NFL 1st round draft picks (2006-2010) as recruits

[dpodoll68]

In an effort to provide some empirical evidence that star rankings do matter, I conducted my own research regarding the 5 most recent NFL drafts, and where first rounders ranked as recruits on Rivals. I used the 2006-2010 NFL Draft 1st rounders, and since nearly all of those picks came from the 2002-2006 recruiting classes, I used those data sets as a reference point.

Before you view the link, there were some basic problems with the data:

- Eight of the 2006 draftees were recruited in 2001, before Rivals assigned star rankings. Given that several of those would have been highly ranked under the Rivals system (Matt Leinart, Jay Cutler, Michael Huff, Joseph Addai) their exclusion actually biases the results in favor of the "stars don't matter" crowd.

- The Rivals database is notoriously buggy technically, especially when you go further back. I was unable to get a count of total 2-stars for any year. I used 1000 for each year as a placeholder, which is likely a very low estimate. Further, I was unable to see exactly how many total 3-stars there were for 2003, as the website would not let me go past 1100 total prospects. Based on that number, there were at least 841, so that is the number I used. Again, both of these problems in data access bias the results in favor of the "stars don't matter" crowd.

- Some of the 2010 draftees were actually recruited in 2007. The number is sufficiently small, and given that I'm using a 5-year average, I feel the numbers will remain close even with disregarding the actual data from the 2007 recruiting class.

All of that being said, here is a link to my results below:

http://tinyurl.com/3t3ncs8

The most pertinent results:

18.06% of 5-star recruits were drafted in the first round
5.32% of 4-star recruits were drafted in the first round
less than 0.98% of 3-star recruits were drafted in the first round
less than 0.30% of 2-star recruits were drafted in the first round

Some other facts to consider:

- Only 2 unranked prospects (Dominique Rodgers-Cromartie and Clay Matthews) were drafted in the first round between 2006 and 2010.

- Only 1 walk-on (Matthews) was drafted in the first round between 2006 and 2010.

- Given that there are roughly 4,080 draft eligible DI-A prospects available each year (34 seniors/juniors per squad * 120 DI-A teams), the odds of being selected in the first round (32/4080) in any given year is approximately 0.78% for draft-eligible DI-A players. Given that 18.06% of 5-star recruits were selected in this five-year period, that means the average 5-star draft-eligible player was approximately 23 times more likely to be selected in the first round than the average draft-eligible DI-A player.

 

[Kuato Lives!]

Great work.

Not sure it really matters but Jason Pierre-Paul was upgraded to a 4 star as JUCO.

 

[Gopherphile]

based on your posts here, i know you consider yourself a smart guy. as such, i'd recommend that you look at your numbers in a different way such as what % of of 1st/2nd rounders were five stars, four stars, etc. the rationale is there are many times the number of 3 star recruits than five star so of course their success rate is going to be diminished. the weight of that denominator needs to be taken into account.

 

[dpodoll68]

based on your posts here, i know you consider yourself a smart guy. as such, i'd recommend that you look at your numbers in a different way such as what % of of 1st/2nd rounders were five stars, four stars, etc. the rationale is there are many times the number of 3 star recruits than five star so of course their success rate is going to be diminished. the weight of that denominator needs to be taken into account.

The data is there to extrapolate however you wish. The numbers you suggest (raw number of 5-stars drafted vs. 3-stars drafted, for example) are meaningless when taken out of context. The fact that there are many times more 3-stars than 5-stars helps to prove the point. If ranking recruits were "strictly for entertainment" and analogous to throwing darts at a board, there would be just as many 2-stars drafted as 3-stars, 4-stars, 5-stars, etc. in any given year, let alone over a 5-year period. The numbers are startingly consistent; number of 5-stars drafted each year: 5, 5, 5, 6, 5. The number of 3-stars drafted each year: 4, 7, 7, 6, 7. Keep in mind that there are approximately 22-23 times as many 3-stars as 5 stars in any given year, yet the raw numbers remain roughly equal.

 

[Killjoy]

dpodoll68

Unfortunately all your excellent work will probably not change anybody's mind. You are going to need to teach a statistics class here if there is any hope of doing this.

The fact is that Las Vegas makes a very good living with odds MUCH SMALLER than the difference in the odds of being drafted by the NFL for five, four, and three and two star recruits.

No people, the star system doesn't guarantee anything but the correlation between the star rankings is very significant. Is it perfect? Of course not. That why coaches do their own analysis.

Since it is a long time before fall camp, you non-believers can continue to argue why this is not true and bore the hell out of the rest of us. Denial is not just a river in Egypt.

