NBA Chemistry Experiments
Home | Teams/Players | Commentary | Premium Services | Contact Us

Game Charting Insights: Contested Rebounds

Rebounds have a wide range in degree of difficulty, and so this exercise is designed to begin to understand the frequency of contested rebounds versus uncontested ones, and to make distinctions between them. We also can look at the impact of shot location on rebounds, shot clock timing, and other facets of hitting the boards

Project scope: we'll be tracking "contestedness levels" as well as shot location and rebound location.

This is admittedly venturing into much more subjective territory, but charters tracked:

A) how many players for each team were in the vicinity of where the ball went for the potential rebound, and
B) of those players how many made an attempt to contest the rebound.

In essence how many guys are in reasonable range to get a chance at it -- eg guys might be posting up on the left side ready to grab the rebound but if the ball bounces out to the right they are not contesting the rebound since they're not close to the ball. On the other hand a player may be in position, blocking out, etc and just failed to jump or had the ball go past them.

Contested Rebound levels and success rates
So our tracking allows us on the simplest level to just measure the competitiveness of any given rebound:

uncontested -- 1 or more players for one team, 0 players in range for the other team
even contest -- the same number of players contested for each team (1 vs 1, 2 vs 2, etc.)
advantaged situation -- one team has more players contending for the rebound (2 vs 1, 3 vs 2, etc.)

Here's how the splits turned out from our sample:

Rebound Situation
Offensive
Rebound
Situation
Frequency
 Uncontested Offensive Rebound 97% 5%
 Advantage Offense (+1 or more players) 50% 6%
 Even Contest 39% 31%
 Advantage Defense (+1 or more players) 32% 28%
 Uncontested Defensive Rebound 1% 30%

Now you may be wondering how can the uncontested rebounds not always go to the team with the only player going for the rebound? and the answer is that you're dealing with the classic dropping the ball out of bounds/stepping out of bounds maneuver.

Aside from that, the points that seem to jump out from this are:

  1. The distributions between the three rebound situations are fairly even...35% are uncontested, 31% are evenly matched in terms of number of players contesting, and 34% where one team has a numbers advantage.
  2. The number of players involved doesn't skew the rebound rates to the degree you might expect -- for instance even with a man advantage in going for the board, the offensive team only had a 50% chance of getting it, and an "even contest" still saw the defense win out 61% of the time. So position matters more than sheer manpower in attacking the glass.
Now does this mean we should chart every rebound and discount all the ones players nab without competition? Not necessarily since there could be reasons why the player got the rebound that are to his credit -- he may have been the only player on the floor who accurately predicted where the rebound would carom to, or the player might be such a formidable glass cleaner that the other team didn't even bother to contest it when they might have with a lesser player.

Still it does seem clear that the range of difficulty when it comes to rebounds is pretty large. A more substantial rebounding rating should then look at the specifics of a rebound and not just count them all as one more tally in the scoresheet.

Another implication may be that not sending anyone to the offensive glass could be a mistake, since even when outnumbered the offensive team grabs the ball at a pretty healthy clip. Of course a true answer to this requires looking at "sending guys to the glass" versus the ensuing possession for the former defensive team in the event your 'sent guys' don't corral the rebound.

Let's move on now to looking at the details of the type of shot versus frequency of being contested.

Rebound Situations by Shot Type


For shot types we often use our trusty "court zones" concept, and for this we ran with the classic fourteen zone scheme.

Basically zones 1 to 5 are three point locations, 6 to 10 are two point zones outside the paint, and zones 11 to 14 are in the paint. For any backcourt events we use a "-1" zone, and once in a while you get a "zone 0" for deep frontcourt events (like a foul crossing halfcourt or something)


We'll break it down here though in the three natural shot type splits --

Three-Pointers -- self explanatory really, zones 1 to 5 on our map
Two Point Jump Shots -- outside of the paint, zones 6 to 10
In the Paint -- zones 11 to 14

...and also show the overall "contested" percentages where each team had at least one player in range to go for the rebound.

THREE-POINTERS
Rebound Situation
Offensive
Rebound
Situation
Frequency
 Uncontested Offensive Rebound 100% 3%
 Advantage Offense (+1 or more players) 56% 15%
 Even Contest 40% 29%
 Advantage Defense (+1 or more players) 31% 26%
 Uncontested Defensive Rebound 1% 27%
 Contested 40% 70%

TWO-POINT JUMP SHOTS
Rebound Situation
Offensive
Rebound
Situation
Frequency
 Uncontested Offensive Rebound 95% 3%
 Advantage Offense (+1 or more players) 49% 11%
 Even Contest 30% 26%
 Advantage Defense (+1 or more players) 29% 26%
 Uncontested Defensive Rebound 1% 35%
 Contested 33% 63%

IN THE PAINT
Rebound Situation
Offensive
Rebound
Situation
Frequency
 Uncontested Offensive Rebound 96% 7%
 Advantage Offense (+1 or more players) 46% 11%
 Even Contest 46% 32%
 Advantage Defense (+1 or more players) 38% 28%
 Uncontested Defensive Rebound 3% 22%
 Contested 43% 71%

So no shock here, but the "in the paint" shots lead to more contested rebounds, and the highest offensive rebound rate, with three-pointers being close, but those pesky two-point jumpers from outside, which we've picked on in other articles, are once again the least effective for the offensive team in terms of producing second chances!

