DraymondGold wrote:OhayoKD wrote:DraymondGold wrote:this is a leap in logic
firstly, the point of scalability is you mantain value on better teams. The 16/17 cavs were on the same level in the playoffs as mj's bulls. the 15 cavs were on the same level as the 89/90 bulls. Jordan getting to the same spot with better teammates isn't a point in his favor.
Here are the top teams by composite (regular-season/playoff) SRS:
1. 1996 Jordan
2. 1991 Jordan
3. 1992 Jordan
4. 1997 Jordan
5. 2016 LeBron
6. 2012 LeBron
7. 1998 Jordan
8. 2013 LeBron
9. 2009 LeBron
10. 2020 LeBron
11. 2017 LeBron
12. 1993 Jordan
13. 2011 Miami Heat
14. 2015 LeBron
So by Overall SRS, Jordan has played on 4 better teams than LeBron, and the top 2 are better by a wide margin. This is a list that weights playoffs highly. Taking playoffs only, the 91 Bulls and 96 Bulls also gain separation from any of LeBron’s teams (including the 16/17 Cavs), and playoff-only SRS is the kind of stat that could overrate the 16 Cavs slightly as it wouldn’t account for opponent injury. You mention the 12 Heat were as good in the playoffs as the 91 Bulls, but unfortunately that’s flat out not true and the margin isn’t insignificant per SRS.
In sum: I’m not sure LeBron has played on as good teams as MJ.
Now you may argue: they’re as good when LeBron’s on the court. They just fall apart when LeBron’s off! But first, it’s worth noting that 89-93 Jordan has a better on-rating than LeBron’s best sample in his second cavs stint.
Second… that’s exactly my point!
Let me introduce a slight change in phrasing for your definition for scalability. I’m not married to this different definition, but it may more clearly indicate the point I’m making. Rather than saying “scalability is how well you perform on better teams” (like resilience is how well you perform against better teams), let’s change the phrasing to: “how well you perform with better teammates.”
Well, when we look at these cases, per above, LeBron faces diminishing returns with better teammates, as shown in having generally worse differential in his on minutes. [I can also link to a more in-depth analysis of how LeBron’s scoring specifically declined with better teammates, and this is a bigger decline than players who are usually seen as more scalable like Curry].
Why does this matter? Well, you’re probably going to need to play with better teammates to improve your chances of winning a championship. If you face less diminishing returns with these teammates, your on-minutes will be far more dominant. If you play with better teammates, your off minutes won’t be as bad. These will help your teams win championships. So if there’s scalability concerns, that would limit LeBron’s ability to play on better teams / with better teammates, which would slightly hurt his championship odds.
I will note that Thinking Basketball has made this direct comparison of how Jordan and LeBron would perform on better teams / with better teammates, and he’s argued that Jordan would perform better. He used available data and film analysis to make the case that Jordan’s more scalable than LeBron. It’s an interesting argument, and perhaps one that would be fun to discuss if we have time (not sure personally), but I just want to say this is a different argument from whether LeBron has scalability concerns at all.
Keep in mind that it's only really close between the two using box-score heavy stuff. In aupm, rapm, pipm, ect, Lebron has a signifcant advantage. And if you go with raw impact stuff the gap only widens. This is true on multiple different rosters in multiple different situations(second stint cavs, heatles, first stint cavs, spacing/no spacing, ect.)
. This one’s interesting. It’s possible box stats have a bias for Jordan.
But I’m not sure it’s true that LeBron has the significant pure plus-minus advantage, at least not at his peak (though maybe over the course of his prime). Jordan’s 3 year playoff AuPM is better than any of LeBron’s runs. Jordan’s 3 year playoff PIPM estimate is better than LeBron’s Miami years (haven’t checked his other years). And Jordan’s partial RAPM sample in 91 is better than LeBron’s RAPM in 13.