 

[Bob_Loblaw]

Good work Dpo, I did a much more lazy attempt a week ago just using this years draft top 15 prospects because I didn't want to go through this but you did a really nice job. This is exactly what I was trying to illustrate, but you did a much better job than myself. Using your data, here is another way to look at it...

If you were a 5 star player you had a 17.7 times better chance of being drafted in the first round than a 3 star kid.

 

[BisonGopher]

In an effort to provide some empirical evidence that star rankings do matter, I conducted my own research regarding the 5 most recent NFL drafts, and where first rounders ranked as recruits on Rivals. I used the 2006-2010 NFL Draft 1st rounders, and since nearly all of those picks came from the 2002-2006 recruiting classes, I used those data sets as a reference point.

Before you view the link, there were some basic problems with the data:

- Eight of the 2006 draftees were recruited in 2001, before Rivals assigned star rankings. Given that several of those would have been highly ranked under the Rivals system (Matt Leinart, Jay Cutler, Michael Huff, Joseph Addai) their exclusion actually biases the results in favor of the "stars don't matter" crowd.

- The Rivals database is notoriously buggy technically, especially when you go further back. I was unable to get a count of total 2-stars for any year. I used 1000 for each year as a placeholder, which is likely a very low estimate. Further, I was unable to see exactly how many total 3-stars there were for 2003, as the website would not let me go past 1100 total prospects. Based on that number, there were at least 841, so that is the number I used. Again, both of these problems in data access bias the results in favor of the "stars don't matter" crowd.

- Some of the 2010 draftees were actually recruited in 2007. The number is sufficiently small, and given that I'm using a 5-year average, I feel the numbers will remain close even with disregarding the actual data from the 2007 recruiting class.

All of that being said, here is a link to my results below:

http://tinyurl.com/3t3ncs8

The most pertinent results:

18.06% of 5-star recruits were drafted in the first round
5.32% of 4-star recruits were drafted in the first round
less than 0.98% of 3-star recruits were drafted in the first round
less than 0.30% of 2-star recruits were drafted in the first round

Some other facts to consider:

- Only 2 unranked prospects (Dominique Rodgers-Cromartie and Clay Matthews) were drafted in the first round between 2006 and 2010.

- Only 1 walk-on (Matthews) was drafted in the first round between 2006 and 2010.

- Given that there are roughly 4,080 draft eligible DI-A prospects available each year (34 seniors/juniors per squad * 120 DI-A teams), the odds of being selected in the first round (32/4080) in any given year is approximately 0.78% for draft-eligible DI-A players. Given that 18.06% of 5-star recruits were selected in this five-year period, that means the average 5-star draft-eligible player was approximately 23 times more likely to be selected in the first round than the average draft-eligible DI-A player.

First of all, thank you for doing this. I appreciate your effort and I also appreciate how open you were about your methodology and the artifacts you observed in compiling the data into an information set.

Now, that said let me just first focus on the essential "punch line" of your post: "Given that 18.06% of 5-star recruits were selected in this five-year period, that means the average 5-star draft-eligible player was approximately 23 times more likely to be selected in the first round than the average draft-eligible DI-A player."

OK. Yes, ok - as in, I accept your hypothesis as solid theory. I am not going to verify your research, I will accept that it is correct.

It appears that if some recruiting service labels you as a "5-star" recruit, this eventually means you have a very significantly increase likelihood to be drafted in the first round than the overall draft-eligible pool.



Now....can you tell me what this has to do with winning college football games?

I assume, given your premise, that the more first round draft picks a team has on its team the more games that team will win.

How did that work for Alabama last year? #3 Dareus, #6 Jones - Ingram still on the board. 2010 results: 5-3 conf record, 4th place in their division.

Hmm...

 

[STPGopher]

Good work Dpo, I did a much more lazy attempt a week ago just using this years draft top 15 prospects because I didn't want to go through this but you did a really nice job. This is exactly what I was trying to illustrate, but you did a much better job than myself. Using your data, here is another way to look at it...

If you were a 5 star player you had a 17.7 times better chance of being drafted in the first round than a 3 star kid.

*Note: Don't read if only want to discuss stars only.

Maybe an alternate view could be "all-star" players verse players that fit systems. I think a good fit is equally if not more important than if a player is all world or not. Many examples of that

Another thing is Collegiate success doesn't necessarily mean you will be a great pro. Way too many reasons for this. Injuries being near the top. I won't discuss that point here except to say that judging a college player by draft status alone may not be the best thing to do.

