The PAWS40 Story Before the 2010 NBA Draft

Arturo Galletti, Andres Alvarez, and I were planning on presenting a wealth of material on the 2010 NBA draft.  But time has run out and our plans were not realized (but we have high hopes for 2011!!!).  At this point, all there is time to do is to present the Position Adjusted Win Score per 40 minute (PAWS40) numbers for the players Yahoo!Sports think will get drafted tonight. (much of ESPN’s coverage is for insiders only)

Before we get to the numbers, let me note that Wins Score per 40 minutes (WS40) for each college player is tracked by DraftExpress.   To calculate PAWS40, one needs to know the average performance at each position.   From 1995 to 2009, the players selected out of college at each position posted the following average WS40:

PG: 7.40

SG: 8.40

SF: 9.95

PF: 12.59

C: 12.32

The average player across all position had a WS40 of 10.17. With these numbers in hand, we can calculate each player’s PAWS40.  For example, point guard John Wall has a WS40 of 7.2 in 2009-10.  So his PAWS40 would be…

John Wall PAWS40 = (7.2–7.4) + 10.17 = 10.0

A mark of 10.0 is just about average for a player selected out of college from 1995 to 2009.  And as I noted back in May, this performance suggests the Wizards needed to put some real thought into the decision to take Wall with the first pick.  Yes, the scouts tell us he is clearly a future star.  But he did not perform this way as a freshman.  Now maybe that was because he was just a freshman.  Or maybe he isn’t quite as good as people believe.  

Just looking at PAWS40 is not going to give you that answer.  As was noted in The Wages of Wins (in the discussion of Wins Produced): “Knowing the value of each player is only the starting point of analysis. The next step is determining why the player is productive or unproductive. In our view, this is where coaching should begin. We think we can offer a reasonable measure of a player’s productivity. Although we have offered some insights into why players are productive, ultimately this question can only be answered by additional scrutiny into the age and injury status of the player, the construction of a team, and the roles the player plays on the floor.

Okay, with that paragraph in mind, let’s look at all the players Yahoo! thinks will get drafted.

Here are the players from the mock of the first round:

And here are the players from the mock of the second round:In Stumbling on Wins (plugged both books in one post!!!), we examine the NBA Draft.  What we found is that some factors that don’t impact future performance (i.e. Final Four appearances, relative height, etc…) do impact where a player is drafted.  And rebounds (which we looked at a number of different ways), a factor that is related to future performance, is not related to where a player is chosen (and again, we looked at rebounds a variety of ways and the story was the same). 

So it appears people in the NBA could do better on draft night.  In fact, we note that if all you looked at were the college numbers, you would project future performance better in the NBA than you would by just noting where a player is selected.  That being said, the link between college numbers and NBA performance is not perfect.  Yes, there is a statistical link.  Players who do well in college tend to do better in the NBA.  But it is possible for players who do well in college to play poorly in the NBA.  And it is possible (although this is not a common occurrence) for a player to play poorly in college and then play well in the NBA.

Again, Arturo, Andres, and I hope to do more before next year’s draft.  At least, I hope by 2011 we can make more definitive statements about each player.  For now, though, hope these numbers help you enjoy the 2010 draft.

- DJ

P.S. Robbie O’Malley – at his blog (which I think has been re-designed) – also provides this analysis.  So one can also go to Robbie’s blog and discuss the 2010 draft.

Arturo Galletti Evaluates 30 Years of the NBA Draft for the Wall Street Journal

David Biderman of the Wall Street Journal has once again referenced Wins Produced in telling a story.  Biderman’s story — Good News, Nets: Third Pick’s Better Than the Second – offers evidence that the second pick in the NBA draft is not necessarily better than the third pick.

As noted, Biderman’s evidence is based on Wins Produced.  And although these numbers started with my calculations, it was the work of Arturo Galletti – as Biderman notes – that allowed us to see the productivity of each draft pick since 1977 (the numbers are after four years in the league).

Biderman’s article is only 300 words long.  And his table only has 13 rows.  So there is a bit more we can say with the numbers Galletti provides.

One issue we have to address is that players taken earlier in the draft will get more minutes.  And the increase in minutes – as published research from Colin Camerer and Roberto Weber (1999) indicates – is independent of player performance.  So if we are going to evaluate the NBA draft, we can’t use aggregate measures.  Top draft picks will produce more wins simply because they will get more minutes.  So to evaluate the productivity of players at each draft position, one should probably consider per-minute (or per 48 minute) performance.

With that in mind, let’s look at the total minutes played, total Wins Produced, and WP48 [i.e. total Wins Produced divided by total minutes played multiplied by 48] at each draft position.

