Bynum Comes Back

Posted on November 13, 2009 by


What a difference a day made
Twenty-four little hours
Brought the sun and the flowers
Where there used to be rain

Okay, it’s actually snowing in Cedar City.  That aside, yesterday is a bit different from today. And it’s not just the weather that changed.  Yesterday I argued that Andrew Bynum isn’t quite back.  Then last night, Mr. Bynum scores 26 points (on just 18 field goal attempts) and grabs 15 rebounds.  His WP48 [Wins Produced per 48 minutes] for the game was 0.496.  And now his season mark is 0.229.  No, he isn’t quite back to what he was in 2007-08.  But after eight games, he’s clearly offering more than what we saw last year.

As a team, the Lakers also look better today. Prior to last night’s game the team’s efficiency differential was 5.4.  Today it’s 7.1, and Pau Gasol hasn’t played yet. 

So what’s the moral of the story?  Writing about stats and basketball before the first ten games has been played is a risky business (even if you are just looking at the generally consistent box score numbers).  One game can have a significant impact on your story.  Yesterday I argued that Bynum wasn’t really much different from what he was last year, and therefore, the Lakers had not really improved enough to repeat as champions. Today… well, Bynum is still not back to what he was in 2007-08 (unless he just keeps doing what he did last night).  But if I were a fan of the Lakers, I would certainly be a bit more optimistic.

– 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.