An analysis of each team’s total titles (Super Bowl wins and pre-1966 championships) vs expected titles. (Packers +7.2, Browns +4.4, Patriots +3.3…Cardinals -3.6)
by Jason Pauley
This analysis shows every current team and their expected championships since their entry into the NFL (or AFL, or AAFC, or APFA), their actual championships, and their over/under vs expected. This data uses some simple assumptions and simple math, to determine each team’s historical performance vs what a random team should have done playing in the same league, against the same amount of teams over the same number of years.
Note: Skip this section and go straight to chart # 2 if you don’t care how the sausage is made.
To illustrate the fairly simple concept I’ll use the Jets and the Cardinals as an example.
The Cardinals franchise has won two championships during the span of their existence, that’s one more than the Jets who have only one Super Bowl / Championship since they entered pro football. For now, the scoreboard tells us:
If you’re reading this, you probably like football enough to know that the Cardinals franchise was one of the original teams; their first year in the league was 1920. You probably also know that the Jets entered the league much later, in 1960. So, with a 40-year head start, the Cardinals should have more championships than the Jets. Okay, easy fix…let’s normalize the data by years in the league. The Cardinals have won two in a span of 102 years, a ratio of 51 years per championship. The Jets have been in the league for 62 years. With this adjustment, the scoreboard still favors the Cardinals, but not as much as the comparison to straight-up titles. Normalizing by year has the Cards only about 18% better.
Cardinals: 51 to 1 ratio
Jets 62:1 ratio
Hold up, though. We normalized by years played, but there’s another important factor at play. In 1947 when the Cardinals last won a championship, there were 10 teams in the league. A random team had a 10% chance of winning it all going into that season. A team playing from 1940 to 1949 should have one championship on average. By the time the Jets won their first and only Super Bowl/Championship there were 26 teams competing for the title; a random team would have a <4% chance of winning it all. Over the entire timespan of the Jets history, they were going up against an average of 27.3 teams per season for the title. The Cardinals had 20% fewer teams vying for the title on a per season basis, with an average of 21.7 over the history of their franchise. Not all seasons are created equal, and this is why I think that a third variable needs to be a part of the equation. In addition to normalizing by years played, every season needs a percentage value that a random team should win the title. That value is 1/X with X being the number of teams in the league. Each team for each year has a 1/X value and the 1/X value accumulates over time from the team’s first year to their last.
For example, a team that plays five seasons; with the first three seasons having 10 teams in the league and the next two seasons having 20 teams in the league would have an expected number of titles of 0.40 (.100 + .100 + .100 + .05 + .05). Earlier, I mentioned that the Cardinals, because of their time in the 30s and 40s, had the benefit of going against far fewer teams than the Jets on a per-year basis. So, when I adjust for this, what does this do to the comparison between the two teams. The Cardinals end up with an expected 5.8 titles based on what a random team would have had in the same timeframe. The Jets have an expected 2.7 titles. Both have underperformed against their expectation, but now the scoreboard shows that the Jets have done better in terms of titles against expected titles:
Jets -1.7 (1 championship, 2.7 expected)
Cardinals -3.8 (2 championships, 5.8 expected)
This Jets-Cards example, although long-winded, shows you how comparing championships alone doesn’t work well (you already knew that), but it also shows that simply normalizing by years in the league has a major flaw. But by combining years in the league with the number of teams in each season, we can have a good indicator of where a team should be in terms of titles and compare to their actual titles for an over/under value.
Here is an example of the Cardinals timeline shown in a graph. You can see that up until 1932, they were ahead of schedule, but have been playing catch-up ever since.
Chart 1: Cardinals Championship timeline (actual vs expected)
What this analysis doesn’t do: This doesn’t evaluate each team going into the season each year and derive an expected championship based on their roster talent vs the rest of the teams in the league for that year. It’s all based on what a random team would be expected to do against a league of other random teams. The other consideration I’m not taking is the number of times in the playoffs as my starting point. That’s a different analysis and would show playoff success rate or a championship to playoff ratio. I am simply calculating what a team is expected to do based on when they were in the league, and how many years they have existed.
