Brunson v Curry

Moderators: Doctor MJ, Clyde Frazier, penbeast0, trex_8063, PaulieWal

lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#41 » by lessthanjake » Fri May 3, 2024 3:50 pm

AEnigma wrote:2023 Steph “on par with Jalen Brunson” Curry: 30 points per 75 possessions on 65.6% true shooting (113 TS+), 42.7% from 3 (118 3P+), 24.8% free throw rate (93 FTr+)

2017 Steph “arguably greatest offensive peak in league history” Curry: 27.4 points per 75 possessions on 62.4% true shooting (113 TS+), 41.1% from 3 (115 3P+), 25.1% free throw rate (93 FTr+)

Different animal indeed. But apparently it is obnoxious to suggest that people seem to care more about on/off splits than about actual production.


Yeah, impact data (including *much* more sophisticated data than raw “on/off splits”) is a big part of the picture in evaluating a player—and, while impact data isn’t perfect, it’s likely significantly better than just the raw box scoring data you list there. And, to take one example, Steph’s EPM in 2022-2023 was his lowest since 2012-2013, and a good deal below his peak years (but still slightly above Brunson’s this year, while Brunson was ahead of Steph’s 2023-2024). Steph’s LEBRON last year was lower than any year since 2012-2013, and, again, was quite a bit below his peak years, while still being above the LEBRON Brunson put up this year (while Steph this year is below Brunson). Steph’s RAPTOR in 2022-2023 was also the lowest he had in any season (note: the data starts at 2013-2014 and ends with last year, so we can’t compare to Brunson here), and quite a bit below his peak years. So yeah, this sort of data indicates that Steph had already notably declined by 2022-2023 and was comparable (though maybe a little better) than Brunson’s 2023-2024, while indicating that Brunson’s 2023-2024 is a bit better than Steph’s 2023-2024.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
AEnigma
Veteran
Posts: 2,853
And1: 4,432
Joined: Jul 24, 2022
 

Re: Brunson v Curry 

Post#42 » by AEnigma » Fri May 3, 2024 4:57 pm

In information conveyed they are not much more sophisticated than the raw splits, no, which is why you can see such significant year-to-year variation in anything that attempts to account for them. 2015-19 Curry is not significantly different from 2021-23 Curry; he just was more of a league relative standout playing with much better lineups. Similarly, Jalen Brunson did not suddenly transform his entire skillset the second Quickley was removed from the team — and while he did successfully increase his overall usage without any drop-off, which is commendable in itself, my hope has always been that people would not assume doing so made him 8.5 points more valuable on-court (to say nothing of the on/off value). But that requires paying attention to the players themselves and how different lineup combinations affect them, and it limits our ability to directly equate players to what they accomplish in their best possible environment, so it unfortunately seems to have fallen out of favour as a real approach.

2024 was Curry’s least productive regular season since… 2022, when he won a title as a top five player. I suppose someone could glance at O-LEBRON and try to tell me 2024 Curry’s gravity declined by 80% while scoring more on equivalent league relative efficiency, and I will treat that with the same degree of seriousness I always have. I just did not expect there to be so much truth to all my jokes about fans losing faith in Curry should he ever suffer the ignominy of being supported by a disproportionately effective bench.

The Knicks were +3.5 with Brunson on the bench until the Quickley trade — giving him, just like last year, a negative on/off. Since the trade, they have been -7.5 with him on the bench and he has an on/off close to Jokic’s. While that makes him literally more valuable, common sense (sadly not so common anymore) should be screaming that trading a backup has little bearing on a player’s quality, and certainly not in the stratosphere of a 20-point swing.
User avatar
cpower
RealGM
Posts: 18,618
And1: 7,202
Joined: Mar 03, 2011
   

Re: Brunson v Curry 

Post#43 » by cpower » Fri May 3, 2024 5:01 pm

AEnigma wrote:2023 Steph “on par with Jalen Brunson” Curry: 30 points per 75 possessions on 65.6% true shooting (113 TS+), 42.7% from 3 (118 3P+), 24.8% free throw rate (93 FTr+)

2017 Steph “arguably greatest offensive peak in league history” Curry: 27.4 points per 75 possessions on 62.4% true shooting (113 TS+), 41.1% from 3 (115 3P+), 25.1% free throw rate (93 FTr+)

Different animal indeed. But apparently it is obnoxious to suggest that people seem to care more about on/off splits than about actual production.

not to blame on certain teammates but Klay and Wiggins has been terrible most of the season...If you give him Hart and Donte warriors would probably a top 3 team this season
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#44 » by lessthanjake » Fri May 3, 2024 5:12 pm

AEnigma wrote:In information conveyed they are not much more sophisticated than the raw splits, no, which is why you can see such significant year-to-year variation in anything that attempts to account for them. 2015-19 Curry is not significantly different from 2021-23 Curry; he just was more of a league relative standout playing with much better lineups. Similarly, Jalen Brunson did not suddenly transform his entire skillset the second Quickley was removed from the team — and while he did successfully increase his overall usage without any drop-off, which is commendable in itself, my hope has always been that people would not assume doing so made him 8.5 points more valuable on-court (so say nothing of the on/off value). But that requires paying attention to the players themselves and how different lineup combinations affect them, and it limits our ability to directly equate players to what they accomplish in their best possible environment, so it unfortunately seems to have fallen out of favour as a real approach.

