Durant is great defensively by some Vantage metrics. Most notably, he is second in the league in points allowed per shot at 0.87 (Green is first at 0.86). He's better than Green and Kawhi on points allowed per screen as well. The only (published) area he is lacking in is turnovers forced. And yet, DRAPM tells a very different story, with Durant not coming close to Green or Kawhi. The most likely reasons for this would be:
(1) Vantage stats are rate stats. That is, volume has not yet been accounted for, and Kawhi and Green presumably have higher "defensive volume".
(2) Green and Kawhi are given tougher assignments which deflate their Vantage stats.
(3) Forcing turnovers has a very large defensive impact if the player isn't gambling.
Any one else have ideas on why Durant shows up so much better under Vantage stats?
Durant is a monster at isolation defense, I think everything supports that.
Lol playoffs have been crazy rapm wise. Curry looked a bit slow to me when he came back, though that just might be me coming up with excuses.
I think I have misunderstood what everyone has been saying, I thought that people were straight up saying draymond>Curry. Impact wise my thing with rapm is I'm not sure if it calculates for fit (the sample size for green with klay is something I think about, without is too small to make conclusions but there are "outliers" in that small sample so I don't take the results seriously, though they favor green)
I honestly do see an arguement that green, in terms of pure minute impact, could be more impactful than curry on this Warriors team, but I don't think that makes him more important, you know what I mean?
On a sidenote, I don't really understand why curry having or not having a higher rapm than green effects him. I mean, JE himself said that the difference isn't reliable. Now, when I read this, what I think is that this means that even rapm can't keep up with the lineup differences. (Or lack of it)
Basically, curry gets some defensive credit because of green, and green gets offensive credit because of curry.
Atmanne wrote:Why does it keep updating if the season's been over for a while? Apparently, it's not including the playoffs because the games played remain the same.
No it is including playoff games, its just not including them in the game count (this is how its been in all 3 seasons). All past complete RPM/RAPM/APM sets include playoff data.
But in his home dwelling...the hi-top faded warrior is revered. *Smack!* The sound of his palm blocking the basketball... the sound of thousands rising, roaring... the sound of "get that sugar honey iced tea outta here!"
Atmanne wrote:Why does it keep updating if the season's been over for a while? Apparently, it's not including the playoffs because the games played remain the same.
No it is including playoff games, its just not including them in the game count (this is how its been in all 3 seasons). All past complete RPM/RAPM/APM sets include playoff data.
I see. Well, it's odd that the game count doesn't change.
What is the main source people go to for RAPM? I always hear people mention it but there are different versions out there.
Is it for the entire season or just the regular season?
Jeremias Engelmann (primary RAPM/RPM guy) posts updates on his twitter and on the APBR board (usually at fpliii and my requests ). I try my best to post every update in the yearly RAPM threads and try to label them as accurately as possible to differentiate between the variations (Single-year vs. multi-year; prior-informed vs. vanilla; etc.).
Any complete set from the past includes Regular Season & Playoffs, current year it usually up-to-date by the week.
Here was my brief-ish summary of the different versions:
Spoiler:
[#1]APM is Simple OLS, exactly as Spaceman described it. Set up every 5on5 matchup, set equal to the scoring margin and solve for each player across the league (I've run it for a few years, its like ~65000 lines of 5on5 matchups). The resulting coefficients (on each player) are the APM values. This needs a very large sample size to say anything of considerable meaning; a single-year APM has large error terms on each coefficient, multi-year (usually 2-year) studies are preferred.
RAPM is essentially the same thing (OLS) with one exception. It introduces what we'll call a "reference matrix", basically each player is given a baseline value, towards which their coefficient will be pulled. I believe this tries to reduce the multicollinearity problem.
In [#2A]vanilla/basic RAPM, every value in the reference matrix is set to 0. The greater the amount of games played, the less weight that reference of 0 has. It is almost the same as APM [#1], but the regression towards 0 in theory reduces the error within a single-season set. It's still fairly volatile, but it's better than APM [#1] is within a single year. There is also [#2B]multi-year RAPM, which just uses a larger number of seasons, with most weight given to the current year and less and less weight given to previous years, reference matrix of 0s.
[#3]Prior-informed RAPM is essentially the best (ITO out-of-sample prediction) version of this family without introducing the box-score. It's built the same way as RAPM, but the reference matrix uses RAPM values from the previous year, instead of all players being set at 0. Again, as the sample size of the season grows, the reference value holds less and less weight. Obviously this only works once we have multiple years of data, in the 1st year, there is no prior. In the 2nd, we have a prior but it is vanilla RAPM [#2A], but by the 3rd year we can use PI RAPM of the previous year to inform the current year.
[#4A]RPM is RAPM, but the reference matrix is made up of SPM values (SPM is again, regression of box-score metrics on a multi-year non-box-score model such as RAPM). There is also [#4B]multi-year RPM, which is the same as multi-year RAPM, except it presumably uses a reference matrix of multi-year SPM values.
There is also [#5]prior informed RPM. Again, same idea as PI RAPM [#3], single-year, reference matrix of prior-year's RPM values.
I don't want to open the can of worms that is the financial collapse, but as far as I understand, it's more along the lines of knowingly turning a blind-eye to what were known as poor models, rather than them just being present/utilized.
But in his home dwelling...the hi-top faded warrior is revered. *Smack!* The sound of his palm blocking the basketball... the sound of thousands rising, roaring... the sound of "get that sugar honey iced tea outta here!"
