Talk Me Down: How is Deng not MVP of the Bulls?

[Doctor MJ]

Great stuff alucryts, and worth noting that what you describe makes more sense with how I've traditionally though of basketball than a defense where Deng is more of a focal point.

[alucryts]

Yeah, it doesn't make much sense for someone like that to be a focal point. I could plan an offense that completely removes a SF if they were the backbone of a team. You cannot remove a power forward or center from the focal point of a great defense unless you have some exotic ass talent. A reason why Deng gets a bulk of the credit is that he primarily plays the big scorers on the other team in this dominating defense and plays them well. Rose's role on defense is more about shutting down team offense than explicitly stopping scoring. If he diverts ball handlers out of passing/driving lanes, 90% of plays are broken and the other team reverts to an isolation. Our defense now has a heavily reduced shot clock with an offense playing into our strength in isolation defense.

[lorak]

Well but also Deng didn't look that great for three years previous to 2010 either. And most Bulls fans think he is essentially the same player (outside a bit of confidence and the threeball I mentioned). His base production really hasn't changed at all, but for up here, down there. His offense in specific has been kind of dreadful this year.

So I'm trying to get at what Deng is doing so amazingly different that he would become so much more valuable than he was in 08, 09, 10.

He plays better defense, because probably Thibo corrected his flaws.
Confidence level also is important and I think Eurobasket helped him a lot. He really looked like LeBron on that Great Britian team.

[alucryts]

Well but also Deng didn't look that great for three years previous to 2010 either. And most Bulls fans think he is essentially the same player (outside a bit of confidence and the threeball I mentioned). His base production really hasn't changed at all, but for up here, down there. His offense in specific has been kind of dreadful this year.

So I'm trying to get at what Deng is doing so amazingly different that he would become so much more valuable than he was in 08, 09, 10.

He plays better defense, because probably Thibo corrected his flaws.
Confidence level also is important and I think Eurobasket helped him a lot. He really looked like LeBron on that Great Britian team.

Thibs didn't correct his flaws as much as he put his talents into a system that makes them SHINE. Imagine Deng with no big man help. He has to guard the pass, shot, and drive. In Thibs system, he can DOMINATE the pass and shot while bothering the drive. His expertise is magnified by the system.

[lorak]

Well but also Deng didn't look that great for three years previous to 2010 either. And most Bulls fans think he is essentially the same player (outside a bit of confidence and the threeball I mentioned). His base production really hasn't changed at all, but for up here, down there. His offense in specific has been kind of dreadful this year.

So I'm trying to get at what Deng is doing so amazingly different that he would become so much more valuable than he was in 08, 09, 10.

He plays better defense, because probably Thibo corrected his flaws.
Confidence level also is important and I think Eurobasket helped him a lot. He really looked like LeBron on that Great Britian team.

Thibs didn't correct his flaws as much as he put his talents into a system that makes them SHINE.

Whatever ;] Thibs helped Deng and that's why he looks better this and last season. But that doesn't change the fact that Deng is more valuable for Bulls than Rose, beacuse of his huge role in Thibs defensive system.

[mysticbb]

Everyone's is a 'personal' belief, your personal belief is just in the infallibility of RAPM as a predictor, at least relative to anything else, while I prefer a more wholesome approach using all metrics we have to evaluate players, including firsthand observations, and understanding that players do not have static values.

You are talking a strawman here. I know that is typical in order to make yourself feel superior in some way, but the reality is, that you are not superior at all. RAPM is a better predictor than any other statistical tool at hands. And no, I don't base my statement alone on that. As I pointed out within the given data set we can see consistent results which are in agreement with RAPM, the reality is not in agreement with your argumentation.

As I said, you are trying to rationalize your personal belief, but I already pointed out a lot of flaws in your idea (like Watson was just hot, Keith Bogans was responsible for the lower +/- numbers, etc.). None of that was refuted by you in any meaningful way. The data also does not show that Deng's performance level depends on Rose much more than vice versa. So, in essence, Deng makes it possible for Rose to play like he does is as true as the other way around.

