Top 150+ GOAT (Peak, Prime, Longevity, Award)

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

User avatar
2klegend
Bench Warmer
Posts: 1,333
And1: 409
Joined: Mar 31, 2016
     

Re: Top 100+ GOAT (Peak, Prime, Longevity, Award) 

Post#61 » by 2klegend » Mon Aug 8, 2016 9:56 am

Quotatious wrote:Calling Detlef Schrempf "just a random average player" is seriously insulting. He averaged 19/9.5/6 on 57% TS one season - that's pretty much a poor man's Larry Bird. Hell, that wasn't even necessarily his best season - he averaged 19/6/4 on 64% TS in '95 (including 51% 3-pt - even with a shortened 3-pt line, that's phenomenal).

He played like with a dumbbell on his nut in the playoff in '95. So I would't that insulting. He is a solid player. Nothing spectacular.
My Top 100+ GOAT (Peak, Prime, Longevity, Award):
viewtopic.php?f=64&t=1464952
Lucky707
Sophomore
Posts: 100
And1: 59
Joined: Jun 09, 2012

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#62 » by Lucky707 » Mon Aug 8, 2016 7:10 pm

What do you think of a very, very small boost for those who have won DPOY?

(I am aware that the award started in 1982 so that would make things VERY complicated)
Arman_tanzarian
Veteran
Posts: 2,577
And1: 2,712
Joined: Dec 27, 2012
     

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#63 » by Arman_tanzarian » Mon Aug 8, 2016 7:17 pm

Great list and well thought out rationale.
Image
User avatar
2klegend
Bench Warmer
Posts: 1,333
And1: 409
Joined: Mar 31, 2016
     

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#64 » by 2klegend » Tue Aug 9, 2016 8:12 am

Lucky707 wrote:What do you think of a very, very small boost for those who have won DPOY?

(I am aware that the award started in 1982 so that would make things VERY complicated)

Like you said, the late award make it tricky to value a DPOY. Then I have to value a ROY, MIP, All-NBA, etc.. I personally don't think those awards are significant to define a player career.
My Top 100+ GOAT (Peak, Prime, Longevity, Award):
viewtopic.php?f=64&t=1464952
trex_8063
Forum Mod
Forum Mod
Posts: 12,495
And1: 8,134
Joined: Feb 24, 2013
     

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#65 » by trex_8063 » Fri Aug 12, 2016 3:15 am

2klegend wrote:.

Joao Saraiva wrote:.


Am quoting you two just because you’ve both recently worked on and expressed interest in these sorts of “all-in-one” formulas.
Not to highjack your thread, but rather by way of sharing (I too have had interest in making formulas to function as a “complete” player greatness measurement).......thought you might be interested in the results of some of my formulas.

The most in-depth one I made attempts to incorporate LOTS of factors. Without totally spilling formula’s guts for inspection, I’ll summarize to say it’s based on the following:
1) Career Production and efficiency: in the “original” (which had actually undergone at least a half-dozen modifications already), this was done with a heavy emphasis on PER (because that was the all-in-one metric I was into at the time, and which produced I think a touch of bias toward volume players)---weighted against mpg---though I also used some cumulative pts/reb/ast (part of my accounting for longevity, which is very important to me), ppg (minimally weighted factor, though this further biases it toward volume scorers), WS, and WS/48 to a smaller degree. PER and WS/48 were also weighted against rs win% (which may have created a touch of rs “winner’s bias” in the “original” formula, but this was part of my accounting for context).
For playoffs I used PER (again, the heaviest input and again weighted against mpg and playoff win%) along with playoff WS and WS/48.
Regular season inputs were factored against total seasons played (another accounting of longevity), and playoffs against total playoff games played.
2) Also included with/without records as a means of measuring impact (did not use RAPM, given the relatively limited time period that’s been available for).
3) Peak season value (as measured by PER and WS/48, weighted against mpg).
4) rDRtg (relative [to league avg] DRtg). For players with seasons prior to ‘74 I used team DRtg as a proxy, making marginal adjustments from the team average based on defensive reputation, subjective opinion, etc. I didn’t want to get too liberal in how much I deviated from the team avg, though; consequently, for example, someone like Bill Russell is likely being undercredited by the DRtg I assigned him (while some of his teammates, such as Bob Cousy or Tom Heinsohn, are likely be over-credited in their DRtg’s).
5) **Awards, honors. This was neither a huge nor small contributing factor to final score. I had a point system for valuation of All-Star nominations, All-NBA honors (different for 1st, 2nd, and 3rd team), All-Defensive honors, DPOY, Scoring champ (minimal for that one, fwiw), MVP; this latter is worth less than some of the other honors because I also included in the formula…..
….6) **MVP Award Shares (this also is a somewhat substantial component toward a player’s final score, more so than awards/honors).
7) ”Title points”. These are accrued for being a member a title team, winning a FMVP (worth same value as a title/ring), or simply making it to the finals (4/10 the value of a title). I didn’t make different values relative to one’s role on a title team, though that may be an adjustment I make on a future version.

**Obviously some previous players are held back by some of these honors not being present in their careers (no defensive honors or FMVP until ‘69--->effects on Bill Russell, for example?; no MVP prior to ‘56--->effects on George Mikan? etc). These are simply considerations I have to hold in the back of my mind when scrutinizing their scores; and I have created hypothetical versions of them to see how the MIGHT have scored if these things had been available.