As one can see, the NBA appears to do a good job identifying wins producers at the top of the draft.  The top five spots in the draft rank among the top seven spots in Wins Produced.  Yes – as Biderman notes – the third spot is somewhat better than the second pick.  But it looks like NBA decision-makers can generally identify players like LeBron James, Tim Duncan, Michael Jordan, Shaquille O’Neal, David Robinson, Charles Barkley, Magic Johnson, Larry Bird, Hakeem Olajuwon, Chris Paul, etc… 

What do these players have in common?  Each player can do many things well.  Consequently, it is easier for decision-makers to see that these players will be able to contribute.

Once we get past the top five slots, though, you then have players who can’t do everything well.  Now decision-makers have to evaluate players who can only score (think Allan Houston) and others who primarily rebound (i.e. Dennis Rodman, Anderson Varejao, Charles Oakley, etc…).  As noted in Stumbling on Wins, decision-makers on draft day clearly favor scorers.  And rebounds in college– again, as noted – do not really impact a player’s draft position.  So we should not be surprised that after we get past the players who do everything (i.e past the first five slots), productive players can be found throughout the first round. 

There are many more stories to be with Arturo’s numbers.  And all these stories will be told eventually.  But not in the next post.  Hopefully the next post will be about the 2010 NBA draft.

- DJ

The Trade of the Century?

Editor’s Note: The recent trade between the Kings and Sixers will be discussed from two perspectives.  The following is from Mike Madden, a fan of the Kings.  An upcoming post – from Sam Cohen –  will discuss this trade from the perspective of the Sixers.

Mike Madden is a noble man living in the slums of Cincinnati. He follows the Kings because they used to be the Cincinnati Royals, and they’re the closest thing Cincinnati has to a basketball team. One day, if he’s up to it, Mike will buy the Kings from the Maloofs and relocate them back to da ‘nati. Until then, he’s stuck hoping Vlade Divac will come back and resurrect the Kings.

The Sacramento Kings recently traded Andres Nocioni and Spencer Hawes to the Sixers for Samuel Dalembert. Seems like a harmless trade on paper- two tall men are swapped, and the Sixers get a solid bench player. Philly also sheds about $3 million in net salary. One would think that the Sixers got the better end of the deal, right?

Wrong.

According to the production posted by each player involved in the trade this past season, it appears as though the Kings got the better end of the deal:

Hawes: -.007 WP48

Nocioni: -.015 WP48

Dalembert: .243 WP48

Hawes and Nocioni combined for -.74 wins, while Dalembert produced 10.75 victories. From 2006-10, Slammin’ Sammy D has been an above average player in terms of Wins Produced, averaging between 7-10 wins per season (with this past year being his best). This is due primarily to his rebounds and blocked shots, as well as his relatively high field goal percentage.

The Kings were eager to get rid of both Nocioni and Hawes. Nocioni demanded a trade out of Sacramento, and is due $13 million over the next two years. His best year was 2006 when he produced 9.38 wins, but his production has dwindled tremendously since then. Hawes was drafted 10th overall in 2007, has been in the negatives in terms of WP48 since then, and doesn’t show many signs of improvement. Once again, the Kings were happy to rid themselves of these two players.

With this trade, this is what the Kings’ roster looks like for next year:

Starting:

PG: Tyreke Evans: .158*

SG: Francisco Garcia: .097

SF: Omri Casspi: .069

PF: Jason Thompson: .044

C: Samuel Dalembert: .243

Bench:

PG: Beno Udrih: .136

SG: Ime Udoka: .132

SF: Donte Green: -.063

PF: Carl Landry: .026

C: Jon Brockman: .252

Projection: 36 wins.

*Evans played the majority of his time at SG, but the team is looking to move him to PG next season.

The Kings also have the fifth pick in the upcoming draft, and with the recent success of Evans, I’m expecting a player who will be an above average rookie and produce between 2-3 wins.

In conclusion, this may not be the trade of the century, but Sacramento did take a step in the right direction. Their record next year will be dependant upon a number of variables; particularly Evans success at point guard, Beno Udrih’s decreased playing time, Dalembert’s consistent production, the rookie, and the development of younger players such as Casspi, Thompson, and Brockman (and Evans, of course). And if the Maloof’s can lure a quality free agent (Joe Johnson?) for next season, don’t be surprised to see the Kings finish above .500.

- Mike Madden

How the Thunder Charged Forward (and How the Charge can Continue)

James Brocato graduated from Washington State University in 2009.  He is currently attending Law School at Gonzaga University.  He grew up a passionate supporter of the Seattle Supersonics, but their relocation to Oklahoma City in 2008 put him in an awkward position.   Failed attempts to root for Phoenix and Portland made him realize that his heart is still with the team he grew up loving, even if they’re not the hometown heroes anymore.

