Here are my thoughts on the questions that I posed before. Looking at the data three teams bring up question marks about their place as a dynasty in the NFL. The Bears of 84-88 who were not classified as a dynasty the first time I looked at the data. The Bills of 88-93 who were counted as a dynasty, yet never won a Super Bowl during this time. And the Rams of 99-01, whose rise to power was dramatic however their team was quickly eclipsed by the Patriots. The Rams also never reached the height that other teams that are considered a dynasty did. Below is a plot of those teams with other relevant teams plotted around them. Each point on the plot is the teams' five year average of playoff performance, with a weighting further rounds more heavily.
1. Can there be more than one dynasty at a time?
I do not think so. I think that there can be two really good teams at the same time, but I don't think that the second best team over a span of time can be called a dynasty. This leads me down a path that I don't particularly like though, since I do not like having two dynasties at the same time I must ignore a team that is building until it is the best team, this means that when I look back at when a team is a dynasty I can not include the previous years that it was building up to be the dynasty team.
The larger implication of one dynasty team at a time has is that it is possible that there could be two teams that trade off Super Bowl wins over a couple of years but only one team can be named the dynasty for that time. This is my main argument against the Bears being a dynasty team. It turns out that during the same years that they were good the 49ers were even better. You can see in the chart above that for every spot along the graph the 49ers were still averaging a better playoff performance for the same time. More about this in question three.
2. How important is the regular season, how important is the playoffs?
I agree with most people here. The regular season is important but performance in the playoffs is what makes the team a legend. Lance made a good point of the 15-1 Vikings from 1998, they had an awesome regular season but history generally pushes that off to the side when talking about the 98 season, we know the Falcons went to the Super Bowl that year, we don't all know the Vikings great regular season.
I have though about adding a small component of the analysis that takes into account how well a team did over their regular season, but since I don't have the data already typed up I believe that it isn’t that important.
3. You can't just look at one year at a time, so how many years do we look at?
This is probably the trickiest question here. First because the span of years to look at changes which teams are considered the best, for example for we look at a four year average the Bears do become the best team for two years. Secondly it really seems like there should be a mathematical way to define how long of a span than we should look at, however I can't think of a way that satisfies me completely.
First let me explain the way which I decided upon 5 years. The initial thoughts were because it just seemed like a good round number to deal with; however hunches are not good since we are trying to let the data do the talking for us. So the idea that I came up with is to look at the length of playoff streaks (years in a row that teams went to the playoffs). Below I made a chart of the length of all the playoff streaks.
This seems to suggest that we should look at 4 or more years since a playoff streak of that length is better than the average streak. However if we look at playoff streaks that allow one year skips (so over four years a team goes to the playoffs 3 times, it still would count as a three year streak), we see that 5 or more years is above the ordinary length of a streak.
I will argue that the "One Skip" is a reasonable way to look at this problem, since we agree that missing the playoffs one year during a dominant stretch should not disqualify a team for being considered a dynasty.
Now if we look at a chart of 5,6 and 7 years we find that there isn't too much change among the three graphs, however the longer the span the smoother the ride at the top seems to be. Also with the 7 year average we see then the Bills and Rams are knocked out of being the top team. Does this give credit to the idea of a 7 year span?
4. Is winning the championship the only quality that makes a dynasty?
When I looked at this data I used a scoring system that favored winning in the playoffs. A team got 1 point for losing the Wildcard game, 2 points for losing the Divisional game, 3 points for losing the Conference Championship game, 5 points for losing the Super Bowl and 7 points for winning the Super Bowl. This was my way to weight the teams performance where getting to the Super Bowl was more important than just getting to the playoffs.
Earl suggested that winning the Super Bowl is the only thing that defines a dynasty team, he suggested that if a team wins it 3 times out of 6 years then that would be grounds for declaring a dynasty. Let's see how a chart of winning the Super Bowl looks (over 6 years):
Using this criterion we find three dynasties, the 49ers the Cowboys and the Patriots. I don't think that anyone would argue that these three teams are dynasties, so this method does work. However I believe that it is just a bit too conservative when looking for the good teams (also it is so straight-forward that it isn't much fun to talk about ^_^).
5. Can a dynasty last for only one season?
Take a look at the Steelers and the Raiders here. Both teams are the best team only one year, the Raiders in 1983 and the Steelers in 2008.
My initial thought was to disallow these two teams from being considered a dynasty because they were on the top for only one year. However, I have thought about this a bit more and I think that they are OK. Each point on the plot is actually 5 years of data combined, so that means that over the last 5 years that team was the best, not just that one year. Even if a team later eclipses the teams performance that should not take away the accomplishments made for those five years. Say the Steelers get overtaken next year by the Colts. That still means that for the last 5 years the Steelers have been the best team. (I have run the numbers and basically the only way the Steelers to lose their top spot is for the Colts to make it to the Super Bowl next year.)
That settles it right? :)