2016 RAPM/RPM/etc. *Full 2016 RS + PS RPM & RAPM Updated 6/24*

[lorak]

KAT very good on offense, but also very bad on defense. Porzignis clearly more valuable than him overall.

Most valuable rookie is Nikola Jokic, well done DEN..

Yes, but he didn't play enough minutes. And i'm not sure if you are sarcastic about Nuggets scouting job, but they have young scout from Europe, which I know, and he is doing great job, so it's not suprising at all he discovered so good player.

[SideshowBob]

Green is a full SD above James/Curry in Single-Year

[SideshowBob]

Kobe only 7th worst on Lakers rosters by single-year. About break-even offense in both sets (probably still some degree of disruption on defenses) but near bottom of the league on defense.

[Dr Spaceman]

Green is a full SD above James/Curry in Single-Year

Have you ever officially given your thoughts on why this is occurring? I'm interested to hear your take.

I'm still rolling with the hypothesis that as GSW is currently defended (i.e. Trap pick and roll, keep the ball away from Steph) Green may legitimately be having more impact on the scoring margin than Steph. That said, if the equilibrium shifted toward more defensive attention for Green, the team as a whole would be clearly worse overall. EDIT: upon reflection this isn't actually what I mean; I mean that teams would be way more successful at stopping Draymond than they are with Steph. I'm not sure what that would do the team results.

I'm just not sure if this makes mathematical sense, and to be honest it doesn't totally *feel* right. Between that and deciding whether Green deserves to be ranked above someone like Paul or Westbrook, it's too much cognitive dissonance for one poster

[Doctor MJ]

What strikes me about Green's numbers when I look at them right now is that the DRAPM makes sense, and that while I think Curry's true edge on offense is bigger than the ORAPM indicates, it does seem plausible that Green's defensive edge may still give him the overall edge even when we factor in all those things we look to factor in in our skepticism.

[SideshowBob]

2014 & 2015 Curry vs. Green

Code: Select all

Season  Set    Player     Off      Def      Net

2014    P-I    Curry    +3.84    -0.49    +4.33
2014    P-I    Green    -0.69    +1.45    +0.76
2014    M-Y    Curry    +4.63    +0.55    +5.18
2014    M-Y    Green    +0.66    +2.71    +3.36
2014    S-Y    Curry    +3.02    +0.43    +3.46
2014    S-Y    Green    +0.66    +2.23    +2.89


Code: Select all

2015    M-Y    Curry    +5.90    +1.04    +6.94
2015    M-Y    Green    +1.78    +4.20    +5.98
2015    S-Y    Curry    +4.39    +2.18    +6.57
2015    S-Y    Green    +2.28    +3.77    +6.04


[SideshowBob]

Green is a full SD above James/Curry in Single-Year

Have you ever officially given your thoughts on why this is occurring? I'm interested to hear your take.

Not really, because I'm still thinking on it, but the simplest explanation is that Green's just providing more impact given everything that's occurred this year. This isn't really that far-fetched - we've seen Odom look better than Bryant on title teams, Ginobili looking as impressive as Duncan as early as 05, Collison outshining Durant/Westbrook, etc. These guys just enable their team's ceilings.

I'm still rolling with the hypothesis that as GSW is currently defended (i.e. Trap pick and roll, keep the ball away from Steph) Green may legitimately be having more impact on the scoring margin than Steph. That said, if the equilibrium shifted toward more defensive attention for Green, the team as a whole would be clearly worse overall.

In what way? If teams game-plan for Green sliding in up top/rolling down the middle and swinging/finishing instead of traping Curry, then Curry wrecks them. But when Curry's on court and Green's off court, this wouldn't happen, Curry would have a harder time, and Green would still be credited in +/-. Curry needs Draymond to have his maximum effect, whereas Draymond has his hands in enough of the GS's pies that he's going to be covering more bases at all times, even if he doesn't do one single thing as good as Curry can.

EDIT: upon reflection this isn't actually what I mean; I mean that teams would be way more successful at stopping Draymond than they are with Steph. I'm not sure what that would do the team results.

Well it could go either way - but the question is what would happen on court? With Curry ON, the answer is that he'd get more space/freedom, and the likely result of that is that he goes off (the extent to which he does determines how much of a shift, if any, we'd see in GS's performance).

