Likelihood of 4 of the 20 best players of all time being from the 60s

Home for all your discussion of basketball statistical analysis.
Post Reply
Dr Positivity
Posts: 298
Joined: Thu Sep 20, 2012 6:44 pm

Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Dr Positivity » Thu Feb 28, 2019 3:41 am

Russell, Wilt, West, Oscar are regularly ranked top 20 all time, which is statistically less likely as:

- Strictly from a timeframe perspective, the NBA has been around 73 years with nearly 6 full decades from 60-now even if one disregards the mostly white first part, so either way 4/20 is a bigger proportion than the 60s era has in NBA history which is closer to 1/6 than 1/5

- An argument I have seen used on another board is that since basketball is more popular now and the population is bigger, a lot more people play it hence the competition to be the best in the world is stronger. So this person argued what Hakeem is more impressive than what Wilt or Russell did even if they dominated their era more, because he was a top 3 player in the world against harder competition in the population.

So this would make it statistically less likely that 4 of the top 20 are from the 60s. However it doesn't work that cleanly. Just because it's less likely doesn't mean it didn't happen. There is some % chance that by 4 of the 20 best players of all time or 4 of the 20 most talented (which would be even more impressive) were in the 60s. For example the 3 best/most talented tennis players of all time playing at the same time is also unlikely, but it appears by fluke that's what happened.

My question is - Can you figure out the statistical odds of the 60s having 4 of the top 20 basketball players? Using population, length of NBA history, etc.

Mike G
Posts: 4413
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Mike G » Thu Feb 28, 2019 1:10 pm

Dr Positivity wrote:
Thu Feb 28, 2019 3:41 am
Can you figure out the statistical odds of the 60s having 4 of the top 20 basketball players?...
You would have to make some assumptions like "is competition equivalent" and then statistically define "top" players.
Or you could do it the old time way and just rank your favorite players.

Russell, Wilt, West, and Oscar all entered the league within half a decade (1957-61) actually, so the odds are even longer.
D Wade has made 6 (of 22) threes in 7 games, but 5 of them were last night. Stuff like this does happen.

Wilt at 7-2 and super-athletic would be a superstar in any era, if his attitude allows it. Russell for a time was thought to be too small to dominate at center in later decades, but he might fit right into the current game. And he had rare mental acuity. West and Oscar had skills and durability, before modern training and medicine.

In regular-season Win Shares, Wilt ranks #2, Oscar 10, Russell 18, and West 20.
In postseason, it's Wilt 6, Russell 9, West 10; Oscar is 50th.
West and Oscar are roughly Curry and Paul. Wilt is Shaq.

Mike G
Posts: 4413
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Mike G » Thu Feb 28, 2019 2:15 pm

Since I have rankings that include postseasons, ABA equivalents, and other era equalizers, I notice some historic discontinuities.
Expanding to top 30, we get players that someone would call top 20, and another cutoff point for perspective. Using year of entry (NBA or ABA), I've grouped them chronologically by within-2yr time of entry; line breaks therefore are 3 or more years gap.

Code: Select all

season1    future top 30
1955-61   Pettit, Russell, Baylor, Wilt, West, Oscar
1970-72   Kareem, Erving
1975      Moses
1980      Magic, Bird
1984-90   Drexler, Jordan, Olajuwon, Barkley, Stockton, Malone, Ewing, Pippen, Robinson
1993      Shaq
1996-99   Garnett, Kobe, Duncan, Nowitzki
2004-09   LeBron, Wade, Paul, Durant, Westbrook
In 3 years, 1984-86, the league got 7 alltime greatests; 4 in 1985 alone.
Before that, just 2 in 8 seasons ('76-83), just Shaq thru half of the '90s.
Only 5 in 22 years, from West/Oscar to the mid 80's! (Maybe 4 more in top 40.)

Curry and Harden are hurtling toward this territory. Of course, these later years will be understating career totals, since we are looking at guys still in their prime.

