share

The Superstar Theory: Revisited and Supercharged, Part 1

Fri, 07/11/2008 - 9:18am

By Robert W. McChesney
NBADraft.net
July 11, 2008

(Three-part series)




Kevin Garnett

When the dust clears on the Boston Celtics unprecedented transformation from 2007 doormat to 2008 powerhouse and NBA champion, attention will rightfully turn to Celtics executive Danny Ainge, and his magnificent work in building this team. It is ironical that Ainge was repeatedly attacked during his first four years as the head of the Celtics for not making deals to make the Celtics more competitive in the short-term. Now it is apparent even to the blind among us that Ainge had a brilliant plan to win an NBA title, and that he understood the necessary foundations for winning an NBA title better than almost any other GM in the game.

In 2008 Ainge became the potential heir to Red Auerbach and Jerry West, the greatest GMs in NBA history.

What did Ainge understand? What did Ainge do? He understood that the basis for winning an NBA title is having a superstar in his prime. Not an all-star, or a bunch of all-stars, but a superstar. There are only a few in the game at any time so they are almost impossible to get. But he went and got one. He did so by carefully developing young players – aside from Paul Pierce the cupboard was bare in 2003 when he took the reins of the team -- and not being obsessed with short-term performance. By 2007 he had the chips to keep Paul Pierce and still get Kevin Garnett, and to surround him Ray Allen. He moved quickly and decisively in some of the most brilliant GM work in any sport in history.

It is astonishing that other teams and fans and pundits have been largely oblivious to what Ainge accomplished and how much he followed in the footsteps of Auerbach and West. In this three-part series I will provide dramatic evidence to explain the genius of Ainge’s work, and to highlight how important having a genuine superstar is to contending for, not to mention winning, an NBA title.

This is not really that difficult a concept. If you think about it, it seems like almost always the team that wins the NBA championship – and the other team that gets to the finals – is led by a player who is one of the very best players in the history of the game. There were only five times in the past 29 seasons that the NBA champion was not led by a player named Bird, Magic, Jordan, Olajuwon, Duncan or O’Neal.

In fact, it seems like if you don’t have one of these guys on your team, no matter how many quality players and all-stars you might have, you simply cannot win an NBA title, or possibly even contend.

It is not like this in baseball or football. There championship teams often do not include the very best players in the game, there are an aggregation of very good all-star caliber players and role players. Likewise, first-ballot Hall of Fame players can play much or all of their careers without ever playing for a title, let alone winning it. Barry Sanders anyone? So imitating baseball or football teams in building up rosters of able quality players, even all-stars, may be a dead-end street, at least if the point of the exercise is to win, or even seriously contend, for a championship.

It is not just that there are fewer players on a basketball team, though that is a significant factor. It is also that by the sport’s very nature, a truly great player establishes his greatness by being the hub of a successful team. The best players weave the most powerful webs. And in crunch time they can take over like in no other sport. In baseball a pitcher can win the Cy Young award on a last place team. That would be unthinkable in the NBA.

The importance of having a player-for-the-ages in his prime to winning an NBA title, or even contending, is astonishing. Yet, as I stated at the top, most fans and many NBA GMs fail to grasp this. Fans root crazily for modest incremental improvements, failing to see that by getting a lesser chance at a high draft pick or by squandering cap space they are only digging their hole that much deeper. GMs similarly obsessed by marginal improvement fail to take steps to improve their chances of actually winning a title.

This is an issue I have been researching for a few years. Two years ago I published a two-part article on NBAdraft.net on the topic: The Gold Medal Superstar Theory and NBA Championships: Part 1, Part 2. The response was positive, including from people in and around the NBA. I also received some valuable criticism, so I have significantly refined and recalibrated the data. I tightened the argument so that the case is considerably more muscular now. If anything, by doing so the relationship between having a superstar and winning an NBA title is made that much stronger. And the implications for how to build an NBA team could not be more dramatic.

