Semi-OT: Hollinger talks about Durant

Welcome to our community

Be a part of something great, join today!

Register Log in
santeesioux

santeesioux

Just keep on scrolling by
Joined
Dec 3, 2008
Messages
10,752
Likes
5,333
Points
113
Durant is the biggest discrepancy I've ever seen. For those who don't know, Durant's on-court vs. off-court numbers from this first two seasons were absolutely horrid -- the Thunder were massively better, by more than eight points per 100 possessions, when he was off the court. We're still learning about adjusted plus-minus stats and the things that they can and can't tell us, but suffice it to say a third straight year of such numbers would cause some very pointed questions to be asked.
joey (tx)


so youre telling me that the Thunder are better off without Durant? Ok youre system just lost all credibility.

John Hollinger (3:36 PM)


Stop sticking words in my mouth. I told you what the facts are -- the team has played much, much better with him off the court. It's not exactly a giant feather in his cap, no? But because the on-court v. off-court stuff is still so new, we're still trying to figure out exactly what it means.
Click to expand...

Yeah, we really missed out, especially with Oden looking good :ohno:
 
The Thunder went on a nice winning streak when Durant was out last year. It could just be that his teammates step up when he's not there instead of leaning on him.
 
Very interesting that Hollinger would admit to being confused by new stats.
 
Last edited:
The history of the NBA is littered with great scorers who never won a ring. Unless he's traded, or OKC gets a dominant rebounder/defender to pair with him, Durant will be another.

I said it when we won the lottery, and I still believe it as much today as I did then. Defense and rebounding wins championships. Oden gives you both. Durant gives you neither.

BNM
 
Minstrel said:
A good article on the disconnect between Durant's terrible +/- and the observational data and production-based data (like PER).
Click to expand...

I love Wayne Winston more the more I read of him:

Knowing that just about any NBA general manager would trade his own children for a prospect of Durant's caliber, I asked Winston if he'd advise his team to accept if the Mavericks were (in some alternate universe) offered Durant for free. "I'd say probably not," he replied. "I would not sign the guy. It's simply not inevitable that he'll make mid-career strides. Some guys do. But many don't, and he'd have to improve a lot to help a team."
Click to expand...

Here's my latest analogy for Durant: a 6'10" Kevin Martin.
 
Winston is one of those guys that thinks Dirk is better than Shaq in his prime. He's one of the worst stat people out there.

The more I read about him the more pointless it is, even Hollinger doesn't buy into this stat completely.

I'd rather go to 82games.
 
Boob-No-More said:
The history of the NBA is littered with great scorers who never won a ring. Unless he's traded, or OKC gets a dominant rebounder/defender to pair with him, Durant will be another.

I said it when we won the lottery, and I still believe it as much today as I did then. Defense and rebounding wins championships. Oden gives you both. Durant gives you neither.

BNM
Click to expand...

Reminds me of a debate several years ago on a baseball site. The author was arguing that Boggs was a "great" player and a HOFer, while Tony Gwynn was not. His proof? Boggs coat-tailed his way to a ring, and Gwynn didn't. You could make the same argument to say that Clyde was a greater player than Stockton.

Great players sometimes don't win rings - crappy players do "win" rings. Titles only measure *team* success...they tell us very little about most individual players.

(and yes, I care more about the titles - I would trade every individual accolade ever won by a Blazer player for a trophy)
 
Rastapopoulos said:
Here's my latest analogy for Durant: a 6'10" Kevin Martin.
Click to expand...

I can see it so far in his career. However, it is clear that he is young enough that there is a good chance he will become more than that.

I think however that if Durant ever had the good luck of being drafted by Portland - Nate would have made him work hard to overcome his issues and play right and while some of our forum would pull their hair out how stupid he (Nate) is - it probably would have made him a much better player...

Somewhat related - why the Thunder (and Memphis) are not going to make the playoffs this year:

http://www.blueblitz.net/2009100924...the-playoffsand-neither-will-the-thunder.html
 
RR7 said:
That was way too many words wasted in an article about why the Thunder and Grizzlies aren't going to make the playoffs.
Click to expand...

Indeed. A couple of thousand words to say "They won't make the playoffs because they both suck."
 
My sig says it all.
 
It seems to me that people are confusing causation with correlation with these numbers.

You can't assume these other players play worse when Durant is on the floor because of Durant. That would be causation.

