"My rule was I wouldn't recruit a kid if he had grass in front of his house.
That's not my world. My world was a cracked sidewalk." —Al McGuire

Sunday, July 29, 2012

Prep Rankings Great Predictor of NBA Future, but Value Add Finds the NBA Surprises Early

This year 12 players were drafted by NBA teams despite not being rated as one of the best high school players out of high school. To continue to hone the system of ranking college players (www.valueaddbasketball.com) and projecting if they can make it in the NBA (www.valueaddbasketball.com/draft), Rob will continue to factor in excellent work such as the precise measurement of how much players will improve every year until they are 25 (http://wagesofwins.com/2012/07/25/age-is-just-a-varible/).

In addition, I did a study this weekend to determine how common it is for a player not in the RSCI top 200 out of high school to get drafted, and reached two conclusions:

1. High school rankings are a very good indicator of whether or not a high school player will be drafted by the NBA.

2. Value Add is very good at spotting unranked high school players early in their college careers when they start to show NBA potential.

First, here is the table I built breaking down the 770 players drafted since the RSCI rankings started in 1998 to show just how good an indicator a players high school ranking in determining if he will be drafted. I use the database at http://www.draftexpress.com/RSCI/, which differs slightly from the original http://www.rscihoops.com/:

RSCI RankTotal 1998-2011Drafted% chance drafted
1 to 912611188% (will be 97%+)
10 to 3029413245%
31 to 401403122%
41 to 704206515%
71 to 1004204410%
101 to 140560428%
141 to 20031783%
Others (est.)104001111%
Foreign151
Pre 1998 Class75
Total drafted 1999 to 2012770

In fact, players ranked in the top 9 of their class almost always makes the NBA. As indicated above, 88% have been drafted. However, 66 of 72 have been drafted since 2003. That number is almost sure to improve to 69 of 72 when James McAdoo, Adonis Thomas and Bryan Nash go early in the 2013 draft. In addition, Samardo Samuels wasn’t drafted but was immediately picked up and has played two solid seasons with the Cavs to really make it 70 of 72, or 97%. That only leaves borderline Kenny Boynton (projected 52nd pick next year), who could make it 71 of 72. It is likely the out-of-shape and troubled Mississippi State center Renardo Sidney will be the only top 9 prospect since 2003 not to make it to the NBA.

RSCI players ranked 10th to 30th are basically a 50-50 shot to go pro, as several such players from the last few classes are still in college and should get the figure up a little from the current 45%.

If you are not rated in the top 30 out of high school, you are really a long shot. Even just going to the next 10 spots – the 31st to 40th ranked players, have only a 22% chance, and so forth, down to just over 1% of the players who are not rated at all in the RSCI but still make it to the draft.

Finding the 1% - Value Add Spots Them Early

So there is plenty of talent to get among the unranked players, but you don’t have time to watch 10,000 players to figure out which ones are really good enough to go to the next level. This entails first trying to evaluate what their stats mean despite them often playing against much weaker opponents if they do not play in a BCS Conference, and then determine if that ability translates to the NBA level.

We recommended picking eventual 2nd round/non-RSCI players Bernard James, Jae Crowder, Orlando Johnson, Marcus Denmon and Kyle O’Quinn this year while not as high on Kevin Murphy and Justin Hamilton and with Darius Johnson-Odom objectively on the border but subjectively a favorite.

But for brevity shape, I want to focus on what the Value Add databases said about the 24 non-RSCI players who have gone in the last five first rounds. Of the 24 non-RSCI players who went in the 1st round during the past five years, Value Add had projected 17 of them early as NBA prospects. 

Value Add sticks by NO on Ezeli, Fredette, W. Johnson, D. Carroll and J. Alexander 

Admittedly, Value Add missed Jason Thompson and JaValle McGee in 2008, but in five other cases non-RSCI players were picked against what the calculations said. I’m afraid Festus Ezeli and Jimmer Fredette are very likely to follow in the footsteps of Wes Johnson, DeMarre Carroll and Joe Alexander as very bad first round picks. In short, if you were not a RSCI ranked player out of high school AND the Value Add databases don’t show you as breaking out statistically, you are unlikely to measure up as a 1st round NBA pick.

Here are the 24 non-RSCI 1st rounders and whether or not Value Add spotted them as prospect early (YES or NO):

2012


1. In 2011 we already had Damian Lillard of Weber State in the top 3% of players, then we had him vaulting to 19th this year and we pegged him as a clear 1st rounder. He backed it up by dominating Summer League for MVP honors. YES

2. In 2010 we already had Andrew Nicholson of St. Bonaventure in the top 10% of players, and he improved in 2011 before shooting up to the 26th best college player and 22nd best prospect in our databases. He backed up his selection as the 16th best draftee in Summer League. YES

3. We had Arnett Moultrie of Mississippi State as the 21st best prospect this year, and applaud the selection but will have to wait to see if he measures up since he was injured for Summer League. YES

4. We did not even have Festus Ezeli (NO) as in the top 15% of all players as junior, and still only had him as the 44th best prospect this year, a pretty shaky first round pick. He played about where we had him in Summer League. NO

