One of the most frustrating aspects of analyzing a sport with many moving parts and player interactions -- like basketball -- is that diminishing returns rear their ugly heads just about everywhere.
Unlike baseball, a sport in which the marginal value of a given action at the individual level is essentially the same at the team level -- upgrading to a new third baseman who hits 10 more home runs than the old one, for instance, will result in 10 more home runs for the team -- adding, say, a prolific individual rebounder does not necessarily mean the team will reap big benefits on the boards.
Why is this? The main reason is that the act of rebounding involves players competing not just with opponents but also with members of their own teams. There are only so many boards to go around on a team, and often seemingly productive rebounders rack up gaudy totals in part by "stealing" boards from teammates, rather than adding extra rebounds to their teams' bottom lines.
We can detect this phenomenon in a number of ways. First, just take a look at the greatest rebounding seasons of all time, and the impact they had on teammates' rebounding numbers.
Since the NBA-ABA merger in 1976-77, there have been 689 instances of a player logging at least 1,500 minutes for a single team in a season while grabbing at least 16 percent of available rebounds while on the floor. If diminishing returns aren't a big factor, we'd predict their teammates to have roughly average rebounding percentages. However, out of those 689 players, only three had teammates who collectively posted an above-average rebounding rate. This is either a remarkably improbable coincidence or strong evidence that at least some of the boards pulled down by big-time rebounders have been taken away from teammates, not opponents.
The phenomenon also occurs when examining the impact of adding a great rebounder to a team. Since the merger, the 165 NBA teams that added a player who met our prior qualifications only improved their rebound percentages by an average of 0.4 points, despite adding rebounders who were, on average, 7 percentage points better than average. In fact, 62 of those 165 teams (38 percent) failed to see their rebounding performance improve at all.
The following five players are examples of this phenomenon -- their individual rebounding numbers are impressive, but they don't seem to add much to their teams' rebounding performance when they're on the floor.