Let's rank the Top 11 GOATs by PER

[dubeta]

Ok so this thread will try and see if statistics is a better indicator of ranking the GOATs. The stat is Player Efficiency Rating, a stat which gauges true player skill and contribution to winning.

Generally the higher the career PER, the more impactful the player was to helping his team win, ie. the BETTER the player he was

On ISH, the consensus TOP 11 are (Even though I don't generally agree with all of them) In no order

Jordan
Kareem
Bird
Magic
Shaq
Kobe
LeBron
Hakeem
Duncan
Russell
Wilt

So lets rank them by PER and see if the rankings make sense

1. Michael Jordan 27.91
2. LeBron James 27.79
3. Shaquille O'Neal 26.43
4. Wilt Chamberlain 26.13
5. Kareem Abdul-Jabbar 24.58
6. Tim Duncan 24.56
7. Magic Johnson 24.11
8. Hakeem Olajuwon 23.59
9. Larry Bird 23.50
10. Kobe Bryant 23.36
11. Bill Russell 18.87

Obviously there are some outliers, as BLOCKS werent recorded in Wilt and Russell's day, meaning their PER may have been underrated

But this seems to be a good indicator to judge players from 1970's- to now

I think these rankings generally make sense, if you disagree with a particular player's ranking, please argue why you feel there is a discrepancy between their PER and what you feel their actual ranking should be

Please no trolling, lets have an intellectual basketball debate

[fpliii]

PER doesn't exist before 1977-78. You need all the stats in a box score to calculate it, and missing stats for seasons prior to 77-78 are zeroed out:


The calcuation of uPER obviously depends on these statistics, so here are my solutions for years when the data are missing:

Zero out three-point field goals, turnovers, blocked shots, and steals.
Set the league value of possession (VOP) equal to 1.
Set the defensive rebound percentage (DRB%) equal to 0.7.
Set player offensive rebounds (ORB) equal to 0.3 * TRB.

source: http://www.basketball-reference.com/about/per.html

[dubeta]

PER doesn't exist before 1977-78. You need all the stats in a box score to calculate it, and missing stats for seasons prior to 77-78 are zeroed out:



source: http://www.basketball-reference.com/about/per.html

I've addressed this in my post, thats why Wilt and Russells may be lower than actual

[fpliii]

I've addressed this in my post, thats why Wilt and Russells may be lower than actual
It hurts Kareem too, since it cuts out most of his prime (his peak year is 76-77). LeBron also hasn't entered his decline phase yet, so his PER will go down.

[dubeta]

It hurts Kareem too, since it cuts out most of his prime (his peak year is 76-77). LeBron also hasn't entered his decline phase yet, so his PER will go down.

I doubt LeBron will go team jumping at the age of 40 like Shaq, or play at 40 like MJ to drastically reduce his PER

LeBron has 2-3 years to INCREASE his PER, and about 2-3 years where it will DECREASE after

So when he retires I think it will be the same it is now

[Sarcastic]

Neil Johnston - 24.69 PER, 9th best all time

Bill Russell - 18.87 PER, 101st all time

[Dragic4Life]

Good thread. Really shows who's the best players in NBA history by discounting team accolades.

[ArbitraryWater]

PER doesn't measure defense though, as illustrated by Sarcastic's post about Bill Russell.

Now how the fvk is Neil Johnston up there?

[Nikola_]

Inb4 Ayotunde Ndiaye whit his PER.

[dubeta]

Theres a reason why I only included the consensus TOP 11 GOATS, and not just rank the Top 11 PERs

Because I want to rank the players that are seen together as the ultimate

Its the same reason Jordan and Kobe fans whine when someone brings in Horry's 7/7 because Horry is tiers below

[stalkerforlife]

Who's PER? Is he good? Can the Magic get him?

[Lebronxrings]

:applause: great thread.

This is well made and has no bias in it (like other threads)

These are facts which you cannot dispute. Whether you hate these players or not, their impact doesn't change and this is an extremely valid contributer towards judging a player. Honestly, this is how players should be judged, by impact rather then rings/petty accomplisments.

[dubeta]

:applause: great thread.

