The Wheel of Fortune Turns on Don Nelson

Posted on April 1, 2010 by


Lenny Wilkens leads all NBA coaches in league history with 1,332 career regular season victories (vs. 1155 losses for a 0.536 career winning percentage).  Don Nelson’s career record, though, currently stands at 1330-1060 (a 0.556 career winning percentage).  So with just three more wins, Nelson will assume the top spot in league history.

Unfortunately for Nelson, this is one of the worst seasons in his career.  After 74 games, the Golden State Warriors record stands at 21-53.  With a 0.284 winning percentage, the Warriors are only on pace to win two more games in 2009-10.  And if that happens, Nelson and Wilkens will finish tied for first after this season.

When we look at the Warriors this season, it looks like this record should have been a bit easier to break. The Warriors efficiency differential – and Wins Produced (as Table One indicates) – suggests the Warriors should have already won 27 games.  So Nelson and the Warriors have been a bit unlucky this year.

Table One: The Golden State Warriors after 74 games in 2009-10

And when we consider the injuries on this team, we see even more misfortune.  Andris Biedrins – 0.200 WP48 [Wins Produced per 48 minutes] – and Anthony Randolph – 0.167 WP48 – have each missed 41 games this year.  Had these players been available the entire season, Nelson would now be adding to his record.

But injuries and bad luck did happen, and therefore, the Warriors may not be able to set the record for Nelson.  Despite this potential failure, there has been some good news in 2009-10.  Stephen Curry – the team’s lottery pick in 2009 – currently leads Golden State in Wins Produced and is posting a 0.124 WP48 [average is 0.100]. 

And Curry is not the only above average rookie. Reggie Williams left VMI in 2008.  And back in May of that year, Erich Doerr (who often does draft analysis at the WoW Journal), had this to say about Williams: NCAA scoring leader Reggie Williams posted a strong pace-adjusted .209 PAWS/M, but played a mere one game against a top 100 team.

Despite strong college numbers, Williams didn’t appear in the NBA until 2010.  And thus far – after just 441 minutes – Williams has posted a 0.146 WP48.  So it looks like the Nelson and the Warriors have found another above average performer.

The play of Williams (if it is for real) and Curry suggests the future in Golden State could look better. If Nelson returns next season, it looks like he could be coaching Curry, Williams, Randolph, and Biedrins.  All of these players are above average performers.  And if these players can be healthy and available, the Warriors should do much better than the 0.284 winning percentage seen this season.

Let me close with two observations. First, Nelson has not always been a victim of bad luck.  As noted last August – in The Squid that Saved Don Nelson (I still love that title) – one can argue that Nelson’s success as a coach was owed to the Milwaukee Bucks selection of Sidney Moncreif.  Had that not happened (and it woudn’t have happened if Dick Vitale and the Pistons made a different choice), Nelson’s career might not have been as notable.

That being said, it should be noted – as mentioned in Stumbling on Wins – Nelson is one of the few coaches we investigated who had a statistically significant impact on player performance.  So Nelson may not be the greatest coach ever (or he may be), but it certainly appears that he has made a difference.  And with a bit of good fortune this year, this particular difference-maker might become the winningest coach in NBA history.

– 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.