You say the Bonner example isn’t congruous because of his minutes. And I’m saying that it is similar because you are telling me snippets of Deng playing a few minutes without Rose per game, are going to tell us how the team does at large for 48 minutes every game without him, including closing every tight game the team gets into. It’s not remotely the same circumstances to draw from. That the team choose to purposely rest Rose probably longer than they needed to because they knew they could run over about 5-7 poor teams isn’t really proving much here.

How can you even believe that the situation for Bonner is in any way similar? Bonner is used in specific matchups, in which he provides an advantage. Thus, it is expected that he will have better +/- numbers. Bonner is not used much when the situation is not there. Bonner has a higher standard deviation of his minutes than Luol deng despite the fact that he plays 18 minutes less. Deng is used in any situation, not just in specific situations in which he clearly has an advantage. Deng is much more versatile and showed that he can impact the game in bigger minutes at a high level, Bonner on the other side did NOT show that all. Ignoring that aspect is either trolling or completely not understanding the context.

The Bulls without Rose performed at the expected level based on statistical analysis. The Bulls without Deng also performed at that expected level. And I'm talking about scoring margin adjusted for the strength of schedule. And again, that does show that Rose has a high impact, just that Deng has a higher impact.

You mentioned Vegas, that’s interesting, be curious to see the source.

http://scores.goldsheet.com/merge/tsnfo ... slist.aspx

I took the spreads, ran a ridge regression only on the games with Rose and Deng included, then with Rose and without Deng in the sample and then with Deng and without Rose. The intercept was 3, which is the HCA.

Last I read about this, none of the stat based models are beating Vegas yet.

In this season so far I have model that has beaten the spread in 57% of the games. A model based on ridge regression of the results + ridge regression of the Vegas spreads is at about 60%. So, Vegas can be beaten by statistical models. The issue is that Vegas is much closer in terms of knowing which player will be in or out, the most people using stats models aren't including adjustments for injuries for specific games. Thus, they have a serious disadvantage.

But historically, simply using the last season records of most teams, and repeating them as a prediction, ends up beating most stat based predictions for team wins.

That is false. The previous season scoring margin used as a predictor gives a RMSE (root mean square error) of 6.32 for the last 11 seasons. Using my boxscore based model for the same test gives 2.75 as the RMSE. Metrics like ASPM or RAPM are in the same region as my model and clearly better than a prediction based on the scoring margin from the previous season.

So, the rest of your post is based on the false idea.

[Rerisen]

As I said, you are trying to rationalize your personal belief, but I already pointed out a lot of flaws in your idea (like Watson was just hot, Keith Bogans was responsible for the lower +/- numbers, etc.). None of that was refuted by you in any meaningful way.

I don't have a big problem with your Bogans adjustment, even though I still distrust his rating based on his ridiculous high minutes last year with the same few players. Watson... the problem there is more the sample size.

However, none of what I wrote about clutch time play (where Deng has faded historically - and that we have no strong data for him without Rose) and the inability of our sample data to actually consist of enough of the the circumstances that we need to know; what the Bulls look like in full games over a long period without Rose, and against more than poor teams, was refuted by you. It can't be. Because it's true, we really don't have minutes from those circumstances in any meaningful amount.

We have a bunch of small minute instances of Deng playing without Rose, with backups against other backups. And then 10 games this year. 10 games which is a laughably small sample size for any serious +/- endeavor.

Your personal belief is that RAPM can nonetheless and regardless of these differing circumstances, predict the future better in any specific case, than any other informed opinion. That's fine, and I don't begrudge you your faith in these numbers, but if that is all you have to espouse over and over, then why even bother engaging in a discussion? You are just telling me what these specific numbers say would happen, but you have to know that is not really what would happen down to the exact point.

You had no counter to my example of LeBron's impact falling greatly in Miami, other than to agree with what I already know, that his circumstances changed greatly.

But that only reinforces my point. Without Derrick Rose on the Bulls for a full year – not 10 minutes a game when he rests vs backups, and Deng plays a couple of those – a similar massive transformation would happen to the Bulls - and to Deng's own individual role. I think very likely creating an opposite of effect of what happened to LeBron. Instead of redundancy of skills, such as in Miami, the Bulls would have a dire drought of skills, of ball handling and shot creation, such that everyone’s impact would reduce on bad fit, especially against stiff competition.

