APBRmetrics

Chronz1 · Joined: 22 May 2006 Posts: 201

Offensive rebound or defensive rebound?

I always used to say offensive rebounds are more valuable because they are harder to achieve but this guy says defensive rebound is just as important because they prevent offensive rebounds.

Another question why is a offensive rebound with .7 on the PER scale wheras a defensive rebound is worth only .3

If they are equally important shouldnt they both be worth .5

edijorj · Joined: 21 Apr 2006 Posts: 30

For a team an orb and a drb are equally important (for the reason you give). However for evaluating an individual, an orb is much more important. Why? Because the regressions say so. At http://www.uncg.edu/eco/rosenbaum/NBA/winval2.htm Dan Rosenbaum finds that for the average player the value of a drb is almost equal to zero. I think more recently he has found a higher value using a slightly different regression model, but the value is still much less than the value of an orb.

An interpretation of these results is that players other than the rebounder (for example the one guarding the player who missed) deserve a very large share of the credit for a drb, so even though a drb is as valuable for the team as an orb, the player who catches a drb deserves much less credit. Another interpretation (not incompatible with the first) is that the number drbs mostly reflect the the game plan (for example the game plan was for Rodman to be the rebounder), but that orbs, because they are so much harder to catch, actually reflect rebounding ability.

I think 0.7 for an orb and 0.3 for a drb is about right. I usually use 0.6 and 0.27.

mrgold · Joined: 29 Apr 2006 Posts: 5

Orebounds are greater than droubounds cuz their harder to get. If a player who grabs an offensive rebound doesnt grab it than the other team will almost certainly gain possesion. However, if a player is not in position to get a defensive rebound, someone else on his team will most likely get it. The value of an individual getting a rebound is equal to the liklihood that their action caused their team to retain possesion as opposed to lose it. Since most rebounds are defensive an offensive rebound is of much more valuable.

cherokee_ACB · Joined: 22 Mar 2006 Posts: 157

This 82games.com article
http://www.82games.com/rebounds.htm
tells that most offensive rebounds are contested, while nearly half of deffensive rebounds are not. This is consistent with the roughly two-to-one ratio between the value of an offensive rebound relative to a deffensive rebound that PER and other rating systems use. I'm not saying that uncontested rebounds have no value (positioning is key), but that a good amount of deffensive rebounds are grabbed essentially because the ball goes to the player hands, and not the other way around. That's why you won't see any player with a rebound rate lower than 3%.

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

Why is an offensive rebound more valuable because it's harder to get? What if what's most valuable for the offense is to get gack on defense and prevent a fastbreak basket instead of expending any effort to get an offensive rebound?

Ed Küpfer · Joined: 30 Dec 2004 Posts: 787 Location: Toronto

Here'a a different way of saying the same thing:

The graph shows the relationship between a player's REB% and his team's. The red points are defensive rebounds, the blue points offensive. The panels divide the data among 9 player positions.

Look at the regression lines. Notice the slope for OR% is higher than that of DR%? That tell us that for each unit increase in a player's OR% he raises his team's OR% by greater amount that an increase in player DR% raises his team's DR% -- that is, a player's effect on his team's rebounding is greater on ORs than DRs. My number's suggest that each OR is twice as valuable as a DR for frontcourt players, more for the other players. (In fact, I see a slope not significantly different from zero on guard's DR%.)
_________________
ed

alphamale · Joined: 03 May 2006 Posts: 67

I don't have any stats to back this up but I've seen Dennis Rodman grab numerous rebounds both offensive and defensive, that weren't contested. He managed to anticipate where the rebound was going and got to that position before anyone else. Of course, there are the usual rebounds that just so happen to come to you and you're in the right place at the right time.

It's kind of like how making uncontested shots is less impressive than making contested shots. But maybe I player just managed to outsmart the defense to be in that position in the first place.

Analyze This · Joined: 17 May 2005 Posts: 364

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

edijorj · Joined: 21 Apr 2006 Posts: 30

deepak · Joined: 26 Apr 2006 Posts: 665

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

Analyze This · Joined: 17 May 2005 Posts: 364

exile · Joined: 05 Feb 2006 Posts: 6 Location: Melbourne, Australia

Speaking of offensive and defensive rebounds, has there been any statistical anaylsis done (82games or otherwise) that tell us where the rebounds are going to? Preferably after a certain shot? (i.e. jumpers from say, 16 feet, often are longer rebounds than shots from 5 feet)
_________________
JBL Fantasy Basketball League: http://www.totaljbl.com

Mike G · Posted: Thu Jun 01, 2006 5:54 am Post subject:

If on 2 consecutive opponent possessions, 2 teammates are in position to get the d-rebound, they will still only total 2 (at most) between them. It's exactly because d-reb are more often 'shared' that they aren't accrued as often as they might be (by a given individual).

This is analagous to the illusion that assists to an inside player are more 'valuable' than those to a perimeter shooter. High-% shots are more likely to be converted -- retroactively, into assists -- and thus the passer in fact gets more credit when he hits the likelier scorers.

Rarely do players 'team up' for an o-reb. There is no sharing, nor deferring to a 'designated rebounder' (Rodman). D-reb totals represent only some fraction of the total reb a player could have if he were greedy.
_________________
`
36% of all statistics are wrong