Kobe Bryant and All-Time Ranking of Players

[Mike G]

"value over playoff starter" which is +2.5 or so.

Looking at last season's playoff stats, players who got >20 mpg and >2.5 BPM number barely 2 per team.
The avg 5th-best 'starter' BPM thus defined was -0.4

The top 80 players in mpg (27+) had avg BPM = 1.6

[kmedved]

Dan Rosenheck at the Economist did a nice writeup on Kobe, looking at RAPM and BPM, noting the playoff splits: http://www.economist.com/blogs/gametheo ... basketball

[BasketDork]

So I borrowed Mike G.'s idea of averaging WS/48, and his formulas for PER W/48 ((PER - 5.75)/92.5) and BPM W/48 ((BPM + 5.14)/51.4), to come up with a W/48 average for all players with 15,000+ career MP, .125+ WS/48, 1.0+ BPM and 15.0+ PER, 130 players in all. I then took those Win Averages, multiplied them by the player's career MP and divided by 48, to come up with a Win Total. Here are the results: (*active)

Code: Select all

1. K.Abdul-Jabbar 257.19
1. K.Malone 230.95
2. M.Jordan 212.74
3. T.Duncan 199.43 *
4. K.Garnett 198.81 *
5. L.James 191.24 *
7. S.O'Neal 182.95
8. J.Stockton 182.60
9. C.Barkley 181.33
10. O.Robertson 180.85
10. D.Nowitzki 177.95 *
11. K.Bryant 177.40 *
12. H.Olajuwon 173.57
13. D.Robinson 170.21
14. R.Miller 157.74
15. J.Kidd 155.24
16. M.Malone 153.99
17. M.Johnson 153.20
18. L.Bird 151.91
19. J.West 151.31
19. P.Pierce 149.72 *
20. G.Payton 148.71
21. C.Drexler 144.14
22. R.Allen 143.96
23. B.Russell 142.20
23. R.Parish 136.64
24. S.Pippen 135.87
25. P.Ewing 127.96
26. J.Havlicek 127.31
26. V.Carter 126.77 *
27. C.Paul 126.45 *
28. S.Marion 126.22
29. P.Gasol 124.95 *
30. A.Dantley 123.23
31. D.Wilkins 122.84
32. Bob Lanier 121.46
32. D.Wade 120.09 *
33. S.Nash 117.22
34. A.Iverson 115.41
35. E.Brand 114.10
36. T.McGrady 110.37
37. J.Erving 110.11
38. Horace Grant 109.51
39. A.English 109.19
40. L.Nance 107.82
41. J.Sikma 107.60
42. Buck Williams 106.87
43. C.Billups 106.73
44. D.Mutumbo 106.69
45. G.Hill 105.20
46. J.Hornacek 104.16
47. D.Howard 102.89 *
48. A.Gilmore 101.92
49. T.Porter 101.35
50. V.Divac 101.16
51. R.Wallace 101.03
52. C.Webber 101.02
53. K.McHale 100.96
54. O.Thorpe 100.13
55. E.Jones 98.68
56. D.Schrempf 98.13
57. C.Mullin 97.09
58. M.Cheeks 95.90
59. C.Bosh 95.48 *
60. T.Parker 95.27 *
61. B.Laimbeer 94.30
62. S.Perkins 94.16
63. B.Wallace 93.85
64. M.Ginobili 93.65 *
65. C.Anthony 92.54 *
66. K.Durant 92.40 *
68. R.Lewis 90.49
69. T.Hardaway 89.45
70. Bob McAdoo 89.23
70. H.Hawkins 88.29
71. D.Cowens 87.03
71. S.Kemp 86.35
72. S.Cassell 86.27
73. A.Iguodala 85.63 *
74. G.Gervin 84.86
75. K.Johnson 84.63
76. M.Camby 84.38
77. A.Mourning 84.09
79. Alvan Adams 82.69
80. J.Worthy 82.13
81. A.Kirilenko 81.12
82. S.Moncrief 80.54
83. Marques Johnson 80.16
84. Dale Davis 77.78
85. C.Boozer 77.51
86. T.Chandler 76.37 *
87. Steve Smith 75.74
88. P.Stojakovic 74.72
89. D.Issel 74.29
90. D.West 73.16 *
91. Deron Williams 72.95 *
92. Gus Williams 71.54
93. Brad Miller 71.21
94. C.Maxwell 70.76
95. L.Johnson 69.72
96. L.Aldridge 69.01 *
97. Bobby Jones 68.69
98. D.Rivers 68.20
99. Hot Rod Williams 67.74
100. M.Price 67.46
101. Dan Roundfield 67.45
102. T.Brandon 66.52
103. John Drew 66.21
104. A.Hardaway 66.09
105. P.Millsap 66.07
106. B.Barry 65.86
107. R.Westbrook 65.15
108. David Lee 64.86
109. B.Daugherty 62.42
110. T.Kukoc 61.78
111. R.Barry 61.73
112. P.Westphal 61.65
113. J.Harden 61.49
114. N.Hilario 61.09
115. S.Curry 60.34
116. M.Gasol 59.41
117. Fred Brown 58.57
119. Y.Ming 58.36
120. G.Arenas 58.21
121. K.Love 55.55
122. Al Horford 55.21
123. J.Noah 54.88
124. Calvin Natt 52.38
125. K.Lowry 52.36
126. Mehmet Okur 50.14
127. Steve Mix 49.72
128. D.Granger 48.22
129. D.Thompson 47.79
130. K.Kittles 44.05

[Mike G]

Hey, that's very nice.
Now, what about playoffs?
Then, what of pre-1974 seasons? What about the ABA?

[DSMok1]

"value over playoff starter" which is +2.5 or so.

Looking at last season's playoff stats, players who got >20 mpg and >2.5 BPM number barely 2 per team.
The avg 5th-best 'starter' BPM thus defined was -0.4

The top 80 players in mpg (27+) had avg BPM = 1.6

Sorry, that was "average playoff starter who makes it to the second round".

[Mike G]

For the record, I would not use Value over Replacement Player for an all-time ranking. I would use value over average, or perhaps even "value over playoff starter" which is +2.5 or so. In other words, I'd look more at players who produced a lot of value towards a championship.

Sorry, that was "average playoff starter who makes it to the second round".

OK, of 80 2nd-round players, just 23 were >2.5 BPM, almost 3 per team.

For the GSW, Curry, Green, Bogut, and Iguodala "produced a lot" toward their title. The others all carry a negative value toward the goal. What does it mean?

For careers >15000 RS minutes, some 90 players have BPM>2.5
Missing the cut are Billups, McHale, Moses, Parish, Ewing, 'Nique, English, Sikma, Mutombo, DJ, Isiah, Dumars, Cheeks, Laimbeer, Cummings, Parker, Fisher, and hundreds of other championship contributors.

Maybe you could add up single season values >2.5, counting other seasons as zeroes in a player's 'over the top' value. Otherwise, you have a preponderance of negative values.

Some high-minute NBA champs who miss the mark: Rodman '98, DJ '86, BScott '88, Parish '84, Parker x4, ACGreen '88, Kareem '88, McHale '84, K Smith x2, Hollins '77, Barnes '15, Odom '10, Sikma '79, A Johnson '99, Wilkes '80, Chalmers '13, Haslem '06, Fisher x2, Nixon '80, Armstrong '93, R Harper '00, Elie '95, Perkins '08, Archibald '81, Dumars '89, ... all with 620+ PO minutes.
A couple of Finals MVP in there.

[BasketDork]

Used the same criteria for the ABA, with an 8,000+ MP cutoff, and incomplete BPM information, 11 players qualify :

Code: Select all

1. J.Erving 82.65
2. A.Gilmore 75.94
3. D.Issel 67.28
4. Jimmy Jones 64.62
5. Jim Eakins 48.42
6. Billy Paultz 47.04
7. G.McGinnis 44.44
8. Dave Robisch 35.50
9. James Silas 34.65
10. G.Gervin 28.26
11. Tom Owens 24.95

[BasketDork]

Playoffs, I was thinking which cutoffs to use, using the same as RS and a 2000+ MP cutoff nets a little over 70 players, and pre-1974 has holes, calculating any sort of SPM in place of BPM would still require estimation.

[BasketDork]

The four NBA/ABA players affected most:

Code: Select all

--------NBA W's---ABA W's---Total W's
J.Erving------110.11-----82.65------192.76
A.Gilmore----101.92-----75.94------177.86
D.Issel--------74.29------67.28-----141.57
G.Gervin------84.86------28.26-----113.12

Wins Bump:

J.Erving 37th to 5th
A.Gilmore 48th to 11th
D.Issel 89th to 23rd
G.Gervin 74th to 36th

[BasketDork]

I think I'm going to delve deeper into this. I'll do the same for Playoffs, with a 2000 minute cutoff, then, I'll go back and scoop up the data on anyone listed at 15000 career MP and include their win totals as well. Cross that with anyone with ABA experience and tack on those numbers. To tackle the pre-1974 problem, easiest thing to do is a WS/48-PER blend and apply to those players minutes, do same thing for Playoffs, with same minute cutoffs for each. Probably post my findings over next couple days.

[BasketDork]

And also go back and scoop up numbers on anyone with 2000+ playoff minutes. This ought to take care of the Rodman's, I.Thomas' of the bunch...

[BasketDork]

So, last week I set out to embark on just a quick and dirty method of coming up with a wins list as a way of ranking players over their whole career. I used Mike G's idea and format to come up with PER Wins, W.S/48, and BPM Wins. I then averaged them together to get a general win total. I figured I'd carry over this research to its own thread, so not to get off the topic of where just Kobe ranks all-time. This list is comprehensive in that, it contains all those with 15,000 MP, 15.0+ PER, .125+ WS/48, and 1.0+ BPM, both collectively and any player who hits one of the three player metric thresholds as well close to 500 players in all. Players who played their entire careers before the BPM era, have only a WS/PER blend, any oldtimer with incomplete BPM 5 years or less, are also only using a WS/PER blend. Here is the original list of 130 players in the "BPM era". I will be adding to this list All other players who have made the cut, then a quick visit with the ABA, and then its on to the Playoffs, where I'm open to any tweaking if anyone has any suggestions?

So I borrowed Mike G.'s idea of averaging WS/48, and his formulas for PER W/48 ((PER - 5.75)/92.5) and BPM W/48 ((BPM + 5.14)/51.4), to come up with a W/48 average for all players with 15,000+ career MP, .125+ WS/48, 1.0+ BPM and 15.0+ PER, 130 players in all. I then took those Win Averages, multiplied them by the player's career MP and divided by 48, to come up with a Win Total. Here are the results: (*active)

Code: Select all

1. K.Abdul-Jabbar 257.19
2. W.Chamberlain 233.31
1. K.Malone 230.95
2. M.Jordan 212.74
3. T.Duncan 199.43 *
4. K.Garnett 198.81 *
5. L.James 191.24 *
7. S.O'Neal 182.95
8. J.Stockton 182.60
9. C.Barkley 181.33
10. D.Nowitzki 177.95 *
11. K.Bryant 177.40 *
12. H.Olajuwon 173.57
13. D.Robinson 170.21
14. R.Miller 157.74
15. J.Kidd 155.24
16. M.Malone 153.99
17. M.Johnson 153.20
18. L.Bird 151.91
19. P.Pierce 149.72 *
20. G.Payton 148.71
21. C.Drexler 144.14
22. R.Allen 143.96
23. R.Parish 136.64
24. S.Pippen 135.87
25. P.Ewing 127.96
26. V.Carter 126.77 *
27. C.Paul 126.45 *
28. S.Marion 126.22
29. P.Gasol 124.95 *
30. A.Dantley 123.23
31. D.Wilkins 122.84
32. Bob Lanier 121.46
32. D.Wade 120.09 *
33. S.Nash 117.22
34. A.Iverson 115.41
35. E.Brand 114.10
36. T.McGrady 110.37
37. J.Erving 110.11
38. Horace Grant 109.51
39. A.English 109.19
40. L.Nance 107.82
41. J.Sikma 107.60
42. Buck Williams 106.87
43. C.Billups 106.73
44. D.Mutumbo 106.69
45. G.Hill 105.20
46. J.Hornacek 104.16
47. D.Howard 102.89 *
48. A.Gilmore 101.92
49. T.Porter 101.35
50. V.Divac 101.16
51. R.Wallace 101.03
52. C.Webber 101.02
53. K.McHale 100.96
54. O.Thorpe 100.13
55. E.Jones 98.68
56. D.Schrempf 98.13
57. C.Mullin 97.09
58. M.Cheeks 95.90
59. C.Bosh 95.48 *
60. T.Parker 95.27 *
61. B.Laimbeer 94.30
62. S.Perkins 94.16
63. B.Wallace 93.85
64. M.Ginobili 93.65 *
65. C.Anthony 92.54 *
66. K.Durant 92.40
68. R.Lewis 90.49
69. T.Hardaway 89.45
70. B.McAdoo 89.23
70. H.Hawkins 88.29
71. D.Cowens 87.03'
71. S.Kemp 86.35
72. S.Cassell 86.27
73. A.Iguodala 85.63 *
74. G.Gervin 84.86
75. K.Johnson 84.63
76. M.Camby 84.38
77. A.Mourning 84.09
79. Alvan Adams 82.69
80. J.Worthy 82.13
81. A.Kirilenko 81.12
82. S.Moncrief 80.54
83. Marques Johnson 80.16
84. Dale Davis 77.78
85. C.Boozer 77.51
86. T.Chandler 76.37 *
87. Steve Smith 75.74
88. P.Stojakovic 74.72
89. D.Issel 74.29
90. D.West 73.16 *
91. Deron Williams 72.95 *
92. Gus Williams 71.54
93. Brad Miller 71.21
94. C.Maxwell 70.76
95. L.Johnson 69.72
96. L.Aldridge 69.01 *
97. Bobby Jones 68.69
98. D.Rivers 68.20
99. Hot Rod Williams 67.74
100. M.Price 67.46
101. Dan Roundfield 67.45
102. T.Brandon 66.52
103. John Drew 66.21
104. A.Hardaway 66.09
105. P.Millsap 66.07
106. B.Barry 65.86
107. R.Westbrook 65.15
108. David Lee 64.86
109. B.Daugherty 62.42
110. T.Kukoc 61.78
111. R.Barry 61.73
112. P.Westphal 61.65
113. J.Harden 61.49
114. N.Hilario 61.09
115. S.Curry 60.34
116. M.Gasol 59.41
117. Fred Brown 58.57
119. Y.Ming 58.36
120. G.Arenas 58.21
121. K.Love 55.55
122. Al Horford 55.21
123. J.Noah 54.88
124. Calvin Natt 52.38
125. K.Lowry 52.36
126. Mehmet Okur 50.14
127. Steve Mix 49.72
128. D.Granger 48.22
129. D.Thompson 47.79
130. K.Kittles 44.05

