RealGM 2023 Top 100 Project - #11 (Stephen Curry)

[One_and_Done]

For the modernists/in a vacuum crowd, the league is currently overflowing with kd or kd+ nba talents. Davis, Giannis, Harden, Kawhi, Curry, and Lebron have all played in a comparably or more talented league and have all been more valuable at their best(lebron somehow is still better). It's early tidings, but Luka already looks like a much better playoff performer impressively taking the Clippers to 7 while Durant was hailed as the best in the world for going 1-3 vs the Bucks.

And if those players had more longevity I would be voting for most of them. The only one I'd have below him if they had comparable longevity would be Harden who has his own issues. The takeaway is all these guys peaked above Kobe too.

[OhayoKD]

For the modernists/in a vacuum crowd, the league is currently overflowing with kd or kd+ nba talents. Davis, Giannis, Harden, Kawhi, Curry, and Lebron have all played in a comparably or more talented league and have all been more valuable at their best(lebron somehow is still better). It's early tidings, but Luka already looks like a much better playoff performer impressively taking the Clippers to 7 while Durant was hailed as the best in the world for going 1-3 vs the Bucks.

And if those players had more longevity I would be voting for most of them. The only one I'd have below him if they had comparable longevity would be Harden who has his own issues. The takeaway is all these guys peaked above Kobe too.

Fair enough.

Will say though. Since you're going to have to consider older players, Kobe kind of obviously translates better than the field. Reductive as it is, bros got a bag and is hard to exploit in a playoff matchup defensively, Can't say the same for Bird or Dirk. You need to make some assumptions for them to translate properly.

Bird also has the worst longetvity and played in the weakest era of those 3 so uh, not really sure why he'd be getting your vote.

[One_and_Done]

I explained in the Bird vs Kobe thread pretty clearly why I'm favouring him. Just because your era is weak, doesn't necessarily mean you can't transcend it.

[OhayoKD]

I explained in the Bird vs Kobe thread pretty clearly why I'm favouring him. Just because your era is weak, doesn't necessarily mean you can't transcend it.

Crazy how Kobe arguably led better offenses than our transcendent king and has a case as both a better era-relative creator and an era-relative scorer.

Why don't you explain in basketball terms how a guy who can't get to the rim, can't get to the stripe, is a sub-par ball-handler, and can (and was) turned into a massive defensive liability in a playoff setting(by offenses not as well suited to punish) is translating better?

Or were you under the impression he was an elite defender because he racked up uncontested defensive rebounds and a block a game next to two bigs?

[eminence]

Voting Post

Vote #1: George Mikan
Vote #2: Jerry West (I expect neither of these guys will wind up in the top 2, but I don't have a strong enough preference between Bird/Kobe to go either way)

Nomination: Oscar Robertson

Mikan - dominated the league for 8 seasons, it's tough to beat that value.

West - relatively close between all of West/Bird/Kobe here (Curry a step off the pace due to longevity), I curve the pre-ABA guys up a bit in longevity, which helps West (helps Mikan too). Feels a bit off voting West, as I prefer Oscar, but that's how the cookie crumbles.

Oscar - good competition here, up next for me is probably Dirk or Karl, though I want to look into Moses a bit more as well (mostly want to look at post-Sixers years, I'm pretty convinced he was quite quality in both Houston/Philadelphia). CP3 in contention vs the older guys, but I'm pretty sure about Dirk over him.

[One_and_Done]

Let's mostly save the Bird arguments for next thread after Curry gets in. More energy needs to be focused on getting some different types of candidates nominared this thread.

On that note Curry has 6 votes, West 3, Kobe 2, Bird 1, Mikan 1. However Bird has by far the most 2nd preferences, so will probably get up next.

Nominations are Oscar 7, D.Rob 4, Moses 1.

[OhayoKD]

So, I didn't respond to this before but I think I should.

While I totally understand why you'd term this concept "resiliency", I would object.

When an opponent commits to putting great pressure on you, it's generally the right thing to do to pass. While that pass will sometimes count for an assist, in general, when the defense commits like this, they're going to knock any holistic production assessment for an individual even though he may well be playing more valuably than if he had insisted on keeping the ball and shooting.

