The Stephen Curry Thread (2015-16 Pt. 2)

[MyUniBroDavis]

(Curry On, Green On, 2525 Mins) ORTG: 120.4, DRTG: 100.4, Net: +20.0
(Curry On, Green Off, 246 Mins) ORTG: 112.0, DRTG: 111.5, Net: +0.5
(Curry Off, Green On, 620 Mins) ORTG: 112.9, DRTG: 101.5, Net: +11.4
(Curry Off, Green Off, 1017 Mins) ORTG: 101.7, DRTG: 112.6, Net: -10.9

Curry's TS% with Green: 67.5%
Curry's TS% without Green: 59.3%

Curry's RPM: 8.33 (4th)
Green's RPM: 9.06 (1st)

Isn't the "Curry On, Green Off" sample size too small? I realize this is fully subjective but still, that's not a whole lot of games.

In terms of the stats that show up, most of it somewhat fits with the "norm"

But his fg percentage from 10-23 feet is 8/27, and his fg outside of 30 feet is 28.9%, compared to his regular averages of 44.5% and 45% from those places.

His TS would go down without green, doubt that it would go down by more than 64 or 65 though

When I look at sample size I just look at the players stats and see if anything is "outside the norm" and how this effects the stats

[mysticOscar]

(Curry On, Green On, 2525 Mins) ORTG: 120.4, DRTG: 100.4, Net: +20.0
(Curry On, Green Off, 246 Mins) ORTG: 112.0, DRTG: 111.5, Net: +0.5
(Curry Off, Green On, 620 Mins) ORTG: 112.9, DRTG: 101.5, Net: +11.4
(Curry Off, Green Off, 1017 Mins) ORTG: 101.7, DRTG: 112.6, Net: -10.9

Curry's TS% with Green: 67.5%
Curry's TS% without Green: 59.3%

Curry's RPM: 8.33 (4th)
Green's RPM: 9.06 (1st)

Isn't the "Curry On, Green Off" sample size too small? I realize this is fully subjective but still, that's not a whole lot of games.

But overall i think its enough to provide a good indicator of how important Dray is to the Warriors Defense as Curry is in offense.

[MyUniBroDavis]

(Curry On, Green On, 2525 Mins) ORTG: 120.4, DRTG: 100.4, Net: +20.0
(Curry On, Green Off, 246 Mins) ORTG: 112.0, DRTG: 111.5, Net: +0.5
(Curry Off, Green On, 620 Mins) ORTG: 112.9, DRTG: 101.5, Net: +11.4
(Curry Off, Green Off, 1017 Mins) ORTG: 101.7, DRTG: 112.6, Net: -10.9

Curry's TS% with Green: 67.5%
Curry's TS% without Green: 59.3%

Curry's RPM: 8.33 (4th)
Green's RPM: 9.06 (1st)

Isn't the "Curry On, Green Off" sample size too small? I realize this is fully subjective but still, that's not a whole lot of games.

But overall i think its enough to provide a good indicator of how important Dray is to the Warriors Defense as Curry is in offense.

I mean, I'm not sure if I could say that even, I mean, I think draymond is a better defender than kawhi to be honest, but even in the wowy data for defense there seem to be outliers (defensive rating goes above 120 with curry and Bogut on the floor, which flat out doesent make sense)

draymonds impact is amazing though, I think he clearly has the second best impact in the league

But like, not to seem close minded, but he isn't better than curry.

[Colbinii]

Isn't the "Curry On, Green Off" sample size too small? I realize this is fully subjective but still, that's not a whole lot of games.

But overall i think its enough to provide a good indicator of how important Dray is to the Warriors Defense as Curry is in offense.

I mean, I'm not sure if I could say that even, I mean, I think draymond is a better defender than kawhi to be honest, but even in the wowy data for defense there seem to be outliers (defensive rating goes above 120 with curry and Bogut on the floor, which flat out doesent make sense)

draymonds impact is amazing though, I think he clearly has the second best impact in the league

But like, not to seem close minded, but he isn't better than curry.

Why are you saying "I mean," and "but like," so much? They are redundant and make your post difficult to read.

Draymond Green has had more impact this season than Curry has. Right now I have them #1/#3, but I firmly believe they will finish #1/#2 at season's end with a championship trophy over their heads.

[cpower]

Draymond Green has not had more impact this season than Curry, not even close. If he had, his BPM and WS/48 would be a lot closer than it looks.

It's mind-blowing that one guy had arguably the GOAT RS and his teammate had more impact than him. lol

[Bourne85]

Draymond Green has not had more impact this season than Curry, not even close. If he had, his BPM and WS/48 would be a lot closer than it looks.

It's mind-blowing that one guy had arguably the GOAT RS and his teammate had more impact than him. lol

Jordan lovers clinging to anything they can. Cant have this Curry dude take MJ off his high horse

[Blackmill]

Draymond Green has not had more impact this season than Curry, not even close. If he had, his BPM and WS/48 would be a lot closer than it looks.

I'm not going to disagree with Curry having more impact than Green, since frankly I agree with you to some extent, but your reasoning is really ill-founded.

Curry has not had more impact this season than Green. If he had, his RAPM would be a lot better than Green's.

I'm sure you see the problem in the above statement. But it's no worse than what you said.

[MyUniBroDavis]

But overall i think its enough to provide a good indicator of how important Dray is to the Warriors Defense as Curry is in offense.

I mean, I'm not sure if I could say that even, I mean, I think draymond is a better defender than kawhi to be honest, but even in the wowy data for defense there seem to be outliers (defensive rating goes above 120 with curry and Bogut on the floor, which flat out doesent make sense)

draymonds impact is amazing though, I think he clearly has the second best impact in the league

But like, not to seem close minded, but he isn't better than curry.

Why are you saying "I mean," and "but like," so much? They are redundant and make your post difficult to read.

Draymond Green has had more impact this season than Curry has. Right now I have them #1/#3, but I firmly believe they will finish #1/#2 at season's end with a championship trophy over their heads.

I have pulled off 2 all nighters in a row for schools and exams so grammar ain't exactly what I'm thinking about right now.

I still don't see proof that draymond has more impact than curry. The only thing we have to go by really is rapm and nbawowy, which obviously has enough outliers with the data that I couldn't consider that a valid arguement at all.

I hope this doesent sound mean, but it annoys me when someone says something without providing any evidence, at least in a conversation like this. It's a valid arguement, dzra and lorak have shown this and have had damn good arguements for it, but I feel like you should provide evidence if you are gonna say something like this.

I mean, just taking out speights shows this. Team stats with curry on the court, speights off of it and draymond off it.

189 minutes, 403 possessions, vs the 250 minutes and 540 possessions with the only thing being curry on and draymond off

Offensive rating goes up to 118.4.

Exactly e opposite for draymond, offense goes up with speights.

The difference is that speights shoots 31.3% at the rim (5/16 from 0-3 feet) and 6/19 in the paint, vs 14/18 inside of 3 feet and 19/26 in the paint overall with draymond.

Now, while you can attribute some of this to playmaking, speights actually attempts a higher percentage of his shots in the paint with curry than with draymond, so I can't say curry makes him worse in that regard. (0-3 feet, I mean come on)

18/65 oh his shots are in the restricted areas 26/65 are in the paint

Vs 16/38 in the restricted area and 19/38 in the paint.

Barbosa is shooting a bad percentage too, but in his "off minutes" speights is 2/18. Taking him off increases points per shot, but decreases points per possessions. Curry has a lot of turnovers in that sample though, although I'd call this an outlier, since I don't think he is gonna not break, but shatter hardens turnover record no matter what.

Then there is Livingston, who also shoots below 42TS with curry.
Team offense goes up to 116.2 without him. With draymond, he shoots 58.6 TS.

Also, in the case of defense, defense with curry on, draymond and Bogut off 108.3

Defense with curry on, draymond off, Bogut on 117.4

I think that since Bogut is a clear plus defender, this should be considered an outlier as well

Alright, so the thing about this is someone might say "oh that's just one outlier" but the fact is it completely breaks his on off data without green on offense, and neither player can say they get worse with curry on the floor, since curry is an amazing playmaker and provides his own spacing as well.

Now, am I saying we can't credit draymond for making speights and Livingston succeed? No. Livingston is debatable, since draymond making him shoot that percentage is completely normal. But considering his career TS is 52.6%, 53.8 excluding his rookie year, I don't think curry, who would still likely have a positive Impact on him, would bring him down to 40s in TS, and low 40s at that.

Speights is a complete outlier on both sides. In one sample he is better than Barkley in the paint and in the other sample it looks like varsity players would give him a run for his money. And then we have Bogut for the defense in terms of outliers.

I believe draymond has superstar impact and is a superstar. But the Warriors still have a decent defense without him most likely, and curry's offensive impact under normal conditions far outshines draymonds defensive impact.

also, keep in mind curry is shooting half of his regular percentage form 30+ with draymond off. No, you don't playmake from 30+ feet.


Overall, I'd think that the Warriors are more of a 117 off 106 def with curry on and draymond off and 112 off and 101 def with draymond on and curry off.

Enough outliers that I don't take the wowy data seriously.

[MyUniBroDavis]

But overall i think its enough to provide a good indicator of how important Dray is to the Warriors Defense as Curry is in offense.

I mean, I'm not sure if I could say that even, I mean, I think draymond is a better defender than kawhi to be honest, but even in the wowy data for defense there seem to be outliers (defensive rating goes above 120 with curry and Bogut on the floor, which flat out doesent make sense)

draymonds impact is amazing though, I think he clearly has the second best impact in the league

But like, not to seem close minded, but he isn't better than curry.

Why are you saying "I mean," and "but like," so much? They are redundant and make your post difficult to read.

Draymond Green has had more impact this season than Curry has. Right now I have them #1/#3, but I firmly believe they will finish #1/#2 at season's end with a championship trophy over their heads.

I have pulled off 2 all nighters in a row for schools and exams so grammar ain't exactly what I'm thinking about right now.

I still don't see proof that draymond has more impact than curry. The only thing we have to go by really is rapm and nbawowy, which obviously has enough outliers with the data that I couldn't consider that a valid arguement at all.

I hope this doesent sound mean, but it annoys me when someone says something without providing any evidence, at least in a conversation like this. It's a valid arguement, dzra and lorak have shown this and have had damn good arguements for it, but I feel like you should provide evidence if you are gonna say something like this.

I mean, just taking out speights shows this. Team stats with curry on the court, speights off of it and draymond off it.

189 minutes, 403 possessions, vs the 250 minutes and 540 possessions with the only thing being curry on and draymond off

Offensive rating goes up to 118.4.

Exactly e opposite for draymond, offense goes up with speights.

The difference is that speights shoots 31.3% at the rim (5/16 from 0-3 feet) and 6/19 in the paint, vs 14/18 inside of 3 feet and 19/26 in the paint overall with draymond.

Now, while you can attribute some of this to playmaking, speights actually attempts a higher percentage of his shots in the paint with curry than with draymond, so I can't say curry makes him worse in that regard. (0-3 feet, I mean come on)

18/65 oh his shots are in the restricted areas 26/65 are in the paint

Vs 16/38 in the restricted area and 19/38 in the paint.

Barbosa is shooting a bad percentage too, but in his "off minutes" speights is 2/18. Taking him off increases points per shot, but decreases points per possessions. Curry has a lot of turnovers in that sample though, although I'd call this an outlier, since I don't think he is gonna not break, but shatter hardens turnover record no matter what.

Then there is Livingston, who also shoots below 42TS with curry.
Team offense goes up to 116.2 without him. With draymond, he shoots 58.6 TS.

Also, in the case of defense, defense with curry on, draymond and Bogut off 108.3

Defense with curry on, draymond off, Bogut on 117.4

I think that since Bogut is a clear plus defender, this should be considered an outlier as well

Alright, so the thing about this is someone might say "oh that's just one outlier" but the fact is it completely breaks his on off data without green on offense, and neither player can say they get worse with curry on the floor, since curry is an amazing playmaker and provides his own spacing as well.

Now, am I saying we can't credit draymond for making speights and Livingston succeed? No. Livingston is debatable, since draymond making him shoot that percentage is completely normal. But considering his career TS is 52.6%, 53.8 excluding his rookie year, I don't think curry, who would still likely have a positive Impact on him, would bring him down to 40s in TS, and low 40s at that.

Speights is a complete outlier on both sides. In one sample he is better than Barkley in the paint and in the other sample it looks like varsity players would give him a run for his money. And then we have Bogut for the defense in terms of outliers.

I believe draymond has superstar impact and is a superstar. But the Warriors still have a decent defense without him most likely, and curry's offensive impact under normal conditions far outshines draymonds defensive impact.

also, keep in mind curry is shooting half of his regular percentage form 30+ with draymond off. No, you don't playmake from 30+ feet.


Overall, I'd think that the Warriors are more of a 117 off 106 def with curry on and draymond off and 112 off and 101 def with draymond on and curry off.

Enough outliers that I don't take the wowy data seriously.

Considering we have 4 players with curry shooting sub 45TS that account for 86 of the 483 shots, I think that there's a problem. Now, the only main outliers are Livingston and speights. But they are pretty big outliers. Big enough that taking them out of the equation equals to a 131 minute sample size, a bit over half, and the offense is above 120 there.

[cpower]

Draymond Green has not had more impact this season than Curry, not even close. If he had, his BPM and WS/48 would be a lot closer than it looks.

I'm not going to disagree with Curry having more impact than Green, since frankly I agree with you to some extent, but your reasoning is really ill-founded.

Curry has not had more impact this season than Green. If he had, his RAPM would be a lot better than Green's.

I'm sure you see the problem in the above statement. But it's no worse than what you said.

RAPM has significant more standard errors than BPM, especially for small sample size. For a full season of minutes, the standard error would be around 2 for BPM. And if one guy has 12.4 BPM and another guy has 5.9 BPM, then the gap is too significant to be ignored.

[MyUniBroDavis]

Draymond Green has not had more impact this season than Curry, not even close. If he had, his BPM and WS/48 would be a lot closer than it looks.

I'm not going to disagree with Curry having more impact than Green, since frankly I agree with you to some extent, but your reasoning is really ill-founded.

Curry has not had more impact this season than Green. If he had, his RAPM would be a lot better than Green's.

I'm sure you see the problem in the above statement. But it's no worse than what you said.

RAPM has significant more standard errors than BPM, especially for small sample size. And if one guy has 12.4 BPM and another guy has 5.9 BPM, then the gap is too significant to be ignored.

Isn't that box score derived? I agree with the sample size arguement. Outliers effect the data way too much for me to take any plus minus regarding those 2 as a end all arguement.

My main thing is that when curry goes in his streaks, like in overtime, it doesent make sense to give green the same credit on offense.

[Doctor MJ]

Draymond Green has not had more impact this season than Curry, not even close. If he had, his BPM and WS/48 would be a lot closer than it looks.

I'm not going to disagree with Curry having more impact than Green, since frankly I agree with you to some extent, but your reasoning is really ill-founded.

Curry has not had more impact this season than Green. If he had, his RAPM would be a lot better than Green's.

I'm sure you see the problem in the above statement. But it's no worse than what you said.

RAPM has significant more standard errors than BPM, especially for small sample size. For a full season of minutes, the standard error would be around 2 for BPM. And if one guy has 12.4 BPM and another guy has 5.9 BPM, then the gap is too significant to be ignored.

While I don't advocate ignoring anything, I don't know why you're bringing BPM and WS/48 into an impact discussion (they aren't impact stats), nor why you'd use them to dismiss a player that we know the box score woefully underrates.

[Blackmill]

I'm not going to disagree with Curry having more impact than Green, since frankly I agree with you to some extent, but your reasoning is really ill-founded.



I'm sure you see the problem in the above statement. But it's no worse than what you said.

RAPM has significant more standard errors than BPM, especially for small sample size. And if one guy has 12.4 BPM and another guy has 5.9 BPM, then the gap is too significant to be ignored.

Isn't that box score derived? I agree with the sample size arguement. Outliers effect the data way too much for me to take any plus minus regarding those 2 as a end all arguement.

My main thing is that when curry goes in his streaks, like in overtime, it doesent make sense to give green the same credit on offense.

BPM is box score derived.

At times, BPM seems to have more trouble picking up offensive impact resulting from elite playmaking than from elite scoring. For instance, Nash appears below Wade & Kobe in many of his prime years.

In general, BPM also falters some in capturing defensive impact. Or at least with power forwards and centers. You can see here a visual of the correlation between BPM and DRAPM for different positions.

[cpower]

I'm not going to disagree with Curry having more impact than Green, since frankly I agree with you to some extent, but your reasoning is really ill-founded.



I'm sure you see the problem in the above statement. But it's no worse than what you said.

RAPM has significant more standard errors than BPM, especially for small sample size. For a full season of minutes, the standard error would be around 2 for BPM. And if one guy has 12.4 BPM and another guy has 5.9 BPM, then the gap is too significant to be ignored.

While I don't advocate ignoring anything, I don't know why you're bringing BPM and WS/48 into an impact discussion (they aren't impact stats), nor why you'd use them to dismiss a player that we know the box score woefully underrates.

BPM is a box score based metric to evaluating basketball players' quality and contribution to the team, I consider that's an impact stat.

Does box score woefully underrates Green? I don't think so, he gets more assists from playing with Curry alone.

[Doctor MJ]

RAPM has significant more standard errors than BPM, especially for small sample size. For a full season of minutes, the standard error would be around 2 for BPM. And if one guy has 12.4 BPM and another guy has 5.9 BPM, then the gap is too significant to be ignored.

While I don't advocate ignoring anything, I don't know why you're bringing BPM and WS/48 into an impact discussion (they aren't impact stats), nor why you'd use them to dismiss a player that we know the box score woefully underrates.

BPM is a box score based metric to evaluating basketball players' quality and contribution to the team, I consider that's an impact stat.

Does box score woefully underrates Green? I don't think so, he gets more assists from playing with Curry alone.

Okay how about this:

Give us a reference to a serious source outside of RealGM referring to impact stats which goes with your thinking.

Short of that though, the use of the term "impact" is in my experience specifically used in contrast to "production" stats as a way of referring to +/--based stats contrasted from "box score".

Re: how box score underrate Green? Uh, the dude is probably the best defensive player on the planet. Tell me what in his box score indicates he's anywhere near that good.

[cpower]

While I don't advocate ignoring anything, I don't know why you're bringing BPM and WS/48 into an impact discussion (they aren't impact stats), nor why you'd use them to dismiss a player that we know the box score woefully underrates.

BPM is a box score based metric to evaluating basketball players' quality and contribution to the team, I consider that's an impact stat.

Does box score woefully underrates Green? I don't think so, he gets more assists from playing with Curry alone.

Okay how about this:

Give us a reference to a serious source outside of RealGM referring to impact stats which goes with your thinking.

Short of that though, the use of the term "impact" is in my experience specifically used in contrast to "production" stats as a way of referring to +/--based stats contrasted from "box score".

Re: how box score underrate Green? Uh, the dude is probably the best defensive player on the planet. Tell me what in his box score indicates he's anywhere near that good.

This is coming from APBRmetrics Author:

1. Best box score metric, has the advantages and disadvantages of box score metrics.
2. When a good quality RAPM is not available or doesn't measure what you want. RAPM is good for a specific situation: estimating true talent level over a period of months and years. If you want a stat that will stabilize quickly over smaller periods of time, BPM is far, far better than RAPM. For instance--who has been playing best this year? 5 games in, BPM can give you a good idea, RAPM cannot. Also, if you don't have access to lineup data (history or other leagues), then BPM is the best option. Even with multi-year samples for estimating true talent level, RAPM can still struggle with multicollinearity issues, hence the need for RPM/xRAPM. Even at the end of this season--if you asked me who played better this year, I think BPM is more accurate than RAPM, and perhaps even xRAPM/RPM (since prior years are included). (Remember, though, BPM doesn't capture defense extremely well.)

[Doctor MJ]

BPM is a box score based metric to evaluating basketball players' quality and contribution to the team, I consider that's an impact stat.

Does box score woefully underrates Green? I don't think so, he gets more assists from playing with Curry alone.

Okay how about this:

Give us a reference to a serious source outside of RealGM referring to impact stats which goes with your thinking.

Short of that though, the use of the term "impact" is in my experience specifically used in contrast to "production" stats as a way of referring to +/--based stats contrasted from "box score".

Re: how box score underrate Green? Uh, the dude is probably the best defensive player on the planet. Tell me what in his box score indicates he's anywhere near that good.

This is coming from APBRmetrics Author:

1. Best box score metric, has the advantages and disadvantages of box score metrics.
2. When a good quality RAPM is not available or doesn't measure what you want. RAPM is good for a specific situation: estimating true talent level over a period of months and years. If you want a stat that will stabilize quickly over smaller periods of time, BPM is far, far better than RAPM. For instance--who has been playing best this year? 5 games in, BPM can give you a good idea, RAPM cannot. Also, if you don't have access to lineup data (history or other leagues), then BPM is the best option. Even with multi-year samples for estimating true talent level, RAPM can still struggle with multicollinearity issues, hence the need for RPM/xRAPM. Even at the end of this season--if you asked me who played better this year, I think BPM is more accurate than RAPM, and perhaps even xRAPM/RPM (since prior years are included). (Remember, though, BPM doesn't capture defense extremely well.)

I notice you're not responding to what I actually said, so you might want to get on that . To respond to the new content:

Box score based stats have better precision, or reliability, than +/- stats
+/- stats have better accuracy, or validity, than box score based stats

The APBRmetric author stating BPM is more "accurate" than RAPM is fundamentally wrong if we're being pedantic, but for the reasons I give BPM certainly is more useful over small enough sample size.

