The History of NBA Champions: A Win Share Story
LeBron or MJ? Basketball fans have been inclined to bring up this topic upon LeBron’s 8th (8th!!!) straight Finals appearance. Golden State’s sweep of King James reminded us all that in spite of one man’s herculean postseason efforts, basketball, in the end, is a team sport. In fact, now the discussion revolves around the Warriors dynasty status. Are they a dynasty? What is the greatest NBA dynasty? What even qualifies as a dynasty?
I’m here to say, “I don’t know!”. But what I do know is that the 2017–2018 edition of the Warriors was probably their second worst championship team yet. Their best was the team that started the dynasty in 2015, and their worst was the Philadelphia Warriors championship in the inaugural season of the Basketball Association of America (BAA), the predecessor of the NBA.
How did I arrive at this conclusion? Win shares.
Win shares attempt to quantify a player’s contribution to their team’s success. The more win shares a player has, the more of an impact that player supposedly has on his team’s performance. I compiled the win share totals for every BAA/NBA/ABA championship winning team and then divided the total up between the starters (top 5 win share leaders for a given team) and bench players (everyone else on the roster). Next, I divided these totals based on the number of players. For the starters, it was by five and for the bench players, it was various numbers based on how many people played on the team. This allowed me to determine the average win shares per starter and average win shares per bench player. I summed up these averages to create the Win Share Sum for each championship team. It seems that for the most part, really good teams have a great starting five and a good bench. Or an insane starting five and an ok/subpar bench. Or an ok starting five, an ok bench, and just got really lucky (or played in the 1940’s).
According to this method, the top 5 all-time NBA teams that won a championship are as follows:
- 1995–96 Chicago Bulls
- 1971–72 Los Angeles Lakers
- 1970–71 Milwaukee Bucks
- 1996–97 Chicago Bulls
- 1991–92 Chicago Bulls
So the top five is pretty boring with repeat franchises from similar eras, so I tweaked it a bit to get something more interesting. I removed the last two Bulls teams to include some different franchises.
- 1995–96 Chicago Bulls
- 1971–72 Los Angeles Lakers
- 1970–71 Milwaukee Bucks
- 2007–08 Boston Celtics
- 1969–70 New York Knicks
A little bit of variety creates some intrigue here since I wouldn’t have included the 2008 Celtics in my list of the best NBA teams of all-time. 95–96 Bulls, check. Any team with Kareem, check. Walt Frazier in his prime, definitely. Maybe it’s because the team hasn’t had time to let legendary tales form? The bench statistics explain a lot. That Celtics team had the best bench of a championship winner since the 1961 Celtics! Also, they arguably had one of the best starting five of any Celtics team to have won a title. In fact, the third best, according to win share averages. Maybe one day I’ll tell my children of the extraordinary triple headed monster that were the 2007–08 Celtics.
Now, onto the worst championship teams!
- 1947–48 Baltimore Bullets
- 1998–99 San Antonio Spurs
- 1957–58 St. Louis Hawks
- 1946–47 Philadelphia Warriors
- 1994–95 Houston Rockets
The Warriors and Bullets were the first two BAA champions, respectively, so I can’t say I’m surprised these teams were not great. Led by Bob Pettit and Cliff Hagan, the Hawks somehow beat the Bill Russell led Celtics, who are probably pretty pissed they didn’t win 10 championships in a row. The remaining two teams are interesting to me, though. Both teams won championships in the 90’s, a decade dominated by Jordan. Curiously, they won when Jordan didn't play. If MJ stays on the Bulls those years…
Another set of lists I think are interesting are the best/worst starting five and the best/worst benches to win a championship.
The best starting five rankings look something like this…
- 1970–71 Milwaukee Bucks
- 1971–72 Los Angeles Lakers
- 1995–96 Chicago Bulls
- 1991–92 Chicago Bulls
- 1996–97 Chicago Bulls
Again, the Michael Jordan dominance continues. The list also looks eerily similar to the top five teams to win a championship ever; the same three Bulls teams and the Kareem-led Bucks/Lakers.
The worst starting five to win a championship are:
- 1947–48 Baltimore Bullets
- 1998–99 San Antonio Spurs
- 1977–78 Washington Bullets
- 1994–95 Houston Rockets
- 1954–55 Syracuse Nationals
The list also looks pretty close to the list of worst teams to win a championship.
Onto the best and worst benches! The BBNH Award (Best Bench in NBA History) goes to…
- Your 1968–1969 ABA Champion Oakland Oaks!