Resistance to adaptation which can result in decreased efficacy isn't resilience so much as stubbornness.

sure. but i also think i would be more sympathetic if steph's assists flew up and his volume went down but his TS% went up, indicating he was foregoing shot attempts in the face of tremendous defensive pressure. then i would just be penalizing him for giving up the ball. but i look at things like the first 3 games of the 2015 finals before the warriors lineup change where both the warriors and steph struggled, look at the struggle for both steph and the warriors against a non-elite defense in the 2016 cavs, look at the rockets switch-everything defense giving the warriors and steph trouble in 2018 (when healthy), look at the same defense giving steph one of his worst series ever in 2019 (with the warriors mostly surviving on KD going off individually). it feels like the warriors biggest struggles are around times when steph himself was limited, indicating to me that the two aren't unrelated and we didn't just see steph putting up lower numbers but his team still cruising.

To me you're saying here that you'd get it if Steph's numbers changed like an on-ball player's numbers would be expected to change if the defensive pressure on him compelled him to pass, but with Steph playing a distinct role where there is no such box score transition you're left feeling like Steph's just getting stopped and thus less valuable.

I will point out his effect is...to a degree, quantifiable. Box-creation is a metric specifically designed to account for that kind of thing and Steph does not look like some era-outlier creator. Passer-rating(an effeciency metric made "to compliment") is predictably lower(indirect creation is less valuable than direct creation, layups>>, height helps), and when we account for contested/uncontested shots, Steph does not generate the same shot-quality players like Lebron and Jokic do.

Steph is an all-time creator due to gravity. A massive goatish outlier who can compensate for any drop in on-ball production? No, not really. And that is reflected in all the not-goat offenses he's led with good to stellar offensive help(and good impact analysis tbh but i guess we'll get to that).

And so we should never expect box score data to effectively evaluate the value of a rover, and we should expect that particularly for this type of player you really can't conclude much at all analytically without impact data.

And of course, if the impact data I saw made Curry look like he really wasn't all that effective in the playoffs, I'd see things differently...but that's really not what the data says.

Depends on what you mean by effective and what you're looking at. Magic, Duncan, Lebron, Kareem, and Russell are off the board so I may not have much reason to press. That said, I will point out Cheema has alleged choker Harden as comparable on more possessions in the playoffs(raptor likes him too), giannis has his metrics, and jokic looks kingly in alot of different things so if you're imagining he's some impact god as has been oft suggested throughout this project...eh

Beyond there's this thing where I just struggle to understand how anyone can think of these Warriors as playoff disappointments if they are taking in the entire picture. We're talking about a team that has won way more championships and playoff series compared to anyone else in the time frame, and who has also upset other teams (by standard measures, SRS, W/L, etc) more than they've been upset.

Well there are two guys with 5 rings to Curry's 4 on the board. I'd also say painting the Rockets series as anything but an underperformance isn't really going to sell with me(and alot of the posters here) and to be clear I have the same thoughts on the Lakers in 2000, the heat vs the pacers, and the bulls vs the knicks. Russell is odd because he keeps getting taken to 7 but he also keeps not losing and then clowns 60's MJ and 60's Hakeem with not great help(feel free to use the much better 71 cast as a reference point instead of the similar 70 roster that was just bad). I am willing to give him some benefit of the doubt.

I feel like people really have run with what happened in a 3-4 loss to the Cleveland Cavaliers and let it define the foundation of their idea of Curry and the team. Oh sure people will also bring up '14-15's statistical struggles, but there we're literally talking about a team that won the championship and won at least 3 games in a row in every series - 3 times to close out the series, 1 time to go up 3-0 and basically clinch the series. If that's a critical part of the argument against them, then to me it shows how hard it is to justify the argument.

Feel like I should point out the logic you've used with 2015(once the Warriors figured it out they had almost no chance of losing) can be applied(to a degree) the other way around. The Cavs started playing faster and hunting steph in matchups in game 3 and were decisively the better team for the rest of the series. Injury context also obviously more balanced(kevin love concussed, misses game, plays subpar most of the series).

Could argue the warriors had been figured out and the Cavs by the end of those playoffs was likely to win future rounds.

There are downsides to this of course that go just beyond the complication that people rightly point out. It means that certain types of basketball players who are in the league really just for their bodies can't seem to fit in. On the other hand though, because NBA scouting is so focused on particular forms of body talent, the Warriors have been able to slot in guys who have something missing as a prospect into great success, and it's not even necessarily the case that they are super BBIQ guys. Gary Payton II I think really exemplifies this. This is a guy who if he had an obvious outlier BBIQ would have been a high drafted prospect, yet despite lacking this, on the Warriors he's been super valuable because of things that have everything to do with Kerr's philosophy and Curry's gravity.