This is well made and has no bias in it (like other threads)

These are facts which you cannot dispute. Whether you hate these players or not, their impact doesn't change and this is an extremely valid contributer towards judging a player. Honestly, this is how players should be judged, by impact rather then rings/petty accomplisments.

:cheers:

Nothing quite like objective discussion

[GrapeApe]

Theres a reason why I only included the consensus TOP 11 GOATS, and not just rank the Top 11 PERs

Because I want to rank the players that are seen together as the ultimate

Its the same reason Jordan and Kobe fans whine when someone brings in Horry's 7/7 because Horry is tiers below

You could conceivably be 7/7 in the finals without having played a minute in any series. This is why it's a stupid argument (well one of the reasons anyway). PER, however, is a statistic based on actual play. By arbitrarily selecting and omitting players in which to compare you are effectively saying that PER is a flawed statistic, and if so what is the point of using it?

[Straight_Ballin]

PER is flawed in that it doesn't take choking in the finals into account. If you have a great 82 game season with good PER, what does it matter when you lose to guys like Leonard and Jason Terry?

It was a good attempt to try and prop Bron up and overlook his failures in the finals which move him outside of the top 10....but such logic has just been exploited.

[dubeta]

You could conceivably be 7/7 in the finals without having played a minute in any series. This is why it's a stupid argument (well one of the reasons anyway). PER, however, is a statistic based on actual play. By arbitrarily selecting and omitting players in which to compare you are effectively saying that PER is a flawed statistic, and if so what is the point of using it?

Ok but the finals argument is still flawed when comparing similar players

Wilt is above kobe in most people's eyes and wilt is 2/6 to kobe's 5/7

So should we scrap that argument as well?

[BlkMambaGOAT]

So lets rank them by PER and see if the rankings make sense

1. Michael Jordan 27.91
2. LeBron James 27.79
3. Shaquille O'Neal 26.43
4. David Robinson 26.18[/B]
5. Wilt Chamberlain 26.13
6. Chris Paul 25.59
7. Bob Petit 25.53
8. Dwayne Wade 25.29
9. Neil Johnston 24.69
10. Charles Barkley 24.63
[/QUOTE]
Fixed oh and lol no matter way the Bran stans try to spin it he still looks like a shit player whose rings are asterisks.:lol :lol :lol :lol Same team with a top-10 all-time player(Wade) but only going 2*/4, I am not impressed:sleeping :sleeping :sleeping


Rankings based purely on PER are 4 phaggits.
Bonus: Durant would rank at #13

[GODbe]

Neil Johnson 6.9ppg career :
Top 9 all time on PER

GOAT gonna GOAT:eek:

[BlkMambaGOAT]

Neil Johnson 6.9ppg career :
Top 9 all time on PER

GOAT gonna GOAT:eek:
MDE, Shaq is secretly jealous of him.

[dubeta]

So lets rank them by PER and see if the rankings make sense

1. Michael Jordan 27.91
2. LeBron James 27.79
3. Shaquille O'Neal 26.43
4. David Robinson 26.18[/B]
5. Wilt Chamberlain 26.13
6. Chris Paul 25.59
7. Bob Petit 25.53
8. Dwayne Wade 25.29
9. Neil Johnston 24.69
10. Charles Barkley 24.63


Wow so Wade, Durant, Barkley, CP3 are all better than Kobe? :applause:

numbers never lie your boy isnt a GOAT

[dubeta]

I try and prop up Kobe objectively, by removing the Non-GOATS, but the stans are bringing in other players

LOL you kobetards are throwing your own idol under the bus

Keep it going :oldlol: :applause:

[GrapeApe]

Ok but the finals argument is still flawed when comparing similar players

Wilt is above kobe in most people's eyes and wilt is 2/6 to kobe's 5/7

So should we scrap that argument as well?

You're missing my point. I'm saying that dismissing 7/7 is not comparable to dismissing PER. You praised PER in the OP as an accurate measure of a player's greatness. Why arbitrarily omit high ranking PER's if it is indeed such a great statistic?