We're only showing the three groups here, but as you might expect, there are differences between a baseline shot and say a similar distance shot from the wings.

Now one supposed quality of great rebounders is knowing where the ball is likely to go if it misses, and we can provide some data on that...

Where missed shots go

We'll again use the fourteen zone scheme for both where a shot was taken from and where it was rebounded, but grouping zones for ease of analysis into:

Long rebounds -- all three point zones (1 to 5) and the straightaway two from the free throw line (zone 8)
Medium rebounds -- the remaining two-point jumper zones (6,7,9,10) and the two high paint zones (11, 12)
Short rebounds -- the two under the basket zones (13, 14)

In addition, we'll track a left/right skew for the medium/short zones, which is:

Left -- zones 6, 7, 11, 13
Right -- zones 9, 10, 12, 14

Shot Zone
Long
Med.
Short
Left
Right
 1 - Left corner 3pt 6% 45% 49% 37% 57%
 2 - Left wing 3pt 7% 54% 39% 42% 51%
 3 - Straightaway 3pt 4% 63% 33% 56% 40%
 4 - Right wing 3pt 12% 56% 32% 50% 38%
 5 - Right corner 3pt 8% 42% 50% 53% 39%
 6 - Left baseline 2pt 4% 32% 65% 49% 48%
 7 - Left wing 2pt 4% 51% 45% 50% 46%
 8 - Straightaway 2pt 4% 41% 55% 43% 53%
 9 - Right wing 2pt 6% 48% 46% 52% 42%
 10 - Right baseline 2pt 2% 41% 57% 37% 61%
 11 - Left side high paint 2% 32% 67% 43% 55%
 12 - Right side high paint 3% 34% 62% 49% 47%
 13 - Left side low paint 4% 22% 74% 59% 37%
 14 - Right side low paint 2% 21% 78% 44% 54%

So yes, the three point shots produce more medium and long rebounds, especially from the wings and straightaway. More importantly there's strong evidence that long range missed shots tend to go beyond the basket, so a left side three point shot ends up in a right side zone and vice versa.

Now our sample sizes aren't great which probably explains some of the inconsistency in certain mirror image zones. Furthermore a bank shot probably has a markedly different "miss spray" then a straight at the hoop attempt. This is something that may be worth addressing over a full season of shots, at which point you can also start digging into the tendencies of specific players.

The "Ultimate Rebounding" rating?
These league wide averages and trends are interesting, but let's face it, everyone wants to get down to the level of which players are monsters at pulling down the key rebound in traffic. Given we've established here and in previous articles that there is a wide range in rebounding scenarios, it would make some sense to separate a player's rebounding opportunities and "credits" by the circumstances.

An interesting parallel may be what is being done in baseball for fielding ratings by our friends at PROTRADE and other places -- with data on the path of every batted ball, strength and height, they can then draw up averages as to how often a ball hit in a certain way has "typically" been converted into an out, a single, a double, etc and grade players versus their performance on "similar" fielding opportunties compared to the rest of the league.

This makes sense in the NBA to some degree since rebounding involves not just leaping and timing, but also positioning, blocking out, judging the flight of the ball, etc which are all captured reasonably well in the above, but as we also mentioned earlier it may not fully reflect everything that's happening. Still, it would be a worthwhile starting point.

You want to see these kinds of individual player rebounding stats? So do we...stay tuned!

Also see:
Game Charting: Floor Locations (03/27)
Game Charting: The Value of a Good Pass (03/20)

Next up: touches and dribbles!


Game Charters for the "Contested Rebound" Project:
Adrian Lawhorn, Andy Davis, B.J. Colby, Bo Schwarz, Brad Burnett, Brendan Gildea, Brett Steele, Cameron Tana, Charles Floyd, Charles Floyd, Chip Crain, Chris Goudey, Craig Ward, Curtis Chody, Dana Henderson, Daniel Kelly, David Mintz, Demetrius Burrows, Dmitri Salcedo, Eric Patten, Eric Wallace, Erich Hothem, Frank Mantesta, Gabe Farkas, Greg Humphreys, Husamettin Erciyes, Jared Wade, Jeff Miller, Jeremy Killey, Jerry Hardin, Joe Malloy, John Magee, Justin Kimbrell, Lorenzo Pascucci, Mark Reyes, Matt Kolsky, Matthew Powell, Miguel Cuaron, Mike Piers, Mike Raak, Mike Wolf, Patrick Sheehy, Paul Bishop, Phil Edwards, Philip Wong, Raj Kannan, Rich Schmidt, Rob Ireland, Rob Stewart, Sandy Weil, Schuyler Sheaffer, Scott Castiglia, Sean Franklin, Susan Nelson, Tom Lore, Wendy Gathers, Zach Ellin

Thanks as always to the noble efforts of the charters -- they are the ones who will be pushing NBA statistical analysis to new heights!

Interested in doing some game charting? Send a message to: charting@82games.com


Rate this Feature
Poor   Fair   Good   Excellent

Enter your comments in the box

Email (optional)

We want your feedback! Tell us your thoughts

Copyright © 2006 by 82games.com, All Rights Reserved