In the 2008-09 season, the Oklahoma City Thunder won 23 games.  In 2009-10, they won 50.  This 27 game swing was good for the best improvement in the NBA over that time period, and an unexpected surprise for a promising young team.  Much of the surprise is not that the team achieved a 50 win season, but when the 50 win season was realized. The Thunder have recently undergone a rigorous rebuilding process with an impressive and confident front office.  The success of this process promises hope for struggling teams everywhere – if the right people are put in charge and the pieces fall into place — a bad team can become good in only a few years.  Of course, a player like Kevin Durant is only available so often.  Regardless, one has to give credit where credit is due.  But, while the front office has clearly done wonders for a team that has only recently gone through hard times, it is important to determine exactly where the team improved, and what it can do to further its progress. So the begging question remains: how did Oklahoma City improve so much in one year? 

The allocation of more minutes in 2009-10 to Oklahoma City’s productive bench players from 2008-09 accounted for just over five extra wins.  This can all be attributed to one player – Thabo Sefolosha.  The rest of the improvement — which accounts for the bulk of the Thunder’s progress — is the increase in production from two starters.  The two players, Kevin Durant and Russell Westbrook, combined to produce 15 more wins than they had produced in the 2008-09 season.  This increase in wins did not come from a dramatic increase in minutes; rather, these two actually improved their per-minute production by doing things like shooting more efficiently and earning more possessions for their team.  This is especially notable in that, as Dave Berri has argued repeatedly in the past, player production is relatively stable over time.  Stumbling on Wins does note that younger players get better with age.  However, for one player to triple his production is very rare.  And for another player — in this case the most productive player on his team during the previous season — to double his production, is also uncommon.

Kevin Durant won the rookie of the year award in 2007-2008, but he was not a good rookie (see Kevin Durant was not the Best Rookie).

However, as disappointing as his rookie campaign was, he improved more than any other player in the NBA over the next two seasons.  Durant prides himself on being a hard worker and determined to win.  His hard work has paid off.  But now he has a great responsibility to maintain his brilliant level of production, as he is the most important piece to a team with big dreams.  In fact, his subpar play in the playoffs (though the sample size is very small) may have been a factor in their first round exit. Durant produced only 0.043 wins per 48 minutes, compared to 0.280 during the regular season. Again, the sample size is very small, so it is hard to draw a worthwhile conclusion from those numbers.  The main point, however, is clear: Durant plays the lead role in the Thunder’s future, and their success will largely depend on his play.

Durant’s success may not be a surprise to anyone (he finished second in MVP voting for the 2009-10 season).  In fact, most of what has been established thus far in this post – the improved play of Westbrook and solid contribution from Sefolosha – is probably not surprising to casual fans. 

What may come a surprise, however, is the very unproductive play of Jeff Green.  Green produced less than two wins in over 3,000 minutes of play time.  He played more minutes than all but six players in the entire NBA, and only produced ONE win.  Scarier still is that this bad season isn’t an anomaly.  He posted nearly identical numbers in 2008-09, and was the worst player in the NBA the previous season.  The reason this is so alarming is that the front office puts a great deal of confidence in Green.  He was drafted with the 5th overall pick in the 2007 NBA draft, a pick that was traded by Boston for Ray Allen.  In addition, they have made no moves to secure a player to start at power forward – a sign of their trust in Green to play the position.  Obviously the coaching staff shares this confidence, or he wouldn’t have seen such a huge volume of playing time.   He is considered (along with Durant and Westbrook) a member of the core of this young team by the front office, coaching staff, and the media.  This is extremely unfortunate, as Green’s poor play may well put a damper on Oklahoma City’s success for years to come.

The best case scenario would be to trade Green, though this is unlikely given what we have just established.  A more foreseeable, yet still positive possibility is that the front office realizes it needs a true power forward (Green is naturally a small forward, but plays PF so he and Durant can be on the floor at the same time).  In this case, a productive power forward could be brought in, and Green could be moved to the sixth man position. This would diminish the effect of Green’s lack of productivity enough to make everyone happy.

While we’re on the subject of players that hurt the Thunder, we might as well mention Nenad Kristic.  Kristic, despite his continued negative production, continues to be employed as Oklahoma City’s starting center.  He has literally lost the Thunder a win over his last two seasons.  However, the poor play of Kristic will probably not be as detrimental to Oklahoma City in the years to come. 