[BlkMambaGOAT]

Fixed oh and lol no matter way the Bran stans try to spin it he still looks like a shit player whose rings are asterisks.:lol :lol :lol :lol Same team with a top-10 all-time player(Wade) but only going 2*/4, I am not impressed:sleeping :sleeping :sleeping


Rankings based purely on PER are 4 phaggits.
Bonus: Durant would rank at #13
^^^^^^^^^

You're right, Neil Johnston is the true GOAT, I have been convinced. Forgive me prophet for my blindless.:bowdown:

I think Lebron's **** is disrupting your cerebrum's connection to your spinal cord.

[dubeta]

You're missing my point. I'm saying that dismissing 7/7 is not comparable to dismissing PER. You praised PER in the OP as an accurate measure of a player's greatness. Why arbitrarily omit high ranking PER's if it is indeed such a great statistic?

Same reason people omit Horrys 7/7 and Parker and Ginobli's 4/5 Norris Cole's 2/3

Using it to compare players similar in ability

[G.O.A.T]

But this seems to be a good indicator to judge players from 1970's- to now



I like PER. I think it's a useful stat, but as a measure of great players it has no place. If you look at each players PER by season, you will see that their highest PER is typically in seasons where they had the least team success. Usually this means that player had to carry too much burden, thus the higher PER. This is the same problem that persists with per game or per 36 averages. They do not account for plays that do not show up in the stat sheet. And when teams have success, there is a multitude of those type of plays.


Neil Johnston - 24.69 PER, 9th best all time

Bill Russell - 18.87 PER, 101st all time

And there is a famous story that the arrival of the later ended the former's career.

[Pointguard]

Point guards are punished by PER. They usually lead the team in turnovers and if they are doing their job right they aren't the lead scorer on the team. There are always more rebounds per game than assist per game, and more blocks than steals per game if you go by leaders. They don't get as many free throws and because they bail the team out their FG% suffers. Nash wasn't in the top ten in either of his MVP runs and was verygood with FT% and FG%.

Chris Paul, especially when he has played too conservative. Both Curry and Westbrook are equal players but don't have his PER. Westbrook definitely outplayed him but I bet Paul had the better PER. Its a tricky thing for a PG. Basically, I'm saying it was harder for Magic than anybody on that list.

[GrapeApe]

Same reason people omit Horrys 7/7 and Parker and Ginobli's 4/5 Norris Cole's 2/3

Using it to compare players similar in ability

You cannot be that dense. You said YOURSELF in the OP that PER is the most accurate measure of a player's greatness (or something to that effect). This has nothing to do with team accomplishments or accolades. This about an individual's playing ability based on a statistic in which YOU cited as accurate. By that logic the top 10 PER's should reflect the 10 best players. Once again you are contradicting yourself.

BTW, I don't think the top 10 PER's are indeed the 10 GOAT. I'm just pointing out OP's usual inconsistencies.

[dubeta]

You cannot be that dense. You said YOURSELF in the OP that PER is the most accurate measure of a player's greatness (or something to that effect). This has nothing to do with team accomplishments or accolades. This about an individual's playing ability based on a statistic in which YOU cited as accurate. By that logic the top 10 PER's should reflect the 10 best players. Once again you are contradicting yourself.

Ok fine

The top 10 PER ratings show the 10 best players in terms of impacting team success throughout their careers (since it is CAREER PER)

Other subjective factors like rings impact other peoples GOAT lists

there, I stand behind the PER principle

Whats the problem?

[GrapeApe]

Ok fine

The top 10 PER ratings show the 10 best players in terms of impacting team success throughout their careers (since it is CAREER PER)

Other subjective factors like rings impact other peoples GOAT lists

there, I stand behind the PER principle

Whats the problem?

No problem here. You just admitted Dwyane Wade is one of the 10 best players in terms of impacting his team's success throughout his career. :cheers:

[knickswoman]

LOL at these nerds

[stalkerforlife]

LOL at these nerds

Daaaaaaaaaaaaamn, a woman owning the OP.

:lebronamazed:

[riseagainst]

Friends of ISH!

As you all know, InsideHoops is the mecca for intelligent basketball discussion. I enjoy this venue because the posters here use advanced statistical models to look "beyond the boxscore". Indeed, there is so much knowledge in literally every thread posted here.