SideshowBob wrote:Jeremias Engelmann (primary RAPM/RPM guy) posts updates on his twitter and on the APBR board (usually at fpliii and my requests ). I try my best to post every update in the yearly RAPM threads and try to label them as accurately as possible to differentiate between the variations (Single-year vs. multi-year; prior-informed vs. vanilla; etc.).
Any complete set from the past includes Regular Season & Playoffs, current year it usually up-to-date by the week.
Here was my brief-ish summary of the different versions:
Spoiler:
[#1]APM is Simple OLS, exactly as Spaceman described it. Set up every 5on5 matchup, set equal to the scoring margin and solve for each player across the league (I've run it for a few years, its like ~65000 lines of 5on5 matchups). The resulting coefficients (on each player) are the APM values. This needs a very large sample size to say anything of considerable meaning; a single-year APM has large error terms on each coefficient, multi-year (usually 2-year) studies are preferred.
RAPM is essentially the same thing (OLS) with one exception. It introduces what we'll call a "reference matrix", basically each player is given a baseline value, towards which their coefficient will be pulled. I believe this tries to reduce the multicollinearity problem.
In [#2A]vanilla/basic RAPM, every value in the reference matrix is set to 0. The greater the amount of games played, the less weight that reference of 0 has. It is almost the same as APM [#1], but the regression towards 0 in theory reduces the error within a single-season set. It's still fairly volatile, but it's better than APM [#1] is within a single year. There is also [#2B]multi-year RAPM, which just uses a larger number of seasons, with most weight given to the current year and less and less weight given to previous years, reference matrix of 0s.
[#3]Prior-informed RAPM is essentially the best (ITO out-of-sample prediction) version of this family without introducing the box-score. It's built the same way as RAPM, but the reference matrix uses RAPM values from the previous year, instead of all players being set at 0. Again, as the sample size of the season grows, the reference value holds less and less weight. Obviously this only works once we have multiple years of data, in the 1st year, there is no prior. In the 2nd, we have a prior but it is vanilla RAPM [#2A], but by the 3rd year we can use PI RAPM of the previous year to inform the current year.
[#4A]RPM is RAPM, but the reference matrix is made up of SPM values (SPM is again, regression of box-score metrics on a multi-year non-box-score model such as RAPM). There is also [#4B]multi-year RPM, which is the same as multi-year RAPM, except it presumably uses a reference matrix of multi-year SPM values.
There is also [#5]prior informed RPM. Again, same idea as PI RAPM [#3], single-year, reference matrix of prior-year's RPM values.
I don't want to open the can of worms that is the financial collapse, but as far as I understand, it's more along the lines of knowingly turning a blind-eye to what were known as poor models, rather than them just being present/utilized.
SideshowBob wrote:Jeremias Engelmann (primary RAPM/RPM guy) posts updates on his twitter and on the APBR board (usually at fpliii and my requests ). I try my best to post every update in the yearly RAPM threads and try to label them as accurately as possible to differentiate between the variations (Single-year vs. multi-year; prior-informed vs. vanilla; etc.).
Any complete set from the past includes Regular Season & Playoffs, current year it usually up-to-date by the week.
Here was my brief-ish summary of the different versions:
Spoiler:
[#1]APM is Simple OLS, exactly as Spaceman described it. Set up every 5on5 matchup, set equal to the scoring margin and solve for each player across the league (I've run it for a few years, its like ~65000 lines of 5on5 matchups). The resulting coefficients (on each player) are the APM values. This needs a very large sample size to say anything of considerable meaning; a single-year APM has large error terms on each coefficient, multi-year (usually 2-year) studies are preferred.
RAPM is essentially the same thing (OLS) with one exception. It introduces what we'll call a "reference matrix", basically each player is given a baseline value, towards which their coefficient will be pulled. I believe this tries to reduce the multicollinearity problem.
In [#2A]vanilla/basic RAPM, every value in the reference matrix is set to 0. The greater the amount of games played, the less weight that reference of 0 has. It is almost the same as APM [#1], but the regression towards 0 in theory reduces the error within a single-season set. It's still fairly volatile, but it's better than APM [#1] is within a single year. There is also [#2B]multi-year RAPM, which just uses a larger number of seasons, with most weight given to the current year and less and less weight given to previous years, reference matrix of 0s.
[#3]Prior-informed RAPM is essentially the best (ITO out-of-sample prediction) version of this family without introducing the box-score. It's built the same way as RAPM, but the reference matrix uses RAPM values from the previous year, instead of all players being set at 0. Again, as the sample size of the season grows, the reference value holds less and less weight. Obviously this only works once we have multiple years of data, in the 1st year, there is no prior. In the 2nd, we have a prior but it is vanilla RAPM [#2A], but by the 3rd year we can use PI RAPM of the previous year to inform the current year.
[#4A]RPM is RAPM, but the reference matrix is made up of SPM values (SPM is again, regression of box-score metrics on a multi-year non-box-score model such as RAPM). There is also [#4B]multi-year RPM, which is the same as multi-year RAPM, except it presumably uses a reference matrix of multi-year SPM values.
There is also [#5]prior informed RPM. Again, same idea as PI RAPM [#3], single-year, reference matrix of prior-year's RPM values.
I don't want to open the can of worms that is the financial collapse, but as far as I understand, it's more along the lines of knowingly turning a blind-eye to what were known as poor models, rather than them just being present/utilized.
But in his home dwelling...the hi-top faded warrior is revered. *Smack!* The sound of his palm blocking the basketball... the sound of thousands rising, roaring... the sound of "get that sugar honey iced tea outta here!"
