Celtics Stat of the Day, 12/12/09

Each day, I'm going to be posting a stat I hope will be of interest and may generate some conversation. Today's stat deals with the only significant statistical category in which the Celtics are trailing opponents this season: offensive rebounding. (That's right, they're even leading in terms of defensive boards, 29.6 to 27.9 per game.)

The Celts are grabbing 8.4 offensive rebounds a game versus 10.6 for opponents. Not a huge difference; but one would like to think they could do at least as well as last season (10.6 Celtics to 10.2 opponents), when they were missing KG for a considerable period.

The big difference is that they miss Leon Powe: Last season, the two Celtics leaders in offensive boards per game were Perk (2.7) and Leon (2.1). This season, the two leaders are Perk (down to 1.7) and KG (1.5).

Now, this is hardly reason for panic. Part of the reason the Celtics get fewer offensive rebounds than opponents is that they don't miss as many shots (40.2 per game versus 44.9 for opponents). Nonetheless, consider the relative efficiency of the Celtics versus opponents in grabbing the available offensive rebounds off their respective boards:

In 2007-08, the Celtics grabbed 26.6% of the available rebounds off their own offensive boards
In 2007-08, the enemy grabbed 25.6% of the available rebounds off their own offensive boards
Advantage 1.0% Celtics

In 2008-09, the Celtics grabbed 27.9% of the available rebounds off their own offensive boards
In 2008-09, the enemy grabbed 24.4% of the available rebounds off their own offensive boards
Advantage 3.5% Celtics

In 2009-10, the Celtics gather 23.0% of the available rebounds off their own offensive boards
In 2009-10, the enemy gathers 26.4% of the available rebounds off their own offensive boards
Advantage 3.4% enemy

In the past five games, KG, while playing very well in most respects, has averaged 0.8 offensive rebounds per game.
In the past five games, the entire bench (other than Tony) has averaged 1.8 offensive rebounds per game.

I notice that Tony at got two offensive rebounds in his 17 minutes last night. Notwithstanding failings he may have, Tony does tend to be active on the offensive boards. One of several advantages of his hopefully playing only the forward spot in relief of Paul is that he'll be closer to those offensive boards. Lord knows the Celts could use some reinforcement there. Maybe Tony's new mantra should be "Slash and crash." Could anything be more straightforward?

And what's up with the dropoff in Perk's offensive rebounding? For one thing, he's too often the only Celtics under the offensive boards. For another thing, I believe he's developing a tendency to retreat on defense a beat before he should...leaving NO Celtics (except possibly for Rondo) competing for the offensive boards.

Maybe the return of Davis will help, although Sheldon Williams actually grabs offensive rebounds this season at a slightly higher rate per minute than Glen did last season.

Finally, gaining on opponents in this category is not a matter only of improving the Celtics' own offensive rebounding. They need to stop foes from gang rebounding on the other end. The Celtics' big men MUST body out opponents, and the Celtics' guards cannot release too quickly in transition because they're missing a lot of seven foot defensive
rebounding opportunities that opponents are gobbling up and converting.

I hope people will weigh in on the offensive rebounding, especially with other suggestions as to how to erase (and hopefully reverse) the Celtics' deficit.

Sam

Sam

I thought that Boston would be a dominant rebounding team where they would give up very few offensive rebounds It has not happened and as you have pointed out there are several combination of things that may impact this stat.

From what I can see the Celtics give up offensive rebounds because they are simply out of position. They have very sophisticated defensive rotation schemes and they are often matched up with different players than they are assigned to guard on defense. I just think they have a time gap between rotating on defense and then putting a body on the closet man regardless of who is assigned to cover him. It is really the only time when you do not have a man other than the one closest to you that wants to get an offensive rebound. It is one of those things that they just need to work on.

As far as them getting offensive rebounds, that is really an energy thing. That's why guys like Powe and Tony Allen are good at that and why Rondo will sneak in there to grab offensive rebounds and also why we may see better productivity when Baby comes back. Another reason but to a lesser degree may be the fact that Boston does not really focus on a low post offense although they score a bunch of points in the paint. A lot of times there just isn't anyone there to grab a rebound.

I think the numbers will even out a bit with offensive rebounds given up when the Celtics do a better job on the defensive glass and when we have another rebounder like baby in the game.

In general the Celtics will become a better team when they rebound the ball better at both ends.

dboss

I've got two angles to take on this subject, so rather than make one huuuuge post, I'm going to break it up into two.

I used basketball-reference.com to see how teams ranked in offensive rebounds per game, and I was surprised to see the Celtics last.

http://www.basketball-reference.com/play-index/tiny.cgi?id=z0tiv

I normally wouldn't put too much emphasis on where a team places in this particular ranking. I consider the stats Sam referenced (percentage of available rebounds) much better considering that a lot of factors can affect how many rebounds are actually available (tempo and shooting percentage, for example). But I can't deny I was shocked to find to find the Celtics dead last.

So I took it further. I imported the data into Excel and added a column for offensive rebounding percentage. Here's the formula I used.

offensive rebounds / (field goal attempts - field goals made)

Field goal attempts minus field goals made equals missed field goals, or in other words, the available rebounds. Sam, is that right? Using that information and that formula, I came up with Boston getting 22.0% of the available offensive rebounds, which differs from what you show (23.0%). Anyway, Here's what I noticed about the ranking:

-- Boston is 29th of 30 teams. Only the Warriors are lower.

-- The other elite teams tend to be down the list. Cleveland is 26th, Orlando is 23rd, Dallas is 21st, and Denver is 19th. Only Atlanta (4th) and the Lakers (10th) are toward the top.