If I may, I would like to make a humble contribution, something I've been working on for many years now in my spare time. It is a very complicated mathematical formula that I believe encapsulates the most important quantities in basketball. I don't want to go into a huge explanation of how I arrived at this, but you can rest assured that the math is solid. The formula is:

SHOTS = FGA + FTA

or, since it is often more useful to see this PER game,

PER = (FGA + FTA) / #GAMES

As you all know, the objective of the game is to put up shots, so I am surprised that my revolutionary formula is not in the common nomenclature. Let us make it so!

This formula passes the "common sense" test as the greatest player in NBA history, Kobe Bean Bryant, routinely leads the league in this category. With Kobe out this year due to injury, we have Kevin Durant leading. From what I understand, he is the leading candidate for the MVP award (I don't know since I stopped watching the game the moment Kobe stopped playing), so again it looks solid.

Let us look at last year's PER for the top MVP candidates:
LeBron: 24.8
Durant: 26.1
Kobe: 28.4

Look at "old" Kobe schooling these youngsters in their prime.

And look at their career highs:
LeBron: 33.4
Durant: 29.9
Kobe: 37.4

Not even close!

I hope this helps illustrate Kobe's superiority to his so-called "competition"; this is especially important for you youngsters who started watching basketball this year and haven't had the fortune of watching the NBA PER leader. But no worries, he'll be back and better than ever next year!

Up next: I will add salary to the equation to show why Kobe is even more productive than these guys. SHOTS + $$ is pretty much what it's all about.

Warmest regards,

Ayotunde Ndiaye

[EDIT!] Due to popular demand, I have submitted to Adam Silver a proposal to make PER a part of the official NBA awards, to hopefully replace the meaningless MVP. I have even designed a trophy and an infographic to show how this might look:

http://oi57.tinypic.com/28qsr9x.jpg

http://oi62.tinypic.com/2chmc8x.jpg

http://www.insidehoops.com/forum/showthread.php?t=324201

:roll:
:roll:
:roll:

[Y2ktors]

PER doesn't exist before 1977-78. You need all the stats in a box score to calculate it, and missing stats for seasons prior to 77-78 are zeroed out:



source: http://www.basketball-reference.com/about/per.html
Indeed.

[kennethgriffin]

