Are we overly penalizing guys on longevity based on era?

[Gooner]

Nobody is talking about the fact that his team has had terrible results without Anthony Davis, which kinda puts LBJ's impact into perspective.

Lakers with LeBron without Davis: +3.98 Net in 4340 minutes
Lakers with Davis without LeBron: -0.83 Net in 3083 minutes

What's this you were saying about impact?

Results tell you more than stats. Lakers missed the playoffs in LeBron's first season before AD came, and they missed the playoffs(and play-in) last season as AD missed most of the games. In the first round of the playoffs 2 years ago AD got injured in game 4 and Lakers got blown out by Phoenix the rest of the way. This season AD was injured and Lakers were out of the playoffs. Then he came back, they made some trades and they started to win games with LeBron being out with the injury. There is enough evidence that shows who is the most impactful player on that team. Last season was an embarrassment for the Lakers, but LeBron's longevity narrative never changed because he put up numbers.

[LA Bird]

Nobody is talking about the fact that his team has had terrible results without Anthony Davis, which kinda puts LBJ's impact into perspective.

Lakers with LeBron without Davis: +3.98 Net in 4340 minutes
Lakers with Davis without LeBron: -0.83 Net in 3083 minutes

What's this you were saying about impact?

Results tell you more than stats. Lakers missed the playoffs in LeBron's first season before AD came, and they missed the playoffs(and play-in) last season as AD missed most of the games. In the first round of the playoffs 2 years ago AD got injured in game 4 and Lakers got blown out by Phoenix the rest of the way. This season AD was injured and Lakers were out of the playoffs. Then he came back, they made some trades and they started to win games with LeBron being out with the injury. There is enough evidence that shows who is the most impactful player on that team. Last season was an embarrassment for the Lakers, but LeBron's longevity narrative never changed because he put up numbers.

Yeah, I am not wasting time with someone who thinks team net rating is not "results"

[70sFan]

That's simply not true. In 129 games with the Nets KD had the highest ppg, field goal percentage, 3 point percentage, free throw percentage, rpg and apg. I'm not making it up, it's easy to check. 29 ppg on 50/40/90 shooting with 7.1 rpg and 5.8 apg, to be exact. Best numbers of his career on a team in the regular season. It's true that in the last 2 postseaons he hasn't been the best, but he had statistically the best one in 2021.

Raw numbers:

2013-16 Durant: 29.1 ppg, 7.7 rpg, 4.9 apg, 1.2 spg, 1.0 bpg, 3.4 tov, 50.6 FG%, 39.7 3P%, 63.8 TS%
2021-23 Durant: 28.8 ppg, 7.1 rpg, 5.7 apg, 0.8 spg, 1.2 bpg, 3.4 tov, 53.7 FG%, 40.8 3P%, 65.6 TS%

"Advanced" numbers:

2013-16 Durant: 28.7 PER, .283 WS/48, 8.3 OBPM
2021-23 Durant: 25.9 PER, .199 WS/48, 6.3 OBPM

The only advantage Brooklyn KD has over OKC version is his improved passing. OKC Durant posted higher ppg in much slower pace and comparable efficiency in far less efficient league. He also posted higher rebounding numbers despite playing more at SF.

You compare 3 years in Nets vs 6 seasons in OKC that include KD's rookie and sophomore seasons, which downgrades his averages. By all accounts, 2013-16 Durant was more productive RS player than Brooklyn KD.

Improved passing should be considered when you talk about "all-around" game. You used the term "comparable efficiency". That's more appropriate for his ppg compared in those 2 periods. Efficiency gap is bigger. Overall, KD's stats with the Nets are the best clearly.

Boxscore production is counted in all these "advanced" stats and you see a massive difference between them.

Efficiency difference basically doesn't exist if you adjust for league average.

There is no reason to believe that Nets KD produced more than OKC Durant.

[trex_8063]

Top 25 guys in games played and season they entered the league, as a proxy for elite longevity - it takes just short of 16 full seasons to crack the list:

Parish 1977
KAJ 1970
Carter 1999
Dirk 1999
Stockton 1985
K. Malone 1986
KG 1996
Moses 1975
Willis 1985
LeBron 2004
Terry 2000
Duncan 1998
Kidd 1995
Miller 1988
Uncle Cliffy 1990
Kobe 1997
Pierce 1999
Payton 1991
Gilmore 1972
Crawford 2001
Buck Williams 1982
Andre Miller 2000
Hayes 1969
Allen 1997
Caldwell Jones 1974

ya, I think everyone on this board knows about Parish but I'm not sure what the context or relevance is to what is in the op. Maybe you could elaborate on it. Also worth noting that Parish didn't play over 32mpg until he was 30 years old and only did 4 times in his career. So could argue that Parish got pretty good load management throughout his career(career avg of 28.4mpg).

I think one way to alleviate [within your own criteria] the concerns regarding longevity as it relates to load-management, etc, is to not ONLY fixate on rate metrics, per-possession impact metrics, and general "how good was he [when on the court]"......but to also consider availability to be on the court (not just in terms of games played, but in terms of quarter-to-quarter and minute-to-minute availability).

To put a vague hypothetical on it, suppose the following:
*Player A has a player [on-court] quality/impact of 9.5 [on a scale of 1-10] and plays 33 mpg for about 65 games per year.
**Player B has a quality/impact of 8.5, but plays 40 mpg for an average 78 games per year.

Who is having the "better" seasons, year to year?
Player A is clearly better on a per-minute basis, but is it by enough to offset the nearly 1000 additional minutes of availability that Player B has?


A poster somewhere on page 2 made the very salient point that the amount of ground covered [on defense in particular] and the effort/energy expended on perimeter defense is FAR more than it was 50, or even 25-30, years ago. Thus comparing minutes isn't quite an apples to apples comparison across eras.
That said, load management is more of a focus today than it was in prior eras, that also cannot be questioned [imo].

At any rate, in effort to alleviate the above concerns regarding load management, it might be worthwhile to look also at MINUTES played, instead of just games played (EDIT: or seasons played). In fact, it may be useful to place MORE focus on the minutes than the games/seasons.
Vince Carter is 3rd all-time in games played; but do the HUNDREDS of games he played as a 15-17 mpg role player in the twilight of his career really matter all that much? I'm as focused and sympathetic toward meaningful longevity as ANYONE on this site.....but even for me, those years don't move the needle much.

And besides, within the umbrella of longevity (or at least meaningful longevity, and total career value), are we not talking about total mileage to a degree?

Below are the top 25 in total minutes played (and year came into pro basketball) [many of the same names as listed above (with a few notable exceptions), though the order changes somewhat]:

1. Kareem Abdul-Jabbar (1969)
2. Karl Malone (1985)
3. LeBron James (2003)
4. Dirk Nowitzki (1998)
5. Kevin Garnett (1995)
6. Jason Kidd (1994)
7. Elvin Hayes (1968) --notice he barely made the top 25 in games
8. Moses Malone (1974)
9. Kobe Bryant (1996)
10. Wilt Chamberlain (1959) - NOTE he isn't even close to the top 100 in games played, but top 10 here
11. John Stockton (1984)
12. Reggie Miller (1987)
13. Tim Duncan (1997)
14. Artis Gilmore (1971)
15. Gary Payton (1990)
16. John Havlicek (1962) - note is outside the top 30 in games played
17. Vince Carter (1999) - note is 3rd in games played
18. Ray Allen (1996)
19. Paul Pierce (1998)
20. Robert Parish (1976) - note is 1st in games played
21. Julius Erving (1971)
22. Joe Johnson (2001)
23. Hakeem Olajuwon (1984)
24. Oscar Robertson (1960) - again, not even close to the top 100 in games played
25. Carmelo Anthony (2003)

One other notable: Kevin Willis was 9th in games played; he's barely in the top 50 in minutes played, though.

Overall, I feel minutes (or at the very least incorporating minutes in addition to games/seasons played) gives a better representation of the effective longevity and/or TOTAL production/value of these guys.

Pretty much all decades with the exception of the 1950s are adequately represented above (Wilt being the only guy who entered the league in the 1950s, though Russell is 35th all-time in minutes).

Here's the distribution by decade entered the league (bearing in mind that the league got MUCH bigger going from 60s to modern era, so we should reasonably expect the representative numbers to get bigger for later decades, as well):
50s - 1
60s - 4
70s - 4
80s - 4
90s - 9
00s - 3

The 00's are potentially curved downward because players who came into the league in the LATE 00's have simply not had enough years to make it possible to crack the list, although Chris Paul is really the only one who MIGHT manage to crack the top 25 before he retires. I think it's safe to say the 00's will top out no higher than 4 of the top 25 [if that].
So that may provide an argument that effective longevity is going DOWN in recent times, although the # for the 90s muddies that a bit.

Other than that uptick in the decade of the 90s [or is that more reflective of the growth in league size??], it's pretty evenly distributed all the way back into the 60s. In fact, you could argue the 60s saw an uptick, too, considering the league was just 8-9 teams for most of the decade, yet it equals the decades after it (even exceeds the 00s for now).

[Doctor MJ]

Should some kind of era based curve be applied when it comes to longevity in the future?

I definitely factor in era context when considering longevity, and you can turn that into a curve as a guide, but there's not just one factor here.

As an example: I find the idea that we'd ever hold Bill Russell's longevity against him to be absurd. As if he had anything left to prove after 11 titles, and as if any team would choose some other player over Russell based on years 14 and beyond.

So for me when looking at someone like Russell in a career comparison, to me the question is really just about his best 13 years compare against the other player's best 13 years.

[frica]

Should some kind of era based curve be applied when it comes to longevity in the future?

I definitely factor in era context when considering longevity, and you can turn that into a curve as a guide, but there's not just one factor here.

As an example: I find the idea that we'd ever hold Bill Russell's longevity against him to be absurd. As if he had anything left to prove after 11 titles, and as if any team would choose some other player over Russell based on years 14 and beyond.

So for me when looking at someone like Russell in a career comparison, to me the question is really just about his best 13 years compare against the other player's best 13 years.

Don't even see how Russel's longevity can be held against him (even if he had something left to prove), he was an MVP-level player in his rookie year and he closed his career with an MVP-level season.
That's more MVP level seasons than Tim Duncan...

[Doctor MJ]

Should some kind of era based curve be applied when it comes to longevity in the future?

I definitely factor in era context when considering longevity, and you can turn that into a curve as a guide, but there's not just one factor here.

As an example: I find the idea that we'd ever hold Bill Russell's longevity against him to be absurd. As if he had anything left to prove after 11 titles, and as if any team would choose some other player over Russell based on years 14 and beyond.

So for me when looking at someone like Russell in a career comparison, to me the question is really just about his best 13 years compare against the other player's best 13 years.

Don't even see how Russel's longevity can be held against him (even if he had something left to prove), he was an MVP-level player in his rookie year and he closed his career with an MVP-level season.
That's more MVP level seasons than Tim Duncan...

That's true, but then again, wasn't really thinking of Russell vs Duncan as a real competition based on in-era dominance. (If someone just thinks Duncan is the better player that's more debatable.)

I think Kareem is the guy that's more the traditional debate here, and I'll add LeBron to the mix as he can specifically argue to have had an MVP-level run longer than Russell's entire NBA career.

[Owly]

Should some kind of era based curve be applied when it comes to longevity in the future?

I definitely factor in era context when considering longevity, and you can turn that into a curve as a guide, but there's not just one factor here.

As an example: I find the idea that we'd ever hold Bill Russell's longevity against him to be absurd. As if he had anything left to prove after 11 titles, and as if any team would choose some other player over Russell based on years 14 and beyond.

So for me when looking at someone like Russell in a career comparison, to me the question is really just about his best 13 years compare against the other player's best 13 years.

Don't even see how Russel's longevity can be held against him (even if he had something left to prove), he was an MVP-level player in his rookie year and he closed his career with an MVP-level season.
That's more MVP level seasons than Tim Duncan...

Rookie as MVP level ... that seems bullish.

The team performance improvement with Russell and Ramsey isn't that great (worse than the Sharman healthy spell). Not a perfect measure by any means but something that might make MVP level a tricky sell.

[Cavsfansince84]

I just want to do a quick shout out to some of the guys this thread was aimed more towards which is guys like Mikan, Arizin, Johnston, Pettit, Oscar, West and Baylor. I think it goes extra for guys who played above the rim somewhat and took a beating on their knees and ankles by doing so. Even Wilt despite his longevity was having major knee issues by like 64(which of course he played through).

[AEnigma]

Lakers with LeBron without Davis: +3.98 Net in 4340 minutes
Lakers with Davis without LeBron: -0.83 Net in 3083 minutes

What's this you were saying about impact?

Results tell you more than stats. Lakers missed the playoffs in LeBron's first season before AD came, and they missed the playoffs(and play-in) last season as AD missed most of the games. In the first round of the playoffs 2 years ago AD got injured in game 4 and Lakers got blown out by Phoenix the rest of the way. This season AD was injured and Lakers were out of the playoffs. Then he came back, they made some trades and they started to win games with LeBron being out with the injury. There is enough evidence that shows who is the most impactful player on that team. Last season was an embarrassment for the Lakers, but LeBron's longevity narrative never changed because he put up numbers.

Yeah, I am not wasting time with someone who thinks team net rating is not "results"

You do not even need net rating to make the point: over the past five years, LeBron has a 68-66 record without Davis, and Davis has a 47-59 record without LeBron.

Facts never matter to people like Gooner. The only goal is finding some slant which can superficially support their gut feelings or reflexive antipathies.

I definitely factor in era context when considering longevity, and you can turn that into a curve as a guide, but there's not just one factor here.

As an example: I find the idea that we'd ever hold Bill Russell's longevity against him to be absurd. As if he had anything left to prove after 11 titles, and as if any team would choose some other player over Russell based on years 14 and beyond.

So for me when looking at someone like Russell in a career comparison, to me the question is really just about his best 13 years compare against the other player's best 13 years.

Don't even see how Russel's longevity can be held against him (even if he had something left to prove), he was an MVP-level player in his rookie year and he closed his career with an MVP-level season.
That's more MVP level seasons than Tim Duncan...

Rookie as MVP level ... that seems bullish.

The team performance improvement with Russell and Ramsey isn't that great (worse than the Sharman healthy spell). Not a perfect measure by any means but something that might make MVP level a tricky sell.

By modern standards? Sure. By 1957 standards? I think you need to present who was better. Cousy won MVP that year, but I am certainly not taking him. Pettit won the prior year, narrowly finished second to Cousy that year, and went seven games against the Celtics in the Finals, so he has the strongest case, but usually we grant more than one “MVP-level” player in a given season. Paul Arizin perhaps. Still, if our standard is a twenty game sample with a strong backup in his place, then someone like Kawhi was never a qualifiedly “MVP level” player in 2017 or 2019 either, nor was Embiid this past year, and I suspect few people would agree with that type of standard.

I do think this board penalize older players too much for staying in college rather than turning professional earlier. For example, Bird was most likely an All-NBA caliber player during college and it doesn't make much sense to have Bird in those years as less valuable than say 97-99 Kobe just because it wasn't professional basketball.

I agree with this to a point, and Bird is a good example here… Kareem probably the best possible example, as his rookie season he was already arguably the best player in the league by the end of the season… but I have found it cuts both ways too, and Bird there is also a quintessential example: he was pretty undeniably better than Magic for at least the first three years of his career, and many would say the first seven, and plenty of people try to use that to give him the advantage… while ignoring he had three (well, two and a half) years on Magic. The comparison of those two looks a lot different when we acknowledge Magic had a gigantic edge when looking at what they were as players before turning 23, and from that lens, you can keep going farther: 1983-86 Magic over 1980-83 Bird is also an effortless case, and even 1987-91 Magic over 1984-88 Bird probably leans Magic for a majority.

Magic in general I think receives the worst of both. He had weak absolute longevity for basically no other reason than pure 1990s bigotry. His NBA career was finished at 31, with five titles, nine conference titles, three Finals MVPs, and three MVPs. Jordan was playing baseball at the same age. If Kareem had been functionally forced into retirement before his 32nd birthday, he would have one title (albeit plus five MVPs and the greatest college career ever).

I am fine making some attempt at assessing college years, but the key point should be that they are not at the expense of NBA production. I am not sure whether 1979 Magic at Michigan State is demonstrably better than 19-year-old Lebron (2005 as a full year rather than a season), but I am confident they were both a hell of a lot better than 19-year-old Jordan or Bird.

[OhayoKD]

I definitely factor in era context when considering longevity, and you can turn that into a curve as a guide, but there's not just one factor here.

As an example: I find the idea that we'd ever hold Bill Russell's longevity against him to be absurd. As if he had anything left to prove after 11 titles, and as if any team would choose some other player over Russell based on years 14 and beyond.

So for me when looking at someone like Russell in a career comparison, to me the question is really just about his best 13 years compare against the other player's best 13 years.

Don't even see how Russel's longevity can be held against him (even if he had something left to prove), he was an MVP-level player in his rookie year and he closed his career with an MVP-level season.
That's more MVP level seasons than Tim Duncan...

That's true, but then again, wasn't really thinking of Russell vs Duncan as a real competition based on in-era dominance. (If someone just thinks Duncan is the better player that's more debatable.)

I think Kareem is the guy that's more the traditional debate here, and I'll add LeBron to the mix as he can specifically argue to have had an MVP-level run longer than Russell's entire NBA career.

The case for anyone vs Russell era-relative is pretty weak yeah. Individual or team.

[OhayoKD]

Nobody is talking about the fact that his team has had terrible results without Anthony Davis, which kinda puts LBJ's impact into perspective.

Lakers with LeBron without Davis: +3.98 Net in 4340 minutes
Lakers with Davis without LeBron: -0.83 Net in 3083 minutes

What's this you were saying about impact?

Is now a good time to remind everyone that (at least by impact) the gap between Lebron's "prime" and everyone else post-merger(Kareem excepted) is as big as his advantage in terms of culminative value?






something something Port, something something Consistency

[eminence]

Nobody is talking about the fact that his team has had terrible results without Anthony Davis, which kinda puts LBJ's impact into perspective.

Lakers with LeBron without Davis: +3.98 Net in 4340 minutes
Lakers with Davis without LeBron: -0.83 Net in 3083 minutes

What's this you were saying about impact?

Is now a good time to remind everyone that (at least by impact) the gap between Lebron's "prime" and everyone else post-merger(Kareem excepted) is as big as his advantage in terms of culminative value?

something something Port, something something Consistency

LeBron leads in both there (I'd say for data-ball vs post-merger, as guys like Magic can't really be compared by similar metrics), but it would seem to me the cumulative lead is notably larger than the 'prime' version.

The first table would imply LeBron having a ~30-40% career value lead on Duncan/KG, with everyone else notably behind that, where the prime lead would be estimated at 5-15% over KG/Duncan (I don't have 5 year possession data to hand). Steph also likely closer than that 30-40% range, though availability brings him down a bit more. Likely others also closer than Duncan/KG for career (CP3 almost for certain), but just looking at that top 10.

[Owly]

Don't even see how Russel's longevity can be held against him (even if he had something left to prove), he was an MVP-level player in his rookie year and he closed his career with an MVP-level season.
That's more MVP level seasons than Tim Duncan...

Rookie as MVP level ... that seems bullish.

The team performance improvement with Russell and Ramsey isn't that great (worse than the Sharman healthy spell). Not a perfect measure by any means but something that might make MVP level a tricky sell.

By modern standards? Sure. By 1957 standards? I think you need to present who was better. Cousy won MVP that year, but I am certainly not taking him. Pettit won the prior year, narrowly finished second to Cousy that year, and went seven games against the Celtics in the Finals, so he has the strongest case, but usually we grant more than one “MVP-level” player in a given season. Paul Arizin perhaps. Still, if our standard is a twenty game sample with a strong backup in his place, then someone like Kawhi was never a qualifiedly “MVP level” player in 2017 or 2019 either, nor was Embiid this past year, and I suspect few people would agree with that type of standard.

Regarding a “need” to present who was better because “we usually grant more than one “MVP-level” player” I see multiple points of possible contention or requirement of further clarification here. Who is the “we”? Why do “ need” to match the standards of the “we”? Is it necessarily the case that it should be a given that there should several “Most Valuable Player” [superlative, singular] standard players (not to say it’s wrong, I get that it’s tiers so matching a range … maybe “MVP-contention” if [i]expecting multiple players, or are we baking in that due to bad voting [as well as other stuff] the average MVP may be a bit worse than the average best player? Personally I wouldn't)? And if I need to provide someone better doesn’t the debate stop being about a standard, couldn’t there not be any such players in a given year? If you think that there are different era standards for “MVP level” does it really make sense to talk about such a standard generically as a level (instead either an absolute standard [with or without the MVP tag, but if “with” then with an assumed calibration to a given era] or a more era specific term (and, fwiw, in reading the above where reacting to a poster about how he’s always MVP level, the standard at least might not be locked in to ’57 league standards).


I’m not sure if I’m following on more Russell specific stuff, but I’ll venture something. Is it an objection to the Celtics strength without Russell? Granting that people can weight this differently (I would say some Russell boosters are more bullish on other, I think cumulatively smaller without samples). I’m not locked in to a really specific weighting. But say Embiid (we'll assume he meets the standard), we know that his team was 10 points better with him on court than off it (even this … noise, caveats etc). I don’t know about better pure impact-y tools yet but I don’t see any reason to use anything worse (in case maybe the case was they held up W-L in his absence, idk).

Russell the best we can do for him is say they went from 4.541667 points margin per game before his arrival to 6.017241 after. That would be including playoffs (or it’s 5.770833) and note here that both playoff opponents had a negative SRS though one could certainly make a case that the with-Martin version of the Hawks were above neutral. That would be to credit Ramsey as granting no improvement and to ignore that the pre-Russell spell can be recognizably split into two chunks: 18 games with Sharman healthy, outscoring opponents by 129 points or 7.166666667 points per game and 6 games which Sharman plays 9 total minutes in one game (the fourth game, a narrow win) over which they are outscored by 20 or -3.333333333 per game (one could make that looks worse by taking out the Sharman 9 minute win).

I think even the best-case interpretation … Ramsey no better than those he replaced (and those he replaced no better in fewer minutes than those they kick out of the rotation), ignoring the Sharman injury, ignoring the cruder W-L picture (I think correctly), including the playoffs where schedule is imbalanced, making the decision after the fact knowing that it makes him look better … it’s not a huge lift. Something that I would say actively harms an “MVP level” case. And I do think the Sharman split and Ramsey are relevant.

Now as above this is one tool. Russell is quite productive (21.1 PER, narrowly 2nd on the team to Cousy, .176 WS/48, narrowly 4th on the team – behind Sharman, Ramsey and Cousy). We assume he was good beyond the boxscore (though that should show up in the impact side stuff). We can expect that he was good. The playoff production numbers, fwiw, do not in this instance tilt upwards as they often would (though a slight reduction – as here – might be thought to be holding steady in real terms, because of the higher standard of average player, though not necessarily in this instance significantly higher standard of opponent) overall, over his career. There’s significant fuzziness given so much less data at the time and that should be assumed to be baked in to what's said, but do a see a bunch of evidence for him at what I would think of as “MVP level” in this year, I don’t, and so I think seeing him as such requires a bullish reading.

And I *think* “a twenty game sample with a strong backup in his place” means that we’re regarding Russell’s backups as “strong”, yes? I suppose that depends in the first instance who you consider the backup … Conley, Risen, maybe some Jungle Jim minutes. Conley is the highest minutes guy overall, I don’t know how those minutes are split over the season. Regardless at a glance I’d regard Houbregs as a strong backup or Share as a strong backup, without a close survey regarding each roster, minutes, notional positions etc. Boston’s backups seem more acceptable than notably strong, are they better than Dukes, Felix, Hopkins … I don’t know (again lots of fuzziness here).

[AEnigma]

Rookie as MVP level ... that seems bullish.

The team performance improvement with Russell and Ramsey isn't that great (worse than the Sharman healthy spell). Not a perfect measure by any means but something that might make MVP level a tricky sell.

By modern standards? Sure. By 1957 standards? I think you need to present who was better. Cousy won MVP that year, but I am certainly not taking him. Pettit won the prior year, narrowly finished second to Cousy that year, and went seven games against the Celtics in the Finals, so he has the strongest case, but usually we grant more than one “MVP-level” player in a given season. Paul Arizin perhaps. Still, if our standard is a twenty game sample with a strong backup in his place, then someone like Kawhi was never a qualifiedly “MVP level” player in 2017 or 2019 either, nor was Embiid this past year, and I suspect few people would agree with that type of standard.

Regarding a “need” to present who was better because “we usually grant more than one “MVP-level” player” I see multiple points of possible contention or requirement of further clarification here. Who is the “we”? Why do “ need” to match the standards of the “we”? Is it necessarily the case that it should be a given that there should several “Most Valuable Player” [superlative, singular] standard players (not to say it’s wrong, I get that it’s tiers so matching a range … maybe “MVP-contention” if [i]expecting multiple players, or are we baking in that due to bad voting [as well as other stuff] the average MVP may be a bit worse than the average best player? Personally I wouldn't)? And if I need to provide someone better doesn’t the debate stop being about a standard, couldn’t there not be any such players in a given year? If you think that there are different era standards for “MVP level” does it really make sense to talk about such a standard generically as a level (instead either an absolute standard [with or without the MVP tag, but if “with” then with an assumed calibration to a given era] or a more era specific term (and, fwiw, in reading the above where reacting to a poster about how he’s always MVP level, the standard at least might not be locked in to ’57 league standards).

Not really seeing how an observation that there is little productive use in quibbling over “MVP-level” markers for a player without express disagreement over the idea that they could have been the best (or second-best) player in the league provoked this level of logorrhoea. If there were no MVP-level players in 1957 then that seems pedantic and uncharacteristically absolutist (in which case, perhaps 1957 Russell is merely at the level of someone like Clint Capela), and if there were only one MVP-level player then it still seems worth asking whether that is particularly distinct from the 1998 dynamics with Duncan weighed against Jordan, Malone, and Shaq.

I’m not sure if I’m following on more Russell specific stuff, but I’ll venture something. Is it an objection to the Celtics strength without Russell? Granting that people can weight this differently (I would say some Russell boosters are more bullish on other, I think cumulatively smaller without samples). I’m not locked in to a really specific weighting. But say Embiid (we'll assume he meets the standard), we know that his team was 10 points better with him on court than off it (even this … noise, caveats etc). I don’t know about better pure impact-y tools yet but I don’t see any reason to use anything worse (in case maybe the case was they held up W-L in his absence, idk).

Seems like something you could check easily rather than dancing around the speculation, but to save time, the 76ers were 10-5 without Embiid, with something like a +7 net rating. And Kawhi famously had similar WOWY results in 2017 and 2019. Yes, we have enough data to be confident that does not too adequately represent Embiid’s real value — but why would we assume an absence in Russell’s case?

Russell the best we can do for him is say they went from 4.541667 points margin per game before his arrival to 6.017241 after. That would be including playoffs (or it’s 5.770833) and note here that both playoff opponents had a negative SRS though one could certainly make a case that the with-Martin version of the Hawks were above neutral. That would be to credit Ramsey as granting no improvement and to ignore that the pre-Russell spell can be recognizably split into two chunks: 18 games with Sharman healthy, outscoring opponents by 129 points or 7.166666667 points per game and 6 games which Sharman plays 9 total minutes in one game (the fourth game, a narrow win) over which they are outscored by 20 or -3.333333333 per game (one could make that looks worse by taking out the Sharman 9 minute win).

I think even the best-case interpretation … Ramsey no better than those he replaced (and those he replaced no better in fewer minutes than those they kick out of the rotation), ignoring the Sharman injury, ignoring the cruder W-L picture (I think correctly), including the playoffs where schedule is imbalanced, making the decision after the fact knowing that it makes him look better … it’s not a huge lift. Something that I would say actively harms an “MVP level” case. And I do think the Sharman split and Ramsey are relevant.

Now as above this is one tool. Russell is quite productive (21.1 PER, narrowly 2nd on the team to Cousy, .176 WS/48, narrowly 4th on the team – behind Sharman, Ramsey and Cousy). We assume he was good beyond the boxscore (though that should show up in the impact side stuff). We can expect that he was good. The playoff production numbers, fwiw, do not in this instance tilt upwards as they often would (though a slight reduction – as here – might be thought to be holding steady in real terms, because of the higher standard of average player, though not necessarily in this instance significantly higher standard of opponent) overall, over his career. There’s significant fuzziness given so much less data at the time and that should be assumed to be baked in to what's said, but do a see a bunch of evidence for him at what I would think of as “MVP level” in this year, I don’t, and so I think seeing him as such requires a bullish reading.

And I *think* “a twenty game sample with a strong backup in his place” means that we’re regarding Russell’s backups as “strong”, yes? I suppose that depends in the first instance who you consider the backup … Conley, Risen, maybe some Jungle Jim minutes. Conley is the highest minutes guy overall, I don’t know how those minutes are split over the season. Regardless at a glance I’d regard Houbregs as a strong backup or Share as a strong backup, without a close survey regarding each roster, minutes, notional positions etc. Boston’s backups seem more acceptable than notably strong, are they better than Dukes, Felix, Hopkins … I don’t know (again lots of fuzziness here).

And the fuzziness is my point. If we are confident Russell comfortably clears the “impact” of someone like Pettit in their respective primes (and maybe you are not confident, but that would certainly make you an outlier here), then a couple of years prior to his prime, still on the best team in the league, I myself would not commit to the idea that rookie Russell would be at any definite disadvantage compared to prime Pettit. And more relevantly, considering the point of comparison here was 1998 Duncan, whether he is at such a disadvantage that 1998 Duncan can be passively accepted as “MVP-level” but saying the same of a title-winning 1957 Russell garners concerted pushback.

[Owly]

By modern standards? Sure. By 1957 standards? I think you need to present who was better. Cousy won MVP that year, but I am certainly not taking him. Pettit won the prior year, narrowly finished second to Cousy that year, and went seven games against the Celtics in the Finals, so he has the strongest case, but usually we grant more than one “MVP-level” player in a given season. Paul Arizin perhaps. Still, if our standard is a twenty game sample with a strong backup in his place, then someone like Kawhi was never a qualifiedly “MVP level” player in 2017 or 2019 either, nor was Embiid this past year, and I suspect few people would agree with that type of standard.

Regarding a “need” to present who was better because “we usually grant more than one “MVP-level” player” I see multiple points of possible contention or requirement of further clarification here. Who is the “we”? Why do “ need” to match the standards of the “we”? Is it necessarily the case that it should be a given that there should several “Most Valuable Player” [superlative, singular] standard players (not to say it’s wrong, I get that it’s tiers so matching a range … maybe “MVP-contention” if [i]expecting multiple players, or are we baking in that due to bad voting [as well as other stuff] the average MVP may be a bit worse than the average best player? Personally I wouldn't)? And if I need to provide someone better doesn’t the debate stop being about a standard, couldn’t there not be any such players in a given year? If you think that there are different era standards for “MVP level” does it really make sense to talk about such a standard generically as a level (instead either an absolute standard [with or without the MVP tag, but if “with” then with an assumed calibration to a given era] or a more era specific term (and, fwiw, in reading the above where reacting to a poster about how he’s always MVP level, the standard at least might not be locked in to ’57 league standards).

Not really seeing how an observation that there is little productive use in quibbling over “MVP-level” markers for a player without express disagreement over the idea that they could have been the best (or second-best) player in the league provoked this level of logorrhoea. If there were no MVP-level players in 1957 then that seems pedantic and uncharacteristically absolutist (in which case, perhaps 1957 Russell is merely at the level of someone like Clint Capela), and if there were only one MVP-level player then it still seems worth asking whether that is particularly distinct from the 1998 dynamics with Duncan weighed against Jordan, Malone, and Shaq.

You were the one who said someone else "had" to show that there were others. I'm not sure why an absolute standard would have Russell at Clint Capela levels, unless perhaps it's absolute time travel, rather than rather than something like X points per game above average starter (or per 100 or whatever). I wasn't responding to anything about rookie Duncan being MVP level.

I’m not sure if I’m following on more Russell specific stuff, but I’ll venture something. Is it an objection to the Celtics strength without Russell? Granting that people can weight this differently (I would say some Russell boosters are more bullish on other, I think cumulatively smaller without samples). I’m not locked in to a really specific weighting. But say Embiid (we'll assume he meets the standard), we know that his team was 10 points better with him on court than off it (even this … noise, caveats etc). I don’t know about better pure impact-y tools yet but I don’t see any reason to use anything worse (in case maybe the case was they held up W-L in his absence, idk).

Seems like something you could check easily rather than dancing around the speculation, but to save time, the 76ers were 10-5 without Embiid, with something like a +7 net rating. And Kawhi famously had similar WOWY results in 2017 and 2019. Yes, we have enough data to be confident that does not too adequately represent Embiid’s real value — but why would we assume an absence in Russell’s case?

On W-L
Why would I? I don't care about it and explicitly avoided using W-L (Boston's RS W-L got worse with Russell, specificity advocate avoiding it in the post, I do so because it's a cruder tool). Why is the worse tool, that was not used against Russell, relevant?

Besides which, It's also not only smaller in raw terms but also more so as a proportion of the season than the full without Russell sample or the Sharman sample (15/82 versus 24/72 or 18/72).

On net rating version
You've said it's easier and should have found it and then given "something like".

"Why would we assume an absence ..." I don't know what that means? If, "Why would we assume the more precise points margin isn't wildly misleading?" As consistently throughout the post, we recognize uncertainty but I don't think "Worse, cruder data was once off in a particular direction" should make that the default.

Russell the best we can do for him is say they went from 4.541667 points margin per game before his arrival to 6.017241 after. That would be including playoffs (or it’s 5.770833) and note here that both playoff opponents had a negative SRS though one could certainly make a case that the with-Martin version of the Hawks were above neutral. That would be to credit Ramsey as granting no improvement and to ignore that the pre-Russell spell can be recognizably split into two chunks: 18 games with Sharman healthy, outscoring opponents by 129 points or 7.166666667 points per game and 6 games which Sharman plays 9 total minutes in one game (the fourth game, a narrow win) over which they are outscored by 20 or -3.333333333 per game (one could make that looks worse by taking out the Sharman 9 minute win).

I think even the best-case interpretation … Ramsey no better than those he replaced (and those he replaced no better in fewer minutes than those they kick out of the rotation), ignoring the Sharman injury, ignoring the cruder W-L picture (I think correctly), including the playoffs where schedule is imbalanced, making the decision after the fact knowing that it makes him look better … it’s not a huge lift. Something that I would say actively harms an “MVP level” case. And I do think the Sharman split and Ramsey are relevant.

Now as above this is one tool. Russell is quite productive (21.1 PER, narrowly 2nd on the team to Cousy, .176 WS/48, narrowly 4th on the team – behind Sharman, Ramsey and Cousy). We assume he was good beyond the boxscore (though that should show up in the impact side stuff). We can expect that he was good. The playoff production numbers, fwiw, do not in this instance tilt upwards as they often would (though a slight reduction – as here – might be thought to be holding steady in real terms, because of the higher standard of average player, though not necessarily in this instance significantly higher standard of opponent) overall, over his career. There’s significant fuzziness given so much less data at the time and that should be assumed to be baked in to what's said, but do a see a bunch of evidence for him at what I would think of as “MVP level” in this year, I don’t, and so I think seeing him as such requires a bullish reading.

And I *think* “a twenty game sample with a strong backup in his place” means that we’re regarding Russell’s backups as “strong”, yes? I suppose that depends in the first instance who you consider the backup … Conley, Risen, maybe some Jungle Jim minutes. Conley is the highest minutes guy overall, I don’t know how those minutes are split over the season. Regardless at a glance I’d regard Houbregs as a strong backup or Share as a strong backup, without a close survey regarding each roster, minutes, notional positions etc. Boston’s backups seem more acceptable than notably strong, are they better than Dukes, Felix, Hopkins … I don’t know (again lots of fuzziness here).

And the fuzziness is my point. If we are confident Russell comfortably clears the “impact” of someone like Pettit in their respective primes (and maybe you are not confident, but that would certainly make you an outlier here), then a couple of years prior to his prime, still on the best team in the league, I myself would not commit to the idea that rookie Russell would be at any definite disadvantage compared to prime Pettit. And more relevantly, considering the point of comparison here was 1998 Duncan, whether he is at such a disadvantage that 1998 Duncan can be passively accepted as “MVP-level” but saying the same of a title-winning 1957 Russell garners concerted pushback.

"On the best team in the league" that were already (so far as we can tell) that without him ...

Cf previous stuff regarding multiple MVP level players each year.

Again, if I saw someone were claiming rookie Duncan as MVP level ... I imagine I'd have queried that too.

I'm out on this.

[AEnigma]

Regarding a “need” to present who was better because “we usually grant more than one “MVP-level” player” I see multiple points of possible contention or requirement of further clarification here. Who is the “we”? Why do “ need” to match the standards of the “we”? Is it necessarily the case that it should be a given that there should several “Most Valuable Player” [superlative, singular] standard players (not to say it’s wrong, I get that it’s tiers so matching a range … maybe “MVP-contention” if [i]expecting multiple players, or are we baking in that due to bad voting [as well as other stuff] the average MVP may be a bit worse than the average best player? Personally I wouldn't)? And if I need to provide someone better doesn’t the debate stop being about a standard, couldn’t there not be any such players in a given year? If you think that there are different era standards for “MVP level” does it really make sense to talk about such a standard generically as a level (instead either an absolute standard [with or without the MVP tag, but if “with” then with an assumed calibration to a given era] or a more era specific term (and, fwiw, in reading the above where reacting to a poster about how he’s always MVP level, the standard at least might not be locked in to ’57 league standards).

Not really seeing how an observation that there is little productive use in quibbling over “MVP-level” markers for a player without express disagreement over the idea that they could have been the best (or second-best) player in the league provoked this level of logorrhoea. If there were no MVP-level players in 1957 then that seems pedantic and uncharacteristically absolutist (in which case, perhaps 1957 Russell is merely at the level of someone like Clint Capela), and if there were only one MVP-level player then it still seems worth asking whether that is particularly distinct from the 1998 dynamics with Duncan weighed against Jordan, Malone, and Shaq.

You were the one who said someone else "had" to show that there were others.

My apologies, I was under the impression that the default nature of a comparisons board would suggest a comparative nature. There are a few ways to do that, and I suggested some worse fitting ones in the alternative to just looking at 1957 specifically, but that seems like a decent starting point — even if the conclusion ends up being “there were only one or zero MVP-level players that year.”

I’m not sure if I’m following on more Russell specific stuff, but I’ll venture something. Is it an objection to the Celtics strength without Russell? Granting that people can weight this differently (I would say some Russell boosters are more bullish on other, I think cumulatively smaller without samples). I’m not locked in to a really specific weighting. But say Embiid (we'll assume he meets the standard), we know that his team was 10 points better with him on court than off it (even this … noise, caveats etc). I don’t know about better pure impact-y tools yet but I don’t see any reason to use anything worse (in case maybe the case was they held up W-L in his absence, idk).

Seems like something you could check easily rather than dancing around the speculation, but to save time, the 76ers were 10-5 without Embiid, with something like a +7 net rating. And Kawhi famously had similar WOWY results in 2017 and 2019. Yes, we have enough data to be confident that does not too adequately represent Embiid’s real value — but why would we assume an absence in Russell’s case?

On W-L
Why would I? I don't care about it and explicitly avoided using W-L (Boston's RS W-L got worse with Russell, specificity advocate avoiding it in the post, I do so because it's a cruder tool). Why is the worse tool, that was not used against Russell, relevant?

Because this is not a private message and people have used crude win/loss to make those assessments — including on this very page, albeit as a false recollection of what the win/loss totals would suggest rather than an honest concern with them. If you want to restrict the future of this discussion to net rating, that is fine.

Besides which, It's also not only smaller in raw terms but also more so as a proportion of the season than the full without Russell sample or the Sharman sample (15/82 versus 24/72 or 18/72).

I do not see too much productivity in splitting hairs over the strength of a 24-game sample and a 15-game sample. Yes, the large one is better. It is within the realm of variation regardless.

On net rating version
You've said it's easier and should have found it and then given "something like".

I was going off memory because Statmuse limits search quantity now, but either way it is +7 net.

"Why would we assume an absence ..." I don't know what that means?

Because we lack on/off data for Russell that could suggest an Embiid-esque contradiction, we seem to be assuming that the result would not be similar.

If, "Why would we assume the more precise points margin isn't wildly misleading?"

I do not see it as particularly precise, no. Were Hakeem and the Rockets radically different in 1992 as compared to 1991? It is a decent indicator, but so is being the best player on a #1 SRS title team in a league with few other outlier talents and predating a run as one of the ~consensus ten best primes ever.

As consistently throughout the post, we recognize uncertainty but I don't think "Worse, cruder data was once off in a particular direction" should make that the default.

It is not once. All data can be noisy, true, but as with Embiid and Kawhi, we can make educated inferences with other data to oppose what might seem to be noise. The fact we cannot do that with Russell (or at least not with the same degree of ease) does not mean we then must blinding trust one point of data which is again relatively at odds with indicators from the rest of his career.

And on that note, the lack of comparative assessment is where it becomes difficult to even have a discussion where “MVP-level” is not defined and where there is not point of comparison by which we can approximate a definition.

Russell the best we can do for him is say they went from 4.541667 points margin per game before his arrival to 6.017241 after. That would be including playoffs (or it’s 5.770833) and note here that both playoff opponents had a negative SRS though one could certainly make a case that the with-Martin version of the Hawks were above neutral. That would be to credit Ramsey as granting no improvement and to ignore that the pre-Russell spell can be recognizably split into two chunks: 18 games with Sharman healthy, outscoring opponents by 129 points or 7.166666667 points per game and 6 games which Sharman plays 9 total minutes in one game (the fourth game, a narrow win) over which they are outscored by 20 or -3.333333333 per game (one could make that looks worse by taking out the Sharman 9 minute win).

I think even the best-case interpretation … Ramsey no better than those he replaced (and those he replaced no better in fewer minutes than those they kick out of the rotation), ignoring the Sharman injury, ignoring the cruder W-L picture (I think correctly), including the playoffs where schedule is imbalanced, making the decision after the fact knowing that it makes him look better … it’s not a huge lift. Something that I would say actively harms an “MVP level” case. And I do think the Sharman split and Ramsey are relevant.

Now as above this is one tool. Russell is quite productive (21.1 PER, narrowly 2nd on the team to Cousy, .176 WS/48, narrowly 4th on the team – behind Sharman, Ramsey and Cousy). We assume he was good beyond the boxscore (though that should show up in the impact side stuff). We can expect that he was good. The playoff production numbers, fwiw, do not in this instance tilt upwards as they often would (though a slight reduction – as here – might be thought to be holding steady in real terms, because of the higher standard of average player, though not necessarily in this instance significantly higher standard of opponent) overall, over his career. There’s significant fuzziness given so much less data at the time and that should be assumed to be baked in to what's said, but do a see a bunch of evidence for him at what I would think of as “MVP level” in this year, I don’t, and so I think seeing him as such requires a bullish reading.

And I *think* “a twenty game sample with a strong backup in his place” means that we’re regarding Russell’s backups as “strong”, yes? I suppose that depends in the first instance who you consider the backup … Conley, Risen, maybe some Jungle Jim minutes. Conley is the highest minutes guy overall, I don’t know how those minutes are split over the season. Regardless at a glance I’d regard Houbregs as a strong backup or Share as a strong backup, without a close survey regarding each roster, minutes, notional positions etc. Boston’s backups seem more acceptable than notably strong, are they better than Dukes, Felix, Hopkins … I don’t know (again lots of fuzziness here).

And the fuzziness is my point. If we are confident Russell comfortably clears the “impact” of someone like Pettit in their respective primes (and maybe you are not confident, but that would certainly make you an outlier here), then a couple of years prior to his prime, still on the best team in the league, I myself would not commit to the idea that rookie Russell would be at any definite disadvantage compared to prime Pettit. And more relevantly, considering the point of comparison here was 1998 Duncan, whether he is at such a disadvantage that 1998 Duncan can be passively accepted as “MVP-level” but saying the same of a title-winning 1957 Russell garners concerted pushback.

"On the best team in the league" that were already (so far as we can tell) that without him ...

In the regular season? Possible, but worth noting the prior season saw them second in a clustered group (all trailing the champion Nationals) and losing in their first postseason round. I think many eras would look positively on a player elevating that to a title, with again Kawhi and the Raptors being a fair analogy.

Cf previous stuff regarding multiple MVP level players each year.

I think taking the position that the 1950s had only one or even none is more “bullish” than placing 1957 Russell on that level relative to his league, and again should merit some attempt at comparison to support that idea. Instead we see a flat rejection off an isolated 24-game WOWY sample.

I am not opposed to the idea that Russell was not “MVP-level” that year, nor am I opposed to using that sample to support the argument. However, making that sample the entirety of the argument falls flat when without any point if comparison he could still qualify as one of the two or three most valuable players in the league, and I think most people tend to equate “second-best in the league” with “MVP-level”. Again, not a given, but any pushback on that seems like it should demand more than what you cared to offer.

[ty 4191]

I think we have already entered into an era where players' primes are extending out in a way they didn't used to with maybe a couple of exceptions.

I do think we absolutely need to adjust for longevity and primes based on era.

Consider a few facts:

From 1959-1960 through 1979-1980, only 4 players played 10,000+ minutes after their age 33 season. The average MP of the top 25 in minutes was 7,188 career MP after their age 33 season.

Between 2003-2004 through 2022-2023, 21 players played 10,000+ minutes after their age 33 season. The average MP of the top 25 was 13,125 MP after their age 33 season.

Yes, there were more teams and players (last 20 years vs. 1960-1980), but, clearly, even after adjusting for that, guys simply last much longer today.

[OhayoKD]

Not really seeing how an observation that there is little productive use in quibbling over “MVP-level” markers for a player without express disagreement over the idea that they could have been the best (or second-best) player in the league provoked this level of logorrhoea. If there were no MVP-level players in 1957 then that seems pedantic and uncharacteristically absolutist (in which case, perhaps 1957 Russell is merely at the level of someone like Clint Capela), and if there were only one MVP-level player then it still seems worth asking whether that is particularly distinct from the 1998 dynamics with Duncan weighed against Jordan, Malone, and Shaq.

You were the one who said someone else "had" to show that there were others.

My apologies, I was under the impression that the default nature of a comparisons board would suggest a comparative nature. There are a few ways to do that, and I suggested some worse fitting ones in the alternative to just looking at 1957 specifically, but that seems like a decent starting point — even if the conclusion ends up being “there were only one or zero MVP-level players that year.”

Seems like something you could check easily rather than dancing around the speculation, but to save time, the 76ers were 10-5 without Embiid, with something like a +7 net rating. And Kawhi famously had similar WOWY results in 2017 and 2019. Yes, we have enough data to be confident that does not too adequately represent Embiid’s real value — but why would we assume an absence in Russell’s case?

On W-L
Why would I? I don't care about it and explicitly avoided using W-L (Boston's RS W-L got worse with Russell, specificity advocate avoiding it in the post, I do so because it's a cruder tool). Why is the worse tool, that was not used against Russell, relevant?

Because this is not a private message and people have used crude win/loss to make those assessments — including on this very page, albeit as a false recollection of what the win/loss totals would suggest rather than an honest concern with them. If you want to restrict the future of this discussion to net rating, that is fine.

Besides which, It's also not only smaller in raw terms but also more so as a proportion of the season than the full without Russell sample or the Sharman sample (15/82 versus 24/72 or 18/72).

I do not see too much productivity in splitting hairs over the strength of a 24-game sample and a 15-game sample. Yes, the large one is better. It is within the realm of variation regardless.

On net rating version
You've said it's easier and should have found it and then given "something like".

I was going off memory because Statmuse limits search quantity now, but either way it is +7 net.

"Why would we assume an absence ..." I don't know what that means?

Because we lack on/off data for Russell that could suggest an Embiid-esque contradiction, we seem to be assuming that the result would not be similar.

If, "Why would we assume the more precise points margin isn't wildly misleading?"

I do not see it as particularly precise, no. Were Hakeem and the Rockets radically different in 1992 as compared to 1991? It is a decent indicator, but so is being the best player on a #1 SRS title team in a league with few other outlier talents and predating a run as one of the ~consensus ten best primes ever.

As consistently throughout the post, we recognize uncertainty but I don't think "Worse, cruder data was once off in a particular direction" should make that the default.

It is not once. All data can be noisy, true, but as with Embiid and Kawhi, we can make educated inferences with other data to oppose what might seem to be noise. The fact we cannot do that with Russell (or at least not with the same degree of ease) does not mean we then must blinding trust one point of data which is again relatively at odds with indicators from the rest of his career.

And on that note, the lack of comparative assessment is where it becomes difficult to even have a discussion where “MVP-level” is not defined and where there is not point of comparison by which we can approximate a definition.

And the fuzziness is my point. If we are confident Russell comfortably clears the “impact” of someone like Pettit in their respective primes (and maybe you are not confident, but that would certainly make you an outlier here), then a couple of years prior to his prime, still on the best team in the league, I myself would not commit to the idea that rookie Russell would be at any definite disadvantage compared to prime Pettit. And more relevantly, considering the point of comparison here was 1998 Duncan, whether he is at such a disadvantage that 1998 Duncan can be passively accepted as “MVP-level” but saying the same of a title-winning 1957 Russell garners concerted pushback.

"On the best team in the league" that were already (so far as we can tell) that without him ...

In the regular season? Possible, but worth noting the prior season saw them second in a clustered group (all trailing the champion Nationals) and losing in their first postseason round. I think many eras would look positively on a player elevating that to a title, with again Kawhi and the Raptors being a fair analogy.

Cf previous stuff regarding multiple MVP level players each year.

I think taking the position that the 1950s had only one or even none is more “bullish” than placing 1957 Russell on that level relative to his league, and again should merit some attempt at comparison to support that idea. Instead we see a flat rejection off an isolated 24-game WOWY sample.

I am not opposed to the idea that Russell was not “MVP-level” that year, nor am I opposed to using that sample to support the argument. However, making that sample the entirety of the argument falls flat when without any point if comparison he could still qualify as one of the two or three most valuable players in the league, and I think most people tend to equate “second-best in the league” with “MVP-level”. Again, not a given, but any pushback on that seems like it should demand more than what you cared to offer.

It's always interesting to see posters take the 24-game sample indicating russ may not have been the most valuable rookie seriously, but then disregard the 82-game sample from 1970 which more or less puts to bed the idea that anyone should be ranking anyone's peak or prime higher outside of time-machine considerations.

[Cavsfansince84]

I do think we absolutely need to adjust for longevity and primes based on era.

Consider a few facts:

From 1959-1960 through 1979-1980, only 4 players played 10,000+ minutes after their age 33 season. The average MP of the top 25 in minutes was 7,188 career MP after their age 33 season.

Between 2003-2004 through 2022-2023, 21 players played 10,000+ minutes after their age 33 season. The average MP of the top 25 was 13,125 MP after their age 33 season.

Yes, there were more teams and players (last 20 years vs. 1960-1980), but, clearly, even after adjusting for that, guys simply last much longer today.

Ya I think this somehow turned into a current player vs 90's guys thread but it was meant to be more about guys who played in the 50's/60's/70's. I used current players to demonstrate simply how multiple guys have had one or more season ending injuries and still are able to play at near mvp level at age 34 plus LeBron but my larger point is that where this could be trending over the next 10-15 years. As in if superstars putting up mvp caliber seasons at age 35-40 becomes something of a norm. Even Kareem for instance probably had his last real mvp caliber season at 33-35(depending how you see it).