Peaks project update: #12

Moderators: Doctor MJ, trex_8063, penbeast0, PaulieWal, Clyde Frazier

Mavericksfan
Senior
Posts: 533
And1: 200
Joined: Sep 28, 2011

Re: Peaks project update: #12 

Post#41 » by Mavericksfan » Mon Aug 5, 2019 4:54 pm

E-Balla wrote:

And then to add to that WS and BPM scale between eras/seasons/teams horribly so the reason you claim to use them doesn't hold up to that either. The highest OBPM in a season prior to 87 is 37th all time. There's only 6 player seasons between the merger and 87 in the top 100 for single season OBPM. I mean it's a stat created using the 01-14 period to create it's statistical weights, it is horrible to use for previous eras given the fact it wasn't made with them in mind at all. It wasn't even really made with individual players in mind. WS on the other hand doesn't even have the same formula year to year so IDK how it can provide context of numbers in between different seasons when it's not even attempting to.


I ignored the other stuff since it’s mostly strawmen or you flat out misunderstanding the argument.

I’m really curious where you got this information. The creator of BPM himself said it’s meant for comparing historic seasons.

www.apbr.org/metrics/viewtopic.php?f=2&t=9689&start=75

Scroll down to see username DSMok1

“Gradually working on this project!

Yes, the idea is to maintain the existing general concept of BPM (i.e. historic applicability, general structure) and significantly improve the handling of outliers.

Thus far I have focused on the linear version of BPM, currently called GmBPM. It should be very stable and should handle outlier numbers very well.

Then I intend to build upon that framework and add nonlinear terms as appropriate to help handle nuances while hopefully not destroying applicability to outlier values (like the existing BPM did).”

This is from June 13th of this year. So either you’re straight up fabricating or your sources are wrong. That alone makes me call into question your understanding of WS as well.

Here’s a detailed interview mentioning that WS was designed with the intention of accounting for era and team

https://www.thescore.com/nba/news/430681

“Corby: Win Shares does well for itself because it's useful and easy to understand conceptually. It's also comprehensive, that is, it includes a player's offense and defense, which lends it some versatility. A fan can reasonably cite Win Shares as an argument for player awards like the All-Star game or MVP, or career distinctions like the Hall of Fame. Another selling point is that it's a low-bias statistic because it corrects for the tendencies of an era or a player's team.”

This is from Justin Corby one of the guys at b-ball reference.

So again you are either blatantly fabricating or your sources are incorrect.

So now that it’s been established that you misunderstood both metrics can you stop dismissing them?
User avatar
E-Balla
RealGM
Posts: 35,828
And1: 25,127
Joined: Dec 19, 2012
Location: The Poster Formerly Known As The Gotham City Pantalones
   

Re: Peaks project update: #12 

Post#42 » by E-Balla » Mon Aug 5, 2019 5:20 pm

Mavericksfan wrote:
E-Balla wrote:

And then to add to that WS and BPM scale between eras/seasons/teams horribly so the reason you claim to use them doesn't hold up to that either. The highest OBPM in a season prior to 87 is 37th all time. There's only 6 player seasons between the merger and 87 in the top 100 for single season OBPM. I mean it's a stat created using the 01-14 period to create it's statistical weights, it is horrible to use for previous eras given the fact it wasn't made with them in mind at all. It wasn't even really made with individual players in mind. WS on the other hand doesn't even have the same formula year to year so IDK how it can provide context of numbers in between different seasons when it's not even attempting to.


I ignored the other stuff since it’s mostly strawmen or you flat out misunderstanding the argument.

I’m really curious where you got this information. The creator of BPM himself said it’s meant for comparing historic seasons.

www.apbr.org/metrics/viewtopic.php?f=2&t=9689&start=75

Scroll down to see username DSMok1
“Gradually working on this project!

Yes, the idea is to maintain the existing general concept of BPM (i.e. historic applicability, general structure) and significantly improve the handling of outliers.

Thus far I have focused on the linear version of BPM, currently called GmBPM. It should be very stable and should handle outlier numbers very well.

Then I intend to build upon that framework and add nonlinear terms as appropriate to help handle nuances while hopefully not destroying applicability to outlier values (like the existing BPM did).”

This is from June 13th of this year. So either you’re straight up fabricating or your sources are wrong. That alone makes me call into question your understanding of WS as well.

Umm... 2001 is a historic season is it not? He says in that thread the data set he uses to test the changes in his formula are 97-16. It's not a stat optimized for the NBA regardless of era, it's optimized for the modern NBA, modern rules, etc. It is made to mirror RAPM through the boxscore and beyond that doesn't have all too much utility.
User avatar
LA Bird
Analyst
Posts: 3,669
And1: 3,465
Joined: Feb 16, 2015

Re: Peaks project update: #12 

Post#43 » by LA Bird » Mon Aug 5, 2019 5:21 pm

Since multiple people have voiced concerns about the non-enforcement of the reasoning rule, from now on, a vote is only valid if there is at least 1 sentence explaining each pick. It doesn't have to be a long explanation. For example, something like this would suffice and would only take a minute to write.

Erving 76 -- A true season for the ages
Curry 17 -- GOAT regular season offense
Walton 77 -- Actually healthy for the playoffs and played 60+ games though still missing a lot


If you are voting for the exact same players as in a previous round, either copy and paste the reasoning or link the previous post completely. Anything along the lines of "see previous thread for explanation" will result in an invalid vote. I can't remember every single post in earlier threads and it would help the discussion if people don't have to dig through earlier threads to find the justification for votes.

Below, I have quoted everybody who has already voted and did not have a sentence for explaining each pick. Please edit your post before the deadline with reasoning for all 3 picks for the vote to count.

Gregoire wrote:.

Morb wrote:.

GeorgeMarcus wrote:.
Mavericksfan
Senior
Posts: 533
And1: 200
Joined: Sep 28, 2011

Re: Peaks project update: #12 

Post#44 » by Mavericksfan » Mon Aug 5, 2019 5:41 pm

E-Balla wrote:Umm... 2001 is a historic season is it not? He says in that thread the data set he uses to test the changes in his formula are 97-16. It's not a stat optimized for the NBA regardless of era, it's optimized for the modern NBA, modern rules, etc. It is made to mirror RAPM through the boxscore and beyond that doesn't have all too much utility.


You really refuse to admit you’re wrong lol.

From the next page
“Since I have decided that BPM will not include PbP/Lineup data (for ease of use in older or non-NBA settings), I can't include on-court/off-court data.

The assist*reb term will definitely not be included in the new version of BPM, at least in its present form. It is very bad in dealing with outliers. Some sort of modification will be required.”

He specifically doesnt use +/- data so it can apply to pre play by play seasons. That’s well before 2001.

You’re literally arguing against the words of the creator of the damn stat lol.

I knew it was a waste of time but I tried to give you the benefit of doubt. You ego is too big.

You win. I won’t ever hold a discussion with you until you become willing to concede things during a discussion.
DatAsh
Senior
Posts: 627
And1: 356
Joined: Sep 25, 2015

Re: Peaks project update: #12 

Post#45 » by DatAsh » Mon Aug 5, 2019 6:04 pm

The value of stats like BPM is that they allow us to assess, measure, and compare box scores far better than the human brain is capable of by looking at the box scores alone.

They allow us to more accurately compare players than if we only had the base box scores, as the human brain isn’t capable of accurately weighting that many separate variables.
DatAsh
Senior
Posts: 627
And1: 356
Joined: Sep 25, 2015

Re: Peaks project update: #12 

Post#46 » by DatAsh » Mon Aug 5, 2019 6:13 pm

Mavericksfan wrote:
E-Balla wrote:Umm... 2001 is a historic season is it not? He says in that thread the data set he uses to test the changes in his formula are 97-16. It's not a stat optimized for the NBA regardless of era, it's optimized for the modern NBA, modern rules, etc. It is made to mirror RAPM through the boxscore and beyond that doesn't have all too much utility.


You really refuse to admit you’re wrong lol.

From the next page
“Since I have decided that BPM will not include PbP/Lineup data (for ease of use in older or non-NBA settings), I can't include on-court/off-court data.

The assist*reb term will definitely not be included in the new version of BPM, at least in its present form. It is very bad in dealing with outliers. Some sort of modification will be required.”

He specifically doesnt use +/- data so it can apply to pre play by play seasons. That’s well before 2001.

You’re literally arguing against the words of the creator of the damn stat lol.

I knew it was a waste of time but I tried to give you the benefit of doubt. You ego is too big.

You win. I won’t ever hold a discussion with you until you become willing to concede things during a discussion.


I'm not sure I agree with you here, but I agree with your previous posts. BPM definitely favors modern players by rewarding things like defensive rebounds and penalizing offensive rebounds, but as long as you're comparing modern players to modern players and past players to past players, it should be very valuable.

The creator did his best to make it comparable across era's, but given that its weights are based off modern RAPM, it's inevitably going to heavily favor modern players.
Mavericksfan
Senior
Posts: 533
And1: 200
Joined: Sep 28, 2011

Re: Peaks project update: #12 

Post#47 » by Mavericksfan » Mon Aug 5, 2019 6:28 pm

DatAsh wrote:
Mavericksfan wrote:
E-Balla wrote:Umm... 2001 is a historic season is it not? He says in that thread the data set he uses to test the changes in his formula are 97-16. It's not a stat optimized for the NBA regardless of era, it's optimized for the modern NBA, modern rules, etc. It is made to mirror RAPM through the boxscore and beyond that doesn't have all too much utility.


You really refuse to admit you’re wrong lol.

From the next page
“Since I have decided that BPM will not include PbP/Lineup data (for ease of use in older or non-NBA settings), I can't include on-court/off-court data.

The assist*reb term will definitely not be included in the new version of BPM, at least in its present form. It is very bad in dealing with outliers. Some sort of modification will be required.”

He specifically doesnt use +/- data so it can apply to pre play by play seasons. That’s well before 2001.

You’re literally arguing against the words of the creator of the damn stat lol.

I knew it was a waste of time but I tried to give you the benefit of doubt. You ego is too big.

You win. I won’t ever hold a discussion with you until you become willing to concede things during a discussion.


I'm not sure I agree with you here, but I agree with your previous posts. BPM definitely favors modern players by rewarding things like defensive rebounds and penalizing offensive rebounds, but as long as you're comparing modern players to modern players and past players to past players, it should be very valuable.

The creator did his best to make it comparable across era's, but given that its weights are based off modern RAPM, it's inevitably going to heavily favor modern players.


No, you’re 100% right. The point of our discussion was E-balla claiming it was never even made to be applied to older generations.

The context that it was used for in the original discussion was Ewing vs D-rob. Two players from a similar era yet it was still dismissed.

I don’t think any stat can be applied directly across eras due to how much the sport has changed over the years. Everything requires context.
DatAsh
Senior
Posts: 627
And1: 356
Joined: Sep 25, 2015

Re: Peaks project update: #12 

Post#48 » by DatAsh » Mon Aug 5, 2019 6:30 pm

Mavericksfan wrote:
DatAsh wrote:
Mavericksfan wrote:
You really refuse to admit you’re wrong lol.

From the next page
“Since I have decided that BPM will not include PbP/Lineup data (for ease of use in older or non-NBA settings), I can't include on-court/off-court data.

The assist*reb term will definitely not be included in the new version of BPM, at least in its present form. It is very bad in dealing with outliers. Some sort of modification will be required.”

He specifically doesnt use +/- data so it can apply to pre play by play seasons. That’s well before 2001.

You’re literally arguing against the words of the creator of the damn stat lol.

I knew it was a waste of time but I tried to give you the benefit of doubt. You ego is too big.

You win. I won’t ever hold a discussion with you until you become willing to concede things during a discussion.


I'm not sure I agree with you here, but I agree with your previous posts. BPM definitely favors modern players by rewarding things like defensive rebounds and penalizing offensive rebounds, but as long as you're comparing modern players to modern players and past players to past players, it should be very valuable.

The creator did his best to make it comparable across era's, but given that its weights are based off modern RAPM, it's inevitably going to heavily favor modern players.


No, you’re 100% right. The point of our discussion was E-balla claiming it was never even made to be applied to older generations.

The context that it was used for in the original discussion was Ewing vs D-rob. Two players from a similar era yet it was still dismissed.

I don’t think any stat can be applied directly across eras due to how much the sport has changed over the years. Everything requires context.


Agreed.
User avatar
E-Balla
RealGM
Posts: 35,828
And1: 25,127
Joined: Dec 19, 2012
Location: The Poster Formerly Known As The Gotham City Pantalones
   

Re: Peaks project update: #12 

Post#49 » by E-Balla » Mon Aug 5, 2019 6:31 pm

DatAsh wrote:
Mavericksfan wrote:
E-Balla wrote:Umm... 2001 is a historic season is it not? He says in that thread the data set he uses to test the changes in his formula are 97-16. It's not a stat optimized for the NBA regardless of era, it's optimized for the modern NBA, modern rules, etc. It is made to mirror RAPM through the boxscore and beyond that doesn't have all too much utility.


You really refuse to admit you’re wrong lol.

From the next page
“Since I have decided that BPM will not include PbP/Lineup data (for ease of use in older or non-NBA settings), I can't include on-court/off-court data.

The assist*reb term will definitely not be included in the new version of BPM, at least in its present form. It is very bad in dealing with outliers. Some sort of modification will be required.”

He specifically doesnt use +/- data so it can apply to pre play by play seasons. That’s well before 2001.

You’re literally arguing against the words of the creator of the damn stat lol.

I knew it was a waste of time but I tried to give you the benefit of doubt. You ego is too big.

You win. I won’t ever hold a discussion with you until you become willing to concede things during a discussion.


I'm not sure I agree with you here, but I agree with your previous posts. BPM definitely favors modern players by rewarding things like defensive rebounds and penalizing offensive rebounds, but as long as you're comparing modern players to modern players and past players to past players, it should be very valuable.

The creator did his best to make it comparable across era's, but given that its weights are based off modern RAPM, it's inevitably going to heavily favor modern players.

Which is my whole point. If the sample used to create the data is 97-16 it's not going to be well geared towards previous eras no matter how much the creator wants it to be.
DatAsh
Senior
Posts: 627
And1: 356
Joined: Sep 25, 2015

Re: Peaks project update: #12 

Post#50 » by DatAsh » Mon Aug 5, 2019 6:34 pm

E-Balla wrote:
DatAsh wrote:
Mavericksfan wrote:
You really refuse to admit you’re wrong lol.

From the next page
“Since I have decided that BPM will not include PbP/Lineup data (for ease of use in older or non-NBA settings), I can't include on-court/off-court data.

The assist*reb term will definitely not be included in the new version of BPM, at least in its present form. It is very bad in dealing with outliers. Some sort of modification will be required.”

He specifically doesnt use +/- data so it can apply to pre play by play seasons. That’s well before 2001.

You’re literally arguing against the words of the creator of the damn stat lol.

I knew it was a waste of time but I tried to give you the benefit of doubt. You ego is too big.

You win. I won’t ever hold a discussion with you until you become willing to concede things during a discussion.


I'm not sure I agree with you here, but I agree with your previous posts. BPM definitely favors modern players by rewarding things like defensive rebounds and penalizing offensive rebounds, but as long as you're comparing modern players to modern players and past players to past players, it should be very valuable.

The creator did his best to make it comparable across era's, but given that its weights are based off modern RAPM, it's inevitably going to heavily favor modern players.

Which is my whole point. If the sample used to create the data is 97-16 it's not going to be well geared towards previous eras no matter how much the creator wants it to be.


It's definitely going to favor modern players over past players. In this case though, Ewing should be subject to the same BPM disadvantages than Robinson is? Do you disagree?

I guess you could argue that the player who plays closer to modern standards should face less of a disadvantage. Not sure who that would be here, though.
DatAsh
Senior
Posts: 627
And1: 356
Joined: Sep 25, 2015

Re: Peaks project update: #12 

Post#51 » by DatAsh » Mon Aug 5, 2019 6:40 pm

Really curious about Chris Paul here. He's arguably the best defensive PG of all time, and probably much better than Oscar.

He's also one of the best offensive players ever. How do y'all view Paul's peak offense relative to Oscar's? Given that Paul was a much better defender, you'd have to have Oscar significantly ahead on offense to rank him above Paul overall.

Am I off base here? I had Oscar well above Paul, but thinking about it now, I'm not so sure.
User avatar
E-Balla
RealGM
Posts: 35,828
And1: 25,127
Joined: Dec 19, 2012
Location: The Poster Formerly Known As The Gotham City Pantalones
   

Re: Peaks project update: #12 

Post#52 » by E-Balla » Mon Aug 5, 2019 6:45 pm

DatAsh wrote:The value of stats like BPM is that they allow us to assess, measure, and compare box scores far better than the human brain is capable of by looking at the box scores alone.

They allow us to more accurately compare players than if we only had the base box scores, as the human brain isn’t capable of accurately weighting that many separate variables.

I still am not getting this, what is the inherent value here? You're all starting from the assumption that a single number can gather all boxscore contributions in the first place and if you aren't tell me why we need to come up with a way to do so?

If player A has a 2.0 BPM and you know nothing else what can you say about that player at all (that you would be comfortable saying as a blanket statement)?

Now if player B averaged 30 ppg on 70 TS% on the other hand (including the full boxscore of course) there's plenty of things you can derive from that information. Now it's to you to convince us that information makes them better at basketball than another player (which is the purpose of projects like this and discussion on this message board in general) but there's actual real, tangible information being given there.
Colbinii
RealGM
Posts: 34,243
And1: 21,859
Joined: Feb 13, 2013

Re: Peaks project update: #12 

Post#53 » by Colbinii » Mon Aug 5, 2019 6:47 pm

DatAsh wrote:Really curious about Chris Paul here. He's arguably the best defensive PG of all time, and probably much better than Oscar.

He's also one of the best offensive players ever. How do y'all view Paul's peak offense relative to Oscar's? Given that Paul was a much better defender, you'd have to have Oscar significantly ahead on offense to rank him above Paul overall.

Am I off base here? I had Oscar well above Paul, but thinking about it now, I'm not so sure.


I have Paul in the same tier as Curry/Magic/West/Oscar for PGs with little separation between all of them.
Mavericksfan
Senior
Posts: 533
And1: 200
Joined: Sep 28, 2011

Re: Peaks project update: #12 

Post#54 » by Mavericksfan » Mon Aug 5, 2019 6:56 pm

DatAsh wrote:Really curious about Chris Paul here. He's arguably the best defensive PG of all time, and probably much better than Oscar.

He's also one of the best offensive players ever. How do y'all view Paul's peak offense relative to Oscar's? Given that Paul was a much better defender, you'd have to have Oscar significantly ahead on offense to rank him above Paul overall.

Am I off base here? I had Oscar well above Paul, but thinking about it now, I'm not so sure.


I think the size advantage really helps Oscar here. Being a better rebounder and able to guard 1-3 mitigates Chris Paul’s elite PG man defense but size issues.

And offensively I think it’s Oscar easily. His relative Ts% are pretty insane in addition to anchoring some top offenses.

I’d probably have Paul somewhere below Wade/T-Mac.

I honestly want to bring up Dirk since Walton is being mentioned but I know my bias may be showing. But the Mavs dominated all year long with Dirk in the lineup and went something like 3-9 without him.
User avatar
E-Balla
RealGM
Posts: 35,828
And1: 25,127
Joined: Dec 19, 2012
Location: The Poster Formerly Known As The Gotham City Pantalones
   

Re: Peaks project update: #12 

Post#55 » by E-Balla » Mon Aug 5, 2019 7:00 pm

DatAsh wrote:
E-Balla wrote:
DatAsh wrote:
I'm not sure I agree with you here, but I agree with your previous posts. BPM definitely favors modern players by rewarding things like defensive rebounds and penalizing offensive rebounds, but as long as you're comparing modern players to modern players and past players to past players, it should be very valuable.

The creator did his best to make it comparable across era's, but given that its weights are based off modern RAPM, it's inevitably going to heavily favor modern players.

Which is my whole point. If the sample used to create the data is 97-16 it's not going to be well geared towards previous eras no matter how much the creator wants it to be.


It's definitely going to favor modern players over past players. In this case though, Ewing should be subject to the same BPM disadvantages than Robinson is? Do you disagree?

Well here we're talking about a stat that doesn't exist to start (BPM for those specific series only), but yes they are subject to the same disadvantages. It won't effect them the same because of their different games though, BPM isn't counting all points equally and the way certain stats that have nothing to do with offense (like steals) affect OBPM can lead to issues in direct comparisons. At that point, you might as well dump BPM and use the boxscore adding context as you go because in a direct comparison of players you have the time to actually do that.

I guess you could argue that the player who plays closer to modern standards should face less of a disadvantage. Not sure who that would be here, though.

The one with the high steal numbers and free throws drawn (neither shot 3s).

And let's not even get into the terribleness of boxscore defensive stats. DBPM, DWS, individual DRTG are all complete abominations wholly unfit for discussion.

Really BPM and those numbers aren't adding context, they're an excuse people use to not add context. Mentally we already know spacing is valuable, efficiency is valuable, etc. why do we need a number that can possibly be completely wrong for a player just to see that? BPM's real application comes in statistical models and predicting future play, in which case getting each individual player right has little application but getting something close to right for everyone on a given team is.
User avatar
E-Balla
RealGM
Posts: 35,828
And1: 25,127
Joined: Dec 19, 2012
Location: The Poster Formerly Known As The Gotham City Pantalones
   

Re: Peaks project update: #12 

Post#56 » by E-Balla » Mon Aug 5, 2019 7:03 pm

Colbinii wrote:
DatAsh wrote:Really curious about Chris Paul here. He's arguably the best defensive PG of all time, and probably much better than Oscar.

He's also one of the best offensive players ever. How do y'all view Paul's peak offense relative to Oscar's? Given that Paul was a much better defender, you'd have to have Oscar significantly ahead on offense to rank him above Paul overall.

Am I off base here? I had Oscar well above Paul, but thinking about it now, I'm not so sure.


I have Paul in the same tier as Curry/Magic/West/Oscar for PGs with little separation between all of them.

I think Paul gets screwed over by what came after 09 a bit because if I'm being honest part of why I'm not putting 08 Paul that high is his durability concerns but now that I'm thinking about it he was still durable as hell in 08.

Still I don't see him as over Kobe so he's not coming for a while.
Colbinii
RealGM
Posts: 34,243
And1: 21,859
Joined: Feb 13, 2013

Re: Peaks project update: #12 

Post#57 » by Colbinii » Mon Aug 5, 2019 7:05 pm

E-Balla wrote:
Colbinii wrote:
DatAsh wrote:Really curious about Chris Paul here. He's arguably the best defensive PG of all time, and probably much better than Oscar.

He's also one of the best offensive players ever. How do y'all view Paul's peak offense relative to Oscar's? Given that Paul was a much better defender, you'd have to have Oscar significantly ahead on offense to rank him above Paul overall.

Am I off base here? I had Oscar well above Paul, but thinking about it now, I'm not so sure.


I have Paul in the same tier as Curry/Magic/West/Oscar for PGs with little separation between all of them.

I think Paul gets screwed over by what came after 09 a bit because if I'm being honest part of why I'm not putting 08 Paul that high is his durability concerns but now that I'm thinking about it he was still durable as hell in 08.

Still I don't see him as over Kobe so he's not coming for a while.


That's fair. I have 08/09 as Kobe's peak and CP3 above 08 Kobe.
User avatar
GeorgeMarcus
Retired Mod
Retired Mod
Posts: 18,876
And1: 24,033
Joined: Jun 17, 2006
     

Re: Peaks project update: #12 

Post#58 » by GeorgeMarcus » Mon Aug 5, 2019 7:25 pm

95 D Rob
16 Curry
76 Dr J
The Legend of George Marcus

"Where I'm from, bullies get bullied." - Zach Randolph
No-more-rings
Head Coach
Posts: 7,104
And1: 3,913
Joined: Oct 04, 2018

Re: Peaks project update: #12 

Post#59 » by No-more-rings » Mon Aug 5, 2019 10:24 pm

DatAsh wrote:Really curious about Chris Paul here. He's arguably the best defensive PG of all time, and probably much better than Oscar.

He's also one of the best offensive players ever. How do y'all view Paul's peak offense relative to Oscar's? Given that Paul was a much better defender, you'd have to have Oscar significantly ahead on offense to rank him above Paul overall.

Am I off base here? I had Oscar well above Paul, but thinking about it now, I'm not so sure.

I’m similar to what Mavericksfan said, it’s a size thing and i’ve never been convinced of Paul being a game changer on defense despite him being intelligent and playing bigger than he actually is.
ardee
RealGM
Posts: 15,320
And1: 5,397
Joined: Nov 16, 2011

Re: Peaks project update: #12 

Post#60 » by ardee » Tue Aug 6, 2019 3:37 am

Same vote as last time.

1. 1977 Bill Walton

Not much needed to be said about this season. I think it's basically equal to peak Russell. Below him defensively but above offensively: I think if I had to pick the second best defensive peak of all time though it would be hard to choose against him here. There was a quarter against the Sixers in the Finals when he took over defensively in a manner comparable to when an offensive player scores 20 in a quarter. Announcers were losing their ****, screaming "the Sixers cannot find a way to score against Bill Walton!!" Unreal awareness on defense, insane rebounder, and one of the top 3 playmaking centers of all time (with Wilt and Jokic). To me his offense is worth more than someone like Alonzo Mourning or most of Patrick Ewing's career outside of his outlier peak. Those guys aren't outlier scorers like Kareem or Shaq, in which case I'd rather have Walton's creation and lower volume than medium volume on not so great efficiency.

2. 2008 Kobe

ardee wrote:Lakers have a 7.4 SRS, 57 wins, no.1 seed.

The standard line-up with everyone healthy was Fisher/Kobe/Radmanovic/Odom/Pau. Pau was only healthy 27 games. Bynum was healthy for 35, and they never played together.

Player by player: Fisher had a good year. 12/23, 44% from the field and 41% from 3. He was still all right on defense. I want you to note his jump in efficiency going from the Jazz to playing with Kobe. This is something that has been seen when many players play with and then without Kobe. He draws so much attention that they see their percentages rank.

Radmanovic was also basically a shooter. He shot 41% from 3, and 44% for the first half of the '09 season. This dipped to 36% when he was traded in the second half, and further to 28 the next season. So elite shooter with Kobe, average to bad without.

Odom was phenomenal that year, no doubts about it, great player all around. The main reason was because we first had Bynum and then Pau to be the second option to Kobe, while Odom was more comfortable as no. 3. His TS% jumped 3.5% from 55 '05-'07, when he was no. 2, to 58.5 in '08, when he was no. 3. In the stretch between Bynum's injury and the Gasol trade when he had to be the no. 2 option again, he shot 42% TS.

Pau was the perfect no. 2 option for Kobe, of course he was, we won 2 titles with him. Remember 2 things though:

1. He played 27 games.
2. As the no. 1 in Memphis, his team was 13-32 before he got traded. They ended up 22-60, so they went from a .280 win pace with him to a .244 win pace without him.

Bynum was also good, however, he wasn't as good as Pau, the numbers spell it out. He played 35 games, and would likely get injured quicker if he

Kobe took this cast to a 7.4 SRS and 57 wins.

I want you to imagine this team with no Kobe.

You'd be starting Fisher/Vujacic/Radmanovic/Odom/27 games of Pau + 35 games of Bynum + 16 games of Turiaf.

The best team would be the one with Pau. Consider, however, like I said, how Pau did on a Memphis team that was poorly built but still had some talent. Their lead scorer was Rudy Gay, who is a flawed player but can at least provide some kind of offense when needed. They had a lights out shooter at the 2 in Mike Miller.

This hypothetical Lakers team built around Pau would have Odom as their second option. Scoring wise, he is worse than Gay for this role. I have already shown he struggles to be consistent in that role. He struggles like that with KOBE as his first option. Pau is a far inferior first option to Kobe and that would put a ton more pressure on Odom. Fisher and Radmanovic can't create, neither can Sasha, and their efficiency dropped heavily when not playing with Kobe.

You can make the argument that this efficiency was due to the triangle partially, and not all Kobe, but the triangle only WORKS when you have an elite perimeter creator like Kobe. So therefore, you can rest assured their efficiency would drop a good bit, if not all the way down to what it was when they didn't play for the Lakers.

So, Pau, inconsistent in the 2nd option role Odom now with the added pressure of playing with a worse no. 1 option than Kobe, and Sasha, Fisher and Radmanovic offering little. I honestly don't see more than .500 in those 27 games and that's being VERY optimstic. In fact, it's more likely to be like 10-11 wins out of 27. The Blazers were a .500 team and they had 2 legit scoring options in Roy and Aldridge surrounded by fitting role players. The Lakers without Kobe are worse then that, even with Pau. Let's call it a push at 12-13 wins in those 27 games.

Bynum's 35 games. Bynum was worse than Pau at everything. He doesn't offer Pau's high-post playmaking. He can still be the main scorer but now Odom has to be the primary creator. More pressure on him. Bynum might get injured from the extra strain. I don't see more than 12-14 wins out of 35. Again, optimstically.

16 games of Turiaf. Odom in the no. 1 role. The team completely falls apart. Maybe 1-2 wins in 16 games.

So essentially, that team in a full season without Kobe wins 25-29 games. They won 57. Kobe was providing roughly ~30 wins of lift.

With this knowledge, it is hard for me to rank Kobe lower than 12 on the all-time peaks. I have Walton at 11, and this is equivalent to the kind of lift we know him to provide.

This was not a good supporting cast. If he had a full season of Pau it would be different, I think the '09 Lakers were great, but 27 games means he was working with a lot less for the rest of the season. It was a good-fitting supporting cast but aside from Pau all the players were supremely dependant on Kobe to do well in their roles.

He took an otherwise lottery team to elite status and put up a historical ORtg for the team when he had Pau.



Through the first 3 rounds of the Playoffs, the Lakers played 3 50 win teams and Kobe averaged 32-6-6 on 60% TS. That is peak Jordan level production against elite opposition. People forget the Jazz were a 7 SRS team and Kobe averaged 33-7-7 against them. People forget he dropped 30 ppg on 53% from the field against the defending champ Spurs while no other star in the series got anything going on that end.

On the weight of the RS and his stunning Playoff performances, Kobe absolutely should not be ranked any lower.

3. 2006 Kobe

Very similar lift to 2008 Kobe. Led a putrid cast to 48 Pythogorean wins, the 7th best SRS and the 8th best ORtg in the league. For reference, the 2006 Lakers performed as well as the 2019 Nuggets relative to the league and the 2019 Sixers offensively. Look at those rosters and then look at the roster Kobe was surrounded with. Tell me he doesn't deserve this spot.

Return to Player Comparisons