Here is the chart showing expected titles in blue and actual titles in yellow. The teams are sorted from best to worst with the best performance relative to expectations at the top. The bar chart to the right shows the gap between the two points.
- The number one team in all-time championships, the Packers (13) is also number one on my list of over/under titles (+7.2). However, in most cases, you will not see the over/under rankings fall in line with total championship volume.
- Any Browns fan who is age 57 or older should feel a little cheated when they look at this metric. The Browns have the 2nd highest Score with +4.4 championships, yet none of those titles came during their lives. The Browns won seven titles in the 40s and ’50s. By their 10th year in the league, they had already won seven with an expected number of 1 for an over/under of +6.
- The Patriots and Bears have similar scores (Pats +3.3, Bears +3.2) but came about those scored in some very different ways with the Bears starting in 1920 and winning early, the Patriots starting in 1960 and ruining most people’s lives in the 2000s and 2010s. Another way to look at a team’s success relative to expected results is the percentage instead of the absolute difference (you’ll see this in the table below). The Patriots six titles vs an expected 2.7 are 124% better than expected. The Bears nine vs 5.8 expected is 54% better than the baseline.
- The last three Super Bowl champions all pulled even or almost even with their expected totals from their most recent SB win. The Chiefs moved to +0.3 (3 actual, 2.7 expected), the Buccaneers moved to +0.5 (2 actual, 1.5 expected), and the Rams -0.4 (4 championships, 4.4 expected)
- The Cardinals, one of the original teams (1920) have, by far, the worst championship history winning only two versus a league-high (tied) expected total of 5.8. Their -3.6 over/under is a full 1.5 worse than the 2nd worst team in the league, the Vikings.
- Many teams would move from negative to positive by winning one more. A few good teams that have a reasonable chance soon are the Titans, Rams, and Bills.
- Some teams who haven’t been in the league long enough to have a full 1.0 expected titles are the Texans (0.6), Ravens (0.8), Panthers (0.9), and Jaguars (0.9). Just one title in the next five years and the Jaguars, Texans, or Panthers are right where they should be.
- The Bengals who just missed out on a Super Bowl win entered the league in 1968. Based on their 54 years, they should be at almost two titles. They came close to cutting the deficit in half this year. Hopefully, they are at the beginning of a run where they can continue their success and get a win or two with Burrow, Chase, and co.
- The Eagles and Lions are tied for the 10th most championships in history with four each. However, because of their long timeline in the league and being the in league with fewer teams in their early years, they are ranked 25th and 26th in over/under against expected titles. Conversely, the Ravens with half as many titles (2) are 9th best all-time for over/under with a +1.2 because they came into the league in 1996 (I know…something, something, Browns).
The table below shows the same data above, but in a table format so you can view the data along with additional detail like the years they won, years in the league, years/titles, etc.
Hopefully, you find this analysis/context useful when looking at teams’ total Super Bowls and pre-1966 championships. Knowing how few teams there were in the league from the 1920s up until the first Super Bowl in the 1966 season, was eye-opening to me. The average number of teams vying for a title in the first 40+ years was only 12. This provides much-needed context to some of the Packers, Browns, Bears, and Giants titles that they accumulated early on. But this metric adjusts for that and the over/under becomes a useful data point. I think it’s a fun way to look at a team’s success and to use in sports debates/discussions. Use this if it’s helpful for your narrative. Ignore it if it hurts your narrative. I hope you find you found some of this data interesting.
Note: For a few years there were separate titles in separate leagues: The NFL and AFL had separate titles from 1960 to 1965. Then in 1966, although they still had an NFL and AFL title, the winners would meet in the Super Bowl to determine the sole champion. From 1946 to 948 the NFL and AAFC had separate titles.