2011 Finals - Mia vs Dal

[Mike G]

His greatest contribution was his offensive boarding. Miami centers rebounded very poorly and Tyson got a bunch of those available boards. He had 24 offensive rebounds in 223 minutes. Dirk had 2 in 241 minutes.
..

Sorry, I don't see how Miami's lack of a good center makes Chandler more valuable. Especially not more valuable than Nowitzki.
Is an offensive rebound as valuable as a defensive rebound plus a point or two?

... it seems like every Miami counterpart outperformed their contributions in the series, making whatever they did "negative".

Other than Wade, we guess.
So again, if the 2 best players in a series happen to play the same position, only one can be said to have played well? Or both were just mediocre?
Doesn't the best defender generally cover the offensively strongest opponent? Whether or not he's the positional 'counterpart'?

[Bobbofitos]

His greatest contribution was his offensive boarding. Miami centers rebounded very poorly and Tyson got a bunch of those available boards. He had 24 offensive rebounds in 223 minutes. Dirk had 2 in 241 minutes.
..

Sorry, I don't see how Miami's lack of a good center makes Chandler more valuable. Especially not more valuable than Nowitzki.

It's a matter of perspective, then. I would say it means Chandler was more valuable in a vacuum. You would say, well, Chandler was expected to beat up on the MIA centers, so his performance did not over exceed pre-series expectations.

Is an offensive rebound as valuable as a defensive rebound plus a point or two?

An offensive rebound is equal to the EV of whatever that missed shot was worth to the new possession. A defensive rebound is equal to the EV of new possession minus the missed shot. In practice it means oboarding is roughly worth double that of dboarding. So, yeah, Tyson doing most of his work on the offensive glass does mean something.

... it seems like every Miami counterpart outperformed their contributions in the series, making whatever they did "negative".

Other than Wade, we guess.
So again, if the 2 best players in a series happen to play the same position, only one can be said to have played well? Or both were just mediocre?

I feel like we've said this, just said it 2 weeks ago in a slightly different manner. To me, it means they aren't the 2 best players in the series! (Unless they're playing each other to a draw and totally going off on the backup)

Over the course of the season against 29 teams 2 different players can be thought of as the best. But in a series between those 2 players - when they are in fact guarding one another - unless you define "best" to mean pure offensive production - those 2 cannot be the best.

Doesn't the best defender generally cover the offensively strongest opponent? Whether or not he's the positional 'counterpart'?

Sure, if you rechart everything correctly, then the counterpart actually means quite a bit more.

To boot, Dirk didn't really guard Bosh much. It was Tyson on Bosh, and he did a fantastic job. Dirk was hidden on Joel or Haslem for much of the series.

Mike, maybe another argument, but what's more valuable:
A player who never scores & never uses any possessions, but also holds his counterpart to the exact same; OR
A player who uses 25% of his team's possessions at a league average clip, but allows his counterpart to do the exact same?

I think you think the latter is more valuable, whereas I think they're pretty identical.

[Mike G]

I would say it means Chandler was more valuable in a vacuum. You would say, well, Chandler was expected to beat up on the MIA centers, so his performance did not over exceed pre-series expectations.

I'd say those factors are irrelevant. Nowitzki was more valuable.

An offensive rebound is equal to the EV of whatever that missed shot was worth to the new possession. A defensive rebound is equal to the EV of new possession minus the missed shot. In practice it means oboarding is roughly worth double that of dboarding. So, yeah, Tyson doing most of his work on the offensive glass does mean something

I've never bought into that accounting. The missed shot and the rebound are separate events, regardless of which team gets the rebound. The rebound is a possession, to whomever gets it.

..Over the course of the season against 29 teams 2 different players can be thought of as the best. But in a series between those 2 players - when they are in fact guarding one another - unless you define "best" to mean pure offensive production - those 2 cannot be the best.

Irrelevant how I define anything. Jordan and Magic, Hakeem and Ewing/Shaq, Duncan vs Shaq, Dirk-Durant... -- these are the best players in their series, whomever they guard.

A player who never scores & never uses any possessions, but also holds his counterpart to the exact same; OR
A player who uses 25% of his team's possessions at a league average clip, but allows his counterpart to do the exact same?

I think you think the latter is more valuable, whereas I think they're pretty identical.

Still not sure if by 'counterpart' you mean the same alleged position or the players guarding one another.
If a guy does not score, why is he guarding a guy who doesn't score? Why is he in the lineup?
What does league average have to do with a series or a game? League avg efficiency might be priceless in a 80-78 game and useless in a 140-138 game.

In general, there's great value in a player who can get a shot. Each player on the floor who cannot do this is something of a liability. And it's exponential: An offense can tolerate one such player, two if there are some great offensive players with him. Beyond that, it's a serious problem. It cuts into the team efficiency.

[EvanZ]

In general, there's great value in a player who can get a shot. Each player on the floor who cannot do this is something of a liability. And it's exponential: An offense can tolerate one such player, two if there are some great offensive players with him. Beyond that, it's a serious problem. It cuts into the team efficiency.

It works the other way too, doesn't it? A team can tolerate one weak defender, and maybe two, if they are surrounded by very strong defenders. Dwight Howard should become much more valuable if he is surrounded by teammates who aren't good defenders. Does eWins capture that?

(This is not to say that Tyson Chandler was more valuable than Dirk. I think it can be argued that he, indeed, was surrounded by better defenders, relative to Dirk being surrounded by relatively worse offensive threats.)

