RealGM 2020 Top 100 Project: #2 (Michael Jordan)

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

limbo
Starter
Posts: 2,029
And1: 1,827
Joined: Jun 30, 2019
 

Re: RealGM 2020 Top 100 Project: #2 

Post#121 » by limbo » Sun Oct 18, 2020 4:43 pm

Ainosterhaspie wrote:I dont see 12. Maybe 10, but I think it's 9: 88-93, 96-98. Still had a losing record in 87. Great players seem to be able to manage winning records with just about any cast. I don't expect a lot come playoffs with a bad cast, but a 4-6 seed, easier first round match-up and first round win seems like something at great player whose game is mature enough should be able to pull off as a baseline.


off the top of my head

Baylor '61
Wilt '63
West '67
Oscar '67, '68, '69
Kareem '75, '76
Barkley '88, '92
Hakeem '88, '89, '92
Kobe '05
KG '06, '07
AD '19
The Master
Junior
Posts: 334
And1: 475
Joined: Dec 30, 2016

Re: RealGM 2020 Top 100 Project: #2 

Post#122 » by The Master » Sun Oct 18, 2020 5:02 pm

Ainosterhaspie wrote:Still had a losing record in 87. Great players seem to be able to manage winning records with just about any cast. I don't expect a lot come playoffs with a bad cast, but a 4-6 seed, easier first round match-up and first round win seems like something at great player whose game is mature enough should be able to pull off as a baseline.

Losing record in 23-team league is something different than nowadays. Bulls had 8th SRS and projected record of 43-39, but they played in much more talented conference, and back then teams played more games within its conference than today. It definitely wasn't all-time great carriage of subpar supporting cast, and that's perhaps one of the most underestimated aspects of LeBron v. Jordan comparison (floor raising in below average environment), but not taking Bulls to 2nd round in much better conference is no story in terms of player's evaluation, especially considering MJ was 'just' top3 player that year.
DQuinn1575
Pro Prospect
Posts: 856
And1: 231
Joined: Feb 20, 2014

Re: RealGM 2020 Top 100 Project: #2 

Post#123 » by DQuinn1575 » Sun Oct 18, 2020 5:22 pm

The Master wrote:
Ainosterhaspie wrote:Still had a losing record in 87. Great players seem to be able to manage winning records with just about any cast. I don't expect a lot come playoffs with a bad cast, but a 4-6 seed, easier first round match-up and first round win seems like something at great player whose game is mature enough should be able to pull off as a baseline.

Losing record in 23-team league is something different than nowadays. Bulls had 8th SRS and projected record of 43-39, but they played in much more talented conference, and back then teams played more games within its conference than today. It definitely wasn't all-time great carriage of subpar supporting cast, and that's perhaps one of the most underestimated aspects of LeBron v. Jordan comparison (floor raising in below average environment), but not taking Bulls to 2nd round in much better conference is no story in terms of player's evaluation, especially considering MJ was 'just' top3 player that year.


86 Bulls were 21-43 without MJ, so 27 win team, they improve 13 games with no other real changes.
Defense goes from worst in league to average.
Start 23 yo Oakley, plus marginal starters in Banks, Paxson, Corzine.
Thought of enough at the time to get MVP votes away from Magic, who was at his peak, and Magic won some votes as lifetime achivement, and beat out Bird in MVP voting.
Really hard not to call his highest scoring season, when playing for an average team, and playing above average defense not a great season.
He's 37/5/5 - Kobe was 35/5/5 in his highest season of WS, VORP, BPM



stats from Basketball-Reference
The Master
Junior
Posts: 334
And1: 475
Joined: Dec 30, 2016

Re: RealGM 2020 Top 100 Project: #2 

Post#124 » by The Master » Sun Oct 18, 2020 5:35 pm

DQuinn1575 wrote:86 Bulls were 21-43 without MJ, so 27 win team, they improve 13 games with no other real changes.
Defense goes from worst in league to average. Start 23 yo Oakley, plus marginal starters in Banks, Paxson, Corzine. Thought of enough at the time to get MVP votes away from Magic, who was at his peak, and Magic won some votes as lifetime achivement, and beat out Bird in MVP voting. Really hard not to call his highest scoring season, when playing for an average team, and playing above average defense not a great season. He's 37/5/5 - Kobe was 35/5/5 in his highest season of WS, VORP, BPM

That regular season was a statistical equivalent of LeBron's 2007/08 season, so I agree with you to some degree, but I think it was one year before beginning of their actual prime. But I see your point.
drza
Analyst
Posts: 3,454
And1: 1,710
Joined: May 22, 2001

Re: RealGM 2020 Top 100 Project: #2 

Post#125 » by drza » Sun Oct 18, 2020 6:29 pm

(ETA: lol, TRex called the thread while I was typing this. I thought about that possibility and went to make a 'placeholder' post, but when I last hit 'preview' I saw that TRex had called the thread a minute earlier, so...yeah. I guess this whole post will be "after the final whistle", as it were. I'll also probably recycle the lion-share of it for my first Russell vs Kareem post next thread, so it can be seen. C'est la vie!)

Count me as another for whom the 2-day window makes things difficult. I understand the reason for it (and am not even arguing against it), I'm just pointing out why in this (and likely future) threads I may miss the actual heart of the argument. I'd intended to go for a big "Jordan vs Russell" post here, with MJ's entry a foregone conclusion (and LeBron already in) I figured this might be my only opportunity to do this. On the other hand, this thread is effectively over so I don't know if it's worth it to put in the time to really dig in at this point. :Shrugs: Let's just free-write and see where this goes.

Michael Jordan vs Bill Russell

I wrote in the #1 thread about some of my general thoughts on Russell, and Jordan vs LeBron. It's always interesting to see how others react to a post after it's left my fingers, and to get the overall feel for the group's mindset on the whole. Thus far, with Russell in particular, I've noticed either a) a willingness to essentially stipulate Russell's in-era impact as (either potentially or definitely) the biggest ever while also giving demerits based on differences in era, or b) some question as to whether Russell really WAS worthy of as much credit as he's given for the Celtics' success due to the lack of granularity of information from the era.

Both mindsets have merit, are complicated, and therefore can be difficult to counter. I spent some time on the first in my first Russell post, so I'll spend some time on the latter here: how confidently can we estimate Russell's impact?

As I mentioned before (and likely will mention again in this project), by first really deep dive into Russell (and most of the historical players before my time) came in the Retro Player of the Year project Doc MJ initiated about a decade ago. Last post I stated this then jumped to the conclusion I reached that Russell really was the main engine behind that era of Celtics dominance, with a reference to the timeline of Celtics' team defenses, and how it correlated with Russell's absence and presence, as the only real support data. I'd done that for speed, and so-as not to bog down the point I was trying to make with words and numbers (yes, shockingly, I'm known to do that :P ). But, interestingly, I was called out for that pretty vigorously. So be it, let's dig a little deeper.

Russell's in-era impact, deeper dive
First, for any that haven't already, I'd strongly urge you to go through that RPoY project. Especially the "early years" (which were actually the later weeks of the project). Because the depth of discussion and data presented in that project, especially for the time in which we did it, was unparallelled in my opinion. It wasn't just a bunch of guys spouting box scores and ringzzz count. We had guys looking up and pasting newspaper articles describing individual games and thoughts at the time. Quotes from relevant books through the years, from NBA players and personnel from the era. Quotes from Russell's books. And a lot of genuine, innovative quantitative analysis that was well beyond the norm for even professional basketball analysts. Outside of having an NBA League Pass or databall-era stats from the 60s, I really felt like those discussions were enough to give me a much clearer insight into previous generations than I'd ever had before. Also, some really cool blogs/careers/analysts spawned from that project. Doc MJ had his blog up and going for awhile, and produced some really good content (that I'm likely to quote at various points throughout this project). And of course, Ben (known here as ElGee) kept going with those new analytics approaches and has put out a ton of good analysis since.

