Tuesday, December 9, 2008

Converting off of Defensive Rebounds

This is completely random, but I was curious so I played around a little bit calculating team offensive success immediately following a defensive rebound. Is a team aided significantly by rebounders who lead a fast break, like Jason Kidd or Lamar Odom? Are there any bigs who are underrated due to their effective outlet passing?

I took a look at the play-by-play data from 2005-2008, and every single defensive rebound recorded during those three seasons. I then removed those that were immediately followed by a timeout or by the end of a quarter, and ended up with a record of defensive rebounds that were followed by either a shot (made or missed), a drawn foul, or a turnover. For this study, we'll consider "made shot" or "drawn foul" to be successes, and we'll take a look at the success rates after defensive rebounds for different players. I'll look team-by-team so as not to have to worry about the quality of a team's offense skewing the success rates. In addition to success, I'll take a look at the average time between the defensive rebound and the next play, to see whose rebounds lead to more fastbreaks and early offense.

For the purposes of this project, I've included only players with 250 or more defensive rebounds in a given season.

Here's some notes from teams that have interesting stories (I'll skip the teams where there wasn't much of interest to take away):

Josh Smith's defensive rebounds consistently lead to much faster action than thos of anyone on his team. For instance, in 2006-2007, Smith's rebounds turned into an offensive play within an average of 9.6 seconds, whereas the rebounds of his teammate Marvin Williams turned aruond in 11.7 seconds. Last year, it was 9.2 seconds for Smith and 10.8 for Williams. The change over the years for Smith, though, has been his success rate. It has climbed each year, from 47% in 05-06 (which was second-lowest on the team) to 53% in 07-08 (highest on the team, and much higher than second place Al Horford, who had a success rate of 49%). So what happened? Well, the big improvements came in FG% after rebounds, which went from 42.3% in 05-06, to 50% two seasons ago, all the way up to 53.9% last season. This ascent mirrors Smith's own FG% on inside shots over the years, from 58.5% in 05-06, to 61% in 06-07, to 61.8% last year. Maybe he just got better at finishing what he started?

Last year Gerald Wallace's rebounds led to much more effective offense than those of any of his teammates, with his success rate at 55% compared to his team's average of 48%. The effectiveness wasn't a result of his necessarily starting fast-breaks, though, as his rebounds didn't cause any faster action than those of his teammates. So what's the story? I have no idea -- if you have any, leave them in the comments.

Andres Nocioni's rebounds result in turnovers more often than those of any of his teammates.

Obviously, when Lebron James gets a rebound, good things happen, and they happen quickly. Lebron has high success rates every year, and his rebounds turn into plays close to two seconds faster, on average, than those of his teammates.

We know they've been a fast-paced team for a few years. What's intriguing here is that center Marcus Camby doesn't slow down the offense at all -- his rebounds turn into offense just as quickly as those of Carmelo Anthony. On a team that likes to get out and run, I could see Marcus Camby being hugely valuable since he is a very good defensive rebounder who is also good at starting the fast break. Without Camby this year, Denver is playing at a much slower pace than they did last year. Meanwhile, the Clippers are struggling, and, by the way, they are playing at close to the same pace as they did last year, despite adding prolific fast-breakers Camby and Baron Davis (discussed below). Maybe they'd be better using their talents if they got out and ran more?

Golden State
We've all seen Baron Davis and Monta Ellis single-handedly start fast-breaks, so it shouldn't be a surprise to see that last season their defensive rebounds turned into offense a lot more quickly than anyone those of anyone else on the team (an average of 7.2 seconds for Davis, and 7.7 seconds for Ellis -- compared to 9.2, 8.2, and 9.7 for Andris Biedrins, Al Harrington, and Stephen Jackson). I did find it odd to see Stephen Jackson's rebounds being not only the slowest on the team last year, but also the least successful, with just a 47% success rate. He is helped a little when you look at points produced, as he likes to shoot those pull up three pointers in transition. So even though his success rate is lower, his rebounds produce points at about the same rate as those of his teammates.

Tracy McGrady is easily the fastest Rocket around, turning his boards into offense in an average of 10.4-10.7 seconds each year (Rafer Alston is also pretty quick, but doesn't grab enough rebounds to qualify). Unfortunately for McGrady, his success rate has dropped each year, from 52% (which led his team and was much higher than the team average of 45%) to 47% to 46% last year.