NOTE: Awards/honors are weighted against a “Positional Era Rating”......an evaluation of the strength of high-end competition at their position (C, F. or G), which I rated each and every year individually by position and overall (by a combination of objective and subjective means).

ALL OTHER FACTORS are weighted against the “Overall Era Rating”, which was also created by sort of adding up the high-level talent in the league year-by-year, though also giving a slight mathematical favoritism to leagues with a larger number of teams. My reason for doing so was assuming larger league represents an increasingly global popularity of the game and larger player pool to sample from, as well as more technically advanced (or “modern”) technique and style of play, and also noting tougher playoff structure (as compared to when six of eight teams in the league would make the playoffs, and a team may only have to win ONE series to make it into the finals).
Ultimately though----for most things which are weighted against the Overall Era Rating----I ended up using a variable to dilute the overall era ratings. I admit this was purely a “massaging the results” adjustment, as the formula initially seemed to otherwise under-credit older players. I somewhat justify the dilution noting that what talent was present was more concentrated in those earlier years. This is perhaps relevant when noting, for instance, that someone like Bill Russell or Wilt Chamberlain didn’t have the “luxury” of---literally every other game---facing opposing centers who were roughly the quality of Len Elmore, Kent Benson, Herb Williams, James Edwards, Dave Corzine, Steve Johnson, Wayne Cooper, Jerome Whitehead, or Tree Rollins (great rim protector, but offensive zero).......such as Moses Malone was facing every other game in the early 80’s.

In summary, the formula (“original” version) derives the largest portions of a player’s score from rs PER weighted against mpg, seasons played, and win% (and diluted era strength rating); MVP award shares (where applicable, and weighted against era strength); and playoff PER weighted against mpg, playoff games played, playoff win%, and diluted era strength rating.
Award points is the next most highly weighted factor (weighted directly against positional era rating), followed closely by “title points” (weighted against diluted era strength rating), and then followed by some cumulative numbers (rs pts/reb/ast, rs WS, playoff WS…….these to better reward meaningful longevity, though the WS also bring in a little more accounting of efficiency), rs and playoff WS/48 (both weighted against mpg and diluted era strength), and a peak rating (as measured by the highest PER*WS/48*mpg product). The least weighted factors are the with/without records, DRtg (individual DRtg can be a bit sketchy, especially when using estimates for pre-1974), and ppg numbers.


I made a second version (titled “Version 16.8.1”) which has the following major alterations from the original:
1) MUCH less rewarding to longevity.
2) PER and WS/48 are accounted only in their relation to a “replacement level player” (in the “original” version a PER of 5.0, for example, would still yield some degree of positive score; in this version it effects the score up or down based on how much above or below replacement level it is). Can outline what “replacement level” parameters I used; Owly helped come up with the figures.
…..And the following minor alterations:
3) Slightly less rs winner’s bias (less accounting for rs win%).
4) Slightly more consideration toward efficiency (by way of greater weight for WS/48); NOTE: PER still weighted more heavily, though.
5) Slight more emphasis toward big-man defenders (recognizing their often huge defensive value as rim protectors (used rebounds as a proxy for this).
6) Slightly less ppg emphasis.


Then I made a third version (“Version 16.8.2”) which has the following major alterations from the original:
1) Much MUCH more consideration toward efficiency (via MUCH heavier weighting for WS/48, now weighted heavier than PER).
2) Much more “Title points” emphasis (really isn’t TOO huge a contributing factor in first two versions; I made it fairly prominent in this one, though).
3) PER and WS/48 are accounted for by their relation to a replacement level player (even better accounting of this than the 16.8.1 version).
4) Significantly greater emphasis on big-man defenders (using rebounds as proxy to even greater degree than in version 16.8.1; also greater weight given to DRtg).
…...and the following minor alterations:
5) Little more emphasis on longevity.
6) Slightly more award/honors emphasis
7) Slightly less MVP award shares emphasis.
8) Somewhat firmer accounting of era strength (i.e. not “diluting” my Overall Era Ratings as much).
9) Slightly less ppg emphasis.


ANYWAY (whew!)......I’ve run 321 players thru all three versions of this formula. And then I amalgamated the results/rankings of the three versions (hoping the strengths/flaws of each will balance each other), finding a player’s average rank among the three. Thought you might be interested in knowing who the top 150 are in this amalgamation (btw, this is NOT the order of my ATL, though it wouldn’t make a bad list imo, especially if you make small adjustments based on my notations below. Anyway, this is something I did for hobby/interest and use as merely one more tool):

