Semi-OT: Hollinger talks about Durant

[andalusian]

Green and Durant play the most minutes on that young squad and Green is lacking just as much on D. They play at pretty much the same time which accounts for part of the problem.

Yeah, Green is not a good defensive player (but his on/off difference is not as bad as Durant's with -5 vs. -8 per 100 possessions). It is interesting to look at their offensive on/off performance - and again, Green seems to contribute more (they are actually + with him on the court offensively, while they are - with Durant. You will notice that Green also has a 29.9% win rate, while Durant is only at 23.6% - for a team that won at a little better than 25% - this again raises eyebrows.

I wasn't talking about LeBron as a rookie, he still sucked defensively his second year just not as much as Durant, maybe. Playing next to Green in that environment, who knows really.

My entire argument that LeBron looked like he played defense from the get-go was based on observation - that was also confirmed by his rookie year statistics. I did not look at his 2nd year stats - but since you mentioned it I went and looked at it - and his relative defensive contribution seems to be about the same - his team was better defensively when he was on the court.

Again - playing defense does not mean he is a defensive stopper - but it means that he looks like an NBA player there paying attention to both sides of the floor.

 

[huevonkiller]

Yeah, Green is not a good defensive player (but his on/off difference is not as bad as Durant's with -5 vs. -8 per 100 possessions). It is interesting to look at their offensive on/off performance - and again, Green seems to contribute more (they are actually + with him on the court offensively, while they are - with Durant. You will notice that Green also has a 29.9% win rate, while Durant is only at 23.6% - for a team that won at a little better than 25% - this again raises eyebrows.

I don't understand why you think +/- is the only figure to be fixated on

Durant plays more minutes than Green, and Green's defensive PER is pretty much just as awful. -5 per 100 possessions is still awful even for the Okc Thunder. If I had to guess it would be a combination of Durant's lack of individual D, combined with Green. That sounds more accurate than to just pin it all on Kevin.

My entire argument that LeBron looked like he played defense from the get-go was based on observation - that was also confirmed by his rookie year statistics. I did not look at his 2nd year stats - but since you mentioned it I went and looked at it - and his relative defensive contribution seems to be about the same - his team was better defensively when he was on the court.

Just off hand without looking at anything, I've heard about Bron's lack of D early on in his career. Even today Pick and Roll in particular still troubles him, so I wasn't too surprised.

I don't think you want to get into this narrow argument about +/-, if you do LeBron is a much worse defender his junior year (-3 points per 100). I think people should take into account a bit of everything, including the details/nuances of these equations. I don't put stock into just one thing.

 

[andalusian]

Yeah that's right Hollinger didn't fully seem to accept +/-.

Well then, what does Defensive rating even mean? That's not exactly a precise stat either.

Most stats are not precise - that's why you have concepts like confidence level and error-margin in statistics. But what they can do, is give you an idea about trends.

For example - I can tell you that I looking at the Blazers +/- defensive and offensive ratings from last year confirm my opinion. For example - I always thought that Aldridge is a much (or at least, so far has been) better defensive player than Roy - but Roy was a better offensive player - their on/off statistics seem to confirm that. It is hard to say for sure because they play different positions - and guard different players - but the numbers seem to confirm it.

If we want more head-to-head comparisons - it was clear that Oden is a much better offensive player than Pryzbilla but it was completely reversed on defense - even if Oden was pretty decent on individual defense against other centers - if you look at the opposing center PER - Oden and Pryzbilla had the exact same defensive PER stat - but Pryzbilla had a much much better DRTG - how can that be? Clearly because Pryzbilla was able to switch much better on pick and rolls, did not get into stupid foul trouble on these pick and roll situations - and the like.

Defensive rating by itself compared across multiple teams and years is clearly an imprecise stat - but when you look at it compared to the rest of your team - you can start noticing some trends...

 

[andalusian]

I don't think you want to get into this narrow argument about +/-, if you do LeBron is a much worse defender his junior year (-3 points per 100). I think people should take into account a bit of everything, including the details/nuances of these equations. I don't put stock into just one thing.

Honestly, I do not know. That's a good point. I just do not have the details about it - maybe he was just because they had a new coach )Mike Brown's first year) - and he did not grasp the changes in the defensive strategy?

I did not pay that much attention to him that year - so I really find it hard to correlate it.

Anyway... let's just say that at this point I think that Durant looks a lot worse on defense than LeBron did in his early years. The good news is that Durant is still young - and maybe having the same coach for the 2nd year will make it easier for him this year...

 

[huevonkiller]

Most stats are not precise - that's why you have concepts like confidence level and error-margin in statistics. But what they can do, is give you an idea about trends.