My exercise is simple: First, I make a list of the very best players in NBA history over the past 52 years, or since the MVP award was introduced in 1956. I use objective criteria, not give my personal opinion. I want the list to be made based upon regular-season performance, to avoid having a list of best players that rewards players for playing on championship teams. That is what I do in what follows in part one of this three-part series.

Once I locate a list of the top NBA players over the past 52 years, I figure out who the two best players have been on each NBA champion, and the best player on each team that lost in the finals, and on the other two teams that lost each year in the conference finals. In other words I determine the best players on each of the NBA’s “final four” every season, as well as the second best player on the championship team. I then see how many of these best players are from the list of best players, and where on the list they can be found. The results are astounding. I do that in part two, which appears in a few days.

The series will conclude in part three with a discussion of how we can predict who will be the great superstars of the coming era. More important, I discuss the best ways teams can go about getting a superstar. It ain’t easy, but it is more than a matter of luck. Danny Ainge demonstrated that.

Part One: Determining the Best NBA Players since 1956

Who are the best players in NBA history? How can we determine it, without it becoming a battle of subjective opinions? In particular, for our purposes, we want to determine the best players in NBA regular season history. If we have a system that rewards players for playoff performances my overall argument will be circular, because players on championship teams will be far more likely to be at the top of the list of best players.

There has been a revolution in the past decade with basketball statistics, and it may well be that we can come up with an objective standard that we can use to apply to all players since 1956 to rank order them. John Hollinger’s PER is the most notable example. But these are not flawless. The most interesting new statistical work is still largely unknown to the public and chronicles how players do offensively and defensively on a touch-by-touch basis. To do this requires videotape, and that means it is impossible to go very far into the past to be comprehensive, not to mention it would be mind-bogglingly time consuming.

So I have tracked down the four most objective criteria I could find. The value of each of these is that they are determined every year immediately after the regular season and before the play-offs. None of these is perfect, but they are as objective a standard as we can hope to find. In a weighted combination, they will give us our list of superstars.

The first and most important criterion is the MVP voting, because this is done explicitly to select the best players in the league for the season. Its strength is that is does not care about positions, so if the best players are all centers, they can get the most votes. The weakness is that players may get privileged a bit too much for regular-season team success, over their actual performance. I have gone through every MVP vote since the award was introduced in 1956 and allocated points to players on the following system:

MVP:

MVP: 15 points

MVP 2nd: 13 points

MVP 3rd: 11 points

MVP 4th: 9 points

MVP 5th: 7 points

MVP 6th-10th: 5 points

MVP 11th-15th: 3 points

So if a player wins two MVPs, finishes third in the MVP voting one time, and finishes 7th and 12th on two other occasions, that player would get 49 MVP points.

The second criterion is all-NBA teams. There were first and second all-NBA teams until 1989, when a third team was added. Because the league has so many more teams in recent decades I do not think it unfair to include the third team all-NBA team in this study. The strength of these all-NBA teams is that they tend even more than the MVP to reward players independent of a team’s success. The weakness of the all-NBA teams is that they are determined by position. This means that only one center can make first-team. We have seen the absurd situation of a center like Bill Russell or Dave Cowens winning the MVP award yet making second-team all-NBA. The all-NBA teams cannot accurately reflect how bigs dominate the sport. A great players like Nate Thurmond never made all-NBA in his career, yet he often finished in the top 10 of MVP voting.

The scoring is pretty straight forward.

All-NBA:

First team all-NBA: 10 points

Second team all-NBA: 5 points

Third team all-NBA: 3 points

 

The third criterion is the first and second all-Defense teams, which the NBA introduced in 1968-69. I decided to add this because the NBA wisely understood that its standard all-NBA team tended to favor players with impressive offensive statistics, even though great defensive players won many games in their own way. I give the all-defensive teams less weight.

All-Defense:

First team all-Defense: 3 points

Second-team all-Defense: 2 points

Because the all-defensive teams did not begin until 1969, I retroactively estimated how players who made these teams after 1969 but who had much or most of their careers before 1969 would have done had the all-defensive teams gone back to 1956. This way players like Russell, West, Havlicek, Chamberlain, Gus Johnson, and Thurmond were not slighted in the rankings. (I put their all-defensive figures in parentheses to indicate they include my estimates.)

