There's a lot of really strong dynasty analysis out there, especially when compared to five or ten years ago. But most of it is so dang practical-- Player X is undervalued, Player Y's workload is troubling, the market at this position is irrational, and take this specific action to win your league. Dynasty, in Theory is meant as a corrective, offering insights and takeaways into the strategic and structural nature of the game that might not lead to an immediate benefit but which should help us become better players over time.

What's the Difference Between Early and Wrong?

A lot of people hate social media for a lot of very good reasons. Probably my least favorite thing about it is the way it tends to distort the conversation; only a small fraction of fantasy football managers are on Twitter, but the overwhelming majority of fantasy football writers are, so insular debates on this tiny corner of the internet are often treated as if they're the biggest issues facing the community today. It's also tempting to just sum up "The Discourse" and pretend that it's compelling content when the reality is everyone who is on Twitter already knows and everyone who's not on Twitter doesn't care.

With that said, I always find social media invaluable just as a chance to expose myself to perspectives I never would have otherwise considered. Such was the case this week when a friend of mine asked something I wouldn't have dreamed was in doubt.

if you advocate buying a player and they disappoint that season before breaking out the following season, were you right?

— Cooper Adams (@CoopsFB) October 12, 2023

In my mind, if you identify a player as a breakout candidate and that player breaks out within a reasonable (if not quite immediate) timeframe, then, of course, you were right about them. How could this be a question? But a significant percentage of respondents (around a third at the time of writing) felt differently, and I wanted to understand why.

As far as I can tell, I think this is a difference between projections and heuristics. Heuristics, you might recall, are simple rules of thumb designed to help you navigate complex environments with lots of uncertainty. I am a big believer in relying heavily on heuristics in dynasty. For instance, a favorite heuristic of mine is "talented players tend to produce eventually". (I call this the Dr. Ian Malcolm hypothesis after a speech by the character from Jurassic Park. You might not see the path from here to there in advance, but life finds a way.)

It's hard to remember now, but when the Titans drafted A.J. Brown in 2019, it was considered the worst possible place he could have landed. Marcus Mariota was playing in the final year of his rookie contract and coming off of his worst season as a pro, and the top backup was 31-year-old Ryan Tannehill, who, in six below-average seasons for Miami, had never supported a fantasy receiver who didn't play in the slot (Jarvis Landry, Brian Hartline).

Only Washington produced fewer fantasy points for its wide receivers the year before, and only then because they were decimated by injuries. (They started four different quarterbacks, and the only wide receiver to start eight or more games was noted bust Josh Doctson, who finished his career with just 1100 yards.)

Additionally, Brown had to contend with Corey Davis, a former 5th-overall pick heading into his third season, and Adam Humphries, newly signed to a 4-year, $36 million contract. By any measure available, Tennessee was the worst-case scenario for a rookie receiver. But it didn't matter; Brown produced the 4th-best rookie season of the last eighteen years (behind Odell Beckham, Justin Jefferson, and Ja'Marr Chase) and ranked 5th in fantasy points over 2019 and 2020.

The path to production wasn't clear (no one had "Mariota gets hurt, Tannehill has one of the most surprising breakouts ever, and A.J. Brown produces despite Tennessee ranking 31st in pass attempts because he averages an eye-popping 12.5 yards per target" on their bingo card). But Brown was a good player, and production tends to find good players.

(The path to production became muddled once again when Tennessee traded Brown to the Eagles, a team that ranked last in the league in pass attempts behind a run-first quarterback with unresolved questions as a passer and that already featured two other high-quality targets in DeVonta Smith and Dallas Goedert. But life... found a way.)

At least, that's how I view fantasy football through the lens of heuristics. If your rule is that good players tend to produce (even if you don't know how it will happen), and you say A.J. Brown is a good player, and then A.J. Brown produces, you get to count that as a win (whether the thing that finally unlocks that production happens in a couple of weeks, a couple of months, or a couple of years).

The "Slop Doesn't Count" School of Fantasy Analysis

Many people don't use heuristics to guide their decisions, though; they make projections instead. And for projections, you don't just get to handwave away the "path from here to there" part. If you have "A.J. Brown is going to be good (because Mariota will turn things around and the Titans will pass more)", you don't get to claim credit if the outcome is "A.J. Brown is good (because Mariota got benched, Tannehill had a monster breakout, and the Titans still barely passed the ball but it didn't matter because Brown put up historic per-attempt efficiency numbers)."

I'm viewing dynasty more like soccer, where if the ball goes in the net, the team scores a point (whether it went there because your team made a brilliant shot or because the other team accidentally scored an own goal). But from a projection lens, dynasty is more like 8-ball-- it doesn't matter whether you sink the ball, if you didn't specifically call the pocket, it doesn't count.