I know that the initial point was and is that stars matter. I don't necessarily disagree. I've just seen too many instances over the years where "mediocre" players that were in a system that allowed them to shine played very well. I also saw great players in the wrong system for their talents stink.

Applying this to our current Gopher program, depth and fit are more important right now. Quality depth(meaning not a huge drop off from 1st to 2nd/ 3rd string can be huge. Example. Give a coach 8+ DL that are decent players that can be substituted freely during the game, and they more times than not will be much better than 3 to 5 all-stars. Quality depth alone over time will help this program greatly. Think of all of the times our D faded in the fourth quarter. It wasn't because they sucked, it was often because they were sucking wind. They were tired! Strength and conditioning can only go so far.

So what is my point in saying this? Depth and fit more than stars is important for this program right now and more attainable. It is my humble opinion that it will be difficult to recruit a full roster of 4 and 5 star players right away. Do we still recruit 4 and 5 star players? Yes absolutely. However even with 4 and 5 star players they need to be a good fit. Plus, there needs to be enough of them to matter.

So what do we do in the meantime? Get the best program players. Coach them up within the system and our program will be headed in the right direction. IMHO this would be a good use of resources. By the time our program gets to fielding teams with solid depth accross the board,our Gophers may be in the position of having multiple 4 and 5 star players that fit and want to sign a Minnesota LOI.

 

[DarrenTheGreek]

I did something similar for a couple years by looking at the all Big Ten 1st and 2nd teams and came to the same conclusion. Here's a copy/paste from part of my findings from the 2008 all Big Ten first and second teams

*****************
Now look at the recruiting totals in the Big Ten from 2004-2008:
Year: 5/4/3/2
2004: 5/35/97/123
2005: 6/38/108/82
2006: 5/50/92/95
2007: 5/43/106/70
2008: 4/48/114/85
TOTAL: 25/214/517/455

5 star recruits made up 10% of the all Big Ten team but only 2% of the total recruits.
4 star recruits made up 17% of the all Big Ten team but only 18% of the total recruits.
3 star recruits made up 50% of the all Big Ten team but only 43% of the total recruits.
2 star recruits made up 23% of the all Big Ten team but only 37% of the total recruits.

It's possible for a 2 star recruit to break through and become a stud but it's far less likely.

 

[Goldmember]

dpdoll, I think you would agree that a program like Minnesota (or Iowa, or Wisconsin for that matter) would dilligently pursue any 4* or 5* recruit alleged to have an interest. For almost every program (Texas and maybe a couple others notwithstanding), these are bluchip recruits. No-brainers.

The recruting classes of the rest of the BCS-conference programs are comprised primarily of 3* and 2* recruits. Accordingly, at least 95% of the value of the data should be determined by how well the services can distinguish between a 3* and a 2*.

I am still unsure how well are these services able to determine who really deserves to be a 3* and who does not. But for most FBS-programs this would seem to be what matters the most when it comes to evaluating the talent of recruits.

 

[josh087]

DP, your "statistics" are totally meaningless because almost nobody recruited Eric Decker or Marion Barber III and they turned out to be great, and every not-highly-ranked High School player in Minnesota is just like these kids and only needs a chance. We should only recruit Minnesota High School players, we are the best!!! Lock down those borders and stop going after highly ranked outstate players!!!

 

[RodentRampage]

Now....can you tell me what this has to do with winning college football games?

I assume, given your premise, that the more first round draft picks a team has on its team the more games that team will win.

How did that work for Alabama last year? #3 Dareus, #6 Jones - Ingram still on the board. 2010 results: 5-3 conf record, 4th place in their division.

Hmm...

You seriously want to go with that?

Alabama was 12-2, ranked #6 in 2008; 14-0, national champions in 2009; and 10-3, ranked #10 in the nation in 2010. Alabama has done awfully well in winning games over the years.

 

[RedPoo]

You seriously want to go with that?

Alabama was 12-2, ranked #6 in 2008; 14-0, national champions in 2009; and 10-3, ranked #10 in the nation in 2010. Alabama has done awfully well in winning games over the years.

I'm not arguing for any particular point here, partially because I have no idea what point is supposedly being argued in this thread. But, as an exercise: As I'm writing this, it looks like we're through two rounds of the draft. Anyone want to guess how many players have been drafted from Oregon, who played in the national championship game last season? How about Clemson, who finished under .500, and exactly .500 in the mighty ACC?


Also, stars obviously generally mean better players. Anyone who doesn't understand that probably doesn't understand how toast works.