Spoiler:
1. Michael Jordan
2. Kareem Abdul-Jabbar
3. Lebron James
**Lebron just barely behind: his average rank was 2.67 (finished 2nd in one version, 3rd in the other two), vs. 2.33 for Kareem (2nd in two versions, 3rd in the other).
4. Wilt Chamberlain
5. Tim Duncan
**Duncan barely behind with average rank of 4.67 vs. the slightly higher 4.33 for Wilt.
6. Shaquille O’Neal
7. Karl Malone
8. Kobe Bryant
9. Magic Johnson
**Kobe avg 8.33 rank to Magic’s 8.67; though I’d note this is with Kobe’s obviously undeserved All-Star nominations in ‘14, ‘15, and ‘16, as well as at least one or two undeserved All-D honors (which I’m obligated to include them all); take those away, however, and their positions would flip-flop (Kobe #9, Magic #8).
10. Larry Bird
11. Hakeem Olajuwon
t12. Dirk Nowitzki
t12. David Robinson
**tied in average rank; Nowitzki narrowly wins tie-breaker by having higher cumulative score (if add up score of all three versions)
t14. Kevin Garnett
t14. Bill Russell
**Tied in average rank; Garnett narrowly wins tie-breaker based on cumulative score. I’m not surprised or discouraged that Russell is placed this low. If stat-keeping had been more complete in his day (namely blocks), I suspect both his PER and WS/48 would be a little higher; and an actual DRtg is likely significantly better than the conservative estimate I’ve given him. Also, were defensive honors and FMVP awarded throughout his career, that too would change things. I’ve pumped a hypothetical Bill Russell (estimating the changes to PER, WS/48, gave a little more liberal DRtg estimation, as well as what defensive honors and FMVP’s he would have won) in the formulas…..this hypothetical would have the 10th-highest average rank (very nearly 9th). And this is before considering Russell’s other “intangible” values.
16. Julius Erving
17. Charles Barkley
18. Jerry West
19. Oscar Robertson
t20. Moses Malone
t20. John Stockton
**Stockton narrowly wins the tie-breaker based on cumulative points.
22. Dwyane Wade
23. Chris Paul
24. Bob Pettit
25. Kevin Durant
26. Artis Gilmore
27. Scottie Pippen
28. Elgin Baylor
29. Patrick Ewing
30. Clyde Drexler
31. Jason Kidd
32. Steve Nash
33. Paul Pierce
34. John Havlicek
35. Rick Barry
36. Dolph Schayes
37. Dan Issel
38. Dwight Howard
39. Robert Parish
40. Pau Gasol
41. Gary Payton
42. Reggie Miller
43. Kevin McHale
44. Allen Iverson
45. Elvin Hayes
46. Chauncey Billups
47. Ray Allen
48. George Mikan
49. Stephen Curry
50. Walt Frazier
51. Tracy McGrady
52. George Gervin
53. Manu Ginobili
54. Tony Parker
55. Isiah Thomas
56. Bob Lanier
57. Adrian Dantley
58. Bob Cousy
59. Dominique Wilkins
60. Russell Westbrook
61. Vince Carter
t62. Horace Grant
t62. Shawn Marion
**tied avg rank; Grant wins tie-breaker of cumulative score.
t64. Chris Bosh
t64. Bob McAdoo
**tied avg rank; Bosh wins tie-breaker.
66. Alonzo Mourning
67. Walt Bellamy
68. Amar’e Stoudemire
69. Carmelo Anthony
70. Kevin Johnson
71. Chris Webber
72. Wes Unseld
73. Dikembe Mutombo
74. Sam Jones
75. James Worthy
76. James Harden
77. Dave Cowens
78. Zelmo Beaty (he’s another that might be worth putting in your formula)
79. Alex English
80. Jack Sikma
81. Shawn Kemp
82. Paul Arizin
83. Elton Brand
84. George McGinnis
85. Grant Hill
t86. Larry Nance
t86. Terry Porter
**Nance wins tie-breaker by negligible margin.
88. Rasheed Wallace
89. Willis Reed
90. Ben Wallace
t91. Jeff Hornacek
t91. Chet Walker
**Hornacek wins tie-breaker.
93. Jerry Lucas
94. Maurice Cheeks
95. Sidney Moncrief
t96. Bailey Howell
t96. Marques Johnson
**Howell wins tie-breaker.
98. Neil Johnston
99. Bobby Jones
t100. Bill Laimbeer
t100. Terry Cummings
**Laimbeer wins tie-breaker.
102. Mel Daniels
103. Sam Cassell
104. Billy Cunningham
t105. Cliff Hagan
t105. Vlade Divac
**Hagan narrowly wins the tie-breaker.
107. Connie Hawkins
t108. Spencer Haywood
t108. Hal Greer
**Haywood narrowly wins tie-breaker.
110. Tim Hardaway
111. Buck Williams
112. Detlef Schrempf
113. Bernard King
114. Anfernee Hardaway
115. Yao Ming
t116. Deron Williams
t116. Kawhi Leonard
**Kawhi very narrowly wins tie-breaker (though fwiw, I don’t think he deserved both DPOY awards; removing one would drop his rank a few spots)
t118. Dennis Johnson
t118. Chris Mullin
**Chris Mullin wins tie-breaker by negligible margin
120. LaMarcus Aldridge
121. Carlos Boozer
122. Dennis Rodman
t123. Bob Dandridge
t123. Gus Williams
**Dandridge very narrowly wins the tie-breaker based on cumulative score.
125. Nate Thurmond
126. Bill Sharman
127. Tiny Archibald
128. Mark Jackson
129. Zach Randolph
130. Walter Davis
131. Blake Griffin
132. Andre Miller
133. Kevin Love
134. Eddie Jones
135. Vern Mikkelsen
136. Lamar Odom
t137. Joe Dumars
t137. Lenny Wilkens
**Joe Dumars narrowly wins the tie-breaker. Note Dumars is drastically underrated by individual DRtg, one of many examples pertaining to my reluctance to put a lot of weight on that metric.
139. Tom Heinsohn
140. Ed Macauley
141. Mark Aguirre
142. Rod Strickland
143. Jason Terry
t144. Marcus Camby
t144. Sam Perkins
**Sam Perkins wins the tie-breaker by a slim margin.
146. Brad Daugherty
t147. Baron Davis
t147. Larry Foust
**Larry Foust wins the tie-breaker
149. Jimmy Jones
150. David West
"The fact that a proposition is absurd has never hindered those who wish to believe it." -Edward Rutherfurd
"Those who can make you believe absurdities, can make you commit atrocities." - Voltaire
User avatar
2klegend
Bench Warmer
Posts: 1,333
And1: 409
Joined: Mar 31, 2016
     

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#66 » by 2klegend » Sat Aug 13, 2016 5:39 am

It is okay to make a list but I don't think the idea of era competition is possible to quantify. I'm also modifying a bit on the Award to include more highlight because I felt certain players are too high (Ginobili) while his accomplishment is not enough to warrant that level of rank. Therefore I'm going to include ALL NBA, All Star, and ALL NBA Defense. Stay Tune..
My Top 100+ GOAT (Peak, Prime, Longevity, Award):
viewtopic.php?f=64&t=1464952
User avatar
2klegend
Bench Warmer
Posts: 1,333
And1: 409
Joined: Mar 31, 2016
     

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#67 » by 2klegend » Sat Aug 13, 2016 9:36 am

I'm modifying my Award calculation to account for All Star, ALL NBA selection. The rationale for All NBA 1st = 2pts and AS = 1pt is simple. There are only 5 can be chosen, instead of 12 for AS selection. So 2nd All NBA receive similar chance to get AS selection.

Title as the Main Focus (best player) = 10pts
Title as the compliment star = 5pts
Title as support role = 2.5
MVP = 10pts
Final MVP = 5pts
All Star = 1pt
All NBA 1st Team = 2pts
All NBA 2nd Team = 1pt
All NBA 1st Defensive Team = 2pts
All NBA 2nd Defensive Team = 1pt

With the change, Lebron has surpass Bill Russell. Kobe and Hakeem are really close but once award were factor in, Kobe got in the Top 10 ahead of Hakeem which I think is correct. Ginobili is also dropping from 21th to 30th. Nash is unchanged despite the new addition, still at 51th. Though, the lowest in the 150 is Spencer Haywood, instead of Tiny Archibald whose moved up from 150th to 149th.
My Top 100+ GOAT (Peak, Prime, Longevity, Award):
viewtopic.php?f=64&t=1464952
User avatar
Joao Saraiva
RealGM
Posts: 13,336
And1: 6,140
Joined: Feb 09, 2011
   

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#68 » by Joao Saraiva » Sat Aug 13, 2016 11:01 am

trex_8063 wrote:
2klegend wrote:.

Joao Saraiva wrote:.


Am quoting you two just because you’ve both recently worked on and expressed interest in these sorts of “all-in-one” formulas.
Not to highjack your thread, but rather by way of sharing (I too have had interest in making formulas to function as a “complete” player greatness measurement).......thought you might be interested in the results of some of my formulas.

The most in-depth one I made attempts to incorporate LOTS of factors. Without totally spilling formula’s guts for inspection, I’ll summarize to say it’s based on the following:
1) Career Production and efficiency: in the “original” (which had actually undergone at least a half-dozen modifications already), this was done with a heavy emphasis on PER (because that was the all-in-one metric I was into at the time, and which produced I think a touch of bias toward volume players)---weighted against mpg---though I also used some cumulative pts/reb/ast (part of my accounting for longevity, which is very important to me), ppg (minimally weighted factor, though this further biases it toward volume scorers), WS, and WS/48 to a smaller degree. PER and WS/48 were also weighted against rs win% (which may have created a touch of rs “winner’s bias” in the “original” formula, but this was part of my accounting for context).
For playoffs I used PER (again, the heaviest input and again weighted against mpg and playoff win%) along with playoff WS and WS/48.
Regular season inputs were factored against total seasons played (another accounting of longevity), and playoffs against total playoff games played.
2) Also included with/without records as a means of measuring impact (did not use RAPM, given the relatively limited time period that’s been available for).
3) Peak season value (as measured by PER and WS/48, weighted against mpg).
4) rDRtg (relative [to league avg] DRtg). For players with seasons prior to ‘74 I used team DRtg as a proxy, making marginal adjustments from the team average based on defensive reputation, subjective opinion, etc. I didn’t want to get too liberal in how much I deviated from the team avg, though; consequently, for example, someone like Bill Russell is likely being undercredited by the DRtg I assigned him (while some of his teammates, such as Bob Cousy or Tom Heinsohn, are likely be over-credited in their DRtg’s).
5) **Awards, honors. This was neither a huge nor small contributing factor to final score. I had a point system for valuation of All-Star nominations, All-NBA honors (different for 1st, 2nd, and 3rd team), All-Defensive honors, DPOY, Scoring champ (minimal for that one, fwiw), MVP; this latter is worth less than some of the other honors because I also included in the formula…..
….6) **MVP Award Shares (this also is a somewhat substantial component toward a player’s final score, more so than awards/honors).
7) ”Title points”. These are accrued for being a member a title team, winning a FMVP (worth same value as a title/ring), or simply making it to the finals (4/10 the value of a title). I didn’t make different values relative to one’s role on a title team, though that may be an adjustment I make on a future version.

