jesterbobman
03-08-2012, 06:12 PM
Starting off by quoting timvp to get your attention:
...Speaking of which, Kawhi Leonard's WP48 is off the charts high right now. WP48 is basically a stat created by a group of economists whose mission was to figure out a stat that most closely correlates with winning basketball games. FWIW, their findings are pretty interesting.
Anyways, here are the current leaders* in WP48:
1. LeBron James: .408
2. Tyson Chandler: .333
3. Chris Paul: .288
4. Joakim Noah: .285
5. Kawhi Leonard: .273
6. Steve Nash: .272
:hat
*at least 500 minutes played
Basically, WP48 uses box score data to predict wins for a season, and it's very accurate for season long wins(explains about 95% of wins). The idea behind it is to find out a way to measure the individual contributions of acts on court(Positive value of a steal, negative value of a turnover etc), sum the individual contributions of players, and then adjust for position(To account for the fact that Centres and Power forwards will have a more positive general value based on rebounding etc).
The basic process treats possesions as a scarce resource, and players get credit for how many possesions they bring(relative to position) and how effectively they use those possesions.(TS%, etc).
For any questions, go here: http://wagesofwins.com/faq/
There is some evidence from the last few drafts that the Spurs are using WP48(or some similar Regression based stat) in their player evaluations for the draft.
Other, Smarter people(Arturo Galetti) than me have made models to predict WP48 performance in the draft. See
Drafts 1995-2010
http://wagesofwins.com/2012/02/13/jeremy-lin-and-the-ghost-of-nba-drafts-past/
2011 Draft
http://wagesofwins.com/2011/06/18/the-2011-win-score-numbers-for-nba-draft-prospects/
The interesting point is the last few years, and the Spurs draft picks, in the period where they've gone towards picking College players in the draft.
Year : Spurs Pick(#) : Best Available in model(Overall Rank in model)
2008:George Hill(26) : George Hill (3rd)
2009: DeJuan Blair (37) : DeJuan Blair(1st)
2010: James Anderson (20): Damion James(1st)
2011: Kawhi Leonard(15) : Kenneth Faried(1st)
In 2010, James Anderson was the 6th best prospect available, behind James, Jarvis Varnado, Tiny Gallon, Landry Fields and Jeremy Lin.
In 2011, Leonard was the 2nd best available(and 2nd overall) behind Faried.
I'm not saying it's a perfect predictor of what the Spurs will do on draft day, or the only thing the use in evaluations of players(in the draft or for trades). But it's a fair whack of circumstantial evidence. So when you are trying to think of what the Spurs are going to do, have a look at win production.
...Speaking of which, Kawhi Leonard's WP48 is off the charts high right now. WP48 is basically a stat created by a group of economists whose mission was to figure out a stat that most closely correlates with winning basketball games. FWIW, their findings are pretty interesting.
Anyways, here are the current leaders* in WP48:
1. LeBron James: .408
2. Tyson Chandler: .333
3. Chris Paul: .288
4. Joakim Noah: .285
5. Kawhi Leonard: .273
6. Steve Nash: .272
:hat
*at least 500 minutes played
Basically, WP48 uses box score data to predict wins for a season, and it's very accurate for season long wins(explains about 95% of wins). The idea behind it is to find out a way to measure the individual contributions of acts on court(Positive value of a steal, negative value of a turnover etc), sum the individual contributions of players, and then adjust for position(To account for the fact that Centres and Power forwards will have a more positive general value based on rebounding etc).
The basic process treats possesions as a scarce resource, and players get credit for how many possesions they bring(relative to position) and how effectively they use those possesions.(TS%, etc).
For any questions, go here: http://wagesofwins.com/faq/
There is some evidence from the last few drafts that the Spurs are using WP48(or some similar Regression based stat) in their player evaluations for the draft.
Other, Smarter people(Arturo Galetti) than me have made models to predict WP48 performance in the draft. See
Drafts 1995-2010
http://wagesofwins.com/2012/02/13/jeremy-lin-and-the-ghost-of-nba-drafts-past/
2011 Draft
http://wagesofwins.com/2011/06/18/the-2011-win-score-numbers-for-nba-draft-prospects/
The interesting point is the last few years, and the Spurs draft picks, in the period where they've gone towards picking College players in the draft.
Year : Spurs Pick(#) : Best Available in model(Overall Rank in model)
2008:George Hill(26) : George Hill (3rd)
2009: DeJuan Blair (37) : DeJuan Blair(1st)
2010: James Anderson (20): Damion James(1st)
2011: Kawhi Leonard(15) : Kenneth Faried(1st)
In 2010, James Anderson was the 6th best prospect available, behind James, Jarvis Varnado, Tiny Gallon, Landry Fields and Jeremy Lin.
In 2011, Leonard was the 2nd best available(and 2nd overall) behind Faried.
I'm not saying it's a perfect predictor of what the Spurs will do on draft day, or the only thing the use in evaluations of players(in the draft or for trades). But it's a fair whack of circumstantial evidence. So when you are trying to think of what the Spurs are going to do, have a look at win production.