Finally, I give an additional 2 points for the player selected Defensive Player of the Year. This award honors the truly dominant game-changing defenders, who were insufficiently recognized by merely being first-team all-Defense. Because this award began in 1982-83, I retroactively award points for it to players who would have been likely winners between 1956-1982. That means: Bill Russell, Jerry West, Wilt Chamberlain, Kareem Abdul-Jabbar, Bill Walton and Dennis Johnson. (Again, where my estimates are added to the real totals, I use parentheses.)

So if a player makes the first-team all-Defense twice and second-team once, he gets 8 all-defense team points. If he was Defensive Player of the Year once, he gets 2 more points, for a grand total of 10.

Using these criteria, I tabulated a point total for every player in the league since 1956, assigning points for every time a player got one of these honors. (Thanks to an extraordinary website, this is not such an arduous task.) The more times a player gets in the top 15 in MVP balloting and the more times a player makes all-NBA and all-Defense teams, the more points a player will accrue. The most points a player could get in a single season is 30: MVP (15 points), first-team all-NBA (10 points), first-team all-Defense (3 points), Defensive player of the year (2 points). This season Kobe Bryant racked up 28 points.

One might quibble with how I weighted the categories, but after considerable debate and input this is the best I could do. I will say that after playing with these numbers for some time, the actual ranking does not change very much if the weighting of these four categories is altered. It is pretty much the same cast of characters in pretty much the same order.

There were a few issues I needed to sort out along the way. First, I decided not to include the ABA, because to do so would mean that there would be almost double the points given for players in the years from around 1968 to 1976, and that would give players from that era an unfair dominance in the rankings. At the same time, to not include those MVPs and all-ABA awards would disadvantage tremendous players like Julius Erving and Rick Barry who spent years dominating that league. They would have certainly dominated the NBA had they been in the NBA during those same years.

I solved this problem by doing the ranking not by the total number of points accrued during a career, but rather by taking a player’s total number of points and dividing it by the number of seasons they played at least 2,000 minutes. (I also counted as seasons years in which they played less than 2,000 minutes, if they were seasons in which the player still received points for all-NBA teams or top-15 MVP votes.) That way the rankings are based on yearly averages rather than total points and do not penalize players who had fewer seasons in the league because of the ABA. It still clobbers Dr. J and Barry because they had monster years in the ABA that would have raised their averages, but it does not clobber them as much as if we simply used raw totals.

Using season averages as the basis of rankings also has the beneficial effect of not penalizing players like Bob Cousy or Bill Sharman or Dolph Schayes who had several great years before 1956.

Most important, by basing it on yearly averages rather than total points, it means we can slot active young superstars right in the heart of the list. They do not need to play ten years to get on the list. That makes it much more useful for the evaluation I do in part three. We can see who the real up-and-coming studs are. (Active players are bold-faced.) Most young stars like Paul, James, Howard and Wade will see their averages shoot up as they rack up several 20 point seasons in a row. If a player fails to make this list by 25 or 26, it is unlikely they will ever be a superstar.

What follows is the list, then, of the 80 best regular season players in the NBA since 1956. Why 80? I cut it off with players who averaged at least 3.7 points per season for the years they played 2,000 minutes. A player who averages 3.7 points per season is someone who during his prime is getting regular recognition as one of the 15 best players in the league. That seems like the ante for admission.

This may seem like an easy target for a player to achieve, but it is not. Only 80 players made the cut. A lot of very good players, perennial all-stars, did not make the cut.

To provide some sense of how elite this list is, consider this: Since 1956, some 320 different players have qualified to play in the annual mid-season NBA all-star game. To discount players who may have made the team on a fluke, there were some 225 players who made the NBA All-Star game at least twice in their careers. Yet nearly two-thirds of those wonderful players do not qualify for this list. Nine players who have made the NBA All-Star team at least seven times – e.g. Ray Allen, James Worthy, Lenny Wilkens, Jack Sikma, Robert Parish, JoJo White and Vince Carter – are not on this list.