**Obviously some previous players are held back by some of these honors not being present in their careers (no defensive honors or FMVP until ‘69--->effects on Bill Russell, for example?; no MVP prior to ‘56--->effects on George Mikan? etc). These are simply considerations I have to hold in the back of my mind when scrutinizing their scores; and I have created hypothetical versions of them to see how the MIGHT have scored if these things had been available.

NOTE: Awards/honors are weighted against a “Positional Era Rating”......an evaluation of the strength of high-end competition at their position (C, F. or G), which I rated each and every year individually by position and overall (by a combination of objective and subjective means).

ALL OTHER FACTORS are weighted against the “Overall Era Rating”, which was also created by sort of adding up the high-level talent in the league year-by-year, though also giving a slight mathematical favoritism to leagues with a larger number of teams. My reason for doing so was assuming larger league represents an increasingly global popularity of the game and larger player pool to sample from, as well as more technically advanced (or “modern”) technique and style of play, and also noting tougher playoff structure (as compared to when six of eight teams in the league would make the playoffs, and a team may only have to win ONE series to make it into the finals).
Ultimately though----for most things which are weighted against the Overall Era Rating----I ended up using a variable to dilute the overall era ratings. I admit this was purely a “massaging the results” adjustment, as the formula initially seemed to otherwise under-credit older players. I somewhat justify the dilution noting that what talent was present was more concentrated in those earlier years. This is perhaps relevant when noting, for instance, that someone like Bill Russell or Wilt Chamberlain didn’t have the “luxury” of---literally every other game---facing opposing centers who were roughly the quality of Len Elmore, Kent Benson, Herb Williams, James Edwards, Dave Corzine, Steve Johnson, Wayne Cooper, Jerome Whitehead, or Tree Rollins (great rim protector, but offensive zero).......such as Moses Malone was facing every other game in the early 80’s.

In summary, the formula (“original” version) derives the largest portions of a player’s score from rs PER weighted against mpg, seasons played, and win% (and diluted era strength rating); MVP award shares (where applicable, and weighted against era strength); and playoff PER weighted against mpg, playoff games played, playoff win%, and diluted era strength rating.
Award points is the next most highly weighted factor (weighted directly against positional era rating), followed closely by “title points” (weighted against diluted era strength rating), and then followed by some cumulative numbers (rs pts/reb/ast, rs WS, playoff WS…….these to better reward meaningful longevity, though the WS also bring in a little more accounting of efficiency), rs and playoff WS/48 (both weighted against mpg and diluted era strength), and a peak rating (as measured by the highest PER*WS/48*mpg product). The least weighted factors are the with/without records, DRtg (individual DRtg can be a bit sketchy, especially when using estimates for pre-1974), and ppg numbers.


I made a second version (titled “Version 16.8.1”) which has the following major alterations from the original:
1) MUCH less rewarding to longevity.
2) PER and WS/48 are accounted only in their relation to a “replacement level player” (in the “original” version a PER of 5.0, for example, would still yield some degree of positive score; in this version it effects the score up or down based on how much above or below replacement level it is). Can outline what “replacement level” parameters I used; Owly helped come up with the figures.
…..And the following minor alterations:
3) Slightly less rs winner’s bias (less accounting for rs win%).
4) Slightly more consideration toward efficiency (by way of greater weight for WS/48); NOTE: PER still weighted more heavily, though.
5) Slight more emphasis toward big-man defenders (recognizing their often huge defensive value as rim protectors (used rebounds as a proxy for this).
6) Slightly less ppg emphasis.


Then I made a third version (“Version 16.8.2”) which has the following major alterations from the original:
1) Much MUCH more consideration toward efficiency (via MUCH heavier weighting for WS/48, now weighted heavier than PER).
2) Much more “Title points” emphasis (really isn’t TOO huge a contributing factor in first two versions; I made it fairly prominent in this one, though).
3) PER and WS/48 are accounted for by their relation to a replacement level player (even better accounting of this than the 16.8.1 version).
4) Significantly greater emphasis on big-man defenders (using rebounds as proxy to even greater degree than in version 16.8.1; also greater weight given to DRtg).
…...and the following minor alterations:
5) Little more emphasis on longevity.
6) Slightly more award/honors emphasis
7) Slightly less MVP award shares emphasis.
8) Somewhat firmer accounting of era strength (i.e. not “diluting” my Overall Era Ratings as much).
9) Slightly less ppg emphasis.


ANYWAY (whew!)......I’ve run 321 players thru all three versions of this formula. And then I amalgamated the results/rankings of the three versions (hoping the strengths/flaws of each will balance each other), finding a player’s average rank among the three. Thought you might be interested in knowing who the top 150 are in this amalgamation (btw, this is NOT the order of my ATL, though it wouldn’t make a bad list imo, especially if you make small adjustments based on my notations below. Anyway, this is something I did for hobby/interest and use as merely one more tool):

Spoiler:
1. Michael Jordan
2. Kareem Abdul-Jabbar
3. Lebron James
**Lebron just barely behind: his average rank was 2.67 (finished 2nd in one version, 3rd in the other two), vs. 2.33 for Kareem (2nd in two versions, 3rd in the other).
4. Wilt Chamberlain
5. Tim Duncan
**Duncan barely behind with average rank of 4.67 vs. the slightly higher 4.33 for Wilt.
6. Shaquille O’Neal
7. Karl Malone
8. Kobe Bryant
9. Magic Johnson
**Kobe avg 8.33 rank to Magic’s 8.67; though I’d note this is with Kobe’s obviously undeserved All-Star nominations in ‘14, ‘15, and ‘16, as well as at least one or two undeserved All-D honors (which I’m obligated to include them all); take those away, however, and their positions would flip-flop (Kobe #9, Magic #8).
10. Larry Bird
11. Hakeem Olajuwon
t12. Dirk Nowitzki
t12. David Robinson
**tied in average rank; Nowitzki narrowly wins tie-breaker by having higher cumulative score (if add up score of all three versions)
t14. Kevin Garnett
t14. Bill Russell
**Tied in average rank; Garnett narrowly wins tie-breaker based on cumulative score. I’m not surprised or discouraged that Russell is placed this low. If stat-keeping had been more complete in his day (namely blocks), I suspect both his PER and WS/48 would be a little higher; and an actual DRtg is likely significantly better than the conservative estimate I’ve given him. Also, were defensive honors and FMVP awarded throughout his career, that too would change things. I’ve pumped a hypothetical Bill Russell (estimating the changes to PER, WS/48, gave a little more liberal DRtg estimation, as well as what defensive honors and FMVP’s he would have won) in the formulas…..this hypothetical would have the 10th-highest average rank (very nearly 9th). And this is before considering Russell’s other “intangible” values.
16. Julius Erving
17. Charles Barkley
18. Jerry West
19. Oscar Robertson
t20. Moses Malone
t20. John Stockton
**Stockton narrowly wins the tie-breaker based on cumulative points.
22. Dwyane Wade
23. Chris Paul
24. Bob Pettit
25. Kevin Durant
26. Artis Gilmore
27. Scottie Pippen
28. Elgin Baylor
29. Patrick Ewing
30. Clyde Drexler
31. Jason Kidd
32. Steve Nash
33. Paul Pierce
34. John Havlicek
35. Rick Barry
36. Dolph Schayes
37. Dan Issel
38. Dwight Howard
39. Robert Parish
40. Pau Gasol
41. Gary Payton
42. Reggie Miller
43. Kevin McHale
44. Allen Iverson
45. Elvin Hayes
46. Chauncey Billups
47. Ray Allen
48. George Mikan
49. Stephen Curry
50. Walt Frazier
51. Tracy McGrady
52. George Gervin
53. Manu Ginobili
54. Tony Parker
55. Isiah Thomas
56. Bob Lanier
57. Adrian Dantley
58. Bob Cousy
59. Dominique Wilkins
60. Russell Westbrook
61. Vince Carter
t62. Horace Grant
t62. Shawn Marion
**tied avg rank; Grant wins tie-breaker of cumulative score.
t64. Chris Bosh
t64. Bob McAdoo
**tied avg rank; Bosh wins tie-breaker.
66. Alonzo Mourning
67. Walt Bellamy
68. Amar’e Stoudemire
69. Carmelo Anthony
70. Kevin Johnson
71. Chris Webber
72. Wes Unseld
73. Dikembe Mutombo
74. Sam Jones
75. James Worthy
76. James Harden
77. Dave Cowens
78. Zelmo Beaty (he’s another that might be worth putting in your formula)
79. Alex English
80. Jack Sikma
81. Shawn Kemp
82. Paul Arizin
83. Elton Brand
84. George McGinnis
85. Grant Hill
t86. Larry Nance
t86. Terry Porter
**Nance wins tie-breaker by negligible margin.
88. Rasheed Wallace
89. Willis Reed
90. Ben Wallace
t91. Jeff Hornacek
t91. Chet Walker
**Hornacek wins tie-breaker.
93. Jerry Lucas
94. Maurice Cheeks
95. Sidney Moncrief
t96. Bailey Howell
t96. Marques Johnson
**Howell wins tie-breaker.
98. Neil Johnston
99. Bobby Jones
t100. Bill Laimbeer
t100. Terry Cummings
**Laimbeer wins tie-breaker.
102. Mel Daniels
103. Sam Cassell
104. Billy Cunningham
t105. Cliff Hagan
t105. Vlade Divac
**Hagan narrowly wins the tie-breaker.
107. Connie Hawkins
t108. Spencer Haywood
t108. Hal Greer
**Haywood narrowly wins tie-breaker.
110. Tim Hardaway
111. Buck Williams
112. Detlef Schrempf
113. Bernard King
114. Anfernee Hardaway
115. Yao Ming
t116. Deron Williams
t116. Kawhi Leonard
**Kawhi very narrowly wins tie-breaker (though fwiw, I don’t think he deserved both DPOY awards; removing one would drop his rank a few spots)
t118. Dennis Johnson
t118. Chris Mullin
**Chris Mullin wins tie-breaker by negligible margin
120. LaMarcus Aldridge
121. Carlos Boozer
122. Dennis Rodman
t123. Bob Dandridge
t123. Gus Williams
**Dandridge very narrowly wins the tie-breaker based on cumulative score.
125. Nate Thurmond
126. Bill Sharman
127. Tiny Archibald
128. Mark Jackson
129. Zach Randolph
130. Walter Davis
131. Blake Griffin
132. Andre Miller
133. Kevin Love
134. Eddie Jones
135. Vern Mikkelsen
136. Lamar Odom
t137. Joe Dumars
t137. Lenny Wilkens
**Joe Dumars narrowly wins the tie-breaker. Note Dumars is drastically underrated by individual DRtg, one of many examples pertaining to my reluctance to put a lot of weight on that metric.
139. Tom Heinsohn
140. Ed Macauley
141. Mark Aguirre
142. Rod Strickland
143. Jason Terry
t144. Marcus Camby
t144. Sam Perkins
**Sam Perkins wins the tie-breaker by a slim margin.
146. Brad Daugherty
t147. Baron Davis
t147. Larry Foust
**Larry Foust wins the tie-breaker
149. Jimmy Jones
150. David West


It's a very solid list.

With such diference between eras that's very difficult to do. Congrats on the project.
“These guys have been criticized the last few years for not getting to where we’re going, but I’ve always said that the most important thing in sports is to keep trying. Let this be an example of what it means to say it’s never over.” - Jerry Sloan
Lucky707
Sophomore
Posts: 100
And1: 59
Joined: Jun 09, 2012

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#69 » by Lucky707 » Sat Aug 13, 2016 11:00 pm

2klegend wrote:I'm modifying my Award calculation to account for All Star, ALL NBA selection. The rationale for All NBA 1st = 2pts and AS = 1pt is simple. There are only 5 can be chosen, instead of 12 for AS selection. So 2nd All NBA receive similar chance to get AS selection.

Title as the Main Focus (best player) = 10pts
Title as the compliment star = 5pts
Title as support role = 2.5
MVP = 10pts
Final MVP = 5pts
All Star = 1pt
All NBA 1st Team = 2pts
All NBA 2nd Team = 1pt
All NBA 1st Defensive Team = 2pts
All NBA 2nd Defensive Team = 1pt

With the change, Lebron has surpass Bill Russell. Kobe and Hakeem are really close but once award were factor in, Kobe got in the Top 10 ahead of Hakeem which I think is correct. Ginobili is also dropping from 21th to 30th. Nash is unchanged despite the new addition, still at 51th. Though, the lowest in the 150 is Spencer Haywood, instead of Tiny Archibald whose moved up from 150th to 149th.


Why are you punishing Russell for not having awards he could not have possibly won?
User avatar
2klegend
Bench Warmer
Posts: 1,333
And1: 409
Joined: Mar 31, 2016
     

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#70 » by 2klegend » Sun Aug 14, 2016 12:49 am

Lucky707 wrote:
2klegend wrote:I'm modifying my Award calculation to account for All Star, ALL NBA selection. The rationale for All NBA 1st = 2pts and AS = 1pt is simple. There are only 5 can be chosen, instead of 12 for AS selection. So 2nd All NBA receive similar chance to get AS selection.

Title as the Main Focus (best player) = 10pts
Title as the compliment star = 5pts
Title as support role = 2.5
MVP = 10pts
Final MVP = 5pts
All Star = 1pt
All NBA 1st Team = 2pts
All NBA 2nd Team = 1pt
All NBA 1st Defensive Team = 2pts
All NBA 2nd Defensive Team = 1pt

With the change, Lebron has surpass Bill Russell. Kobe and Hakeem are really close but once award were factor in, Kobe got in the Top 10 ahead of Hakeem which I think is correct. Ginobili is also dropping from 21th to 30th. Nash is unchanged despite the new addition, still at 51th. Though, the lowest in the 150 is Spencer Haywood, instead of Tiny Archibald whose moved up from 150th to 149th.


Why are you punishing Russell for not having awards he could not have possibly won?

The only award he could have won is Final MVP and All Defensive 1st Team up to 1969. Consider he got the easy path to the NBA Final, playing just 2 series, I think it balance out.
My Top 100+ GOAT (Peak, Prime, Longevity, Award):
viewtopic.php?f=64&t=1464952
JordansBulls
RealGM
Posts: 60,466
And1: 5,344
Joined: Jul 12, 2006
Location: HCA (Homecourt Advantage)

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#71 » by JordansBulls » Sun Aug 14, 2016 5:54 am

Lucky707 wrote:
2klegend wrote:I'm modifying my Award calculation to account for All Star, ALL NBA selection. The rationale for All NBA 1st = 2pts and AS = 1pt is simple. There are only 5 can be chosen, instead of 12 for AS selection. So 2nd All NBA receive similar chance to get AS selection.