For example - I can tell you that I looking at the Blazers +/- defensive and offensive ratings from last year confirm my opinion. For example - I always thought that Aldridge is a much (or at least, so far has been) better defensive player than Roy - but Roy was a better offensive player - their on/off statistics seem to confirm that. It is hard to say for sure because they play different positions - and guard different players - but the numbers seem to confirm it.

If we want more head-to-head comparisons - it was clear that Oden is a much better offensive player than Pryzbilla but it was completely reversed on defense - even if Oden was pretty decent on individual defense against other centers - if you look at the opposing center PER - Oden and Pryzbilla had the exact same defensive PER stat - but Pryzbilla had a much much better DRTG - how can that be? Clearly because Pryzbilla was able to switch much better on pick and rolls, did not get into stupid foul trouble on these pick and roll situations - and the like.

Defensive rating by itself compared across multiple teams and years is clearly an imprecise stat - but when you look at it compared to the rest of your team - you can start noticing some trends...

Yeah I'll agree with you there man. Sometimes one can be mislead by playing against backups/semi-garbage time, so a large size of minutes helps me comprehend the game better.

Shannon Brown for instance seems hard to figure out. I want to see how he does over the course of a season. I don't question his defense just how long he can maintain that deep perimeter shooting. Little things like that are troublesome, but comparing starters on the same team is easier for me.

 

[huevonkiller]

Honestly, I do not know. That's a good point. I just do not have the details about it - maybe he was just because they had a new coach )Mike Brown's first year) - and he did not grasp the changes in the defensive strategy?

I did not pay that much attention to him that year - so I really find it hard to correlate it.

Anyway... let's just say that at this point I think that Durant looks a lot worse on defense than LeBron did in his early years. The good news is that Durant is still young - and maybe having the same coach for the 2nd year will make it easier for him this year...

Not as easy to track defense in general, probably the most complicated issue in sports. +/- has its merits sometimes.

Hopefully Durant puts on some more weight I suppose? Don't know if it is that simple though. I wonder when he'll finally become average on that end.

 

[JayJohnstone]

+/- has huge swings.

The off-court minutes are only about 1/2 a season of basketball. When Durant's +/- goes positive and then to superstar levels, who will even know if it's because of random noise, the fact that he was 19/20 without a pro-body, b/c the GM has teammates that finally better complement him, or b/c he plays smarter. It's probably some combination of all of the above. But it's going to happen.

You'd have 4 guys standing around watching Zbo run down the shot clock forcing a shot...

And why would Zbo still have to be on the team?

 

[Deebag]

What the hell is up with some NBA player's twitter posts? Durant and Oden to be specific. They're both about 21 years old but you surely can't tell from their twitters and blogs. I mean, I know 21 is young, but at the same time, it's really not.

 

[jlprk]

Jason Quick loves a childlike personality. He doesn't understand a 33-year-old, 10-year veteran. That's why Blazer players take so many lazy years to develop--if they're the hard-driving types, they have always been run out of town by the Oregonian.

 

[oldmangrouch]

Jason Quick loves a childlike personality. He doesn't understand a 33-year-old, 10-year veteran. That's why Blazer players take so many lazy years to develop--if they're the hard-driving types, they have always been run out of town by the Oregonian.

So Quick's man-crush on Damon was some kind of pseudo-pedophillia?

You may be on to something!

 

[STOMP]

You need to respond to his criticisms of defensive PER. Does Hollinger (who invented PER, right?) even acknowledge the existence of such a thing? Seriously, if a team played zone permanently, would it mean ANYTHING?

solid point. When playing man D it's normal for guys to switch on who they're guarding several times in a single defensive set.

Defensive PER is a poor measure of whats what.

STOMP

 

[STOMP]

btw... judging strictly from the team USA clips we saw this summer it looked like Durant had stepped up his defensive intensity quite a bit. Harden gives them a second scorer so maybe he'll be able to shift his focus a bit this season.

STOMP

 

[huevonkiller]

solid point. When playing man D it's normal for guys to switch on who they're guarding several times in a single defensive set.

Defensive PER is a poor measure of whats what.

STOMP

Defensive Rating also has issues, they are both team-based.

 

[andalusian]

More stuff popped up today. This is truly an interesting subject for anyone that is interested in basketball statistics, imho.

http://www.basketballprospectus.com/article.php?articleid=732

 

[PtldPlatypus]

Defensive Rating also has issues, they are both team-based.

Yes, but defensive rating is a measure of what an entire team does against your team while you play, whereas defensive PER is just a measure of what players at a particular position do. Opposing teams' performance is of much more significance than a single player's.

 

[huevonkiller]

Yes, but defensive rating is a measure of what an entire team does against your team while you play, whereas defensive PER is just a measure of what players at a particular position do. Opposing teams' performance is of much more significance than a single player's.

An opposing team is not guarded by just one player, which is my point. On/Off court numbers were also discussed before.