NBA Rookies and their 'Win Production'

| by David Berri

Once again an award has been given by the NBA. And once again, the leading scorer eligible for the award is the winner. Tyreke Evans – who led all rookies with 1,450 points scored – has been named Rookie of the Year. Stephen Curry – who finished second in scoring – finished second in voting. And Brandon Jennings – the third leading scorer – finished third.

When we move past this one element in the box score and consider a player’s production of wins we see that Evans and Curry are… okay, they are still the top rookies.  Curry produced 9.1 wins with a 0.151 WP48 [Wins Produced per 48 minutes]. And Evans produced 8.9 wins with a 0.159 WP48.  So one could argue that either player was the most productive rookie, and therefore, this is not quite the familiar story we always see. Yes, the scorers were given the award.  But since a scorer can produce wins, the writers would have come to the same conclusion whether they looked at scoring totals or all the elements of the box score.

At least, that’s true if you just look at Evans and Curry.  If we move further down the list we see two scorers who are not as productive. The first is Jennings.  Of those receiving any consideration for this award, Jennings is the only one still playing.  Across the regular season, though, he only produced 3.8 wins with a 0.067 WP48.  Yes, Jennings can score.  But he is not yet producing wins in large quantities (at least, not in the regular season).

Jennings, though, is much more productive than Jonny Flynn. With the 6th pick in the 2008 NBA draft, the Minnesota Timberwolves selected Flynn. And although he finished fourth in scoring in this rookie class – and earned two third place votes for Rookie of the Year — Flynn finished the season with -1.4 Wins Produced and a -0.030 WP48. 

It’s interesting to note that every rookie who received consideration for this award was drafted in the first round.  DeJuan Blair produced 7.9 wins with a 0.255 WP48. And Jonas Jerebko produced 6.0 wins with a 0.128 WP48. Neither player, though, received even a third place vote for Rookie of the Year.

The voters focus on first round picks led me to wonder: How does the first round of the 2009 draft look now that we have seen these players for one year? For an answer we turn to Table One.

Table One: Reviewing the First Round of the 2009 Draft

From Table One we can see the Wins Produced and WP48 of each player taken in the first round (with the exception of those that did not play).  As one can see, only four of the rookies taken in the top ten finished in the top ten in Wins Produced.  Blake Griffin and Ricky Rubio didn’t get a chance to play.  And the minutes of Hasheem Thabeet and Jordan Hill were limited.  But DeMar DeRozen and Flynn, didn’t perform well.  And given what each player did in college, this is not surprising.

When we look at the entire first round we do not see much of a link between where a player was drafted and his first year performance.  In this small sample, the correlation coefficient is only -0.22.  This is actually a similar result to what we report in Stumbling on Wins (where we looked at a much larger sample of players across more than just the first season of a player’s career).   NBA productivity and draft position simply do not have a very strong link.

This is similar to what we see for quarterbacks  in the NFL. The reasons, though, are different.  Much of what we know about a quarterback before the draft (i.e. most college stats and the combine data) is not correlated with future performance. This means choosing quarterbacks on draft day is really quite difficult (more on this in a near future post). There is a relationship, though, between what players do in college and what they do in the NBA.  It is not a perfect relationship.  But there is a statistical link.  Despite the link between college and professional performance, though, NBA decision-makers make more mistakes than expected on draft day.

Much of this is due to a very familiar story.  Yes, too much emphasis is placed on scoring.  Scorers like Jonny Flynn tend to go first.  And these players don’t always produce many wins.

Let me close by noting again that scorers can certainly produce wins.  Evans and Curry were the most productive rookies.  And LeBron James – a very prolific scorer – was the both the Most Productive Player (MPP) and MVP. More on that story in a future post.

- DJ

Our research on the NBA was summarized HERE.


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.