2024 was Curry’s least productive regular season since… 2022, when he won a title as a top five player. I suppose someone could glance at O-LEBRON and try to tell me 2024 Curry’s gravity declined by 80% while scoring more on equivalent league relative efficiency, and I will treat that with the same degree of seriousness I always have. I just did not expect there to be so much truth to all my jokes about fans losing faith in Curry should he ever suffer the ignominy of being supported by a disproportionately effective bench.

The Knicks were +3.5 with Brunson on the bench until the Quickley trade — giving him, just like last year, a negative on/off. Since the trade, they have been -7.5 with him on the bench and he has an on/off close to Jokic’s. While that makes him literally more valuable, common sense (sadly not so common anymore) should be screaming that trading a backup has little bearing on a player’s quality, and certainly not in the stratosphere of a 20-point swing.


Yes, a guy in his mid-30’s having steadily declining impact numbers must just be because of changing lineup combinations (including in metrics that aim to control for quality of other players on the court) and not at all because he’s getting old and not as good as before. Of course!
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
AEnigma
Veteran
Posts: 2,853
And1: 4,432
Joined: Jul 24, 2022
 

Re: Brunson v Curry 

Post#45 » by AEnigma » Fri May 3, 2024 5:40 pm

Amazing how a single offseason can make one player 60% worse because of “age” in a way that corresponds better to painfully obvious team changes than it does to anything they are doing on the court, while another player three years older and much more visibly declined can be adamantly equated to their prime form. Very principled.
Doc MJ wrote:This is one of your trademark data-based arguments in which I sigh, go over to basketballreference, and then see all the ways you cherrypicked the data toward your prejudiced beliefs rather than actually using them to inform you
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#46 » by lessthanjake » Fri May 3, 2024 8:03 pm

AEnigma wrote:Amazing how a single offseason can make one player 60% worse because of “age” in a way that corresponds better to painfully obvious team changes than it does to anything they are doing on the court, while another player three years older and much more visibly declined can be adamantly equated to their prime form. Very principled.


Huh? He doesn’t look 60% worse, nor is his decline in impact numbers at all out of whack with his decline in box numbers. His decline in EPM between 2022-2023 and 2023-2024 was 19%, and his decline in LEBRON was 38%. And that’s while purely box-based stats like Basketball Reference BPM and Thinking Basketball BPM (i.e. both not directly affected by lineup combinations or on-off) has him going down 31% and 29% respectively. It’s all a pretty consistent picture of decline, regardless of whether or not the metrics are using impact data or not.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
AEnigma
Veteran
Posts: 2,853
And1: 4,432
Joined: Jul 24, 2022
 

Re: Brunson v Curry 

Post#47 » by AEnigma » Fri May 3, 2024 9:18 pm

lessthanjake wrote:
AEnigma wrote:Amazing how a single offseason can make one player 60% worse because of “age” in a way that corresponds better to painfully obvious team changes than it does to anything they are doing on the court, while another player three years older and much more visibly declined can be adamantly equated to their prime form. Very principled.

Huh? He doesn’t look 60% worse, nor is his decline in impact numbers at all out of whack with his decline in box numbers. His decline in EPM between 2022-2023 and 2023-2024 was 19%, and his decline in LEBRON was 38%.

Crossed wires while typing. 60% as good, 40% worse. Either way, hardly feasible when watching, and also very much “out of whack” with impact measures. +2.51 RAPM in 2021 (#11) to +3.65 RAPM in 2022 (#2nd) to +2.4 RAPM in 2023 (#18) to +0.46 RAPM this year (#152). Really tracks well with the eye test, huh. No box score contradictions there.

And that’s while purely box-based stats like Basketball Reference BPM and Thinking Basketball BPM (i.e. both not directly affected by lineup combinations or on-off) has him going down 31% and 29% respectively. It’s all a pretty consistent picture of decline, regardless of whether or not the metrics are using impact data or not.

Both use team quality as a measure, which as far as I know requires the involvement of other player beyond Curry. And there too, I would not say 30% decline over an offseason is remotely supportable by the film, no.

2022 is a demonstrably worse box input year than 2021/23. It is more on par with this year, and to focus on BPM, a shift from 5.8 to 5.2 is reasonable enough as a season-to-season variation. But whereas 2022 was consistently measured as the highest pure “impact” season since Durant left, regardless of the individual decline in production, Curry’s “impact” this go around may as well be marginal. Same scoring (arguably better), but now a 50% decline in LEBRON.

Or to look at 2021, we have Curry’s comfortable third place BPM season after 2015/16, but much more distant by raw impact, and conparatively neutered by the metrics directly pulling raw impact. Maybe you have forgotten all the Warriors fans lamenting how Wiseman and Oubre were horrific vampires on Curry’s impact that year. Or maybe you simply shrugged and decided that random intangibles were the reason why he was not a top 5 player that year despite all the gaudy offensive production. Just like you will shrug if Curry plays exactly the same — hell, judging by 2022, maybe plays worse — but sees an uptick in all these spreadsheet outputs because the team struggles more without him.

I feel like every month you get more removed from being able to discuss any aspect of the sport sincerely. Age matters until it does not. Theoretical defence matters until it does not. And lineups? Well, they matter when we occasionally care to use a cudgel about “fit” or similarly arbitrary assessments of “talent”, but otherwise, nah, not that important, no need to give it any real thought — after all, the player is a year older! Why even review film, think about scheme, or consider team structure, when it is so much easier to reactively grasp at whatever nebulous principle seems like it might momentarily justify those outputs as a real encapsulation of a player. :-?
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#48 » by lessthanjake » Fri May 3, 2024 11:50 pm

AEnigma wrote:
lessthanjake wrote:
AEnigma wrote:Amazing how a single offseason can make one player 60% worse because of “age” in a way that corresponds better to painfully obvious team changes than it does to anything they are doing on the court, while another player three years older and much more visibly declined can be adamantly equated to their prime form. Very principled.