Why you think Luol Deng's impact would remain the same in this vastly different circumstance is baffling. Deng has only played at this level even by RAPM for *2 years*, while the 3 years prior to that he was rated significantly worse, while being virtually the same player in every other way commonly measured! It's not like he is a 10 year superstar with flatline mega impact throughout his career. We've already seen great variation just in his last 4 years.

Your response to why I don't buy this data as a good predictor (in this specific example, not RAPM in general; we are hardly talking swapping bit player around) is to accuse me of trolling.... because someone doesn't share your opinion! Nice.

That is false. The previous season scoring margin used as a predictor gives a RMSE (root mean square error) of 6.32 for the last 11 seasons. Using my boxscore based model for the same test gives 2.75 as the RMSE. Metrics like ASPM or RAPM are in the same region as my model and clearly better than a prediction based on the scoring margin from the previous season.

That's great if you have a system that can make money, I hope you are using it. But my comment was based on ESPN's blog last year which tracked dozens of predictions and systems, and only Hollinger and a hybrid formula beat Vegas.

http://espn.go.com/blog/truehoop/post/_ ... beat-vegas

....

We can go on repeating the same things, but I think the main points have been made for why we see things the way we do. And actually this reminds me of our last discussion, I think about Hinrich's value. He's since fallen off a cliff, that RAPM probably also didn't see coming, but well....

I would rather move in a productive direction, and ask you the question posted at the top of the page. Do you have Luol Deng as a top 10 player in the NBA? He has been by RAPM the last two years. I would be afraid/fascinated to see your top 20, but won't ask for that. I suspect based on what you have espoused in this thread it would have to be pretty much RAPM by order, with perhaps some lower minute guys kicked out.

The MVP and SPM ratings in your sig link respect Derrick quite a bit more. You must put some merit in other metrics if you are pointing people at those numbers.

[mysticbb]

I don't have a big problem with your Bogans adjustment, even though I still distrust his rating based on his ridiculous high minutes last year with the same few players.

For the adjustment I used -15.3 per 36 minutes, that is really low. And it should be clear, if you think Rose is effected negatively so is Deng.

However, none of what I wrote about clutch time play (where Deng has faded historically - and that we have no strong data for him without Rose) and the inability of our sample data to actually consist of enough of the the circumstances that we need to know; what the Bulls look like in full games over a long period without Rose, and against more than poor teams, was refuted by you. It can't be. Because it's true, we really don't have minutes from those circumstances in any meaningful amount.

I agreed with that view. ;)

The worst teams in terms of closing out close games are still winning about 20% of those close games. That makes around 4 wins less than expected. It is not that much of a deal. The majority of the games is decided earlier and not in crunch time anyway.

Your personal belief is that RAPM can nonetheless and regardless of these differing circumstances, predict the future better in any specific case, than any other informed opinion.

You are talking nonsense here, because that is not what I said. RAPM is just in agreement with the data set we have; it is in agreement with the reality, and thus it can give us a pretty good impression about that. And that is NOT completely independent of the circumstances, in fact I had a big discussion with Wayne Winston about that as he declared Deng the MVP of the league based on his WinVal rating (APM derivate). I told him that the numbers are influenced by the circumstances and I would NEVER say that this is not true. So, don't make up stuff. The context is important.
The issue I have with your explantion is that the contextual changes should have been seen in the data. But all we see is the predicted drop in the performance level of the Bulls. And the drop is big, the Bulls are going from a contender (being pretty much the best team in the league with about 67 wins) to just being an expected 50 win team (add the 4 less wins due to bad performance in close games, if you like) without Rose. How is that not a big enough drop? What do you expect? Explain, what kind of performance level drop would be justified by your personal belief here?
With Rose and without Deng it would be 46 wins expected. So, if we add those expectations for clutch I introduced, we pretty much end up with a tie. So, please, how far are we apart here?

