Are these Celtics the worst offensive rebounding team in history?

[BballTim]

The problem with offensive rebounding stats is that the sample size is always so small. I mean, the difference between the top and the bottom team in offensive rebounding (Chicago and us, respectively), is 6. The difference between 1st and 20th is 3.

There are just so few offensive rebound opportunities in any basketball game, that a difference of one or two boards makes a huge impact on the percentage/efficiency numbers, that may be out of proportion to its actual impact on the game.

  The way you have to look at it is each offensive rebound is basically an extra possession, and teams average close to a point per possession. The Knicks are 22nd in offense, another three offensive rebounds and they might be 8th or so in offense.

No, an ORB is not an 'extra possession'.  It is simply a continuation of the current possession.

  Obviously this is true in terms of points per possession measurements. But my point was that it gives you another crack at scoring on your possession. Getting more offensive rebounds will increase your scoring efficiency.

Yes, indeed, IF you grab an ORB, it is without question a good thing and indeed increases your chances to score on that possession (versus zero if you do not grab it).  The strategic question though, is whether it is worth the _effort_ to try to grab it.   As I indicated, most team's have about a 25% chance of grabbing an ORB.  The best teams have about 30%.  That's your odds of getting it back on an ORB once the ball leaves your hand on the shot (if it doesn't go in).   Alternatively, if you pursue a tight transition D strategy that forces low percentage shots, your chance of getting it back before they score is significantly higher.

And getting it back before they score is equally without question a good thing as well.

  Again, this is clearly true, but somewhat unrelated to my point that an extra 3-6 offensive rebounds can have a significant effect on your relative offensive efficiency. Whether it makes strategic sense to pursue those offensive rebounds is another issue.

This gets to the fundamental game-value difference between offensive rebounds and defensive rebounds.  A DRB is 100% the stoppage of a current possession with zero points.  An ORB is just a continued _chance_ to score on the current possession.

  This is fairly nebulous. When you get an offensive rebound you prevent the opponent from getting a defensive rebound from that shot. You're preventing them from realizing that game-value, which should have equal value for you.

No, nothing nebulous about the difference.  There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play).  Points surrendered on defensive possessions are equally as important as points scored on offensive possessions.  Thus a 100% stoppage of a defensive possession with no points is worth more game value than 50-60% chance to still score on an offensive possession.

  First of all, basketball's something of a zero sum game in many respects. Defensive rebounds end possessions, offensive rebounds prevent the opponent from having an offensive possession. In reality, it's the same for either team.

  Also, you state "There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play)". This is true in terms of offensive efficiency. But if you look at scoring chances (fga + .44 * fta) you see that, while the Celts have about the same number of possessions, their opponents have 4-5 more scoring chances a game. That means, in order to win, we need to get more points from about 85.5 scoring chances than our opponent gets from 90.

[indeedproceed]

Statistically, what constitutes the 'end of a possession'?

For individual players, possessions are measured as a counting number between assists, turnovers, FGA's, and FTA's.

For teams, wouldn't it be the same, except without measuring assists as a counting number signifying the end of a possession?

So, basketball-wise, an OReb isn't an 'extra' possession, but is it statistically? You just attempted a field goal (possession), and you got a rebound that can only end in a FGA, turnover, or foul shots (possession). So that's 2, right?

[alajet]

The one thing that ORB% does tend to correlate with is low FG%.  Teams that miss a lot of shots tend to 'compensate' by fighting for rebounds to improve their possession scoring efficiency.

Plus, there are more opportunities for ORB if there are more misses.

We are a high FG% high TO team, both of which mean less misses to rebound.

I'm not the one to throw this out, but JVG recently stated this held no relation at all. Don't remember which game it was. To be honest, to some degree, I'm with this idea.

[BballTim]

Statistically, what constitutes the 'end of a possession'?

For individual players, possessions are measured as a counting number between assists, turnovers, FGA's, and FTA's.

For teams, wouldn't it be the same, except without measuring assists as a counting number signifying the end of a possession?

So, basketball-wise, an OReb isn't an 'extra' possession, but is it statistically? You just attempted a field goal (possession), and you got a rebound that can only end in a FGA, turnover, or foul shots (possession). So that's 2, right?

  For a team, a possession ends when the other team gets the ball, whether it's a rebound, made basket or turnover. Calculating possessions for a team is generally something like fga + (.44 * fta) + turnovers - offensive rebounds. So if your team gets the ball from the opponent 5 times, has no turnovers and gets 3 offensive rebounds, they can get 8 shots in those 5 possessions. If they make 4 of those 8 shots (all 2 pointers) they'd end up with 8 points on 5 possessions, or 1.6 ppp.

  For a player a possession generally ends in a fga, fta or turnover, assists aren't generally counted as possessions.

[guava_wrench]

No, nothing nebulous about the difference.  There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play).  Points surrendered on defensive possessions are equally as important as points scored on offensive possessions.  Thus a 100% stoppage of a defensive possession with no points is worth more game value than 50-60% chance to still score on an offensive possession.

You don't see how this is apples and oranges?

Your '100% stoppage' is actually only stopping them from having the '50-60% chance' to score by retaining the ball for another scoring attempt. Let them keep the ball and they still have just as much chance of not scoring as you would when you get your offensive rebound. This is why what you are saying is nebulous.

Likewise, not getting the offensive rebound means a 100% chance of you not scoring on that possession in the exact same way as your 100% stoppage.

[banty19]

The Celtics lack of offensive rebounds is a strategy!

Heinsohn has said this many times on air. He may be getting old, but he still knows the game and still has inside info.

They'd rather be set up defensively than go for a few extra possessions.

[mmmmm]

  First of all, basketball's something of a zero sum game in many respects. Defensive rebounds end possessions, offensive rebounds prevent the opponent from having an offensive possession. In reality, it's the same for either team.

I think this is the part that you and guava keep sticking on.  The second part of the second sentence is simply NOT true EXCEPT in the last possession of a period.  An offensive rebound does NOT 'prevent the opponent from having an offensive possession'.  The opponent WILL still eventually get the ball back.  An offensive rebound does not change that.  It merely prolongs your current possession.   

 Also, you state "There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play)". This is true in terms of offensive efficiency. But if you look at scoring chances (fga + .44 * fta) you see that, while the Celts have about the same number of possessions, their opponents have 4-5 more scoring chances a game. That means, in order to win, we need to get more points from about 85.5 scoring chances than our opponent gets from 90.

Scoring chances are, indeed increased by ORBs. But THAT is its own form of zero sum game because you cannot have an ORB unless you first had a miss. The 'miss' was a scoring chance that netted zero points.  So the 'extra scoring chances' produced by ORBs are partially fools gold.

Ultimately, what matters is points produced per possession.  Whether that takes 3 shots and 2 ORBs or just one shot is irrelevant.

Strategically, because defenses - even poor rebounding defenses - will tend to grab the vast majority of missed shots, you are best served if your _first_ shot of each possession is as high percentage a shot as possible.  Because even if you are the best offensive rebounding team in the league, the odds are you won't get the rebound.

[BballTim]

  First of all, basketball's something of a zero sum game in many respects. Defensive rebounds end possessions, offensive rebounds prevent the opponent from having an offensive possession. In reality, it's the same for either team.

I think this is the part that you and guava keep sticking on.  The second part of the second sentence is simply NOT true EXCEPT in the last possession of a period.  An offensive rebound does NOT 'prevent the opponent from having an offensive possession'.  The opponent WILL still eventually get the ball back.  An offensive rebound does not change that.  It merely prolongs your current possession.   

  What you're saying is only true in the statistical world. Look at the reality though.

  When you get an offensive rebound, you prevent the team from getting a defensive rebound. If the opponent gets the defensive rebound, they get a possession. So preventing a defensive rebound prevented that possession from occurring. Instead of defending that possession and getting the ball back to try and score, you immediately get to try and score.

