Getting Too Much to the Point in Memphis

Posted on December 3, 2007 by


The Memphis Grizzlies were the worst team in the NBA in 2006-07.  And this was not an accident.  Not only did this team lead the NBA in losses, it also posted the worst efficiency differential [offensive efficiency minus defensive efficiency] in the Association.

When we look at Wins Produced, we can see why this team played so poorly. 

In 2005-06 the Grizzlies received 42.0 Wins Produced from Pau Gasol (14.1 Wins Produced), Mike Miller (10.9 Wins Produced), Shane Battier (8.7 Wins Produced), and Eddie Jones (8.3).  The remainder of the team only produced 8.9 wins. 

Last season the Grizzlies received 8.1 wins from everyone not named Gasol, Miller, Battier, and E. Jones; a mark fairly similar to what the rest of the Grizzlies did the year before. But in 2006-07 Battier gave this team nothing because he was traded to Houston. E. Jones only produced -0.3 wins because he was first hurt and then released (E. Jones did produce 3.8 wins for the Heat after he left Memphis).  Gasol and Miller did combine for 19.7 wins, but when the rest of the roster produces fewer than ten wins; your team is going to struggle.

Getting the Point in the Draft

The reward for this struggle was a high lottery pick.  As noted, Gasol and Miller were the primary producers of wins in 2006-07.  Gasol plays center-power forward and Miller plays shooting guard-small forward.  In other words, the Grizzlies’ top two players can contribute at four of the five positions on the court.  Obviously this tells us the Grizzlies should have focused on acquiring a point guard in the draft.

There is only one problem with this argument.  In 2006 the Grizzlies selected Kyle Lowry with the 24th pick in the NBA draft.  Last season Lowry produced just 1.1 wins, but that was because he only played 175 minutes.  In these minutes he posted a 0.292 WP48 [Wins Produced per 48 minutes].  This is well above average (average is 0.100), and suggested that point guard was not an area of weakness on this team.

Of course Lowry only played 175 minutes due to injury.  Is this enough time to evaluate a player?  What if he can’t come back from injury?

With such questions in the air, the Grizzlies decided to select point guard Mike Conley in the 2007 draft. When we look at the college numbers from 2006-07 we see that Conley was the most productive point guard selected out of the NCAA in the draft.  So if you were going to take a point guard, Conley looked like a pretty good choice.

And thus far in 2007-08, Conley has lived up to this expectation, posting a stellar 0.245 WP48. Such a mark leads all point guards playing for Memphis, so it looks like taking Conley was a wise selection.

Okay, there’s a bit more to the story.  Conley got hurt after only playing 72 minutes.  So he really hasn’t played enough for us to evaluate his performance.

And there is one more problem.  Lowry has clearly recovered from his injury.  Thus far in 2007-08, Lowry has produced 1.6 wins and posted a 0.185 WP48. In addition, Damon Stoudamire – now in his 13th NBA season – has produced 1.1 wins with a 0.164 WP48 in 322 minutes at the point guard position.  With such production at this position, it’s going to be difficult finding time for Conley when he returns from injury. 

I would note that eventually Conley can certainly take over for Stoudamire.  Again, Stoudamire has been playing since 1995 (a draft I reviewed in October).  In his career he had produced 56.9 wins with a 0.097 WP48. In sum, Stoudamire has been an average point guard in his career.  And he has been remarkably consistent.  His lowest WP48 was in 2002-03 when his production stood at 0.067.  His best mark was 2004-05 when his mark was 0.117.  

So Stoudamire is average, which is not bad (not good, but not bad).  Plus he is old.  If Conley can produce, and it looks like he can (although we don’t know that for sure), he will eventually take Stoudamire’s job. 

But can Conley supplant Lowry?  And should he? 

Looking at the Grizzlies Today

When we look at the Grizzlies roster, we see that this team needs help.  Yes, with the improved play at the guard position, the Grizzlies have clearly improved.  We can see this when we look at Table One, which gives us two projections of how many wins we can expect this team to have by the end of the season.

Table One: Projecting Memphis in 2007-08

If we assume each player on Memphis will play as he did last year (except for the rookies and Casey Jacobsen), the Grizzlies should expect to win 44 games this season.  The improvement we see is from Kyle Lowry playing major minutes (and offering a 0.292 WP48 which is probably unrealistic) and Juan Carlos Navarro taking over at shooting guard.  Navarro is only posting a 0.078 WP48, which is below average.  But he replaced Dahntay Jones and Eddie Jones, two players who were in the negative range last season.  In other words, replacing players who played very, very bad with a player who is just a bit below average can lead to a few more wins. 

Again, the 44 win projection assumes players did what they did last season.  If we look at what these players are actually doing this year, we can expect this 6-10 team to win 38 games.  This team has an efficiency differential of -1.0 (last season it was -5.3), so we should not be surprised to see such a projection for a team that is in last place and has only won 38% of its contests. 

Although we can see the Grizzlies have clearly improved, they have not improved as much as the first projection suggested. And that’s because Pau Gasol has not performed very well this season.  Entering this season Gasol had produced 64.4 wins in his six year career with a 0.200 WP48.  This year his WP48 is only 0.059. And when you look at the individual stats you see a decline across the board.  I am not sure what has happened to Gasol (injury perhaps?) but maybe the folks at 3 Shades of Blue can shed light on this decline.

The decline in Gasol’s performance has been offset by improved play from Miller, Stoudamire, Stromile Swift, and Rudy Gay.  Miller has played very well before, so his performance might continue.  As noted, though, Stoudamire has never played this well.  And the same story can be told for Swift.  Meanwhile, Gay is still below average, he just isn’t as far below average as he was his rookie season.

Summarizing the Story

Let me try and summarize the story.  The Grizzlies do have some talent.  Miller is good. And Gasol should be good.  Furthermore, and the point of this story, is that this team may have two very good young point guards.  But both Lowry and Conley are two small to move to shooting guard.   So what should this team do in the future?

1. The team could keep both point guards.  Unfortunately there are only 48 minutes per game at the position, so that means leaving some talent on the bench.

2. The team could trade one of the point guards. That might be a problem.  Lowry’s strength is rebounding and getting to the free throw line.  He is below average as a scorer and passer.  So his trade value might be low.  And I am fairly certain the Grizzlies didn’t draft Conley in the 2007 lottery to just trade him away.

Right now this team doesn’t need to make this decision.  As I keep saying with each of these columns early in the season, “it’s early in the season”.  So all the numbers I cited for 2007-08 could change, and thus, so will the decisions facing this team.

Still, if the numbers for Lowry and Conley hold, Memphis will eventually have to make a decision about who to start and who ultimately to keep.  At this point only one thing is certain. The team is not going to keep starting Stoudamire.

– DJ

Our research on the NBA was summarized HERE.

The Technical Notes at 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.