[Mike G]

It works the other way too, doesn't it? A team can tolerate one weak defender, and maybe two, if they are surrounded by very strong defenders...

Heck yeah.

..Dwight Howard should become much more valuable if he is surrounded by teammates who aren't good defenders. Does eWins capture that?

Actually, I think better co-defenders make your own defense more valuable. This is especially true if you are a weak defender, but perhaps also if you are elite.
I don't know who is a weak defender for Orlando. In general, if you play good D you get more minutes. In these minutes, you score and rebound and etc.
In eWins, points are scaled to opponent points, rebounds to opponent rebounds. If you are in the game and producing, your team's strong defense scales your productions upward.
Dwight gets minutes even when his offense isn't clicking. So he goes on accruing rebounds, blocks, etc. Unless their bench is very thin, weaker defenders shouldn't be on the floor when they aren't producing.

[Bobbofitos]

I would say it means Chandler was more valuable in a vacuum. You would say, well, Chandler was expected to beat up on the MIA centers, so his performance did not over exceed pre-series expectations.

I'd say those factors are irrelevant. Nowitzki was more valuable.

And I say Tyson was more valuable. It's a circular argument. In the MIA-DAL series, Tyson defended better and rebounded much better. Dirk's offense wasn't enough to offset this.

An offensive rebound is equal to the EV of whatever that missed shot was worth to the new possession. A defensive rebound is equal to the EV of new possession minus the missed shot. In practice it means oboarding is roughly worth double that of dboarding. So, yeah, Tyson doing most of his work on the offensive glass does mean something

I've never bought into that accounting. The missed shot and the rebound are separate events, regardless of which team gets the rebound. The rebound is a possession, to whomever gets it.

You call it accounting, I call it the proper way to convert each PBP action into an EV shift of points for/points against.

All rebounds are attributable to the creator of the shot (on offense) or the contestor of the shot (on defense). Rather than giving full credit to one or the other, every possession off a certain possession type has value depending on the shot clock. Once the shot is taken the defender gets credit for the difference between the new possession types, just as a rebound swings a possession differently.

..Over the course of the season against 29 teams 2 different players can be thought of as the best. But in a series between those 2 players - when they are in fact guarding one another - unless you define "best" to mean pure offensive production - those 2 cannot be the best.

Irrelevant how I define anything. Jordan and Magic, Hakeem and Ewing/Shaq, Duncan vs Shaq, Dirk-Durant... -- these are the best players in their series, whomever they guard.

Sure, they are the best players, but that doesn't mean they *all* have the best series. That's the core argument here - in no way am I saying Chandler was more important to the Mavs for the season. (He wasn't, although he was their 2nd best player) But for this series, he was.

A player who never scores & never uses any possessions, but also holds his counterpart to the exact same; OR
A player who uses 25% of his team's possessions at a league average clip, but allows his counterpart to do the exact same?

I think you think the latter is more valuable, whereas I think they're pretty identical.

Still not sure if by 'counterpart' you mean the same alleged position or the players guarding one another.

We can say either allegedly guarding one another or, more precisely, actually guarding one another. In this case, I mean, guarding one another.

If a guy does not score, why is he guarding a guy who doesn't score? Why is he in the lineup?

It's just a hypothetical. He's doing such a great job on defense that he doesn't even allow his opponent to shoot. If you're more comfortable with a closer-to-practical example, we could've swapped 15% on lg average offensively/defensively.

What does league average have to do with a series or a game? League avg efficiency might be priceless in a 80-78 game and useless in a 140-138 game.

I figured you'd get a little caught up over this. It doesn't matter, but the point is you're allowing the same efficiency both ways.

In general, there's great value in a player who can get a shot. Each player on the floor who cannot do this is something of a liability. And it's exponential: An offense can tolerate one such player, two if there are some great offensive players with him. Beyond that, it's a serious problem. It cuts into the team efficiency.

This isn't really news to me. What is puzzling to me, however, is why you don't hold the same truism on the opposite end of the floor.

[Bobbofitos]

It works the other way too, doesn't it? A team can tolerate one weak defender, and maybe two, if they are surrounded by very strong defenders.

Is this true?

[Crow]

It could be tested using EZPM D100, Defensive Rating, defensive RAPM and maybe other measures if that player level information was matched up with season summary lineup defensive efficiency data and then average team defensive efficiency performance was calculated for the different combinations of strong and weak defensive players.

From previous brief checking of raw and Adjusted lineup +/- I thought I saw surface indications that 2 weak defenders (by defensive Adjusted +/- and / or reputation) does generally tend to cause significant problems. Not sure if strong players can save them or not, didn't focus on that.

[huevonkiller]

If you need a special circumstance to be "more valuable", then you're not more valuable. Playing a weaker defense and not getting double teamed, that doesn't indicate MVP status. That's just abusing a serendipitous matchup and ignoring why the APBR exists.


Wade had a great series because the Mavs and the matchups allowed him to that. Against Chicago he wished he could play like LeBron did against Dallas, that's how bad he was. It is all relative to the circumstances on the court. An MVP is the best overall player after facing multiple elite teams and outperforming his peers. Sample Size is the ultimate truth.

Kobe Bryant played possessed against the Phoenix Suns in 2010, because they're porous defensively. Conversely he was mediocre against the 2001 Sixers but was just as valuable as Shaq in the playoffs, if not more. The MVP shouldn't be decided by just getting lucky.