2011

5. While a great college player, Jimmer Fredette (NO) was a worst prospect than all five other non-RSCI 1st round picks behind him in this draft. As a pure college player, he was the 15th best in the country, but with every NBA Indicator red flag – way overused (2nd most possessions), poor defensive rebounder, barely above average in steals despite playing guard. This is just one where you yell, “Don’t take him in the first round!” but clearly a great guy, so we hope he recovers from a miserable first year. NO

6. In 2010 as a FRESHMAN Alec Burks of Colorado was already the 67th best player in the country, and he had already moved up to 30th his sophomore year before going 12th in the draft and having a strong first year. YES

7. We had Nikola Vucevic of USC in our top 3% as a sophomore, and as a junior he was up to 18th before being drafted 16th and having a strong rookie season. YES

8. Value Add spotted Kenneth Faried at Morehead State three years out when in 2009 when he was the 54th best player. He then improved to 36th in 2010 and was the 9th best player before entering the draft. As a rookie, Hollinger at ESPN ranks him as the 20th best NBA player this year – not 20th best rookie, but 20th best player. YES

9. Norris Cole was in the top 7% already as a sophomore, and the 3rd best Value Add player his final year before being taken by the Heat and getting a title his rookie year. YES

10. Believe it or not, Jimmy Butler was actually higher in our database than Lazar Hayward in Lazar’s final year, ranking as the 5th best player in the country and finishing slightly lower his senior year. ESPN subjectively ranked him as the 23rd best rookie this year, with a big role to play. YES

2010

11. Certainly Wes Johnson (NO) had a great one year at Syracuse, but his Value Add numbers were so bad the previous season at Iowa State – not even in the top 5% with poor steals and defensive rebounds, that we just could not have seen him coming as a No. 4 pick. NO

12. Value Add pegged Gordon Hayward of Butler as the 26th best player in the country his FRESHMAN year, and he was even better his sophomore year before going 9th in the draft. Definitely a 1st rounder. YES

13. Value Add already had Paul George in the top 4% as a freshman even though he was playing for a Fresno State team that wasn’t even in the top 200, so this is the kind of player Value Add will find and project out to an NBA player early in their career in obscure locations. YES

14. Larry Sanders was in the top 10% of players as a sophomore and then was one of the top 1% of all defensive rebounders and a top 100 player his junior year. YES

15. Value Add spotted Trevor Booker of Clemson four years out as he was in the top 2% of all players as a freshmen and went onto have one of the greatest careers in the Value Add database. He stayed in the top 2% his sophomore year and then was the 7th best player as a junior and 15th best player as a senior – giving great indication of 1st round pick and great season he had this year in the NBA. YES

16. Another player spotted three years out as a top 100 player already as a freshman was Dominique Jones of USF, who stayed there and shot up to 11th his final year. YES

17. While he hasn’t had the big success in the league so far, Lazar Hayward’s steady improvement his final three seasons from 150th to 90th to 34th in the Value Add database coupled with just missing the top 2% in steals and top 3% in defensive rebounding did point to him as a draftee. YES

2009 

18. Value Add spotted Stephen Curry of Davidson as the greatest mid-major player in the database and already in the top 3% of all players as a freshman. He went onto have perhaps the greatest 3-year career in the database, ranking as the 5th best player in the country his sophomore year and 4th best his junior year, before being taken 7th in the draft and currently rating as one of the 30 best players in the NBA according to NBA/Hollinger. YES

19. Eric Maynor was not nearly as good as Curry, but he was in the top 100 in Value Add his last three years and had one great year as the 6th best player his senior year. He has struggled as a pro, but would have projected as a late first rounder. YES

20. DeMarre Carroll (NO) was in the top 4% at Vanderbilt, and then repeated that at Missouri before shooting up to be the 19th best player his final year. Can’t say he was spotted early though. NO

2008 

21. Joe Alexander (NO) wasn’t even one of the top 100 college players in the country in the Value Add database either of his seasons, so his selection as the No. 8 pick remains one of the most confusing picks we’ve seen and his almost immediate exit from the NBA one of the most predictable events we’ve seen. That being said, the next two were also not seen by Value Add and they both were great this year. NO

22. Jason Thompson (NO-BUT BACKED IT UP) of Rider ranked outside the top 100 as well, but has proven to be a very strong NBA player. NO

23. JaVale McGee’s (NO-BUT BACKED IT UP) action was limited enough at Nevada to go unnoticed by Value (just top 15% of all players), so we did not see that pick, or really strong performance this year that may put the gun antics with Gilbert Arenas behind him. NO

24. Value Add sees even IUPUI, picking George Hill as just outside the top 3% in 2007 before rising to be the 9th best player in the country his final year. He backed up the selection on the NBA court today. YES

Certainly Value Add points out some other prospects that upon observation cannot make the jump. In addition to Value Add picks like O'Quinn, Crowder, Nicholson and Lillard, Central Connecticut's Ken Horton jumped out as a great college player in 2011 and we calculated him as a borderline 57th pick - but when scouts looked at him he wasn't ready.  Likewise TJ McConnell has been one of the top 14 players his first two years despite playing out of the spotlight at Duquesne, but in watching him he doesn't look athletic enough to play at the NBA-level even though Arizona apparently agreed with Value Add as far as his Value Add and picked him up.  But the point is that while we know to watch the RSCI 5-stars and 4-stars through their college careers, Value Add almost always identifies the late bloomers that may also land in the NBA.

No comments: