A New Method of Evaluating Franchise Dominance

By Andrew Puopolo and Benedict Brady

In sports bars around the country, people are constantly arguing the merits of their respective sports teams. Fans of the New York Yankees and Boston Celtics will often cite their superiority in the number of Championships won while fans of teams like the Edmonton Oilers will cry foul, claiming that they have been in the NHL for a much shorter period. Furthermore, fans of the Los Angeles Lakers will argue that the Celtics won all their NBA Championships in the 1960s when there were less than 10 teams in the NBA. This got us thinking, is there a better way to determine which franchises are the best at winning championships?

In theory, we would like to devise a formula that would control for the length of time the franchise has existed (winning two championships in 20 years should be better than winning three in 100 years) and number of teams in the league competing for each championship (winning a title with eight teams should be worse than winning one with 30). To account for this, we assigned each championship a value. This value is y/30 where y is the number of teams in the league in that particular season. We are fully aware that the NFL currently has 32 teams (and not 30), but we are interested in comparing across sports so we wanted to standardize it somehow. This may make give the NFL data a less obvious interpretation, but since most of the leagues have 30 teams this is an obvious choice. Under this, the Boston Celtics winning the 1964 NBA Championship was worth 9/30 of a Championship while the Rangers winning the Stanley Cup in 1994 was worth 26/30 of a Championship. From this, we were able to calculate the number of adjusted championships for each team.

The next issue to tackle is the longevity of teams in the league. We decided to adjust for this by dividing the number of years the team has been in the league. This gives the average championships per year in a 30 team league. We then multiplied by 100 to represent this value as a percent (essentially the percent chance this team would win a championship in a 30 team league). On average, each team will have an overall index of 3.33. We then ranked the teams by this index and produced the below table of all the teams in pro sports that had an index greater than 5 (1.5x expectation).

Editors Note: Since the winner of the 1925 Stanley Cup was a non-NHL team (based on the way the Stanley Cup was set up at the time), we started counting Stanley Cup wins (and seasons in the league) from the start of the 1925/26 season. As a result, we had to take away 1 Stanley Cup from the Montreal Canadiens and 2 from the Toronto Maple Leafs.

Somewhat surprisingly, we find that the Los Angeles Lakers are the best sports franchise in history while the Montreal Canadiens, despite their 22 championships, languish in 11th place. The Toronto Maple Leafs, who have won 11 Stanley Cups, do not even appear on this list. This graph also highlights the lack of parity present in the NBA compared to other sports leagues that Andrew studied back in December. Of the top 8 overall teams, 5 are NBA teams which implies a certain level of dominance amongst an elite group of teams (Celtics in 60s/80s, Lakers in 80s/00s, Bulls in 90s, Spurs in 00s and Heat in 10s). It is important to note that it is also harder to get an outlier result given a larger number of years played, so we would expect the older teams and leagues to have more middling results all things held equal. Now, we will look at it broken down sport by sport. Only teams that have won at least one title will appear on the chart.

NFL

Editors Note 2: The Baltimore Ravens are considered a separate franchise than the Cleveland Browns since the original Cleveland Browns franchise was officially deactivated in 1996 and reinstated in 1999

This data set is largely uninteresting given that the first Super Bowl was held in 1967, when there were already 24 teams in the league. However, the Baltimore Ravens, by virtue of their Super Bowls in 2000 and 2013, combined with their relatively short run in the league, have shot up to the number 3 position ahead of the Cowboys.

NBA

This is a much more interesting chart than the NFL one. The Spurs and Heat are very highly ranked after winning many championships in recent years. The Celtics, due to their championships being less valuable than the Lakers, fell all the way to number 3. The concentration of NBA Championships is extremely concentrated, so after the top 5 the list is heavily populated by teams with one title (like the Cavs) or teams that have been around for an extremely long time (like the Warriors).

MLB

In baseball, we can see some of the teams that have been around for less time have risen to the top, leading to some variance issues. Other than that though, a lot of the top teams have been around for the duration of the league. This is somewhat to be expected given that the teams that have been around the longest are the teams in some of the biggest markets (New York, Boston, Los Angeles, San Francisco and Philadelphia) in the country and they are the ones expected to perform better than a uniform distribution.

NHL

The data for the NHL is undoubtedly the most interesting. The gold standard for NHL franchises is generally considered to be the Original 6, but of those teams only the Montreal Canadiens appear in our top 5. Instead, two franchises that dominated hockey during the 1980s, the New York Islanders and the Edmonton Oilers appear in our top 5 as do the Pittsburgh Penguins, who have had a lot of success in recent years. The NHL data definitely has the most variation between total championships and indices but that is mainly due to the fact that for more than a quarter of the NHL’s history there were only 6 teams who hogged the championships.

About the author

harvardsports

View all posts

8 Comments

  • Why is winning a title with eight teams worse than winning a title with 30? I would argue that if talent is equally distributed amongst a small number of teams, then each team would be much stronger. Spreading the same talent amongst 30 teams simply means teams with diluted talent and lesser quality.

    • Hi Robert,

      Thanks so much for reaching out to us. The reason why we adjusted to take into account the number of teams in the league has nothing to do with talent distribution, as what you point out is entirely correct. If there are 8 teams in a league, that team (in theory) has a 1 out of 8 chance of winning the Championship while in a 30 team league that chance is 1 out of 30. It is clear that it is much easier for a team to win a title with less teams in a league. If there were only one team, then that team would win the title every single year! I hope that helps explain why we adjusted for the number of teams in the league. If you have any more questions, don’t hesitate on reaching out to us!

      -Andrew Puopolo

        • Why is that horrible logic? Makes a lot of sense to me. As Harvard Sports said, winning in a league of 1 is a given. Winning in a league of 2 is pretty easy. Winning in a league of 6 is harder. Winning in a league of 30 is obviously a lot harder than that. Yes, of course some teams are more likely to win than others. The whole point is that some franchises have done a better job over time in putting together those teams. And this is a good way to measure that, adjusting credit for the impressiveness of each title according to league size.

      • Thanks for the answer. While I agree with you theoretically, I find that in any given season, regardless of the sport, all teams in a given league do not have an equal chance of winning the title. Because talent distribution will always be skewed one way or another, there will always be teams at the bottom that are simply fodder for whatever teams are stronger. Now, accounting for this particular imbalance in a study like yours is probably impossible, so your method is probably the best way to do it.

Leave a Reply

Your email address will not be published. Required fields are marked *