Battling Under the Boards: Measuring Attempted Rebounds

facebooktwitterreddit

Jan 10, 2015; Philadelphia, PA, USA; Indiana Pacers center Roy Hibbert (55) battles for a loose ball with Philadelphia 76ers guard Michael Carter-Williams (1) during the third quarter at the Wells Fargo Center. The Sixers won the game 93-92. Mandatory Credit: John Geliebter-USA TODAY Sports

Following up from our recent quick look at some of the SportVU-based rebounding data, here are some further insights into rebounding, this time focusing more on the individual level. For starters, here are the percentage of available rebounds contested by qualifying players[1. Minimum 400 minutes with a given team, though this amount will be raised as the season goes along so that the chart captures the top 250-300 players in terms of minutes]. Note that this uses the SportVU definition of “contesting” a rebound, which is being within 3.5 feet, roughly arms’ length, of the ball when the rebound is secured. This almost certainly undercounts what we’d describe as “going to the boards” watching the game, but comparing this number to the number of available rebounds will serve as some degree of proxy for comparison among players.

There’s nothing terribly surprising at the top end of the spectrum – Tyson Chandler, Reggie Evans, Omer Asik, Andre Drummond and Kevin Love are certainly names one would think of in terms of dedicated defensive rebounders, and while Kevin Garnett might be a surprising inclusion, he has remained extremely active on the defensive glass even as age takes much of the rest of his game. Similarly, Evens, Tyler Hansbrough, Tristan Thompson and Zach Randolph are names to be expected near the top of the offensive rebound attempt lists.

Since data is public for both this year and last, it’s possible to compare players who have improved or declined the most in terms of rebounding attempts. This isn’t necessarily a “better” or worse thing. For example, Josh Smith was more likely to have chances of rebounds during his aborted sting in Detroit this season than he was last season because he was playing nearer the basket, especially defensively. LeBron James shows a big decline in rebound chances, but that was to be expected moving to a team where he went from being perhaps the best rebounder in the rotation to third or fourth on this Cavs team. Among big minute players, here are the players who have increased their defensive rebound chances the most:

Of particular note is Hibbert. Savagely maligned at times for his lack of individual rebounding numbers last year, Indiana remains an excellent defensive rebounding club[2. Second to Charlotte in DREB% on missed FGs] despite losing both of their excellent wing rebounders from last season in Paul George and Stephenson, while David West has also missed a ton of time this season. The most natural explanation is now that the team around him demands Hibbert pursue rebounds himself rather than simply clear the way for his teammates to secure them, he his doing so.

Another bit of fine slicing is thanks to Justin Willard from The Analytics Game, who suggested breaking out missed FTs from missed field goals. This makes a great deal of sense as shots from the floor are more than twice as likely to have been offensive rebounded than free throw misses.[Thru Friday’s games, 26.1% on missed FGs vs. 11.5% on missed FTs.] While this makes for a better understanding of rebounding in the flow of play, it also revealed what has to be an oddity, in terms of Oklahoma City’s ability to offensive rebound it’s own free throw misses, even compared to the other top teams on this measure:

Serge Ibaka individually has more offensive rebounds off of free throws than five teams (Atlanta, Charlotte, Cleveland, Golden State and Portland) as the Thunder account for nearly 10% of the total offensive rebounds off of missed FTs in the NBA this season.[3. Unfortunately, it’s not yet practicable to break these numbers out for individual players, as the public data does not differentiate between a rebound chance from a missed FG or FT, so while we can identify rebounds a player actually gets by type, it’s not possible to determine what kinds of rebounds they were contesting but losing out on. Perhaps with the release of the full visualization data, someone will come up with a way to measure this, but it’s not exactly a top priority for study.]

Finally, on the team level, I’m developing a working theory as to some causes of Minnesota’s poor transition defense. One possibility I’m examining is the frequency with which their guards attack the glass. This theory is, well, more than plausible:

A next step for investigation is examining these metrics with some sort of positional average which could indicate why, for example, Chris Bosh’s rebounding has gone up less than I expected or whether Brook Lopez is slacking on the board or just not good at winning rebounding battles.