First off, I'm loving that the conversation about Draymond Green is advancing beyond the "how can a glorified role player be that good" level that it rested for much of the year, and is starting to move into deeper examinations of his strengths and weaknesses and how his impact might really be super-star level in its own right. Green is rapidly developing into one of my favorite players, as I really enjoy the way that he approaches the game. And as an analyst, I really like that he's making his impact in non-traditional ways and making us think.
therealbig3 wrote:Ok, but then this is where I'm at specifically with a Draymond Green vs Kevin Garnett comparison:
Spoiler:
My opinion on Kevin Garnett has shifted quite dramatically since I've joined, and I've got him quite high in my personal rankings now, and the realization I came to was:
Kevin Garnett is not an elite 1 on 1 scorer, and he certainly falls short of someone like Tim Duncan in this regard. However, unless you are literally one of the handful of super elite big man scorers (Shaq, Barkley, Dirk, Kareem, Hakeem), which Tim Duncan certainly was not, it just doesn't make much sense to use your big man as a volume scorer...it's a last resort that's just not very efficient offense. Duncan's Spurs during his prime were merely ok on offense, and they got better once Parker and Manu got more involved. The best Spurs offenses have featured Duncan as a finisher on offense, rather than an initiator.
And this is what got me thinking about the ideal non-elite big man scorer...what should they excel in? And this is where KG shines, and why I've become so high on him. Because when you start thinking about using most great big men optimally, KG trumps pretty much everyone.
You want great defense first and foremost...KG is the best defensive player I've ever seen...this is supported by a lot of +/- data, as well as qualitative analysis that points to the conclusion that yes, KG is a monster defensively, and is certainly as good as anyone post-Russell, at least.
Offensively, if you're not going to run the offense through him as a volume scorer, then this big man should be able to complement great offense in every way. This is done through being able to handle the ball, pass, shoot, and finish...he'd be providing dynamic playmaking, versatility, spacing, and offensive gravity towards the rim. Again, KG did this as well as anybody. So for me, his non-elite 1 on 1 scoring is NOT a big deal to me, because except for some rare exceptions, most big men aren't elite 1 on 1 scorers and shouldn't be relied on to do so for volume. This is why KG has an advantage over the Tim Duncans and the Karl Malones and the David Robinsons...when used optimally, none of those guys can provide what KG does, and this largely explains why KG was a +/- king, even towards the end of his career in Boston.
But then we get to Draymond Green. He's quite close to KG defensively. He's probably the closest thing we've got actually. Insane versatility, quickness, length, and instincts. He's been the anchor of one of the league's best defenses for the last 2 years. Can't say enough about him defensively.
And then we get to offense, and basically all of those things that I give credit to KG for in terms of being the ultimate "glue guy" and complementing the offense in every way he can outside of 1 on 1 scoring...Green does all of those things as good or better. He's a better shooter. He's a better ball handler and passer, and thus a more dynamic playmaker both in transition and in the half court. He's got insane gravity moving towards the rim.
The one thing on offense that peak KG has over Green is...1 on 1 scoring. Is that really something to give much credit for when KG's entire argument against everyone else is that you don't need him to be a 1 on 1 scorer? Honestly, it's pretty irrelevant to me, because KG isn't best used that way. And at the other non-scoring things on offense...Green is just better. And defensively, I would give KG the edge based on pure size, but then again, Green is even more versatile, because he can legitimately defend guards for extended periods of time, while the best of KG was more so about being able to keep up with them on switches, not necessarily play them as well as a primary defender.
So...why exactly peak KG over Draymond? Seems to me that when you look at everything besides 1 on 1 scoring, which was never KG's strength and isn't really the reason why so many people consider him amazing, Green matches up very favorably to Garnett, with some clear advantages in terms of shooting and playmaking. And to consider Green over Garnett seems like heresy at first, but to me, it's almost like Green is a KG that dialed back the scoring even more and improved on certain key areas that make him even more useful as an offensive piece.
And the RAPM isn't exactly a landslide in KG's favor here either, which means that it might not be so crazy to feel that Green matches up just fine...
04 KG: +8.6 08 KG: +8.1
16 Green: +7.93
And if he matches up to peak/prime KG as the uber-portable superstar that can make some really great things happen when surrounded with some talent, he can technically match up with almost anyone, since KG had one of the greatest peaks of all time.
This is an interesting, and IMO a natural comparison to make. Green is one of the closest that I've seen in terms of impacting the game across-the-board the way that Garnett does, and they also appear to share a lot of intangible similarities such as the way that they approach leadership and the way they use trash-talking as a weapon. As far as your particular points, this is how I'd weigh in:
1) Defense. As you pointed out, Draymond is a defensive beast and a lot of that comes from some of the things that made Garnett special defensively. Green's lateral mobility, his defensive intelligence, his help defense, his vocal defense and his rotations are all very strong and his length is deceptive for a man his height. However, I don't get the sense that he's really on the level of Garnett at his best, defensively. Garnett excelled in all of those same things defensively, but arguably to a larger extent than Green. On top of that, Garnett did this at 7-1 with stupidly long arms. That really can't be overstated. Green's deceptive length allows someone with traditional small forward size to operate as a 4/5 in today's game. But Garnett's obvious length was standard deviations beyond the norm, even for a 4/5, and allows him to change the game defensively both out on the court and near the rim in ways that Green isn't equipped to do. Someone (was it you?) posted the defensive RAPM scores for Duncan and Garnett up until 2007 in a thread, and Green measured out somewhere near the middle of that grouping which is outstanding for anyone. However, that still put him solidly behind the best defensive scores that we saw from Garnett in Minnesota, and WAY behind the scores he was putting up in Boston. While Green is an excellent defender...arguably the best in the current NBA, it'd be a mistake IMO to even intimate any type of equality or even really any sort of legitimate comparison with Garnett's level.
