Beasley or Boozer?

Posted on July 30, 2009 by


A few days ago John Hollinger of made the following observations (Insider Access Required):

1. Michael Beasley averaged 22.4 points per 40 minutes with a 17.29 PER.

2. This is the same PER Boozer recorded. 

3. Therefore, “upgrading from Beasley to Boozer has no impact whatsover. None. Zero.”

4. Odom has a 16.60 PER so he is not as good as Beasley.

5. Adding Odom is not an improvement over Beasley.

Hollinger’s analysis obviously depends upon PERs.  If we look at WP48 [Wins Produced per 48 minutes] we see a very different story.  Here is what each player did last year at power forward:

Michael Beasley: 0.049 WP48

Carlos Boozer: 0.201 WP48

Lamar Odom: 0.220 WP48

An average player posts a 0.100 WP48.  So Beasley was below average last year while both Boozer and Odom performed at a level that doubled the output of an average power forward.

When we turn to the individual statistics we can see why these differences are so large:

Table One: Comparing Beasley, Boozer, and Odom

Table One reports what each player did with respect to each box score statistic in 2008-09.  Let’s focus first on Beasley.  Relative to an average power forward Beasley was above average with respect to scoring and personal fouls.  With respect to all other statistics Beasley was below average.  So why does PERs rate Beasley so highly?  As I have noted in the past, PERs will reward a player for taking more shots.  Of these three players, Beasley took the most field goals per 48 minutes.  As a consequence, he managed to score 26.8 points per 48 minutes.  But Beasley’s shooting efficiency was only slightly above average.  So his scoring was not a tremendous help to the Heat.  And he cost the team by his inability to do anything else. 

When we look at Odom and Boozer we see two players who are at least as efficient as Beasley with respect to scoring.  Both Odom and Boozer also contribute with respect to rebounds, steals, and assists.  Consequently, Boozer and Odom make larger contributions to team success. 

So would Boozer and Odom help the Heat?  Well, here is what the first and second string (depth chart from ESPN) on the Heat did last year:

First String

PG: Mario Chalmers [5.5 Wins Produced, 0.101 WP48]

SG: Dwyane Wade [22.2 Wins Produce, 0.350 WP48]

SF: James Jones [-0.3 Wins Produced, -0.026 WP48]

PF: Udonis Haslem [5.0 Wins Produced, 0.093 WP48]

C: Jermaine O’Neal [-1.1 Wins Produced, -0.067 WP48]

Second String

PG: Chris Quinn [1.2 Wins Produced, 0.060 WP48]

SG: Daequan Cook [0.4 Wins Produced, 0.009 WP48]

SF: Yakhouba Diawara [-1.5 Wins Produced, -0.083 WP48]

PF: Michael Beasley [2.1 Wins Produced, 0.049 WP48]

C: Jamaal Magloire [0.8 Wins Produced, 0.052 WP48]

These ten players combined to produce 34.2 wins last season.   The team won 43 games, but it has lost Shawn Marion (in the Jermaine O’Neal trade) and Jamario Moon (signed by Cleveland).  Marion and Moon combined to produce 9.4 wins last season.  For the Heat to make the playoffs again this production needs to be replaced.

One option would be to acquire Odom and Boozer.  Odom produced 10.6 wins last year and has posted 0.184 WP48 for his career.   Boozer was hurt much of last season.  For his career, though, he has produced 74.9 wins and posted a 0.256 WP48. 

Now Odom turns 30 this fall and Boozer is prone to injury. So both players come with question marks.  What isn’t a question mark (at least, I don’t think this is a question mark) is the value of what each player has done in the past.  And if we look at more than scoring, it seems clear that Odom and Boozer offered more than Beasley last season.

– DJ

The WoW Journal Comments Policy

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.