There are two reasons for this: 1.He is not perceived as that good of a player, and 2. The success of Serge Ibaka, particularly in the playoffs, might soon earn him the starting position, and more of Kristic’s minutes. 

As for the offseason, Sam Presti’s job is clear: get a good big man! Oklahoma City’s backcourt is great.  Their swingman is fantastic.  Their big men are terrible.  Fans can keep their fingers crossed that Serge Ibaka’s development will follow suit with Durant’s and Westbrook’s.  Chances are, though, that it won’t.  Additionally, it is very improbable that one of the Thunder’s four late first round picks will develop into a star.

Fortunately, there is hope.  The Thunder have $17 million free in the salary cap this offseason ($22 million in the unlikely event that Kristic doesn’t resign with his player option).  This is enough to sign a huge free agent.  Of those available, the most desirable would be David Lee, Carlos Boozer, or Chris Bosh.  Bosh is out of the picture, he already narrowed down his list to five, and the Thunder didn’t make the cut. Boozer, although extremely desirable, is also probably unattainable, signs point to him seeking a larger market.  Lee, on the other hand, would be a perfect match.  In fact, theoretically, if Lee simply replaced Kristic’s minutes, one would expect the Thunder to produce around 60 wins.  Giving Lee the same amount of minutes he received last season, which would involve replacing all of Kristic’s minutes at center, and some of Collison’s at power forward, would render the Thunder a 64 win team.  However, before people jump to conclusions and assume adding Lee would make Oklahoma City the best team in the NBA, there are two important points to note:

1. This conclusion is based on Wins Produced from the 2009-10 season, and this may not remain constant in the 2010-11 season (though, as mentioned earlier, evidence supports it to stay at least relatively constant), and

2. Diminshing returns – a real but small effect (as noted in The Wages of Wins and Stumbling on Wins) — may decrease Lee’s production, or the production of someone else, if he were to be added to the team.  Specifically, only so many rebounds can be grabbed each game.  Oklahoma City is already one of the premier rebounding teams in the NBA, and a great deal of Lee’s production comes from rebounding. Still, the addition of Lee, or Boozer, certainly could make the Thunder one of the best teams in the league.

-James Brocato

Not So Final Thoughts on the 2010 Finals

Here are the final Wins Produced numbers for the 2010 NBA Finals.

And for those who are interested, here are the WP48 marks for players who played at least 10 minutes in Game Seven:

  • Pau Gasol: 0.452
  • Derek Fisher: 0.283
  • Lamar Odom: 0.116
  • Kobe Bryant: 0.110
  • Ron Artest: 0.095
  • Andrew Bynum: 0.004
  • Jordan Farmar: -0.625

 

  • Glenn Davis: 0.326
  • Rajon Rondo: 0.307
  • Paul Pierce: 0.157
  • Rasheed Wallace: 0.061
  • Kevin Garnett: -0.008
  • Ray Allen: -0.186

And here are some thoughts on these numbers (and there will be more on the NBA Finals in a later post):

  • One does not need a sophisticated model to see that Kobe Bryant struggled in Game Seven.  His amazing rebounding effort was just enough to barely move him into the above average range.
  • Pau Gasol — who was the Most Productive Player for LA in the Finals — was once again the key player for the Lakers.
  • Andrew Bynum, though, didn’t do much.
  • However, Kevin Garnett did even less.  Across the last two games, KG played 69 minutes and grabbed only nine rebounds.  Across all seven games he was above average.  But when the Celtics needed one more win, KG didn’t show up.  And that means I blame KG for my fourth place finish in the True Hoop Smackdown.  Had KG done more (enough for the Celtics to win Game Six), I would have finished in second (I might have more on that contest later as well).
  • Rajon Rondo was again amazing.  He led all players in the playoffs in Wins Produced (and I will have more on this later as well).
  • And finally, Big Baby played well in the NBA Finals. Does this mean he will play well going forward?  I would not bet on this outcome.  But a productive Big Baby is suddenly a possibility (while before it seemed more like an impossibility).

Okay, that is all I wanted to say tonight.  We do have a number of posts planned. 

  • James Brocato has a column on the Oklahoma City Thunder which should be posted on Saturday.
  • Michael Madden and Sam Cohen will be discussing the Samuel Dalembert trade from the perspective of the Kings and Sixers.
  • And Devin Dignam has submitted a post on the Toronto Raptors.

I also want to alert people to “Wins Produced” (or WP inspired analysis) on other sites:

Let me close by noting again that I will be offering more on the Finals soon.  And there is also going to be some analysis on the NBA Draft as well.  Not sure if the bulk of our draft coverage will appear before or after the event, though.

- DJ