A Crisis of Criteria leading up to the 2023 Top 100

[Doctor MJ]

Hey y'all,

So I wanted to share with people where my head is at pertaining to Top 100 criteria, and hopefully get into fruitful conversation on the topic.

I'll first note that Tim made a thread in this vein already that was great, but this is something I'm just saying is a crisis for me, so I wanted to just start fresh speaking for myself.

Here's the essence of the "crisis" as I see it:

1. After decades and decades of relatively stable strategic equilibrium in the NBA, now we're in the midst of a massive shift which can be summed up fairly well by the change in ORtg.

When we look at bk-ref's year's ORtg's, we see a league that hovered within a 6 point range (102.5 to 108.5) from '78-79 to '15-16, and in the past few years we've now climbed to the point where we're approaching another 6 points beyond that.

2. If this were something caused by a 2016 rule change that simply made scoring easier, it would straight forward to at least diagnose the dilemma in an era-separated way, but the reality is that the critical rule change happened in 1979, and it just took decades for teams to embrace an adaptation which allowed that change to transform the league.

3. Adding to the analytic challenge here is that the fact that the new paradigm shift doesn't effect all players in the same way. Shouldn't be unexpected that this is so - different players have different talents and skills - but it leads to this fundamental issue:

Player A had a better career than Player B based on standards for achievement, but Player B likely would have had the superior career if coaches at the time had properly internalized the paradigm shifts we now see as clear cut, even without any further rule changes.

4. When we go to do all-time player comparisons, we can acknowledge various criteria that will lead to different ranked lists...but when doing something like the Top 100, we are choosing to put forward one particular criteria above all others, and this raises the question of which criteria we should use.

5. The most straight forward approach, I'd argue, is one that looks to completely ignore how "good" any given league is, and focus only on dominance within one's own era. But while a particular user is free to use this approach, that's not what the norm has ever been since I first participated in 2006.

6. Rather, there's always some form of era-adjustment going on which leads to - among other things - the ability for the group to leave George Mikan out of the top tier of candidates.

7. While this has always seemed pretty reasonable to me, the question of how individual posters are doing these sort of collaborations hasn't been front and center. Sometimes we describe the process as simply looking to somehow estimate how good the competition of any season's league was, and apply that as a degree of difficulty bonus/penalty relative to a guy in another era.

8. What I've always tried to do though is to try to do an analysis that actually looks at the basketball strengths and weaknesses of each player relative to each other, and imagine what they'd do in the other player's era. There are a lot of little issues to think through when you do this, and you're never going to do it perfectly, but at the very least, you're thinking about basketball details when you're making you choice between basketball players.

9. But when we experience a strategic paradigm shift that makes the player who was worse into the player who would have been better, now we end up with a problematic loop:

Historical A had a better career than Historical B
Modern C is better than Historical A at the modern game.
Historical B is better than Modern C at the modern game.

B > C > A > B

How do you decide who should be ranked first?

10. I think most have concluded already that there's no way to do it, and so you have to focus on a broader degree-of-difficulty mixed with in-era dominance, and I'm thinking I'm going to have to go in this direction myself despite my misgivings, and if I do, it will have some pretty profound changes on my voting compared to what I've done in the past.

11. Case in point, I've had Garnett ahead of Duncan on my Top 100 list since the 2014 edition, but I think it likely I'll go back to having Duncan ahead of Garnett when we do this in the 2023 edition.

Would I draft Duncan over Garnett? No.
Am I more impressed with Duncan than Garnett? No.
Do I think Duncan was better at basketball than Garnett? No.

But...

Does Duncan rank higher than Garnett based on how I evaluate season by season achievement accumulated over the course of a career? Yes.

And to put this in something a bit more concrete, I'm now going to share some historical spreadsheets I've made.

Before I do, note that I'd love to talk more about the details in these spreadsheets, but I'm not interested in being on trial for "bias". I do the best I can as I go through these analyses, but make no claim about being definitively right. Further, while it would be great if I either a) had a perfect memory or b) took really detail notes, but I don't. So I can try to explain why I put things as I did, but I won't necessarily have a ready and detailed answer for any particular spreadsheet field.

First spreadsheet: A POY-shares style document based on my own opinions for POY, OPOY, DPOY & COY going all the way back to '43-44. (ftr, I've grayed out '43-44 & '44-45 because in retrospect I've concluded that '45-46 - the first season after World War II - was really the better line of demarcation.)

Doc's Ballon d'Orange

Second spreadsheet: With the idea of Player of the Decade, a common issue is that people insist on focusing on stuff with the same second-to-last digit ('80s, '90s, etc), when I think the more meaningful thing is to do a running POD analysis. Anyone who stands out as the best over any 10-year run is a POD in my book.

So in this spreadsheet you'll see PODs (as well as OPODs, DPODs & CODs) that go from '45-46 to '54-55 all the way to '12-13 to '21-22 for men's basketball, as well as sheets for women's basketball going back to the founding of the WNBA.

I know not everyone cares about women's basketball, but from my perspective it represents a nice parallel analysis for comparison.

