The-Power wrote:Valid and good thoughts. I'll just respond to some of them.
OhayoKD wrote:Curry's scoring advantage is baked into the the "winning" stuff which favors lebron even though, at least theoretically, this year should suppress that(if you buy Lebron is especially dependent on spacing, and can't function as well next to heliocentric playmakers like Westbrook(this is basically the main justification analysts have for not taking his apparent impact at face-value)).
I think it's fair to be a bit more cautious about single-year plus-minus data on the whole, especially when we're talking about players with many years worth of data that paint a clear picture.
That's fair enough. 1 year samples are not optimal for any metric, even "stabilized" ones, so I can't really have an issue with a cautious approach there. That said...
Well, let's just assume that all the advantages above are true. They were true in years before, too. Yet especially over the past couple years – at least if I recall correctly – the gap between Curry and LeBron in your preferred metrics was smaller than today. This is despite LeBron seemingly falling off more in terms of boxscore composites than Curry.
Health. The explanation you're looking for is probably health. Lebron posted arguably the best impact stuff of either of the two since he came to LA when healthy in the 2020 regular season. He looked like one of the most valuable players in the league in 2019 before injured(iirc he was actualy 2nd behind Giannis in various all-in=ones before his injury vs the Warriors). He was also looking like the most valuable player in the league in 2021 before getting hurt.
2023 is the healthiest Lebron Laker year outside of 2020. They look comparable over these last few years without health accounted for, so it stands to reason a season where Lebron is healthier than usual and is as available as Steph might favor Lebron. It's also worth noting this "closeness" really only is present in the regular season. In the playoffs Lebron's teams consistently get better along with his own metrics, and when you account for minutes played playoff Lebron is well ahead of playoff Steph(he looks more valuable per-minute too outside of 2015).
Another factor to consider is colinearity. Draymond has actually been at or ahead of Curry in terms of plus/minus since he emerged as a co-star under Kerr. Steph's minutes are heavily tied to Draymond which we'd expect to juice Curry's numbers(best players share court together, than team without those best players gets dramatically worse). Lebron has not had his minutes tied to Davis to a similar degree(some of that is staggering, some of that is injury), so Lebron would see a similar emperical boost.
I don't have an issue focusing on a larger sample as opposed to a one-year one, but I think healthy Lebron still has an advantage there and Lebron has been as healthy as Steph this year
None of the meta-thoughts presented above can account for that because they are not limited to this year. I'm also not sure whether it's true that historically, LeBron's impact-relative-to-boxscore numbers are greater than Curry's. Do you have anything to back that up? Because if not, there is an obvious issue with the archetype comparison as, again, this is not a one-year perspective.
I'd say we have
lots backing that up. I compared helio and non-helio offenses in the post I linked above relative to their box stuff(including the exceptional results we see with college-age Lebron), but let's start with a quick profile of Lebron's impact profile using lineup-adjusted and raw data(the former is less noisy, but when we're trying to compare impact between Lebron and Curry, lineup-adjusted stuff doesn't really make much sense to use in 1-year comps because they will misattribute value to role players once you hit a certain treshold):
https://forums.realgm.com/boards/viewtopic.php?p=104353387#p104353387It's a long post but you can skip to the bottom or control+f "IMPACT BREAKDOWN" for a quick summary
What should jump out immediately is there is a massive discrepancy between Lebron's box-profile and Lebron's impact profile. By box, 2016 Steph is arguably better than any lebron regular season and miles ahead of Lebron's 15-17. By impact, Lebron's 09/10 are on a completely different level, and 12, 13, 15-17, and 20 are all comparable(you can get an advantage or disadvantage depending on what you're using). The whole concept of Curry as a "GOAT-lvl" regular season player is basically built around him being "stat-breaking" in 2016, but 2016 isn't nearly as remarkable looking when you take an impact approach with the best KG, Duncan, and Shaq years all looking comparable and the best Lebron years looking a tier above.
If you want to look at adjusted metrics, Lebron is basically a tier onto himself while Curry isn't particularly special relative to data-ball contemporaries like Giannis, KG, Shaq, Jokic, Chris Paul, and Duncan. In fact the closest player in rs impact stuff to Lebron is not Curry but KG and even then he really doesn't come that close:
2018 is the 25th season of league-wide plus-minus data, which covers nearly 40 percent of the shot-clock era and touches 12 of the top-20 players on this list. None have achieved LeBron’s heights: He holds four of the top-five scaled APM seasons on record, and six of the top eight. Since 2007, 10 of his 11 years land in the 99th percentile.
Most RAPM-Data sets have Lebron miles ahead of anyone post-1997. Using Ben Taylor's own 5-year RAPM, the gap between Lebron and 2nd place KG is as big as the gap between KG and 7th place Steve Nash. Looking at his prime as a whole...
You have to go back to Kareem for any semblance of a comparable value-profile and I think Russell is the only player with a strong argument along those lines. On the other hand, taking a box-approach, keeping in mind we can only make like for like comparisons starting at the mid 70's, Lebron's best regular seasons are not outliers relative to the best stuff from his contemporaries including Steph.
This isn't to say you should always assume Lebron is underrated via box or that Curry is always overrated(Curry's box-aggregates stopped looking so gaudy), but the offensive and defensive results for Lebron's teams have generally outpaced his box-portfolio and there are things Lebron does on both ends which are not captured by box-metrics of any kind.
I also don't really see any basis to suggest that Curry's archetype limits his offensive impact compared to others and that his boxscore numbers overstate his impact. This is also where archetype-discussions become flawed, because we have never seen someone quite like Curry and even of the more ‘traditional’ archetype, there are so few at the elite level that it's tough for me to throw them into one box and derive generalized statements from it.
[/quote]
Curry potentially off-sets the weakness of the other "non-helios" with being able to create alot without completing a final pass, but I think there's film, theoretical, and empirical basis to assume that being a lesser ball-handler, a lesser passer, and a lesser floor general is going to be disadvantageous. I think an archetype which does less at an elite-level than another archetype is generally going to be less valuable.
Regardless, I think I've provided some
basis in the posts I've linked/this thread. Feel free to scrutinize/challenge that.
