Monday, March 12, 2012

NBA Championship Title Leaders Adjusted for Era

Introduction


NBA titles leaders are usually trotted out in discussions of the best players in the history of the league. Bill Russell regularly comes up in this respect because the media love to mention that he won 11 out of 13 titles when he was in the league. He's often called the best player or one of the top three best in the history of basketball for that fact alone. However, there were less than ten teams in the league during most of his playoff run. Even conferences today are much larger. In order to contextualize title leaders from different eras, you need to make some simple adjusments.


Methodology


In order to adjust championship titles, first I found the number of teams every year of the league. This was pretty easy, and there are some surprising numbers. There were only eight teams throughout much of the 60's, but after the NBA and ?BL merged in 1950 there were 17 teams. The league soon dropped some teams, but for most of its history new organizations were steadily added until the mid-90's as the Charlotte Bobcats are the only new one.

Next I put together a list of every player for a championship winner since the first title in 1947 when the league was known as Basketball Association of America (BAA). I included everyone who tallied playing time in the playoffs. Unfortunately, that does not include bench guys who didn't receive any minutes, but I don't think it actually matters in the greater context of this study. Teams can give rings to pretty much whoever they want -- even guys who operate the shot clocks have gotten them -- yet I think when looking at who's collected the most titles I think it's more proper to only include the guys who actually played in the playoffs. There are extenuating circumstances like Caron Butler with the Mavericks last year as he got injured before the title run, but ultimately it doesn't affect much of the top standings.

The first method of adjusting titles is to simply rate a title based on how many teams there are in the league. Winning an eight-team league is much easier; there's no debate there. I adjusted each title win to present standards meaning a win in a ten-team league is worth 10/30 or 0.33 titles. I know some will argue that winning a title before the league was "diluted" is more impressive, but you also have an influx of international players and a much larger US population. Basketball also wasn't very popular until the 80's. Another consideration: expansion teams typically take guys from each team, so the level of parity isn't heavily distorted in the years with new teams as one might think. Then a player's adjusted title for each year is added together.

The second method attempts to reward guys who contribute to winning teams as opposed to role players who ride the coattails of better ones to a title. For production, I used Win Shares as it helps guys who play large minutes and those that lead their teams. It's not perfect, obviously, but it separates Hakeem Olajuwon from, say, Rick Fox well. That year's entire playoff Win Shares were used instead of only the finals because it's a larger sample set and producing for the entire playoff run is more important. Then you take the percentage of that player's Win Shares out of his team's for the playoffs and multiply it with that season's adjusted title (teams/30) for the WS title score. Last step is to add a player's WS title scores for each championship year.

Results


In the table below, you can see how the total adjusted titles compare to the traditional total titles leaders. Instead of hall of fame center Bill Russell it's ... Robert Horry. It's surprising, but in some ways it isn't. In a previous study I found that he's the greatest playoff performer in terms of increasing his PER score from the regular season. He's been compared to the NBA's Forrest Gump, always in the right place at the right time, winning championships with Olajuwon, Shaq, Kobe, and Duncan, providing big shots in big moments. A critic of the adjusted titles method would say Horry at the top is enough evidence to sink the list, but he has the most titles out of any non-60's and 70's era Celtic, and the only player who's won multiple titles on three different teams.

More appropriately, Michael Jordan is second on the list with his teammate Scottie Pippen. I think there are many people who consider winning 11 titles in Russell's era than 6 in Jordan's, but there's enough reason to overturn that reasoning. One is that, obviously, it's easier to win when there are less teams in the league, which is what the adjusted titles does, but it's also an era with more rules to spread talent among teams evenly and one with heavier scouting and advanced defense. The Bulls weren't exactly stacked either; Jordan deservedly receives a lot of credit.

The leaders for the adjusted total titles are mostly Lakers and Bulls players at the top, while for the total titles it's Celtics and a few Lakers. One could argue that adjusting for the number of teams in that season is unfair to winners, but again I'd argue that looking at total titles it's unfair to those in the present.