Followed by:
2. 1960–61 Boston Celtics
3. 2014–15 Golden State Warriors
4. 2007–08 Boston Celtics
5. 1963–64 Boston Celtics
Man, what is it with Boston and having great teams and even better benches! As a Sixers fan, I’m pissed.
The worst bench squad to win a championship was the 1967–68 Pittsburgh Pipers. The remaining teams in order are the 1947–48 Baltimore Bullets, 1950–51 Rochester Royals, 1957–58 St. Louis Hawks, and 1948–49 Minneapolis Lakers.
I also created another statistic called the Win Share Ratio. This statistic is the Top 5 Win Share/Bench Win Share and it can help determine which teams were carried by their starting five or which teams had the most bench help.
The teams carried most by their best players are:
- 1950–51 Rochester Royals
- 1955–56 Philadelphia Warriors
- 1967–68 Pittsburgh Pipers
- 1970–71 Milwaukee Bucks
- 1957–58 St. Louis Hawks
The top team-oriented teams are:
- 2013–14 San Antonio Spurs
- 2017–18 Golden State Warriors
- 2004–05 San Antonio Spurs
- 1994–95 Houston Rockets
- 2014–15 Golden State Warriors
I also wanted to see what decades had the highest average Win Share Sum among the championship winners. These are the decade rankings from best to worst (average in parentheses).
- 60’s (10.42)
- 80’s (10.39)
- 90’s (10.35)
- 00’s (10.23)
- 70’s (9.76)
- 10’s (9.64)
- 50’s (9.11)
- 40’s (7.19)
The best winners from each decade are the 1948–49 Minneapolis Lakers, 1952–53 Minneapolis Lakers, 1961–62 Boston Celtics, 1971–72 Los Angeles Lakers, 1985–86 Boston Celtics, 1995–96 Chicago Bulls, 2007–08 Boston Celtics, and the 2014–15 Golden State Warriors. The Lakers and Boston really need to chill.
After looking at this histogram above, I realized these win shares look very similar to a normal distribution.
Now, in a normal distribution, as shown in the table above, 68% of the data is within one standard deviation from the mean (or average). 95% of the data is within two standard deviations from the mean and 99.7% of the data is within three standard deviations of the mean. What’s the standard deviation, you ask? Standard deviation measures how far each data point is from the mean and it lets us know how varying the data set is. Lower standard deviations indicate a data set that is closer to the average.
The mean Win Share Sum is 9.88 and the standard deviation for the data is 1.44. Using these numbers, we can create a unique normal distribution for the data!
We can use this distribution to calculate what percentile each championship team falls in. We can also use this to see how current (or past) NBA teams stack up to champions. Since I’m a 76ers fan and they actually did well this season, I will use them to demonstrate this point. The Sixers’ Win Share Sum this season was 8. With a calculator, we can find that this is in the 10th percentile. The Sixers, had they won the NBA Finals :( , would have been better than only 10% of NBA Champions. In other words, they would have been statistically one of the worst teams to win the title. The Warriors, by comparison, were better than 20% of NBA Champions. Not great, but they were obviously a better team. Were they the best team this year though? The Houston Rockets actually had a Win Share Sum of 10.58 which is better than 69% of previous NBA title winners. Many people picked the Warriors based on experience, coaching, and a proven track record in the playoffs, but at the start of the postseason, some picked the Rockets to win it all. Statistics can’t explain everything so our model failed to predict the outcome here, but if the Rockets had beaten the Warriors, I think most would agree there would be a parade in Texas.
What about the record-breaking 2015–16 Warriors? They had a Win Share Sum of 11.32!!! This would have made them better than 84% of previous NBA Champions and solidified themselves among teams like the 1990–91 Bulls and 1999–00 Lakers. Some things can’t be explained by statistics. Things like LeBron James.
Which brings me to my final point. There is no doubt that LeBron James and Michael Jordan are two of the greatest basketball players we’ve ever seen (although if I’m being honest a G.O.A.T. conversation that doesn’t include Wilt Chamberlain and Kareem Abdul-Jabbar is a flawed conversation). However, most of Michael Jordan’s teams were better than LeBron’s teams. Is this a testament to LeBron’s solo ability or evidence of Jordan’s ability to make everyone around him excel? The G.O.A.T. debate will rage on until the NBA ceases to exist or until Ben Simmons wins 6 rings and the debate process repeats. Yeah, I said process… you gotta problem with that?
Data provided by Basketball Reference.
Basketball Reference. NBA & ABA Champions. Philadelphia: Sports Reference LLC, 1947–2018. Web. 27 Jun 2018. https://www.basketball-reference.com/playoffs/