You had no counter to my example of LeBron's impact falling greatly in Miami, other than to agree with what I already know, that his circumstances changed greatly.

Yeah, and instead of taking the next logical step and assume that I don't look at that without context, you proceded and claimed I would ignore context. ;)

But that only reinforces my point. Without Derrick Rose on the Bulls for a full year – not 10 minutes a game when he rests vs backups, and Deng plays a couple of those – a similar massive transformation would happen to the Bulls - and to Deng's own individual role. I think very likely creating an opposite of effect of what happened to LeBron. Instead of redundancy of skills, such as in Miami, the Bulls would have a dire drought of skills, of ball handling and shot creation, such that everyone’s impact would reduce on bad fit, especially against stiff competition.

The thing is, it didn't happen in those 10 games without Rose. What do you expect? That the Bulls will just play against better than average teams all of the sudden without Rose? It makes no sense. The Bulls will play the majority of their games against worse than average teams, the same like they are doing with Rose. The Bulls will still play the great teams as often and will very likely lose the most of those games without Rose. That's what the results are telling us, that is what RAPM is telling us and that is what your personal belief is telling you too. So, where exactly is the problem? Do you expect the Bulls to play a much stronger schedule just because Rose isn't there?

Why you think Luol Deng's impact would remain the same in this vastly different circumstance is baffling. Deng has only played at this level even by RAPM for *2 years*, while the 3 years prior to that he was rated significantly worse, while being virtually the same player in every other way commonly measured! It's not like he is a 10 year superstar with flatline mega impact throughout his career. We've already seen great variation just in his last 4 years.

Well, and I'm really surprised that you as a Bulls fans aren't able to remember those specific seasons in which Deng's numbers aren't good. Can you imagine that Deng's performance level might be effected by injury problems? I would assume that the performance level is not independent and is effected by injuries. And yeah, Deng in 2007, a year without injuries is basically as good as the last 1.5 seasons. Funny thing is that Deng's RAPM correlates greatly with the amount of games played by him. ;)

That's great if you have a system that can make money, I hope you are using it.

Actually, I haven't bet on anything. It is the first season in which I tested that at all. But maybe I should start. You need a system being able to beat Vegas in at least 54% of the games in order to make money. So, it is pretty hard overall.

But my comment was based on ESPN's blog last year which tracked dozens of predictions and systems, and only Hollinger and a hybrid formula beat Vegas.

http://espn.go.com/blog/truehoop/post/_ ... beat-vegas

Those are seasons based predictions and the biggest issue for most is predicting the minutes distribution. Also, Hollinger is better than most at predicting the playing level of the rookies, because he can use the ESPN database for the college stats. Most other people don't have the luxury of that.

This season so far Hollinger again beats Vegas, but so does RAPM and 4 other predictions (listed on APBR board after half the season was played out). Vegas is pretty good at that, but I also doubt that there are just using the "eye test". ;)

We can go on repeating the same things, but I think the main points have been made for why we see things the way we do. And actually this reminds me of our last discussion, I think about Hinrich's value. He's since fallen off a cliff, that RAPM probably also didn't see coming, but well....

Hinrich fall off the cliff after his latest injuries. So, if you want to say that RAPM can't predict injuries, I'm with you on this one.

The MVP and SPM ratings in your sig link respect Derrick quite a bit more. :wink: You must put some merit in other metrics if you are pointing people at those numbers.

Those are boxscore based numbers, it should be clear who is winning the boxscore battle between Rose and Deng. Deng's strength is not captured well by the boxscore numbers, but better by non-boxscore based models. Overall the best rating would be the combination of both. A regression based combination (offense comes out as equally weighted, defense by RAPM is weighted twice as much as by SPM) would give the following Top20 per 100 possessions:

Code: Select all

Name                 Team  Min   Off   Def   Off+Def
LeBron James         MIA     1138   5.4   2.1   7.5
Chris Paul           LAC      902   4.7   0.8   5.5
Dirk Nowitzki        DAL      944   2.9   2.4   5.2
Dwight Howard        ORL     1266   1.7   3.0   4.8
Kevin Durant         OKC     1211   4.1   0.3   4.5
Steve Nash           PHO      960   4.5  -0.2   4.4
Luol Deng            CHI     1036   1.2   3.2   4.3
Dwyane Wade          MIA      740   4.0   0.3   4.3
Kevin Garnett        BOS      855   1.3   2.9   4.3
LaMarcus Aldridge    POR     1122   2.1   2.1   4.2
Tim Duncan           SAS      871   1.4   2.7   4.1
Derrick Rose         CHI      852   3.1   0.9   4.0
Kyle Lowry           HOU     1098   2.0   1.8   3.7
Paul Millsap         UTA      962   2.8   0.9   3.7
Kobe Bryant          LAL     1221   3.8  -0.1   3.7
Kevin Love           MIN     1241   3.2   0.3   3.6
Vince Carter         DAL      679   1.4   2.1   3.5
Ryan Anderson        ORL     1002   2.4   1.0   3.4
Josh Smith           ATL     1126   0.8   2.5   3.3
Chris Bosh           MIA     1129   1.7   1.5   3.2


I would use the minutes played in order to assign an overall value to each player. So, I would place Deng slightly over Rose here. But as I said, for Deng the predictions with RAPM are better than with SPM, while for others it is the other way around.
The retrodiction test with that combination for the last 4 years gave a lower error than for either metric alone. Thus it should be better at predicting overall.

[BradPiff]

No it's

1. Their defense and Rebounding
2. Rose
3. Deng

[alucryts]

Whatever ;] Thibs helped Deng and that's why he looks better this and last season. But that doesn't change the fact that Deng is more valuable for Bulls than Rose, beacuse of his huge role in Thibs defensive system.

I agree that Thibs helped him, but if you read my posts Deng does not have a huge role. His role is on par with Brewer's and Rose's in our defense. If you are willing to say that Deng is more valuable because of his defense, you should also have Deng as you DPOY. In order to affect a defense to a magnitude that makes Deng more valuable than Rose from his position, he would have to be an absolute monster on defense. Also, simply based on our system, you would also have to place Noah over Deng and Rose. The notion that Deng is more valuable than Rose is not correct. I believe the reason that RAPM or whichever stat says it is so is because of a flaw in separating out credit/blame on the defensive end. Deng may be the one contesting the shot, but Noah is the reason he can play in the guys face in the first place.

[mysticbb]

alucryts, if Deng is not so important and can easily be replaced by other players, why do you think Deng plays the most minutes? And why is the Bulls defense with Deng and without Noah better than vice versa (much better)?
I think even with you looking closely at that, you aren't able to really make out the difference for the Bulls defense. When Deng is on the court overall the defensive mistakes are lower, because Deng is helping out much more timely than anybody else on the Bulls. Deng is the guy with least amount of defensive mistakes, and the rate of the mistake is basically as important as scoring efficiency for the team on the other end of the court. A difference, which might be not that obvious, can make a big difference for the overall team result. We are talking about a difference of 0.15 points per possession between the best and the worst defense in the league. To see that kind of difference, you need to look at every second really closely and look at all 5 players at the same time.

So, no, Noah is NOT the most important player to the Bulls defense. With Asik or Gibson at C the Bulls are better defensively than with Noah at C. The results are showing that the Bulls in the 7 games without Deng played 2.8 points worse defensively than average (based on the offensive strength of the opponents), in the games with Deng the Bulls played them 6.3 points better than average. And I know that people like to throw in "sample size" here, but in order to come up with the same defensive strength without Deng, the Bulls would need to play 75 games with 7.1 points better defensively than average in addition to those 7 games. How likely do you think is that? And your defensive MVP played 33 minutes during those 7 games in average. How can you think that Noah is so important, when he does not make that difference, while the absence of Deng had a big effect on the defensive strength? Don't you think Thibodeau plays Deng so many minutes, because he knows that Deng is the catalyst for the defensive strength?