The correlation here says that when Kevin Durant is on the floor with that select combinations of a very small subset of all basketball players on a particular team, those combinations of players on that team tend to be play poorer then they do when Durant is not on the floor. They don't say anything else. Not only that, they don't give you any reasons for this correlation. There could be any number of factors that account for this that having nothing at all to do with Durant. Maybe the coach's playcalling is different with him off the floor. Maybe other teams play better defense when he is on the floor as opposed to off.

There is a reason for the phrase Lies, Damn Lies and Statistics.

Even that Winston guy admitted his stuff should only be used as a guideline and not gospel.

The world is full of nonsensical correlations. IIRC the book Freakanomics is full of them.
 
LittleAlex said:
It seems to me that people are confusing causation with correlation with these numbers.

You can't assume these other players play worse when Durant is on the floor because of Durant. That would be causation.

The correlation here says that when Kevin Durant is on the floor with that select combinations of a very small subset of all basketball players on a particular team, those combinations of players on that team tend to be play poorer then they do when Durant is not on the floor. They don't say anything else. Not only that, they don't give you any reasons for this correlation. There could be any number of factors that account for this that having nothing at all to do with Durant. Maybe the coach's playcalling is different with him off the floor. Maybe other teams play better defense when he is on the floor as opposed to off.

There is a reason for the phrase Lies, Damn Lies and Statistics.

Even that Winston guy admitted his stuff should only be used as a guideline and not gospel.

The world is full of nonsensical correlations. IIRC the book Freakanomics is full of them.
Click to expand...

It could be, and it also very easily could be that other players are not able to get into a flow and rhythm with him on the court. That other players stand around and watch when he is on the floor, and that they are not allowed to play to the top of their abilities because of him.
 
LittleAlex said:
It seems to me that people are confusing causation with correlation with these numbers.

You can't assume these other players play worse when Durant is on the floor because of Durant. That would be causation.

The correlation here says that when Kevin Durant is on the floor with that select combinations of a very small subset of all basketball players on a particular team, those combinations of players on that team tend to be play poorer then they do when Durant is not on the floor. They don't say anything else. Not only that, they don't give you any reasons for this correlation. There could be any number of factors that account for this that having nothing at all to do with Durant. Maybe the coach's playcalling is different with him off the floor. Maybe other teams play better defense when he is on the floor as opposed to off.

There is a reason for the phrase Lies, Damn Lies and Statistics.

Even that Winston guy admitted his stuff should only be used as a guideline and not gospel.

The world is full of nonsensical correlations. IIRC the book Freakanomics is full of them.
Click to expand...

But is this nonsensical? That's the question. It's definitely a correlation but, at a trivial level, everything is. Even something as obvious as a person throwing a ball and the ball moving through the air is a correlation...it's just a correlation with such a predictable result that we feel comfortable saying that the throw caused the ball sailing through the air.

+/- (or, rather, the superior Adjusted +/-) is a lot more abstract, so it's not clear what the causation is. But that doesn't mean the data is valueless. The question, really, is how predictive it is. Does this tell us anything about how teams will perform in the future with Durant (or any other player, like Battier--an adjusted +/- superstar)?

That's hard to say and there's plenty of controversy. The downside of it is that it's an indirect measure of the player, rather than a direct measure like ones that use tangible production (points, rebounds, assists, etc). The upside is that it takes into account defense, any intangibles...anything that impacts the score should be affecting +/-.

Like anything else, it's a good idea to be skeptical but not dismiss it. I think it's an interesting measure to look at alongside everything else. When one doesn't know the full value of a measure, it's best to view it as piece of a the uncertain task of evaluating players...not the end-all/be-all or completely useless.
 
LittleAlex said:
It seems to me that people are confusing causation with correlation with these numbers.
Click to expand...

I can see this argument after one year. But when it happens for two years in a row... the red flags are there.

I guess that we will have a much better idea if it happens for the 3rd year in a row... - but it could be argued that in a way it already happened in college - the LMA Texas team (year before) and the DJ Texas team (year after) were both better than the Durant Texas team..
 
Durant came out of the game when they were losing by 20 and scrubs from both teams played even quite a bit I would guess
 
Minstrel said:
But is this nonsensical? That's the question. It's definitely a correlation but, at a trivial level, everything is. Even something as obvious as a person throwing a ball and the ball moving through the air is a correlation...it's just a correlation with such a predictable result that we feel comfortable saying that the throw caused the ball sailing through the air.

