And now back to our regularly scheduled program...
One way to look at the value of a player, in terms of rebounding, is the difference in team rebounding rate with him on and off the floor. You can think of this like a +/- for rebounding. The focus of most discussions/debates on rebounding (see here or here for recent examples) tends to be on forwards and centers, simply because they get the most rebounds. As it's often good practice to test assumptions, I wanted to look at the effects of on/off rebounding rates for guards, to see what effect, if any, they actually do have. It turns out, somewhat to my surprise, that the "marginal" value of guard rebounding for many teams is quite high (as can be seen from the tables). Note that I keep using the word value and not ability. It is very important to distinguish these terms, because they are not necessarily the same, in terms of on/off rates. Here are some of the factors that could generate varying on/off rates (there are surely many others):
- The player is genuinely a "good" rebounder (relative to other guards in the league).
- The player has a "poor" backup (relative to other guards in the league).
- Conversely, the player may have a good backup(s), so his on/off rates appear lower.
- A starter is surrounded by better rebounders relative to a backup who comes in with the second unit.
Here, (1) represents the player's ability, whereas (2) and (3) represent, respectively, high and low marginal value. (4) is a confounding factor that mucks up value and ability simultaneously, and should be controlled for. It should be clear why it is so important to differentiate the two concepts. Right? Ok, so let's get to the data. The following table lists the composite on/off rebounding (offensive + defensive) rates in 2009-10 for all guards who played over 800 minutes during the season (Note: a value of +1% would mean that the team rebounding rate drops by 1 percentage point when that player is off the floor):
On/Off rates were collected from BasketballValue.com.
|Ranking||Player||Team||Total Rebounding Value|
|52||Murray, Ronald (Flip)||CHA||-0.5%|
Having seen the data, one is tempted to conclude that Jrue Holiday is the "best" rebounding guard in the league, and that Ben Gordon is the "worst" rebounding guard. Of course, I wouldn't do that, as I've just explained above how these rates reflect marginal value, and not necessarily player ability. I'll save those debates for future posts, when I have more data to back up my conclusions. Finally, it should be obvious that everything I have said and done here can (and should) be done for other positions. Comments? Suggestions? Fire away...