I'll also note though how he basically tells you that BPM is will tell you very little about Green's defense, and you were questioning how such stats could underrate Green. Really until we get player tracking stats to be really, really good, DRAPM is a far better way of judging defensive impact than any box score based metric.

[cpower]

Okay how about this:

Give us a reference to a serious source outside of RealGM referring to impact stats which goes with your thinking.

Short of that though, the use of the term "impact" is in my experience specifically used in contrast to "production" stats as a way of referring to +/--based stats contrasted from "box score".

Re: how box score underrate Green? Uh, the dude is probably the best defensive player on the planet. Tell me what in his box score indicates he's anywhere near that good.

This is coming from APBRmetrics Author:

1. Best box score metric, has the advantages and disadvantages of box score metrics.
2. When a good quality RAPM is not available or doesn't measure what you want. RAPM is good for a specific situation: estimating true talent level over a period of months and years. If you want a stat that will stabilize quickly over smaller periods of time, BPM is far, far better than RAPM. For instance--who has been playing best this year? 5 games in, BPM can give you a good idea, RAPM cannot. Also, if you don't have access to lineup data (history or other leagues), then BPM is the best option. Even with multi-year samples for estimating true talent level, RAPM can still struggle with multicollinearity issues, hence the need for RPM/xRAPM. Even at the end of this season--if you asked me who played better this year, I think BPM is more accurate than RAPM, and perhaps even xRAPM/RPM (since prior years are included). (Remember, though, BPM doesn't capture defense extremely well.)

I notice you're not responding to what I actually said, so you might want to get on that . To respond to the new content:

Box score based stats have better precision, or reliability, than +/- stats
+/- stats have better accuracy, or validity, than box score based stats

The APBRmetric author stating BPM is more "accurate" than RAPM is fundamentally wrong if we're being pedantic, but for the reasons I give BPM certainly is more useful over small enough sample size.

I'll also note though how he basically tells you that BPM is will tell you very little about Green's defense, and you were questioning how such stats could underrate Green. Really until we get player tracking stats to be really, really good, DRAPM is a far better way of judging defensive impact than any box score based metric.

Green's defense has been well measured in DBPM, only 1 point less than DRPM but I believe they also credited him more on offensive end. For example, Green is a 39% three point shooter this season, how much of that was due to playing with the presence of Curry? His Tov% has been horrible at 21%, but there is no way to punish that in RAPM/RPM . I think they are severely underrates Curry's offensive impact in RAPM/RPM as well.

[Doctor MJ]

Green's defense has been well measured in DBPM, only 1 point less than DRPM but I believe they also credited him more on offensive end. For example, Green is a 39% three point shooter this season, how much of that was due to playing with the presence of Curry? His Tov% has been horrible at 21%, but there is no way to punish that in RAPM/RPM . I think they are severely underrates Curry's offensive impact in RAPM/RPM as well.

Okay stop it now. Just because a player looks like a good player by a certain metric doesn't mean it's measuring him well. BPM attaches weight to box score stats. There are no box score stats that judge man defense or heady plays that don't lead to that player getting a block, steal, or rebound. Therefore DBPM in theory might be overrating or underrating Green, all we know for sure is that we can't have any confidence that it's getting things right.

As far as having no way to punish a guy for a specific issue in RAPM, that's not true. RAPM literally factors everything in. It's not a perfect stat but it most definitely factors in everything that occurs out on the floor without bias toward glamour or easy quantification.

[cpower]

Green's defense has been well measured in DBPM, only 1 point less than DRPM but I believe they also credited him more on offensive end. For example, Green is a 39% three point shooter this season, how much of that was due to playing with the presence of Curry? His Tov% has been horrible at 21%, but there is no way to punish that in RAPM/RPM . I think they are severely underrates Curry's offensive impact in RAPM/RPM as well.

Okay stop it now. Just because a player looks like a good player by a certain metric doesn't mean it's measuring him well. BPM attaches weight to box score stats. There are no box score stats that judge man defense or heady plays that don't lead to that player getting a block, steal, or rebound. Therefore DBPM in theory might be overrating or underrating Green, all we know for sure is that we can't have any confidence that it's getting things right.

As far as having no way to punish a guy for a specific issue in RAPM, that's not true. RAPM literally factors everything in. It's not a perfect stat but it most definitely factors in everything that occurs out on the floor without bias toward glamour or easy quantification.

Please enlighten me how Nikola Jokic is an more impactful player than Kevin Durant.