J. of Basketball Studies
Power of Parity
|NBA Preview Pages|
A RECENT ESPNet poll asked whether one would rather build a franchise around 23 year-old Rasheed Wallace or 21 year-old Kevin Garnett. It closed with about 85% percent favoring Garnett.
Wallace has a reputation as a malcontent. Garnett does not. Wallace played on a good team that got very little press. Garnett got named to the All Star Team along with teammate Tom Gugliotta because his team did much better than expected. Wallace signed a contract extension for a measely $13M/year. Garnett turned down a contract worth $16M/year before signing for $20M/year.
At this point, the results of the ESPN poll, whether accurate or not, has been heavily influenced by what the media has told you.
Here are some facts:
Beyond the facts, we can guess about the future of these players or we try to predict it. Over the summer, a reader named Guy Tyler approached me with a method he developed for predicting the career of players. Though the method has helped Guy win his fantasy league three years in a row, it is nowhere near as advanced as any of us would like. Still, it gives us something to argue about...
The method projects the number of minutes played per game and something called PAR -- points plus assists plus rebounds -- per 48 minutes. It uses the stats from one year and a player's age to predict the next year. It doesn't explicitly predict the number of points a player will score or their turnovers or their assits.... you get the idea. It simply predicts PAR/48 with advancing age and that PAR/48 is meant as an approximate value of the player.
Guy demonstrated the method using Kevin Garnett and I will reproduce that here.
|Kevin Garnett's Projected Future|
Overall, these numbers are not particularly good. They don't approach those of some great players like Olajuwon, Drexler, Malone, or even the not-so-great Tom Chambers. Guy suggests that his method may not be accurate because he didn't develop the formulas using players like Garnett. That is certainly true -- there weren't many 19 year old rookies before Garnett -- but that doesn't allow us to dismiss the results completely.
Wallace better fits the profile of the method since he came into the league as a 21 year-old. The method predicts the following future for him:
|Rasheed Wallace's Projected Future|
Basically, the method projects Wallace to be slightly less effective than Garnett over their careers. These sort of numbers justify the contracts that they received.
...But Is It Right?...
Despite the general conclusions of the method, I have two concerns. First of all, the method only marginally increases the accuracy of prediction over a fairly random estimate of a player's career projection. Second, the thing that it predicts -- PAR -- is not a great indicator of winning.
Finally, Wallace really is a much more polished offensive player than Garnett. He does not play defense as well and that is crucial for a center. But Wallace has offensive moves that he can go to in order to score. Garnett does not yet have this. He may not develop it either. In short, injuries and attitude aside, Wallace is a better player right now. In the long run, I don't know if those can be put aside.
Neither of these guys are as good as Shaquille O'Neal, whose predicted future looked like this after his first two years:
Shaquille O'Neal's Predicted Future
After First Two Years
So why do they make as much money as O'Neal?