2) Offense. This is the more interesting area to compare, because as you pointed out it can be argued that Green is in some ways an offensive evolution for the way that Garnett played offense: Garnett was the closest thing to a point-big man that we've ever really seen. While Green, on the other hand, I'd argue actually is the real point guard for the Warriors. Getting point guard-type contributions from a big man is, as I've pointed out before, one of the tried-and-true methods for getting a mega offensive impact out of a big. And I'd agree that Green, being more effective at contributing point guard skills than any big man ever, is better suited to making a large impact in that particular way than Garnett (or any other big).
However, I do think that your stated approach (e.g. if not the volume scoring hub, individual scoring is irrelevant) is, to me, an oversimplification here. You state that Garnett as an iso scorer isn't as good as Duncan's, but since Garnett is better at the other types of offense you put him over Duncan, so you feel you should do the same for Green. However, any differences between Garnett and Duncan as effective iso scorers is razor thin and very much in the noise. I've posted on this many times, but over their decade-long primes Garnett and Duncan averaged almost exactly the same scoring volume and efficiency over the regular season (down to the decimal point) over 800+ games. And in the postseason, if you acknowledge in any way the calibers of their teams, the calibers of their opponents, and their performances against common opponents it is extremely arguable that Garnett and Duncan are also extremely similar in practice as iso scorers in the post-season as well. This is another topic, and not relevant enough that I think we should re-examine it in too much detail here, so suffice it to say that even if you consider Duncan to be better, I think that, conservatively, there is not a large degree of difference between Garnett and Duncan as iso scorers.
Green, on the other hand, is significantly behind both of them as iso scorers. And that is significant, because it limits the number of situations in which he could equal Garnett's level of offensive impact. You pointed out that if your big man isn't a dominant offensive player then you'd want him to be able to do the things to support a great offensive player, and I can agree with that to some extent. However, in Green's case, the reason that he's having such a mega offensive impact isn't primarily because of his spacing (which is nice), but because he actually IS the point guard. And this impact is maximized on the Warriors, because they actually NEED him to play point guard to allow Steph (and Klay) to optimize their volume shooting/scoring game.
However, on any team with a traditional point guard, Green's level of offensive impact would be lessened. He'd still be an offensive plus, as a spacing big is almost always a positive, as is a big that can facilitate ball movement. Those are always good skills to have. But if he's playing with Nash...or Kidd...or Chris Paul...or Dre' Miller...or John Wall...or really ANY traditional point guard that isn't maximized as an off-ball threat, then Draymond loses what makes his offensive impact the most special. And he doesn't have the counter-move of being able to operate as an effective volume scorer (the way that both Garnett and Duncan demonstrably were able to do, on championship caliber teams). This is a clear weakness for him.
Similarly, if Green were on a team that needed him to be a volume scorer...which isn't really a unique situation nor one that precludes successful basketball, Green isn't equipped. Duncan and Garnett have both been the leading scorers on multiple playoff runs that extended to at least the Conference Finals and many times into championships. As far as Duncan, not only could Green not have provided sufficient offense for the 2003 Spurs, he also couldn't have provided sufficient offense for any of the other Spurs teams from 1999 - 2008. While (early) Robinson or (later) Ginobili/Parker were strong contributors to the Spurs' offense, they worked as a team offense because Duncan's ability to volume score complimented their skills. While Green's facilitation is a positive, it wouldn't have been sufficient in that example. And the same is true of Garnett's Wolves and Celtics. The Wolves teams that were winning 50+ games regularly absolutely required Garnett to be the primary scorer in addition to his facilitation skills, and pushing that combo further towards facilitation without the iso scoring punch would have been a huge negative on those teams. The '04 Wolves were a Cassell injury away from a very legitimate shot at a title built around Garnett's ability to carry a mega offensive load...with Green in his place, there is a dramatic falloff for those Wolves that would have made it a challenge for them to make the playoffs out West. And this speaks to...
3) Portability and scaleability: Green has a game that, depending on how you define the terms, can be considered very portable and very scaleable. It's portable because things like elite defense, spacing, ball-movement, pick-setting, leadership and high-motor are valuable in any context. Similarly, his impact can be considered scaleable because he is currently producing very arguably the highest level of impact in the league on the best team in the league. Using the words in those ways, his game is very portable and scaleable.
However, when compared to Garnett, it eventually shows up that this isn't a complete usage of the terms. It's somewhat analogous to how Green is having an elite defensive impact compared to his peers, but not when compared to the best of Garnett. Because in Garnett's case, his portability and scaleability were for remaining the biggest impact in the league (with some of the highest marks ever recorded) across a huge variety of circumstances. Garnett led the NBA in overall RAPM in 2003 on a team that required him to be a unipolar offensive creator (heavy on both team creation and individual scoring creation) and the defensive anchor (one of the better in the league). In 2004 the situation was completely different...he was the leading scorer in a more bi-polar offense (more heavily as a finisher, still with a good amount of team creation) and even more of a defensive anchor (arguably the best in the league, in a league that included peak all-history defenders like Big Ben and Duncan as the competition)...and he was still putting up best-in-the-NBA (and near best in recorded history) RAPM scorers. Then, by 2008, Garnett was in a completely different and in many ways opposite situation, where he was called upon to be strong on offense but in more of a finishing role (still led a championship team in scoring on their postseason march, finished top-20 in NBA in offensive RAPM) but had to amp up his individual defensive impact into the all-history level...and again he did this, putting up the best RAPM score in the league.