Huh? He doesn’t look 60% worse, nor is his decline in impact numbers at all out of whack with his decline in box numbers. His decline in EPM between 2022-2023 and 2023-2024 was 19%, and his decline in LEBRON was 38%.

Crossed wires while typing. 60% as good, 40% worse. Either way, hardly feasible when watching, and also very much “out of whack” with impact measures. +2.51 RAPM in 2021 (#11) to +3.65 RAPM in 2022 (#2nd) to +2.4 RAPM in 2023 (#18) to +0.46 RAPM this year (#152). Really tracks well with the eye test, huh. No box score contradictions there.


This is just a sleight of hand, because no one is basing their views on single-season RAPM. As I’ve said many times, Single-season RAPM is largely garbage, precisely because it is way too noisy and way too dependent on on-off. As I already showed, Steph’s drop in better single-season impact metrics was consistent with the drop in box metrics.

And that’s while purely box-based stats like Basketball Reference BPM and Thinking Basketball BPM (i.e. both not directly affected by lineup combinations or on-off) has him going down 31% and 29% respectively. It’s all a pretty consistent picture of decline, regardless of whether or not the metrics are using impact data or not.

Both use team quality as a measure, which as far as I know requires the involvement of other player beyond Curry. And there too, I would not say 30% decline over an offseason is remotely supportable by the film, no.

2022 is a demonstrably worse box input year than 2021/23. It is more on par with this year, and to focus on BPM, a shift from 5.8 to 5.2 is reasonable enough as a season-to-season variation. But whereas 2022 was consistently measured as the highest pure “impact” season since Durant left, regardless of the individual decline in production, Curry’s “impact” this go around may as well be marginal. Same scoring (arguably better), but now a 50% decline in LEBRON.


Actually in Thinking Basketball BPM, Steph’s BPM was 3.4 this year, 4.8 last year, and 4.7 in 2021-2022, so that did not have 2022 as more on par with this year. For BBREF BPM, it’s true that the 2022 season was closer to 2024 than it was to 2023, but this year was still below the 2021-2022 season in that measure. Overall, 2022 was above 2024 in box measures.

Of course, as you allude to, we all are aware that Steph had a really significant cold shooting streak for a lot of the 2021-2022 regular season. So that regular season looking worse in box metrics than the next year isn’t a big surprise. The fact that it still looked good in impact metrics is a reflection of the fact that his defensive impact was higher than normal that year—something that well reflected popular eye-test consensus about his defense that year. What happened with his offensive impact tracked quite well with what we see in the box metrics. For instance, in O-EPM, Steph’s 2022 season was his worst season since 2012-2013 and was the same as his 2024 season. You see a similar thing with LEBRON. There, Steph’s 2022 season was his worst overall since 2012-2013, which was on the back of a big drop in O-LEBRON. His O-LEBRON in 2022 was worse than the next season too (and not a lot better than 2024), but his D-LEBRON was good compared to the surrounding years, so his total LEBRON was still higher than 2023 (but worse than 2021, due to the offensive drop).

Basically, you raise issue with his impact in 2022 being higher than 2024 even though he had a relatively weak 2022 regular season in box terms, but if we dig into it we see that the primary driver of that was his noted defensive superiority that year—which, again, was something people actively talked about throughout that year. His *offensive* impact metrics look very similar between 2022 and 2024, with 2022 looking slightly better, which is consistent with his box metrics being slightly better in 2022 than 2024 as well. The data really doesn’t lead to the inferences you are suggesting.

Or to look at 2021, we have Curry’s comfortable third place BPM season after 2015/16, but much more distant by raw impact, and conparatively neutered by the metrics directly pulling raw impact. Maybe you have forgotten all the Warriors fans lamenting how Wiseman and Oubre were horrific vampires on Curry’s impact that year. Or maybe you simply shrugged and decided that random intangibles were the reason why he was not a top 5 player that year despite all the gaudy offensive production. Just like you will shrug if Curry plays exactly the same — hell, judging by 2022, maybe plays worse — but sees an uptick in all these spreadsheet outputs because the team struggles more without him.


Congratulations, you’ve discovered the fact that impact measures and box measures don’t always align perfectly in how they see players’ seasons! What a discovery! It’s a real mystery how a box measure might be more favorable to seasons where a player was on a team where they had greater responsibility to put up bigger box numbers because they no longer had a major scorer next to them. Apparently this must somehow mean that the impact measures are garbage when a player you don’t like did extremely well in them for many years?!

Impact measures such as EPM and LEBRON and others generally capture Steph having a pretty clear career arc where he became a genuinely good player in 2011-2012, had a prime that started in 2013-2014, and a peak that was 2015-2017, with the other years of his prime before and after all being pretty similar, followed by a steady decline in recent years. That’s broadly corroborated by box metrics as well, with a little bit more noise surrounding having an uptick in scoring responsibility due to lesser team talent (and even that depends on the box metric, with that not being seen in Thinking Basketball’s BPM). Not only are the impact metrics and box measures broadly corroborative of each other regarding Steph, but the years of his prime, peak, and decline are all wholly unsurprising as a career arc in terms of his age. I don’t understand why you think there’s anything in his career arc that would somehow suggest Steph’s amazing impact data was a mirage. You seem to just want to believe it and to have convinced yourself that it must be true.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
homecourtloss
RealGM
Posts: 10,804
And1: 17,770
Joined: Dec 29, 2012

Re: Brunson v Curry 

Post#49 » by homecourtloss » Sat May 4, 2024 12:43 am

AEnigma wrote:Amazing how a single offseason can make one player 60% worse because of “age” in a way that corresponds better to painfully obvious team changes than it does to anything they are doing on the court, while another player three years older and much more visibly declined can be adamantly equated to their prime form. Very principled.


Also how that same argument, i.e., age, isn’t used for another player. Hmmmm.
lessthanjake wrote:Kyrie was extremely impactful without LeBron, and basically had zero impact whatsoever if LeBron was on the court.

lessthanjake wrote: By playing in a way that prevents Kyrie from getting much impact, LeBron ensures that controlling for Kyrie has limited effect…
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#50 » by lessthanjake » Sat May 4, 2024 1:05 am

homecourtloss wrote:
AEnigma wrote:Amazing how a single offseason can make one player 60% worse because of “age” in a way that corresponds better to painfully obvious team changes than it does to anything they are doing on the court, while another player three years older and much more visibly declined can be adamantly equated to their prime form. Very principled.


Also how that same argument, i.e., age, isn’t used for another player. Hmmmm.


What in the world are you referring to? This is just a completely false insinuation. You’re obviously suggesting that I don’t talk about LeBron’s age as a mitigating factor we must take into account. But that’s just 100% false, and is refuted by just a boatload of posts of mine.

For instance, take the following quote from me from a week or so ago:

Spoiler:
I don’t think these years for LeBron really matter in any negative way on their own. They’re really just icing on the cake, where they can be a positive if he has success but can’t really be a negative if he doesn’t (because if he doesn’t have success then it’s still the case that other great players were either retired or not as good at this point anyways).

. . .

In a vacuum, LeBron doing that now isn’t really a negative to me, because I think one would still take a 39-year-old LeBron that pouts when things go badly over another 39-year-old player in NBA history. It doesn’t make him compare negatively to other players at his age.


Or how about another:

Spoiler:
After that, he had a few more consistent runs, followed by a bit of understandable inconsistency at a late age.


There’s many other examples I could give if I dug through my posts. You’ve invented a non-existent double-standard to try to criticize.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
AEnigma
Veteran
Posts: 2,853
And1: 4,432
Joined: Jul 24, 2022
 

Re: Brunson v Curry 

Post#51 » by AEnigma » Sat May 4, 2024 1:07 am

lessthanjake wrote:
AEnigma wrote:
lessthanjake wrote:Huh? He doesn’t look 60% worse, nor is his decline in impact numbers at all out of whack with his decline in box numbers. His decline in EPM between 2022-2023 and 2023-2024 was 19%, and his decline in LEBRON was 38%.

Crossed wires while typing. 60% as good, 40% worse. Either way, hardly feasible when watching, and also very much “out of whack” with impact measures. +2.51 RAPM in 2021 (#11) to +3.65 RAPM in 2022 (#2nd) to +2.4 RAPM in 2023 (#18) to +0.46 RAPM this year (#152). Really tracks well with the eye test, huh. No box score contradictions there.

This is just a sleight of hand, because no one is basing their views on single-season RAPM. As I’ve said many times, Single-season RAPM is largely garbage, precisely because it is way too noisy and way too dependent on on-off.

No, of course not, who would do that. But single season metrics that draw from overall team performance and lineup results? Sign me up!

As I already showed, Steph’s drop in better single-season impact metrics was consistent with the drop in box metrics.

If your definition of “consistency” extends only as far as “there was in fact a decline of some amount”.

And that’s while purely box-based stats like Basketball Reference BPM and Thinking Basketball BPM (i.e. both not directly affected by lineup combinations or on-off) has him going down 31% and 29% respectively. It’s all a pretty consistent picture of decline, regardless of whether or not the metrics are using impact data or not.

Both use team quality as a measure, which as far as I know requires the involvement of other player beyond Curry. And there too, I would not say 30% decline over an offseason is remotely supportable by the film, no.

2022 is a demonstrably worse box input year than 2021/23. It is more on par with this year, and to focus on BPM, a shift from 5.8 to 5.2 is reasonable enough as a season-to-season variation. But whereas 2022 was consistently measured as the highest pure “impact” season since Durant left, regardless of the individual decline in production, Curry’s “impact” this go around may as well be marginal. Same scoring (arguably better), but now a 50% decline in LEBRON.

Actually in Thinking Basketball BPM, Steph’s BPM was 3.4 this year, 4.8 last year, and 4.7 in 2021-2022, so that did not have 2022 as more on par with this year.

Never said it did. What I said was that the idea has Steph dropped off that hard was untenable by anyone invested in watching him play.

or BBREF BPM, it’s true that the 2022 season was closer to 2024 than it was to 2023, but this year was still below the 2021-2022 season in that measure. Overall, 2022 was above 2024 in box measures.

You have a gift for writing explanations for the sake of explaining nothing. At no point did I express confusion as to whether metrics preferred 2022. Either you continue to not care about tracking the conversation, or you are deliberately trying to maneuver past it.