With Curry off the court, its hard to say, we need to hone-in on how teams defend Draymond with Curry on the bench a bit more. However, the results of RAPM seem to be nudging in the direction that seems to be managing to exert a similar level of pressure on opposing defenses

Generally speaking though, we'd expect that over ~79 games + some similar experience from last year, NBA defenses have already trended towards the equilibrium that would yield them the best results on average. What kind of exact data do we have on how Curry's being defended when handling on the P&R? Is it a healthy balance of switch/trap/1on1 or is it leaning in towards trap (my perception leads me to believe the latter, but I can't take that as gospel)? That would at least be indicative of what the current equilibrium is and then we could further hypothesize about how Draymond's presence is affecting that equilibrium and which direction that influence would go if the balance tipped in a different way.

I'm just not sure if this makes mathematical sense, and to be honest it doesn't totally *feel* right. Between that and deciding whether Green deserves to be ranked above someone like Paul or Westbrook, it's too much cognitive dissonance for one poster

I think if you're criteria is impact given the situation you have to be leaning Green. If you're focused on goodness across thinkable scenarios you would probably slightly side with the other two. Personally I have Westbrook/Paul ahead but even then I think Green's not more than a point worse than them, and given the almost certain level of error/bias in my subjectivity, plus the +/- data would lead me to believe that its entirely possible that Green is possibly as good or better than them, even in a vacuum.

[theonlyclutch]

KAT very good on offense, but also very bad on defense. Porzignis clearly more valuable than him overall.

Most valuable rookie is Nikola Jokic, well done DEN..

Yes, but he didn't play enough minutes. And i'm not sure if you are sarcastic about Nuggets scouting job, but they have young scout from Europe, which I know, and he is doing great job, so it's not suprising at all he discovered so good player.

He played like ~300 minutes less than Porzingis for the whole year, that's plenty enough minutes,

No sarcasm w.r.t Jokic here, DEN really hit it out of the park

[lorak]

Most valuable rookie is Nikola Jokic, well done DEN..

Yes, but he didn't play enough minutes. And i'm not sure if you are sarcastic about Nuggets scouting job, but they have young scout from Europe, which I know, and he is doing great job, so it's not suprising at all he discovered so good player.

He played like ~300 minutes less than Porzingis for the whole year, that's plenty enough minutes,

I'm talking about MPG.

[Dr Spaceman]

Chiming into the Green/Curry discussion. I think that I'm one of the few (am I the only?) really active posters that doesn't have any problem at all with the concept that Green might be as important to the Warriors as Curry. I'm pretty thoroughly not convinced that Curry is having the GOAT peak (in the last thread on it, I had him fighting for top-10). And I'm also pretty thoroughly open to the concept that Green's level of impact might be more translatable to more teams in the current NBA than Curry's.

Now that I've blasphemed and pissed off the majority of the people on here, including some of the ones that I respect the most, I'll continue. I'm not going to go too deep here because I literally don't have time for the type of marathon post that I like to do. But I'm going to take an entirely different tact, and focus entirely on Green here.

With RAPM, WOWY, and all of the attempts that we've made in various projects to try to estimate a player's non-boxscore impact, there are some pretty clear trends about the types of players that have huge impacts. We see great point guards having bigger offensive impact than their box scores suggest, great defensive bigs having bigger impact than their boxscore stats suggest. We see spacing showing up well, especially among bigs. These are three of the biggies, as far as unexpected impact, with a player showing strength in any one of them often enough to boost them surprisingly high in the impact studies.

And Green shows up well in ALL THREE of them!

I And-1'd someones post recently...maybe RonnyMac?, that pointed that out succinctly. But I don't think that concept is getting nearly enough run around here. Draymond Green is essentially playing point guard for the Warriors, from the big-man positions. He is a very credible DPoY candidate, by both reputation and measured impact. AND he's a 40% shooter from downtown, again from the big man slots.

Let's take two of those skills, big man as an offensive hub from the high-post, and defensive player-of-the-year caliber defensive impact. Using B-R's season-finder (for forward-centers with at least 5 assists per game), this is the list of players in NBA history that could fit that description:

Bill Russell
Wilt Chamberlain
Kareem Abdul-Jabbar
BIll Walton
Kevin Garnett
Joakim Noah (2014)
Draymond Green

And that's pretty much it. Three are some names on the list I'm not fully familiar with, but this is at least the majority of the names that could realistically be argued as a high-post offensive hub big man with DPoY caliber defense.

Now, the point here is obviously not to say that Draymond Green is the GOAT of his generation the way some of those other names were. But on the flip side, we have people (again, posters I really respect) doing mental gymnastics to figure out a reason for Curry having to be the biggest impact player for these Warriors. We're turning the numbers upside down. We're trying to break it down by competition level. We're arguing over who played with better tertiary players.

And again, you all know I'm all about digging into the context to tell the story. But in this case, most of the posts aren't looking for full context...they're trying to make the numbers fit a preconceived notion. And even doing that, from what I've seen of the arguments, all of this digging isn't producing anything all that convincing.