Draymond's skillset would have positive utility in almost any situation, which makes his game very portable. And his skillset, in his current situation, gives him arguably the best impact in the league on the best team in the league, which connotes scaleable. But what hasn't been demonstrated, and what (when compared to Garnett's game) seems unlikely, is that Draymond would be able to maintain his argument as best impact in the league under such a variety of situations. As pointed out above, he's a strong defender, but I don't see his defense as scaleable up to all-history level if the situation requires it. Draymond's spacing impact would always be a positive, but if he's not required to be the team's the point guard I don't think his offensive impact remains at nearly the level that he currently enjoys. Someone (RonnyMac?) brought up an Horry comparison for Green that, if he were not the point guard, would be more apt. Horry made a career of having good +/- scorers with that spacing, but he wasn't fighting for the top of the league. And again, if Green were called upon to be the primary scoring threat for a team which, again, Garnett (or Duncan) was asked to do repeatedly on contending-caliber teams, I see no evidence that he'd be able to answer that call.
4) Bottom line. So, yeah. I absolutely champion Green as one of the biggest impact players in the NBA right now. I think he deserves much more credit for the Warriors' success than he's received to date, that he is very arguably as valuable (possibly more valuable) to what the Warriors are doing than Curry is. That he'd be a big positive on any team in the NBA. And that he should be getting more talk in the MVP vote as opposed to being "just" an All Star level sidekick, which is what he is currently most often considered. Green is fast developing into one of the most fun players in the NBA.
MyUniBroDavis wrote: JE himself said that the difference isn't reliable.
He said: "Not very reliable, but the difference is *relatively* large. Far from "statistical significant" though."
So we can't draw definitive conclusion based on RAPM alone, but it is important piece of information, which we can't ignore if it's consistent with other data points.
MyUniBroDavis wrote: JE himself said that the difference isn't reliable.
He said: "Not very reliable, but the difference is *relatively* large. Far from "statistical significant" though."
So we can't draw definitive conclusion based on RAPM alone, but it is important piece of information, which we can't ignore if it's consistent with other data points.
Isn't the only other data point basically raw plus minus, in terms of calculation, since the data used is the same.
drza wrote:First off, I'm loving that the conversation about Draymond Green is advancing beyond the "how can a glorified role player be that good" level that it rested for much of the year, and is starting to move into deeper examinations of his strengths and weaknesses and how his impact might really be super-star level in its own right. Green is rapidly developing into one of my favorite players, as I really enjoy the way that he approaches the game. And as an analyst, I really like that he's making his impact in non-traditional ways and making us think.
therealbig3 wrote:Ok, but then this is where I'm at specifically with a Draymond Green vs Kevin Garnett comparison:
Spoiler:
My opinion on Kevin Garnett has shifted quite dramatically since I've joined, and I've got him quite high in my personal rankings now, and the realization I came to was:
Kevin Garnett is not an elite 1 on 1 scorer, and he certainly falls short of someone like Tim Duncan in this regard. However, unless you are literally one of the handful of super elite big man scorers (Shaq, Barkley, Dirk, Kareem, Hakeem), which Tim Duncan certainly was not, it just doesn't make much sense to use your big man as a volume scorer...it's a last resort that's just not very efficient offense. Duncan's Spurs during his prime were merely ok on offense, and they got better once Parker and Manu got more involved. The best Spurs offenses have featured Duncan as a finisher on offense, rather than an initiator.
And this is what got me thinking about the ideal non-elite big man scorer...what should they excel in? And this is where KG shines, and why I've become so high on him. Because when you start thinking about using most great big men optimally, KG trumps pretty much everyone.
You want great defense first and foremost...KG is the best defensive player I've ever seen...this is supported by a lot of +/- data, as well as qualitative analysis that points to the conclusion that yes, KG is a monster defensively, and is certainly as good as anyone post-Russell, at least.
Offensively, if you're not going to run the offense through him as a volume scorer, then this big man should be able to complement great offense in every way. This is done through being able to handle the ball, pass, shoot, and finish...he'd be providing dynamic playmaking, versatility, spacing, and offensive gravity towards the rim. Again, KG did this as well as anybody. So for me, his non-elite 1 on 1 scoring is NOT a big deal to me, because except for some rare exceptions, most big men aren't elite 1 on 1 scorers and shouldn't be relied on to do so for volume. This is why KG has an advantage over the Tim Duncans and the Karl Malones and the David Robinsons...when used optimally, none of those guys can provide what KG does, and this largely explains why KG was a +/- king, even towards the end of his career in Boston.
But then we get to Draymond Green. He's quite close to KG defensively. He's probably the closest thing we've got actually. Insane versatility, quickness, length, and instincts. He's been the anchor of one of the league's best defenses for the last 2 years. Can't say enough about him defensively.
And then we get to offense, and basically all of those things that I give credit to KG for in terms of being the ultimate "glue guy" and complementing the offense in every way he can outside of 1 on 1 scoring...Green does all of those things as good or better. He's a better shooter. He's a better ball handler and passer, and thus a more dynamic playmaker both in transition and in the half court. He's got insane gravity moving towards the rim.