On that note:
Basically, you raise issue with his impact in 2022 being higher than 2024 even though he had a relatively weak 2022 regular season in box terms,

Slick mischaracterisation, but no. Pay better attention.

Of course, as you allude to, we all are aware that Steph had a really significant cold shooting streak for a lot of the 2021-2022 regular season. So that regular season looking worse in box metrics than the next year isn’t a big surprise. The fact that it still looked good in impact metrics is a reflection of the fact that his defensive impact was higher than normal that year—something that well reflected popular eye-test consensus about his defense that year. What happened with his offensive impact tracked quite well with what we see in the box metrics. For instance, in O-EPM, Steph’s 2022 season was his worst season since 2012-2013 and was the same as his 2024 season. You see a similar thing with LEBRON. There, Steph’s 2022 season was his worst overall since 2012-2013, which was on the back of a big drop in O-LEBRON. His O-LEBRON in 2022 was worse than the next season too (and not a lot better than 2024), but his D-LEBRON was good compared to the surrounding years, so his total LEBRON was still higher than 2023 (but worse than 2021, due to the offensive drop)… but if we dig into it we see that the primary driver of that was his noted defensive superiority that year—which, again, was something people actively talked about throughout that year.

Imagine that, you did exactly what I predicted: gestured vaguely at defence as if the real point of separation was Curry suddenly becoming an all-defensive guard for one year, rather than having a decent defensive year buoyed by his best defensive cast since at least 2017 and the best overall cast since Durant left.

His *offensive* impact metrics look very similar between 2022 and 2024, with 2022 looking slightly better, which is consistent with his box metrics being slightly better in 2022 than 2024 as well. The data really doesn’t lead to the inferences you are suggesting.

You have seemingly made no actual effort to discern what I am suggesting.

Portending:
Or to look at 2021, we have Curry’s comfortable third place BPM season after 2015/16, but much more distant by raw impact, and comparatively neutered by the metrics directly pulling raw impact. Maybe you have forgotten all the Warriors fans lamenting how Wiseman and Oubre were horrific vampires on Curry’s impact that year. Or maybe you simply shrugged and decided that random intangibles were the reason why he was not a top 5 player that year despite all the gaudy offensive production. Just like you will shrug if Curry plays exactly the same — hell, judging by 2022, maybe plays worse — but sees an uptick in all these spreadsheet outputs because the team struggles more without him.

Congratulations, you’ve discovered the fact that impact measures and box measures don’t always align perfectly in how they see players’ seasons! What a discovery! It’s a real mystery how a box measure might be more favorable to seasons where a player was on a team where they had greater responsibility to put up bigger box numbers because they no longer had a major scorer next to them. Apparently this must somehow mean that the impact measures are garbage when a player you don’t like did extremely well in them for many years?!

Thanks for broadcasting that you read absolutely nothing I wrote about Brunson. :roll:

Impact measures such as EPM and LEBRON and others generally capture Steph having a pretty clear career arc where he became a genuinely good player in 2011-2012, had a prime that started in 2013-2014, and a peak that was 2015-2017, with the other years of his prime before and after all being pretty similar, followed by a steady decline in recent years. That’s broadly corroborated by box metrics as well, with a little bit more noise surrounding having an uptick in scoring responsibility due to lesser team talent (and even that depends on the box metric, with that not being seen in Thinking Basketball’s BPM). Not only are the impact metrics and box measures broadly corroborative of each other regarding Steph, but the years of his prime, peak, and decline are all wholly unsurprising as a career arc in terms of his age. I don’t understand why you think there’s anything in his career arc that would somehow suggest Steph’s amazing impact data was a mirage.

Because what you constantly describe is not merely an arc; what you regularly describe as it pertains to Steph’s “peak” is a narrow mountain.

These are arcs:
Image

No one is saying Steph did not decline. The question is how much. To be clear, I think even DARKO skews a bit too steep as a consequence of its plus/minus weighing (see: Kobe), but at least they are still leaving “declined Curry” ahead of most star guards.

People who care about real assessment will say Curry has declined a bit, because that is what has happened. He was a better scorer last year, but he is still one of the league’s best. What they will not do is try to invent a 30-40% drop-off not reflected in individual production or film.

You seem to just want to believe it and to have convinced yourself that it must be true.

And as always we end on the projection. What you want is a stunning peak, and the more you lean into measures emphasising prime on/off or plus/minus, the easier that becomes. But doing so inevitably also requires portraying 2021-23 Steph as a shadow of his former self, and the problem is nothing in his individual play remotely justifies that to any sufficient degree. So instead of being honest about the environmental influences on these metrics, we need to pretend massive age drop-offs exist where none do, that rotations do not matter, that team quality does not matter, that individual production does not matter, that film does not matter, and that your once favourite player really only had eight years of any particular worth — but oh, those first five years sure were unprecedented, huh.

Thus, for our viewing enjoyment, this thread presents us with the contrast between the Warriors fan who wants to dismissively say Steph is over the hill after a year offensively on par with his title-winning season two years ago, and the Warriors fan who has paid enough attention to see a guard given the ultimate opportunity to showcase “impact” and think, “Man, Steph could have really shined in that environment; too bad he is stuck with this awkward and rapidly degrading team instead.”
Doc MJ wrote:This is one of your trademark data-based arguments in which I sigh, go over to basketballreference, and then see all the ways you cherrypicked the data toward your prejudiced beliefs rather than actually using them to inform you
DorianRo
Senior
Posts: 722
And1: 619
Joined: Apr 20, 2023
       

Re: Brunson v Curry 

Post#52 » by DorianRo » Sat May 4, 2024 1:07 am

Ohh Brunson is gonna be a way better player than Steph. LOL. He just needs a better team around him.. Hell arguably he already is better. He can just do more offensively
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#53 » by lessthanjake » Sat May 4, 2024 1:37 am

AEnigma wrote:What you want is a stunning peak


I don’t want a stunning peak. I *got* a stunning peak. It happened. And you’re obviously just very invested in trying to handwave it away—likely because that stunning peak resulted in Steph being the clear impact king during his prime years, over a still-prime LeBron.