[alucryts]

alucryts, if Deng is not so important and can easily be replaced by other players, why do you think Deng plays the most minutes? And why is the Bulls defense with Deng and without Noah better than vice versa (much better)?
I think even with you looking closely at that, you aren't able to really make out the difference for the Bulls defense. When Deng is on the court overall the defensive mistakes are lower, because Deng is helping out much more timely than anybody else on the Bulls. Deng is the guy with least amount of defensive mistakes, and the rate of the mistake is basically as important as scoring efficiency for the team on the other end of the court. A difference, which might be not that obvious, can make a big difference for the overall team result. We are talking about a difference of 0.15 points per possession between the best and the worst defense in the league. To see that kind of difference, you need to look at every second really closely and look at all 5 players at the same time.

So, no, Noah is NOT the most important player to the Bulls defense. With Asik or Gibson at C the Bulls are better defensively than with Noah at C. The results are showing that the Bulls in the 7 games without Deng played 2.8 points worse defensively than average (based on the offensive strength of the opponents), in the games with Deng the Bulls played them 6.3 points better than average. And I know that people like to throw in "sample size" here, but in order to come up with the same defensive strength without Deng, the Bulls would need to play 75 games with 7.1 points better defensively than average in addition to those 7 games. How likely do you think is that? And your defensive MVP played 33 minutes during those 7 games in average. How can you think that Noah is so important, when he does not make that difference, while the absence of Deng had a big effect on the defensive strength? Don't you think Thibodeau plays Deng so many minutes, because he knows that Deng is the catalyst for the defensive strength?

I already answered your first question in one of my posts. I never said he was easily replaced; that is what you are assuming.

Adding on to my post about the Bull's defense in detail, the spot where Deng is so important is the type of players he can play in ICE. He is the only player on our team that can actively guard the pass and shot of someone like Lebron, Danny Granger, Paul Pierce, etc. No one else on our team has the combination of length and understanding of how to play ICE like Deng.

Our defense is better without Noah and with Deng because of the bench mob. Taj and Omer together do as good a job on defense as Noah does. I think that penetration against the Bulls is more difficult with Noah in the game, but Taj/Omer are better at doing something about the penetration when it happens. On the Bulls, removing Noah/Boozer for the backups is a defensive upgrade because teams are no longer able to move Noah to the weak side and attack Boozer in ICE. While he is improving his play, Boozer still does not match Noah, Taj, or Omer in ICE defense.

Where is Deng helping exactly within the Bull's defensive scheme? His job for most of the time is not to help but to make the job of those who help HIM easier. To say that Deng is the defensive MVP is only something you can get with burying your opinion in stats because within the system he is not the defensive MVP. Noah is the best defensive player on the Bulls roster. Next to Noah is the combination of Taj/Asik. After that comes Deng then Brewer and Rose.

[Concept Coop]

alucryts, if Deng is not so important and can easily be replaced by other players, why do you think Deng plays the most minutes? And why is the Bulls defense with Deng and without Noah better than vice versa (much better)?

Surely you know the answer to this question. Omer is one of the best defensive centers in the NBA - not exactly a fair baseline to establish value over replacement.

And yes, Omer has an argument for being the better defensive player. But, he is such a liablity in other aspects, that he can't stay on the floor for extended minutes. If you're not on the floor, you're not being valuable.

Noah can be - and is - the most important defensive player on the Bulls, while his backup is better, per minute played. Over the span of a game, on a greater scale, the season, Noah has a bigger impact on the Bulls' defensive unit than Asik and Gibson, because he justifies being on the court.

Alucryts' points were accurate: Noah is the most important player in Thibs' defensive system.

[mysticbb]

I already answered your first question in one of my posts. I never said he was easily replaced; that is what you are assuming.

So, what you are saying is that in a specific defensive situation Deng can't be replaced and thus the Bulls defense suffers greatly?

Our defense is better without Noah and with Deng because of the bench mob. Taj and Omer together do as good a job on defense as Noah does. I think that penetration against the Bulls is more difficult with Noah in the game, but Taj/Omer are better at doing something about the penetration when it happens. On the Bulls, removing Noah/Boozer for the backups is a defensive upgrade because teams are no longer able to move Noah to the weak side and attack Boozer in ICE. While he is improving his play, Boozer still does not match Noah, Taj, or Omer in ICE defense.