The one thing on offense that peak KG has over Green is...1 on 1 scoring. Is that really something to give much credit for when KG's entire argument against everyone else is that you don't need him to be a 1 on 1 scorer? Honestly, it's pretty irrelevant to me, because KG isn't best used that way. And at the other non-scoring things on offense...Green is just better. And defensively, I would give KG the edge based on pure size, but then again, Green is even more versatile, because he can legitimately defend guards for extended periods of time, while the best of KG was more so about being able to keep up with them on switches, not necessarily play them as well as a primary defender.
So...why exactly peak KG over Draymond? Seems to me that when you look at everything besides 1 on 1 scoring, which was never KG's strength and isn't really the reason why so many people consider him amazing, Green matches up very favorably to Garnett, with some clear advantages in terms of shooting and playmaking. And to consider Green over Garnett seems like heresy at first, but to me, it's almost like Green is a KG that dialed back the scoring even more and improved on certain key areas that make him even more useful as an offensive piece.
And the RAPM isn't exactly a landslide in KG's favor here either, which means that it might not be so crazy to feel that Green matches up just fine...
04 KG: +8.6 08 KG: +8.1
16 Green: +7.93
And if he matches up to peak/prime KG as the uber-portable superstar that can make some really great things happen when surrounded with some talent, he can technically match up with almost anyone, since KG had one of the greatest peaks of all time.
This is an interesting, and IMO a natural comparison to make. Green is one of the closest that I've seen in terms of impacting the game across-the-board the way that Garnett does, and they also appear to share a lot of intangible similarities such as the way that they approach leadership and the way they use trash-talking as a weapon. As far as your particular points, this is how I'd weigh in:
1) Defense. As you pointed out, Draymond is a defensive beast and a lot of that comes from some of the things that made Garnett special defensively. Green's lateral mobility, his defensive intelligence, his help defense, his vocal defense and his rotations are all very strong and his length is deceptive for a man his height. However, I don't get the sense that he's really on the level of Garnett at his best, defensively. Garnett excelled in all of those same things defensively, but arguably to a larger extent than Green. On top of that, Garnett did this at 7-1 with stupidly long arms. That really can't be overstated. Green's deceptive length allows someone with traditional small forward size to operate as a 4/5 in today's game. But Garnett's obvious length was standard deviations beyond the norm, even for a 4/5, and allows him to change the game defensively both out on the court and near the rim in ways that Green isn't equipped to do. Someone (was it you?) posted the defensive RAPM scores for Duncan and Garnett up until 2007 in a thread, and Green measured out somewhere near the middle of that grouping which is outstanding for anyone. However, that still put him solidly behind the best defensive scores that we saw from Garnett in Minnesota, and WAY behind the scores he was putting up in Boston. While Green is an excellent defender...arguably the best in the current NBA, it'd be a mistake IMO to even intimate any type of equality or even really any sort of legitimate comparison with Garnett's level.
2) Offense. This is the more interesting area to compare, because as you pointed out it can be argued that Green is in some ways an offensive evolution for the way that Garnett played offense: Garnett was the closest thing to a point-big man that we've ever really seen. While Green, on the other hand, I'd argue actually is the real point guard for the Warriors. Getting point guard-type contributions from a big man is, as I've pointed out before, one of the tried-and-true methods for getting a mega offensive impact out of a big. And I'd agree that Green, being more effective at contributing point guard skills than any big man ever, is better suited to making a large impact in that particular way than Garnett (or any other big).
However, I do think that your stated approach (e.g. if not the volume scoring hub, individual scoring is irrelevant) is, to me, an oversimplification here. You state that Garnett as an iso scorer isn't as good as Duncan's, but since Garnett is better at the other types of offense you put him over Duncan, so you feel you should do the same for Green. However, any differences between Garnett and Duncan as effective iso scorers is razor thin and very much in the noise. I've posted on this many times, but over their decade-long primes Garnett and Duncan averaged almost exactly the same scoring volume and efficiency over the regular season (down to the decimal point) over 800+ games. And in the postseason, if you acknowledge in any way the calibers of their teams, the calibers of their opponents, and their performances against common opponents it is extremely arguable that Garnett and Duncan are also extremely similar in practice as iso scorers in the post-season as well. This is another topic, and not relevant enough that I think we should re-examine it in too much detail here, so suffice it to say that even if you consider Duncan to be better, I think that, conservatively, there is not a large degree of difference between Garnett and Duncan as iso scorers.
Green, on the other hand, is significantly behind both of them as iso scorers. And that is significant, because it limits the number of situations in which he could equal Garnett's level of offensive impact. You pointed out that if your big man isn't a dominant offensive player then you'd want him to be able to do the things to support a great offensive player, and I can agree with that to some extent. However, in Green's case, the reason that he's having such a mega offensive impact isn't primarily because of his spacing (which is nice), but because he actually IS the point guard. And this impact is maximized on the Warriors, because they actually NEED him to play point guard to allow Steph (and Klay) to optimize their volume shooting/scoring game.
However, on any team with a traditional point guard, Green's level of offensive impact would be lessened. He'd still be an offensive plus, as a spacing big is almost always a positive, as is a big that can facilitate ball movement. Those are always good skills to have. But if he's playing with Nash...or Kidd...or Chris Paul...or Dre' Miller...or John Wall...or really ANY traditional point guard that isn't maximized as an off-ball threat, then Draymond loses what makes his offensive impact the most special. And he doesn't have the counter-move of being able to operate as an effective volume scorer (the way that both Garnett and Duncan demonstrably were able to do, on championship caliber teams). This is a clear weakness for him.