In the list of 80 there is a huge drop-off from the top to the bottom. Therefore, I break the 80 superstars into three groups: the gold medal superstars; the silver medal superstars and the bronze medal superstars. By doing so we can see who among the superstars are carrying the most weight.

There are two major problems with basing the list on annual averages rather then career totals points: First, players with short but productive careers get favored over players with longer and almost as productive careers. This would distort the list. Without some qualification, Bill Walton would rank as the very best player in NBA history, because for two seasons Bill Walton was unstoppable. But does anyone think Bill Walton is the best player in NBA history?

So I solved this problem by making the number of qualifying seasons at seven for a player to make the top of the list. Those players whose annual totals qualify but who do not have seven qualifying seasons, like Bill Walton, are on the list, but they are ranked after those players who did have seven qualifying seasons. I keep active players in the main list, under the assumption that they will all eventually get seven qualifying seasons. No reason to penalize their ranking.

Second, players who extended their careers into their late 30s see their season averages decline, and this pushes them down the list below where they would have been had they retired on their 36th birthday. Does anyone really think Michael Jordan was any less a dominant superstar from 1984-98 because he came back for two seasons as he approached 40?

I solved this problem by not including as seasons those years players played after their 36th birthday, unless they were seasons in which the player received some points by my above categories; in other words, seasons with over 2,000 minutes counted for players over 36 as long as the players were in or near their primes. Players like Jordan, Kareem, Ewing and Havlicek are not penalized for having such long careers.

Even with these qualifications, I do not think the list is perfect. There are players not on this list I think are better than some of the players on it, especially toward the bottom. But that is unavoidable. Part of the problem is that the four categories I use are not perfect. Sportswriters are far from perfect as voters. Players get penalized when they play in the same era as other great players at their position. They are simply the only objective standards we have. This list is as objective as it gets.

Nor do I think this can be regarded as statistically the final word: that, for example, Jerry West is not necessarily a better player than Magic Johnson because he has a higher score.

Nevertheless, I think we can use this system to separate these players from the balance of the league, and then among these 80 players, to create three general clusters: gold medal, silver medal, and bronze medal superstars.

The 80 Greatest Players in NBA History since 1956:

Gold Medal Superstars

 


PLAYER:
ALL-NBA PTS

MVP
PTS

DEF.
PTS

TOTAL
PTS / YRS
AVG
PTS
1. Michael Jordan
105
+
128
+
29
=
262 / 11
23.8
2. Kareem Abdul-Jabbar
150
+
185
+
(35)
=
370 / 17
21.8
3. Tim Duncan
100
+
109
+
30
=
239 / 11
21.7
4. Bill Russell
70
+
146
+
(62)
=
276 / 13
21.2
5. Larry Bird
95
+
127
+
6
=
228 / 11
20.7
6. Wilt Chamberlain
85
+
127
+
(28)
=
240 / 13
18.5
7. Jerry West
110
+
89
+
(40)
=
239 / 13
18.4
7. Shaquille O’Neal
99
+
116
+
6
=
221 / 12
18.4
7. Bob Cousy
70
+
59
+
0
=
129 / 7
18.4
7. Bob Pettit
95
+
84
+
(5)
=
184 / 10
18.4
11. Magic Johnson
95
+
115
+
0
=
210 / 12
17.5
12. Kobe Bryant
74
+
68
+
22
=
164 / 10
16.4
13. Elgin Baylor
100
+
74
+
0
=
174 / 11
15.8
13. Karl Malone
123
+
118
+
11
=
252 / 16
15.8
15. Oscar Robertson
100
+
91
+
(25)
=
216 / 14
15.4
15. David Robinson
62
+
86
+
22
=
170 / 11
15.4
17. Hakeem Olajuwon
84
+
86
+
27
=
197 / 14
14.1
18. LeBron James
30
+
39
+
0
=
69 / 5
13.8
19. Kevin Garnett
61
+
75
+
27
=
163 / 12
13.6

Qualify But Below Seven Season Minimum:

20. Bill Walton
15
+
28
+
(10)
=
53 / 2
26.5
21. Dolph Schayes
40
+
42
+
0
=
82 / 6
13.7

 

COMMENT: I made the cut-off for gold-medal status at 13.5 points, because a player that averages that during his career is basically going first-team all-NBA and getting top 10 MVP votes a clear majority of the years when they are healthy. They are among the five best players in the league throughout much of their career. There are generally only three or four players of this caliber playing in their prime in the league at any time. Players averaging over 20 points per year are especially mind-blowing. This could almost be a special “platinum” level. These guys were almost always among the two or three best players in the league for virtually the entirety of their healthy careers. And as we will see in part two, that means the teams these players led were almost always the teams contending for the NBA championships.

 

I suspect the only major surprise is Bill Walton, and this is only a surprise for those too young to remember Walton from his Portland days. For two seasons, before he turned 26, Bill Walton dominated the NBA like few other players ever have. Had he remained healthy the history of NBA basketball in the 1980s would have been very, very different. Portland would have given the Lakers an annual war like the 76ers did for the Celtics. It would have been heaven on earth, and the 80s already was heavenly.

 

The balance of the list includes 20 players who would be on just about any conceivable list of the 24 best players in the NBA since 1956. Julius Erving would be among the Gold Medal superstars, averaging 15.3, if his ABA years were included. That LeBron James is in this group at age 23 is mind-blowing. The only question is how close to the top he will finish.

 

Based on this portion of the list alone, this statistical system seems to be turning up the right names.




Silver Medal Superstars


PLAYER:
ALL-NBA PTS

MVP
PTS
DEF.
PTS

TOTAL
PTS / YRS
AVG
PTS
22. Julius Erving
60
+
63
+
0
=
123 / 10
12.3*
23. Dirk Nowitzki
51
+
58
+
0
=
109 / 9
12.1
24. Steve Nash
41
+
54
+
0
=
95 / 8
11.9
25. George Gervin
60
+
57
+
0
=
117 / 10
11.7
26. Moses Malone
60
+
93
+
5
=
158 / 14
11.3
27. Charles Barkley
78
+
77
+
0
=
155 / 14
11.1
28. John Havlicek
75
+
31
+
(37)
=
143 / 14
10.2
29. Willis Reed
30
+
37
+
3
=
70 / 7
10.0
30. Tracy McGrady
41
+
38
+
0
=
79 / 8
9.9
31. Jason Kidd
55
+
44
+
22
=
121 / 13
9.3
32. Walt Frazier
50
+
12
+
21
=
83 / 9
9.2
33. Nate Archibald
40
+
33
+
0
=
73 / 8
9.1
34. Rick Barry
55
+
26
+
0
=
81 / 9
9.0
34. Gary Payton
51
+
46
+
29
=
126 / 14
9.0
34. Patrick Ewing
40
+
53
+
6
=
99 / 11
9.0
36. Scottie Pippen
46
+
36
+
28
=
110 / 13
8.5
38. Allen Iverson
48
+
51
+
0
=
99 / 12
8.3
39. Chris Paul
10
+
13
+
2
=
25 / 3
8.3
40. John Stockton
59
+
46
+
10
=
115 / 14
8.2
40. Dave Cowens
15
+
48
+
(9)
=
72 / 9
8.0
42. Ben Wallace
21
+
15
+
25
=
61 / 8
7.6
43. Bob McAdoo
15
+
46
+
0
=
61 / 8
7.6
44. Dwyane Wade
13
+
13
+
2
=
28 / 4
7.0

Qualify But Below Seven Season Minimum:

45. Sidney Moncrief
30
+
29
+
18
=
77 / 6
12.8
46. Bill Sharman
45
+
8
+
0
=
53 / 5
10.6
47. David Thompson
20
+
21
+
0
=
41 / 4
10.3
48. Paul Westphal
35
+
5
+
0
=
41 / 5
8.2
49. Gus Johnson
20
+
5
+
(20)
=
45 / 6
7.5

 

*If Erving’s ABA years were counted, his average would be 15.3, putting him in the gold medal category.

