I’m not a hardcore stats geek. I find Sabermetrics and the like interesting, but I don’t devote my life to it. That being said, there is one statistic that absolutely drives me out of my mind – rebound margin. That’s one of the stats that is very frequently talked about when people are handicapping basketball and trying to sound smart. It is also an incredibly stupid statistic that doesn’t paint an accurate picture of what a team is actually capable of. At best it is misleading, and at worst it is totally deceptive. If you are relying on rebound margin to make your picks then you aren’t making the best betting decisions you can. By understanding the problems with the stat you can see what the issues are and understand how you can do better.

First, a quick definition. Rebound margin is simply the difference between the number of rebounds that a team averages per game, and the number of rebounds they allow their opponents to average per game. The theory goes that when you are out-rebounding your opponents on a consistent basis you are controlling the ball and are going to enjoy success. That’s sounds pretty good. So, why is this such a useless stat? Let’s take a look:

Combines offensive and defensive rebounds – Offensive and defensive rebounds are very different things, yet rebound margin doesn’t distinguish between them at all. A defensive rebound is unquestionably a good thing – it means that the offensive team missed on a chance to score a basket. There is much more of a grey area surrounding offensive rebounds, though, as we will see in a little while. Rebound margin values both types of rebounds equally, so it lacks meaning as a result. More significantly, by treating both types equally you can’t get a sense of how teams are out-rebounding their opponents. For example, they could be extremely good at offensive rebounding but hopeless at defensive rebounding, or vice versa. You couldn’t tell which is the case by looking at their rebounding margin. A good stat is one that tells a story and paints a clear picture of what a team or a player is good at and where they are vulnerable. Rebound margin forces you to draw too many conclusions based on guesses or bad information because it is incomplete and imprecise.

Perimeter teams can be punished – There is nothing wrong with a team that chooses to base their offense around the perimeter. As long as they can shoot well they are going to have some success. In fact, a lot of teams – especially in college basketball – don’t contend offensive rebounds at all. Their strategy is to set up the best shot they can, and if they miss it they are well positioned to get back on defense and take care of business there. Those teams are going to be punished in the calculation of rebound margin. Not only are they not going to get many offensive rebounds, but they will also give up more than their share of defensive rebounds. If you relied on rebound margin if not looking closer at the team then you would assume that they are a poor team. That is absolutely not the case – not necessarily, anyway. If you don’t serve up this stat with a heavy dose of context then you can come to conclusions that just aren’t accurate.

Poor shooting teams can generate more rebound opportunities – You can’t get an offensive rebound on a shot that scores. That means that a team that is a good shooting squad is going to be at a disadvantage to a team that is inaccurate in terms of offensive rebounding opportunities, and therefore potentially in terms of rebounding percentage. Any stat that has the potential to reward poor execution is a dubious one.

Highly affected by pace – When a team plays a high tempo style of play on both ends of the court and they are a decent rebounding team then they are going to get a whole lot of opportunities to rebound. If they are able to set the pace against opponents then they are going to be able to shine against teams that aren’t as comfortable playing fast. If a team likes to play at a much slower tempo them they are going to have less opportunities, and it will be harder for them to open up a wide rebound margin in a game. It would be okay that the faster paced team would have a wider rebound margin than a slower paced team of similar rebounding skill if playing at a slow pace was inferior to playing at a fast pace. It obviously isn’t, though, so rebound margin fails here as well.

Depends on opponent play – The best statistics are ones that can’t be widely affected by the opponents that are played. Rebound margin is highly dependent on opponents. If a team has a big, interior-focused team then they would likely shine in this statistic against a three-guard, perimeter based team. They would be far less impressive in rebound margin if their schedule featured other bigm rebound minded teams. You have to be extremely careful when looking at rebound margin to see who they have played against – especially in college basketball. Any statistic that needs that much extra interpretation to tell its true meaning is not particularly useful.

So, what is better? We’ve looked at some issues with this statistic, and there are many more. People talk about the statistic as being particularly insightful and effective, but it has far too many issues to be trusted. Instead, you need to look at statistics that isolate just one portion of the game, and tell a story based on that alone. For example, if you want to look at rebounding then defensive rebound percentage is far superior to rebound margin. To calculate that you take the number of defensive rebounds a team has and divide it by the sum of the number of defensive rebounds plus the number of offensive rebounds the opponent has. By just focusing on defensive rebounds instead of all rebounds you can get a much more accurate picture of what a team is capable of, and you can more meaningfully interpret numbers that come as a surprise at first glance.