Well, Draymond plays a role here as an on-court captain, but yes. The assumption that Steph's box-production would increase in a "conventional offense' is dubious imo. We saw Steph in 2014. I do not think he became a completely different player in terms of capability to produce in a given context. His impact was pretty good though.

It's not so problematic if you talk about it one stat at a time "scoring volume resiliency" "TS% resiliency", in part because it reaches for less, but when you put it all under one umbrella and then use the term you literally penalized guys for making the right call at times.

And of course those who know me probably no where I'm going with this: I think Impact Resilience is more the thing to focus on here, both because Impact is in the end what matters, and it has no preferences as to whether a guy helps his team by volume scoring, playmaking, defense, or harder to see and quantify measures.

Now as I say this, in the context of a project like this, I'm also less interested in Resilience than I often would be. How Great you are is first and foremost about what you actually do when it matters, not by how that compares to how do other times.

so i know i brought up some of this a long time ago in maybe thread #3 or #4 and you responded with a really good post and i never got a chance to respond, but i'll bring up what i brought up then. when steph's impact seemingly always looks good, whether he plays well or not (i'll say more when i respond to DraymondGold), it makes me question the value of the numbers and whether we're not just getting some weird lineup/draymond effect in the numbers and not really impact, per se.

I appreciate you being open about your concerns here. It's absolutely worth talking through.

On the broadest level I think the thing to remember is this:

The only reason why we shouldn't take +/- data to be THE defining estimation of player value in that context is noise.
That noise is a very real and massive concern...but when you're talking about something that's "seemingly always" happening, it starts becoming very problematic to chalk it up to noise.

Eh, you are neglecting what is also "always" true is that Draymond looks like a super-star by impact without Steph. The sole exception is 2020 where there's a bunch of context(and if you want to ignore all that...steph looked worse before closing down). Steph also has always played the vast majority of his minutes with draymond at the point he started generating historically great impact(as opposed to simply one of the best in the league).

Moreover this is not an argument from absence

-> Lineup adjusted stuff is way lower on Steph than plus-minus is and those are still only really approximations
-> We saw Steph pre-draymond becoming a starter. He would not be ranked here based on what we saw there.
-> Steph's on/off spikes with dray and falls without

This is the reverse of what we see with say Duncan who played way more minutes than his best teammates and had to then play with their **** replacements. on/off is volatile, 1-year will say player x or y was really good. And then over extended samples Duncan looks alot better with lineup-adjustment, these other on/off monsters mysteriously have no effect on the team when they're out(drob, manu) and these "real best players' mysteriously do way worse when duncan is completely out of the picture(a better drob isn't getting anywhere near 1999, and then the spurs fall off a cliff in 2000 with duncan out)

I don't see anything problematic here. Whenever we isolate for the variable in question, Steph looks worse(okay fine sacremento king game, but there's more where the opposite happens). Fp4 can probably be stronger here. I'd even say it's probable Steph's numbers are inflated just like it's probable Duncan's are suppressed.

Spoiler:

Just in terms of zooming in to a specific period though, I feel like the '21-22 regular season is the most obvious thing to look at so that's what I'll do.

Let's note up front that, in going from '20-21 to '21-22, the Warriors shot a worse 3PA%:

'20-21: 37.6%
'21-22: 36.4%

This isn't anything huge - the whole league shot worse in '21-22 than '20-21 - but I think it's noteworthy because it tells us that there's a limit to the the potential for shooting luck. Not that it can't exist, but this isn't a situation where the Warriors as a whole literally became considerably more successful at shooting 3's the next year.

This of course is despite the fact that that Wiggins & Poole saw their 3P% go up incrementally, the abyssmally low Oubre was no longer on the team, and Klay came back to the team for about half a year while he wasn't at his best, he still shot the long ball well by all reasonable standards.

We can do more analysis of the 3-ball here, but I'd suggest that it may not be as weird as folks may have assumed it was. A team getting it's players more used to playing this way while jettisoning bad shooters who weren't gaining traction with it, and getting a great shooter back part way through the season managing to not quite shoot from downtown as well as before, isn't necessarily that crazy.

Nevertheless, the team's ORtg goes, so what's driving that? Among the 4 factors, it's clearly the Offensive Rebounding %.

In '20-21 the Warriors were last in the league at 17.9%.
In '21-22 the Warriors were 18th in the league at 22.8%.