Similarly, if Green were on a team that needed him to be a volume scorer...which isn't really a unique situation nor one that precludes successful basketball, Green isn't equipped. Duncan and Garnett have both been the leading scorers on multiple playoff runs that extended to at least the Conference Finals and many times into championships. As far as Duncan, not only could Green not have provided sufficient offense for the 2003 Spurs, he also couldn't have provided sufficient offense for any of the other Spurs teams from 1999 - 2008. While (early) Robinson or (later) Ginobili/Parker were strong contributors to the Spurs' offense, they worked as a team offense because Duncan's ability to volume score complimented their skills. While Green's facilitation is a positive, it wouldn't have been sufficient in that example. And the same is true of Garnett's Wolves and Celtics. The Wolves teams that were winning 50+ games regularly absolutely required Garnett to be the primary scorer in addition to his facilitation skills, and pushing that combo further towards facilitation without the iso scoring punch would have been a huge negative on those teams. The '04 Wolves were a Cassell injury away from a very legitimate shot at a title built around Garnett's ability to carry a mega offensive load...with Green in his place, there is a dramatic falloff for those Wolves that would have made it a challenge for them to make the playoffs out West. And this speaks to...
3) Portability and scaleability: Green has a game that, depending on how you define the terms, can be considered very portable and very scaleable. It's portable because things like elite defense, spacing, ball-movement, pick-setting, leadership and high-motor are valuable in any context. Similarly, his impact can be considered scaleable because he is currently producing very arguably the highest level of impact in the league on the best team in the league. Using the words in those ways, his game is very portable and scaleable.
However, when compared to Garnett, it eventually shows up that this isn't a complete usage of the terms. It's somewhat analogous to how Green is having an elite defensive impact compared to his peers, but not when compared to the best of Garnett. Because in Garnett's case, his portability and scaleability were for remaining the biggest impact in the league (with some of the highest marks ever recorded) across a huge variety of circumstances. Garnett led the NBA in overall RAPM in 2003 on a team that required him to be a unipolar offensive creator (heavy on both team creation and individual scoring creation) and the defensive anchor (one of the better in the league). In 2004 the situation was completely different...he was the leading scorer in a more bi-polar offense (more heavily as a finisher, still with a good amount of team creation) and even more of a defensive anchor (arguably the best in the league, in a league that included peak all-history defenders like Big Ben and Duncan as the competition)...and he was still putting up best-in-the-NBA (and near best in recorded history) RAPM scorers. Then, by 2008, Garnett was in a completely different and in many ways opposite situation, where he was called upon to be strong on offense but in more of a finishing role (still led a championship team in scoring on their postseason march, finished top-20 in NBA in offensive RAPM) but had to amp up his individual defensive impact into the all-history level...and again he did this, putting up the best RAPM score in the league.
Draymond's skillset would have positive utility in almost any situation, which makes his game very portable. And his skillset, in his current situation, gives him arguably the best impact in the league on the best team in the league, which connotes scaleable. But what hasn't been demonstrated, and what (when compared to Garnett's game) seems unlikely, is that Draymond would be able to maintain his argument as best impact in the league under such a variety of situations. As pointed out above, he's a strong defender, but I don't see his defense as scaleable up to all-history level if the situation requires it. Draymond's spacing impact would always be a positive, but if he's not required to be the team's the point guard I don't think his offensive impact remains at nearly the level that he currently enjoys. Someone (RonnyMac?) brought up an Horry comparison for Green that, if he were not the point guard, would be more apt. Horry made a career of having good +/- scorers with that spacing, but he wasn't fighting for the top of the league. And again, if Green were called upon to be the primary scoring threat for a team which, again, Garnett (or Duncan) was asked to do repeatedly on contending-caliber teams, I see no evidence that he'd be able to answer that call.
4) Bottom line. So, yeah. I absolutely champion Green as one of the biggest impact players in the NBA right now. I think he deserves much more credit for the Warriors' success than he's received to date, that he is very arguably as valuable (possibly more valuable) to what the Warriors are doing than Curry is. That he'd be a big positive on any team in the NBA. And that he should be getting more talk in the MVP vote as opposed to being "just" an All Star level sidekick, which is what he is currently most often considered. Green is fast developing into one of the most fun players in the NBA.
But he's still not KG.
I'm literally tempted to get all of ur Garnett related posts and put them in a group and add it on a shrine and send it to Garnett/spam his insta with ir or something.
But in his home dwelling...the hi-top faded warrior is revered. *Smack!* The sound of his palm blocking the basketball... the sound of thousands rising, roaring... the sound of "get that sugar honey iced tea outta here!"
Not a good sign he said RAPM is "real adjusted plus minus". Kinda makes it seem uninformed.
MyUniBroDavis wrote: he was like YALL PEOPLE WHO DOUBT ME WILL SEE YALLS STATS ARE WRONG I HAVE THE BIG BRAIN PLAYS MUCHO NASTY BIG BRAIN BIG CHUNGUS BRAIN YOU BOYS ON UR BBALL REFERENCE NO UNDERSTANDO
bondom34 wrote:Not a good sign he said RAPM is "real adjusted plus minus". Kinda makes it seem uninformed.
Yeah that jumped at me immediately. Plus a lot of references to Dave Berri, which is sad because I'm pretty sure that dude is the epitome of regressive stat-centric bball analysis.
But in his home dwelling...the hi-top faded warrior is revered. *Smack!* The sound of his palm blocking the basketball... the sound of thousands rising, roaring... the sound of "get that sugar honey iced tea outta here!"
My god. This is comically bad. I'm enjoying the fact that Berri and Chaikin are now tied to this amateur. Back when I first joined RealGM I also joined APBRmetrics, and back then it was dominated by guys like them who were insistent you could basically tell everything you needed from box score stats and that +/- stats didn't make any sense.