Title as the Main Focus (best player) = 10pts
Title as the compliment star = 5pts
Title as support role = 2.5
MVP = 10pts
Final MVP = 5pts
All Star = 1pt
All NBA 1st Team = 2pts
All NBA 2nd Team = 1pt
All NBA 1st Defensive Team = 2pts
All NBA 2nd Defensive Team = 1pt

With the change, Lebron has surpass Bill Russell. Kobe and Hakeem are really close but once award were factor in, Kobe got in the Top 10 ahead of Hakeem which I think is correct. Ginobili is also dropping from 21th to 30th. Nash is unchanged despite the new addition, still at 51th. Though, the lowest in the 150 is Spencer Haywood, instead of Tiny Archibald whose moved up from 150th to 149th.


Why are you punishing Russell for not having awards he could not have possibly won?

It balances out because no one else had only 2 series to win for a title. And that for 8 of his.
Image
"Talent wins games, but teamwork and intelligence wins championships."
- Michael Jordan
Jboogz23
Sophomore
Posts: 231
And1: 163
Joined: May 12, 2013
 

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#72 » by Jboogz23 » Sat Feb 4, 2017 6:39 am

I utilized 2k's formula and applied it to a comparison of MJ and LBJ. Note, this is all for purely statistical purposes. I did NOT use any of the post season coefficients. Really enjoyed doing this so thanks 2K. My numbers may be a bit similar/off compared to your chart. I rounded 2 decimal places.


Code: Select all

   Year   Reg. Season   Playoffs   Peak (Combined)
Michael Jordan   
   1985 (1)   115.06      113.82      114.44
   1986 (2)   90.62      125.19      107.91
   1987 (3)   122.02      119.06      120.54
   1988 (4)   149.58      134.20      141.89
   1989 (5)   149.92      148.00      148.96
   1990 (6)   137.86      156.34      147.10
   1991 (7)   144.58      160.47      152.53
   1992 (8)   125.32      121.53      123.43
   1993 (9)   129.12      135.76      132.44
   1995 (10)   77.89      100.00      88.95
   1996 (11)   131.59      122.45      127.02
   1997 (12)   115.86      118.38      117.12
   1998 (13)   97.64      109.40      103.52
   2002 (14)   62.82      N/A         62.82
   2003 (15)   59.47      N/A         59.47


Career Total    
         1,706.35 (113.76)     1,664.60 (128.05)     1,748.14 (120.91)


First 13 Years   
         1,584.06 (121.85)   1,664.60 (128.05)   1,625.85 (124.95)


Peak (1991)   
         144.58      160.47      152.53

1st 3-Year Prime (88-90) Avg.
         146.27      147.69      146.98

2nd 3-Year Prime (91-93) Avg.
         133.03      137.63      135.33

3rd 3-Year Prime (96-98) Avg.
         114.70      116.28      115.49

5-Year Prime (88-92) Avg.
         142.07      142.97      142.52

7-Year Prime (88-93, 96) Avg.
         138.28      139.82      139.05

10-Year Prime (87-93, 96-98) Avg.
         130.35      132.76      131.56

Career Avg.    
         119.42      128.73      124.08




Code: Select all

   Year   Reg. Season   Playoffs   Peak (Combined)
LeBron James   
   2004 (1)   59.50      N/A         59.50
   2005 (2)   112.96      N/A         112.96
   2006 (3)   122.89      99.36      111.13
   2007 (4)   108.65      110.36      109.51
   2008 (5)   134.53      124.31      129.42
   2009 (6)   155.74      160.68      158.21
   2010 (7)   150.37      137.64      144.01
   2011 (8)   122.88      110.28      116.58
   2012 (9)   143.66      140.02      141.84
   2013 (10)   150.95      132.78      141.87
   2014 (11)   128.72      139.97      134.35
   2015 (12)   109.34      121.97      115.66
   2016 (13)   125.32      152.55      138.94
   2017 (14)   116.83      


Career Total 
         1,742.34 (124.45)   1,429.92 (130.00)      1,616.42 (127.13)


First 13 Years   
         1,625.51 (125.04)   1,429.92 (130.00)   1,613.98 (127.52)


Peak (2009)   
         155.74      160.68      158.21

1st 3-Year Prime (08-10) Avg.
         146.91      153.18      150.04

2nd 3-Year Prime (12-14) Avg.
         140.71      138.45      139.58

3rd 3-Year Prime (15-17) Avg.
         117.51      136.52      127.01

5-Year Prime (09-13) Avg.
         144.39      138.84      141.62

7-Year Prime (08-14) Avg.   
         140.95      137.57      139.26

10-Year Prime (08-17) Avg.
         134.42      137.55      135.99

Career Avg.   
         124.19      131.83      128.01
Lucky707
Sophomore
Posts: 100
And1: 59
Joined: Jun 09, 2012

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#73 » by Lucky707 » Sun Oct 8, 2017 7:08 am

Any chance of updating this for 2017? I'm honestly finding that the RealGM list after the Top 30-35 or so just starts to completely lose direction compared to both this list and ESPN's.
Jboogz23
Sophomore
Posts: 231
And1: 163
Joined: May 12, 2013
 

Re: Top 150+ GOAT (Peak, Prime, Longevity, Award) 

Post#74 » by Jboogz23 » Sun Oct 8, 2017 8:01 am

Crazy how similar this top 10 is to the Realgms top 10

Return to Player Comparisons