Big jump, who is getting those boards for the Warriors? I think looking per 100 possessions (for guys who significant minutes) makes the most sense here:

In '20-21:
1. Kevon Looney 4.7
2. James Wiseman 3.1
3. Kelly Oubre 2.3
4. Eric Paschall 1.9
5. Andrew Wiggins 1.6

1. Kevon Looney 5.9
2. Otto Porter 3.0
3. Gary Payton II 2.8
4. Nemanja Bjelica 2.6
5. Jonathan Kuminga 2.2

You see the difference in focus? Obviously Looney is doing more offensive rebounding and the rest of the top 5 guys are brand new. Not clear here is that despite the fact Wiggins is dropping out of the Top 5, his OReb/100 goes up, as do Draymond & JTA who are the next guys on the '20-21 list, and even Curry's rate goes up.

So if we're looking for a change in focus that helped the Warriors, I think the thing to look at is the offensive rebounding, which I don't think "luck" is a good explanation for. This was a strategic shift.

One might want to say that since the guys getting these numbers are not Curry, Curry should be seen as lucky...but I think it's important to remember that the Warriors were not a big team, and their main offensive rebounder is a 6'9" who isn't all that athletic.

I would suggest that this offensive rebounding shift had everything to do with taking advantage of the spacing provide by the team's outside shooting. Curry's not the only part of that, but he was certainly a big part of it.

And of course, then there's the spacing impact on other guys shooting. Here I'll highlight Payton again.

This is a guy with a career 59.3 FT%, who to this point in his career at age 30 has made more 3's in '21-22 than he has in the entire rest of his career combined, and did so with a decent %. And of course, that's not the really noteworthy thing so much as his dunks, which were not coming from him driving against the teeth of the defense, but rather were about the defense leaving him with space that allowed teammates to find him for the easy jam.

Okay, think I'll leave it at that for now.

Fair points. Will note I find this significantly more compelling as a defense of Steph as opposed to a certain thinking basketball screenshot...

[rk2023]

Nominations are Oscar 7, D.Rob 4, Moses 1.

Am curious the case for Robinson over Nowitzki.

[One_and_Done]

I've previously nominated Dirk. He's not getting traction.

[eminence]

Nominations are Oscar 7, D.Rob 4, Moses 1.

Am curious the case for Robinson over Nowitzki.

I'm on the side of Dirk over Robinson, but it feels like mostly a hypothetical/skillset type of argument. Dirk peaked as a clear MVP level player, and had a pretty nice aging curve, it feels like one would have to rate Robinson as essentially the GOAT peak/prime to overcome the clear gap in time played.

I'm interested to see a direct comp from a Robinson supporter as well.

[rk2023]

Nominations are Oscar 7, D.Rob 4, Moses 1.

Am curious the case for Robinson over Nowitzki.

I'm on the side of Dirk over Robinson, but it feels like mostly a hypothetical/skillset type of argument. Dirk peaked as a clear MVP level player, and had a pretty nice aging curve, it feels like one would have to rate Robinson as essentially the GOAT peak/prime to overcome the clear gap in time played.

I'm interested to see a direct comp from a Robinson supporter as well.

One advantage in Robinsons' favor - As one who is quite confident ranking Dirk ahead - is having close to the best first 3 seasons all-time [I take Kareem a clear #1 in this regard, and see Robinson competing with the likes of Wilt, Duncan, & Oscar - perhaps Shaq though I'm not high on 1993]. I mentioned yesterday at first, but I see 1990-96 at an MVP level for the most part - with the exception of missing time in 1992. I surely like Dirk's supporting years better, but that leaves me with 7 >= ~MVP tier seasons for the both of them (05-11 for Dirk).

[ceiling raiser]

I'm a little surprised Magic fell all the way to 10 (even with Garnett making the top 10 more crowded). Life has gotten in the way so I haven't had a chance to read many arguments, but I still find his offensive impact to be truly elite putting him slightly ahead of some two way all time greats. I'll go back and read the threads when I have a chance.

Magic and Bird were savant level decision makers which is why I think they're still clear cut top 10 players regardless of average at best longevity. So while my gut is to still go with Bird here, I totally get the argument for Curry who's one of my favorite players. He has a relatively full career we can now judge vs these guys. 

Then there's kobe, who i've had right outside the top 10 for a while now. Taking a look at his career again, he still has the longevity edge but not quite as significant as I remember. Those last few seasons post achilles were rocky to say the least. Going into the project I had bird/kobe/curry in that order so now I have some thinking to do. 

