Two years ago the Pistons won 59 games and advanced to the Eastern Conference finals.  At the start of last year, though, Detroit sent Chauncey Billups to the Denver Nuggets for Allen Iverson.  Although the Pistons claimed this was done to enhance the team’s chances in 2008-09, it was suspected the acquisition of Iverson was all about salary space.

When the 2008-09 season ended it was clear that Iverson hadn’t helped much. But the salary cap space promised hope.  Such hope, though, was dashed when the Pistons took this space and invested in Ben Gordon and Charlie Villanueva.  When those acquisitions were announced it seemed clear the Pistons would struggle in 2009-10. 

Soon after the season started, the grim expectations were realized.  And now after 59 games – and after the Pistons politely declined a gift wrapped victory from the Golden State Warriors —  it seems like a good time to check in on the depths the new Pistons have reached.

Table One: The Detroit Pistons after 59 games in 2009-10

Table One reports both what the Pistons have done so far and what we could have expected given the performance of the veterans last year.  Looking at these numbers one can see…

• the Pistons have only won 21 games this season.  Had these veterans maintained what we saw last year, the Pistons would have only won 26 games.  In other words, the Pistons are supposed to be bad (although not quite this bad).
• much of this small decline can be tied to declines seen in the play of Gordon and Villanueva.  Had these players maintained what they did last year the Pistons would have still been below average.  But with these players offering even less, the Pistons are now awful.
• it would be much worse (much worse than awful?) if Ben Wallace hadn’t signed for $1.3 million.  Wallac has already produced 9.0 wins.   The other Pistons have only produced 11.7 wins.  In other words, without Wallace the Pistons would be only on pace to win 16 games.
• given the age of Wallace, the Pistons are in trouble going forward.  After Big Ben the Pistons have only two above average players (Jonas Jerebko and Tayshaun Prince).  And neither player is well above average. Given what the Pistons have committed in salaries next season, it looks like help is only going to come in the draft. 
• the draft has been somewhat mixed.  Rodney Stuckey is still a below average player.  And passing on DeJuan Blair for DaJuan Summers is still hard to believe.  But the choice of Jerebko gives Pistons fans (of which I am one) some hope.

So in two years the Pistons have gone from title contenders to lottery hopefuls.  And given that so much of this team’s meager success is linked to Big Ben, the future looks bleak. Maybe, though, the draft will go well and Joe Dumars can make some trades this summer.  In other words, maybe some different decisions can be made and different – and more productive – players can come to Detroit. 

- DJ
The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

Apparently there is some conflict between Paul Westphal and his team.  After Spencer Hawes and other members of the Kings criticized Westphal’s rotations, the head coach responded by reducing Hawes’ minutes to zero.  Since Hawes has been the starting center for this team for much of the season, some might see the deactivation of Hawes as an action that makes Sacramento a worse team.  But the numbers don’t seem to point in that direction.

Hawes was selected by the Kings with the 10^th pick in the 2007 NBA draft.  In his first two seasons he has posted the following numbers:

2007-08: -0.9 Wins Produced, -0.046 WP48 [Wins Produced per 48 minutes]

2008-09: -1.0 Wins Produced, -0.021 WP48

So Hawes was in the negative range his first two season.  This season – as noted in Table One — Hawes has improved.  After 57 games he has produced -0.2 wins and posted a -0.006 WP48.  Yes, he has improved but he is still not in positive territory.

Table One: The Sacramento Kings after 57 games in 2009-10

If we look at the individual numbers, we see that Hawes began his career with a level of shooting efficiency that was below average. These numbers have improved, but still remain below par.  When we turn to rebounding we see that again he was below average in 2007-08.  These numbers, though, have just gotten worse.

So sitting Hawes is not really a bad strategy.  But that is perhaps not the point of Hawes’ critique.  Is there a problem with Westphal’s rotations?

One way of addressing this is to look at how the performance of individual players has changed from last season to this season.  And as we can see in Table One, of the players who have played significant minutes, only Andres Nocioni has declined.  Most players on the team – like Hawes – have gotten slightly better.