Meanwhile, every impact study we've ever conducted would strongly suggest that a big man that you could run the offense through from out-top, that played DPoY defense, AND provided excellent big-man spacing would be an off-the-charts impact player. And, lo-and-behold, Draymond Green fits all of those categories and has been measuring out as an off-the-charts impact player for two straight seasons now. Again, this isn't rigorous by any means, but...maybe we should at least give some consideration to the possibility that Green really might be who he seems to be measuring out to be.

Just for clarity's sake:

If you were on an opposing coaching staff, would you feel comfortable with the statement "we should focus more energy on stopping Green than we do Curry?"

[NyCeEvO]

I guess my issue with regards to solely pointing to RAPM as to why Curry can't be in the discussion of GOAT peaks is...it's one stat, in which he looks very good in anyway.

And I kind of feel that RAPM tends to assign credit to Green that should go to Curry. I think Green's role exists because of Curry, and I don't think he would look anywhere close to as good as he does right now without Curry. I know there are small sample sizes of data that says Green has actually done well without Curry, but over a larger sample size, I believe he would see a big dropoff.

For example, his 3pt shooting. He shoots around 40%. But his shots are also WIDE OPEN. Because everyone is paying attention to Curry. If Green's shots are more contested, he probably won't shoot 40% anymore. He probably won't have as many 4 on 3 opportunities anymore (which means he's not going to average over 7 apg anymore). His ability to run a PnR as the ball handler isn't going to be as successful anymore, because defenses can focus more on stopping him, rather than the screener (usually Curry).

I think it's fair to assume that maybe Green is stealing some of that "impact" from Curry in terms of the numbers.

I also think it's kind of weird how the box score tends to get totally dismissed. I like +/-, but I think the box score has value too.

EDIT: that's not to say that I think Green is necessarily overrated. I think he's fantastic. And the portability of a Draymond Green is duly noted. I certainly think he's in the conversation of top 5 this season. IDK if I'd vote for him that high, but he's without a doubt a top 10 player for me.

But which version of RAPM are you referring to though because there are different types. We have single-year RAPM, multi-year RAPM, xRAPM, NPI RAPM, et al. Then when you're using PI RAPM, how many years of data are we incorporating into the calculation?

All of those factors and nuances change the outcomes of the numerical value given for each player.

[MyUniBroDavis]

Chiming into the Green/Curry discussion. I think that I'm one of the few (am I the only?) really active posters that doesn't have any problem at all with the concept that Green might be as important to the Warriors as Curry. I'm pretty thoroughly not convinced that Curry is having the GOAT peak (in the last thread on it, I had him fighting for top-10). And I'm also pretty thoroughly open to the concept that Green's level of impact might be more translatable to more teams in the current NBA than Curry's.

Now that I've blasphemed and pissed off the majority of the people on here, including some of the ones that I respect the most, I'll continue. I'm not going to go too deep here because I literally don't have time for the type of marathon post that I like to do. But I'm going to take an entirely different tact, and focus entirely on Green here.

With RAPM, WOWY, and all of the attempts that we've made in various projects to try to estimate a player's non-boxscore impact, there are some pretty clear trends about the types of players that have huge impacts. We see great point guards having bigger offensive impact than their box scores suggest, great defensive bigs having bigger impact than their boxscore stats suggest. We see spacing showing up well, especially among bigs. These are three of the biggies, as far as unexpected impact, with a player showing strength in any one of them often enough to boost them surprisingly high in the impact studies.

And Green shows up well in ALL THREE of them!

I And-1'd someones post recently...maybe RonnyMac?, that pointed that out succinctly. But I don't think that concept is getting nearly enough run around here. Draymond Green is essentially playing point guard for the Warriors, from the big-man positions. He is a very credible DPoY candidate, by both reputation and measured impact. AND he's a 40% shooter from downtown, again from the big man slots.

Let's take two of those skills, big man as an offensive hub from the high-post, and defensive player-of-the-year caliber defensive impact. Using B-R's season-finder (for forward-centers with at least 5 assists per game), this is the list of players in NBA history that could fit that description:

Bill Russell
Wilt Chamberlain
Kareem Abdul-Jabbar
BIll Walton
Kevin Garnett
Joakim Noah (2014)
Draymond Green

And that's pretty much it. Three are some names on the list I'm not fully familiar with, but this is at least the majority of the names that could realistically be argued as a high-post offensive hub big man with DPoY caliber defense.