  Of course I realize that in terms of calculating possessions for a statistical comparison that offensive rebound extends the current possession, but do you see any errors in the above paragraph? You're getting caught up in semantics.

 Also, you state "There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play)". This is true in terms of offensive efficiency. But if you look at scoring chances (fga + .44 * fta) you see that, while the Celts have about the same number of possessions, their opponents have 4-5 more scoring chances a game. That means, in order to win, we need to get more points from about 85.5 scoring chances than our opponent gets from 90.

Scoring chances are, indeed increased by ORBs. But THAT is its own form of zero sum game because you cannot have an ORB unless you first had a miss. The 'miss' was a scoring chance that netted zero points.  So the 'extra scoring chances' produced by ORBs are partially fools gold.

Ultimately, what matters is points produced per possession.  Whether that takes 3 shots and 2 ORBs or just one shot is irrelevant.

  Ultimately, what matters is scoring more points than your opponent. Points produced per possession is merely a way of quantifying how well you score. Also, you can't have an offensive rebound if you don't miss a shot, but you can miss a shot without getting an offensive rebound. Getting an extra shot because you got an offensive rebound isn't fools gold because you'll generally average a certain amount of points off of those offensive rebounds. Those points are real.

Strategically, because defenses - even poor rebounding defenses - will tend to grab the vast majority of missed shots, you are best served if your _first_ shot of each possession is as high percentage a shot as possible.  Because even if you are the best offensive rebounding team in the league, the odds are you won't get the rebound.

  Wait, so let me get this straight. You're saying that teams should try and score when they shoot because they might not get an offensive rebound? That's brilliant. I wonder if there's a way we could pass that insight on to Doc, the Celts would be practically unbeatable if they followed that strategy.

[mmmmm]

  First of all, basketball's something of a zero sum game in many respects. Defensive rebounds end possessions, offensive rebounds prevent the opponent from having an offensive possession. In reality, it's the same for either team.

I think this is the part that you and guava keep sticking on.  The second part of the second sentence is simply NOT true EXCEPT in the last possession of a period.  An offensive rebound does NOT 'prevent the opponent from having an offensive possession'.  The opponent WILL still eventually get the ball back.  An offensive rebound does not change that.  It merely prolongs your current possession.   

  What you're saying is only true in the statistical world. Look at the reality though.

  When you get an offensive rebound, you prevent the team from getting a defensive rebound. If the opponent gets the defensive rebound, they get a possession. So preventing a defensive rebound prevented that possession from occurring. Instead of defending that possession and getting the ball back to try and score, you immediately get to try and score.

  Of course I realize that in terms of calculating possessions for a statistical comparison that offensive rebound extends the current possession, but do you see any errors in the above paragraph? You're getting caught up in semantics.

 Also, you state "There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play)". This is true in terms of offensive efficiency. But if you look at scoring chances (fga + .44 * fta) you see that, while the Celts have about the same number of possessions, their opponents have 4-5 more scoring chances a game. That means, in order to win, we need to get more points from about 85.5 scoring chances than our opponent gets from 90.

Scoring chances are, indeed increased by ORBs. But THAT is its own form of zero sum game because you cannot have an ORB unless you first had a miss. The 'miss' was a scoring chance that netted zero points.  So the 'extra scoring chances' produced by ORBs are partially fools gold.

Ultimately, what matters is points produced per possession.  Whether that takes 3 shots and 2 ORBs or just one shot is irrelevant.

  Ultimately, what matters is scoring more points than your opponent. Points produced per possession is merely a way of quantifying how well you score. Also, you can't have an offensive rebound if you don't miss a shot, but you can miss a shot without getting an offensive rebound. Getting an extra shot because you got an offensive rebound isn't fools gold because you'll generally average a certain amount of points off of those offensive rebounds. Those points are real.

