http://www.basketballanalyticsbook.com/2015/03/13/updated-college-prospect-ratings-cpr/
Russell: 10.9
Okafor: 9.2
Looney: 8.2
Turner: 7.1
Towns: 5.8
Jones: 5.6
Wood: 5.6
Portis: 5.5
Johnson: 5.1
Hunter: 4.7

A player above 10.0 has never been a bust since these stats were recorded (other than Oden who played well but was always injured). Towns has rapidly climbed throughout the year.

Box score analytics?

VCoug wrote:Nice to see Towns moving up their board. I wish they had a historical database so we could compare previous drafts and players.

So do I. Stephen Shea is doing the calculating himself and apparently it takes a long time to go back to get all the data from players.
He does discuss previous players' data here and in many other articles though:
http://www.basketballanalyticsbook.com/2014/12/11/midseason-college-prospect-rating-cpr-for-2015-draft-class/
Towns is moving up but he's still way behind Okafor and Russell. In the end, it's just one more source, not the definitive source.

Anthony Davis was one of the highest ever. Note that he's in the 10+ group. Shea describes the full formula in his book, though. He has a PhD and I can't imagine that he would forget that the data need to be adjusted for playing time. He has written that it's hard to interpret Towns' #s for the reasons you described.

yellowboy90 wrote:

Bonn1997 wrote:Anthony Davis was one of the highest ever. Note that he's in the 10+ group. Shea describes the full formula in his book, though. He has a PhD and I can't imagine that he would forget that the data need to be adjusted for playing time. He has written that it's hard to interpret Towns' #s for the reasons you described.

Maybe I am not reading it right or he adjusted it. I'm just trying to get a better understanding.

As suggested above, CPR downplays inconsistency and focuses on elite potential as captured through elite college performance.

CPR uses the following inputs: projected NBA position, age, class (freshman, sophomore, etc.), 3PA, 3PM, FTA, FTM, ORB, DRB, AST, STL, BLK, and PTS. For the performance stats (3PA, 3PM, etc.), full game-log data is used.

This is why I asked about if less minutes impact his rating.

OK - then I don't know the answer to your questions. Where did you get that table from? Was that at the end of the season? Something doesn't look right. Maybe there's an adjustment. He's listed AD as 10+ in all the examples I've seen.

yellowboy90 wrote:

Bonn1997 wrote:

yellowboy90 wrote:

Bonn1997 wrote:Anthony Davis was one of the highest ever. Note that he's in the 10+ group. Shea describes the full formula in his book, though. He has a PhD and I can't imagine that he would forget that the data need to be adjusted for playing time. He has written that it's hard to interpret Towns' #s for the reasons you described.

Maybe I am not reading it right or he adjusted it. I'm just trying to get a better understanding.

As suggested above, CPR downplays inconsistency and focuses on elite potential as captured through elite college performance.

CPR uses the following inputs: projected NBA position, age, class (freshman, sophomore, etc.), 3PA, 3PM, FTA, FTM, ORB, DRB, AST, STL, BLK, and PTS. For the performance stats (3PA, 3PM, etc.), full game-log data is used.

This is why I asked about if less minutes impact his rating.

OK - then I don't know the answer to your questions. Where did you get that table from? Was that at the end of the season? Something doesn't look right. Maybe there's an adjustment. He's listed AD as 10+ in all the examples I've seen.

I got that from the older posts. I belive this is from his introductory post of CPR.

http://www.basketballanalyticsbook.com/2014/06/04/college-prospect-ratings-cpr/

Also, I did a little more research and...

http://www.basketballanalyticsbook.com/2015/02/20/updated-midseason-college-prospect-ratings-cpr-russell-and-okafor-on-top/

While we’re on Towns, the remarkable collection of talent in Kentucky diminishes all of their players’ CPR scores. CPR looks for transcendent performances, instances where players have taken over college games. Kentucky doesn’t need any 1 of their players to ever do that. (I’m currently working on possible revisions to CPR that would better account for a situation like Kentucky’s.)

Yeah, I noticed that too. Shea readily discusses the strengths and limitations of all his methods.

i'm not a big fan of his methodology... it certainly is interesting... i'm of the opinion that bad games do matter and a gunner who shoots a ton but has high variance is going to look really good according to this .. which is why rj hunter ranks so high...

an elite player is pretty obvious to suss out even by casual observation... the box score will always reflect this of course so it's not something that proves it one way or another...

players bust for all sorts of reasons... adjusting for class is probably the best part of his methodology.... as a freshman, everyone you play against is older than you... as a senior.. everyone is younger... but you also need to go a bit further and adjust for age and a lot of it isn't readily available unless you're closely following a prospect... oj mayo was actually 20 as a freshman... tim duncan was 16... that's a HUGE difference...

i like weiland since he also does positional adjustments and i think that's the blind spot for all these other draft projections... shea's projections and you can include pelton's in there also... use the same stats weighted the same across positions... this tells you how good they are now but doesn't do well in how they will translate to the pros at their position... assist to TO means one thing for a center and means something else for a pg for example... that's the major reason why his projections are very accurate and he's been doing this for over 10 years...

i've been trying to run my own analysis using a part of weiland's methodology but accounting for age and strength of schedule... the most difficult part is getting accurate data from colleges since there isn't reliable data in one spot.. especially minutes played data... so i definitely sympathize with shea and his issues...