
quote:

Edit: Also, if you want to do a real "scientific" analysis why not judge the star system based on what it is actually suppose to judge in terms of contributions on the collegiate level. Compare stars to starts and stats in college not whether or not a player in drafted in the NFL.
we now know that at least some sites justify rankings on "NFL potential". It would be nice is high school kids were only judged by their contribution at the next level (college), but that wrench is now in the works.
Posted by The Mick
Member since Oct 2010
43038 posts
Posted on 1/10/14 at 8:13 am to
Not sure what the point is here, but the article shows that 5 stars get drafted at a hihger % than 4 stars, then 3 stars, etc... (meaning 5 stars have the best college careers, then 4 stars, etc)

WHAT ARE WE SAYING HERE?
Posted by Woody
Member since Nov 2004
2452 posts
Posted on 1/10/14 at 8:39 am to
Look, I don't think star ratings are the ultimate predictor as to how a player will perform in college and ultimately the NFL, but that article talks in circles, and he ultimately proved that recruiting services are relatively accurate when you project the numbers out to the NFL draft.

The author's numbers actually prove the point opposite of the one he is trying to make. Using his numbers, there were 200 five stars over a seven year period out of 2.1 million players (300,000 per year as he assumed). Over 50% of those 200 were drafted, 53% to be exact. I would say that picking roughly 106 players out of 2.1 million that are likely to make it to the biggest stage is pretty good, especially when you consider the human element and the infinite number of variables to consider with these 18 year old kids.

However, the question as to whether players are rated highly based on the quality of teams pursuing the players is a valid one. I would argue that the credit for identifying the talent goes to the coaches and recruiters, not the recruiting services.

Stats bro.
Posted by doubleb
Baton Rouge
Member since Aug 2006
35822 posts
Posted on 1/10/14 at 8:49 am to
quote:

Edit: Also, if you want to do a real "scientific" analysis why not judge the star system based on what it is actually suppose to judge in terms of contributions on the collegiate level. Compare stars to starts and stats in college not whether or not a player in drafted in the NFL.


So a kid that starts for Tulane is good as a kid that starts for Alabama? I wouldn't think so 99 out of 100 times. So How can you grade a kid high just because he starts.

Players being drafted into the NFL and actually playing in the NFL is more of a judge of ability than just whether or not you start on a college team.

Posted by lynxcat
Member since Jan 2008
24118 posts
Posted on 1/10/14 at 9:32 am to
Pretty terrible analysis. Great example of using statistics to tell the story you want it to.
Posted by ShamelessPel
Metairie
Member since Apr 2013
12719 posts
Posted on 1/10/14 at 9:53 am to
quote:

Pretty terrible analysis. Great example of using statistics to tell the story you want it to.


Somehow, the small sample size of 5 star talent that goes to school = the chances you have of making the pros from that school.

I get how they could make the argument that the recruiting services know very little in reality. The argument that the school attended makes no difference in getting drafted is absurd.
Posted by glaucon
New Orleans, LA
Member since Aug 2008
5292 posts
Posted on 1/10/14 at 10:09 am to
quote:

So a kid that starts for Tulane is good as a kid that starts for Alabama? I wouldn't think so 99 out of 100 times. So How can you grade a kid high just because he starts.

Players being drafted into the NFL and actually playing in the NFL is more of a judge of ability than just whether or not you start on a college team.


The question is whether or not stars reflect the quality of a recruit and whether or not they will contribute at the collegiate level. Yes, any individual start for Alabama means more than a start for Tulane but in the aggregate of all college football players, I don't think it means all that much as we are just trying to find out if a higher ranking from a recruiting service reflects better performance at the college level. If 4 star players have more starts and better stats than 3 star players, it doesn't really matter where they play as for as the stats are concerned. That said, if you really want to be nit-picky through, I wouldn't be opposed to something like a qualifier based on winning percentage or something like that.
first pageprev pagePage 1 of 2Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram