the Polo Grounds

Back to the forum index

some more on the NFL draft & Rivals ratings

by Jay ⌂, San Diego, Tuesday, April 19, 2011, 13:27
edited by Jay, Thursday, March 08, 2012, 18:12

We had a nice thread going a few weeks ago about Rivals star ratings and the NFL draft. I'd crunched some numbers here:

Wanted to get back to this and put up some more data. Here's one cut of it. Open to other suggestions on how to slice'n'dice it.

Here are the ratings of players drafted from 2007-2010, broken down by positions. (The second tab has some preliminary stuff on looking at it by round.)

I'll keep adding to this.



here's an interesting view on it

by Jay ⌂, San Diego, Wednesday, April 20, 2011, 09:19 @ Jay

If you line up the average rankings by position, you get this, from highest star rating to lowest.

Pos	Avg Stars

QB	3.16

all DL	3.13	(DT 3.23, DE 3.05)

RB	3.07

WR	2.98

LB	2.98

TE	2.87

all OL	2.80	(OT 2.85, OC 2.85, OG 2.69)

DB	2.80

This probably dovetails with some other similar type of studies, but it shows that Rivals is better at identifying potentially good QBs and DTs moreso than other positions.

Is this because the potential of QBs and DTs are more "evident" earlier on?



that aligns with our yearly BGS recruiting post

by Pat, Right behind you, Wednesday, April 20, 2011, 10:09 @ Jay

(from 2008)

2. The hardest positions to evaluate appear to be the offensive line and cornerback.

Michael made this point two years ago, and so far it has generally held true. Look through the offensive lineman on this list, as well as the previous two lists and you see a very mixed bag of starters, career backups, and victims of the injury bug. Of course this is the case with all positions, but moreso with offensive lineman. The only real OL "star" from these lists is Justin Blalock, although Herman Johnson and Jeff Byers still have the chance to really make a name for themselves. (And while this is more trivia than an indictment of these rankings, OL is the only position that has failed to generate a 1st round NFL pick from all of the players and positions represented on the past three Top 50 lists.)

Cornerback rankings are a bit more precient than OL, but nowhere near what you would expect for a top 50 list. Each year a number of guys are billed as the next great lockdown corner, but they rarely pan out in college. Is it that hard to evaluate cornerbacks? Do high school highlight tapes accurately showcase a cornerback's skills? Or, perhaps, are the great college corners actually playing quarterback and wide receiver in high school?

3. Keep your Top 50 defensive tackle eligible, and he's likely to be a star.

There's really no middle ground: the defensive tackles in the Top 50 either turn out to be All-American caliber stars, or complete busts (due in large part to academics or off-the-field issues). In a way, I suppose that supports the theory that defensive tackle is one of the easiest positions to rank. There are only so many kids each year that can run down quarterbacks while outweighing them by 100 pounds. They tend to stick out like a sore thumb on recruiting tapes and that makes them easy to pin near the top of the rankings. It seems that if a program can keep a Top 50 tackle hitting the books and out of trouble, odds are he'll live up to his billing.


Chart of relationship between recruiting and the draft

by suave_andrew, Wednesday, April 20, 2011, 01:55 @ Jay

Made this about a year ago. This chart simply looks at the the correlation between a long-run amount of players drafted out of a program vs that program's avg star-rating over that time. The y-axis is just the total, or aggregate, number of players taken out of that program - which round they were taken isn't looked at.

It's an average of the four year averages and if I were to do that again, I'd just use a weighted average instead as it'd be more accurate. Either way, I still think it's quite descriptive: how well a team recruits (going by avg star rating) has a near direct and statistically significant relationship with number of draft picks they produce. Which makes sense from an input vs output point of view. I believe the variance comes from how well coaches develop that recruited talent over the course of time; or to put it another way, the value they add to their inputs.


charts, draft


original source

by Jay ⌂, San Diego, Friday, July 15, 2011, 10:53 @ suave_andrew
edited by Jay, Tuesday, February 26, 2013, 13:13

Didn't want to lose the link to this fine post of yours on Clashmore Mike.

and a timely excerpt:

This second graph shows the trend when you combine the number of draft picks from CMU and Cincinnati. The red bars indicates Kelly's influence, and when combined with the overall trend, shows rather definitively that Kelly's influence can be largely attributed as a main driver in the increased number of draft picks.




great chart -- I remember this one

by Jay ⌂, San Diego, Wednesday, April 20, 2011, 06:58 @ suave_andrew

[ No text ]


(viewing suggestion)

by Jay ⌂, San Diego, Tuesday, April 19, 2011, 13:31 @ Jay

To collapse the Google headers in the embedded sheet, go to the "View" menu item and select "Hide Controls." That should make it more legible.

I don't know why Google Docs disabled the ability to hide them automatically when embedding, but they did.

387056 Postings in 33495 Threads, 205 registered users, 60 users online (2 registered, 58 guests)
powered by my little forum