Indeed, Boozer is the weak link here. Well, we can adjust for that by assuming Noah's worse OnCourt rating is completely caused by the higher minutes he has to play with Boozer in comparison to Deng. After such an adjustment (assuming -10.3 for Boozer on defense alone!) the difference is still 3.5 points (defensive RAPM has 4.1 difference between Noah and Deng). Honestly, even with the biggest possible negative influenced assumed for Boozer, the Bulls with Noah on the court are still 3.5 points per 100 possessions worse than with Deng, while that does not take away the fact that the Bulls defense without Deng is much worse than the defense without Noah.

Where is Deng helping exactly within the Bull's defensive scheme?

Not in a scheme, but in a sense to cover up for defensive mistake by his teammates. That is not so suprising giving his length and footwork while being closer to the ball in many cases. Deng is just in better position, but also seems to recognize those things faster than other players while reacting in a better way.
In addition his lower mistake rate will make it easier for his teammates to play within the defensive scheme, that is also a reason Thibodeau let him play more with different players on the court. It is just easier for most player to play with Deng on the defensive end.

His job for most of the time is not to help but to make the job of those who help HIM easier.

Yes, but that hardly covers all plays. ;)

To say that Deng is the defensive MVP is only something you can get with burying your opinion in stats because within the system he is not the defensive MVP. Noah is the best defensive player on the Bulls roster. Next to Noah is the combination of Taj/Asik. After that comes Deng then Brewer and Rose.

Sorry, alucryts, at one point we have to take the results into account, otherwise we are making them meaningless. You are trying to completely ignore the facts here while trying to rely entirely on your belief that you can completely accurate evaluate defensive value of each player. Don't you think when the results are speaking against you, something might be wrong here with your assessment? Assuming the real results are completely off is a rather weird view.
At one point you have to test your hypothesis in a real environment and predict the real outcome. So, what would be your predictions in terms of value for Noah and Deng for the defense? And why does your value and the reality do not agree with each other?

Btw, you might want to check last season when Noah missed 34 games. How much did the Bulls defense suffer? ;)

Concept Coop, your post is covered by my answer to alucryts already.

[alucryts]

Sorry, alucryts, at one point we have to take the results into account, otherwise we are making them meaningless. You are trying to completely ignore the facts here while trying to rely entirely on your belief that you can completely accurate evaluate defensive value of each player. Don't you think when the results are speaking against you, something might be wrong here with your assessment? Assuming the real results are completely off is a rather weird view.
At one point you have to test your hypothesis in a real environment and predict the real outcome. So, what would be your predictions in terms of value for Noah and Deng for the defense? And why does your value and the reality do not agree with each other?

Btw, you might want to check last season when Noah missed 34 games. How much did the Bulls defense suffer?

Concept Coop, your post is covered by my answer to alucryts already.

This exact same argument can be said from my point of view to you. Also, the Bull's defense will suffer, but it won't suffer greatly. You seem to have an issue with the fact that when a player is removed like Noah, Deng, Asik, Rose, etc that the backup cannot in any magnitude fill in. Kurt Thomas did a great job in our system. He was no Noah, but he was good.

Before this gets lost, what is the standard deviation in the numbers you are posting, and if available, what is the error involved in these stats when put to a prediction test? I don't know how many times I've asked for this in this thread.

Your problem is one of "correlation does not imply causation". These stats can just as easily be saying that Deng is simply a part of great defensive lineups as much as he is causing great defensive lineups. Are you implying that he is causing the great defense, and that he should be DPOY up their with Dwight Howard? I find that most every argument in plus minus is centered around the idea that correlation is causation. This is a huge misuse and abuse of the stat. How do you separate correlation and causation for these stats? You have to take into account other factors besides plus minus. When ANY of these factors are considered, the logic swings to correlation and not causation. Plus minus is the only stat out there that points to causation for Deng.

[Dr Positivity]

MVP is a tricky concept. I don't know if it's as easy as "This guy being on the court boosts their success" the most. It seems to me like if it's accepted that's true for Deng, you can still make a case that the reason that's true is Rose being on the team, and thus Rose is still the straw stirring the drink, or w/e

[mysticbb]

This exact same argument can be said from my point of view to you.