Fast forward through the years and you see that +/- stats came to not only dominate APBRmetrics, but also the ranks of guys who went on to be hired by NBA teams. Granted I believe Chaikin has done work for NBA teams, but it's astonishing to me that all these years later he still hasn't moved forward even one level of abstraction.
Berri it doesn't surprise me, but then he doesn't understand his own stats despite being an academic leveraging the Ivory Tower of CSU Bakersfield for far more than it's worth.
But still, when I tee off on those guys it's nothing like this writer. How on earth did he become convinced that RAPM was based off RPM? Only way that could happen is if someone was explaining things to him and he was so confused he could even take notes properly.
It's not quite as embarrassing as the first Bleacher Report article I ever read - a guy evaluating the drafting of John Wall by the Wizards by meditating on the close relationship he and Michael Jordan would have...years after Jordan left Washington for Charlotte - but it's mind-blowing to me that any human being exists who would dare write such inflammatory negativity about something they clearly don't have the tools to even read and understand. A writer like this, short of a massive philosophical change coming soon, will probably spend their whole life getting angry about things they don't understand and typing visceral sanctimony for the masses to choke on.
bondom34 wrote:Not a good sign he said RAPM is "real adjusted plus minus". Kinda makes it seem uninformed.
Yeah that jumped at me immediately. Plus a lot of references to Dave Berri, which is sad because I'm pretty sure that dude is the epitome of regressive stat-centric bball analysis.
Yup, Berri's the worst.
He makes really basic errors while claiming to know better than everyone else because he's an academic at a podunk university. When people challenge him, he used every kind of excuse to refrain from actual debates and will ban people from his site simply for asking questions. He taps into some people's need to feel like they are part of a revolution that the rest of the world can't see...when in reality the revolution is happening everywhere and these people are on an island rejecting new ideas.
Also, Berri is how I came to realize that Malcolm Gladwell is a nightmare. I'd only heard Gladwell talk about stuff I wasn't an expert in before and thought it sounded pretty good. But when Gladwell championed someone in NBA analysis, he chose Berri. To me that's not a mistake you can make even once. Either you champion the right rebel, or you listen to everyone. If you attach yourself to the dumbest guy in the room it's basically clear that you're always just falling for a simplistic narrative, in this case the rebel nerd in a room of jocks, when in reality the room is full of other nerds far more sophisticated than this nerd who simply don't have a need to call everyone else an idiot.
Doctor MJ wrote:My god. This is comically bad. I'm enjoying the fact that Berri and Chaikin are now tied to this amateur. Back when I first joined RealGM I also joined APBRmetrics, and back then it was dominated by guys like them who were insistent you could basically tell everything you needed from box score stats and that +/- stats didn't make any sense.
Fast forward through the years and you see that +/- stats came to not only dominate APBRmetrics, but also the ranks of guys who went on to be hired by NBA teams. Granted I believe Chaikin has done work for NBA teams, but it's astonishing to me that all these years later he still hasn't moved forward even one level of abstraction.
Berri it doesn't surprise me, but then he doesn't understand his own stats despite being an academic leveraging the Ivory Tower of CSU Bakersfield for far more than it's worth.
But still, when I tee off on those guys it's nothing like this writer. How on earth did he become convinced that RAPM was based off RPM? Only way that could happen is if someone was explaining things to him and he was so confused he could even take notes properly.
It's not quite as embarrassing as the first Bleacher Report article I ever read - a guy evaluating the drafting of John Wall by the Wizards by meditating on the close relationship he and Michael Jordan would have...years after Jordan left Washington for Charlotte - but it's mind-blowing to me that any human being exists who would dare write such inflammatory negativity about something they clearly don't have the tools to even read and understand. A writer like this, short of a massive philosophical change coming soon, will probably spend their whole life getting angry about things they don't understand and typing visceral sanctimony for the masses to choke on.
I tried engaging him on RCF, no response yet.
It's weird, it almost comes off as some misguided bitterness towards Iggy winning FMVP over Lebron last year due to raw +/-
One of the problems you run into is multicollinearity. If 80% of the time when Draymond Green is on the floor, Stephen Curry is on the floor, there’s a strong suspicion that Curry’s greatness will make Green look good simply because he was on the floor when Curry was draining 35 foot jumpers.
Want a good example how that works? Think back to last year’s Finals. Andre Igoudala won the Finals MVP because he was +62 during the six-game series (while Curry was +55 and averaged 42.5 mpg to Iggy’s 37 mpg).
That beat out LeBron James’ 35.8 ppg, 13.3 rebounds and 8.8 assists, double what Iggy produced in each, the first time in Finals history one player has led both teams in all three categories.
Iggy played a great series and was instrumental in the Warriors win, but there is no way in Hell he should win the MVP over the best single series performance in history. Might as well change the trophy title to “Most Valuable Guy on Winning Team” because that’s what the trophy means now. Thanks Real Plus Minus!
I mean what??
But in his home dwelling...the hi-top faded warrior is revered. *Smack!* The sound of his palm blocking the basketball... the sound of thousands rising, roaring... the sound of "get that sugar honey iced tea outta here!"
haha, LeBron is my guy, but come on, that's beyond ridiculous, like, no one can seriously consider himself a basketball connoisseur while uttering such statements.
). I try my best to post every update in the yearly RAPM threads and try to label them as accurately as possible to differentiate between the variations (Single-year vs. multi-year; prior-informed vs. vanilla; etc.). 