Anyways, your insistence that Steph is not meaningfully less good than before is just the result of you not realizing that, absent sudden declines caused by catastrophic injury, declines are usually subtle and actually *typically* don’t look *super* different than before, because the differences are just an accumulation of little things that can be individually hard to catch. The data does catch it though, and, as I’ve shown, it has caught it in both box and impact metrics in similar ways. Again, though, there really weren’t “massive” drop-offs here. The decline in impact data has been pretty steady year to year, with each recent year generally chipping away a bit more. And, indeed, impact metrics still indicate he’s a great player—just not as good as before (for instance, he still was 12th in EPM this season). As I said, none of this is surprising given his age, nor is any of his career trajectory surprising in terms of impact and box data. It’s all consistent with a very normal career trajectory (and particularly so for someone who struggled with injuries in his early years). He was a good player but not a superstar in his early 20s, then began his prime around age 25, had his peak years from ages 26-28, with the rest of his prime seasons before and after being similar to each other, and then has had a steady decline during his mid-30s. This is a really normal trajectory that is generally consistent with the data, and yet you seem super invested in acting like, contrary to essentially every player in NBA history, he’s not meaningfully less good in his mid-30’s than he was in earlier years and that his impact metrics in the past looking better was an undeserved mirage, and it’s his mid-30s impact data that is perhaps more indicative of his quality in his prime. Just absurd pretzel twisting.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
AEnigma
Veteran
Posts: 2,853
And1: 4,432
Joined: Jul 24, 2022
 

Re: Brunson v Curry 

Post#54 » by AEnigma » Sat May 4, 2024 3:26 am

lessthanjake wrote:
AEnigma wrote:What you want is a stunning peak

I don’t want a stunning peak. I *got* a stunning peak. It happened. And you’re obviously just very invested in trying to handwave it away—likely because that stunning peak resulted in Steph being the clear impact king during his prime years, over a still-prime LeBron.

Except it did not, and everyone outside of your Bay Area bubble recognises it did not, and unfortunately for us all you have given every indication that you can never, ever, ever get over it. :cry:

Anyways, your insistence that Steph is not meaningfully less good than before is just the result of you not realizing that, absent sudden declines caused by catastrophic injury, declines are usually subtle and actually *typically* don’t look *super* different than before, because the differences are just an accumulation of little things that can be individually hard to catch.

… Read this back to yourself. I realise literally all of that. That is my exact stance. You are the one taking a position that he has experienced a significant drop-off based entirely on abstractions; I am saying that those “little things” are why on a better team his “impact” would only look like 90% of what he was a couple of years ago.

The data does catch it though,

Such that he is now barely a neutral player? No, of course not, we know that is absurd. So instead what we do is smush that together with poor team results and a mild individual drop and pretend now all that together means he may very well be just a run of the mill all-star because his team is not going anywhere regardless. As always, no real analysis. Just blind consumption of whatever spreadsheet suits you best.

and, as I’ve shown, it has caught it in both box and impact metrics in similar ways.

“Similar” here too meaning “all declines are similar by virtue of being declines”.

Again, though, there really weren’t “massive” drop-offs here.

Such an honest description of a marginally perceptible 30-40% decline over a matter of months.

The decline in impact data has been pretty steady year to year,

And yet the impact correlates more closely with the fluctuations of his team than of his own performance.

based with each recent year generally chipping away a bit more. And, indeed, impact metrics still indicate he’s a great player—just not as good as before (for instance, he still was 12th in EPM this season).

And 30th in LEBRON among those with 1000 minutes played, and then if we care to look at per game “impact” (anathema to the Curry gospel, I know), even by EPM he is suddenly less valuable than players like Fred VanVleet. I would ask whether that matches your eye test, but by now I know there is no point, because there is none at play and your commitment to defending him from that type of implication left with Durant anyway.

As I said, none of this is surprising given his age,

Uh huh, he is older, therefore regardless of what he does, we cannot believe him to be this good still. We do not look, we just know that they are much worse now and must have been so much better before.

nor is any of his career trajectory surprising in terms of impact and box data. It’s all consistent with a very normal career trajectory (and particularly so for someone who struggled with injuries in his early years). He was a good player but not a superstar in his early 20s, then began his prime around age 25, had his peak years from ages 26-28, with the rest of his prime seasons before and after being similar to each other, and then has had a steady decline during his mid-30s. This is a really normal trajectory that is generally consistent with the data,

This would be correct if it were coming from someone who had any sense of his actual trajectory rather than the trajectory of some fictionalised program. But of course it is easy for an abstraction to look correct. Bit different when you specify to, “Overnight he became the greatest player ever for five years, then his team dropped off and he coincidentally became a run-of-the mill all-NBAer while playing pretty close to the same as usual.”

and yet you seem super invested in acting like, contrary to essentially every player in NBA history, he’s not meaningfully less good in his mid-30’s than he was in earlier years and that his impact metrics in the past looking better was an undeserved mirage,

Yay, more gestures at being old therefore by rule not good, please please please please do not actually check to see what how he is performing and next to whom.

and it’s his mid-30s impact data that is perhaps more indicative of his quality in his prime.