I've long been a Kobe skeptic, but I think it's time to reevaluate him. Other than the soft skillset stuff, it's worth noting Kobe has an advantage over Bird based on some of Ben Taylor's work in SRS-championship correlation studies.

I think Bird's skillset is interesting, but I wonder if the top 10 GOAT offensive candidate status is gifted despite mediocre impact from the signals we do have, and big question marks as a creator, scorer, and defender.

[Doctor MJ]

I feel like people really have run with what happened in a 3-4 loss to the Cleveland Cavaliers and let it define the foundation of their idea of Curry and the team. Oh sure people will also bring up '14-15's statistical struggles, but there we're literally talking about a team that won the championship and won at least 3 games in a row in every series - 3 times to close out the series, 1 time to go up 3-0 and basically clinch the series. If that's a critical part of the argument against them, then to me it shows how hard it is to justify the argument.

Feel like I should point out the logic you've used with 2015(once the Warriors figured it out they had almost no chance of losing) can be applied(to a degree) the other way around. The Cavs started playing faster and hunting steph in matchups in game 3 and were decisively the better team for the rest of the series. Injury context also obviously more balanced(kevin love concussed, misses game, plays subpar most of the series).

Could argue the warriors had been figured out and the Cavs by the end of those playoffs was likely to win future rounds.

Ah, makes sense to bring up. So, some thoughts:

1. I do think we should consider very strongly how series ended when talking about how close the series was, but when evaluating a player's career, I think we should be very careful about letting one series dictate too much, particularly when the key opposing player is someone who has already been accepted higher in the project.

2. Smaller thing but: When we're talking about the Cavs winning the last 3 to take the series, we should also note that Game 7 was very, very close. So this wasn't a series where beyond a certain point one team just had the other licked.

so i know i brought up some of this a long time ago in maybe thread #3 or #4 and you responded with a really good post and i never got a chance to respond, but i'll bring up what i brought up then. when steph's impact seemingly always looks good, whether he plays well or not (i'll say more when i respond to DraymondGold), it makes me question the value of the numbers and whether we're not just getting some weird lineup/draymond effect in the numbers and not really impact, per se.

I appreciate you being open about your concerns here. It's absolutely worth talking through.

On the broadest level I think the thing to remember is this:

The only reason why we shouldn't take +/- data to be THE defining estimation of player value in that context is noise.
That noise is a very real and massive concern...but when you're talking about something that's "seemingly always" happening, it starts becoming very problematic to chalk it up to noise.

Eh, you are neglecting what is also "always" true is that Draymond looks like a super-star by impact without Steph. The sole exception is 2020 where there's a bunch of context(and if you want to ignore all that...steph looked worse before closing down). Steph also has always played the vast majority of his minutes with draymond at the point he started generating historically great impact(as opposed to simply one of the best in the league).

Moreover this is not an argument from absence

-> Lineup adjusted stuff is way lower on Steph than plus-minus is and those are still only really approximations
-> We saw Steph pre-draymond becoming a starter. He would not be ranked here based on what we saw there.
-> Steph's on/off spikes with dray and falls without

This is the reverse of what we see with say Duncan who played way more minutes than his best teammates and had to then play with their **** replacements. on/off is volatile, 1-year will say player x or y was really good. And then over extended samples Duncan looks alot better with lineup-adjustment, these other on/off monsters mysteriously have no effect on the team when they're out(drob, manu) and these "real best players' mysteriously do way worse when duncan is completely out of the picture(a better drob isn't getting anywhere near 1999, and then the spurs fall off a cliff in 2000 with duncan out)

I don't see anything problematic here. Whenever we isolate for the variable in question, Steph looks worse(okay fine sacremento king game, but there's more where the opposite happens). Fp4 can probably be stronger here. I'd even say it's probable Steph's numbers are inflated just like it's probable Duncan's are suppressed.

Doesn't feel like I'm neglecting anything that was presented in the context I was responding to.

He was concerned about Curry looking great by +/- even when his production wasn't the best, and I spoke to that.
I've also reiterated that I'm higher on Green than most people, so I'm not looking to knock him.

To your point about Green always looking great too, a few thoughts:

1. I'm not sure I really see this as a concern. They are both great players, so shouldn't they both look great? I understand a feeling that if you don't get enough Off time you'll never really know how independent the two players are, but it's not 2014 any more.