[Mike G]

Yes, what to do about playoffs? The easy thing would be to count them as regular games -- the "icing on the cake" approach. But is "longevity" over-emphasized in this approach?

If you don't want to try and create "wins above replacement" of some sort, you might boost the value of playoff performance, as a sort of proxy for "performance under pressure". Then, sticking with "wins" as a measure, what do you do?

Suppose you think playoff games/minutes/wins are 3 times as important (or indicative of value) as regular season. If you just multiply playoff totals by 3, your final result is an inflated wins total.
However, we know that players average about 7 to 7.5% of their total minutes in playoffs, varying only slightly due to era. Relative to that standard, you could reduce the value of RS minutes, such that (PO+RS) minutes sum to the same total.

Current players in their careers have had about .0732 of their minutes in PO (before this season).
Using that standard, you could multiply RS minutes and wins by .872, PO by 2.617, and you'd end up with the same total of minutes and wins for all players.
.872 * 3 = 2.617

Players with >.073 of their wins in playoffs would see their "wins" increase. Those with less than the avg would drop.
It's well believed that RS games are in fact worth less than the avg (including playoffs) NBA game. Shaq would typically take 10-15% of the season off, then dominate the postseason. Popovich regularly runs his guys a lot less in the RS.

[Statman]

Yes, what to do about playoffs? The easy thing would be to count them as regular games -- the "icing on the cake" approach. But is "longevity" over-emphasized in this approach?

The list will get dominated by current players - where it's much easier to make the playoffs & get more playoff wins that end up being fairly meaningless.

Here's my solution - I weight the playoff wins & run playoff WAR as if it were a regular season. I weight heavily at the championship level.

I give each playoff team a playoff weighted (or adjusted, or whatever you want to call it) wins total. Any win that doesn't clinch a series counts the same as the regular season, 1. Every win that clinches a series doubles in value depending on what round we're in:

round 1: 2
conference semis: 4
conference finals: 8
championship: 16.

So, last season - I run WAR on the Warriors regular season at 67 wins. I run WAR on their playoff totals at 42 "adjusted" wins.

Team playoff adjusted wins 2015
Warriors 42
Cavs 25
Rockets 13
Hawks 12
Clips 8
Bulls, Wiz, Grizz 7
Spurs 3
Bucks, Nets 2
Mavs, Blazers 1
Celtics, Raptors, Pelicans 0

1951 NBA (20 playoff games played TOTAL for the entire playoffs
Lakers (8 actual playoff wins): 33 (5 none clinching wins at 1 each, clinching wins of 4+8+16)
Caps 16
Knicks 6
Royals 5
Bullets 1
Warriors, Stags, Bombers 0

Anyway - this system makes going far in the playoffs much more important, especially winning the championship. I believe the final player results "fit" the perceptions of hard core basketball historians (who very much value championships) better than any other variation I have been able to come up with.

[Mike G]

The list will get dominated by current players - where it's much easier to make the playoffs & get more playoff wins that end up being fairly meaningless.

Well, it's easier to make the playoffs if you're a current player rather than retired. Ha.
But 14/30 teams don't get any playoff games. Some years ago, only 2 of 8 teams failed to get in.

As I mentioned, players of all eras wind up averaging about 7-7.5% of their career minutes in playoffs. There are no meaningless playoff wins. If you win a playoff series, you get another series of 4-7 games, and a chance at yet more.

Haven't done this for a while, so here it is by 7-year intervals:
Careers starting in the intervals shown; unweighted avg PO/T minutes; and (total playoff min)/(total min.)

Code: Select all

start yr   avg    tot
< 1957    .075   .081
1958-64   .079   .086
1965-71   .072   .076
1972-78   .065   .070
1979-85   .070   .078
1986-92   .067   .075
1993-99   .071   .078
2000-06   .072   .076
2007-11   .072   .072

These summarize the 700+ players in my big list. When you include lesser players, the ratios are likely smaller.
The totals column is bigger than unweighted avg, because the best players have longer careers and get into more than their share of playoffs. Apparently, that's less true of late.