LA Lakers
I expected to see Lamar Odom's name more prominently here, but in fact Kobe Bryant has been the Laker who converts his rebounds into offense most quickly and most effeciently for the last three years. Of interest to this year's Lakers' squad is the case of Pau Gasol in Memphis -- Gasol is one of the few bigs in the league who led his team in creating early offense off of defensive rebounds, turning his rebounds into offense faster than even the gaurds on his team. This year, it appears that he has continued that tradition, constantly looking for the outlet once he's secured a rebound -- but I'll wait to see the numbers before drawing firm conclusions.

James Posey
Posey gets his own entry here because he showed similar results while playing for different teams. For two years in Miami and one year in Boston, Posey's defensive rebounds were the slowest on the team to turn into offense, and had the lowest success rate.

New Jersey
Jason Kidd definitely likes to get out and run, creating offense off of his rebounds within 8 to 8.5 seconds each year (around 1 to 2 seconds faster than his teammates). But it's Vince Carter who has the highest success rate over that period. Kidd, a great passer in the open-court, might have just needed some more teammates who could finish on the break (Carter can finish and draw fouls on his own in the open court, so was less held back by the lack of quality teammates).

Like Kidd, Chris Paul gets out off of rebounds much faster than any of his teammates, although Paul has a much higher success rate than his teammates, also. Peja Stojakovic, meanwhile, turns his boards into offense a whopping 2.5 seconds more slowly than Paul.

Hedo Turkoglu wins the speed and efficiency contests here, and it makes sense from watching games -- a lot of times after he grabs a rebound he turns into the de facto point guard, bringing the ball up the court and initiating the offense. The numbers show that this helps the Magic, which would partly explain why the Magic offense was so much more efficient when he was on the court last year.

Good things happen when Brandon Roy gets rebounds!

San Antonio
Like Gasol, Tim Duncan is another of the rare bigs who generates fast and efficient offense off of his defensive rebounds. So, add that to the list of good things you can say about Duncan's game.

Well, that was fun! Hopefully this will be the beginning of being able to look at offense and defense together. Most of the statistics available deliniate a very clear boundary between the analysis of offensive efficiency and defensive efficiency. Sometimes on television broadcasts we get to see statistics like "points off of turnovers," but I think that in some cases looking at points off of defensive rebounds might be valuable. The above exercise shed some additional light, I think, on a couple of areas:

- Point guards who can rebound are additionally valuable since they can so effectively generate offense off of those rebounds -- for instance, Jason Kidd, Chris Paul, and Baron Davis above (similarly effective but not mentioned: Gilbert Arenas and Steve Nash, who tend to have lower rebound totals, so can't make this feature as much a part of their game).
- Marcus Camby, Tim Duncan, and Pau Gasol are special in their ability to quickly turn rebounds into offense through ball security and effective outlet passing. We should be able to take a look after this year to see if Kevin Love, he of the world-famous outlet pass, measures up.
- Maybe there is some value after all to athletic 3/4 tweener/hybrid types. Josh Smith, Gerald Wallace, and Hedo Turkoglu all proved to be very effective at both rebounding the ball on the defensive end and immediately turning those rebounds around into points on the offensive end.


  1. sorry if this is a dumb question, but are you counting "made shot" and "drawn foul" for the fastbreaking team as a whole, or only by the original defensive rebounder? you mention both looking for the outlet and the one-player rebound-to-offense scenarios.

    i'd be interested to see how the data stacked up via position, as well as by individual player. theoretically, the fastest way to transition from a defensive rebound into offense is passing, rather than dribbling. this would lead one to think that the big men looking for a streaking guard could turn boards into points the fastest? although it can probably be assumed that guards are more likely to field a rebound further away from the opponent's basket, giving them a headstart down the court.

  2. I'm counting them for the team as a whole, and for exactly the reasons you're mentioning. I was hoping to find the best outlet passers this way, but as you've pointed out, more often than not (other than the exceptions mentioned in the post), it's guards whose rebounds result in their teams' having the quickest offense -- probably largely due to where they are on the court when they get rebounds, as well as their ability to run up the court without having to stop and look for a guard (or an outlet recipient downcourt) once they've secured a rebound.

    I think presenting the data by position is a good idea, I'll try and put that up soon.

    In some cases, like that of Josh Smith, I mentioned the one player rebound-to-offense situations because subjectively, I've noticed that that is a big part of his game, and I'm guessing that that is what is driving the low average time that his team takes to convert his defensive rebounds into offense. But the numbers presented in all cases are for the team's success and turnaround time, after an individual's rebound.

  3. McFruity, I wish we could watch bball together again! those were the dayz.