2. Because Curry is better on offense and Green better on defense there isn't the same conceptual blurring we'd normally have. Green isn't making Curry the best shooter in history, and I really doubt Curry is making Green the best defender of this era.

2. From '14-15 to '18-19, Curry played 2549 minutes with Dray on the bench. In that time, Curry was a +12.8 per 100. So, even before we see the Warriors falling apart without Curry in '19-20, I think we saw plenty to make clear that everything didn't fall apart without Green.

Re: Duncan played had to play minutes without his best teammates, so data is suppressed where Curry is inflated. Depending on what you mean, I'm not so sure of the general principle here. If you just mean that's the case for the raw +/-, I get the idea, but not with regressed data, and Ginobili, Robinson & Kawhi look great by Playoff RAPM. Are you arguing that Duncan's RAPM is suppressed here?

Re: DRob isn't getting anywhere near 1999 without Duncan. The reverse is true too, no?

Re: Spurs fall of a cliff without Duncan in 2000. Eh, just so we're clear:

In the 8 games without Duncan in that regular season, the Spurs went 5-3, with Robinson having a raw +/- of +58.
In the Playoff series against Phoenix where the Spurs lost 3-1, Robinson had a raw +/- of +13.

I really don't think there's any reason to use that data to doubt Robinson. I mean, you can knock Robinson to a degree for only playing 32 MPG in '99-00 in the RS and 38.8 in the PS, but I really don't think any kind of +/- is going to tell you that Robinson was an impact mirage based on that data.

Also, just looking at the Ginobili vs Duncan WOWY stuff over the course of their careers beginning in '02-03 I get:

Duncan & Ginobilis: +8.94
Duncan no Ginobili: +0.62
Ginobili no Duncan: +5.69
No Duncan or Ginobili: -8.01

Ginobili sure looks good. Am I missing something?

I mean, when it comes to Duncan vs Ginobili, I'm not trying to argue for Ginobili here because of the minutes issue you've already described, but I don't really see any lens that makes Robinson & Ginobili stop looking fantastic.

[eminence]

Am curious the case for Robinson over Nowitzki.

I'm on the side of Dirk over Robinson, but it feels like mostly a hypothetical/skillset type of argument. Dirk peaked as a clear MVP level player, and had a pretty nice aging curve, it feels like one would have to rate Robinson as essentially the GOAT peak/prime to overcome the clear gap in time played.

I'm interested to see a direct comp from a Robinson supporter as well.

One advantage in Robinsons' favor - As one who is quite confident ranking Dirk ahead - is having close to the best first 3 seasons all-time [I take Kareem a clear #1 in this regard, and see Robinson competing with the likes of Wilt, Duncan, & Oscar - perhaps Shaq though I'm not high on 1993]. I mentioned yesterday at first, but I see 1990-96 at an MVP level for the most part - with the exception of missing time in 1992. I surely like Dirk's supporting years better, but that leaves me with 7 >= ~MVP tier seasons for the both of them (05-11 for Dirk).

I'd be inclined to give Dirk MVP level for '03 as well, clear best player on a 60 win team, near his box peak, may have been his +/- peak, had the squad in position to compete for a title before getting injured. League recognition certainly wasn't there yet though.

[HeartBreakKid]

Do any blokes want to elaborate on Curry > West?

[ceiling raiser]

Anyhow, I thought I voted, but I don't see my post in this thread. Primary vote is for Steph Curry.

Curry vs Bird. This doesn't really seem to be a great matchup for Bird IMO. I think Curry was clearly better on both sides of the ball, and has better longevity in terms of Weak MVP+ seasons. I also don't think Bird translates from a modernism POV. Feels like we've been taking for granted his playmaking ability and shooting ability as outlier-level for a while. If we don't, his scoring and defense aren't enough for him to be a top 30 guy. No solid impact signals really, either.

Curry vs Bryant. Kobe is being put in a bad situation here, but he actually was a solid offensive player. Top 5 in a modern league on that end for a few years, which is more than we can say for anybody here except Curry. Being outclassed by Curry is no more of an insult than Curry being outclassed by LeBron.

Curry vs Mikan. Have not seen a strong case made. Absence of evidence != evidence of absence but there is nothing to suggest to me he deserves to be considered among post-segregation players.

