There’s some evidence that there’s diminishing returns in rebounds, specifically defensive rebounds. A monster rebounder coming to a lineup may not have a 1 to 1 effect in adding rebounds over the guy he replaced. For example, if Harrington and his 6 rpg gets replaced by Biedrins and his 10 in the same time, the Warriors wouldn’t necessarily see a 4 rpg increase. There’s some evidence that it would be smaller, that some of the rebounds Biedrins gets would be rebounds someone else would otherwise get with a weaker rebounder in the middle. This may explain some of the ‘emptiness’ of some rebounds, but there’s much more to it.

Things like this seem to be true with many of the superexceptional rebounders. But the reasons for this could be many. It may be that when a team can put a Dwight Howard or a Ben Wallace of few years back in the middle, they tend not to surround him with other good rebounders, but take advantage of his skills by running out a smaller rest of the lineup. Use someone like that and you may favor two smaller forwards for the offensive punch they provide. It may not be that there’s diminishing returns so much as that coaches favor balance to the lineup.

The real issue with Murphy was that there did appear to be negative returns on his rebounds at the team level. I suspect this is because he was such a poor defender that there were too many made baskets by the opposition. He’s a good rebounder in getting those boards that are available, but in giving up so many easy shots, he decreases the pool of boards available. He also appears to take boards that would otherwise go to teammates as the rebound rates declines as well.

On the value to the team of a defensive vs. offensive board, they are on average equal. There is no real difference in terms of win probability. The offensive board negates a missed shot, keeping a possession alive while a defensive board starts a possession with a zero recorded for the opponent. Exactly equal in terms of what they mean for relative possession usage. Since possessions, by definition, are essentially equal between teams in a game, this makes complete sense. A possession ‘wasted’ by the opposition (which is what a defensive rebound means) is equivalent a possession wasted by your own team, which is what a defensive rebound by the opposition means. An offensive rebound means your opponent didn’t acquire that defensive board, meaning you retain that value rather than giving it away. It’s entirely logical and also appears to be empirically true when regressing rebound totals of either type against winning percentage.

Offensive rebounds may appear to be more impressive because they often lead to easy putbacks and points. These points though are recorded on their own. For Biedrins, who hits most of his shots inside, it probably does help to have the high number of O boards because he’s converting 60% of these into an immediate 2 points. But that’s not the rebound, that’s the shot. A guy who gets packed in the paint more often or a guy who seems to miss layups regularly doesn’t realize this potential if he takes the shot. It’s best to think about the two issues, the rebound and the subsequent shot separately as they do seem to be skills that aren’t always linked.

I think Murphy’s problems on the offensive glass were largely because it’s tough to board when you’re camped out looking for your three-ball to fall. Biedrins doesn’t have ‘range.’ I suspect that this may be a good thing if his ‘range’ meant that he’s see a drop in rebounds in favor of taking jumpers that he’d hit less than 50% of the time.