Really? The results are not in agreement with what I'm saying?

Before this gets lost, what is the standard deviation in the numbers you are posting, and if available, what is the error involved in these stats when put to a prediction test? I don't know how many times I've asked for this in this thread.

Standard error for RAPM does not make any sense at all. Ridge regressions adds a bias here and the variance is greatly lowered by the bias. So, while an error could be calculated via bootstrap, it basically can't be interpreted in any meaningful way.
For the games I could give you a standard deviation, which can be compared to a typical standard deviation for a 82 games sample in order to get an impression about the issue of the sample size here. For the 7 games without Deng the standard deviation of the adjusted DRtg was 5.3, for all 82 games last season the standard deviation was 6.3. So, the Bulls in those 7 games without Deng played basically as consistent defensively as during the 82 games from last season.

Your problem is one of "correlation does not imply causation".

You don't say? ;)

These stats can just as easily be saying that Deng is simply a part of great defensive lineups as much as he is causing great defensive lineups.

Yes, both things are possible, and we can seperate that to a degree by looking at the lineup performances.

http://stats-for-the-nba.appspot.com/pm/122.html

Those are the numbers for Deng when playing with the same 4 other teammates as the listed players on the court in comparison the lineup performance with said player instead. So, what do those results tell you?

Are you implying that he is causing the great defense, and that he should be DPOY up their with Dwight Howard?

Yes, I would say that, but Deng also missed games, which has a negative effect on his overall value.

I find that most every argument in plus minus is centered around the idea that correlation is causation.

Yes, you might find that, but be aware of the fact that we can explain why. And I tried to explain that for numerous years now, getting into heated discussion why Deng is worth his salary and not some useless scrub when a lot of people (including many other Bulls fans) disagreed with me.
I would say that we have a case here of cognitive biases, in which many people are trying to rationalize their personal beliefs while dismissing the facts. You specifically are running into a confirmation bias, you are looking specifically to find a reason that Deng is not that important. Seriously, your notion that Noah is the defensive MVP of the Bulls is pretty much hilarious, because the Bulls without Noah didn't suffer at all defensively, in fact they played better defense. So, either the results are wrong or the value you want to apply to Noah is not correct.
The value is also determined by looking at the possibility to replace a certain player. And so far I see the Bulls being able to replace Noah much better than Deng. And that has to be seen in the context that smaller players are usually easier to replace, because there are more smaller players available. That alone should make you questioning your conclusion.

[alucryts]

If stats come to the conclusion that Deng is as defensively as valuable as Dwight Howard (and the only thing separating them is games missed), then you are interpreting the stats incorrectly; it is as simple as that. Deng is a great defensive player, but he is no DPOY.

It is very easy to remove Deng from the defense on offense. If he was truly the cause of the great defense, offensive teams would remove him from the play and go to work on our defense. You can easily remove one perimeter player's impact from a defense. Teams don't remove perimeter players from defense because they are not the cause of great defense; they only help it.

[drza]

Can I just say that I've been applauding and eating popcorn while watching this discussion between allucryts and mysticbb. Really high quality support from both sides, and making for a good exchange of ideas.

[drza]

It is very easy to remove Deng from the defense on offense. If he was truly the cause of the great defense, offensive teams would remove him from the play and go to work on our defense. You can easily remove one perimeter player's impact from a defense. Teams don't remove perimeter players from defense because they are not the cause of great defense; they only help it.

Are you sure this is true? Hasn't your stance been that Deng is the Bull most likely to be covering the LeBron's, the Durant's, the Pierce's, the Grangers, etc. on defense? If that's the case, the only way for an offense to remove Deng is to essentially remove one of their best (if not the best) offensive option from their game plan? And if they did, wouldn't their overall offensive performance likely suffer in that instance?

In other words, haven't you set up a case where either Deng could make a really large defensive impact on his man, or he could make a similar (maybe even bigger) defensive impact by completely removing his opponent from the offense's game plan? Or have I misinterpreted what you mean here?