Strategically, because defenses - even poor rebounding defenses - will tend to grab the vast majority of missed shots, you are best served if your _first_ shot of each possession is as high percentage a shot as possible.  Because even if you are the best offensive rebounding team in the league, the odds are you won't get the rebound.

  Wait, so let me get this straight. You're saying that teams should try and score when they shoot because they might not get an offensive rebound? That's brilliant. I wonder if there's a way we could pass that insight on to Doc, the Celts would be practically unbeatable if they followed that strategy.

a) What I'm saying is true in the real world.  There is no separate 'statistical world'. 

b) You are the one trying to contrive a new meaning for the term 'possession' - please don't accuse me of arguing with semantics.

c) The point of my last comment goes directly to the strategy because you want to _prevent_ your opponent from taking high percentage shots.  Thus, transition D.  Force them to take perimeter jump shots instead of transition layups.

d) For your viewing pleasure:  tonight's game against the Heat, who 'won' the ORB battle.

http://www.nba.com/games/20120410/BOSMIA/gameinfo.html

[azzenfrost]

We shot 60+% and they shot 40+%. The Celtics will let them win the offensive rebound battle and bury them everywhere else.

[indeedproceed]

  For a team, a possession ends when the other team gets the ball, whether it's a rebound, made basket or turnover. Calculating possessions for a team is generally something like fga + (.44 * fta) + turnovers - offensive rebounds. So if your team gets the ball from the opponent 5 times, has no turnovers and gets 3 offensive rebounds, they can get 8 shots in those 5 possessions. If they make 4 of those 8 shots (all 2 pointers) they'd end up with 8 points on 5 possessions, or 1.6 ppp.

Ah, TP. didn't even think to just subtract off rebounds. Solves the entire issue.

[BballTim]

  First of all, basketball's something of a zero sum game in many respects. Defensive rebounds end possessions, offensive rebounds prevent the opponent from having an offensive possession. In reality, it's the same for either team.

I think this is the part that you and guava keep sticking on.  The second part of the second sentence is simply NOT true EXCEPT in the last possession of a period.  An offensive rebound does NOT 'prevent the opponent from having an offensive possession'.  The opponent WILL still eventually get the ball back.  An offensive rebound does not change that.  It merely prolongs your current possession.   

  What you're saying is only true in the statistical world. Look at the reality though.

  When you get an offensive rebound, you prevent the team from getting a defensive rebound. If the opponent gets the defensive rebound, they get a possession. So preventing a defensive rebound prevented that possession from occurring. Instead of defending that possession and getting the ball back to try and score, you immediately get to try and score.

  Of course I realize that in terms of calculating possessions for a statistical comparison that offensive rebound extends the current possession, but do you see any errors in the above paragraph? You're getting caught up in semantics.

 Also, you state "There are a finite number of possessions in each game and you will each have almost the exact same number of them (about 90 each at the pace the Celtics play)". This is true in terms of offensive efficiency. But if you look at scoring chances (fga + .44 * fta) you see that, while the Celts have about the same number of possessions, their opponents have 4-5 more scoring chances a game. That means, in order to win, we need to get more points from about 85.5 scoring chances than our opponent gets from 90.

Scoring chances are, indeed increased by ORBs. But THAT is its own form of zero sum game because you cannot have an ORB unless you first had a miss. The 'miss' was a scoring chance that netted zero points.  So the 'extra scoring chances' produced by ORBs are partially fools gold.

Ultimately, what matters is points produced per possession.  Whether that takes 3 shots and 2 ORBs or just one shot is irrelevant.

  Ultimately, what matters is scoring more points than your opponent. Points produced per possession is merely a way of quantifying how well you score. Also, you can't have an offensive rebound if you don't miss a shot, but you can miss a shot without getting an offensive rebound. Getting an extra shot because you got an offensive rebound isn't fools gold because you'll generally average a certain amount of points off of those offensive rebounds. Those points are real.

Strategically, because defenses - even poor rebounding defenses - will tend to grab the vast majority of missed shots, you are best served if your _first_ shot of each possession is as high percentage a shot as possible.  Because even if you are the best offensive rebounding team in the league, the odds are you won't get the rebound.