Crow
Posts: 6146
Joined: Thu Apr 14, 2011 11:10 pm

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Crow » Sun Mar 03, 2019 3:24 am

I am generally not interested in the far past NBA or era comparison; but if someone used the latest software to analyze historical gametape I might have some curiosity.

Calculate average speed & acceleration of the top players vs. average opponents. Degree of shot contest. Height of rebound capture. What it took to get a foul call.

Mike G
Posts: 4413
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Mike G » Sun Mar 03, 2019 8:26 pm

Is the trend toward better eFG% more a result of rules changes, or has shooting mostly gotten better?
League eFG% cratered below .480 from 1998 to 2004. ORtg <105 in that era.
Now eFG% is .524 and ORtg = 110, both alltime highs. It looks to me like a wealth of talented scorers that the defenses cannot contain.
3FG% has stopped rising and is at 2000 level, but frequency continues to increase. Now twice as many as in 2006, which was itself the best offensive season since the shortened-arc years. Players hit contested 3's regularly now.

Crow
Posts: 6146
Joined: Thu Apr 14, 2011 11:10 pm

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Crow » Mon Mar 04, 2019 4:01 am

Certainly the freedom of movement rules and the 3 point shot taking revolution are big factors. Improved natural shooting ability may have a little to do with it; but looking at the long history, I doubt that is as significant as the other two.

pdevos
Posts: 34
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by pdevos » Fri Apr 19, 2019 2:14 pm

Freedom of movement, 3pt emphasis, no hand checking, allowance/very liberal interpretation of the "gather" process for euro steps and 2-3 steps in general, and now players can turn their wrists, pause, and still dribble mid dribble process. This allows for incredible control and sharp movements while also pausing defenders (who also can't hand check) giving them the big advantage for eFG%. Players in the 60s especially had to shoot almost immediately after gathering, this was a HUGE advantage for bigs as you couldn't euro-step or any other 'continued' process after gathering. These are tremendous advantages for defensive bigs in the 60s, somewhat into the 70s, and now today's rules, are incredible advantages for perimeter players. ergo, we haven't seen a "post" slash "defensive big" win MVP since Duncan & Garnett's run from 2002 to 2004. We may never see it again. Maybe if Embiid is ever able to play a healthy 70+ games. Even so, the eurostep liberties, initiated offensive contact advantages for offensive players (Giannis, Harden, LeBron, even Embiid, etc) just give so much possibility to perimeter playmakers.

That said, outliers are outliers as Mike laid out in his 3pt shooting Dwyane Wade example. That 1984 class is still ridiculous. Depending on where you rank Hakeem, Barkley, and Stockton. Probably best all-time single year addition to the NBA. Something could probably be said for the ABA merger with the NBA in 1976 where you had Gilmore, Erving, Gervin, etc join the NBA in one season.

Sorry, long winded... Absolutely can see 4 of the top 20 players still come from the 60s. Context of rules, playing 3-4 night back-to-backs, performance relative to competition (several SDs better than competition), and the fact we can attribute many efficiency metrics to rule changes (3pt shot, other rules, emphasis, how traveling is called, etc). It's similar to how the Fosbury Flop affected high jumping. The idea the genetics of players today is so much better than the 1960s is false, ask any evolutionary biologist or psychologist and they'll laugh you out of the room saying that's not enough time for genetic mutations to occur. Nutrition advantages, sure. Training advantages, sure. But these are great for median cases. You don't get genetic advantages where they are "prevalent" enough to produce very atheltic 7'2 275+pound people (e.g. Wilt Chamberlain) with greater occurrences. Or Havlicek's lung capacity. Or even heart size's. Pete Maravich died at 40 due to heart issue. My cousin, former NBA player Alec Kessler, a much more recent player, also died of such a case. Len Bias, Reggie Lewis, etc. There are build in genetic things that give advantages/disadvantageous to players. Shooting, it's something where metacognitively some people can tie cognitive "too long...too short...too left..." and tie micro adjustments in their physical mechanics to that mental compute. Some never will be able to do that. So practice can help a little, but not enough to make Shaq shoot 85% FT or Ben Simmons hit 35% 3s. But Oscar Robertson, top 5 FG%, FT% champ during the 1960s...shooting 3s today? You betcha. I'd bet 40%+ easily based on seeing his shot distances while playing + statistical correlations for someone shooting that well in those areas.

This doesn't even account for the racial issues the black players faced back then. DeMar DeRozan, Kevin Love, and other players have come forward and shown how much psychological depression and other issues affect their play. Imagine if Jerry West had more help even for his personal issues? How about Larry Bird skipping college for a season because he couldn't handle how big Bloomington, IN (Indiana University) was at age 18? How much more normalized that would have been today! Amazing changes. So today we get a lot more people not missing for edge cases, medical, psychological, awareness (NBA is an international game and well paid). But that helps increase chances of a few outliers (still very rare) and mostly will see the effects in the "mid-tier" talent increase. But we've also seen a big change in rules, coaching philosophy where Bill Russell types and even large 7'+ players aren't as prized.

The big problem might be in how you assess your "top players"...if you highly value "defense" and "team player" and "sacrificed" for the team... Bill Russell might be your #1 best player all time. But, also, the very definition of "team player" (no I in team), "sacrificed" for the team suggests he wasn't the best individual player he could be. But even so, and I realize I may get stoned for saying this, but in that light... Bill Russell -- when adjusting for pace, FG% relative to league and putting him in today's game -- the best modern day comparison I have is Ben Wallace. Bill would average 9-13 ppg, 12-16 rpg, 1.5-2.2 spg, 3.5-5.5 apg, and 3.5-5.5 bpg. Except for the assists, Ben Wallace is a very fair comparison. And Ben Wallace, tho not NBA MVP during those 7 dominant years in Detroit, he was no doubt the team MVP. He finished 3 times in the top 10 of MVP voting (a proxy for "best-ness") suggesting he was 3x a top 10 player. He was All-NBA 5x (2nd: 3x, 3rd: 2x) so by that proxy, a top 10 player 3x, top 15 2x -- I realize it's positional, so it's not a "pure" top X player comparison -- but just trying to get as many data points as possible to proxy that "top X" position per season. Very few people would put a guy who averages 11 ppg an 14 rpg and 4 bpg as a top 20 NBA player. Well, adjustments wise, that's Ben Wallace 2.0, aka Bill Russell if you do the linear weights/regressions.

Oscar**(Correction noted), West, and Chamberlain had more complete games. All were considered "All league defenders" at some point in their careers as well as leaders offensively. Just have to question if West, a 6'2" guy (with incredibly long arms) and a great outside jump shot, cough Steph Curry, cough Steph Curry... would be great today as West's stats do translate well with all adjustments made to complete the assize.

I think the beauty of "top lists" is you can try to get as many factors as you want and then try to weight them. I personally love it when people go beyond the anecdotal arguments..."I saw...." and team accomplishments "11 rings...6 rings..." and try to put those things in context as well. But this post is getting long or I'd put that list out as well -- factors I like to consider and try to quantify.

Edits: a few poor spellings & as Mike pointed out, Oscar was NOT known as a stellar defender.
Last edited by pdevos on Fri Apr 19, 2019 11:22 pm, edited 4 times in total.

pdevos
Posts: 34
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by pdevos » Fri Apr 19, 2019 2:17 pm

I apologize, I didn't really get into a "statistical likelihood" in my last answer. TL;DR you have to first figure out the factors that make a player a "top X player" and then assize how likely they are to have occurred. Then measure those things, quantifiably, within an era and cross era. Very tough to do. But would happily try to think about those factors. If that's the direction the OP want(ed) to go here. I apologize, I didn't mean to hijack the post, if I deviated at all from the intended direction.

Mike G
Posts: 4413
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Mike G » Fri Apr 19, 2019 4:49 pm

That was a great post! But I never heard of Oscar being a great defender. West in his last season got an awesome 3.6 Stl/100 poss, while Oscar got 1.4

While the NBA-feeding population has grown since the early '60s (southern blacks, internationals), the occurrence of coordinated 7-footers probably has not kept up with that expansion. Along with the value of the 3-point shot, there just isn't the proportional increase in defense over the years.

Rules changes and allowances seem to favor dribbling and driving, and this helps open up the outside shot. But it sure looks as if shooting itself has gotten better. Steph Curry is no longer totally better than everyone else, at his own level. He's just one of several up there.

When Oscar arrived hitting 50% of his shots, Bob Cousy was the default top guard, and he never reached 40%.
Oscar developed some but by age 28 or so, when maximum expansion should have made for easy pickings, his numbers just declined. West did the opposite, statistically peaking right up to the end.

pdevos
Posts: 34
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by pdevos » Sat Apr 20, 2019 12:02 am

Trying to stay on topic vs making this a "top X list" -- 4 of top 20 players all-time playing in the 60s. Yes. I'd also submit Elgin Baylor and Nate Thurmond for debate.

Mike -- Great correction on Oscar, yes, not known as a good defender. I think many of his teams were very poor in Defense actually now that I think of it. (http://www.backpicks.com/2017/12/21/bac ... robertson/)

Are there more than 4 players from the 60s?

The question, for me becomes, how do you rank somewhat or very strongly "one-dimensional" players in your ranking?

For instance, a few names that might be [or would be] in the top 20 for many people -- if a guy can only take over one phase of the game. e.g. Bill Russell (defense), Oscar (offense), Dirk Nowitzki (offense), Steve Nash (offense)...

I would concede a top defender 60s and 70s (pre gather step, euro step changes and 3pt line) had way more impact on the games then vs now. Probably even up to the late 1980s. (Mark Eaton? Bobby Jones?)

But just how high can these one dimensional guys be ranked? Can they still crack the top 20? Especially if they're Ben Wallace offensively? Or if they're Bill Russell offensively? Or Nate Thurmond offensively?

If Bill Russell is top 20 or top 10 or "top x" and Nate Thurmond is regarded as the better defender*** by top offensive players who played against Bill Russell can we also not say Thurmond is right in the same ballpark ranking wise as Russell?

How high does Nate Thurmond Rank?

When I watch highlights -- some Warriors highlights, but mostly Cavaliers & Bulls games (old Nate, 33-34 years old) he had a super smooth jumper from 15-18 feet and was a pretty crafty passer despite not being heavy volume or in the case of the Cavs/Bulls games, not heavily involved in the offense. That said, his career FG% was just 42%. Russell's was 44%. I do think Nate had much better range. His FT% was 67% compared to 56% for Russell. Basically trying to figure how much better offensively he was, e.g. if he really had a jump shot where 42% seasons aren't that bad for a mid-range jump shooter.

***Wilt and Kareem both said Nate was the best defender they ever played. Wilt played prime Russell. Kareem played old Wilt, prime Bill Walton, young Hakeem Olajuwon, prime Mark Eaton, prime Artist Gilmore, prime Willis Reed, prime Elmore Spencer, prime Caldwell Jones, etc. So it's a pretty solid proxy for comparing defenders. So was Thurmond the better defender between him and Russell? Thurmond was a "top 50 player" (1996)...

e.g. Cavs Playoffs vs Celtics
https://www.youtube.com/watch?v=gCeuA42oOJQ

pdevos
Posts: 34
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by pdevos » Sat Apr 20, 2019 12:40 am

Also curious, where Rick Barry would rank for many of you?

I realize his best years were in the 70s. But he did play (3 strong seasons) in the 60s. I've seen a lot of public lists having him from 21st to mid 40s.

Mike G
Posts: 4413
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Mike G » Sat Apr 20, 2019 1:28 am

Between 1965 and 1975, Thurmond played 2000+ minutes each year, and in that 11 years he was 11th in FGA, NBA+ABA.
100 players attempted at least half as many FG, and Nate was 6th from the bottom in eFG% at .423
Among centers, it's even worse -- Most FGA in the era and worst FG% among the top 30 in FGA in that period.
http://bkref.com/tiny/ZlhOY

Bill Russell's FG% landed in the top 5 in the NBA in his first 4 seasons. Yeah, 45% was pretty great in the 50's.
He never improved, though, so he just shot less.

Win Shares doesn't think these two are very comparable.
http://bkref.com/tiny/uX6Fd
Russell with more than twice the RS WS, big lead in WS/48, .193 to .104 -- Nate a barely average player.
In playoffs, the difference is just as stark -- .178 to .90 ws/48 -- Russ with more than 5x the PO WS total.

In 1967, the MVP voting was Wilt-Nate-Russ. Other than that, Nate got some 8th-11th places.
Russell won 5 MVP's and was top 4 in the vote every year after his rookie season.

pdevos
Posts: 34
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by pdevos » Sat Apr 20, 2019 4:30 pm

Pretty solid case. I need to revisit Win Shares. Looks like it's not based entirely on a team win total.
https://www.basketball-reference.com/about/ws.html

Win Shares

It is somewhat interesting to me that at age 33, Nate's DWS (Chicago) was his 4th best DWS season. So 7 seasons worse prior with a much younger Nate Thurmond defensively. The difference could be because of better stat tracking (DReb, Blk, Stl)? But Bob Love (2nd team Def) and Jerry Sloan (1st team Def) also on that Bulls team that CHI finished 2nd in DRtg.

Either way, in the value comparison to Bill Russell defensively, could probably safely assume Russell's defense would rise with that tide [of better granular defensive stats] as well.

One thing that I'd have to confirm and I'm pretty sure is true, that a defensive center has very high impact on DRtg.

https://www.basketball-reference.com/pl ... mna01.html

Career and Win Shares

Warriors were awful year before Nate arrived then became top defensive team in rookie year. He didn't start tho. After that 3rd defense is common, but also a few mid tier finishes. Rick Barry really helped them jump in wins when he was around Warriors were in the Finals or competitive.

Warriors team's Nate was on were pretty bad save for Rick Barry. Nate, played with Wilt Chamberlain for 2.5 years.

All-NBA Teammates of the Top 1960s Centers

Nate's All-NBA Teammates (playing with top 10 players)
Wilt Chamberlain - 2.5x
Rick Barry - 4x

Nate Thurmond - 6.5 All-NBA teammates in 14 seasons. W-L 513-450 (53.2%)
Bill Russell - 25 All-NBA teammates in 13 seasons. W-L 690-273 (71.7%)
Wilt Chamberlain - 10 All-NBA teammates in 14 seasons. W-L 672-373 (64.3%)

Maybe could figure out a Wins to All-NBA teammates ratio. Not sure how telling that would be for expected Win % tho. And Win shares are pretty a solid case (Justin Kubatko's math I believe and it is not "pre-learned" bias of other years IIRC).

Mike G
Posts: 4413
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Likelihood of 4 of the 20 best players of all time being from the 60s

Post by Mike G » Sun Apr 21, 2019 1:36 am

Win Shares really dislikes a low shooting% with high shot volume.
As a consequence, Thurmond never led his team in WS; a couple of close 2nds, a lot of 3rd and 4th. Even playing a lot of minutes.
http://bkref.com/tiny/E9b5k
I don't know why he shot so much and so badly. He was 6-11 and towered over most others.

Since 1965, there are 57 instances of a center taking 800 shots in a season with an eFG% less than .450. Nine of them are Thurmond.
http://bkref.com/tiny/z4rfJ

Oh, and Russell led his team, the best team, in WS for 10 straight seasons. Bad FG% and all.
http://bkref.com/tiny/lLCxq
In 1964 his WS were more than any two teammates'.

Post Reply