What does that mean?  Well, our null hypothesis would be that Westphal messing around with the player’s minutes would cause per-minute performance to decline.  Just looking at the numbers (which is really not a substitute for more systematic analysis where we control for factors that cause performance to change, but this is all I am willing to do for a blog post) it doesn’t appear that Westphal’s rotations are hurting this team. 

Although Westphal is probably not hurting the team, it also doesn’t appear that this team is much better this year (this is typical, since most coaches don’t appear to help their team much).  Very early in the season, though, a different story was told.  After nine games the team was 5-4, and as noted as the time, Jason Thompson was a big reason why the Kings looked so competitive.  Since that start, though, the Kings have been 13-35.  And Thompson looks very much like he did last year. 

Such a story highlights the problem with looking at numbers over a small sample. It is very easy for these numbers to lead one to incorrect conclusions (but early in the season, such numbers are really hard to resist).

Basketball numbers over a larger sample, though, are generally more reliable.  Looking back at Table One we can see the story the bigger sample is telling.  The Kings currently employe a few above average players.  This list includes Tyreke Evans, Omri Casspi, Ime Udoka, and Jon Brockman.  Three of these players are rookies, so the Kings look like they have added some pieces.  But at the moment, none of these players are far removed from average.  For the Kings to really make strides in the standings, one or two players are going to have to produce wins in double-figure quantities.  Currently, no player on this team is on pace to produce ten wins.

Until players improve – and/or better players are added –Westphal is probably going to have trouble finding a line-up that can be consistently successful. 

There is some hope that the recent transactions will help.  Carl Landy and Dominic McGuire were productive last season.  Thus far in Sacramento, though, these players haven’t produced.  Again, though, our current sample is really very small (so drawing inferences at this point is not a good idea).  If we look at both player’s careers, it does look like each can be productive. 

So going forward, the Kings will have a rotation that includes six productive players (Evans, Casspi, Landry, Udoka, McGuire, and Brockman).  If one or more of these players could get substantial minutes and/or up their WP48 numbers above 0.200, the Kings could be a competitive team in 2010-11.  As for 2009-10, though, it looks like players, coaches, and fans in Sacramento are not going to be as happy as they would like.  And that story is going to be the same regardless of how Westphal assigns minutes.

 - DJ
The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

Before the season started Chris Mannix of Sports Illustrated.com offered the following assessment of the Milwaukee Bucks: Milwaukee lost its second- and third-leading scorers (Richard Jefferson and Charlie Villanueva), while its No. 1 scorer (Michael Redd) and top rebounder (Andrew Bogut) are returning from major injuries. And it’s possible that part-time starting point guard Ramon Sessions could walk as a free agent. Coach Scott Skiles will keep the defense playing at a high level, but points will be hard to come by.

Mannix ranked the Bucks as the 28^th best team in the NBA.  In other words, Mannix thought that only two teams (the Nets and Kings) should be ranked below the Bucks.

Mannix was not alone in this assessment.  In the NBA preview of ESPN the Magazine (insider access required), the Bucks were listed as the 14^th best team in the Eastern Conference.  Like Mannix, only the Nets were ranked lower by the experts at ESPN.

After 56 games, though, the Bucks’ record stands at 28-28.  And Milwaukee’s efficiency differential (offensive efficiency minus defensive efficiency) of 0.6 suggests that this team is slightly above average.

Certainly Milwaukee isn’t great.  But why isn’t this team awful?

It’s easy to focus on Brandon Jennings, who currently leads Milwaukee in scoring and shot attempts.  Jennings was selected with the 10^th pick in the 2009 NBA draft.  But it was unclear how good he would be as a rookie.  Jennings skipped college and did not exactly post amazing numbers in his one year in Europe. 

At the start of the season it looked like teams had a made a mistake in passing on Mr. Jennings. But after an amazing start his overall productivity has slipped quite a bit (despite ranking second among all rookies in scoring). As Table One reveals, after 56 games Jennings has posted a 0.059 WP48 [Wins Produced per 48 minutes] and currently only ranks 6^th on the team in Wins Produced.  Although not bad for a rookie, it is not enough production to transform a bad team into the average squad that currently resides in Milwaukee. In sum, Mr. Jennings is not the answer.