  Wait, so let me get this straight. You're saying that teams should try and score when they shoot because they might not get an offensive rebound? That's brilliant. I wonder if there's a way we could pass that insight on to Doc, the Celts would be practically unbeatable if they followed that strategy.

a) What I'm saying is true in the real world.  There is no separate 'statistical world'. 

b) You are the one trying to contrive a new meaning for the term 'possession' - please don't accuse me of arguing with semantics.

c) The point of my last comment goes directly to the strategy because you want to _prevent_ your opponent from taking high percentage shots.  Thus, transition D.  Force them to take perimeter jump shots instead of transition layups.

d) For your viewing pleasure:  tonight's game against the Heat, who 'won' the ORB battle.

http://www.nba.com/games/20120410/BOSMIA/gameinfo.html

  Again, all I said was that getting 3-6 more offensive rebounds a game can have a significant effect on your relative offensive efficiency. I never said that the Celts should forgo their transition defense to get more offensive rebounds, or that concentrating on offensive rebounds was a sound strategy. In fact I've mentioned the fact that the Celts are a top transition defense team in most of the threads about our offensive rebounding woes over the last 2-3 years, and that's the tradeoff that they make. And I wasn't trying to derive a new meaning for "possession", in fact I used the term "scoring chance" to avoid that.

  Most people know that the Celts don't get offensive rebounds because they're getting back on defense and not attacking the glass after they shoot. Everyone knows that you should try and force your opponent to take perimeter jumpers (in particular long 2 point jumpers) whenever possible.  You're simply stating the obvious.

[Greenbean]

The problem with offensive rebounding stats is that the sample size is always so small. I mean, the difference between the top and the bottom team in offensive rebounding (Chicago and us, respectively), is 6. The difference between 1st and 20th is 3.

There are just so few offensive rebound opportunities in any basketball game, that a difference of one or two boards makes a huge impact on the percentage/efficiency numbers, that may be out of proportion to its actual impact on the game.

 

It's nowhere close to a small sample size.  This is not an individual game stat.  It's an entire season.  Actually, it's multiple entire seasons.  This is our 2nd consecutive season of being historically awful on the offensive glass.  We haven't fielded a top 10 offense since our ORB% fell off the face of the earth post-Powe.

It's just extraordinarily frustrating that a team with tremendous defense and great shooting has been held back by our historic incompetence on the offensive boards.

I dont get your correlation.

We have great defense, and great shooting.

What holds us back is poor defensive reebounding and turnovers...the same things that have held us back post 07-08.

Maybe offensive rebounding would get us some of the possesions back that we lose to GIVING UP offensive boards and turnovers, but it isnt the root of the issue.

You should check out the data on basketball-reference.  In 07/08, we were the 2nd most turnover-prone offense and the 8th best defensive rebounding team.  Last year we were tied for 3rd most turnover-prone and and the 9th best defensive rebounding team.  Both teams were elite defensively.

The difference is offensive rebounding.  In 07/08 and 08/09 we had top 10 offenses that  could put the hurt on opponents in multiple ways: shooting, slashing and pounding the boards (Powe was great at it, PJ Brown was good at it and Perk/Baby could occasionally help) - we were always turnover-prone.  In the years since, we've still been able to shoot and slash, but we've lost all, and I mean all, ability to crash the offensive boards.

That single factor has played a greater role than any other in reducing this team to offensive mediocrity (and outright ineptitude this year), which in turn has reduced our chances at a title.

We werent good offensive rebounders in 07-08 or 08-09 statistically (bottom third).

Still dont get the correlation.

The difference in our team offensively has been our decrease in efficiency and also major decrease in pace. This is likely a result of our main scorers getting older. Offense goes stale with age, defense ages like wine.

If you want a better stat to correlate, look at free throw attempts...we a re at the bottom of the league this year.

We are top 5 in field goal percentage.