Rank
Player
Adjusted total titles
Last title year
Rank
Player
Total titles
Last title year
1
Robert Horry
6.70
2007
1
Bill Russell
11
1969
2
Michael Jordan
5.60
1998
2
Sam Jones
10
1969
2
Scottie Pippen
5.60
1998
3
Tom Heinsohn
8
1965
4
Derek Fisher
4.90
2010
3
K. C. Jones
8
1966
4
Kobe Bryant
4.90
2010
3
Satch Sanders
8
1969
6
Ron Harper
4.83
2001
3
John Havlicek
8
1976
6
Steve Kerr
4.83
2003
7
Jim Loscutoff
7
1964
8
Dennis Rodman
4.63
1998
7
Frank Ramsey
7
1964
9
Kareem Abdul-Jabbar
4.37
1988
7
Robert Horry
7
2007
10
Tim Duncan
3.93
2007
10
Bob Cousy
6
1963
11
Shaquille O'Neal
3.90
2006
10
Kareem Abdul-Jabbar
6
1988
12
Magic Johnson
3.80
1988
10
Michael Jordan
6
1998
12
Michael Cooper
3.80
1988
10
Scottie Pippen
6
1998
14
Horace Grant
3.67
2001
14
George Mikan
5
1954
14
John Salley
3.67
2000
14
Jim Pollard
5
1954
14
Will Perdue
3.67
1999
14
Slater Martin
5
1958
17
Bill Russell
3.43
1969
14
Larry Siegfried
5
1969
18
Robert Parish
3.27
1997
14
Don Nelson
5
1976
19
John Havlicek
3.23
1976
14
Michael Cooper
5
1988
20
Sam Jones
3.17
1969
14
Magic Johnson
5
1988
21
Kurt Rambis
3.07
1988
14
Dennis Rodman
5
1998
22
Bruce Bowen
2.97
2007
14
Ron Harper
5
2001
22
Manu Ginobili
2.97
2007
14
Steve Kerr
5
2003
22
Tony Parker
2.97
2007
14
Kobe Bryant
5
2010
25
Brian Shaw
2.90
2002
14
Derek Fisher
5
2010
25
Devean George
2.90
2002
26
Vern Mikkelsen
4
1954
25
Jud Buechler
2.90
1998
26
Frank Saul
4
1954
25
Luc Longley
2.90
1998
26
Bill Sharman
4
1961
25
Randy Brown
2.90
1998
26
Jamaal Wilkes
4
1985
25
Rick Fox
2.90
2002
26
Robert Parish
4
1997
25
Toni Kukoc
2.90
1998
26
Kurt Rambis
4
1988
32
Sam Cassell
2.80
2008
26
John Salley
4
2000
33
Mario Elie
2.77
1999
26
Horace Grant
4
2001
34
B.J. Armstrong
2.70
1993
26
Will Perdue
4
1999
34
Bill Cartwright
2.70
1993
26
Tim Duncan^
4
2007
34
James Edwards
2.70
1996
26
Shaquille O’Neal
4
2006
34
John Paxson
2.70
1993




34
Scott Williams
2.70
1993




34
Stacey King
2.70
1993






One problem with the above approach is that role players can rack up titles playing with all-time greats (Robert Horry, Steve Kerr, Horace Grant, etc.) Perhaps a better one is to look at who contributes to the total, and award credit based on total production during the playoffs. The simplest method would be using points, and then more advanced versions add rebounds, assists, blocks, steals, etc. The logical conclusion is to use a method something that takes into account as much information as is possible. Thus, you give championship title credit based on how many wins shares a popular produces compared to the rest of the team. 

The winner by this method is none other than Michael Jordan, and he wins in a landslide. He has 42 more percent than the second ranked player, Kobe Bryant. His lead is enormous, and it's apparent why -- he was the overwhelming force on six championships in the modern era. What Jordan did in the regular season was enough for him to be legendary, but in the post-season he made his name transcendent. Moving down the list the right names are near the top than in the other lists. Robert Horry is still ranked high, but he was a pretty good player, especially for the Rockets. I'm amazed Fisher is ranked so high, however. I tried to season with that by realizing there is one whole Win Share to go around, and a starter can easily take 0.1 of those Win Shares if he gets major minutes even without taking a ton of shots. Fisher won five titles, and on average  he was decent so he totaled over half a title Win Share. 