In fact, let's start with a quote that originally came from one of Russell's books but that I encountered on Doc MJ's blog, A Substitute for War:

"Bill understood that Wilt’s game was more vertical, that is, from the floor to the basket. Wilt’s game was one of strength and power…Bill’s game was built on finesse and speed, what he called a horizontal game, as he moved back and forth across the court blocking shots, running the floor, and playing team defense."


This concept of Horizontal Defense is key, here, because a) it tends to get overshadowed by the blocked shot and b) it's part and parcel of why I believe Russell's defense would've been hugely impactful, even today. Russell is one of (the?) innovators for jumping to block shots, and every mining of available game film yields cartoonish blocked shot rates for Russell in his career (like 8-9 blocks/game, regularly, type cartoonish). However, it's the horizontal element...the covering of large areas of the court, the help defense, the ability to blow-up the early attempts at the pick-and-roll that made Russell a 1-of-1 defender in his era and would still be successful today.

People tend to focus on shot-blocking...so-called "rim protecting", as the primary defensive attribute of great defensive anchors. I'd argue that's completely not so, that in every era the horizontal defending big men have been more valuable defenders, and that in the current era it's still the case.

Wilt may have blocked as many shots as Russell, but Russell was the defensive force.

Kareem was a great shot-blocker, but Walton's ability to cover the court defensively (when healthy) allowed him to have a much greater impact at that end of the court.

Mark Eaton won DPoY awards, but give me the greater mobility and court coverage of Hakeem or Robinson any day.

Kevin Garnett, Ben Wallace and even Tim Duncan had the mobility and covered a wider area of the court than 3-time DPoY Dwight Howard would a few years later, setting them apart.

And Draymond Green was dominating defensive RAPM numbers a few years back and may deserve more than the one DPoY award he received.

As for putting some numbers to it. Well, first, I'd recommend you checking out ElGee's writeup of Russell in his rankings list ( https://backpicks.com/2018/04/02/backpicks-goat-3-bill-russell/ )...it's like a smorgasbord of videos, scouting, analysis and unique analytics. But focusing in on the numbers, despite the lack of databall-era stats, there are some compelling cases to be mae numerical for Russell's dominance:

1) As posted before, the Celtics' team defense was a historic dynasty, and the time scale of that dominance correlated almost exactly with Russell's career: https://elgee35.wordpress.com/2010/12/31/bill-russells-defensive-impact/

2) A quote from that ElGee writeup I just linked:
"He didn’t miss much time in his career, so WOWY numbers are hard to come by. Journalists and teammates always claimed that the Celtics fell apart without him; Boston was a 35-win team (-1.9 SRS) in 28 games he missed from 1958-69, and for the other 915 games of his career they played at a 59-win pace (6.4 SRS). This is a tiny piece of evidence – the years are spread out, teams change, and so on — but it echoes the same story as Russell’s other value signals."


3) Another passage from that link, talking playoffs defenses, that references a chart on team playoff results each year of each big man's career, where Russell has by-far the best score (7.7) where the next two best scores were the Spurs teammates (Robinson 5.1/Duncan 5.0) and Kareem was further down the list at 1.7, with Dirk at the bottom of the list at -1.5 (to give a sense of scale):

In the postseason, the defensive domination rolled on. Below, I’ve compared Russell’s playoff defenses to those of other all-time big men. The gray bubbles in Russell’s column are the Celtics individual performance in each year. Note that Boston never had a subpar defensive postseason with Russell, and that its worst playoff runs were clustered at the end of his career as a he slowed down


4) Most here are familiar with the concept of WOWY, and Ben's done a lot of stuff with regressed progressions of that concept (again, I recommend everyone checking out that work). But he also includes some work that he attributes to Zachary Stone, called Game-level adjusted Plus-Minus (GPM). Here's one of his brief writeups about GPM:
Spoiler:
Building off of the the same idea, Backpicks reader Zachary Stone has tackled historical games with a slightly different approach that I’ll call GPM (Game-level adjusted Plus-Minus). GPM is more analogous to “pure” RAPM in that each game result is a row in the equation, whereas WOWYR combines games and weights the lineups. The details of Zach’s version of GPM:

It uses only players who played at least 25 minutes per game during a season, so those games where Draymond Green is ejected early still count as a game played for him.
It uses a “replacement” player cutoff of 260 games. (The other studies below use 82 games.)
It’s run on data from 1957-2017.
(Technical detail: This version of GPM chose a lambda using the computationally expensive generalized cross-validation, not the chunkier k-fold method used for WOWYR in Part III.)
But there’s still the issue of time to consider. We don’t want the model thinking that Michael Jordan in his Wizard years is actually the Michael Jordan. So Zach ran the regression in 10-year slices, from 1957-66, 1958-67, 1959-68 and so on, and then grabbed each player’s best 10-year run. Finally, he scaled the results to allow for apples to apples comparisons across eras.

In theory, this will yield a better ballpark of those players with relatively consistent 10-year primes.


The standard, necessary disclaimer: this work is all done on much less granular data than the databall era, so where the stories don't mesh, I'd tend to give more weight to present-day analysis for present-day players than something like GPM or WOWYR. On the other hand, this approach gives us some quantified measures we can use that get the present and past players at least reasonably onto a common ground. Anyway, top GPM scores in history:

T1st: Russell, David Robinson & Steve Nash: 9.4
4th: Magic Johnson: 8.3
8th: Michael Jordan: 7.6

Now, to be fair and not just be cherrypicking, the same chart has his scaled WOWYR data, in which Jordan comes out higher:
1st: Magic Johnson (9.3)
4. Michael Jordan (8.3)
9. LeBron James (7.8)
15. Russell (5.9)

The point here was in no way to use these metrics to try to pinpoint exactly where Russell rates vs Jordan historically. There's not enough accuracy for that level of delineation, IMO. Instead, with the examples I've shown here, I wanted to point out that there is actually a pretty in-depth set of quantitative analysis out there that is growing in sophistication that shows in a variety of ways that a) Russell really was, measurably, the catalyst of all that Celtics success...that it's not just nostalgia and quotes (though there's that too), but that every quantified approach we take yields similar results. Also, b)Russell's individual impact, when compared quantifiably, even with attempts made at normalizing the playing field, shows up at the very elite end of NBA history.

Russell vs Jordan (I'll try to actually write a bit about Jordan into this section, :D )

So, this thread has long-sense been called while I was meandering through my Russell analysis. Plus, that got WAY TL;DR. So, I'll cut the Jordan portion of this way short. Instead, I'll talk about why I rank Russell higher. As I've pointed out before, I see both Russell and Jordan as clearly the big impact players of their generations, and the analytics aren't (IMO) granular enough to make a compelling case for one over the other. So, like I did with Jordan and LeBron, I'll shift this discussion to archetypes. Russell (the dominant defensive anchor big man and secondary offensive threat) and Jordan (the dominant scoring/creating wing w/ 2-way strong 1-on-1 D) define two of the most successful archetypes.

But, as I pointed out with Jordan vs LeBron, while players of his archetype can lead the league in offensive impact stats (like ORAPM) for a given year, it's typically a different offensive archetype (the team offense creator/floor general) that typically posts the historic impact numbers and/or leads the offensive dynasties. On defense, on the other hand, there's no question that Russell's archetype is the historic leader.

Also, as I've discussed in previous projects and likely will speak more of moving forward here, there is more to impact than just magnitude. Looking at a player's RAPM for a given year, for example, may let you know the size of their biggest impact at a given slice of time, which also corresponds to a given situation. However, concepts like Portability, Scaleability and Scarcity are also vital in determining just how valuable a given player is. The dominant scoring/do everything/strong 1-on-1 defensive wing is very valuable, but a) often isn't as portable because to be maximized the offensive system must (rightly) be built to suit their strengths, b) sometimes isn't as scaleable because the fit of offensive personnel to maximize them has to be more specialized (e.g. pairing such a player with other ball-needy scorers leads to diminishing returns, even if all players involved are excellent) and c) isn't as scarce of a resource, because there are almost always more wings capable of excellent offense than there are bigs capable of producing defensive impact on anywhere near the same scale.