Doc's Ballon d'Orange PODs

Along those same lines, the last spreadsheet I've done something similar for tennis. Note that my tennis analysis here is going solely by performance in major tournaments, and is intended to be entirely separate from my own personal opinions about specific tennis players. While I actually have more competitive background in tennis than basketball, there are reasons why historical analysis in tennis is more problematic than it is for basketball.

I'll note that I do separate analysis out here by playing surface.

I should also note that in the pre-Open era, I ignore the Grand Slams in men's tennis and focus on the pro majors, because that's where the best players were. By contrast in the women's game, the top players typically stayed as amateurs until the Open era.

Tennis GS Pods

Alright, I'll leave it at that and hope this leads to good discussion.

[eminence]

Very long, very well thought out stuff, and all I can really say is - yeah, I basically agree, lol.

It's a position I've been gradually shifting towards over the last few years, but I'm heavily heavily weighting own-era dominance at this point (my own opinion of dominance, which does sometimes disagree with the consensus). If I think things are truly truly even I'll go with the guy I feel was from a 'stronger' era, but otherwise I try to keep era shift type stuff out of my arguments. For some past players it's very difficult (impossible) to view them from the right frame that isn't clouded by modern basketball, but I try my best.

I look forward to reading through the Ballon d'Orange sheets sometime later

[MyUniBroDavis]

Hey y'all,

So I wanted to share with people where my head is at pertaining to Top 100 criteria, and hopefully get into fruitful conversation on the topic.

I'll first note that Tim made a thread in this vein already that was great, but this is something I'm just saying is a crisis for me, so I wanted to just start fresh speaking for myself.

Here's the essence of the "crisis" as I see it:

1. After decades and decades of relatively stable strategic equilibrium in the NBA, now we're in the midst of a massive shift which can be summed up fairly well by the change in ORtg.

When we look at bk-ref's year's ORtg's, we see a league that hovered within a 6 point range (102.5 to 108.5) from '78-79 to '15-16, and in the past few years we've now climbed to the point where we're approaching another 6 points beyond that.

2. If this were something caused by a 2016 rule change that simply made scoring easier, it would straight forward to at least diagnose the dilemma in an era-separated way, but the reality is that the critical rule change happened in 1979, and it just took decades for teams to embrace an adaptation which allowed that change to transform the league.

3. Adding to the analytic challenge here is that the fact that the new paradigm shift doesn't effect all players in the same way. Shouldn't be unexpected that this is so - different players have different talents and skills - but it leads to this fundamental issue:

Player A had a better career than Player B based on standards for achievement, but Player B likely would have had the superior career if coaches at the time had properly internalized the paradigm shifts we now see as clear cut, even without any further rule changes.

4. When we go to do all-time player comparisons, we can acknowledge various criteria that will lead to different ranked lists...but when doing something like the Top 100, we are choosing to put forward one particular criteria above all others, and this raises the question of which criteria we should use.

5. The most straight forward approach, I'd argue, is one that looks to completely ignore how "good" any given league is, and focus only on dominance within one's own era. But while a particular user is free to use this approach, that's not what the norm has ever been since I first participated in 2006.

6. Rather, there's always some form of era-adjustment going on which leads to - among other things - the ability for the group to leave George Mikan out of the top tier of candidates.

7. While this has always seemed pretty reasonable to me, the question of how individual posters are doing these sort of collaborations hasn't been front and center. Sometimes we describe the process as simply looking to somehow estimate how good the competition of any season's league was, and apply that as a degree of difficulty bonus/penalty relative to a guy in another era.

8. What I've always tried to do though is to try to do an analysis that actually looks at the basketball strengths and weaknesses of each player relative to each other, and imagine what they'd do in the other player's era. There are a lot of little issues to think through when you do this, and you're never going to do it perfectly, but at the very least, you're thinking about basketball details when you're making you choice between basketball players.

9. But when we experience a strategic paradigm shift that makes the player who was worse into the player who would have been better, now we end up with a problematic loop:

Historical A had a better career than Historical B
Modern C is better than Historical A at the modern game.
Historical B is better than Modern C at the modern game.

B > C > A > B

How do you decide who should be ranked first?

10. I think most have concluded already that there's no way to do it, and so you have to focus on a broader degree-of-difficulty mixed with in-era dominance, and I'm thinking I'm going to have to go in this direction myself despite my misgivings, and if I do, it will have some pretty profound changes on my voting compared to what I've done in the past.

11. Case in point, I've had Garnett ahead of Duncan on my Top 100 list since the 2014 edition, but I think it likely I'll go back to having Duncan ahead of Garnett when we do this in the 2023 edition.

Would I draft Duncan over Garnett? No.
Am I more impressed with Duncan than Garnett? No.
Do I think Duncan was better at basketball than Garnett? No.

But...

Does Duncan rank higher than Garnett based on how I evaluate season by season achievement accumulated over the course of a career? Yes.

And to put this in something a bit more concrete, I'm now going to share some historical spreadsheets I've made.

Before I do, note that I'd love to talk more about the details in these spreadsheets, but I'm not interested in being on trial for "bias". I do the best I can as I go through these analyses, but make no claim about being definitively right. Further, while it would be great if I either a) had a perfect memory or b) took really detail notes, but I don't. So I can try to explain why I put things as I did, but I won't necessarily have a ready and detailed answer for any particular spreadsheet field.

First spreadsheet: A POY-shares style document based on my own opinions for POY, OPOY, DPOY & COY going all the way back to '43-44. (ftr, I've grayed out '43-44 & '44-45 because in retrospect I've concluded that '45-46 - the first season after World War II - was really the better line of demarcation.)

Doc's Ballon d'Orange

Second spreadsheet: With the idea of Player of the Decade, a common issue is that people insist on focusing on stuff with the same second-to-last digit ('80s, '90s, etc), when I think the more meaningful thing is to do a running POD analysis. Anyone who stands out as the best over any 10-year run is a POD in my book.

So in this spreadsheet you'll see PODs (as well as OPODs, DPODs & CODs) that go from '45-46 to '54-55 all the way to '12-13 to '21-22 for men's basketball, as well as sheets for women's basketball going back to the founding of the WNBA.

I know not everyone cares about women's basketball, but from my perspective it represents a nice parallel analysis for comparison.

Doc's Ballon d'Orange PODs

Along those same lines, the last spreadsheet I've done something similar for tennis. Note that my tennis analysis here is going solely by performance in major tournaments, and is intended to be entirely separate from my own personal opinions about specific tennis players. While I actually have more competitive background in tennis than basketball, there are reasons why historical analysis in tennis is more problematic than it is for basketball.

I'll note that I do separate analysis out here by playing surface.

I should also note that in the pre-Open era, I ignore the Grand Slams in men's tennis and focus on the pro majors, because that's where the best players were. By contrast in the women's game, the top players typically stayed as amateurs until the Open era.

Tennis GS Pods

Alright, I'll leave it at that and hope this leads to good discussion.

I think the paradigm shift would help a lot of guys a lot more uniformally that 3pt shooters good 3pt shooters bad, to me it more so would mean that some guys that were underutilized in some regards would be better nowadays if that makes sense

I do think some guys, particularly perimeter players, get helped more, but even today we see guys that aren’t optimized have far worse impact than otherwise imo

I feel most people do a mixture of how good they find someone in an absolute sense + era dominance/achievements, and different criteria is more so the blend of how much they weigh one vs the other

[Heej]

Interesting points being raised here. I tend to sit out the projects as I prefer lurking and letting discussions develop but this piqued my interest. I'd like to put some "names to faces" in a sense

Let's pretend Player A is a juiced up Moses Malone, Player B is Jerry West, and player C is Dirk. My personal take is that I don't think it changes all that much when trying to bring Super Saiyan Moses into the modern era. Even though rule changes would benefit Jerry West's individual game more, SSJ Moses would similarly benefit from better shooting and more skilled teammates 1-4. And IQ is simply immutable across eras. A lot of the reasons these guys were so great was just purely mental.

Even if West catches a 30% boost to his effectiveness in the modern era, I would expect SSJ Moses to benefit like 15-20% just from having better spacing and role players that can do more. I don't think that really changes rankings much as I wouldn't take Dirk over either and nowadays with player empowerment SSJ Moses likely forces himself to a good enough supporting cast somewhere that Jerry West doesn't necessarily overtake him either way.

[eminence]

Interesting points being raised here. I tend to sit out the projects as I prefer lurking and letting discussions develop but this piqued my interest. I'd like to put some "names to faces" in a sense

Let's pretend Player A is a juiced up Moses Malone, Player B is Jerry West, and player C is Dirk. My personal take is that I don't think it changes all that much when trying to bring Super Saiyan Moses into the modern era. Even though rule changes would benefit Jerry West's individual game more, SSJ Moses would similarly benefit from better shooting and more skilled teammates 1-4. And IQ is simply immutable across eras. A lot of the reasons these guys were so great was just purely mental.

Even if West catches a 30% boost to his effectiveness in the modern era, I would expect SSJ Moses to benefit like 15-20% just from having better spacing and role players that can do more. I don't think that really changes rankings much as I wouldn't take Dirk over either and nowadays with player empowerment SSJ Moses likely forces himself to a good enough supporting cast somewhere that Jerry West doesn't necessarily overtake him either way.

SSJ Moses sounds like Wilt to me.

[Heej]

Interesting points being raised here. I tend to sit out the projects as I prefer lurking and letting discussions develop but this piqued my interest. I'd like to put some "names to faces" in a sense

Let's pretend Player A is a juiced up Moses Malone, Player B is Jerry West, and player C is Dirk. My personal take is that I don't think it changes all that much when trying to bring Super Saiyan Moses into the modern era. Even though rule changes would benefit Jerry West's individual game more, SSJ Moses would similarly benefit from better shooting and more skilled teammates 1-4. And IQ is simply immutable across eras. A lot of the reasons these guys were so great was just purely mental.

Even if West catches a 30% boost to his effectiveness in the modern era, I would expect SSJ Moses to benefit like 15-20% just from having better spacing and role players that can do more. I don't think that really changes rankings much as I wouldn't take Dirk over either and nowadays with player empowerment SSJ Moses likely forces himself to a good enough supporting cast somewhere that Jerry West doesn't necessarily overtake him either way.

SSJ Moses sounds like Wilt to me.

Lool yea but that wouldn't even be fair then. I was thinking make him juuust good enough to be better than Jerry West for comparison's sake. Either way though I'm not personally sure how much of a difference it makes because everyone's gonna catch a boost in some ways. Like my initial thought was post players take a hit nowadays given how much they get fouled and swarmed down low, but then I remembered how much they used to hold post players back offensively back in the day as far offensive fouls and bumping yourself open to create space.

That's why you can't even say "oh Player B shot better so he's instantly better in the modern era" without accounting for the fact that some guys like LeBron would've had to completely change their physicality to adapt to the ruleset back then and some of those paint beasts with poor outside shooting that you think take a hit being transported nowadays are suddenly much more free to use their physicality albeit against much stronger and athletic modern defenders.

Trying to over-fit or adjust for differences like that is a tall task unless you truly account for every factor. And if that's the route you wanna take then people should honestly chime in on all the pros and cons in terms of rules from each era so that you at least have some kind of consistency as far as how different archetypes scale.

My opinion though is that IQ and mentality trump all at the end of the day and you'd likely find that the various archetypes don't actually scale all that differently from different eras to now when you add in how much more physical offensively players are allowed to be nowadays vs during the 50s-70s

Makes me think guys like Wilt are a bit underrated if anything but most people assume he would scale worse nowadays

[OhayoKD]

Historical A had a better career than Historical B
Modern C is better than Historical A at the modern game.
Historical B is better than Modern C at the modern game.

B > C > A > B

How do you decide who should be ranked first?

Tbh, I think there's a potential option C you might want to consider:

what if we averaged everyone over all of nba history? as in, you get time machined to every season and have your value averaged over all years. impossible to do, but would be my preferred method. russell's chart of relative value vs year is probably a slowly declining relative value line from his time to the present. most guys would probably have peak relative value in their own time (thus, why we think of them as being so good) and might be a little bumpy as the game changes over time. i wonder whose relative value line is the flattest? lebron?

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

yes, if he stepped out of a time machine, he would be like an alien and unstoppable. until rick mahorn got tired of it and forearm shivered him. but several weeks later when steph recovered, he would go back to dominating.

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

[Heej]

Historical A had a better career than Historical B
Modern C is better than Historical A at the modern game.
Historical B is better than Modern C at the modern game.

B > C > A > B

How do you decide who should be ranked first?

Tbh, I think there's a potential option C you might want to consider:

what if we averaged everyone over all of nba history? as in, you get time machined to every season and have your value averaged over all years. impossible to do, but would be my preferred method. russell's chart of relative value vs year is probably a slowly declining relative value line from his time to the present. most guys would probably have peak relative value in their own time (thus, why we think of them as being so good) and might be a little bumpy as the game changes over time. i wonder whose relative value line is the flattest? lebron?

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

yes, if he stepped out of a time machine, he would be like an alien and unstoppable. until rick mahorn got tired of it and forearm shivered him. but several weeks later when steph recovered, he would go back to dominating.

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

That's a pretty neat idea. You'd have to really kinda outline the differences in each era tho in order to fairly assess that. That being said I'd imagine Steph's relative era graph starts off slow to where he's probably a Pistol Pete type character early on, and then eventually turns into a nuke. I'd imagine he'd be pretty ridiculous in the run and gun 80s and the 90s with the soft off-ball defense they played back then and the stiffs that would be in hell trying to defend Steph in space 30 feet from the hoop.

I could see his off-ball game shining more in that era because nowadays people foul so much off the ball (and in the post!!). His ultimate peak I'd still have to imagine should coincide with today though just because he has more talented players and coaches to work with nowadays but I doubt it sees any drop off really until the 70s where dribbling rules were tighter and the 3 point line wasn't really a thing.

[f4p]

Makes me think guys like Wilt are a bit underrated if anything but most people assume he would scale worse nowadays

my only problem with era translation is when people only translate in one direction. to 2023. this isn't optimal/superior basketball. there's no such thing. it's just the way we play today. in 20 years, things may change that make us wonder why the play style ever got the way it did in 2023, much as we watch a 1980/90's game and can't understand why they don't understand that the white line on the court 23'-9" from the basket gives you one extra point if you shoot from behind it. you can't just slide some big man from the 1980's to today and wonder how they would play, but then not ask how a perimeter guy from today would look if they were taking 1.5 3's per game and having to build their entire game on mid-range jumpers and having to constantly feed the post, even to inferior players, because that's just how it was. if there was a way to essentially take an average of how you feel they would translate to every era, then i could see how that would be fair. but that's also pretty difficult.

but then i'm not high on translation anyway. the time machine argument i think is ridiculous on the face of it, but even trying to scale how you think they would be with training and time to adapt is difficult. the players played when they played. they solved the problems of their day. they didn't develop skills from ages 12-13 on to try to make their game as palatable as possible to a 2023 audience or the game of the 2023 nba. they tried to win championships in 1973 and 1987 and 1999. based on the style of play that their teammates and coaches would be playing. i understand trying to look at the overall skillset of a player and think what they would do in a different setting, but now you're introducing a hypothetical into your analysis, and weighing it however you want to with nothing to confirm or reject your hypothesis. maybe the player translates just how you think. maybe they spend ages 12-20 developing all sorts of skills they never did because necessity is the mother of invention and they were dominating without developing those skills.

[penbeast0]

I will say that while I do adjust for era strength, it's more based on Expansion/Player base rather than subjective ideas of how basketball was played in that era. Mainly. But I do think there are some other things that weaken basketball team like the increased use of cocaine and players jumping leagues in the 70s that led to lower team cohesiveness and the like. Those are relatively minor however.

I am not lowering a player like Kareem in my ratings because in today's league, post play has been devalued. I'm not lowering a player like MJ because in the 60s, defenses would have been cutting his feet out from under him on dunks leaving him injured like Baylor/Hawkins/Gus Johnson/etc. I'm not lowering Steph Curry because he wouldn't be as effective in an era where 3 point shooting isn't allowed or coaches discourage it. And I'm certainly not lowering minority players because they wouldn't have been allowed to play and star in the 40s and 50s. I do notice that most people (not the best posters like Doc) only look across eras at how stars of other eras would function today and not how today's stars would function in an era with no weight training, PEDs, canvas shoes, uneven floors with no give, etc.

I have also disagreed with Doc on the idea that the ABA or early 3 point NBA era would have benefitted that much from the modern offenses built on spamming 3's. NO ONE in those eras was making 3's at modern rates of effectiveness; kids didn't grow up shooting 3's constantly, and without that focus on the shot, while you had 3 point specialists like Les Selvage, he was shooting .316 from there (which makes him 14th most efficient qualifying 3 point shooter in ABA history according to BBR-com). At that rate, the shot was not as efficient as slashing or post play though I agree it was probably underused by most ABA teams.

So, I am one of those posters who look at in-era dominance and to the extent I use portability, it's only in era portability. My era adjustment is not based on how players would do in other eras with other rule sets and other coaching mindsets because I think it's just too removed from reality to be reasonably quantified or even estimated.

I apologize if that's not what you meant Doc, don't mean to be tilting at straw men here. If it's not please correct me.

[f4p]

Historical A had a better career than Historical B
Modern C is better than Historical A at the modern game.
Historical B is better than Modern C at the modern game.

B > C > A > B

How do you decide who should be ranked first?

Tbh, I think there's a potential option C you might want to consider:

what if we averaged everyone over all of nba history? as in, you get time machined to every season and have your value averaged over all years. impossible to do, but would be my preferred method. russell's chart of relative value vs year is probably a slowly declining relative value line from his time to the present. most guys would probably have peak relative value in their own time (thus, why we think of them as being so good) and might be a little bumpy as the game changes over time. i wonder whose relative value line is the flattest? lebron?

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

yes, if he stepped out of a time machine, he would be like an alien and unstoppable. until rick mahorn got tired of it and forearm shivered him. but several weeks later when steph recovered, he would go back to dominating.

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

That's a pretty neat idea. You'd have to really kinda outline the differences in each era tho in order to fairly assess that. That being said I'd imagine Steph's relative era graph starts off slow to where he's probably a Pistol Pete type character early on, and then eventually turns into a nuke. I'd imagine he'd be pretty ridiculous in the run and gun 80s and the 90s with the soft off-ball defense they played back then and the stiffs that would be in hell trying to defend Steph in space 30 feet from the hoop.

I could see his off-ball game shining more in that era because nowadays people foul so much off the ball (and in the post!!). His ultimate peak I'd still have to imagine should coincide with today though just because he has more talented players and coaches to work with nowadays but I doubt it sees any drop off really until the 70s where dribbling rules were tighter and the 3 point line wasn't really a thing.

but that's basically a time machine argument. no one shot 30 feet from the basket back then. they barely shot from the 3 point line, and usually only when wide open. steph would have grown up then playing the way most people do. look at larry bird. he had a 3PAR+ of 321 in 1986. that's 50% higher than steph's peak of 195. meaning he deviated from the league 50% more than steph ever did, with a whole half decade of dominance behind bird to give him any freedom he wanted. and he took 2.4 3PA/gm. so maybe steph even surpasses bird and gets up 3 or 3.5 per game. but that's much less era relative dominance compared to today, where his 3 point shooting is by far his biggest step up on anyone in a comparsion. i think steph with his outlier 3 point shooting and russell with his outlier interior defense are probably 2 of the most steeply monotonically increasing (steph)/decreasing (russell) era-relative dominance guys you can find starting from the beginning of the nba to today if we were doing my "average across all of history" idea. that's not to say they wouldn't be good, just less good relative to their era.

[f4p]

to get more to the point of this thread, my biggest concern with the top 100 is that everyone uses such different criteria to begin with. granted, some people would say that's a feature, not a bug, but some people are longevity junkies and some people aren't. while this tends to show the biggest difference with the modern guys, where we have to wait until like the 2029 Top 100 to finally say Jokic is really good, it still makes even the GOAT part of the discussion really trying. or some people are portability junkies and some people just go by what happened. you basically get people talking past each other because they aren't even evaluating the same way.

i realize some level of "total career value" is going to win the day because staying good longer just seems obviously better, but i'm not sure that should always be the question, as more outside things can affect longevity versus how many can affect your peak/prime level of play. and it always feels a little weird with the modern guys, even if you can technically say "well, they haven't actually done it yet so why consider it?" feels better to just throw in some conservative assumptions (i.e. level of play, not assuming they will win 3 titles or anything) and knock the guy in future lists if it doesn't play out/bump them up if they exceed what we conservatively expected.

[Heej]

Tbh, I think there's a potential option C you might want to consider:

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

That's a pretty neat idea. You'd have to really kinda outline the differences in each era tho in order to fairly assess that. That being said I'd imagine Steph's relative era graph starts off slow to where he's probably a Pistol Pete type character early on, and then eventually turns into a nuke. I'd imagine he'd be pretty ridiculous in the run and gun 80s and the 90s with the soft off-ball defense they played back then and the stiffs that would be in hell trying to defend Steph in space 30 feet from the hoop.

I could see his off-ball game shining more in that era because nowadays people foul so much off the ball (and in the post!!). His ultimate peak I'd still have to imagine should coincide with today though just because he has more talented players and coaches to work with nowadays but I doubt it sees any drop off really until the 70s where dribbling rules were tighter and the 3 point line wasn't really a thing.

but that's basically a time machine argument. no one shot 30 feet from the basket back then. they barely shot from the 3 point line, and usually only when wide open. steph would have grown up then playing the way most people do. look at larry bird. he had a 3PAR+ of 321 in 1986. that's 50% higher than steph's peak of 195. meaning he deviated from the league 50% more than steph ever did, with a whole half decade of dominance behind bird to give him any freedom he wanted. and he took 2.4 3PA/gm. so maybe steph even surpasses bird and gets up 3 or 3.5 per game. but that's much less era relative dominance compared to today, where his 3 point shooting is by far his biggest step up on anyone in a comparsion. i think steph with his outlier 3 point shooting and russell with his outlier interior defense are probably 2 of the most steeply monotonically increasing (steph)/decreasing (russell) era-relative dominance guys you can find starting from the beginning of the nba to today if we were doing my "average across all of history" idea. that's not to say they wouldn't be good, just less good relative to their era.

Eh I mean you have guys like Pistol Pete who was allowed by his pops to Chuck up shots from like 25+ feet during his LSU days. If a guy could shoot the long bomb like that no one was going to have an aneurysm to stop it if he's just proving himself in practice every single day. But I get your point tho. There's a lotta weird nuances then if you were to just try and do thought experiments on "what if such and such player was born 30 years earlier or later" lmao. Interesting point tho that those two probably see steeper drop offs but I mean looking at what Reggie Miller was able to do I doubt Steph's drop off is all that drastic either.

[capfan33]

Tbh, I think there's a potential option C you might want to consider:

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

That's a pretty neat idea. You'd have to really kinda outline the differences in each era tho in order to fairly assess that. That being said I'd imagine Steph's relative era graph starts off slow to where he's probably a Pistol Pete type character early on, and then eventually turns into a nuke. I'd imagine he'd be pretty ridiculous in the run and gun 80s and the 90s with the soft off-ball defense they played back then and the stiffs that would be in hell trying to defend Steph in space 30 feet from the hoop.

I could see his off-ball game shining more in that era because nowadays people foul so much off the ball (and in the post!!). His ultimate peak I'd still have to imagine should coincide with today though just because he has more talented players and coaches to work with nowadays but I doubt it sees any drop off really until the 70s where dribbling rules were tighter and the 3 point line wasn't really a thing.

but that's basically a time machine argument. no one shot 30 feet from the basket back then. they barely shot from the 3 point line, and usually only when wide open. steph would have grown up then playing the way most people do. look at larry bird. he had a 3PAR+ of 321 in 1986. that's 50% higher than steph's peak of 195. meaning he deviated from the league 50% more than steph ever did, with a whole half decade of dominance behind bird to give him any freedom he wanted. and he took 2.4 3PA/gm. so maybe steph even surpasses bird and gets up 3 or 3.5 per game. but that's much less era relative dominance compared to today, where his 3 point shooting is by far his biggest step up on anyone in a comparsion. i think steph with his outlier 3 point shooting and russell with his outlier interior defense are probably 2 of the most steeply monotonically increasing (steph)/decreasing (russell) era-relative dominance guys you can find starting from the beginning of the nba to today if we were doing my "average across all of history" idea. that's not to say they wouldn't be good, just less good relative to their era.

Just wanted to chime in that I've toyed with this general idea in the past of just averaging out impact across each decade for players and I do think at a minimum it's a useful thought exercise both in evaluating players as well as the differences between eras.

[Cavsfansince84]

Eh I mean you have guys like Pistol Pete who was allowed by his pops to Chuck up shots from like 25+ feet during his LSU days. If a guy could shoot the long bomb like that no one was going to have an aneurysm to stop it if he's just proving himself in practice every single day. But I get your point tho. There's a lotta weird nuances then if you were to just try and do thought experiments on "what if such and such player was born 30 years earlier or later" lmao. Interesting point tho that those two probably see steeper drop offs but I mean looking at what Reggie Miller was able to do I doubt Steph's drop off is all that drastic either.

Well I think its been a slow transition from guys taking 3's only when open(which was most of the 80's) to guys starting to take 3's off the dribble with a guy near them(more in the 90's) to guys like Vince who started launching them from 3+ ft behind the line whenever they felt like it(2000's). So I think the game needed to go through all those steps to get to where we are now where pretty much every high scoring wing shoots 5-6 a game and some will shoot 10+. Players had to get used to it and so did coaches. I don't think you could drop Steph off in the 80's and think he'd just naturally expect to be allowed to shoot 10 3's per game. It wouldn't enter into his mind to do that and as a pg it was way rarer back then to see one average over about 22ppg. Most were still under 20ppg back then and were expected to always try and get others involved first.

[OhayoKD]

Tbh, I think there's a potential option C you might want to consider:

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

That's a pretty neat idea. You'd have to really kinda outline the differences in each era tho in order to fairly assess that. That being said I'd imagine Steph's relative era graph starts off slow to where he's probably a Pistol Pete type character early on, and then eventually turns into a nuke. I'd imagine he'd be pretty ridiculous in the run and gun 80s and the 90s with the soft off-ball defense they played back then and the stiffs that would be in hell trying to defend Steph in space 30 feet from the hoop.

I could see his off-ball game shining more in that era because nowadays people foul so much off the ball (and in the post!!). His ultimate peak I'd still have to imagine should coincide with today though just because he has more talented players and coaches to work with nowadays but I doubt it sees any drop off really until the 70s where dribbling rules were tighter and the 3 point line wasn't really a thing.

but that's basically a time machine argument.

And?

steph would have grown up then playing the way most people do.

Well I'm interested in evaluating Steph, not fan-fiction not-steph, so...

i think steph with his outlier 3 point shooting and russell with his outlier interior defense are probably 2 of the most steeply monotonically increasing (steph)/decreasing (russell) era-relative dominance guys you can find starting from the beginning of the nba to today if we were doing my "average across all of history" idea.

Steph pre-three point line is interesting though Unibro's made his case he'd still be broken. Post-three point line, Curry should probably be league breaking and then gradually gets league breaking as basketball progresses.

That's a pretty neat idea. You'd have to really kinda outline the differences in each era tho in order to fairly assess that. That being said I'd imagine Steph's relative era graph starts off slow to where he's probably a Pistol Pete type character early on, and then eventually turns into a nuke. I'd imagine he'd be pretty ridiculous in the run and gun 80s and the 90s with the soft off-ball defense they played back then and the stiffs that would be in hell trying to defend Steph in space 30 feet from the hoop.

I could see his off-ball game shining more in that era because nowadays people foul so much off the ball (and in the post!!). His ultimate peak I'd still have to imagine should coincide with today though just because he has more talented players and coaches to work with nowadays but I doubt it sees any drop off really until the 70s where dribbling rules were tighter and the 3 point line wasn't really a thing.

but that's basically a time machine argument. no one shot 30 feet from the basket back then. they barely shot from the 3 point line, and usually only when wide open. steph would have grown up then playing the way most people do. look at larry bird. he had a 3PAR+ of 321 in 1986. that's 50% higher than steph's peak of 195. meaning he deviated from the league 50% more than steph ever did, with a whole half decade of dominance behind bird to give him any freedom he wanted. and he took 2.4 3PA/gm. so maybe steph even surpasses bird and gets up 3 or 3.5 per game. but that's much less era relative dominance compared to today, where his 3 point shooting is by far his biggest step up on anyone in a comparsion. i think steph with his outlier 3 point shooting and russell with his outlier interior defense are probably 2 of the most steeply monotonically increasing (steph)/decreasing (russell) era-relative dominance guys you can find starting from the beginning of the nba to today if we were doing my "average across all of history" idea. that's not to say they wouldn't be good, just less good relative to their era.

Eh I mean you have guys like Pistol Pete who was allowed by his pops to Chuck up shots from like 25+ feet during his LSU days. If a guy could shoot the long bomb like that no one was going to have an aneurysm to stop it if he's just proving himself in practice every single day. But I get your point tho. There's a lotta weird nuances then if you were to just try and do thought experiments on "whone wgat if such and such player was born 30 years earlier or later" lmao. Interesting point tho that those two probably see steeper drop offs but I mean looking at what Reggie Miller was able to do I doubt Steph's drop off is all that drastic either.

Gotta do the thought experiment to make things "fair" lol. Only valid means of player comparison are the ones that skew towards older ones(era-relativity cough)

[trex_8063]

Cool post, some stuff I'll have to think about, probably come back to.

But I was leafing through the first spreadsheet and---not to "put you on trial" or derail, but...---am I interpreting things correctly that you think LeBron was only the 3rd-best player on the '11 Heat and that Dwight Howard wasn't even a top 5 in the league?

[Doctor MJ]

I think the paradigm shift would help a lot of guys a lot more uniformally that 3pt shooters good 3pt shooters bad, to me it more so would mean that some guys that were underutilized in some regards would be better nowadays if that makes sense

I do think some guys, particularly perimeter players, get helped more, but even today we see guys that aren’t optimized have far worse impact than otherwise imo

I feel most people do a mixture of how good they find someone in an absolute sense + era dominance/achievements, and different criteria is more so the blend of how much they weigh one vs the other

First thing I'd say:

To the extent that everyone is helped the same amount, I'm considering that effectively normalized away. Hence, while everyone benefits from, say, the added spacing of 3-point shooting, I think it's meaningful to be considering who is helped more, and as a short hand, I'll just say those are the ones being helped.

I will say that I think that great outside shooters are generally going to gain more from the recognition that 3-point shooting should be the foundation of all basketball offenses in a mature basketball space than the average poor outside shooter, but I'll acknowledge that other types of players are going to get helped as well, and that the realization that humans in general can be great at shooting 3's rather than the assumption that a few chosen ones can do it could lead at some point to reduction in the value of great shooting.

Re: the previously underutilized would be better if that makes sense. Absolutely.

[Doctor MJ]

Interesting points being raised here. I tend to sit out the projects as I prefer lurking and letting discussions develop but this piqued my interest. I'd like to put some "names to faces" in a sense

Let's pretend Player A is a juiced up Moses Malone, Player B is Jerry West, and player C is Dirk. My personal take is that I don't think it changes all that much when trying to bring Super Saiyan Moses into the modern era. Even though rule changes would benefit Jerry West's individual game more, SSJ Moses would similarly benefit from better shooting and more skilled teammates 1-4. And IQ is simply immutable across eras. A lot of the reasons these guys were so great was just purely mental.

Even if West catches a 30% boost to his effectiveness in the modern era, I would expect SSJ Moses to benefit like 15-20% just from having better spacing and role players that can do more. I don't think that really changes rankings much as I wouldn't take Dirk over either and nowadays with player empowerment SSJ Moses likely forces himself to a good enough supporting cast somewhere that Jerry West doesn't necessarily overtake him either way.

Glad for your participation and feel free to speak more to your basketball beliefs, but I disagree with the idea that, in essence, ordering of players will stay similar enough in different basketball context that it wouldn't make much of a difference.

I will say that if we're going to put names to faces, I'd be thinking more about West vs Oscar than West vs Moses. Guys who were contemporary rivals of relatively similar roles where one guy seems to have had the more accomplished NBA career (Oscar) but the other was better at basketball (West) are the thing that really bugs me.

[Doctor MJ]

Historical A had a better career than Historical B
Modern C is better than Historical A at the modern game.
Historical B is better than Modern C at the modern game.

B > C > A > B

How do you decide who should be ranked first?

Tbh, I think there's a potential option C you might want to consider:

what if we averaged everyone over all of nba history? as in, you get time machined to every season and have your value averaged over all years. impossible to do, but would be my preferred method. russell's chart of relative value vs year is probably a slowly declining relative value line from his time to the present. most guys would probably have peak relative value in their own time (thus, why we think of them as being so good) and might be a little bumpy as the game changes over time. i wonder whose relative value line is the flattest? lebron?

I think Steph might be one of the big beneficiaries of this actually. To quote PF4 again:

yes, if he stepped out of a time machine, he would be like an alien and unstoppable. until rick mahorn got tired of it and forearm shivered him. but several weeks later when steph recovered, he would go back to dominating.

Honestly I feel Steph would just kill the 2000's, 90's, 80's, 70's ect, so while I currently have him outside of the top 10 using an era-relative framework, I wouldn't really protest him being put ahead of anyone he hasn't been directly outperformed by.

I appreciate the suggestion for other approaches on this but my initial thought on your suggestion doesn't seem appealing to me.

A key part of the reason why I'm considering backing off from an approach of a "who would be better if they were on teams with comparable fit and an environment that takes advantage of what we now know about basketball" is how hard it is to do. The approach you say seems to be looking to presume that I could do this for all environments. Seems harder, not easier.

I also don't like the idea that my evaluation of a guy - say Russell - will keep changing with time even if nothing further changes about the NBA simply because eventually the modern NBA will be represented by 50-100 years instead of 5-10. A continued re-evaluation of players relative to state-of-the-art strategy is very worthwhile to me, just really hard, but having a guy's ranking change without the state-of-the-art changing drains some of the meaning for me.

I do agree with you about Steph benefitting from this approach at least back through '79-80. I'd expect him getting hurt a great deal by the 3-point shot going away in the eras before that.