+/- (or, rather, the superior Adjusted +/-) is a lot more abstract, so it's not clear what the causation is. But that doesn't mean the data is valueless. The question, really, is how predictive it is. Does this tell us anything about how teams will perform in the future with Durant (or any other player, like Battier--an adjusted +/- superstar)?

That's hard to say and there's plenty of controversy. The downside of it is that it's an indirect measure of the player, rather than a direct measure like ones that use tangible production (points, rebounds, assists, etc). The upside is that it takes into account defense, any intangibles...anything that impacts the score should be affecting +/-.

Like anything else, it's a good idea to be skeptical but not dismiss it. I think it's an interesting measure to look at alongside everything else. When one doesn't know the full value of a measure, it's best to view it as piece of a the uncertain task of evaluating players...not the end-all/be-all or completely useless.
Click to expand...

I didn't say to discount it entirely. What I was commenting on was the idea that this result shows conclusively that Durant is a fraud or overrated. It does nothing of the sort. The guy who came to this result gives the impression that he does believe his model is the end all be all of player evaluation. In this he is flat out wrong.


The ball being thrown is a bad example. We know through observable data that if a person throws a ball it will move through the air. That is causation. The thrower caused the ball to move through the air. As long as we can observe the entire event and all it's parameters we can be sure the throw caused the ball to move through the air. Causation also pretty much ensures correlation so of course there is a 100% correlation between throwing a ball and that ball moving through space.

What we have here are two complex mathimatical models that are producing incompatable results. Sure the models attempt to abstract things, but they are still only models.
 
There are some players who don't seem to be able to make the players around them any better but are very good in their own right. Durant may be one... Iverson, Outlaw, Zach Randolph etc may be others. Basketball is a team game... and while all those players excel indivdually, I don't think they add a whole lot otherwise. If you play the right kind of team basketball it doesn't matter if one player can go 1 on 5 and create their own shot with great success.

The current Blazers seem to be a great example of a good team that plays good team basketball. I would say Outlaw and Bayless are the two that are struggling most to make others better.
 
LittleAlex said:
The ball being thrown is a bad example. We know through observable data that if a person throws a ball it will move through the air. That is causation. The thrower caused the ball to move through the air. As long as we can observe the entire event and all it's parameters we can be sure the throw caused the ball to move through the air. Causation also pretty much ensures correlation so of course there is a 100% correlation between throwing a ball and that ball moving through space.
Click to expand...

It's not a bad example, it's a philosophical point. All "causation" is actually just correlation that has such a high level of predictive value that we feel secure in saying A causes B. It's theoretically possible that the thrower does not cause the ball to fly through the air...that the throwing motion and the ball moving are unrelated but just happen to occur together every time we observe it. (Theoretically possible but, of course, vanishingly unlikely.)

The reason I brought it up is that there is no objective point at which correlation turns into causation. All we have is more predictive correlations and less predictive correlations. When we think we can explain a highly predictive correlation, we call it causation.

So saying that +/- or Adjusted +/- is "just a correlation" isn't that useful an analysis. Everything is just a correlation. The question is how predictive it is (ideally, we'd also like an expanation for why it is predictive...but even without that, high predictive value is still the key).

What we have here are two complex mathimatical models that are producing incompatable results. Sure the models attempt to abstract things, but they are still only models.
Click to expand...

True, but so is all of human knowledge. We know no perfect, foundational truths...we only have models for how the world works. And we keep looking for better models.
 
Boob-No-More said:
The history of the NBA is littered with great scorers who never won a ring. Unless he's traded, or OKC gets a dominant rebounder/defender to pair with him, Durant will be another.

I said it when we won the lottery, and I still believe it as much today as I did then. Defense and rebounding wins championships. Oden gives you both. Durant gives you neither.

BNM
Click to expand...

...HONK ONCE!!! :cheers:
 
Paxil said:
There are some players who don't seem to be able to make the players around them any better but are very good in their own right. Durant may be one... Iverson, Outlaw, Zach Randolph etc may be others. Basketball is a team game... and while all those players excel indivdually, I don't think they add a whole lot otherwise. If you play the right kind of team basketball it doesn't matter if one player can go 1 on 5 and create their own shot with great success.