So. With both Russell and Jordan, we saw that we could build all-forever-level dynasties in-era around their strengths. However, counter to the narrative that minimizes Russell because of how his impact may or may not translate through eras, I'd actually argue that Russell would fit MORE easily onto MORE championship-level teams than Jordan would have through history. Dominant big man defense is almost entirely portable...every team benefits from the addition of a player like that, even a team that already has a player like that (see the Robinson/Duncan Spurs). For that reason, dominant big man defense is almost entirely scaleable...that player's addition yield team results much closer to linear than adding a dominant offensive player to an already good offense. And again, historically that type of player whose defensive impact is anywhere near max offensive impact is just a much rarer commodity.

And I'll add something else...I think Russell's offensive "limitations", in some way, actually work to his favor in the portability/scaleability conversation compared to someone like Kareem or Wilt. Because he gives you all that additive value on defense, but doesn't have requirements that take anything off the table on offense. In other words, if you wanted to build a championship team around Russell, the fact that his offensive contributions (offensive rebounding, plus passing, decision-making, setting picks, etc) don't require much while only facilitating what the team wants to do can IMO have just as positive an effect as a shooter/floor-spacer type without any friction/diminishing returns/conflict with what your great offensive teammates need to make the offense truly great. This doesn't hold in every instance, of course...a team without a lot of offense-generating talent may need (and be able to succeed uniquely because of) the elite offense of an all-time great offensive big (e.g. Hakeem in the mid-90s, Shaq's Lakers, some Kareem teams, etc. But on a lot of teams, if the big man isn't going to be the offensive anchor (which usually is for the best) then I think Russell's skillset on that end is actually a plus. Counterintuitive, I'm sure I'll be called on this, and I'm sure I'll be saying more about it moving forward.

(Finally) Conclusion

Long story short (heh), I think Russell and Jordan both had maximized impacts in their careers. But I think Russell is a bit more valuable overall in more situations and is also a scarcer resource, and that ultimately was enough to get my vote.

Vote (I know it's too late, but for posterity. Also, like last time, limiting my vote to players that have gotten legitimate mention thus far):

1. Russell
2. Jordan
3. Duncan
Creator of the Hoops Lab: tinyurl.com/mpo2brj
Contributor to NylonCalculusDOTcom
Contributor to TYTSports: https://www.youtube.com/playlist?list=PLTbFEVCpx9shKEsZl7FcRHzpGO1dPoimk
Follow on Twitter: @ProfessorDrz
Owly
Analyst
Posts: 3,164
And1: 1,629
Joined: Mar 12, 2010

Re: RealGM 2020 Top 100 Project: #2 

Post#126 » by Owly » Sun Oct 18, 2020 6:58 pm

drza wrote:2) A quote from that ElGee writeup I just linked:
"He didn’t miss much time in his career, so WOWY numbers are hard to come by. Journalists and teammates always claimed that the Celtics fell apart without him; Boston was a 35-win team (-1.9 SRS) in 28 games he missed from 1958-69, and for the other 915 games of his career they played at a 59-win pace (6.4 SRS). This is a tiny piece of evidence – the years are spread out, teams change, and so on — but it echoes the same story as Russell’s other value signals."

On this one, because I tend to see it used, what would be the justification for excluding '57? It is the largest "without" sample (indeed near matching the combined others), otoh, it is perhaps the only one in which the team is somewhat prepared to play without him (and there is an argument that he is an extreme player, like a Rodman, around whom you consider your pieces so as to optimize the whole). It is, for what it's worth, also without Ramsey, whom I would posit is an upgrade on the wings they played with during that spell.

Perhaps I am missing something. When used it should be noted that a disproportionate amount of the sample comes from his rookie year. Or it could be listed separately. But I've never, to my recollection, seen the logic given for excluding it.

Return to Player Comparisons