Is Age Finishing Shaq?

Posted on December 10, 2009 by


There is one certainty in sports.  For every athlete, a day will come when a player’s production declines.  The decline doesn’t always happen in a consistent pattern (i.e. ups and downs are still possible at an advanced age).  And significant drop-offs don’t happen at the same age for every player.  All we know is that someday, every athlete stops helping.*

To illustrated, consider the WP48 [Wins Produced per 48 minutes] marks Kareem Abdul-Jabbar posted after the age of 30.

30: 0.394

31: 0.348

32: 0.340

33: 0.274

34: 0.206

35: 0.182

36: 0.143

37: 0.225

38: 0.145

39: 0.115

40: 0.045

41: -0.056

An average player posts a WP48 of 0.100.  Kareem was three times the average player prior to 33 years of age.  He was still twice the average player before age 35.  At the age of 36 his WP48 numbers were still above average, but clearly Kareem was no longer outstanding.

But then at age 37, Kareem was suddenly twice the average again. This spike, though, was temporary. His performance at age 38 looked quite similar to what we saw at age 36. And at age 39 he was just barely above average.  In his last season – at age 41 – Kareem’s performance dipped into the negative range.  Yes, one of the greatest centers to play the game was suddenly reduced to Spencer Hawes (Hawes posted a -0.021 WP48 last season) once he passed 40 years of age. 

Now let’s consider the WP48 numbers of Shaquille O’Neal after the age of 30.

30: 0.321

31: 0.287

32: 0.292

33: 0.220

34: 0.106

35: 0.116

36: 0.167

Like Kareem, Shaq entered his 30s as a player producing at a rate well beyond what we see from an average player.  By his mid-30s, though, Shaq was no longer the same player.  Yes, he was still above average.  But the 20-something Shaq that dominated the NBA had vanished.

Although Shaq has clearly declined, the Cleveland Cavaliers still took a chance on acquiring his services.  This chance was motivated by the age and health-status of Ben Wallace.  Last season Big Ben posted a 0.159 WP48.  But his health limited his availability in the regular season and his production in the playoffs. 

So this past summer, Wallace was sent to the Phoenix Suns for Shaq.  It was suspected that Big Ben would retire.  But after being released by Phoenix, Big Ben signed with the Pistons and is now leading Detroit in Wins Produced.

Meanwhile, the Cavaliers decided to take a chance on Shaq.  Yes, it’s likely that people in Cleveland knew that “someday” O’Neal would stop helping.  But it was hoped that “someday” would happen after the 2009-10 season (and after a championship parade in Cleveland).

Unfortunately, there’s evidence that “someday” is happening in Cleveland. 

Table One: The Cleveland Cavaliers after 22 games in 2009-10

Table One reports the Wins Produced of the Cleveland Cavaliers after 22 games.  In addition, it reports what we could have expected from this team had player performance not changed from what we saw last season.  With respect to most players, the assumption of constant performance doesn’t lead us too far astray.  There are, though, two notable exceptions. 

Before getting to Shaq, let’s talk about Zydrunas Ilgauskas.  Last year, Ilgauskas posted a 0.093 WP48 [a mark very close to average]. This season – at the age of 34 – Ilgauskas is posting a -0.051 WP48.  Yes, Ilgauskas is now a 41 year-old Kareem.  Consequently, Ilgauskas is on pace to produce 5.5 fewer wins than what his performance last year would suggest.

For much of his career, Ilgauskas has been an above average (i.e. WP48 in excess of 0.100) but not outstanding performer (i.e. WP48 less than 0.200).  As noted above, though, Shaq has often been amazing.  And even as age has taken its toll, he has still maintained an above average WP48 mark.

At least, until the 2009-10 season began.  Shaq’s WP48 mark this season has only been 0.017.  So at the age of 37, Shaq is a bit worse than the 40-year old Kareem. 

Again, the impact of age is not constant.  So it’s possible Shaq – and Ilgauskas – will bounce back.  But if they don’t, the Cavaliers are simply not going to be what I envisioned before the season started. 

As Table One notes, had everyone maintained what we saw last year, the Cavs would currently be on pace to win 69.3 wins.  Such a record would make the Cavs a clear contender to win a title (which is what I suggested before the season started).  The declines we see with respect to Shaq and Ilgauskas, though, takes 10 wins off of the Cavs projected totals. And that clearly drops the Cavs behind the Celtics and Lakers.

Let me close with two observations.  First – as Table Two suggests – the problems for O’Neal and Ilgauskas are primarily related to shooting efficiency.  Most other numbers for these players haven’t changed from last year.  But shooting efficiency from the field – and for Shaq, also from the line – has declined for both players. And consequently, overall productivity has declined considerably.

Table Two: Shaquille O’Neal and Zydrunas Ilgauskas

Secondly, some people might remember that I have already commented on the Cavaliers troubles.  This first comment, though, came after just seven games (so the sample was even smaller than what we have today).  And at that time, Shaq was playing better (although Ilgauskas was still very bad).  Since that time, though, Shaq has missed time due to injury.  Of course, injuries are part of the problem of age.  But if Shaq can recover and start producing at the level we saw last year, Cleveland can get better.  If not, a season of promise in Cleveland may end up a very big disappointment.

– DJ

* – we talk about age and performance in the NBA in more detail in our next book.

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.