PER IS RETARDED


1. Michael Jordan* 27.91
2. LeBron James 27.79
3. Shaquille O'Neal 26.43
4. David Robinson* 26.18
5. Wilt Chamberlain* 26.13
6. Chris Paul 25.59
7. Bob Pettit* 25.35
8. Dwyane Wade 25.29
9. Neil Johnston* 24.69
10. Charles Barkley* 24.63
11. Kareem Abdul-Jabbar* 24.58
12. Tim Duncan 24.56
13. Kevin Durant 24.53
14. Magic Johnson* 24.11
15. Karl Malone* 23.90
16. Hakeem Olajuwon* 23.59
17. Larry Bird* 23.50
18. Dirk Nowitzki 23.48
19. Kobe Bryant 23.36
20. Oscar Robertson* 23.17
21. Yao Ming 23.02
22. Kevin Garnett 22.92
23. Jerry West* 22.89
24. Elgin Baylor* 22.69
25. Moses Malone* 22.31
26. Tracy McGrady 22.13
27. Dwight Howard 22.09
28. Amar'e Stoudemire 22.03
29. Dolph Schayes* 21.98
30. Julius Erving* 21.97
31. John Stockton* 21.83
32. George Gervin* 21.74
33. Bob Lanier* 21.69
34. Dominique Wilkins* 21.56
35. Clyde Lovellette* 21.55
36. Adrian Dantley* 21.51
37. Pau Gasol 21.49
38. Harry Gallatin* 21.48
39. Manu Ginobili 21.46
40. Alonzo Mourning* 21.24
41. Carmelo Anthony 21.17
42. Clyde Drexler* 21.07
43. Patrick Ewing* 21.01
44. Dan Issel* 20.99
45. Chris Webber 20.94
46. Allen Iverson 20.92
47. Al Jefferson 20.84
48. John Drew 20.74
49. Bob McAdoo* 20.73
50. Elton Brand 20.72
51. Kevin Johnson 20.70
52. Chris Bosh 20.62
53. George Yardley* 20.51
54. Ed Macauley* 20.39
55. Paul Pierce 20.36
56. Larry Foust 20.19
57. Rick Barry* 20.16
58. Artis Gilmore* 20.16
59. Marques Johnson 20.11
60. Vince Carter 20.03
61. Kevin McHale* 20.02
62. LaMarcus Aldridge 19.98
63. Steve Nash 19.95
64. Larry Nance 19.92
65. Alex English* 19.87
66. Walt Bellamy* 19.84
67. Bob Cousy* 19.76
68. Terrell Brandon 19.69
69. David Thompson* 19.68
70. Carlos Boozer 19.66
71. Paul Arizin* 19.66
72. Mark Price 19.62
73. Gilbert Arenas 19.57
74. Zach Randolph 19.56
75. Cliff Hagan* 19.50
76. Sam Cassell 19.48
77. Michael Redd 19.48
78. Paul Westphal 19.43
79. Billy Cunningham* 19.38
80. Robert Parish* 19.22
81. David Lee 19.22
82. Bernard King* 19.18
83. Walt Frazier* 19.12
84. Deron Williams 19.12
85. Bailey Howell* 19.11
86. Tony Parker 19.08
87. Shawn Kemp 19.08
88. Walter Davis 19.07
89. Shawn Marion 19.03
90. Grant Hill 19.03
91. Mark Aguirre 19.02
92. Shareef Abdur-Rahim 19.02
93. Paul Millsap 18.92
94. Gary Payton* 18.88
95. Bill Russell* 18.87
96. Jerry Lucas* 18.86
97. David West 18.86
98. Brad Daugherty 18.85
99. Chauncey Billups 18.82
100. World B. Free 18.78
101. Chris Mullin* 18.78
102. Andrei Kirilenko 18.75
103. Stephon Marbury 18.74
104. Sidney Moncrief 18.69
105. Sam Jones* 18.68
106. Scottie Pippen* 18.63
107. Tim Hardaway 18.61
108. Vern Mikkelsen* 18.60
109. Willis Reed* 18.57
110. Ray Allen 18.56
111. George McGinnis 18.55
112. Gus Williams 18.51
113. Zydrunas Ilgauskas 18.49
114. Pete Maravich* 18.38
115. Kiki Vandeweghe 18.37
116. Reggie Miller* 18.36
117. Steve Francis 18.35
118. Alvan Adams 18.32
119. Terry Cummings 18.29
120. Bill Sharman* 18.21
121. Antawn Jamison 18.14
122. Isiah Thomas* 18.11

[oarabbus]

PER doesn't exist before 1977-78. You need all the stats in a box score to calculate it, and missing stats for seasons prior to 77-78 are zeroed out:



source: http://www.basketball-reference.com/about/per.html


What if you calculate the PER statistics assuming

1) They had league average steals and blocks
2) They were in the 95th percentile all time steals and blocks
3) A best case scenario i.e. >4 or 5 blocks/game

[JT123]

Daaaaaaaaaaaaamn, a woman owning the OP.

:lebronamazed:
Not a woman, just a transgender fggt who used to go by the name knicksman.

[Asukal]

dubeta trying to grow a brain still an idi0t. :oldlol:

[fpliii]

What if you calculate the PER statistics assuming

1) They had league average steals and blocks
2) They were in the 95th percentile all time steals and blocks
3) A best case scenario i.e. >4 or 5 blocks/game
Not sure. PER is calculated based on league averages in a given year. First uPER is calculated (unadjusted), and then after adjusting for pace, it's normalized so that the average player in the league in that particular season.

Just looking at the formula though:

source: http://www.basketball-reference.com/about/per.html

there's a factor of:

VOP * DRB% * BLK

in there. So I guess if you figure other guys weren't really blocking shots, you can just multiply that factor by the pace adjustment, and normalize by league average.

If we want a quick estimate, we can just look at guys who were in the league in both 72-73 and 73-74 who were at the top of the block leaderboard in 73-74 (first year after it became an official stat), and see how much their PERs changed.