Also never said this.

What I am saying is that anyone bothering to watch him would not be seeing a mild decline between those eras. It is nothing to do with which is “more” indicative. Players are less than the product of their best environments, just as they are more than the product of bad environments. What is indicative is what the player does. But the more we focus on what the player actually does, the less you get to inflate him based on everything he does not control, and that is evidently more important to you than watching a player you will otherwise defend rabidly and seeing that he has been changing very gradually. Normal fanbase would celebrate, but normal fanbases are not this at odds with what everyone else can see.

Just absurd pretzel twisting.

Love this tradition. Excited to see what projection will be next.
Doc MJ wrote:This is one of your trademark data-based arguments in which I sigh, go over to basketballreference, and then see all the ways you cherrypicked the data toward your prejudiced beliefs rather than actually using them to inform you
falcolombardi
General Manager
Posts: 8,502
And1: 6,025
Joined: Apr 13, 2021
       

Re: Brunson v Curry 

Post#55 » by falcolombardi » Sat May 4, 2024 3:43 am

lessthanjake wrote:
AEnigma wrote:What you want is a stunning peak


I don’t want a stunning peak. I *got* a stunning peak. It happened. And you’re obviously just very invested in trying to handwave it away—likely because that stunning peak resulted in Steph being the clear impact king during his prime years, over a still-prime LeBron.


Are you so invested to put curry over lebron in an arbitrary stat for an arbitrary time window that you are now trying to downplay and diminish your own favorite player?
capfan33
Senior
Posts: 710
And1: 571
Joined: May 21, 2022
 

Re: Brunson v Curry 

Post#56 » by capfan33 » Sat May 4, 2024 3:52 am

falcolombardi wrote:
lessthanjake wrote:
AEnigma wrote:What you want is a stunning peak


I don’t want a stunning peak. I *got* a stunning peak. It happened. And you’re obviously just very invested in trying to handwave it away—likely because that stunning peak resulted in Steph being the clear impact king during his prime years, over a still-prime LeBron.


Are you so invested to put curry over lebron in an arbitrary stat for an arbitrary time window that you are now trying to downplay and diminish your own favorite player?


I mean, he's not the goat point guard, he's not the best offensive player ever for either peak or prime if you value the playoffs at all, gotta get him in there somehow.
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#57 » by lessthanjake » Sat May 4, 2024 4:25 am

AEnigma wrote:.


I am definitely not going to go through and respond to like 12 different comments, the vast majority of which say nothing that I haven’t already addressed in prior responses. Others can read my prior posts and see what the refutation of virtually all of this would be.

Anyways, what I will say is: The bottom line is that, as much as it obviously would pain you to admit it, Stephen Curry in his prime was demonstrably the impact king over a still-prime LeBron. I have demonstrated this quite rigorously in many posts in the past. In his mid-30’s he is unsurprisingly no longer the impact king. No one in their right mind would expect him to be. Your seeming attempt to suggest him no longer being the impact king in his mid-30’s shows he was never *really* the impact king in his earlier years is just plainly silly. He was the impact king of that era, and it wasn’t even particularly close. He is in many ways a very similar player now, but lots of little things go with age—quickness, burst, stamina, reaction times, vision, etc., not to mention just the accumulation of knocks and injuries over the years taking their toll. Those sorts of things together end up being significant for everyone, and Steph is clearly no different—with his impact and box numbers having notably declined from his younger years, as we would expect. This idea that we wouldn’t expect this scale of decline due to subtle things is just silly. Players routinely have significant jumps (both increases and decreases) in impact when they are near the start and end of their prime. Just to take an analogous player, Ray Allen’s LEBRON went down 42% from age 34 to age 35, despite his box numbers looking extremely similar and the team itself being similar (and even doing better). The metric was in significant part detecting Allen’s age-related decline—with lots of subtle factors surely getting worse and reducing impact. Your entire argument seems premised on the idea that something perfectly normal is actually abnormal, followed by you drawing the most negative possible inference from this silly idea, in order to get to a convenient conclusion. Finally, I’ll note that your seeming assumption that Steph’s decline is not a significant cause of his team as a whole not doing as well is bizarre. Steph not being as good is a major factor in his team not doing as well, and yet you’re somehow essentially taking his team not doing as well as evidence that he hasn’t really declined much!

EDIT: By the way, I find it amusing that you’re acting like data that really just shows Steph going down by about like two points in impact in the last couple years is some massive decrease that we must inherently be skeptical of, while having also just made repeated sarcastic responses in another thread when I suggested that a point or two per game of positive shooting variance is significant. I guess a couple points of impact is either massive or laughably small depending on what’s convenient for you. Of course, if we’re actually trying to be consistent, the reality is that that kind of change in impact is always a significant amount but not so massive a difference that it would be implausible for age to account for it. My arguments across both discussions have been entirely consistent with that (indeed, in the other thread, I specifically said I think other advantages prime LeBron had over current LeBron would’ve more than made up for it), while your posts…really have not, and have instead exhibited a real double standard that you somehow feel warranted in accentuating with personal attacks in both threads.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
lessthanjake
Sixth Man
Posts: 1,629
And1: 1,393
Joined: Apr 13, 2013

Re: Brunson v Curry 

Post#58 » by lessthanjake » Sat May 4, 2024 4:31 am

falcolombardi wrote:
lessthanjake wrote:
AEnigma wrote:What you want is a stunning peak


I don’t want a stunning peak. I *got* a stunning peak. It happened. And you’re obviously just very invested in trying to handwave it away—likely because that stunning peak resulted in Steph being the clear impact king during his prime years, over a still-prime LeBron.