Table One: The Milwaukee Bucks after 56 games in 2009-10

For an answer, let’s look at the five players who are producing more than Jennings.  Of Milwaukee’s 29.2 Wins Produced, 24.3 can be traced to the play of Andrew Bogut, Carlos Delfino, Ersan Ilyasova, Luc Mbah a Moute, and Luke Ridnour.  Of these four, the only surprise is Ilyasova.  Back in 2005, the Bucks selected the 18-year old Ilyasova in the second round of the draft.  He debuted in 2006-07, and as one can see, played quite poorly.  In 2009-10 – at the age of 22 – Ilyasova is now an above average power forward.  With the addition of Ilyasova the Bucks now have an impressive frontcourt.  Bogut is clearly the most productive player on the team at center.  And at the two forward spots the team has Ilyasova, Delfino, and Mbah a Moute.

At guard, though, Milwaukee has problems.  The only above average performer is Ridnour.  It’s possible for Salmons to offer something, although that hasn’t happened yet.

Obviously a team that is hovering around the average mark has weaknesses.  But these weaknesses are not the focus on this discussion.  What we wish to understand is why the Bucks were underestimated before the season started. 

When we look at what this team’s veterans did in the past we see a team that could have expected to win 25 of their first 56 games.  In other words, even if Ilyasova hadn’t improved this team wouldn’t have ranked among the league’s worse.  Again we wonder why this wasn’t seen.

One suspects that the problem lies in the evaluation of players like Delfino, Ridnour, and Mbah a Moute.  The former was quite productive – from a Wins Produced perspective — in both 2006-07 and 2007-08.  And both Ridnour and Mbah a Moute were above average last season.  None of these players, though, are scorers.  Given the importance placed on scoring totals, we shouldn’t be surprised that the Bucks were undervalued.

For more on this topic – and anything else related to the Milwaukee Bucks — one should read the Courtside Analyst (formally Bucks Diary).  Ty Willihnganz consistently offers NBA analysis that is well worth reading. 

- DJ
The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

Dwyane Wade sent his team a message during All-Star weekend:

“I’m in Miami,” he told a gaggle of reporters at All-Star weekend. “This is where I want to be.”

But he did not just leave it there. “I want to make sure,” he added, “that we do everything we can to make sure we build a winning program. We do that … I’m satisfied.”

In case there was any misinterpretation, a reporter asked if the Heat needed to add some pieces to entice him to stay.

“Yeah, we need to add pieces!” Wade said, busting into laughter at the obviousness of the situation. “Aint’ no question about that!”

Wade says the Heat have not had an elite roster in years. “We’re not one of the top teams in the League,” he explains. “You want to have an opportunity every year. You’re not Magic, you’re not Michael, you’re not Larry. You’re not going to win the championship every year. But you do want to have an opportunity to compete for that. I feel like we haven’t had the opportunity since ’06. I feel like, while I’m in my prime, before my prime leaves me in about four or five years, I want to make sure that I can give my all to the organization.”

So Wade’s message is…

• he wants to play in Miami.
• Miami needs help if it’s going to contend
• he is currently in his prime, and he expects that prime to last four or five years.

It’s not possible to comment on Wade’s desire to stay or leave Miami. But the other two contentions can be addressed.

After 57 games the Heat have posted a 29-28 record.  If we turn to efficiency differential (offensive efficiency minus defensive efficiency), we see a mark of 1.5.  That mark is consistent with a team that would win about 31 of their first 57 games.  So the Heat’s record underestimates slightly the quality of this team.  If this team was a serious contender, though, it would need to be on pace to surpass 50 wins; and whether we look at won-loss record of the team’s differential it is clear that the Heat are not a serious contender.

When we turn to the productivity of the individual players – reported in Table One – we can see where the Heat come up short.

Table One: The Miami Heat after 57 games in 2009-10

The top player on the Heat is obviously Wade.  Wade has already produced 11.2 wins this season, or about one-third of the team’s totals.  After Wade, though, there isn’t much.  Quentin Richardson, Udonis Haslem, and Dorell Wright are each above average; and combined they have produced one more win than Wade.  After this trio, the remainder of the team is posting WP48 [Wins Produced per 48 minutes] marks below 0.100 (the mark of an average player).  That list of below average players include Rafer Alston, who is putting up numbers as a starter for the Heat similar to what he did in New Jersey. Alston is now 33 years of age, so despite the willingness of teams to place him in the starting line-up, it looks like the end is near for him.