Now, the point here is obviously not to say that Draymond Green is the GOAT of his generation the way some of those other names were. But on the flip side, we have people (again, posters I really respect) doing mental gymnastics to figure out a reason for Curry having to be the biggest impact player for these Warriors. We're turning the numbers upside down. We're trying to break it down by competition level. We're arguing over who played with better tertiary players.

And again, you all know I'm all about digging into the context to tell the story. But in this case, most of the posts aren't looking for full context...they're trying to make the numbers fit a preconceived notion. And even doing that, from what I've seen of the arguments, all of this digging isn't producing anything all that convincing.

Meanwhile, every impact study we've ever conducted would strongly suggest that a big man that you could run the offense through from out-top, that played DPoY defense, AND provided excellent big-man spacing would be an off-the-charts impact player. And, lo-and-behold, Draymond Green fits all of those categories and has been measuring out as an off-the-charts impact player for two straight seasons now. Again, this isn't rigorous by any means, but...maybe we should at least give some consideration to the possibility that Green really might be who he seems to be measuring out to be.

Just for clarity's sake:

If you were on an opposing coaching staff, would you feel comfortable with the statement "we should focus more energy on stopping Green than we do Curry?"

I think it's just that certain lineups, like the sbds, need green more than curry, if you know what I mean.

Otoh, some people have gone into lineups. While rapm is supposed to fix this, I feel like greens minutes without a "go to scorer" in klay or curry are so low that they don't really "show" in his rapm numbers, you know what I mean?

I mean, the-power made a post on game planning, and it does make sense that the Warriors have more plays without curry than without draymond. It's kinda like a just in case... Kinda thing, so that if curry is out they still have plays they can run. While draymond is great and all, he doesent warrant that on offense, if you know what I mean. I think that 2 things we have to consider are that a bad game from curry, despite it happening rarely, impact wise, effects the team more than a bad game from draymond. Otoh, draymonds impact is a lot more consistent, because he will rarely go 35 minutes with bad defense, and a lot of his impact is just his spacing honestly.

I mean, looking at each of the games curry hasn't played while draymond has, the Warriors had offensive rtgs of

110.8 against the Hawks
114.3 vs the rockets
97.6 against the mavericks
98.8 against the rockets
121.4 against the rockets

Small sample size

Now, looking at each of these games, klay shot pretty badly in most of them, which is honestly expected. Some games were overboard but his shooting going down and his volume/shot difficulty increasing isn't shocking at all

Looking at the games the Warriors did well on offense

Against the Hawks, they had 18 offensive rebounds. They also had relatively few turnovers, at 11, while the Hawks had 17 (idk how many were live ball, but we know how that effects opposing offense)

Klay shot 16/27 against the rockets in their 4 point win

34.2% offensive rebound rate against the rockets in game 2. They scored 17 points off of these.

In 3 of these 5 games, green had a net rating that was tied for the lowest or second lowest on his team for players with over 25 minutes, the 2 games this didn't happen were the Atlanta game (net rating of 10, second best on his team though 2 players with 8 less minutes had ratings of 9, and klay had a rating of 2 despite going 8/27, though all of their ratings would have gone up as well I guess if klay made his shots)

And the Dallas game (only 3 players played more than 25 minutes, -11)

And whenever curry and draymond play together, net rating wise, I can't help but feel that a bad draymond performance can be drowned out by a good curry performance than vice versa

Small sample size but iirc, JE himself said that the rapm comparison between curry and draymond would not be reliable because of sample size. The thing that was notable was the difference in stats.

And tbh in terms of sample size, when curry was on the floor with Bogut, who obviously is a good defender, the opponents offensive rating jumped up to the 120s so that would effect both of their rapms. Also, greens stats without klay and curry were solid, which wouldn't be something I'd knock him for if the "stats" shown were, well, not full of noise.

Next year, I hope someone does something like blackmill said in his box score thread so we can finally clarify this.

[lorak]

Green without Curry is better in every box score category except of FG%, that includes better 3p% with increased volume as well as tov are lower with more asisst.

On the other hand Curry is way worse shooter without Green, his 3p% is down by almost 8 percentge points (so maybe we should start talking about how having elite playmaker like Green helps Curry?), tov are also higher, with the same amount of ast, but bigger volume.

And Green's +/- w/o Curry is better than Steph's w/o Draymond.

So maybe instead of using small sample excuse (while BTW these samples are not so small, more than 10 full games) try to find resonable explanation why results over and over again favour Green. And by resonable I mean something better than "GSW have more plays w/o Green" - as there is no way we can check if that's true or not.

[cpower]

Green without Curry is better in every box score category except of FG%, that includes better 3p% with increased volume as well as tov are lower with more asisst.