Curry vs West. West actually has a decent impact footprint, but I haven't seen a sustained range from his shot in limited footage. Obviously, little incentive to step out 22+ feet when it's still worth two points, but in order to project a three-point skillset out for a guy, need to see at least 2-3 of these a game.

Alternate vote is Kobe Bryant. I don't think he is the 12th best player of all-time, but that's not the question here. The question is, of this subset of players, is he the best of the remainder. As the most modern guy with the fewest question marks, even with mediocre impact by top 25 GOAT standards, he is my pick here.

Nomination is David Robinson. I think Dirk would be an interesting discussion, but given how much I value defense, he shouldn't be far removed from Hakeem and Garnett.

[Proxy]

I'd be inclined to give Dirk MVP level for '03 as well, clear best player on a 60 win team, near his box peak, may have been his +/- peak, had the squad in position to compete for a title before getting injured. League recognition certainly wasn't there yet though.

What are your thoughts on Dirk's playoffs on/off looking fairly unimpressive until the mid-late 2000s?

There's definitely some noise there in the on/off, but the team's net rating when he's on court collapsed as well in that early stretch, I tend to believe it's in part due to some weaknesses in his game - ex: not punishing wings to the same degree he did in the late 2000s so teams were more comfortable with that strategy and reducing some of his off ball value, as well as his defense issues, but that alone doesn't quite explain a drop as severe as his IMO

I definitely agree with that being a MVP-level regular season, but i'm not sure if that version of Dirk was consistently playing at that same MVP level in the playoffs

[lessthanjake]

I will point out his effect is...to a degree, quantifiable. Box-creation is a metric specifically designed to account for that kind of thing and Steph does not look like some era-outlier creator. Passer-rating(an effeciency metric made "to compliment") is predictably lower(indirect creation is less valuable than direct creation, layups>>, height helps), and when we account for contested/uncontested shots, Steph does not generate the same shot-quality players like Lebron and Jokic do.

Steph is an all-time creator due to gravity. A massive goatish outlier who can compensate for any drop in on-ball production? No, not really. And that is reflected in all the not-goat offenses he's led with good to stellar offensive help(and good impact analysis tbh but i guess we'll get to that).

I don’t want to go down a rabbit hole, but I think this once again gets to the really important issue you need to understand about these Backpicks stats. They are just formulas to fit hand-counted data in the aggregate. They can basically just be thought of like a line of best fit. And a line of best fit does not accurately describe every individual data point—in fact, we would expect that it would not actually precisely describe almost any of them and would be far away from describing some outliers. It’s just that, in the aggregate, it fits fairly well. So we should be really careful in taking the output of something like this for an individual player as the gospel and basing an entire argument around precise values in these stats, because it’s just a rough approximation that we’d fully expect to be wrong to at least some degree and could well be quite inaccurate for a given data point.

And, to that end, I doubt Box Creation accurately values Steph’s creation, as Steph is an outlier in relevant ways that the model doesn’t really have a way of accounting for. The model does try to get at creation through gravity—which is good. The way it does this is through looking at scoring and “three-point proficiency.” That three-point proficiency term is basically just a combination of three-point % and three-point volume. That all sounds good and would capture some of Steph’s playmaking, of course. But would it capture everything that makes Steph’s off-ball playmaking so great? I really don’t think so. For instance, there’s no term here to account for a player’s max effective range—which is a super important element in terms of gravity, since the further out past the three-point line that defenders have to guard you, the more space you create for teammates. And, of course, Steph is pretty unique at being able to make deep threes (though there are one or two others these days who do it too), and has to be defended further out than virtually anyone else. The “three-point proficiency” term in the Box Creation formula doesn’t take this into account, so his Box Creation number doesn’t account for the fact that he’s pulling defenders out 10+ feet behind the three-point line—something that has a *massive* playmaking effect. Similarly, Steph is unique in the amount of off-ball movement he does, and Box Creation doesn’t really have a way of accounting for that. The formula knows he scores a lot and it knows that he’s a great three-point shooter, so it knows that he can exert a lot of gravity. But it doesn’t know that he’s exerting that gravity more consistently than other players, because of his almost constant off-ball movement. This is a major reason why the model doesn’t realize that Curry creates more than Damian Lillard does (not to pick on Dame—he’s a great player): It has no way of understanding the unique amount of off-ball pressure that Steph is exerting on the defense beyond that which a player who scores and shoots threes fairly similarly does.

Eh, you are neglecting what is also "always" true is that Draymond looks like a super-star by impact without Steph. The sole exception is 2020 where there's a bunch of context(and if you want to ignore all that...steph looked worse before closing down).