When we look at all these numbers, it appears Wade’s argument has support.  Prior to the trade deadline there was much talk that help was on the way, and one player that was mentioned was Amare Stoudemire.  The addition of Stoudemire would have given the Heat two starting All-Stars.  But it doesn’t look like that move would have transformed the Heat into title contenders.

To see this point, let’s imagine the Heat replaced their current starting power forward – Michael Beasley – with Stoudemire.   Currently Stoudemire is posting a 0.142 WP48.  If the Heat had received this level of production from Beasley (current WP48 of 0.087), Miami could have expected to win about two more games at this point in the season.  Yes, Stoudemire is above average. But he’s not one of the top players in the game and therefore he wouldn’t have provided Wade with the help he desires.

What about Antawn Jamison?  Jamison posted a 0.108 WP48 [and he is very old].  That’s slightly more than Beasley and somewhat less than Stoudemire (the difference between Stoudemire and Jamison is not quite as great as some might think).  So again, this move doesn’t help much.

Okay, what about Carlos Boozer?  Boozer’ sWP48 is above 0.250.  Replacing Beasley with Boozer would net the Heat 6.1 additional wins after 57 games.  And that would be enough to transform the Heat into a 53 win team across an 82 game schedule.  So Boozer would have helped quite a bit, although the Heat still wouldn’t have been as good as LeBron James and the Cavaliers.

Now salary cap restrictions prevented the Heat from swapping Beasley directly for Stoudemire, Jamison, or Boozer.  This exercise, though, reveals that replacing one below average starter with an All-Star is not enough to transform the Heat into title contenders.  At least, that’s true if the All-Star isn’t named LeBron or CP3.

So the Heat – as currently assembled – definitely are not as good as Wade would like.  And it’s probably going to take more than one move to change that reality. 

Let me close by commenting on Wade’s last point.  Wade argues that he is currently in his prime, and his prime-time is going to last another four or five years.  Right now Wade is 28 years old.  In general, we start to see noticable declines in player performance when an NBA player approaches 30 years of age.  And by the time a player is 32 or 33 – as we see with Rafer Alston — typically their performance is not close to what we saw when a player was in his prime.  Of course, any individual player can defy these trends for a period of time (although not forever).  So it’s possible that Wade is correct.  But in general, one should expect a 28 year-old basketball player to offer less and less as he ages (BTW the peak is around 24 and 25, but the performance at 28 is not much different from the peak).

And Wade’s history suggests some red flags. In three of his first six seasons he failed to play more than 61 games.  Plus, he has never played 80 games in a season (and he is currently hurt again).  Given this injury history and his age, perhaps the Heat should be thinking about whether they want to commit significant dollars to Wade. Yes, Wade is the most productive shooting guard in the game.  But this is probably not going to be true throughout his next contract, a contract that will pay Wade like he is still the very best shooting guard.  So for the Heat – or any other team looking to sign Wade to a new contract – his age and injury history should be something to think about.  Again, it’s possible for Wade to defy these trends.  The history of the league – and his own history – suggests that Wade’s primetime might not be as long as he thinks.

- DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.

Jerry Sloan – as noted back in October of 2008 – often tells the players on his team to focus on something besides their own scoring.  And players who go against this advice often suffer Sloan’s wrath.

Although a number of players have resisted such advice, Ronnie Brewer seemed to take it to heart.  Across his career Brewer has taken a below average number of shots from the field. Consequently, Brewer hasn’t been much of a scorer.  But when we look at his overall production,  we see that Brewer is consistently above average with respect to shooting efficiency, steals, and turnovers.  And when we put all these numbers together in terms of Wins Produced, we see in 2008-09 – as the following list of top shooting guards from that season reveals – Brewer ranked among the best in the game.

