All-Season Player of the Year Discussion thread

[The-Power]

Korver showed that if a team focuses on him, he's what was many thought. Teams never focused on Green in the same way, so we never saw it.

That's why Green isn't an offensive anchor, nobody denies that. But 1) Green makes most of his money based on defense and it's no surprise that the adjustment by the coaching staff which ultimately killed the Cavs was centered around Green who allowed them to make this particular adjustment in the first place and 2) Green proved that he had enormous impact on a team with an offensive anchor and even though his impact on offense (and to a lesser extent on defense) isn't easily portable doesn't mean it didn't exist, is worth less or can't be reproduced under similar circumstances. That's the whole point. You completely miss it by constantly focussing on how a player would fare in a different environment instead of looking at what's actually there: significant impact.

[Texas Chuck]

I don't think you are completely on an island. I think people simply are talking about two different things. You seem focused on just who is flat the better player regardless of where they happen to be playing. Dr. S seems to be focused on the impact they had in the role they found themselves in.

In a POY type discussion I tend to side with the latter. What did they really do this year. In a more esoteric discussion of simply who is the better player then I'd include more of the former.

Now where I disagree is with the actual impact being assigned to some of these players--Korver moreso than Green for me. I always am careful when a player has been around the league for a long time playing big minutes in a similar role and all of sudden has a massive jump in a stat like RAPM at the same time the team he plays for takes a massive jump. I just can't rush to the conclusion that Korver is somehow responsible for this in a way he never has before.

I think he benefits from a beautiful system with a starting lineup that fits together like a glove and the fact that his particular piece of that puzzle can't be replaced by anyone on the Hawks bench. But that if the Hawks had a decent bench shooter to come in when he sits that Korver would more correctly be seen as a cog in the machine.

[bondom34]

I don't think you are completely on an island. I think people simply are talking about two different things. You seem focused on just who is flat the better player regardless of where they happen to be playing. Dr. S seems to be focused on the impact they had in the role they found themselves in.

In a POY type discussion I tend to side with the latter. What did they really do this year. In a more esoteric discussion of simply who is the better player then I'd include more of the former.

Now where I disagree is with the actual impact being assigned to some of these players--Korver moreso than Green for me. I always am careful when a player has been around the league for a long time playing big minutes in a similar role and all of sudden has a massive jump in a stat like RAPM at the same time the team he plays for takes a massive jump. I just can't rush to the conclusion that Korver is somehow responsible for this in a way he never has before.

I think he benefits from a beautiful system with a starting lineup that fits together like a glove and the fact that his particular piece of that puzzle can't be replaced by anyone on the Hawks bench. But that if the Hawks had a decent bench shooter to come in when he sits that Korver would more correctly be seen as a cog in the machine.

Korver showed that if a team focuses on him, he's what was many thought. Teams never focused on Green in the same way, so we never saw it.

That's why Green isn't an offensive anchor, nobody denies that. But 1) Green makes most of his money based on defense and it's no surprise that the adjustment by the coaching staff which ultimately killed the Cavs was centered around Green who allowed them to make this particular adjustment in the first place and 2) Green proved that he had enormous impact on a team with an offensive anchor and even though his impact on offense (and to a lesser extent on defense) isn't easily portable doesn't mean it didn't exist, is worth less or can't be reproduced under similar circumstances. That's the whole point. You completely miss it by constantly focussing on how a player would fare in a different environment instead of looking at what's actually there: significant impact.

OK, I think this is where I'm going off the rails/just not seeing things the same as everyone. This to me makes more sense, and I see it differently but get where you're coming from.

[Jaivl]

Except Pipp was better across the board other than rebounding in every one of those seasons.

And defense.

Draymond Green played defense this year on the level of Peak Pippen.

You misunderstood me.

[The-Power]

OK, I think this is where I'm going off the rails/just not seeing things the same as everyone. This to me makes more sense, and I see it differently but get where you're coming from.

Glad we clarified that. And for the record, I have absolutely no problem with that you're applying different criteria when you rank players. But we need to make sure that everyone is aware of the applied criteria and one must understand where others are coming from in order to have a proper discussion.

[bondom34]

OK, I think this is where I'm going off the rails/just not seeing things the same as everyone. This to me makes more sense, and I see it differently but get where you're coming from.

Glad we clarified that. And for the record, I have absolutely no problem with that you're applying different criteria when you rank players. But we need to make sure that everyone is aware of the applied criteria and one must understand where others are coming from in order to have a proper discussion.

I can completely respect that, and honestly, in the end, its why I come to the PC board instead of the GB for these discussions. I can take being misunderstood, I can even take being wrong, but the general mudslinging is what I don't like. Here everyone is all class, and very well thought out which is appreciated.

[Doctor MJ]

[

Hmm. Where to start here? I suppose the biggest thing is that the following things seems to be things you believe:

1) You don't have issues favoring offense vs defense.
2) You just see Green & Kawhi as players who clearly are not of top tier value - you actually in this post imply Westbrook's impact is vastly greater than Green's.
3) Those two players finished Top 2 in DPOY voting, but I haven't seen you object so hard to that.

The last of the might have just been something I missed, so feel free to elaborate on that. Short of that though, you're essentially saying "It's not that I have an offensive bias, it's just that I happen to think that the top offensive players right now are vastly more important than the defensive players." Does that not sound odd to you?

Speaking more generally, one thing that seems clear is that whereas I tend to view a team as being of two halves, and having foundations for each side, you tend to view that team in terms of one foundation. So OKC has Westbrook, and so without Durant that means they are building everything they do around Westbrook whether it's a strength of Westbrook's, or a weakness.

Let's switch away from Westbrook because that's a sore spot between us and focus on Curry here. I'll fully grant that Curry is the primary player that the Warriors are built around, and that part of that means trying to find a place of value for him on defense. But the Warriors had the best defense in the league this year, that was at least as important as their offense to their success, and Curry is nothing like the main engine that makes that work. The fact that Curry is the offense's engine thus to me seems entirely insufficient when it comes to talking about the Warriors' success on the other end of the floor, and the ones who are truly responsible to me deserve a ton of credit. What of that do you object?

RE: Kawhi, I don't have the same feeling there. I see him borderline top 10, you can call him either way. Green I don't view nearly as highly as I think one you can build around, the other not.

On to the rest...

Again I'm not saying this about O vs. D, I'm saying this as a "again if you look at things as a whole other than "Well RAPM says this" then Green isn't nearly a top 10 player. So I ran some NBAWowy numbers using Green and Curry:

Both on: 17.4 net rating
Curry off: 6.2
Green off: 5.1

So I'm seeing the co-linearity issue come up here. Again, I'm not saying there's something wrong w/ Green, but he's not this player who you can build around, and I stand by the idea that if Korver had a better situation around him, we'd be doing the same. It seems odd to me that someone could be fooled by a metric, yet still believe this steadfastly in it, as I know how intelligent everyone here is. There's way too much noise to genuinely believe this. I'm saying you move Green and put him on say the Clippers, that he's not even in this discussion. He's not a guy you build around, he's a guy you build with. And again, Leonard isn't that.

Re: Kawhi. Am I confused? Didn't you just say that you don't see Green or Leonard as being up there with Westbrook & Wall? Well if I'm confused, I'm confused. Fine to focus on Green if that's the real point of disagreement.

Re: Not saying about O vs D. Okay, and I'll note that you've still not really given any reason why the guy who is the vastly superior defensive player and has the best defensive regression data around should be seen as suspect in part because he plays with a great shooter, which to me would be a great way you'd actually demonstrate that this wasn't about O vs D oddly enough. Because as of right now, it just seems like you're assuming Green's defensive data has everything to do with Curry's presence, and for the life of me, I can't see why you'd assume that.

Re: "seeing the collinearity come up". Huh? What exactly in the WOWY numbers makes you see collinearity? What would you expect to see if both guys were in fact having huge impact exactly? I'm not saying that collinearity can't be an issue here, but I don't know how you divine that by raw on/off numbers.

I'll add one more thing: Because we're talking about an offensive oriented player and a defensive oriented player, to me it makes sense in looking at all of this to focus on the data split between those two domains. By no means will I claim that I've done exhaustive research here, but what's clearly the case when you look at it is that the offense is doing better when Curry's on the floor than when Green's on the floor...and the defense is doing better when Green's on the floor than when Curry's on the floor. That the two players end up with near identical Raw +/- then is clearly something of a coincidence, as opposed to something cause by such heavy collinearity that there was no way to separate the two.

Re: not someone you build around, he's someone you build with. I understand what that feels like a meaningful statement, but I'm not sure that it is. In practice, you build around a guy because you KNOW what he can do for you and know that it is huge. If a guy later on turns out to have a huge impact you didn't expect, why is that seen as lesser?

I get that you think that in other contexts Green doesn't have the same value, but why precisely do you see this being the case? I fully agree that had the ball bounced differently Green would never have had the impact he had this year, but I don't necessarily see that as a reason to knock the value he's actually having.

I suppose I would ask: What is it about Golden State's setup that let's Green be the ultra-valuable player he is for them that you think would be hard for another team to make happen if they actively saw the importance of making it happen? Can we at least agree that whatever it is that is, it doesn't include Curry's defense?

I'll add one more thing: I think it makes sense to be skeptical of what Green is truly brining to the table defensively because the Warrior defense was already quite good last year when he played so much less. Can he really be the DPOY if the team could do so well without him?

I think it's something really worth talking about, and I don't want people to assume that I've already made up my mind on that. I'm listening.

[bondom34]

The issue I'm seeing w/ colinearity is simple enough, and I'm starting to come around on Green bordering top 10, but its still an issue for me that he's really not a focal point in my mind on the team. The issue is that if you put him and Curry on court, you get huge net ratings for the plus, but remove either one and its a huge drop. They're feeding from each other, and the credit needs to be parsed fairly. It seems people are willing to give Curry a ton of it, so afterward you're saying Green doesn't get nearly as much to me, or you're overattributing things. I keep coming back to the point of removing a guy and replacing him w/ a replacement level player (which was the WB comp's reason). W/ Dray you still get a pretty high level GSW team. W/ some others you see a drop to irrelevance.

But again, I'll just let it go, its none of my business how others vote, and Green had a great season, I just don't view him as a top 10 player in the league as he's not a guy you build around, he's a guy you build with.

[bondom34]

[

Hmm. Where to start here? I suppose the biggest thing is that the following things seems to be things you believe:

1) You don't have issues favoring offense vs defense.
2) You just see Green & Kawhi as players who clearly are not of top tier value - you actually in this post imply Westbrook's impact is vastly greater than Green's.
3) Those two players finished Top 2 in DPOY voting, but I haven't seen you object so hard to that.

The last of the might have just been something I missed, so feel free to elaborate on that. Short of that though, you're essentially saying "It's not that I have an offensive bias, it's just that I happen to think that the top offensive players right now are vastly more important than the defensive players." Does that not sound odd to you?

Speaking more generally, one thing that seems clear is that whereas I tend to view a team as being of two halves, and having foundations for each side, you tend to view that team in terms of one foundation. So OKC has Westbrook, and so without Durant that means they are building everything they do around Westbrook whether it's a strength of Westbrook's, or a weakness.

Let's switch away from Westbrook because that's a sore spot between us and focus on Curry here. I'll fully grant that Curry is the primary player that the Warriors are built around, and that part of that means trying to find a place of value for him on defense. But the Warriors had the best defense in the league this year, that was at least as important as their offense to their success, and Curry is nothing like the main engine that makes that work. The fact that Curry is the offense's engine thus to me seems entirely insufficient when it comes to talking about the Warriors' success on the other end of the floor, and the ones who are truly responsible to me deserve a ton of credit. What of that do you object?

RE: Kawhi, I don't have the same feeling there. I see him borderline top 10, you can call him either way. Green I don't view nearly as highly as I think one you can build around, the other not.

On to the rest...

Again I'm not saying this about O vs. D, I'm saying this as a "again if you look at things as a whole other than "Well RAPM says this" then Green isn't nearly a top 10 player. So I ran some NBAWowy numbers using Green and Curry:

Both on: 17.4 net rating
Curry off: 6.2
Green off: 5.1

So I'm seeing the co-linearity issue come up here. Again, I'm not saying there's something wrong w/ Green, but he's not this player who you can build around, and I stand by the idea that if Korver had a better situation around him, we'd be doing the same. It seems odd to me that someone could be fooled by a metric, yet still believe this steadfastly in it, as I know how intelligent everyone here is. There's way too much noise to genuinely believe this. I'm saying you move Green and put him on say the Clippers, that he's not even in this discussion. He's not a guy you build around, he's a guy you build with. And again, Leonard isn't that.

Re: Kawhi. Am I confused? Didn't you just say that you don't see Green or Leonard as being up there with Westbrook & Wall? Well if I'm confused, I'm confused. Fine to focus on Green if that's the real point of disagreement.

Re: Not saying about O vs D. Okay, and I'll note that you've still not really given any reason why the guy who is the vastly superior defensive player and has the best defensive regression data around should be seen as suspect in part because he plays with a great shooter, which to me would be a great way you'd actually demonstrate that this wasn't about O vs D oddly enough. Because as of right now, it just seems like you're assuming Green's defensive data has everything to do with Curry's presence, and for the life of me, I can't see why you'd assume that.

Re: "seeing the collinearity come up". Huh? What exactly in the WOWY numbers makes you see collinearity? What would you expect to see if both guys were in fact having huge impact exactly? I'm not saying that collinearity can't be an issue here, but I don't know how you divine that by raw on/off numbers.

I'll add one more thing: Because we're talking about an offensive oriented player and a defensive oriented player, to me it makes sense in looking at all of this to focus on the data split between those two domains. By no means will I claim that I've done exhaustive research here, but what's clearly the case when you look at it is that the offense is doing better when Curry's on the floor than when Green's on the floor...and the defense is doing better when Green's on the floor than when Curry's on the floor. That the two players end up with near identical Raw +/- then is clearly something of a coincidence, as opposed to something cause by such heavy collinearity that there was no way to separate the two.

Re: not someone you build around, he's someone you build with. I understand what that feels like a meaningful statement, but I'm not sure that it is. In practice, you build around a guy because you KNOW what he can do for you and know that it is huge. If a guy later on turns out to have a huge impact you didn't expect, why is that seen as lesser?

I get that you think that in other contexts Green doesn't have the same value, but why precisely do you see this being the case? I fully agree that had the ball bounced differently Green would never have had the impact he had this year, but I don't necessarily see that as a reason to knock the value he's actually having.

I suppose I would ask: What is it about Golden State's setup that let's Green be the ultra-valuable player he is for them that you think would be hard for another team to make happen if they actively saw the importance of making it happen? Can we at least agree that whatever it is that is, it doesn't include Curry's defense?

I'll add one more thing: I think it makes sense to be skeptical of what Green is truly brining to the table defensively because the Warrior defense was already quite good last year when he played so much less. Can he really be the DPOY if the team could do so well without him?

I think it's something really worth talking about, and I don't want people to assume that I've already made up my mind on that. I'm listening.

]The issue I'm seeing w/ colinearity is simple enough, and I'm starting to come around on Green bordering top 10, but its still an issue for me that he's really not a focal point in my mind on the team. The issue is that if you put him and Curry on court, you get huge net ratings for the plus, but remove either one and its a huge drop. They're feeding from each other, and the credit needs to be parsed fairly. It seems people are willing to give Curry a ton of it, so afterward you're saying Green doesn't get nearly as much to me, or you're overattributing things. I keep coming back to the point of removing a guy and replacing him w/ a replacement level player (which was the WB comp's reason). W/ Dray you still get a pretty high level GSW team. W/ some others you see a drop to irrelevance.

But again, I'll just let it go, its none of my business how others vote, and Green had a great season, I just don't view him as a top 10 player in the league as he's not a guy you build around, he's a guy you build with.

[Doctor MJ]

This Green talk is interesting. One thing I take issue with is this idea that he was Golden State's secondary playmaker. Sorry, but I'm just not buying that. If we simply look at raw assist totals maybe, but when you look closer he's not more of a playmaker than Livingston or Iggy or even Bogut really. He moves the ball okay for a PF but people are over-reacting to his playoff assist numbers more than anything here. He still turns the ball over a lot.

The reason his FGA attempts and assist numbers are up in the playoffs is much more a reflection of how teams defended GSW and Curry in the playoffs. He found himself with the ball a lot more and not by Kerr's design, but by the design of the opposing coaches. And the reason for that is he's just not that great an offensive player.


Now defensively I agree he is among the handful of guys able to impact the game defensively the most. His strength, mobility, and defensive intelligence allow him to do so many things defensively. But in our rush to love the guy and explain why his RAPM numbers are off the chart, let's not make him into something he's not.

What kind of data would you like to see to convince you otherwise? Clearly you think I"m overreacting to the playoffs, so let me just pull up from the regular season.

# of Passes Per Game:
Curry 56.1
Green 51.6
Livingston 36.1
Iguodala 33.2

Points created by Assists per Game:
Curry 18.2
Green 9.1
Livingsston 7.8
Iguodala 7.1

Now here's what I will say: Obviously when Livingston is on the court he's a higher primacy playmaker than Green. Playmaking is the entire reason why Livingston is in the league and he's more of a natural at it than anyone else on the team including Curry. But in practice, over the course of a typical game, Green is passing the ball more and creating more assist-points than anyone else other than Curry, and he's doing this while playing the 4 and focusing on defense as much as possible. This "playmaking 4" aspect of his play then isn't there by some kind of default. The team has clearly decided it needs playmaking out there and that Green is the one to give it. Quibble over semantics if you want, dude deserves SOME kind of recognition for what he's doing on this front.

[Doctor MJ]

This Green talk is interesting. One thing I take issue with is this idea that he was Golden State's secondary playmaker. Sorry, but I'm just not buying that. If we simply look at raw assist totals maybe, but when you look closer he's not more of a playmaker than Livingston or Iggy or even Bogut really. He moves the ball okay for a PF but people are over-reacting to his playoff assist numbers more than anything here. He still turns the ball over a lot.

The reason his FGA attempts and assist numbers are up in the playoffs is much more a reflection of how teams defended GSW and Curry in the playoffs. He found himself with the ball a lot more and not by Kerr's design, but by the design of the opposing coaches. And the reason for that is he's just not that great an offensive player.


Now defensively I agree he is among the handful of guys able to impact the game defensively the most. His strength, mobility, and defensive intelligence allow him to do so many things defensively. But in our rush to love the guy and explain why his RAPM numbers are off the chart, let's not make him into something he's not.

I missed this in all the conversation as well, yea I'm not buying it either. He's maybe a third/fourth playmaker, I'd put Klay ahead of him as well as Curry/Livingston/Iggy at the least, likely even Barnes.

I really wonder what it is people are seeing here.

Klay and Barnes each make about half the total passes Green does in addition to getting less assists. It's not normal for a power forward to be so heavily involved. What exactly would you call what Green is doing if not playmaking?

[bondom34]

Maybe "shot creating" would be a sort of different way to put it? He's not a guy you give the ball to and say to do something, which is a more common way of "play making" to be thought of, but again, don't know. I'll let this to the remaining voters to not bias anyone.

[Doctor MJ]

I
I'll end this post with a simplified example: Player A has an impact of +5 on any team in the league, regardless of the roster. Player B has an average impact of +3 around the league, but there are five teams in the league in which his impact is +7 because he can be used the way he's most effective there. Who is more valuable in general? Who has to be ranked higher in POY-rankings if Player B happens to play for one of those five teams? Pick your place.

Good post in general, I'll focus on your hypothetical as I think that's your intention:

I think the key question is probably: Are there 5 teams who he can have his peak impact because he depends on extremely rare, hard to get, types of teammates...or are there 5 teams because 25 coaches/organizations either 1) haven't had reason to optimize for a player like him or 2) are too dumb to understand what the player can do?

The former possibility is one in which you have to hold it against the player in question at least in terms of considering how fortunate he is.

The latter possibility is one in which shouldn't be held against the player at all. If a coach decided to turn Shaq into a 3-point shooter, Shaq would suck. That ain't Shaq's fault though.

This is not to say I'm advocating we rank players based on something other than what they actually did this year of course - I think I've been pretty clear you have to focus on what really happened - I'm just saying that while I think it's fine to consider a player's "good luck" in his current situation to a degree, any worldview on this that essentially punishes a guy for being a spearhead is to my mind clearly the wrong way to go. Up to each voter whether they think that's what's happening of course, but I don't understand a rationale that holds it against a guy that in other situations the coach would be too stupid to use that guy properly.

[RebelWithACause]

The issue I'm seeing w/ colinearity is simple enough, and I'm starting to come around on Green bordering top 10, but its still an issue for me that he's really not a focal point in my mind on the team. The issue is that if you put him and Curry on court, you get huge net ratings for the plus, but remove either one and its a huge drop. They're feeding from each other, and the credit needs to be parsed fairly. It seems people are willing to give Curry a ton of it, so afterward you're saying Green doesn't get nearly as much to me, or you're overattributing things. I keep coming back to the point of removing a guy and replacing him w/ a replacement level player (which was the WB comp's reason). W/ Dray you still get a pretty high level GSW team. W/ some others you see a drop to irrelevance.

But again, I'll just let it go, its none of my business how others vote, and Green had a great season, I just don't view him as a top 10 player in the league as he's not a guy you build around, he's a guy you build with.

Collinearity problem is a fine thing to point out Bondom.

Can you tap in on defense at this point, and how you weigh defense / offense?
All this focal point kind of thing gives the impression you focus a lot more on the franchise tag and less on defense.

By this very defintion you bascially erase all great defenders from yearly Top10 discussion, if they are no stars on offense. Russell, B. Wallace, Mutombo etc. Or is it just a problem with horizontal defenders in particular, more perimeter oriented guys that offer less rim protection?

[Doctor MJ]

And also, to the above portability point, I'd argue that one could replace Westbrook w/ a replacement level player and see a much greater drop than that seen w/ Green. That's what I'm saying, and that's why to me he has no spot here. Impact is not equal to RAPM data, nor is player ranking. Nobody's putting Korver in the top 10, though he was talked that way a month ago, why? Because his team lost and people realized that he wasn't portable, and wasn't as impactful as they believed due to it. The same works w/ Green. That doesn't mean they're any less of a player, it just means they aren't really belonging here. Unless of course you believe Green does (which I don't), in which case I expect you would rank Korver similarly, along w/ others previously mentioned.

What I said all regular season about both Korver and Green was essentially: They are definitely having impact right now, let's see how it holds up against playoff-level strategy and intensity.

While we can quibble about context and luck, what we saw in the playoffs was:

A determined defense does seem to be able to to get passed the spacing benefits Korver provides. More effort in the right places seems to be all that's required.

But on Green, we saw nothing of the sort. I was among the very few (evidently) that was concerned about Memphis, because they represented a team with true size. I had similar thoughts about Cleveland as well. In both cases what we saw is that those mismatches caused problems initially, but the team was able to adapt, and that adaptation didn't involve diminishing Green's role by any stretch of the imagination.

We've been talking in this thread about the concept of resiliency, about how a player does when everyone is trying to attack his effectiveness. And while I'll readily agreed that Curry is more proven on this front than Green, the fact is teams have been trying everything and the kitchen sink against Warriors on all fronts, and that absolutely includes dealing with the guy who almost won DPOY on that #1 ranked defense. I find myself asking: What exactly was it people were waiting for Warrior opponents to do against Green that they didn't try to do in the playoffs?

[RebelWithACause]

Maybe "shot creating" would be a sort of different way to put it? He's not a guy you give the ball to and say to do something, which is a more common way of "play making" to be thought of, but again, don't know. I'll let this to the remaining voters to not bias anyone.

Playmaking does not really involve shot creation, I mean it can, but would you call Boris Diaw, who is a great playmaker, a shot creator?

[Doctor MJ]

Then why have the curry pick and rolls involved draymond so much.

Because of who is defending Draymond primarily. They obviously want to involve a big defensively(like every team) so they aren't going to use their perimeter players. The only other choice really is Bogut, but obviously Draymond's ability to shoot from distance makes him the better option plus of course Bogut plays so many fewer minutes.

I'm not saying he's a bad offensive player or a horrible decision maker. But he's not really the playmaker people are making him out to be either. I mean do people honestly watch the Warriors play and think Green is a better playmaker than Iggy or Livingston? Because I do not. And that's why I think much of this talk is a direct result of his inflated assist numbers this post-season.

Now I feel differently from bondom in that I think you can start to make a case that Green had one of the 10 best seasons this year. I probably wouldn't put him quite that high myself, but I think its a very reasonable position. But let's talk about the player he actually is. And playmaking just isn't a big part of that.

So your take is:

"He's not really the 2nd best playmaker on the Warriors, he's simply the guy whose combination of attributes make him the guy that they choose to put the ball in his hands to make a ton of passes.'

Correct?

I don't disagree at all. In terms of raw playmaking ability Livingston is way above Green. But most of the time Livingston is on the bench, and so the team depends on other players like Green to support Curry quite a bit.

