What impact metrics show MJ as a GOAT candidate?

[ShaqAttac]

Part 1: WOWY and WOWYR Discussion

Again we go...again

'with' the player and compare them to games when the team played 'without' the player.

When we want to find a player's adjusted plus minus, we take the raw on/off of them and all their teammates and opponents and apply a regression.
When we want to find a players' adjusted WOWYR, we take the raw WOWY data for them and all their teammates and opponents and apply a regr

We don't only look at the "on" of on/off best"
Again, ... this just isn't true. Have you actually checked the links I sent?

From the official adjusted WOWY source:
Prime Jordan adjusted WOWY rank: 4th all time
Prime Russell adjusted WOWY rank: 30th all time
Prime Kareem adjusted WOWY rank: 29th all time
Prime LeBron adjusted WOWY rank: 8th all time

[/quote][/quote]
NGL this enterin seppeku territory

bro explained why he think russell loook better. dont be a clock now

[OhayoKD]

Part 1: WOWY and WOWYR Discussion

Again we go...again

'with' the player and compare them to games when the team played 'without' the player.

When we want to find a player's adjusted plus minus, we take the raw on/off of them and all their teammates and opponents and apply a regression.
When we want to find a players' adjusted WOWYR, we take the raw WOWY data for them and all their teammates and opponents and apply a regr

We don't only look at the "on" of on/off best"
Again, ... this just isn't true. Have you actually checked the links I sent?

From the official adjusted WOWY source:
Prime Jordan adjusted WOWY rank: 4th all time
Prime Russell adjusted WOWY rank: 30th all time
Prime Kareem adjusted WOWY rank: 29th all time
Prime LeBron adjusted WOWY rank: 8th all time

NGL this enterin seppeku territory

bro explained why he think russell loook better. dont be a clock now

Could you not? These debates get heated enough without the suicide references. Also...formatting

(To dray) I started a response but I'll need some time to finish due to some projects going. That being said, I may as well mention the WOWYR stuff for Kareem might also be higher since Ben messed up on his WOWY stuff(and 70's corrections actually flip things between Jordan and Jabbar):

Okay, but do remember that there's that soft middle to Kareem's career where the Lakers were going nowhere, and know that at least with Ben's prime WOWYR metric Kareem doesn't look as good as Jordan. Now, much as I respect Ben, I wouldn't treat a number like this as a definitive answer, however I'd be real careful about running with an in-head-WOWY if it tells you something counter to what the data has told him.

And of course, if you've done something more quantitative, or you see a specific issue with Kareem's data along these lines, please elaborate.

I am not interested in another MJ vs LBJ discussion, but I will bite here.

I think it's very important to be careful to call anything "going nowhere":

1. Lakers finished with the best RS record in 1977, Jordan neve did that before 1991.

2. Lakers finished with +2.95 SRS in 1979, which is better than Bulls in any year in 1987-90 period outside of 1988. It happened in the smaller, more balanced league as well.

3. Lakers lost to two future champions in the playoffs during 1977-79 period. The other time, they lost to future finalists in a 3 games series. That's the same level of playoff success as 1987-89 Bulls.

If you want to say that they did nothing during that period, then I'm afraid you should say the same for Jordan's whole career before 1990.

About WOWY - Jordan's biggest samples don't show him as the better one than Kareem (from Ben's database):

1986 Jordan: +2.0 SRS change, 1.2 WOWY score
1995 Jordan +2.7 SRS change, 1.9 WOWY score

1975 Kareem: +7.1 SRS change, +3.6 WOWY score

I'm afraid Ben's database has an error with 1978 sample, as it shows as clear negative for Kareem, despite all the calculations I made and his own words in Kareem profile:

At the beginning of the ’78 season, Kareem cold-cocked Bucks center Kent Benson and missed substantial time with another broken hand. However, it’s hard to infer much from the injury since LA fired off two trades around that period.10 With Jabbar — and ignoring all the other lineup activity — the Lakers played like a 53-win team (4.1 SRS) in ’78. With a similar roster in ’79 (minus Charlie Scott), LA ticked along at a 50-win clip when healthy (3.1 SRS). Below, I’ve plotted the ’78 team’s performance in 21 games without Kareem, in which the Lakers played at a 36-win pace (-1.7 SRS) after a major offensive drop-off.

Which shows a +5.8 SRS change again. The biggest samples we have show Kareem having a clear advantage. We can also look at the more nuanced samples, when a player even joins or leaves his team:

1984 Bulls without MJ: -4.7 SRS, 27 wins
1985 Bulls with MJ: -0.5 SRS, 38 wins
Change: +4.2 SRS and +11 wins

1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

1995 Bulls without MJ: +3.8 SRS, 52 wins pace
1996 Bulls with MJ: +11.8 SRS, 72 wins
Change: +8 SRS, 20 wins

I wouldn't include 1998-1999, because the whole team changed, including a coach.

1969 Bucks without Kareem: -5.1 SRS, 27 wins
1970 Bucks with Kareem: +4.3 SRS, 56 wins
Change: +9.4 SRS, 29 wins

1975 Bucks with Kareem: +2.6 SRS, 49 wins pace
1976 Bucks without Kareem: -1.6 SRS, 38 wins
Change: +4.2 SRS, 11 wins

1975 Lakers without Kareem: -3.9 SRS, 30 wins
1976 Lakers with Kareem: +0.2 SRS, 40 wins
Change: +4.1 SRS, 10 wins

The difference is that Kareem left Bucks in a trade, which means that Lakers gave a lot of value to Bucks. Jordan samples are clean, as Jordan didn't go to the Bulls in exchange.

I don't know, I don't see the case for MJ > Kareem in terms of WOWY.

[DraymondGold]

Part 1: WOWY and WOWYR Discussion

NGL this enterin seppeku territory

bro explained why he think russell loook better. dont be a clock now

Could you not? These debates get heated enough without the suicide references. Also...formatting

(To dray) I started a response but I'll need some time to finish due to some projects going. That being said, I may as well mention the WOWYR stuff for Kareem might also be higher since Ben messed up on his WOWY stuff(and 70's corrections actually flip things between Jordan and Jabbar):

Okay, but do remember that there's that soft middle to Kareem's career where the Lakers were going nowhere, and know that at least with Ben's prime WOWYR metric Kareem doesn't look as good as Jordan. Now, much as I respect Ben, I wouldn't treat a number like this as a definitive answer, however I'd be real careful about running with an in-head-WOWY if it tells you something counter to what the data has told him.

And of course, if you've done something more quantitative, or you see a specific issue with Kareem's data along these lines, please elaborate.

I am not interested in another MJ vs LBJ discussion, but I will bite here.

I think it's very important to be careful to call anything "going nowhere":

1. Lakers finished with the best RS record in 1977, Jordan neve did that before 1991.

2. Lakers finished with +2.95 SRS in 1979, which is better than Bulls in any year in 1987-90 period outside of 1988. It happened in the smaller, more balanced league as well.

3. Lakers lost to two future champions in the playoffs during 1977-79 period. The other time, they lost to future finalists in a 3 games series. That's the same level of playoff success as 1987-89 Bulls.

If you want to say that they did nothing during that period, then I'm afraid you should say the same for Jordan's whole career before 1990.

About WOWY - Jordan's biggest samples don't show him as the better one than Kareem (from Ben's database):

1986 Jordan: +2.0 SRS change, 1.2 WOWY score
1995 Jordan +2.7 SRS change, 1.9 WOWY score

1975 Kareem: +7.1 SRS change, +3.6 WOWY score

I'm afraid Ben's database has an error with 1978 sample, as it shows as clear negative for Kareem, despite all the calculations I made and his own words in Kareem profile:

At the beginning of the ’78 season, Kareem cold-cocked Bucks center Kent Benson and missed substantial time with another broken hand. However, it’s hard to infer much from the injury since LA fired off two trades around that period.10 With Jabbar — and ignoring all the other lineup activity — the Lakers played like a 53-win team (4.1 SRS) in ’78. With a similar roster in ’79 (minus Charlie Scott), LA ticked along at a 50-win clip when healthy (3.1 SRS). Below, I’ve plotted the ’78 team’s performance in 21 games without Kareem, in which the Lakers played at a 36-win pace (-1.7 SRS) after a major offensive drop-off.

Which shows a +5.8 SRS change again. The biggest samples we have show Kareem having a clear advantage. We can also look at the more nuanced samples, when a player even joins or leaves his team:

1984 Bulls without MJ: -4.7 SRS, 27 wins
1985 Bulls with MJ: -0.5 SRS, 38 wins
Change: +4.2 SRS and +11 wins

1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

1995 Bulls without MJ: +3.8 SRS, 52 wins pace
1996 Bulls with MJ: +11.8 SRS, 72 wins
Change: +8 SRS, 20 wins

I wouldn't include 1998-1999, because the whole team changed, including a coach.

1969 Bucks without Kareem: -5.1 SRS, 27 wins
1970 Bucks with Kareem: +4.3 SRS, 56 wins
Change: +9.4 SRS, 29 wins

1975 Bucks with Kareem: +2.6 SRS, 49 wins pace
1976 Bucks without Kareem: -1.6 SRS, 38 wins
Change: +4.2 SRS, 11 wins

1975 Lakers without Kareem: -3.9 SRS, 30 wins
1976 Lakers with Kareem: +0.2 SRS, 40 wins
Change: +4.1 SRS, 10 wins

The difference is that Kareem left Bucks in a trade, which means that Lakers gave a lot of value to Bucks. Jordan samples are clean, as Jordan didn't go to the Bulls in exchange.

I don't know, I don't see the case for MJ > Kareem in terms of WOWY.

Take your time! No rush at all

The possible Kareem mistake is interesting -- seems like we could just ask Ben to check? If he fixes it, that would allow him to update the database to show the true value, or if he finds there's no mistake, that would give us clearer reasoning to where the disparity comes from.

As to the wider point, I have no problem taking Kareem over Jordan in individual raw WOWY samples. There's no complete consensus across all stats as to who is the exact best player. The point I was trying to make is that there are other WOWY-based and WOWY-like stats that favor Jordan over Kareem et al. Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years. If someone wanted to make a WOWY-based argument for Jordan as the GOAT over Russell/Kareem/LeBron, they could.... even if you prefer the WOWY data that goes the other way.

[Djoker]

NGL this enterin seppeku territory

bro explained why he think russell loook better. dont be a clock now

Could you not? These debates get heated enough without the suicide references. Also...formatting

(To dray) I started a response but I'll need some time to finish due to some projects going. That being said, I may as well mention the WOWYR stuff for Kareem might also be higher since Ben messed up on his WOWY stuff(and 70's corrections actually flip things between Jordan and Jabbar):

I am not interested in another MJ vs LBJ discussion, but I will bite here.

I think it's very important to be careful to call anything "going nowhere":

1. Lakers finished with the best RS record in 1977, Jordan neve did that before 1991.

2. Lakers finished with +2.95 SRS in 1979, which is better than Bulls in any year in 1987-90 period outside of 1988. It happened in the smaller, more balanced league as well.

3. Lakers lost to two future champions in the playoffs during 1977-79 period. The other time, they lost to future finalists in a 3 games series. That's the same level of playoff success as 1987-89 Bulls.

If you want to say that they did nothing during that period, then I'm afraid you should say the same for Jordan's whole career before 1990.

About WOWY - Jordan's biggest samples don't show him as the better one than Kareem (from Ben's database):

1986 Jordan: +2.0 SRS change, 1.2 WOWY score
1995 Jordan +2.7 SRS change, 1.9 WOWY score

1975 Kareem: +7.1 SRS change, +3.6 WOWY score

I'm afraid Ben's database has an error with 1978 sample, as it shows as clear negative for Kareem, despite all the calculations I made and his own words in Kareem profile:



Which shows a +5.8 SRS change again. The biggest samples we have show Kareem having a clear advantage. We can also look at the more nuanced samples, when a player even joins or leaves his team:

1984 Bulls without MJ: -4.7 SRS, 27 wins
1985 Bulls with MJ: -0.5 SRS, 38 wins
Change: +4.2 SRS and +11 wins

1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

1995 Bulls without MJ: +3.8 SRS, 52 wins pace
1996 Bulls with MJ: +11.8 SRS, 72 wins
Change: +8 SRS, 20 wins

I wouldn't include 1998-1999, because the whole team changed, including a coach.

1969 Bucks without Kareem: -5.1 SRS, 27 wins
1970 Bucks with Kareem: +4.3 SRS, 56 wins
Change: +9.4 SRS, 29 wins

1975 Bucks with Kareem: +2.6 SRS, 49 wins pace
1976 Bucks without Kareem: -1.6 SRS, 38 wins
Change: +4.2 SRS, 11 wins

1975 Lakers without Kareem: -3.9 SRS, 30 wins
1976 Lakers with Kareem: +0.2 SRS, 40 wins
Change: +4.1 SRS, 10 wins

The difference is that Kareem left Bucks in a trade, which means that Lakers gave a lot of value to Bucks. Jordan samples are clean, as Jordan didn't go to the Bulls in exchange.

I don't know, I don't see the case for MJ > Kareem in terms of WOWY.

Take your time! No rush at all

The possible Kareem mistake is interesting -- seems like we could just ask Ben to check? If he fixes it, that would allow him to update the database to show the true value, or if he finds there's no mistake, that would give us clearer reasoning to where the disparity comes from.

As to the wider point, I have no problem taking Kareem over Jordan in individual raw WOWY samples. There's no complete consensus across all stats as to who is the exact best player. The point I was trying to make is that there are other WOWY-based and WOWY-like stats that favor Jordan over Kareem et al. Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years. If someone wanted to make a WOWY-based argument for Jordan as the GOAT over Russell/Kareem/LeBron, they could.... even if you prefer the WOWY data that goes the other way.

Of course and there is enough roster continuity to include neighboring seasons in some cases. For instance the 1994 Bulls season w/o MJ is often compared to the 1993 season with MJ which is logical because they are consecutive seasons but the 1993 season is the Bulls worst season of all of their title runs. The lack of motivation during that regular season is well documented whereas the 1994 Bulls had a chip on their shoulder to prove they were good without Jordan. There is reasonable roster continuity to pool say 1991/1992/1993 together and compare vs. 1994. That may give a better picture of MJ's WOWY during that period.

[DraymondGold]

Could you not? These debates get heated enough without the suicide references. Also...formatting

(To dray) I started a response but I'll need some time to finish due to some projects going. That being said, I may as well mention the WOWYR stuff for Kareem might also be higher since Ben messed up on his WOWY stuff(and 70's corrections actually flip things between Jordan and Jabbar):

Take your time! No rush at all

The possible Kareem mistake is interesting -- seems like we could just ask Ben to check? If he fixes it, that would allow him to update the database to show the true value, or if he finds there's no mistake, that would give us clearer reasoning to where the disparity comes from.

As to the wider point, I have no problem taking Kareem over Jordan in individual raw WOWY samples. There's no complete consensus across all stats as to who is the exact best player. The point I was trying to make is that there are other WOWY-based and WOWY-like stats that favor Jordan over Kareem et al. Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years. If someone wanted to make a WOWY-based argument for Jordan as the GOAT over Russell/Kareem/LeBron, they could.... even if you prefer the WOWY data that goes the other way.

Of course and there is enough roster continuity to include neighboring seasons in some cases. For instance the 1994 Bulls season w/o MJ is often compared to the 1993 season with MJ which is logical because they are consecutive seasons but the 1993 season is the Bulls worst season of all of their title runs. The lack of motivation during that regular season is well documented whereas the 1994 Bulls had a chip on their shoulder to prove they were good without Jordan. There is reasonable roster continuity to pool say 1991/1992/1993 together and compare vs. 1994. That may give a better picture of MJ's WOWY during that period.

Yeah, great point about the limitations of just looking at 93 vs 94 in isolation!

That's one of the concerns I have with single samples of raw WOWY -- your sample is more limited, and there's no adjustment for any context... e.g. if your team was coasting one year, but weren't the year prior.

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

But! What if we add broader context?
1992 Bulls: +10.07 SRS, 66 Wins
1993 Bulls Playoffs : +10.08 SRS
So when they Bulls aren't coasting, they're pretty consistently a 10+ SRS team. In fact they look significantly better still if you look at the 92 playoffs or 91 team (which again is why Jordan looks better in 10-year WOWY and WOWYR samples).

Change from 94 Bulls without Jordan to non-coasting 92/93-Playoffs Bulls with Jordan: +7.2 SRS
Change from 94 Bulls without Jordan to half-coasting 92/93 average Bulls with Jordan: +4.8 SRS

So in both cases, whether we take the semi-coasting average of 92 or 93 or just look at when the Bulls weren't coasting, Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY), 1970 sample doesn't account for SRS inflation with the rapid expansion of the league (which would overrate 1970 Kareem), while Jordan's sample doesn't account for the fact that you can get diminishing returns at the upper end of SRS (which would cause us to underrate Jordan if we just looked at the 92/93-playoff Bulls' SRS).

[OhayoKD]

2015 Cavs with no Kyrie and Love? Seriously?

They were a +10 PSRS without kyrie and love and swept a 60 win team to make the finals(55 win srs iirc). Don't think the "seriously?" is warranted for that specific comparison.

[falcolombardi]

Take your time! No rush at all

The possible Kareem mistake is interesting -- seems like we could just ask Ben to check? If he fixes it, that would allow him to update the database to show the true value, or if he finds there's no mistake, that would give us clearer reasoning to where the disparity comes from.

As to the wider point, I have no problem taking Kareem over Jordan in individual raw WOWY samples. There's no complete consensus across all stats as to who is the exact best player. The point I was trying to make is that there are other WOWY-based and WOWY-like stats that favor Jordan over Kareem et al. Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years. If someone wanted to make a WOWY-based argument for Jordan as the GOAT over Russell/Kareem/LeBron, they could.... even if you prefer the WOWY data that goes the other way.

Of course and there is enough roster continuity to include neighboring seasons in some cases. For instance the 1994 Bulls season w/o MJ is often compared to the 1993 season with MJ which is logical because they are consecutive seasons but the 1993 season is the Bulls worst season of all of their title runs. The lack of motivation during that regular season is well documented whereas the 1994 Bulls had a chip on their shoulder to prove they were good without Jordan. There is reasonable roster continuity to pool say 1991/1992/1993 together and compare vs. 1994. That may give a better picture of MJ's WOWY during that period.

Yeah, great point about the limitations of just looking at 93 vs 94 in isolation!

That's one of the concerns I have with single samples of raw WOWY -- your sample is more limited, and there's no adjustment for any context... e.g. if your team was coasting one year, but weren't the year prior.

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

But! What if we add broader context?
1992 Bulls: +10.07 SRS, 66 Wins
1993 Bulls Playoffs : +10.08 SRS
So when they Bulls aren't coasting, they're pretty consistently a 10+ SRS team. In fact they look significantly better still if you look at the 92 playoffs or 91 team (which again is why Jordan looks better in 10-year WOWY and WOWYR samples).

Change from 94 Bulls without Jordan to non-coasting 92/93-Playoffs Bulls with Jordan: +7.2 SRS
Change from 94 Bulls without Jordan to half-coasting 92/93 average Bulls with Jordan: +4.8 SRS

So in both cases, whether we take the semi-coasting average of 92 or 93 or just look at when the Bulls weren't coasting, Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY), 1970 sample doesn't account for SRS inflation with the rapid expansion of the league (which would overrate 1970 Kareem), while Jordan's sample doesn't account for the fact that you can get diminishing returns at the upper end of SRS (which would cause us to underrate Jordan if we just looked at the 92/93-playoff Bulls' SRS).

It has been a while taking a break off real gm for career stuff and first thingh i hop into is a lebron vs jordan debate, nice

Being serious tho, this is a interesting debate but i have some thinghs to question here

Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years

This is somethingh i am gonna talk about quickly, but i am not a big fan of wowyr for similar reasons to those exposed by other posters in older threads.

The small tiny samples across multiple years are incredibly noisy when aggregated together so the "sample" part of sample size is fairly questionable and for most players the "size" part is still way too small compared to single season worth of wowy (read somethingh like 93 vs 94 bulls which gives me way better and bigger data to evaluate 93 jordan to the point i see wowyr as a bit worthless in comparision)

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

How does this account for pippen amd grant missing 10 games each in 94? The difference is also even smaller if you look at post season srs in 94 vs 93 or 91-93

Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY)

The lakers sold out to get kareem so comparing their lift as if they had addes him outright is a fair bit misleading here.

Also as you notice yourself, rookie kareem has a -higher- wowy than peak jordan,why not compare kareem actual best wowy signal to jordan best rather than putting jordan best signal against one that is not kareem best?

rookie kareem beating out peak jordan wowy is very relevant in a wowy discussion. The actual player here with the best wowy year is kareem

Also remember than a year like 76 kareem was -not- straight up addition of kareem being added without any loss (lakers sold out their team to get kareem) which muds the 76 as a wowy sample

As he is not actually being added to the 75 lakers straight up, but to a much more weak and dismantled team than the 75 lakers

[ShaqAttac]

Of course and there is enough roster continuity to include neighboring seasons in some cases. For instance the 1994 Bulls season w/o MJ is often compared to the 1993 season with MJ which is logical because they are consecutive seasons but the 1993 season is the Bulls worst season of all of their title runs. The lack of motivation during that regular season is well documented whereas the 1994 Bulls had a chip on their shoulder to prove they were good without Jordan. There is reasonable roster continuity to pool say 1991/1992/1993 together and compare vs. 1994. That may give a better picture of MJ's WOWY during that period.

Yeah, great point about the limitations of just looking at 93 vs 94 in isolation!

That's one of the concerns I have with single samples of raw WOWY -- your sample is more limited, and there's no adjustment for any context... e.g. if your team was coasting one year, but weren't the year prior.

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

But! What if we add broader context?
1992 Bulls: +10.07 SRS, 66 Wins
1993 Bulls Playoffs : +10.08 SRS
So when they Bulls aren't coasting, they're pretty consistently a 10+ SRS team. In fact they look significantly better still if you look at the 92 playoffs or 91 team (which again is why Jordan looks better in 10-year WOWY and WOWYR samples).

Change from 94 Bulls without Jordan to non-coasting 92/93-Playoffs Bulls with Jordan: +7.2 SRS
Change from 94 Bulls without Jordan to half-coasting 92/93 average Bulls with Jordan: +4.8 SRS

So in both cases, whether we take the semi-coasting average of 92 or 93 or just look at when the Bulls weren't coasting, Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY), 1970 sample doesn't account for SRS inflation with the rapid expansion of the league (which would overrate 1970 Kareem), while Jordan's sample doesn't account for the fact that you can get diminishing returns at the upper end of SRS (which would cause us to underrate Jordan if we just looked at the 92/93-playoff Bulls' SRS).

It has been a while taking a break off real gm for career stuff and first thingh i hop into is a lebron vs jordan debate, nice

Being serious tho, this is a interesting debate but i have some thinghs to question here

Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years

This is somethingh i am gonna talk about quickly, but i am not a big fan of wowyr for similar reasons to those exposed by other posters in older threads.

The small tiny samples across multiple years are incredibly noisy when aggregated together so the "sample" part of sample size is fairly questionable and for most players the "size" part is still way too small compared to single season worth of wowy (read somethingh like 93 vs 94 bulls which gives me way better and bigger data to evaluate 93 jordan to the point i see wowyr as a bit worthless in comparision)

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

How does this account for pippen amd grant missing 10 games each in 94? The difference is also even smaller if you look at post season srs in 94 vs 93 or 91-93

Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY)

The lakers sold out to get kareem so comparing their lift as if they had addes him outright is a fair bit misleading here.

Also as you notice yourself, rookie kareem has a -higher- wowy than peak jordan,why not compare kareem actual best wowy signal to jordan best rather than putting jordan best signal against one that is not kareem best?

rookie kareem beating out peak jordan wowy is very relevant in a wowy discussion. The actual player here with the best wowy year is kareem

Also remember than a year like 76 kareem was -not- straight up addition of kareem being added without any loss (lakers sold out their team to get kareem) which muds the 76 as a wowy sample

As he is not actually being added to the 75 lakers straight up, but to a much more weak and dismantled team than the 75 lakers

wait rook cap beats mj?

aslso it looks like gold gassed mjs number and he still took the l. thats p brutal

[Djoker]

Take your time! No rush at all

The possible Kareem mistake is interesting -- seems like we could just ask Ben to check? If he fixes it, that would allow him to update the database to show the true value, or if he finds there's no mistake, that would give us clearer reasoning to where the disparity comes from.

As to the wider point, I have no problem taking Kareem over Jordan in individual raw WOWY samples. There's no complete consensus across all stats as to who is the exact best player. The point I was trying to make is that there are other WOWY-based and WOWY-like stats that favor Jordan over Kareem et al. Jordan's adjusted WOWYR and GPM both put him over Kareem for their best 10 years. If someone wanted to make a WOWY-based argument for Jordan as the GOAT over Russell/Kareem/LeBron, they could.... even if you prefer the WOWY data that goes the other way.

Of course and there is enough roster continuity to include neighboring seasons in some cases. For instance the 1994 Bulls season w/o MJ is often compared to the 1993 season with MJ which is logical because they are consecutive seasons but the 1993 season is the Bulls worst season of all of their title runs. The lack of motivation during that regular season is well documented whereas the 1994 Bulls had a chip on their shoulder to prove they were good without Jordan. There is reasonable roster continuity to pool say 1991/1992/1993 together and compare vs. 1994. That may give a better picture of MJ's WOWY during that period.

Yeah, great point about the limitations of just looking at 93 vs 94 in isolation!

That's one of the concerns I have with single samples of raw WOWY -- your sample is more limited, and there's no adjustment for any context... e.g. if your team was coasting one year, but weren't the year prior.

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

But! What if we add broader context?
1992 Bulls: +10.07 SRS, 66 Wins
1993 Bulls Playoffs : +10.08 SRS
So when they Bulls aren't coasting, they're pretty consistently a 10+ SRS team. In fact they look significantly better still if you look at the 92 playoffs or 91 team (which again is why Jordan looks better in 10-year WOWY and WOWYR samples).

Change from 94 Bulls without Jordan to non-coasting 92/93-Playoffs Bulls with Jordan: +7.2 SRS
Change from 94 Bulls without Jordan to half-coasting 92/93 average Bulls with Jordan: +4.8 SRS

So in both cases, whether we take the semi-coasting average of 92 or 93 or just look at when the Bulls weren't coasting, Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY), 1970 sample doesn't account for SRS inflation with the rapid expansion of the league (which would overrate 1970 Kareem), while Jordan's sample doesn't account for the fact that you can get diminishing returns at the upper end of SRS (which would cause us to underrate Jordan if we just looked at the 92/93-playoff Bulls' SRS).

Good post and worth noting that a +7 SRS lift to an already good team is pretty insane. A few guys in this thread are pulling hard for rookie Kareem who improved his team by +9 SRS but that was a putrid team before he joined them. Improving a -5 SRS to a +4 SRS team is IMO less impressive than improving a +3 SRS team to a +10 SRS team even though the first lift is greater in an absolute sense.

And mind you I think it's clear that Kareem is one of the impactful players ever at his peak. But so is Jordan probably even more so.

[falcolombardi]

Of course and there is enough roster continuity to include neighboring seasons in some cases. For instance the 1994 Bulls season w/o MJ is often compared to the 1993 season with MJ which is logical because they are consecutive seasons but the 1993 season is the Bulls worst season of all of their title runs. The lack of motivation during that regular season is well documented whereas the 1994 Bulls had a chip on their shoulder to prove they were good without Jordan. There is reasonable roster continuity to pool say 1991/1992/1993 together and compare vs. 1994. That may give a better picture of MJ's WOWY during that period.

Yeah, great point about the limitations of just looking at 93 vs 94 in isolation!

That's one of the concerns I have with single samples of raw WOWY -- your sample is more limited, and there's no adjustment for any context... e.g. if your team was coasting one year, but weren't the year prior.

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

But! What if we add broader context?
1992 Bulls: +10.07 SRS, 66 Wins
1993 Bulls Playoffs : +10.08 SRS
[b]So when they Bulls aren't coasting, they're pretty consistently a 10+ SRS team[/b]. In fact they look significantly better still if you look at the 92 playoffs or 91 team (which again is why Jordan looks better in 10-year WOWY and WOWYR samples).

Change from 94 Bulls without Jordan to non-coasting 92/93-Playoffs Bulls with Jordan: +7.2 SRS
Change from 94 Bulls without Jordan to half-coasting 92/93 average Bulls with Jordan: +4.8 SRS

So in both cases, whether we take the semi-coasting average of 92 or 93 or just look at when the Bulls weren't coasting, Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY), 1970 sample doesn't account for SRS inflation with the rapid expansion of the league (which would overrate 1970 Kareem), while Jordan's sample doesn't account for the fact that you can get diminishing returns at the upper end of SRS (which would cause us to underrate Jordan if we just looked at the 92/93-playoff Bulls' SRS).

Good post and worth noting that a +7 SRS lift to an already good team is pretty insane. A few guys in this thread are pulling hard for rookie Kareem who improved his team by +9 SRS but that was a putrid team before he joined them. Improving a -5 SRS to a +4 SRS team is IMO less impressive than improving a +3 SRS team to a +10 SRS team even though the first lift is greater in an absolute sense.

And mind you I think it's clear that Kareem is one of the impactful players ever at his peak. But so is Jordan probably even more so.

So some thinghs here about that +7 number

First that is really weird to use the 93 bulls playoffs improvement as their "real level" cause they improved in the playoffs after coasting. Then not use the 94 bulls playoffs (where they also improved a lot) as their "real level" too

The 94 bulls actually had a +8.9 postseason srs which is almost the same as their 93 seasom +10 post season srs ( +1 difference)

The 94 bulls also missed 20 combined games from their two stars and played a +4.7 srs when healthy in the regular season (+1.5 difference with the 93 bulls with jordan) and in a very generous best case scenario a (+5.3 difference with even the 92 bulls regular season )

If i average the 94 bulls (+4.7 at full strenght in regular season and +8.9 in playoffs) vs the 92 reg season + 93 playoffs combination draymomd used (and please notice i am already picking and choosing the parts that help jordan more) the gap is only 5 points

That is not goat level.

Even by you guys own approach as it is below other all time greats lift in either absolute terms or in "ceiling raising" situations


I also dont get the "improving a good team is harder" part in relation to kareem, who led a goat level team in the 71 bucks so he was not exactly lacking in "ceiling raising" either compared to jordan while also having better "floor raising" lift as evidenced by their 60~ win pace without oscar in 1972

Or 08/09 garnett who had a similar lift from +3.4 to +9 and he is not even among the goat candidatws short list yet matches jordan here

i could also bring up other cases of lift like 2015 lebron cavs +10 postseason srs with a lot less talent and that the 91 bulls (kirye and love hurt) which strikes me as a even more extreme example of "ceiling raising" considering the floor it came off.

[Djoker]

Yeah, great point about the limitations of just looking at 93 vs 94 in isolation!

That's one of the concerns I have with single samples of raw WOWY -- your sample is more limited, and there's no adjustment for any context... e.g. if your team was coasting one year, but weren't the year prior.

What if we just compare 93 vs 94 Bulls? Per OhayoKD/70sFan:
1993 Bulls with MJ: +6.2 SRS, 57 wins
1994 Bulls without MJ: +2.9 SRS, 55 wins
Change: +3.3 SRS, +2 wins

But! What if we add broader context?
1992 Bulls: +10.07 SRS, 66 Wins
1993 Bulls Playoffs : +10.08 SRS
[b]So when they Bulls aren't coasting, they're pretty consistently a 10+ SRS team[/b]. In fact they look significantly better still if you look at the 92 playoffs or 91 team (which again is why Jordan looks better in 10-year WOWY and WOWYR samples).

Change from 94 Bulls without Jordan to non-coasting 92/93-Playoffs Bulls with Jordan: +7.2 SRS
Change from 94 Bulls without Jordan to half-coasting 92/93 average Bulls with Jordan: +4.8 SRS

So in both cases, whether we take the semi-coasting average of 92 or 93 or just look at when the Bulls weren't coasting, Jordan gets a higher individual raw WOWY sample than 75–76 Kareem (though they're lower than 1970 Kareem). Of course, the 1975-76 sample involved other moves which would change the value (another limit of looking at unadjusted WOWY), 1970 sample doesn't account for SRS inflation with the rapid expansion of the league (which would overrate 1970 Kareem), while Jordan's sample doesn't account for the fact that you can get diminishing returns at the upper end of SRS (which would cause us to underrate Jordan if we just looked at the 92/93-playoff Bulls' SRS).

Good post and worth noting that a +7 SRS lift to an already good team is pretty insane. A few guys in this thread are pulling hard for rookie Kareem who improved his team by +9 SRS but that was a putrid team before he joined them. Improving a -5 SRS to a +4 SRS team is IMO less impressive than improving a +3 SRS team to a +10 SRS team even though the first lift is greater in an absolute sense.

And mind you I think it's clear that Kareem is one of the impactful players ever at his peak. But so is Jordan probably even more so.

So some thinghs here about that +7 number

First that is really weird to use the 93 bulls playoffs improvement as their "real level" cause they improved in the playoffs after coasting. Then not use the 94 bulls playoffs (where they also improved a lot) as their "real level" too

The 94 bulls actually had a +8.9 postseason srs which is almost the same as their 93 seasom +10 post season srs ( +1 difference)

The 94 bulls also missed 20 combined games from their two stars and played a +4.7 srs when healthy in the regular season (+1.5 difference with the 93 bulls with jordan) and in a very generous best case scenario a (+5.3 difference with even the 92 bulls regular season )

If i average the 94 bulls (+4.7 at full strenght in regular season and +8.9 in playoffs) vs the 92 reg season + 93 playoffs combination draymomd used (and please notice i am already picking and choosing the parts that help jordan more) the gap is only 5 points

That is not goat level.

Even by you guys own approach as it is below other all time greats lift in either absolute terms or in "ceiling raising" situations


I also dont get the "improving a good team is harder" part in relation to kareem, who led a goat level team in the 71 bucks so he was not exactly lacking in "ceiling raising" either compared to jordan while also having better "floor raising" lift as evidenced by their 60~ win pace without oscar in 1972

Or 08/09 garnett who had a similar lift from +3.4 to +9 and he is not even among the goat candidatws short list yet matches jordan here

i could also bring up other cases of lift like 2015 lebron cavs +10 postseason srs with a lot less talent and that the 91 bulls (kirye and love hurt) which strikes me as a even more extreme example of "ceiling raising" considering the floor it came off.

Ok you've gone a bit off tangent here compared to what was discussed. Yes Kareem was a fantastic ceiling raiser but the discussion was about rookie Kareem's WOWY and how much he improved the Bucks upon joining the team. What Kareem did from 1969 to 1970 lifting a -5 SRS team to a +4 SRS team is clearly floor-raising not ceiling-raising. By definition, ceiling raising will lead to smaller SRS improvements many times much smaller.

As for ceiling-raising, an older Jordan also had the Bulls at a 59-win pace (31-12 W/L) from 1996-1998 in games Pippen missed. He matched 1972 Kareem stride for stride. One day I'm going to dive deeper into all the numbers.

Either way there are so many confounding variables and noise with ON-OFF, WOWY, WOWYR and whatever other stat to really push the GOAT narrative in one direction or another. Depending on how you slice the numbers you can make different conclusions. I don't find the impact numbers for Jordan good enough to definitively put him over other GOAT candidates (e.g. Kareem, Lebron) and not definitively bad enough to put him below either.

The boxscore metrics on the other hand, Jordan totally dominates in.

[AEnigma]

Jordan has unequivocally much less of a “box score edge” than Lebron has an impact edge, especially when it matters — and that is with the box score metrics severely falling short of capturing defence and obviously being affected by the quality of opponent defences (where Lebron generally had much tougher opposition than Jordan).

Playoff Jordan, 1987-98: 11.2 BPM, .258 WS/48 over 172 games at 41.7 minutes per game
Playoff Lebron, 2009-20: 10.6 BPM, .261 WS/48 over 214 games at 40.9 minutes per game

Maybe by “box scores” you meant “points per game.”

[Djoker]

Jordan has unequivocally much less of a “box score edge” than Lebron has an impact edge, especially when it matters — and that is with the box score metrics severely falling short of capturing defence and obviously being affected by the quality of opponent defences (where Lebron generally had much tougher opposition than Jordan).

Playoff Jordan, 1987-98: 11.2 BPM, .258 WS/48 over 172 games at 41.7 minutes per game
Playoff Lebron, 2009-20: 10.6 BPM, .261 WS/48 over 214 games at 40.9 minutes per game

Maybe by “box scores” you meant “points per game.”

You must realize you're comparing what is essentially Jordan's entire career (minus first 7 games) vs. Lebron's prime...

And MJ still has a lead in BPM which is the best aggregate box stat.

[AEnigma]

Jordan has unequivocally much less of a “box score edge” than Lebron has an impact edge, especially when it matters — and that is with the box score metrics severely falling short of capturing defence and obviously being affected by the quality of opponent defences (where Lebron generally had much tougher opposition than Jordan).

Playoff Jordan, 1987-98: 11.2 BPM, .258 WS/48 over 172 games at 41.7 minutes per game
Playoff Lebron, 2009-20: 10.6 BPM, .261 WS/48 over 214 games at 40.9 minutes per game

Maybe by “box scores” you meant “points per game.”

You must realize you're comparing what is essentially Jordan's entire career (minus first 7 games) vs. Lebron's prime...

… It is literally twelve postseasons for both of them, across thirteen years, for their age 24-35 postseasons. The fact that essentially constitutes Jordan’s entire postseason career (including a notably larger sample of games) is to Lebron’s credit, not Jordan’s.

I have said before the Jordan stans just blurt out the first thought that pops into their heads to explain away any inconveniences to their preferred narrative. Glad to see that continues to hold true.

And MJ still has a lead in BPM which is the best aggregate box stat.

Yeah that is why I do all my assessments by glancing at the BPM leaderboards. Really makes analysis easy. 11.2 > 10.6 wrap it up boys, Basketball-Reference has spoken.

[VanWest82]

Jordan has unequivocally much less of a “box score edge” than Lebron has an impact edge, especially when it matters — and that is with the box score metrics severely falling short of capturing defence and obviously being affected by the quality of opponent defences (where Lebron generally had much tougher opposition than Jordan).

Playoff Jordan, 1987-98: 11.2 BPM, .258 WS/48 over 172 games at 41.7 minutes per game
Playoff Lebron, 2009-20: 10.6 BPM, .261 WS/48 over 214 games at 40.9 minutes per game

Maybe by “box scores” you meant “points per game.”

You must realize you're comparing what is essentially Jordan's entire career (minus first 7 games) vs. Lebron's prime...

… It is literally twelve postseasons for both of them, across thirteen years, for their age 24-35 postseasons.

I have said before the Jordan stans just blurt out the first thought that pops into their heads to explain away any inconveniences to their preferred narrative. Glad to see that continues to hold true.

And MJ still has a lead in BPM which is the best aggregate box stat.

Yeah that is why I do all my assessments by glancing at the BPM leaderboards. Really makes analysis easy. 11.2 > 10.6 wrap it up boys, Basketball-Reference has spoken.

No, but you only pick playoffs so you don't have to contend with the fact that MJ played hard all the time whereas Lebron didn't. Also, nice job with the cutoff points so you didn't have to include either MJ's ridiculous 86 Celtics series or Lebron's lackluster 08 playoffs.

[AEnigma]

I pick playoffs because a.) the playoffs are what matter, and b.) people like you conflate box score production with “playing hard”.

[VanWest82]

I pick playoffs because a.) the playoffs are what matter, and b.) people like you conflate box score production with “playing hard”.

Actually I "conflate" playing defense with playing hard and not just being a DH. I must be off my rocker again.

[AEnigma]

If you valued defence then you would not be backing Jordan in the first place.

In any case, maybe if Jordan had toned down his “playing hard” in the regular season, then he would not have needed to retire multiple times.

[VanWest82]

"Maybe if Jordan had toned down his playing hard..."

Uh huh.

[falcolombardi]

Jordan has unequivocally much less of a “box score edge” than Lebron has an impact edge, especially when it matters — and that is with the box score metrics severely falling short of capturing defence and obviously being affected by the quality of opponent defences (where Lebron generally had much tougher opposition than Jordan).

Playoff Jordan, 1987-98: 11.2 BPM, .258 WS/48 over 172 games at 41.7 minutes per game
Playoff Lebron, 2009-20: 10.6 BPM, .261 WS/48 over 214 games at 40.9 minutes per game

Maybe by “box scores” you meant “points per game.”

You must realize you're comparing what is essentially Jordan's entire career (minus first 7 games) vs. Lebron's prime...

And MJ still has a lead in BPM which is the best aggregate box stat.

Is not lebron fault his prime is as long as jordan full bulls career