1. It goes both ways. The warriors are going to be underrated based on not going all-out in the regular season, especially in 2017. Even in 2016 their starters played way less minuites during the regular season. The cavs and the warriros both weren't really going all-out save for one or two matchups in the playoffs. Ultimately even with the injury to curry, the warriors beat an okc team that played 67 win basketball when healthy and decisively beat down the 70 win srs spurs. And in the final the series was pretty close to a tie. Then the cavs seem to have improved the following postseason. Probably fair to have the cavs as a 66-70 ish level playoff team even if you take their srs at face value.
Overall probably does increase the gap for the bulls but if we're talking the championshp probability, gaps in regular season score for a top seed probably don't make a signifcant difference. Like, what situation are you imagining where 67+ win playoff basketball isn't good enough for the title? Is cieling raising defined as your value on a team capable of beating the 17 warriors?
Like really the cavs were probably coasting for the first three rounds(their relative defensive rating increasing dramatically as they played stronger offenses supports this) and they were blowing teams of comparable srs to the first three point bulls by similar point differentials. Then they push a 67 win team without co-stars in 2015 and beat a 70 win team the next year. Then they get even better in 2017.
2. When bosh wade and lebron were in the lineup the 12 heatles went 8-1 winning by a margin of 15 points vs the 48 srs celtics and by a margin of 9 points over the 58 win pace thunder. That seems 91-level to me. Now consider that the co-stars aren't supposed to be good fits, their spacing was pretty bad, and that they were dealing with constant injury thoughout. Don't think 2012 really supports your theory that well. They had lots of issues/fit stuff to work out and yet when it came to time to win they crushed everyone when they shared the court. Notably they were much more dominant than the 13 heat in the playoffs vs comparably strong opponents. Despite the 13 heat having alot more spacing and lebron supposedly being alot more portable.
In 2011 lebron had one of the worst series of his career in the final for a close loss to a team that had smoked three 55+ win teams before them. Before that they were operating at a pretty similar level to the first three peat bulls. From reporting that was when their system was completely improvised and the spacing was pretty bad. Idrk if it makes sense to extrapolate that lebron
at his best gets worse team results with better teammates. The 2012 heat with wade/bosh/lebron sharing the court were way more dominant than the 2013 heat despite
a. The 2013 heat having much better spacing
b. lebron allegedly being much easier to build around due to his shooting improvement
The 2011 heat were on pace for a better postseason than the 2013 heat until a lesser lebron had an outlier bad performance.
The 2020 Lakers were waaay more dominant than the 2013 heat in the playoffs. Was that because lebron was way better?
Is "lebron's teams get worse when the teammates get better" really the right conclusion here? Idk.
3. I'm specifically arguing that being "more scalable" does not mean "better cieling raiser". To be a better cieling raiser you have to be (generally) more valuable on a certain treshold of team. A player can be less scalable AND a better cieling raiser if his value is still higher than whoever you're comparing him to.
4. I think looking at how the individual years in general instead of trying to force a comparison between specific years makes this clearer:
Lebron's 09, 10, 12, 13, 16 and 17 are at +10, +8, +7.5, +7, +7.5, and +7.7 respectively.
You get 5 different seasons which would all boost the average of MJ's 89-91 peak.
[/quote]
Mj's average aupm for that three year stretch is +7 his average bpm is +7.9. The average between the two(what ben uses for lebron's years) would be +7.35. In the quoted section you get
6 years in a 9 year span that beat that. Two which beat any mj year outright.
Mj's "parital" rapm samples peak at +7, lebron has three different seasons that go at +10 and one that goes at +9.
And then just using the sample without mregulization(which tends to overdistribute value to role players), one player is taking a 27 win team to 48-50 wins while the other has a 5-25 team winning 60+ in his first cleveland stint, a 40 win team to 60 in his miami stint, and a 25-30 win team to 50-60 in his second cleveland stint.
Unless I'm missing something the only measures their best years lineup favorably or comparably for mj is box-score aggregates(jordan generally gets an rs edge, lebron gets a playoff edge, though per has them even in the rs and favors ps lebron) and ben's bpm where jordan has better individual years but lebron has the best one year peak.
5. It's probably worth noting jordan's own indivudal impact stuff drops as of 91 and he's less effecient at basically everything in 91 relative to 90. I'll let you decide if that's jordan getting worse, if that's a scalability thing,, or a bit of both