Dwyane Wade: 22.3 Wins Produced

Brandon Roy: 15.3 Wins Produced

Kobe Bryant: 15.0 Wins Produced

Mike Miller: 13.9 Wins Produced

Joe Johnson: 8.6 Wins Produced

Ronnie Brewer: 8.5 Wins Produced

Last season Brewer posted a 0.156 WP48 [Wins Produced per 48 minutes], a mark that actually exceeded what we saw from Joe Johnson.  This season Brewer is still posting above average numbers.  As Table One reveals, after 53 games Brewer had 4.9 wins and posted a 0.141 WP48.  Such production ranked fifth on the Jazz, and suggested it was wise for this team to employ Brewer as Utah’s starting shooting guard. 

Table One: The Utah Jazz after 53 games in 2009-10

Despite such wisdom for 53 games, though, the Jazz decided to go in a very diferent direction.  After defeating the New Orleans Hornets on Wednesday night (the team’s 11^th victory in 12 games), Brewer was sent the next day to the Memphis Grizzlies.  In return, the Jazz received a protected first round pick in 2011.  Here is how this trade was justified by Kevin O’Connor (Utah’s general manager).

O’Connor says the Jazz have some depth at Brewer’s position, so they could afford to trade Brewer, their first-round pick from the 2006 draft. Brewer is making $2.7 million this season and will be a restricted free agent this summer.

“We felt like we gain an asset for drafts coming up,” Jazz general manager Kevin O’Connor said Thursday. “We felt like we had a lot of players that were similar.”

So according to O’Connor, Utah already had players like Brewer.  And if we focus on scoring, this appears to be true.  Per 48 minutes, Brewer was scoring 14.8 points this season. The other potential shooting guards on the team – Wes Matthews and C.J. Miles – were scoring 18.2 and 20.4 points per 48 minutes.  Of course, part of the difference is that these players were taking more shots per 48 minutes.  If we focus on all that these players do – in other words, look beyond scoring — it’s clear that Miles and Matthews are not going to make up for the loss of Brewer.  Both players are below average this season, and the combined productivity of these players is in the negative range.

One suspects that Brewer’s contributions were simply not evaluated correctly by O’Connor. Scoring is easy to see and appreciate in a basketball game.  The object of the game is to put the ball in the hoop, so we tend to undervalue the people who are not doing this often.  But for the ball to go into the hoop, other stuff has to happen.  Steals and turnovers are important.  And when it comes to shooting, it’s efficiency that matters.  In other words, not making mistakes (committing turnovers and missing shots) is important.  But it is hard to see a player not make mistakes.  Consequently, Brewer is now going to avoiding mistakes for the Memphis Grizzlies.

It’s unlikely that Memphis will make the playoffs this season.  And furthermore, it’s unlikely that Memphis is going to start Brewer.  The Grizzlies starting shooting guard is O.J. Mayo, a player who scores 17.7 points per game (with a 0.075 WP48 after 54 games).  Yes, Brewer offers more production overall, but again, scoring is what dominates perceptions in the NBA (again we see this when we look at the free agent salaries, the NBA draft, the allocation of minutes, and the assignment of awards). 

So Brewer has learned a valuable lesson. Don’t listen to everything your coach says. When he recovers from his injury he will find himself coming off the bench for an inferior (although improving) team.  In essence, Brewer has been demoted.  All of this because he helped his team win by doing something other than scoring.

Let me close by noting that I obviously agree with Deron Williams. Here is how it was reported he reacted to the trade in the Salt Lake City Tribune (one of the papers delivered to my house each morning).

A day after the Jazz traded his self-described “little brother,” Deron Williams sharply questioned Friday the direction the team was going after the deal that sent Ronnie Brewer to Memphis for a protected first-round draft pick.

“I think if we’d make a trade it would be something a little different than that,” Williams said at the pregame shootaround. “You look at all the teams that are getting better around the West and we essentially get worse, if you ask me.”

Williams didn’t hide his frustration in talking to reporters and broadcasts, starting out by saying, “I really ain’t got nothing much good to say about the trade,” and declaring it was “pretty safe to say” his feelings were shared among the team.

Asked whether the Brewer trade affected his thinking about his long-term future in Utah, Williams uttered words that could reverberate in the team’s front office: “That’s why I signed a three-year deal.”

So apparently Williams doesn’t think Brewer will be easy to replace.  And it’s safe to say, if Williams departs the Jazz, he won’t be easy to replace either.

- DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.