That’s not true. Of course, we’ve seen Draymond without Steph for an entire season (by the way: Saying “Steph looked worse before closing down” that season is weird when we realize Steph played a grand total of 139 minutes of basketball that season). Draymond had little if any positive impact in 2019-2020, even on a bad team. He was ranked 365th in the league in NBAshotcharts RAPM that season—with a negative score. He had a negative RAPTOR, ranking 128th in the league. EPM is more charitable to him, at least giving him a positive score and ranking him 101st in the NBA. He was ranked 114th in the NBA in RPM. So that not only wasn’t “super-star by impact without Steph,” it wasn’t even really *positive* impact in a season without Steph.

More generally, in the last ten seasons, if we only look at non-garbage-time minutes Steph has been off the court, the Warriors have been -4.04 without Draymond and +0.21 with Draymond. That’s definitely solid impact without Steph, but it doesn’t look like “super-star by impact.” And we can see one reason why this is the case: When Steph isn’t on the court, Draymond is a liability offensively. Indeed, in Steph “off” minutes in that timeframe, the Warriors’ offensive rating has been *worse* when Draymond is on. That comes in contrast to Steph’s “on” minutes, where Draymond being on increases the team’s offensive rating. Steph’s greatness unlocks Draymond as a competent offensive player, but he is not one without Steph and that contributes to him definitely *not* “looking like a super-star by impact without Steph.”

Moreover this is not an argument from absence

-> Lineup adjusted stuff is way lower on Steph than plus-minus is and those are still only really approximations
-> We saw Steph pre-draymond becoming a starter. He would not be ranked here based on what we saw there.
-> Steph's on/off spikes with dray and falls without

It is objectively false that “Steph’s on/off spikes with dray and falls without.” We’ve been over this before, so I really have no idea why you’re still saying this. In the last decade in RS+Playoffs, if we only look at minutes that Draymond is on, Steph has a +12.35 on-off. If we only look at minutes that Draymond is off, Steph has a +12.22 on-off. So Steph’s on-off is essentially completely unaffected by whether Draymond is on or off the court. It is just false to say “Steph’s on/off spikes with dray and falls without” and I’ve already pointed that out to you before.

On the second bullet, FYI, in 2013-2014 (i.e. “pre-draymond becoming a starter”), Steph was already being ranked incredibly highly in impact. He was 1st in AuPM/g, 2nd in RPM, 2nd in RAPTOR, 2nd in EPM, 4th in Engelmann RAPM, etc.

And the idea that “lineup adjusted stuff is way lower on Steph than plus-minus is” is quite odd when we realize how well he does in impact metrics in general—being clearly the highest rated player overall by impact metrics in the last decade. But I know you just refuse to admit that that’s true and will keep saying that it’s not despite a mountain of evidence to the contrary, so I digress.

[eminence]

I'd be inclined to give Dirk MVP level for '03 as well, clear best player on a 60 win team, near his box peak, may have been his +/- peak, had the squad in position to compete for a title before getting injured. League recognition certainly wasn't there yet though.

What are your thoughts on Dirk's playoffs on/off looking fairly unimpressive until the mid-late 2000s?

There's definitely some noise there in the on/off, but the team's net rating when he's on court collapsed as well in that early stretch, I tend to believe it's in part due to some weaknesses in his game - ex: not punishing wings to the same degree he did in the late 2000s so teams were more comfortable with that strategy and reducing some of his off ball value, as well as his defense issues, but that alone doesn't quite explain a drop as severe as his IMO

I definitely agree with that being a MVP-level regular season, but i'm not sure if that version of Dirk was consistently playing at that same MVP level in the playoffs

I see it as a valid concern, with some truth to it, but I don't find it overly concerning compared to the nomination competition here (other than Oscar, whom I voted for). Karl Malone, David Robinson, Chris Paul, unfortunately not paragons of playoff play (CP3 due to injury issues as much as anything). KD/Moses maybe? But KD I see with a lot of the same problems and never really solving them on his own. Moses I need to look at a bit more (any strong feelings on his seasons after leaving the Sixers?).

On the actual voting ballot it would hurt him compared to that comp.

[One_and_Done]

Do any blokes want to elaborate on Curry > West?

Curry was the gamebreaking, defence warping driver of GOAT offences in the modern game. West was not even the best player in a relatively amatuer league.