On the other hand Curry is way worse shooter without Green, his 3p% is down by almost 8 percentge points (so maybe we should start talking about how having elite playmaker like Green helps Curry?), tov are also higher, with the same amount of ast, but bigger volume.

And Green's +/- w/o Curry is better than Steph's w/o Draymond.

So maybe instead of using small sample excuse (while BTW these samples are not so small, more than 10 full games) try to find resonable explanation why results over and over again favour Green. And by resonable I mean something better than "GSW have more plays w/o Green" - as there is no way we can check if that's true or not.

Green without Curry:
53%TS in 437 mins (drops 9.7%)
Curry without Green
60%TS in 241 mins (drops 10%)
Their offensive production drops at the same rate, while green played almost the double minutes without Curry. The lineups are very different so it's hard to come to any conclusions.

[lorak]

TS includes FT% and neither of them can have incluence on how other is shooting FTs. And Curry improves a lot from the line wihtout Green. Besides 3p% is the best indicator here, because it depends the most on creation from others (more than 2p shots).

[cpower]

TS includes FT% and neither of them can have incluence on how other is shooting FTs. And Curry improves a lot from the line wihtout Green. Besides 3p% is the best indicator here, because it depends the most on creation from others (more than 2p shots).

again Curry was 34/88(38%) shooting threes in those 240 mins, playing mainly with Mo, Barnes, Iggy, Barbosa. He also put up 2 threes at the end of each 1st and 3rd quarter, mostly from half court. He had to play a lot of late clock, contested long threes and the sample size isn't big enough to mean anything.

[colts18]

450 minutes is small for any plus/minus analysis. There is a lot of randomness in full season results which means there is a ton more randomness in a sample that is a fraction of a full season.

88 3 point attempts is a small sample too. Assuming that Curry's true 3P% is 45%, you would expect to have a 38% 3P% or less about 14% of the time.

[lorak]

"A lot" so how much exactly in comparison to his numbers with Green? And bTW, such roster means he had very good spacing. And 88 threes isn't small sample. His drop off here - almost 8 percentage points - is significant.

[colts18]

"A lot" so how much exactly in comparison to his numbers with Green? And bTW, such roster means he had very good spacing. And 88 threes isn't small sample. His drop off here - almost 8 percentage points - is significant.

It's not significant. Using a Binomial distribution, you would expect a 45% 3P shooter to make 34/88 or less about 14% of the time.

[MyUniBroDavis]

Green without Curry is better in every box score category except of FG%, that includes better 3p% with increased volume as well as tov are lower with more asisst.

On the other hand Curry is way worse shooter without Green, his 3p% is down by almost 8 percentge points (so maybe we should start talking about how having elite playmaker like Green helps Curry?), tov are also higher, with the same amount of ast, but bigger volume.

And Green's +/- w/o Curry is better than Steph's w/o Draymond.

So maybe instead of using small sample excuse (while BTW these samples are not so small, more than 10 full games) try to find resonable explanation why results over and over again favour Green. And by resonable I mean something better than "GSW have more plays w/o Green" - as there is no way we can check if that's true or not.

Since this came right after mine I'm gonna assume you are referring to me.

1.

Again, this was me using the-powers third point in his long post a few pages ago. I can't remember what it said but it said something about how using it in spots vs using it for a whole game and how sustainable it is.


2.

Curry in this sample size attempted 88 threes in 241 minutes.
That's 13.1 threes per 36 minutes, he is averaging 11.8 per 36 this year
As for sample size

Corner threes. 7 attempts, 2/7 (keep in mind, he Has a career average of 52.5% from there)

24-29 feet
68 attempts, 41.2% 28/68

30+ feet
18 attempts, 30%
6/18

For the season

48.2% from the corner
44.8% from 24-29 feet
46.7% from 30+ feet

So the corner, he shot 7. That's sample size (and no, he isn't shooting less of them than usual either)
From normal three point range it's a 3.5% drop
From 30+ feet it's a 16.7% drop
So unless green is getting him open looks from 30 feet I don't think it's as big of a drop as you are saying.

Curry's turnover percentage is 11%, lower than his season average
Assist rate is 35.3%, slightly higher than his season average
As for sample size,

Curry with Bogut on the floor and draymond off 1.18 defensive rtg
Curry with Bogut off the floor and draymond off 1.063 defensive rtg

So either Bogut is the worst defender in the NBA
Or it's sample size


Adding on to that a lot of people have said curry hasn't played with good centers
In 61 minutes with ezili

Off rtg 117.1
Def rtg 95.3

And in 106 minutes with klay

117.5 off rtg
110.3 def rtg