One theme we return to again and again in The Wages of Wins is the importance of reference points. Whether we are discussing competitive balance in sports or the consistency in player performance across time, we had to be careful to note our reference point when we made statements like, “baseball had a relatively high level of competitive balance in the latter 20th century”, or “basketball players are relatively consistent in performance across time.” Each of these statements required a reference point for the statements to have any meaning.
Two days ago I took a first glance at the 2006 rookie class. Here were a few first impressions:
1. Rookies were not playing many minutes.
2. The rookies who did play did not appear to play very well.
3. Lottery picks in 2006 did not play nearly as well as their draft position would suggest.
All of this suggested that the draft in 2006 was not working out as well as the teams who participated would like. But was the 2006 draft unusual? Or is what we see in 2006 the typical pattern?
To answer these questions we need a reference point. One possible reference point is what we observed with respect to the 2005 draft class.
Last season 23 rookies played at least 1,000 minutes in the NBA. As this table shows, only six of these rookies posted a Wins Produced per 48 minutes [WP48] that was above average (average is 0.100). Leading the way was Chris Paul, whose performance was one of the ten best seen in the league. As noted last May, Paul’s productivity far eclipsed everyone else in the 2005 draft.
Trailing Paul in second place in this class was the surprising David Lee. Lee was the last player chosen in the first round in 2005. When one considers the early returns on Renaldo Balkman, it appears Isiah Thomas has a knack for finding productive role players (more on that story tomorrow).
After Lee we see Andrew Bogut, the first player chosen in the draft. Although Bogut was above average his freshman season, he and Paul were the only lottery picks to post above average performances. Deron Williams appears to be coming on in his sophomore season, but a number of the other lottery picks have yet to generate a substantial return.
Let’s compare what we have learned from the 2005 draft with the early returns on the 2006 class. For the 2006-07 season we see only sixteen rookies who are currently on pace to play 1,000 minutes. And of these players, seven have a WP48 that is below zero. So relative to last year, this draft class looks a bit worse. It is still early in the season, though, and one would hope that the rookies this year will get better as the season progresses.
What about lottery picks? Lottery picks in 2006 are generally not playing well. And only two lottery picks in 2005 finished their first season with an above average WP48. So lottery picks performing poorly is a pattern we see in our data set, which so far consists of a whopping two observations.
What of other seasons? Do we see these patterns repeat themselves?
I think I will save the answers to these questions for another day. Obviously two data points are not enough to reach definitive conclusions. As we look back at other draft classes, though, we might be able to learn something useful for the 2007 NBA draft (then again, maybe we won’t).
One last note before ending this post. So far rookies, in general, appear to perform at a level that is below average. Obviously at some point some of these players must get better. When does that typically happen? I have participated in some research on the topic, but I am not at all sure about the answer. My hope is that one of my co-authors, who is currently on a very extended vacation (and I am not naming names, but he knows who he is), will get back to work soon and help me figure out the answer. When that happens, I can share what we find.
Our research on the NBA was summarized HERE.
Wins Produced and Win Score are Discussed in the Following Posts