[Doctor MJ]

The issue I'm seeing w/ colinearity is simple enough, and I'm starting to come around on Green bordering top 10, but its still an issue for me that he's really not a focal point in my mind on the team. The issue is that if you put him and Curry on court, you get huge net ratings for the plus, but remove either one and its a huge drop. They're feeding from each other, and the credit needs to be parsed fairly. It seems people are willing to give Curry a ton of it, so afterward you're saying Green doesn't get nearly as much to me, or you're overattributing things. I keep coming back to the point of removing a guy and replacing him w/ a replacement level player (which was the WB comp's reason). W/ Dray you still get a pretty high level GSW team. W/ some others you see a drop to irrelevance.

But again, I'll just let it go, its none of my business how others vote, and Green had a great season, I just don't view him as a top 10 player in the league as he's not a guy you build around, he's a guy you build with.

bondom I'm sorry if you feel ganged up on or something like that. I'll add that the recent batch of posts I did, I did replying in chronological order, so if you don't want to address any of this stuff, it's fine.

But here what we're talking about is a technical term you brought up, and to me your use of it seems questionable. Not trying to rub your nose in it, I just think it's worth getting this stuff straight.

Collinearity is caused by guys playing very few minutes a part. As such, the APM/RAPM/etc data has to allocate between them based on a very small sample size.

But when this happens, the issue isn't that both guys get credit, it's that one guy tends to look very good and the other guy very bad. To me the worst-case scenario for this was the '05-06 Pistons with the pure APM (which gets hit worst by this). Interestingly, 82games seems to not have that particular study up any more, but long story short:

Pistons have HUGE raw +/- for their starting 5.
Those guys played big minutes and played together a lot.
Richard Hamilton's APM as a result was wretched.

Why? Because in the few minutes they played apart, the ones with Hamilton did bad, and so the allocation of credit swung away from him in a way that no human could justify.

So yeah, when you're using the similarity of Curry & Green's RAPM data rather than the dissimilarity of that data to talk about collinearity, it's odd to me. Again, not that collinearity can't be an issue here, it's just that I don't see what in the data is making you reach for that particular explanation as the tell-tale signs for me aren't there.

Speaking more generally, as I alluded to in another recent post: I get if you're concerned with giving Green a ton of defensive credit based on the fact that defenses tend to work more as a symbiotic unit, or based on the fact the defense was already good last year when Green played far less. It's not that I fundamentally am certain that your instincts are wrong here, it's just that I literally don't have any evidence to support that intuition at this time. It's not about there being 1 stat supporting Green, it's about there being 0 stats I feel I can point to in the other direction to combat the following things:

1) The Warriors have a great defense.
2) Among the Warriors who play huge minutes (Curry, Thompson, Green), everyone basically agrees it's Green who takes on the biggest burden for the team on defense.
3) The regression data loves it some Draymond Green no matter how I twist it and turn it.
4) Player tracking data for Green seems to say great things too.

A phenomenon exists that cries out for an explanation, and both the qualitative and quantitative stuff seems to give an explanation in Green's direction. I need something significant therefore to make that something other than the most likely explanation.

[Doctor MJ]

Maybe "shot creating" would be a sort of different way to put it? He's not a guy you give the ball to and say to do something, which is a more common way of "play making" to be thought of, but again, don't know. I'll let this to the remaining voters to not bias anyone.

This is something of an ambiguity I'll admit.

I think where I really chafe here is that it's not like there's some guy on the Warriors they really ask to do that when Curry's in the game other than Curry. Thompson has a massive lead on Green for points per touch, but he also has a massive lead on Curry. On the team he only trails Speights. That ordering isn't there because Thompson's making plays, it's because he's the main guy others are making plays FOR. Fine to call him the 2nd best scorer on the team as a result of the skills he has that make that a reasonable plan, but not cool to grant him the secondary title of "playmaker" along with it.

[bondom34]

My "credit" statement was somewhat wondering as well b/c at times we've shifted credit around GSW as I'd mentioned in the past, but I suppose I misunderstood colinearity. The fact that these 2 play a ton of minutes together, and many less apart, would to me cause an issue in that you wouldn't be able to parse them apart in lineups/plus minus numbers. That to me seems something that could be an issue in general.

Sorry again that I derailed this whole thing, but just was maknig a point I felt strongly about earlier, and am a little to tired of trying to make. Again, I'm not saying a bad thing about Green, I just don't see him in the same way I see the others mentioned as far as impact on the season, clearly I'm in a miinority and that's cool.