COMMENT: This group rounds out the NBA top 49. Generally the silver medal players, averaging at least 7 points per season, were first or second team all-NBA much of their careers and received frequent recognition in MVP voting. During much of their careers, these are players who would be considered among the 10 best in the league. Several of the silver medal players have periods in their careers when they rank in the top 3-5 in the league; they simply do not have this status extend for their entire careers, like the gold medal superstars. These are all sure-fire Hall-of Famers, legends, and all-time greats who had stellar careers.

Bronze Medal Superstars


PLAYER:
ALL-NBA PTS

MVP
PTS
DEF.
PTS

TOTAL
PTS / YRS
AVG
PTS
50. Alonzo Mourning
15
+
30
+
6
=
55 / 8
6.9
51. Grant Hill
30
+
31
+
0
=
61 / 9
6.8
52. Billy Cunningham
35
+
18
+
0
=
53 / 8
6.6
52. Elvin Hayes
45
+
44
+
4
=
93 / 14
6.6
52. Amare Stoudemire
20
+
13
+
0
=
33 / 5
6.6
55. Spencer Haywood
30
+
15
+
0
=
45 / 7
6.4
56. Dominique Wilkins
36
+
40
+
0
=
76 / 12
6.3
57. Kevin Johnson
23
+
19
+
0
=
42 / 7
6.0
58. Isiah Thomas
40
+
31
+
0
=
71 / 12
5.9
59. Pete Maravich
30
+
14
+
0
=
44 / 8
5.5
59. Dwight Howard
13
+
7
+
2
=
22 / 4
5.5
61. Dave Bing
25
+
28
+
0
=
53 / 10
5.3
62. Chris Webber
28
+
29
+
0
=
57 / 11
5.2
62. Jerry Lucas
40
+
12
+
0
=
52 / 10
5.2
62.

Dennis Rodman

6
+
14
+
27
=
47 / 9
5.2
65. Anfernee Hardaway
23
+
16
+
0
=
39 / 8
4.9
65. Clyde Drexler
26
+
33
+
0
=
59 / 12
4.9
67. Bernard King
28
+
24
+
0
=
52 / 11
4.7
67. Tim Hardaway
28
+
19
+
0
=
47 / 10
4.7
69.

Jermaine O’Neal

11
+
11
+
0
=
22 / 5
4.4
70. Nate Thurmond
0
+
23
+
(20)
=
43 / 10
4.3
71. Dennis Johnson
15
+
12
+
(28)
=
55 / 13
4.2
72. Marques Johnson
20
+
13
+
0
=
33 / 8
4.1
73. Sam Jones
9
+
21
+
0
=
30 / 8
3.8
73. Tommy Heinsohn
20
+
10
+
0
=
30 / 8
3.8
73. Gilbert Arenas
11
+
8
+
0
=
19 / 5
3.8
76. Kevin McHale
10
+
12
+
15
=
37 / 10
3.7
76. Chauncey Billups
8
+
10
+
4
=
22 / 6
3.7

Qualify But Below Seven Season Minimum:

78. Mark Price
19
+
20
+
0
=
39 / 6
6.5
79. George McGinnis
15
+
10
+
0
=
25 / 5
5.0
80. Cliff Hagan
10
+
14
+
0
=
24 / 6
4.0


COMMENT: Here, too, we see a list of Hall-of-Famers and legends though there is a notable drop from the gold and silver medal categories. The drop-off is most apparent in the MVP voting. Averaging at least 3.7 points means that these players all got recognition annually of being among the 15 best players in the game during a majority of their careers.

So there we have as close to an objective list of the best players in the NBA since 1956 as I can imagine.

This is rarified air, and this is where you must go if you want to find the key to winning NBA championships.

RSS: Syndicate content