Proper definition of Longevity

[migya]

Longevity must have a proper definition in analysis of nba players. 1000 games at Allnba level and 200 extra games at starter level is only marginally better than 1000 games only.


Example:

RS:
Player 1 - 1300gm, 35mins, 19pts, 49fg%, 10reb, 4ast, 1stl, 1.5blk, 2tos, 110OR, 100DR, 23PER, 55ts%, 95ows, 90dws, 185ws, .190ws/48, 5.5bpm, 90vorp

Prime - 1100gm, 37mins, 21pts, 50fg%, 11reb, 4ast, 1.5stl, 1.5blk, 2.5tos, 110OR, 95DR, 24PER, 56ts%, 95ows, 80dws, 175ws, .210ws/48, 6.5bpm, 88.3vorp


Player 2 - 1000gm, 35mins, 22pts, 52fg%, 11reb, 3ast, 1.5stl, 2blk, 2.5tos, 115OR, 100DR, 26PER, 58ts%, 95ows, 80dws, 175ws, .240ws/48, 7bpm, 80vorp


PS:
Player 1 - 150gm, 37mins, 18pts, 50fg%, 10.5reb, 3.5ast, 1stl, 1.5blk, 2.5tos, 105OR, 100DR, 22PER, 54ts%, 7.5ows, 9.5dws, 17ws, .150ws/48, 5bpm, 10vorp

Prime - 120gm, 38mins, 20pts, 48fg%, 11reb, 3.5ast, 1.5stl, 1.5blk, 2.5tos, 105OR, 95DR, 22PER, 54ts%, 6.5ows, 8.5dws, 15ws, .160ws/48, 5.5bpm, 9vorp


Player 2 - 120gm, 34mins, 18pts, 50fg%, 11reb, 2.5ast, 1stl, 3blk, 2.5tos, 110OR, 95DR, 23PER, 57ts%, 7.5ows, 9.5dws, 17ws, .190ws/48, 6bpm, 8.5vorp



In this case, Player 1 has considerably more longevity in total games but the overall numbers and metrics are considerably less than Player 2. When looking at prime years though, there is only a slight amount more games for Player 1 than Player 2 and the comparison is closer but Player 2 is better.

More total games in this case did not mean that player was better. This utmost significant in player comparisons and resulting player rankings.

[rk2023]

A lot of people now regard longevity as how many years you log at minimum at a Strong MVP / All-NBA / All-Star threshold of impact. It's subjective, but I don't see many on here using counting stats or minutes / years played in order to drive their view on longevity.

[70sFan]

Nobody defines longevity as number of games played...

[Dutchball97]

I'd have to disagree with the other responses. When I praise Kareem's longevity I'm not just talking about 70-82 for example. For overall longevity I definitely think things like games and minutes played is the first thing to look at. Then for meaningful longevity I look at how much value they've accumulated over the years statistically.

Maybe I just didn't get the memo and I've been arguing a completely different longevity than everyone else but I can't say I've seen many people mean "length of prime" when talking about longevity.

That's also why it's important to specify the period of time you're referencing when using terms like peak, prime and longevity since clearly not everyone uses the same definition.

[dygaction]

Since we often compare ATGs' longevity, not role players like Danny Green vs. Robert Horry, I feel playoffs should be weight higher.
If a player failed to make or missed playoffs due to injury that season, the value of the regular season should be taken at a discounted value (again for ATG players). It's like when the war starts, a state-of-the-art tank locked in the garage is useless. Playoffs' games might be added to regular season games with a weighing factor (1 to 2 or 3?), as a deeper run is a great toll to the body and mind. For example, LeBron played 266 playoff games, that's more than 3 full season worth of high intensity games, or about all regular season games (or 2000 more minutes more than) Kawhi played in the past 7 years.

[70sFan]

I'd have to disagree with the other responses. When I praise Kareem's longevity I'm not just talking about 70-82 for example. For overall longevity I definitely think things like games and minutes played is the first thing to look at. Then for meaningful longevity I look at how much value they've accumulated over the years statistically.

Maybe I just didn't get the memo and I've been arguing a completely different longevity than everyone else but I can't say I've seen many people mean "length of prime" when talking about longevity.

That's also why it's important to specify the period of time you're referencing when using terms like peak, prime and longevity since clearly not everyone uses the same definition.

With Kareem, I think everybody includes all of his seasons from 1970-87 period for longevity, but I doubt anyone cares that he played full season in 1989 when he was washed up.

It's not about the length of prime, but we shouldn't give Kobe bonus for playing in 2014-16 period when he clearly didn't do anything meaningful in that period.

[Dutchball97]

I'd have to disagree with the other responses. When I praise Kareem's longevity I'm not just talking about 70-82 for example. For overall longevity I definitely think things like games and minutes played is the first thing to look at. Then for meaningful longevity I look at how much value they've accumulated over the years statistically.

Maybe I just didn't get the memo and I've been arguing a completely different longevity than everyone else but I can't say I've seen many people mean "length of prime" when talking about longevity.

That's also why it's important to specify the period of time you're referencing when using terms like peak, prime and longevity since clearly not everyone uses the same definition.

With Kareem, I think everybody includes all of his seasons from 1970-87 period for longevity, but I doubt anyone cares that he played full season in 1989 when he was washed up.

It's not about the length of prime, but we shouldn't give Kobe bonus for playing in 2014-16 period when he clearly didn't do anything meaningful in that period.

That's why I often reference WS for meaningful longevity. Sure it's not the most accurate etc etc but it goes back to the 40s and the general ranges where players are rated here are rarely completely off. It's also not as definite as determining whether a season counts for longevity or not with no inbetween. Kareem's 89 season might have barely increased his value but 2.9 extra WS is at least something compared to Kobe's -0.6 WS over his last 3 seasons.

[70sFan]

I'd have to disagree with the other responses. When I praise Kareem's longevity I'm not just talking about 70-82 for example. For overall longevity I definitely think things like games and minutes played is the first thing to look at. Then for meaningful longevity I look at how much value they've accumulated over the years statistically.

Maybe I just didn't get the memo and I've been arguing a completely different longevity than everyone else but I can't say I've seen many people mean "length of prime" when talking about longevity.

That's also why it's important to specify the period of time you're referencing when using terms like peak, prime and longevity since clearly not everyone uses the same definition.

With Kareem, I think everybody includes all of his seasons from 1970-87 period for longevity, but I doubt anyone cares that he played full season in 1989 when he was washed up.

It's not about the length of prime, but we shouldn't give Kobe bonus for playing in 2014-16 period when he clearly didn't do anything meaningful in that period.

That's why I often reference WS for meaningful longevity. Sure it's not the most accurate etc etc but it goes back to the 40s and the general ranges where players are rated here are rarely completely off. It's also not as definite as determining whether a season counts for longevity or not with no inbetween. Kareem's 89 season might have barely increased his value but 2.9 extra WS is at least something compared to Kobe's -0.6 WS over his last 3 seasons.

Yeah, as much as I dislike win shares it's a good starting point for looking for what I'd call a meaningful longevity.

[migya]

A lot of people now regard longevity as how many years you log at minimum at a Strong MVP / All-NBA / All-Star threshold of impact. It's subjective, but I don't see many on here using counting stats or minutes / years played in order to drive their view on longevity.

Many do. For example, I don't see Garnett as having great longevity. He was significant from 98 or 99 to 2012. That's not as much as a number of other alltime greats.

The total number of games does count as it shows how much the player actually played. Durant may have played for many years now but his missed alot and his total games isn't large. A player's value is in level of play for amount played. The same impact for 1100 games is better than for 1000 games.

[70sFan]

Many do. For example, I don't see Garnett as having great longevity. He was significant from 98 or 99 to 2012. That's not as much as a number of other alltime greats.

That's 15 seasons, how is that not great? How many top 20 players have 15 seasons you'd call "significant? I will use your own list of top 20 players you posted some time ago (without order):

Russell: didn't play 15 seasons
Wilt: didn't play 15 seasons
Kareem: better longevity
Jordan: played 15 seasons total, but two of them were in Washington (they weren't significant) and in 2 of them he missed over 60 RS games
James: better longevity
Duncan: better longevity
Hakeem: I guess 1985-99, so similar longevity
Shaq: 1993-06 with missing a lot of games, I wouldn't call that better
Magic: didn't play 15 seasons
Bird: didn't play 15 seasons

K. Malone: better longevity
West: didn't play 15 seasons
Robertson: didn't play 15 seasons
Robinson: didn't play 15 seasons
Kobe: you may argue similar with 1999-13
Baylor: didn't play 15 seasons
Moses: probably 1977-90, so less than 15 seasons but I guess we can call it similar
Barkley: played total of 16 seasons, don't see his rookie year and last year as relevant here, also missed a lot of time in 1994-99
Ewing: Ewing was done after 1999, so it gives him 14 seasons at most, I guess you can call it similar

Out of your (old) top 20, only Kareem, James, Duncan and Malone have clearly more "significant" seasons. You can add Wilt, Hakeem, Kobe and Moses for arguable, but that's still only 8 out of 19, less than a half. We're comparing him to the best players ever, so being among the best in this group certainly makes his longevity "great".

[Jaivl]

And that's without taking into account that most would go 97-13 (17 relevant years) on KG.

[Joao Saraiva]

For me longevity comes up with the sum of value the player adds per year. MJ with 15 years might have more longevity than Vince Carter, who played a ton but was not contributing nearly as much in a ton of the years.

I literally calculate that. Now, I don't adjust per era so I take it as guidance, not as something that is the truth and not up for changes. But if gives an idea.

[70sFan]

For me longevity comes up with the sum of value the player adds per year. MJ with 15 years might have more longevity than Vince Carter, who played a ton but was not contributing nearly as much in a ton of the years.

I literally calculate that. Now, I don't adjust per era so I take it as guidance, not as something that is the truth and not up for changes. But if gives an idea.

I think that's a fair way to look at this, but the author of the PO usually talks about longevity only among top 20-25 players ever. These players are all amazing in their primes, so the number of good seasons is certainly a decent way to start.

Vince by top 20-25 standards has very few "relevant" seasons, so it would go in line with what you say.

[Joao Saraiva]

For me longevity comes up with the sum of value the player adds per year. MJ with 15 years might have more longevity than Vince Carter, who played a ton but was not contributing nearly as much in a ton of the years.

I literally calculate that. Now, I don't adjust per era so I take it as guidance, not as something that is the truth and not up for changes. But if gives an idea.

I think that's a fair way to look at this, but the author of the PO usually talks about longevity only among top 20-25 players ever. These players are all amazing in their primes, so the number of good seasons is certainly a decent way to start.

Vince by top 20-25 standards has very few "relevant" seasons, so it would go in line with what you say.

Yeah welll I like to do this formulas as indicators. For example, Kobe played 19 seasons. And while young Kobe hads value (the foruma is cumulative) it doesn't necessarily mean his 19 years are enough to surpass MJ's longevity despite having less NBA years.

Same for Shaq for example. I haven't gona arround the formulas for a long time but...

[migya]

For me longevity comes up with the sum of value the player adds per year. MJ with 15 years might have more longevity than Vince Carter, who played a ton but was not contributing nearly as much in a ton of the years.

I literally calculate that. Now, I don't adjust per era so I take it as guidance, not as something that is the truth and not up for changes. But if gives an idea.

I think that's a fair way to look at this, but the author of the PO usually talks about longevity only among top 20-25 players ever. These players are all amazing in their primes, so the number of good seasons is certainly a decent way to start.

Vince by top 20-25 standards has very few "relevant" seasons, so it would go in line with what you say.

Garnett's "relevant seasons" are mostly lower level than most of the top 15-20 alltime players.

1997-98 -
RS - 20.4PER, 52.7ts%, .143ws/48, 4.2bpm, 5.0vorp

PS - 5gm, 15.5PER, 50ts%, -0.002ws/48, 0.9bpm, 0.1vorp

1998-99 -

RS - 22.4PER, 49.3ts%, .146ws/48, 5.3bpm, 3.3vorp

PS - 4gm, 24.1PER, 48.8ts%, .135ws/48, 8.1bpm, 0.4vorp


His prime really isn't 98 or 99.


Chris Paul had better numbers every season from his rookie season 2006 to 2022.

Durant had better numbers every season from his third season 2010 to 2022.

Barkley had better numbers every season from his second season 1986 to his second last season 1999.

Olajuwon had better numbers every season from his rookie season 1985 to 1999.

Nowitzki had better numbers every season from 2001 to 2014.

Stockton had better numbers every season from 1987 to his last season 2003.

Even Ginobili had better numbers every season from 04 to 14.


Garnett's metrics just aren't comparable to the alltime greats. He had 5 or 6 seasons at comparable level to the greats but that's it.

[Joao Saraiva]

For me longevity comes up with the sum of value the player adds per year. MJ with 15 years might have more longevity than Vince Carter, who played a ton but was not contributing nearly as much in a ton of the years.

I literally calculate that. Now, I don't adjust per era so I take it as guidance, not as something that is the truth and not up for changes. But if gives an idea.

I think that's a fair way to look at this, but the author of the PO usually talks about longevity only among top 20-25 players ever. These players are all amazing in their primes, so the number of good seasons is certainly a decent way to start.

Vince by top 20-25 standards has very few "relevant" seasons, so it would go in line with what you say.

Garnett's "relevant seasons" are mostly lower level than most of the top 15-20 alltime players.

1997-98 -
RS - 20.4PER, 52.7ts%, .143ws/48, 4.2bpm, 5.0vorp

PS - 5gm, 15.5PER, 50ts%, -0.002ws/48, 0.9bpm, 0.1vorp

1998-99 -

RS - 22.4PER, 49.3ts%, .146ws/48, 5.3bpm, 3.3vorp

PS - 4gm, 24.1PER, 48.8ts%, .135ws/48, 8.1bpm, 0.4vorp


His prime really isn't 98 or 99.


Chris Paul had better numbers every season from his rookie season 2006 to 2022.

Durant had better numbers every season from his third season 2010 to 2022.

Barkley had better numbers every season from his second season 1986 to his second last season 1999.

Olajuwon had better numbers every season from his rookie season 1985 to 1999.

Nowitzki had better numbers every season from 2001 to 2014.

Stockton had better numbers every season from 1987 to his last season 2003.

Even Ginobili had better numbers every season from 04 to 14.


Garnett's metrics just aren't comparable to the alltime greats. He had 5 or 6 seasons at comparable level to the greats but that's it.

Like I said the formula is not a dogma or something. I have to check where KG ends up with it, but I remember the formula not liking KG or Kobe as much as I do. Nothing wrong with that, it's just numbers after all that need some context. Also the formulas look less favorable to determine defensive impact, even tough they're there I trust the numbers a lot less.

[Owly]

I think longevity reasonably enough is used, and perhaps primarily used as how long you played. Some will mean in the first instance, minutes, others years.

In use in player ranking type exercises it might sometimes be used as a shorthand for "longevity of quality", which then leads into even more complex notions like value added, championship probability added etc where specific yearly values above some baseline are added to accumulate positive value through combining a high level of play and sustaining that level.

As ever clarity of intent is key. My preferences are implied above and I think are clear, but if people try to be clear (however they choose), then there's a better chance of better dialogue and it minimizes the chance talking past one another.

I will say WS tilts too pro-longevity (most egregiously notable at seasons of low value) due to too low a baseline for this purpose. Assuming for the sake of argument it was otherwise a perfect measure of player value, the '93 Mavs come out a net positive of +4.1 WS. I would say the average player (or more precisely, the average minute played) on that team adds no value to my all time rankings, no matter how deep I'm going.

[70sFan]

For me longevity comes up with the sum of value the player adds per year. MJ with 15 years might have more longevity than Vince Carter, who played a ton but was not contributing nearly as much in a ton of the years.

I literally calculate that. Now, I don't adjust per era so I take it as guidance, not as something that is the truth and not up for changes. But if gives an idea.

I think that's a fair way to look at this, but the author of the PO usually talks about longevity only among top 20-25 players ever. These players are all amazing in their primes, so the number of good seasons is certainly a decent way to start.

Vince by top 20-25 standards has very few "relevant" seasons, so it would go in line with what you say.

Garnett's "relevant seasons" are mostly lower level than most of the top 15-20 alltime players.

1997-98 -
RS - 20.4PER, 52.7ts%, .143ws/48, 4.2bpm, 5.0vorp

PS - 5gm, 15.5PER, 50ts%, -0.002ws/48, 0.9bpm, 0.1vorp

1998-99 -

RS - 22.4PER, 49.3ts%, .146ws/48, 5.3bpm, 3.3vorp

PS - 4gm, 24.1PER, 48.8ts%, .135ws/48, 8.1bpm, 0.4vorp


His prime really isn't 98 or 99.


Chris Paul had better numbers every season from his rookie season 2006 to 2022.

Durant had better numbers every season from his third season 2010 to 2022.

Barkley had better numbers every season from his second season 1986 to his second last season 1999.

Olajuwon had better numbers every season from his rookie season 1985 to 1999.

Nowitzki had better numbers every season from 2001 to 2014.

Stockton had better numbers every season from 1987 to his last season 2003.

Even Ginobili had better numbers every season from 04 to 14.


Garnett's metrics just aren't comparable to the alltime greats. He had 5 or 6 seasons at comparable level to the greats but that's it.

I tried many times, I'm done now.