Are you so invested to put curry over lebron in an arbitrary stat for an arbitrary time window that you are now trying to downplay and diminish your own favorite player?


He’s not my favorite player, and I’m only invested in accurately describing reality. Steph Curry is in his mid-30s and is very predictably not as good as he was before. I feel like it’s crazy that people are seemingly so desperate to discredit his prior achievements that they are incredulously arguing that a player in his mid-30’s must not have meaningfully declined.
OhayoKD wrote:Lebron contributes more to all the phases of play than Messi does. And he is of course a defensive anchor unlike messi.
User avatar
AEnigma
Veteran
Posts: 2,853
And1: 4,432
Joined: Jul 24, 2022
 

Re: Brunson v Curry 

Post#59 » by AEnigma » Sat May 4, 2024 5:05 am

lessthanjake wrote:
AEnigma wrote:.

I am definitely not going to go through and respond to like 12 different comments, the vast majority of which say nothing that I haven’t already addressed in prior responses. Others can read my prior posts and see what the refutation of virtually all of this would be.

Yep, which is why I am always thrilled with your willingness to put your manipulations on record.

Anyways, what I will say is: The bottom line is that, as much as it obviously would pain you to admit it, Stephen Curry in his prime was demonstrably the impact king over a still-prime LeBron. I have demonstrated this quite rigorously in many posts in the past.

With a façade of a demonstration everyone outside the fanbase recognises as manure. And maybe eventually you will stumble upon a manipulation that does not immediately singe the hair off everyone’s nostrils, but until then, looks like you are going to be stuck with the fumes.

In his mid-30’s he is unsurprisingly no longer the impact king. No one in their right mind would expect him to be. Your seeming attempt to suggest him no longer being the impact king in his mid-30’s shows he was never *really* the impact king in his earlier years is just plainly silly.

It is also an invented position by someone who consistently would rather bluster on about nothing than read what was written.

He was the impact king of that era, and it wasn’t even particularly close.

Nothing sadder than logging on a basketball forum to lie.

He is in many ways a very similar player now, but lots of little things go with age—quickness, burst, stamina, reaction times, vision, etc., not to mention just the accumulation of knocks and injuries over the years taking their toll. Those sorts of things together end up being significant for everyone, and Steph is clearly no different—with his impact and box numbers having notably declined from his younger years, as we would expect.

More of the usual abstractions.

This idea that we wouldn’t expect this scale of decline due to subtle things is just silly. Players routinely have significant jumps (both increases and decreases) in impact when they are near the start and end of their prime.

Right, like Jalen Brunson leaping directly into an unparalleled new prime the second Quickley was traded.

Cannot believe you decided to wholly embrace my joke description. Chris Paul joined the team, and he made Steph a meaningfully worse player. The ultimate revenge.

Just to take an analogous player, Ray Allen’s LEBRON went down 42% from age 34 to age 35, despite his box numbers looking extremely similar and the team itself being similar (and even doing better). The metric was in significant part detecting Allen’s age-related decline—with lots of subtle factors surely getting worse and reducing impact. Your entire argument seems premised on the idea that something perfectly normal is actually abnormal, followed by you drawing the most negative possible inference from this silly idea, in order to get to a convenient conclusion.

Please read your posts before submitting; this level of self-obliviousness is going to give me a stroke if I see much more of it.

“This player looked the same. But actually he was 40% worse in ways I cannot articulate in any real sense. How convenient and abnormal and negative it is for you to deny that.”
Spoiler:
AEnigma wrote:Seasonal variance and age curves do not result in these massive impact fluctuations or drops the way APM metrics may suggest. If a player’s production holds firm, that should clue you in to there being an alternative explanation. The best players tend to hold onto more their value than what “impact” and production dips may otherwise indicate. There inevitably is a time where that stops being true, but people seem weirdly eager to get ahead of themselves on those declines. Again often a case of vaguely gesturing at minor declines and overselling their real effect on a team.
Finally, I’ll note that your seeming assumption that Steph’s decline is not a significant cause of his team as a whole not doing as well is bizarre. Steph not being as good is a major factor in his team not doing as well, and yet you’re somehow taking his team not doing as well as evidence that he hasn’t really declined much!

What is bizarre is how every other Warriors fan has been able to see the decline around Steph. Not you, though. You understand the long game: all team results belong entirely to Steph.
Doc MJ wrote:This is one of your trademark data-based arguments in which I sigh, go over to basketballreference, and then see all the ways you cherrypicked the data toward your prejudiced beliefs rather than actually using them to inform you
User avatar
spree8
RealGM
Posts: 14,324
And1: 6,392
Joined: Jun 05, 2001
     

Re: Brunson v Curry 

Post#60 » by spree8 » Sat May 4, 2024 5:34 am

MiamiBulls wrote: In Brunson's case, outright grifting for Fouls calls.


No he doesn’t. I’ll let Coach Nick and NBA Head Ref Ronnie Nunn explain…


;si=1uGV7XvXFPqD1CP5
Image

Return to Player Comparisons