-- The top ten includes only four winning teams (Atlanta, Houston, Portland, and the Lakers).

I'm not sure what all that means, other than to say offensive rebounding percentage does not correlate to winning percentage as a general rule. My guess is that it can be important based on a team's philosophy. For example, a big factor will be which a team thinks is more important -- crashing the offensive boards or getting back on defense.

I found the information puzzling, to say the least. I think rebounding is critical to a team's success, so I'm at a loss to explain what's going on. I thought it would be a more straightforward deal -- teams that win tending to rebound well -- but it seems more like analyzing a complex system with many variables, like the weather.

Anyway, I have some thoughts about offensive rebounding as it relates to an individual player, and I'll write those in a separate post.

Outside

A few thoughts about individual player performance when it comes to offensive rebounding. I can think of five primary factors.

1. The player's position on the court as a result of their team's offensive scheme. If Rondo drives and Ray's job is to be in the corner for an outlet, Ray can't be expected to get many rebounds if Rondo puts up a shot.

2. The player's assigned responsibilities once a shot is taken. If your job is to get back on D, you're not going to get offensive rebounds. If your job is to box out the other team's best rebounder, your job is to take him out of the play and let your teammates get the rebound.

3. The type of shots a team takes. If your team takes a lot of threes, rebounds can go over a wider area of the court, and you'll see guards on those teams get more rebounds. If your team shoots more in the paint, you'll see big men get more rebounds.

4. How well the player recognizes rebounding opportunities (applies to both offensive and defensive). Russell is a great example of this. He got a lot of rebounds by going to the spot where a rebound was most likely to go based on the type of shot taken and even the shooter's tendencies.

5. How much effort the player puts into pursuing rebounds (applies to both offensive and defensive). A good example is Orlando. In the game last night against Phoenix, Howard was relentless in the second half going after rebounds. Near the end of the game, Howard was guarding someone on the perimeter instead of Stoudemire, and Howard's player took a three from the corner, so he wasn't under the basket. As Jeff Van Gundy pointed out, everyone else on the Magic sort of stood there because they are so used to Howard getting all the rebounds, so Stoudemire came in unimpeded, got the rebound, and scored the winning basket. Howard works for his rebounds; the other Magic players get the rebounds that come to them. Dennis Rodman is another example of a player who was willing to expend the effort to rebound.

Assuming that a player isn't out of position to get rebounds based on factors 1 and 2 above, it really gets down to being smart enough and hungry enough to go get the rebound. It's what I've called "the Schooner effect." My friend has a standard poodle named Schooner, and that dog lives to chase the ball. When you go into the house, Schooner brings you the ball. If you're out in the yard, Schooner brings you the ball. If you relent and actually play with Schooner, there is nothing on this Earth that will deter Schooner from getting that ball. A truly great rebounder (offensive or otherwise) needs to employ the Schooner effect.

One last note is that there are some players who get more offensive rebounds because they're lousy shooters and miss lots of shots from close range. Last night, Kevin Love of Minnesota had eight offensive rebounds out of 19 total, and he made only three out of 14 shots. Yes, he expends the effort to go get the ball, but if you miss from three feet, not surprisingly, you're in great position to rebound your own brick. One reason Perkins' offensive rebounding is down this year is that he's shooting better (he leads the league at 65%).

Outside

Outside, you've given us a lot of food for thought. To bolster your comment about offensive rebounding effectiveness depending partly on the kinds of shots taken, during this season (before tonight's game):

31.5% of the Celtics' MISSED field goal attempts were three-pointers.

26.2% of opponents' MISSED field goal attempts were three-pointers.

Thus the chances of a long rebound on Celtics' misses is higher than the chances of a long rebound on opponents' misses. Typically Perk is the only Celtics big man under the boards, and he's directly under the basket. so many offensive rebound opportunities go over his flailing one hand.

By the way, my count of available offensive rebounds is a little different from yours. I simply add together Celtics offensive rebounds and opponents' defensive rebounds.

Sam

As always, statistics and confounders

Good stuff Outside.

Sam, I'm not sure why the stats aren't the same. They should be. Maybe one of them includes team rebounds and the other doesn't. Team rebounds are those awarded to a team but not an individual player. For example, when a defensive player fumbles the ball out of bounds, that is a team rebound for the offensive team.

Nice thread Sam. Interesting perpectives on offensive rebounding. I know that boxing out can't be as critical a factor at the offensive end as the defensive end but is it a factor at all? And, if so, is it possible that Glen, being a wide body, factors in somehow because he's good at getting between other big bodies and the offensive rebound, even if he doesn't get the rebound himself?

Outside,

I believe team rebounds represent the difference. I don't consider them "available" to either team. That would explain why you have the lower percentage because the divisor for your percentaging is larger than mine. Not enough to be of any significance though.

How tall is Schooner? Just asking? Oh, and does he like the color green?

Sam

LACelt,

I think of offensive rebounding as placing a premium on agility/hops/height/individuality and defensive rebounding as placing a premium on boxing out/positioning/toughness/teamwork. I know Glen grabs slightly fewer offensive rebounds and many fewer defensive rebounds than Sheldon (see tomorrow's Stat of the Day), but I'd guess his boxing out ability would matter more on the defensive end.

Interesting question.

Sam

Sam wrote:Outside,

How tall is Schooner? Just asking? Oh, and does he like the color green?

Sam

Being a standard poodle, Schooner is about 3 ft tall at the head, give or take some poodle curls, and can jump over 6 ft walls with ease. And considering that green tennis balls are his favorite, I'd say he likes green a lot (even if dogs are color blind).

Outside