Actually, offensive rebounds probably don’t cost much

Henry Abbot, senior writer at ESPN and founder of the TrueHoop blog and network (of which Roundball Mining Company is proud to be a part) recently published an interesting article entitled “Offensive rebounds come with a cost.” In it, in response to John Hollinger’s observation that last season’s Boston Celtics were the worst offensive rebounding team in league history. he poses an intriguing question:

Let’s say Doc Rivers told his bigs to crash the offensive boards more. They would certainly get more easy putbacks in particular and more field goal attempts for their team generally. More shots per possession. Better offensive efficiency overall.

But at what cost? How much would the team be giving up on all those possessions where the bigs didn’t get the rebound. That’s the most common outcome, right? In those cases, the bigs would simply not be back which hurts the team, for sure. Does it hurt the team more than now-and-again offensive rebounds are worth?

He went on to point out that the three teams last season which had the fewest shots per possession were Boston, Miami and Oklahoma City, summarizing that “a good chunk of the very best teams in the league, including both Finals competitors, simply don’t seem to place much of a priority on the offensive glass.”

On its face, there is a logic to this that seems completely reasonable. If you’re not back on defense, how can you defend? In fact, I did not question it at all at first. My thoughts immediately went to the implications of this concept for the Denver Nuggets. They place a premium on scoring at the rim, which most definitely includes second chance points scored off offensive rebounds.They also are seeking to improve their defense, which slipped last season.

Mulling his over, I first wondered aloud on twitter whether the (apparent) defensive harm resulting from prioritizing offensive boards might be minimized in Denver’s case due to the speed and athleticism of its frontcourt – or in other words, whether it would be offset by their ability to get back more quickly than many other teams’ bigs, even after going for offensive boards

I quickly got a reply from twitter user Ty Hicks (@Ty_Hicks), who wasn’t buying any of it. He pointed out that the Thunder and Heat were “hardly the worst in the league” on the offensive glass – 10th and 18th in offensive rebound rate, respectively – and that an alternative explanation for their low shots per possession was the fact that “they miss less than other teams.

These arguments raised some valid questions. Would a closer look reveal that there really is any meaningful correlation between a sustained effort to crash the offensive boards and poor defense?

Although the concept does make sense on paper and is appealing in its symmetry, after digging a little deeper into the data I found that I must humbly and respectfully disagree with Henry Abbot and accept that the answer to this question is most likely, “No.”

Using statistics from Hoopdata.com, I put together the chart below. It graphs the relationship between defensive efficiency and offensive rebound rate for every team in the NBA from 2006-07 to 2011-12.

As you can see, the results are quite evenly distributed across the board. There do not seem to be any significantly concentrated clusters of results in the areas where we might expect to see them (very good and very bad offensive rebounding teams with inversely bad/good defensive efficiency) if this hypothesis were true. Additionally, while the linear trend line does show a nominally upward (ie. worse) inclination in defensive efficiency as offensive rebound rate improves, it is very slight. I’m not a statistician, but I would presume that this effect is so minimal as to be insignificant, and within error bars.

I certainly do not claim that this quick analysis is thorough or sufficient enough to completely falsify the hypothesis that a team’s defense will take a hit if they focus too much on the offensive glass. Perhaps a more detailed look involving transition defense specifically, or a larger data set from a longer span of seasons would turn up different, more positive results.

But provisionally, based on this first survey of the correlation between the two factors, it does not appear that going after those offensive boards has a measurably negative impact on defense. Surely not enough of one, at least, to forgo the benefits that offensive rebounding reaps.

So Faried, McGee, and the rest of you Nuggets hungry for those offensive boards, have no fear. Crash that glass. But after that, please, don’t forget to run back.

Follow me on Twitter
The following two tabs change content below.
Joel is a long time Denver Nuggets (and Broncos) fan from Colorado who's been living in Japan since the mid-90s, and blogging about the Nuggets since 2008. You can contact and follow him on Twitter: @denbutsu.
  • Wally

    Great article, it was a nice read. I agree 100% with your premise above, especially for our team with Kenneth “The Manimal” Faried being a offensive rebounding stalwart.

  • Andrew K

    Very interesting analysis. Especially in the Nuggets’ case, I would not change their strategy of crashing the offensive boards.

  • Sam

    Interesting… is there a way you can do the same analysis while eliminating some of the other variables that might influence defensive efficiency?

    • https://twitter.com/denbutsu denbutsu

      At MySynergy.com they track offense and defense by type of play. Doing the same kind of comparison as I did above, but instead of using defensive efficiency, using defense in transition (probably points per possession allowed in transition) would more accurately zoom in on the specific kind of defensive breakdowns that would result from not getting back quickly. For example, a good offensive rebounding team might give up too many transition points but also be really great at halfcourt defense. Their overall defensive efficiency would end up looking kind of average, and mask the harm done by going after offensive boards.

      Unfortunately, I don’t have Synergy, so I’ll have to leave that to others if they want to pursue it.

  • SmokinNugs

    My exact thoughts when I saw Abbot was basing it off of fewest shots per possession and not actual offensive rebounding!

  • Aaron

    True, True. . Offensive execution is why Miami and OKC were in the Finals, and the reason why a less-talented group such as the aging Celtics get as far as they do.

    Karl categorically just wants to get up more shots, so we’re never going to be high there. But Offensive boards don’t have much to do it. OKC doesn’t need offensive boards because Westbrook and Durant are running back after MADE baskets.

  • DH

    Let’s say that offensive rebounds do have a small inverse effect on defensive efficiency. One thing that’s being ignored is how much offensive rebounds increase offensive efficiency. It seems to me that offensive rebounds often lead to high-percentage shots – like putbacks and open 3’s. I would think that this would more than offset the negative impact on defense. Of course, a lot depends on how many (and which) players you send to the glass. You have to be judicious about it, of course.

  • Evan Woodruff

    Who is “Joel”?

    • https://twitter.com/denbutsu Joel

      “Joel” is “denbutsu”. And both of them are me. I decided to drop the handle, and I’m sorry for causing some confusion. It just turned out to be best for me to use my real name, and I wanted to go ahead and make that change on the sooner side.

  • Misty

    There’s a great book out there called “How to Lie with Statistics.” I’ve taught statistics at the college level.* Most of these “statistical” analyses that you see from sportswriters are filled with errors and assumptions. It is extraordinarily difficult to show cause and effect when there are so many uncontrolled variables. ESPN is the best (worst?) at drawing conclusions with no real valid statistical support.

    The best example of this? As mentioned in the comments section – OKC & Miami have the fewest shots per possession because they make a higher percentage of shots – yet someone tried to draw a correlation between shots per possession and offensive rebounding.

    In my mind that is, well, offensive.

    *I actually did my senior thesis on analyzing NBA data – looking for verifiable indicators of which team (given their stats) would win any given match-up. The only two factors that had any statistical validity were home court advantage and win-loss record.