The current Blazers seem to be a great example of a good team that plays good team basketball. I would say Outlaw and Bayless are the two that are struggling most to make others better.
Click to expand...

...precisely why TO must go, just like zBO!!! [throw in J.B. if you have to]
 
Sam Smith's take on Durant:

http://www.nba.com/bulls/news/smith_090727.html

Durant as good--or better?!?!--than LeBron?

-- It’s being whispered now among NBA types because, I believe, under a secret NBA/ESPN provision you are not permitted to say aloud anything that might be interpreted to suggest LeBron James isn’t the best thing to ever happen in anyone’s life. But sitting in the stands in Las Vegas at the USA Basketball mini-camp and hearing NBA coaches and general managers, the gasps have been for Oklahoma City’s Kevin Durant. I’ve now heard more than one say Durant could be a better player than James given Durant’s combination of amazing size at about 6-10 and pure shooting stroke. And he’s still 20. The feeling is Durant is on the brink of being a 30 per game scorer. Plus, Durant is a serious worker. The Bulls' Derrick Rose has become close with Durant and the two have worked out together considerably this summer. Said Rose, who is five days younger than Durant: “KD is great. He’s real talented. I look at him as like a big brother to me even though we’re the same age. He’s been in the league longer. He’s a great talent, 6-10 guy who can dribble, shoot like a guard. He’s going to be a great talent in the league.”
 
Bill Simmons' take on Durant:

http://sports.espn.go.com/espn/page2...simmons/090212

Group C: "It Makes Us Angry That You'd Even Ask"


8. Kevin Durant
Put it this way: I had a lot to do this week. I am trying to finish a book. I had to finish this column. My wife and daughter returned from a trip late Monday night and I wanted to see them. I am flying to Phoenix on Thursday and had to take care of all the dumb stuff people have when they are about to leave for a trip. On Tuesday night, under normal circumstances, I would have rather given myself a two-foot-long paper cut instead of going to the Staples Center and spending three hours with Lakers fans. I went there anyway. Only three teams would have dragged me out of the house: The LeBrons, the Celts, and Durant's Future Former Team. That's the list.

(Before you make fun of my bromance with KD, explain how a kid who just turned 20 four months ago has basically been averaging a 29-8 with 49-88-45 percentages and this isn't a national story. I mean, you should be tired of hearing about Durant by now. By the way, he's averaging 31.9 ppg, 8.1 rpg and 45.4 mpg over the past 12 games, six of them wins.

Here's his game log if you don't believe me. And check those numbers compared to a 21-year-old LeBron in Year 2, or Kobe's third season when he turned 20. Hmmmm. What do you think Durant has to do to crack one of the first three segments on "PTI"? Score 40 for five straight games? I'm just curious. Again, he's 20. He's can't drink yet. This whole paragraph is making me feel bromantic.)
 
There is also the obvious possibility that OKC has a poorly contructed roster, and that until PJ was fired, they were poorly coached. Playing 2 SFs (Durant and Green) and a revolving door at center just might not be the best way to win games. When Durant is out, maybe they are forced to discard "Nellie ball" and play a more conventional game. Until we see what Durant can do playing next to a real PF and a decent center, much of the bashing is premature. I'm betting that if Durant got to play alongside someone like Joel, a lot of this debate would suddenly sound silly.
 
I'm not saying that Durant is a bad player, i'm just saying that he is getting pretty overrated with the talk of him eventually being as good as Lebron.
 
santeesioux said:
I'm not saying that Durant is a bad player, i'm just saying that he is getting pretty overrated with the talk of him eventually being as good as Lebron.
Click to expand...

They talk about LeBron being better at a young age than Jordan was, and use the stats to show it's a reasonable argument.

Durant's stats are up there with LeBron's, but he's a year younger than LeBron was (in their 2nd seasons).

FWIW, I pegged Durant to be a 6'10" Kobe like player before he is through. This was at the time of the draft. I remember having an AIM chat with gambitnut around the time of the draft and I suggested that Portland draft Durant and play him at SF, keep ZBo as a backup or starting PF, and play LMA at C. I also remember how the situation smelled like Bowie/Jordan all over again. Though I did peg Oden to be potentially a Patrick Ewing type.
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 1 (members: 0, guests: 1)
Back
Top