Rank
Player
Total adjusted Win Shares titles
Last title year
1
Michael Jordan
1.650
1998
2
Kobe Bryant
1.164
2010
3
Tim Duncan
1.139
2007
4
Scottie Pippen
1.035
1998
5
Shaquille O'Neal
1.026
2006
6
Kareem Abdul-Jabbar
0.847
1988
7
Robert Horry
0.763
2007
8
George Mikan
0.684
1954
9
Larry Bird
0.684
1986
10
John Havlicek
0.645
1976
11
Magic Johnson
0.633
1988
12
Sam Jones
0.618
1969
13
Derek Fisher
0.552
2010
14
Bill Russell
0.549
1969
15
James Worthy
0.539
1988
16
Horace Grant
0.508
2001
17
Pau Gasol
0.498
2010
18
Tony Parker
0.497
2007
19
Hakeem Olajuwon
0.481
1995
20
Tom Heinsohn
0.478
1965
21
Manu Ginobili
0.442
2007
22
Dennis Johnson
0.412
1986
23
Jamaal Wilkes
0.395
1982
24
Toni Kukoc
0.394
1998
25
Ron Harper
0.367
2001
26
Kevin McHale
0.333
1986
27
Jim Pollard
0.332
1954
28
Michael Cooper
0.326
1988
29
Isiah Thomas
0.326
1990
30
Dwyane Wade
0.324
2006
31
Norm Nixon
0.323
1982
32
Dennis Rodman
0.318
1998
33
Byron Scott
0.309
1988
34
David Robinson
0.294
2003
35
Joe Dumars
0.276
1990
36
Rick Fox
0.263
2002
37
Robert Parish
0.253
1997

As a tangent, the "record" for highest proportion of Win Shares during a title playoff run is George Mikan in 1949 with an astounding 53.2 percent. Cliff Hagan in 1958 is next with 45.7 percent. Shaq has the modern record at 35.6 percent in 2000 with Tim Duncan close behind in 2003. On another tangent, the way Wins Shares are calculated a player can a negative number, which means he didn't even contribute to nothing; he's worse than that. Teammates Jim Cleamons and John Trapp hold the dubious distinction of lowest Win Shares for the playoffs while still winning a title with -0.018 in 1972 for the Lakers. Darko Milicic, by the way, wasn't far away at -0.011. Take that LeBron.

Awarding titles based on how many competitors you have also brings about the possibility to see adjusted title leaders by team. The Celtics are obviously hurt the most in this because they won most of theirs in the first couple decades of the league. The table below shows the adjusted leaders, and the Lakers take the top spot in this method. Even if you discount the Minneapolis Lakers years they still have the edge over the Celtics. Once again basketball fans are reminded that the titles are nearly all held by a few teams at the top. The NBA has been about dynasties, not parity and competitive balance. 

Team
Total title shares
Lakers
11.20
Los Angeles Lakers
9.27
Celtics
7.90
Bulls
5.60
Spurs
3.93
Pistons
2.70
Minneapolis Lakers
1.93
Rockets
1.80
76ers/Nationals
1.37
Warriors
1.20
76ers
1.10
Knicks
1.03
Heat
1.00
Mavericks
1.00
Blazers
0.73
SuperSonics
0.73
Washington Bullets/Wizards
0.73
Bucks
0.57
Royals/Kings
0.37
Baltimore Bullets*
0.27
Hawks
0.27
Nationals
0.27

Italicized: Championships separated from franchise by city
*: Defunct team (those Bullets aren't the same organization as the Washington one)

Conclusion


People saying Bill Russell is the greatest of all-time because of his titles need take his era into context. This is all without judging 60's era basketball based on its unathletic play, the lack of defense -- there were only eight teams for many of the years he won a championship, and there's no arguing that doesn't make it easier to win a title. They were definitely a great team and historically important, but be wary of making grand assessments about players and teams from that time. Russell's eleven titles when properly valued are less impressive than Jordan's six with the Bulls, Duncan's four with the Spurs, Kobe's five with the Lakers, and Shaq's four. Arguing otherwise is based on the false assumption that all championships are created equal.


Update: Bonus! Picture form. Click to enlarge and share with your friends at school. They're be so envious.

No comments:

Post a Comment