Dynasty Rankings: Defining a Dynasty

The New England Patriots have been atop the league for the entire 21st Century. From the scrappy underdogs that took down the Greatest Show on Turf, through the 16-0 juggernauts, up to 28-3 and beyond, there has not been a franchise more aligned with victory than the Patriots. But now, they suddenly seem mortal -- they were taken apart by the Titans in the wild-card round, and now face an uncertain future without their Hall of Fame quarterback under center. The departure of Tom Brady ends the Patriots' dynasty … right?

The Kansas City Chiefs are ascendant, celebrating their first Super Bowl victory since 1969. Andy Reid has turned the Chiefs into a perennial double-digit-win team, Patrick Mahomes has already earned the first of what promises to be many MVPs, and the team is rocketing up towards surefire superstardom. If anything, the team seems poised to just get better and better over the next few years, as their young talent continues to mature and develop. The 2020s will belong to Kansas City, home of the next major NFL dynasty … right?

An NFL dynasty is very difficult to define in real, concrete terms. It's one of those things where you know it if you see it -- it's a subjective label, pinned to teams based on how much they seem like the measuring stick by which the rest of the league compares themselves. Furthermore, the application of the label varies a ton from person to person. Were the Peyton Manning Colts a dynasty, battling out with Tom Brady's Patriots for control if the AFC, or were they just challengers who could never dethrone the king? Were the Steel Curtain Steelers the only dynasty in the 1970s, or were the Dallas Cowboys, Oakland Raiders, and/or Miami Dolphins on their level? Do the 1990s Cowboys count as a dynasty despite their relatively short-lived period of success? For that matter, is the Patriots dynasty a 20-year long uninterrupted reign, or two smaller reigns divided by Brady's ACL tear and the brief period of time when Mark Sanchez was a thing?

With the NFL's 100th season in the books, it seems as good a time as any to try to answer those questions in a systematic way. Plus, hey, more things to argue about in what is turning into a very strange spring and summer are always a good thing!

There have been many attempts to try to define a dynasty before, some of them really quite good. Brad Oremland created a system to evaluate teams in eight-year chunks, which produces a very solid list, even if it can't quite wrap around the entirety of the Bill Walsh 49ers or Bill Belichick Patriots. Over on 538, Neil Paine used their ELO rankings and a two-Super Bowl entry requirement to calculate the dynasties that performed most above average. There are dozens of ways to try to untangle this particular knot.

One of the key things to take into account when trying to grapple with dynasties is the idea that it's not a binary concept. There is no "you must be this successful to wear the crown" line; a point where one win makes you a dynasty on par with the best of the best and one loss makes you an afterthought. Instead, it's a sliding scale, where teams can look more or less like the platonic ideal of dominance. No matter where you draw the line, I think we can all agree that the Belichick Patriots, what with their six Super Bowl rings and their 17 division titles and their stranglehold over the AFC, are more dynasty-esque than the Peyton Manning Colts, who won one Super Bowl, went to another, and shattered offensive records. Those Colts, in turn, are closer to a dynasty than, say, the Joe Flacco Ravens, who won a Super Bowl and had some very solid seasons before descending into mediocrity. And those Ravens are certainly more of a dynasty than the Jim Harbaugh 49ers, who never could bring a title home and ended up collapsing under their own weight. Exactly where you draw the line between dynasty, near-dynasty, and just a run of good results is ultimately a subjective choice, but we can agree on a sort of rough order to put teams in.

So, what we need, then, is a systematic way of assigning values to a team's run of success, with teams building up scores based on how long and how high their regimes can go. And I want to be as value-agnostic as possible in this process; of course, we're going to have to make judgment calls throughout to determine how much a Super Bowl is worth versus an undefeated season and so on and so forth, but I want to give teams as many paths to dynasty points as possible. That means we're not going to have any requirement for titles won to qualify for points. While the best teams are going to win a ton of championships, I do want to give teams like the '70s Vikings and '90s Bills a chance to compete -- after all, aren't four consecutive Super Bowl appearances more impressive than one Super Bowl win and a bunch of missed playoff chances? I also don't want to define a team's run as lasting any set period of time. The '90s Cowboys won three Super Bowls in four years and then burnt out; the '70s Cowboys didn't win as many titles but stayed near the top of the league for much longer. One is going to score higher than the other, but I don't want to artificially benefit anyone by assuming what a great team's path looks like. This means that we'll need a way to define the start and end of a run, as well.

To do this, we'll turn to the godfather of sabermetrics, Bill James. Back in 2012, James wrote a piece to define the (at the time) 37 baseball dynasties, developing an accounting system to give credit for various accomplishments, creating a "dynastic running score" to evaluate runs of seasons, and coming up with a way to determine when that run ends. Perfect! That's exactly what we need. All we need to do is take his rules, cross out the word "baseball" and write in "football", divide things by 10 to go from 162-game seasons to 16-game seasons, and then crunch the numbers. And then we get…

Garbage. That is, unless you consider the '80s Rams a dynasty-level team; after all, they usually finished above .500 and made the playoffs year after year as the second-best team in the West! No, it turns out baseball is different from football, so we'll have to take James' idea as a jumping-off point rather than gospel. This isn't the first time someone's tried to do this, of course; Joe Dimino had his own take on the system the year after James' came out, and the concept of dynasty points are at the heart of Oremland's system, too. Our version will be different, as we'll end up making different decisions about what to include or not include, but good teams will bubble to the top no matter which accounting system you use.

However, here at Football Outsiders, we have our own stats to help us sort the proverbial wheat from the chaff. Dynasty points are a counting stat; they measure the quantity of dynastic goodness a team put up -- how many championships, how many division titles, how many wins and losses. This is all very good and very important, but not all champions are made equal. Both the 1991 Redskins and the 2011 Giants ended their seasons hoisting the Lombardi Trophy, but that 1991 Washington team would have run circles around the 2011 Giants. Washington finished their season with a 56.9% DVOA; the Giants with an 8.5% mark. While those teams had similar outcomes to their seasons, their quality was obviously wildly different, and any list of dynasties should take that into account.

What we are unveiling this month, then, is really twofold. In this article, we're going to be listing every team that has ever racked up at least 10 of our dynasty points -- all the dynasties, near-dynasties, and could've-been dynasties that ever played the game. We'll show which teams had the most quantity of greatness -- the loaded trophy cases, the division title banners, and the boxscores to back up their run of dominance. This is the list of the biggest dynasties the sport has ever seen.

And then, in a series of articles that will be released over the course of this month, we'll look at each one of those dynastic teams, and blend in our DVOA (or estimated DVOA, for teams before 1985), producing a ranking that factors in a team's quality beyond simple wins and losses. Short-lived dynamos will pass long-term compilers, and we'll eventually crown the best run any NFL football team has ever had. That will be the list of the best dynasties the sport has ever seen.

I'm sure this will be entirely uncontroversial and at the end, everyone will nod their heads and agree on everything.

Here are links to the write-ups on all 56 of the dynasties that we ranked:

• No. 1-10 (May 28)
• No. 11-20 (May 26)
• No. 21-30 (May 21)
• No. 31-40 (May 19)
• No. 41-50 (May 14)
• No. 51-56 (May 12)

Dynasty Points

So, how does a team get these dynasty points, then?

Teams get credit for what they do in a single season. This is strictly based on wins, losses, and titles, not taking into account degree of difficulty or quality of opponent. A team can't score in multiple categories; if they qualify for six points, they don't also qualify for five, four, three, and so on. The highest score counts.

A team earns 6 points if they win a championship while finishing with a winning percentage of .8125 or higher. (Thresholds for win-loss percentages will be explained later.) This is the highest score a team can earn in any one season; the result of stomping through the regular season with little resistance and going on to win the championship at the end of the season. Recent teams to earn six points include the 2017 Philadelphia Eagles, the 2016 New England Patriots, and the 2013 Seattle Seahawks.

A team earns 5 points for any other championship win. Recent teams to earn five points include the 2019 Kansas City Chiefs, the 2018 New England Patriots, and the 2015 Denver Broncos.

A team earns 4 points for losing a championship game while finishing with a .8125 record or higher -- the great seasons that fell short at the end. Recent teams to earn four points include the 2019 San Francisco 49ers, the 2018 Los Angeles Rams, and the 2017 New England Patriots.

There are two different ways to earn 3 points. The first is to lose a championship without hitting that .8125 mark. Recent teams to earn three points that way include the 2016 Atlanta Falcons, 2014 Seattle Seahawks, and 2012 San Francisco 49ers.

The second way to earn 3 points is to win your division with a .8125 record or higher, while still losing before a championship game. In 2019, the Baltimore Ravens, Green Bay Packers, and New Orleans Saints all earned three points this way.

There are two different ways to earn 2 points. The modern way of doing it is to win your division and advance beyond the wild-card round, be it by a first-round bye, by beating your first-round opponent, or by playing before the NFL introduced the wild-card round in 1978. In 2019, the Houston Texans were the only team to score two points this way. In 2018, it was earned by the Kansas City Chiefs, Dallas Cowboys, and New Orleans Saints.

The second way to earn 2 points is to finish with a .8125 record or higher without winning your division. This has never happened since the 2002 realignment; can you imagine a 13-3 team somehow not finishing atop a four-team division? It did, however, happen from time to time in previous divisional alignments, so it stays in as mostly a historical way to earn points. The 1999 Titans would have been here, except they went on to go to the Super Bowl and earn more points. Instead, the last team in the NFL to score here was the 1967 Baltimore Colts, who finished 11-1-2 but still finished behind the Rams in the Coastal Division; they didn't even get to go to the playoffs because they lost the old "point differential in head-to-head games" tiebreaker, which was the rule at the time instead of the modern head-to-head record tiebreaker. Teams like the 1968 Kansas City Chiefs and the 1963 Green Bay Packers also fall here.

Any other playoff berth earns you 1 point. It doesn't matter if you lose in the wild-card round or advance as far as the conference championship; if you don't have a great record and don't make the Super Bowl, you end up with one point. In 2019, the New England Patriots, Buffalo Bills, Tennessee Titans, Philadelphia Eagles, Minnesota Vikings, and Seattle Seahawks all earned one point this way.

You also earn 1 point for finishing with a .643 record or above, even if you fail to make the playoffs -- 11-win teams still get credit nowadays even if they end up short. This is mostly a catchall to save really good teams who get blocked out of the playoff race because of weird tiebreakers or unfair division settings; it will almost certainly become a non-factor now that the NFL is adding two extra playoff teams to the mix. The only two teams to earn one point here in the 16-game era were the 2008 New England Patriots and the 1985 Denver Broncos, both of whom missed the playoffs despite going 11-5. Before playoff expansion, however, this was much more frequent; 13 teams earned the bonus in the '70s alone.

Ending a Dynasty

So that's how you gain points -- but how do you lose them, and when do you stop accumulating them? If things never reset to zero, teams like the Bears would be sitting on hundreds of dynasty points at this point, and figuring out when dynasties begin and when they end would be hopeless.

A dynasty begins in the first year a team earns any dynasty points. That starts their running total, which will continue to climb as they rack up achievements. It ends when their total goes back to zero -- or, rather, when their total goes to zero, their reign retroactively ends the last time they had a qualifying season. A team can be sent back to zero in a couple of ways.

A team loses two dynasty points in any year where they fail to earn any -- there are no "zero-point" years. If they also have a losing record, they lose three points instead. You can't drop below zero, but you can sure wipe out a few years of wild-card exits with one bad season.

You can also go straight back to zero if you have multiple non-qualifying seasons in a row. In the original Bill James model, this required three seasons with no dynasty points, but that is way, way too soft for NFL teams. The average baseball career lasts 5.6 years; the average football career lasts 3.5. A three-year window ends up connecting teams which have little to do with one another; part of what makes a dynasty a dynasty is continuity on the roster and the coaching staff. A baseball career lasts almost exactly 1.5 times as long as a football career, so we can cut that requirement from three seasons to two. That gets us most of the way there.

We do need to make one more tweak, however. One-point seasons can artificially extend a team's reign; it doesn't feel right for a 9-7 wild-card round exit every couple of years to keep an era together. So, for our purposes, two zero-point years count as a reset. Alternatively, a team can replace one (but not both!) of those zero-point years with a year that qualifies for 1 point in only one way -- which, in modern times, means an uneventful playoff berth. At that point, their score resets to zero, and they have to start from scratch yet again.

An example might come in handy here. I'm a 49ers fan, so there's obviously nothing in the world I enjoy more than marking the suffering of my franchise is a numerically consistent manner. Let's take their 21st-century history, and work out their various eras -- which, conveniently, showcase the three different ways a team can fail to hit 10 dynasty points.

The 49ers were a 12-4 wild-card team in 2001, which earned them one point. The next year, they won the NFC West at 10-6, which earned them two more, bringing their total to three. But in 2003, they slipped to 7-9 -- that's a losing record, costing them three points and bringing them back down to zero. So that 2001-2002 Jeff Garcia/Terrell Owens 49ers squad scores three dynasty points, and doesn't qualify for our main table.

The 49ers don't score any more points until Jim Harbaugh comes around in 2011, where they go 13-3 and win the NFC West -- that's three points right there. 2012 goes even better, as they go 11-4-1 but get to the Super Bowl … and lose. That's three more points, for a running total of six. In 2013, they go 12-4 but finish second behind the Seahawks in the NFC West, giving them one more point, for a running total of seven. In 2014, they peter out to an 8-8 record, which does not qualify for dynasty points and costs them two, bringing their running total back down to five and counting as their first strike. And then in 2015, Jim Tomsula's squad finishes 5-11, which is their second strike, thus ending the run. So, that means the Jim Harbaugh era goes from 2011-2013 (the last year they earned points), and finishes with a dynasty score of seven points, their high-water mark in the era. They also do not qualify for our main table.

With a 13-3 record and a Super Bowl loss in 2019, Kyle Shanahan's 49ers (2019-2019) are sitting on four points of their own. That era isn't over yet, and we don't know when it will be. If the 49ers do not earn any dynasty points over the next two years, then their run will retroactively end in 2019. But we won't know that for sure until the 2021 season is over; a bad year in 2020 would not be enough in and of itself to end the current 49ers era. You can only really pinpoint the end of a team's run in retrospect.

The Dynasties

That's enough methodology. You're here for the big table with all the teams in it, I'm sure.

In the NFL's first 100 seasons, 56 different teams earned dynasty scores of at least 10 points. The following table lists them all. It includes the number of seasons each era lasted, their high point in dynasty points, and their win-loss record. Then you have their championships, including conference and divisional titles for modern teams. Finally, you have their DVOAs -- both the average across the entire run of the era, as well as the average of their top five seasons. We remember great teams by their high points, after all, so that last column gives a better look at which teams shined the brightest. Teams with runs of less than five seasons have 0.0% seasons added as needed to penalize their top-five DVOA.

You may also notice that DVOA only goes back to 1985, and yet about two-thirds of the teams on the list predate that. To fill in these gaps, we go to two sources. A few years ago, Andreas Shepard produced estimated DVOA for all seasons from 1950 to 1988. This has a correlation of about 0.95 with actual DVOA, so that's good enough as an estimate for team quality back through the era of free substitution. Before that, we don't even have enough data to run Shepard's estimations, so we instead use Pro Football Reference's Simple Rating System -- a rating that takes into account average point differential and strength of schedule, both of which we have going all the way back to the beginning. In recent years, it has about a 0.90 correlation with DVOA, so we can use that as a rough guess of team strength in the earliest days of the league. When you start talking about "average performance" in a league where teams fold, merge, and form at the drop of a hat, you're getting into murky waters anyway; this is about as good as an approximation as we are going to get.

For one system stretching across multiple eras, season formats and levels of professionalism, that's not half-bad. You can quibble some over the tail end of some of the decade-plus levels of success, and there will be people who feel the ten teams on the table without any championships don't belong in a discussion of the best teams of all time. In general, however, you'd be hard-pressed to come up with a better list of the top runs in NFL history.

The question becomes how many dynasty points are needed to turn a team into a full-fledged dynasty. If you're a big-tent guy and think all 56 teams qualify, you're talking about an average of four dynasties active at any one given time. That does seem like a lot; if you go into a season with four teams atop the Super Bowl odds, you'd most likely be talking about a league without a dominant force, not one with four dominant forces. Teams like the '90s Chiefs or '00s Seahawks were good, regular playoff contenders, and deserve to be honored and respected. But I don't think many people were holding Matt Hasselbeck or Steve Bono up as the standard by which all other teams must be measured.

If you're a small-tent guy, you might draw the dynasty line at 25 points, which limits things to a very prestigious club indeed. Yes, the Purple People Eater Vikings slip in without a title to their name, and you somehow get four teams in the '70s squeezing their way into the list (the '70s were a very unbalanced time in the NFL), but all in all, it's a club of the elites of the elites. Keeping the line there means there's an average of about one dynasty active at any given time, and everyone else is struggling to catch up to them.

As I said above, I don't view dynasties as a binary yes/no thing; it's more of a gradual trend from sure-fire, no-doubt dynasty down to more questionable candidates as you slide down the list. If I had to draw a line somewhere, I might go down to 19, sliding the line right between the early '70s Dolphins and the 0-4 Super Bowl Bills. I think that's the point where you start going from teams that dominated the top of the league to more conference- or division-specific dominance, or teams that couldn't keep their quality up over extended periods of time. Drawing the line there gives you an average of just over one and a half dynasties in action at any given point in time, which feels just about right -- the Cowboys/Steelers battles in the '70s, the 49ers/Cowboys clashes in the early '90s, or the Colts/Patriots battles in the '00s are all examples of multiple great teams crashing into one another, and it's alright to say that both of them are dynasties, even if one ended up outshining the other when all was said and done.

So we have our list of dynasties and near-dynasties. But we promised not just a list of teams from most to least dynasty-like; we said we were going to use DVOA to rank them by quality as well. We'll do that, as well as a deeper dive into the ins and outs of the greatest teams, over the next couple of weeks, covering each and every team listed here, why they end up where they do, and what could have happened to boost them up the rankings.

Clarifications

Alright, let's explain some of the choices we made up there, and what exactly does and does not count.

First of all, what counts as a championship? Obviously, the Super Bowl counts, as does the NFL Championship between 1932 and 1965. Those are the easy ones. The AFL Championship is a tougher nut to crack. To say the 1960 AFL was on par with the NFL is just not true; it was basically a league full of expansion teams. To give the Oilers as much credit for winning the AFL Championship that year as the Eagles get for winning the NFL Championship doesn't really reflect the true level of accomplishment there. However, the AFL improved quickly, taking advantage of an under-utilized talent pool (read: signing players out of historical black colleges and universities) and competing with the NFL for draft picks thanks to owners with deeper pockets than their NFL counterparts. The end result is that the AFL did reach parity, or something approaching it, by the time of the 1970 merger. The problem is figuring out when that happened, and how to appropriately weight the AFL title games up until that point. This is a hard problem, and we're going to solve it by not worrying about it. We are going to treat the AFL Championship on par with the NFL Championship. If you want to subjectively bump down the '60s Oilers, Chargers, or Bills as a result, you have an argument to do so -- and we'll point that out when we dive into each individual team.

We are not, however, counting the AAFC Championship from the 1940s -- we're looking at NFL dynasties, not professional football dynasties, and the AAFC never got on par with the quality of the NFL at the time. The AAFC title goes alongside the USFL and WFL titles as minor leagues we're not going to concern ourselves with. This affects the Cleveland Browns of that era, who were the dominant team in the AAFC and then turned right around and became the dominant team in the NFL; you can subjectively bump them up a few notches if you'd like. It also affects the San Francisco 49ers, who were the clear second-best team in the AAFC, and then turned right around and … didn't do much of note in the NFL for seven years. They're slightly less hard done by, there.

We're also crediting the NFL champion from 1920 to 1931 as the winner of a championship game, even though none actually occurred -- they're the champs, playoff game or no playoff game. Slightly more controversially, we're crediting the runners-up from 1920 to 1931 as the loser of a fictitious championship game. We want to give out those three- and four-point scores to teams from the '20s, and we're not going to let the unorganized manner of the league at the time stop us! This mostly affects the Chicago Bears (five-time runners-up from 1920 to 1926), so you can mentally bump them down a notch if you don't think that's a fair judgment.

Finally, the reasoning for the .8125 and .643 are simple. The former is a 13-3 record in a 16-game season; the latter a 9-5 record in a 14-game season. Since the 1970 NFL merger, these roughly define the cutoff points for the top 5% and top 25% of teams. Those are solid cutoffs for differentiating between above-average teams, good teams, and great teams, and to award extra points to put them above other teams with the same level of postseason success. When we move to a 17-game season, those marks will stay the same, requiring a 14-3 or 11-6 record to be earned.

Problems

No system is perfect, and this one is no exception. Trying to fit 100 years of history into one system is a difficult task. Comparing the semi-pro, unorganized league of the 1920s to today's billion-dollar industry isn't comparing apples to oranges; it's comparing apples to a slab of bacon. I did experiment with different weights for different eras, but any improvements there were minor at best, and the relative simplicity of one system for all seasons won out in the end.

Because so many of the metrics for success are based on playoff runs and division titles, it is slightly easier to earn points today than it was at the beginning of the league. Today, the NFL boasts eight divisions, but the NFL didn't bother dividing their teams up at all until they split into Eastern and Western Divisions in 1933. It wasn't until 1967 that the NFL split into four divisions and introduced an extra round of the playoffs, and you can see that effect in the table above.

The average era on that table that started in 1967 or later lasted 7.5 seasons and scored 17.8 dynasty points; teams whose run started before 1967 averaged 6.2 seasons and 17.0 dynasty points. In a different environment, some of those Packers, Bears, and Giants gaps in the '30s and '40s could have been bridged by winning a weak division or pulling off a divisional round playoff upset. Those opportunities didn't exist then, and so you do get some of those choppier dynasties in ye olden days. These extra opportunities probably helped some of the more modern teams scrape onto the bottom of the table with ten-point reigns, as well -- since the 1990 expansion of the playoffs, the average team on the table has a 6.5-year reign and 15.7 dynasty points. Some of that may be from smaller divisions or more playoff slots. Some of it, however, may be due to the salary cap and free agency -- the Legion of Boom would have likely stuck together for a decade in the '70s, rather than leaving to find better deals elsewhere. Part of the fun in doing an exercise like this is seeing different trends throughout the years, and at least a decent chunk of the differences you can find scrolling through the table chronologically comes from how different franchises were set up at various points over the previous 100 years.