So, the top 10 in BPG in 73-74, and the changes in their PERs:

Elmore Smith (4.9): 14.5->15.9 (ppg went down 6 though)
Kareem Abdul-Jabbar (3.5) 28.5->24.4 (ppg went down 3, rpg went down too)
Bob McAdoo (3.3) 16.8->24.7 (became a much better player in his second year though)
Elvin Hayes (3.0) 17.5->18.2 (went up by 3.6 rpg, fg% dropped though)
Bob Lanier (3.0) 22.1->23.9 (trb dropped a bit, but fg% went up)
Nate Thurmond (2.9) 17.3->15.0 (ppg, rpg dropped)
Gar Heard (2.8) 14.9->16.0 (ppg and rpg increased)
Sam Lacey (2.3) 13.8->17.5 (ppg, rpg, apg increased slightly)
Clifford Ray (2.2) 15.7->15.0 (ppg and rpg increased, but so did his minutes)
George Johnson (1.9) 15.9->14.4 (small dropoffs across the board)

Steals and offensive boards also were recorded for the first time in 73-74, so they had an effect.

Just eyeballing though, I'd say:

1) PER is unchanged, since right now they have league average as well.
2) In that case, they'd look something like the top 10 guys here. Taking into account all the other factors listed above for those guys, it looks like a difference of +1.5 or +2.0.
3) Looks like a difference of +3.0 or +4.0, if everything else is unchanged.

[wally_world]

I doubt LeBron will go team jumping at the age of 40 like Shaq, or play at 40 like MJ to drastically reduce his PER

LeBron has 2-3 years to INCREASE his PER, and about 2-3 years where it will DECREASE after

So when he retires I think it will be the same it is now

Already doing it in his prime :oldlol:

So what you're saying is, MJ and Shaq's PER should be higher if not for the fact they jumped teams or played till their bodies broke down? So shouldn't they be rated higher than they are? Or it just goes to show PER isn't a good measure at all?

[Psileas]

Not sure. PER is calculated based on league averages in a given year. First uPER is calculated (unadjusted), and then after adjusting for pace, it's normalized so that the average player in the league in that particular season.

Just looking at the formula though:

source: http://www.basketball-reference.com/about/per.html

there's a factor of:

VOP * DRB% * BLK

in there. So I guess if you figure other guys weren't really blocking shots, you can just multiply that factor by the pace adjustment, and normalize by league average.

If we want a quick estimate, we can just look at guys who were in the league in both 72-73 and 73-74 who were at the top of the block leaderboard in 73-74 (first year after it became an official stat), and see how much their PERs changed.

So, the top 10 in BPG in 73-74, and the changes in their PERs:

Elmore Smith (4.9): 14.5->15.9 (ppg went down 6 though)
Kareem Abdul-Jabbar (3.5) 28.5->24.4 (ppg went down 3, rpg went down too)
Bob McAdoo (3.3) 16.8->24.7 (became a much better player in his second year though)
Elvin Hayes (3.0) 17.5->18.2 (went up by 3.6 rpg, fg% dropped though)
Bob Lanier (3.0) 22.1->23.9 (trb dropped a bit, but fg% went up)
Nate Thurmond (2.9) 17.3->15.0 (ppg, rpg dropped)
Gar Heard (2.8) 14.9->16.0 (ppg and rpg increased)
Sam Lacey (2.3) 13.8->17.5 (ppg, rpg, apg increased slightly)
Clifford Ray (2.2) 15.7->15.0 (ppg and rpg increased, but so did his minutes)
George Johnson (1.9) 15.9->14.4 (small dropoffs across the board)

Steals and offensive boards also were recorded for the first time in 73-74, so they had an effect.

Just eyeballing though, I'd say:

1) PER is unchanged, since right now they have league average as well.
2) In that case, they'd look something like the top 10 guys here. Taking into account all the other factors listed above for those guys, it looks like a difference of +1.5 or +2.0.
3) Looks like a difference of +3.0 or +4.0, if everything else is unchanged.

In general, according to PER, the value of 1 turnover = 1 steal (in absolute terms, obviously), regardless of pace and 1 turnover is equal to around 0.7 blocks, with pace only affecting this ratio minimally. With this in mind, I guess that young Kareem's actual PER would rise by a little (say, by around 1), Wilt's by somewhat more (maybe 2-3) and Russell's by even more (maybe 3-5), due to lower ball possession, thus less TO's (and probably more steals, as well).