It's a foreign mindset to me, but I understand the purpose. If your process was bad, then any good outcomes were merely favorable luck.

In September 2020, Josh Jacobs was the #8 ranked dynasty running back behind Christian McCaffrey, Saquon Barkley, Ezekiel Elliott, Alvin Kamara, Clyde Edwards-Helare, Dalvin Cook, and Derrick Henry. I wrote that I thought he should rank 3rd on that list (behind McCaffrey and Barkley); after all, he was young (22), talented (former first-round pick), and productive (averaged 100 yards per game as a rookie, drafted as the 8th running back off the board per ADP).

You could argue that call was wrong: Jacobs fell to RB20 in dynasty by September 2021 and all the way to RB29 by September 2022. Anyone who bought Jacobs when I wrote that undoubtedly bought at his peak.

You could also argue that call was right: Jacobs finished as the 8th highest-scoring RB in 2020, 12th in 2021, and 3rd (with a dominant, league-winning performance) in 2022. Among the backs available in September 2020, only three scored more points over the next three years than Jacobs (Ekeler, Henry, and Kamara), and only four command more in trade today per FantasyCalc, which tracks completed trades in actual dynasty leagues (McCaffrey, Taylor, Ekeler, and Barkley). If we were re-ranking dynasty running backs for September 2020 with perfect hindsight, Jacobs would almost certainly be the 3rd or 4th back (after McCaffrey, Ekeler, and possibly Taylor).

But whether a prediction counts as correct or incorrect isn't what I wanted to talk about today. I'm (obviously) prone to geeking out over theoretical debates of little practical importance. But in this case, I got so wrapped up in the theory that I initially missed a crucial factor-- a factor that virtually never gets discussed in dynasty coverage, a factor that I wanted to give its due today.

Early is always wrong if you're gone by the time it turns right.

The Half-Life of Dynasty Leagues

One of the most fun parts of dynasty is imagining possible futures, especially futures in which we dominate in perpetuity. But for many leagues, there is much less future than we might think. Some leagues burn out in a conflagration after a controversial move or decision. Others slowly freeze from dwindling interest until there's not enough momentum left to keep them moving. As the great American poet Robert Frost once wrote of his home dynasty league:

Some say the world will end in fire,
Some say in ice.
From what I’ve tasted of desire
I hold with those who favor fire.
But if it had to perish twice,
I think I know enough of hate
To say that for destruction ice
Is also great
And would suffice.

But a league folding isn't the only way that "late" can become "too late". Even in healthy leagues, there's typically some annual turnover. My oldest dynasty league is in its 17th season; the longest uninterrupted stretch featuring the same managers was just five years.

(Do you recognize that some managers abandon their teams but are convinced that it could never be you? Two of the last three vacancies in my home league arose after the long-time manager passed away unexpectedly. The third manager married a woman with teenage children and quit fantasy football entirely so he could spend more time with them. Life rarely takes the path we expect.)

It's hard to estimate just how common it is for leagues to fold or managers to leave; for one thing, the hobby was much smaller and more decentralized fifteen years ago, so the pool of leagues that could have possibly lasted fifteen seasons is a small fraction of all leagues available. For another, those leagues wouldn't necessarily be representative; the last fifteen years saw several unique challenges that impacted league survival rates, especially the lockout in 2011 and the Coronavirus pandemic in 2020.

But I'm not one to let a lack of data stop me from speculating irresponsibly, so I wanted to look at my own dynasty history to get a rough idea of survival rates.

I have joined four dynasty leagues in my life. One folded after a single season (fire), one made it two years before collapsing (ice), and two are still going strong at 11 and 17 years. Here, we encounter our first complication: what are the odds a league folds based on my history? Half of the leagues I've joined have barely gotten off the ground, which implies a 50% overall failure rate. This certainly overestimates things. I've completed 29 seasons, and after 27 of them, the league has continued, which implies a 7% annual failure rate. This is likely an underestimate.

(A league is more likely to still be going in 2030 if it was started in 2010 than it was in 2020; from what I've seen, leagues that fail typically fail quickly, so once you're past the first five years or so, the risk likely declines.)

Let's try a different approach that will hopefully combine the risk of league collapse with the risk of manager departure. What are the odds that a person managing a team today will not be managing that same team one year from now? Again using my history as a guide, dynasty leagues I've participated in featured 308 distinct "franchise-seasons". After 20 of them, the league folded. After 14 of them, the manager stepped away from his or her team. As a result, in 11% of all team-seasons I've witnessed, the manager was not around to benefit from the next season. If we ignore the collapsed leagues, there's been about a 5% chance per team per year that a manager will be replaced.

Using the 11% figure, there is about a 50% chance that a manager will still be managing his or her team after six seasons and only a 31% chance that they'll still be in charge after ten. Using the more generous 5% figure, the average tenure for a dynasty manager would be closer to 13 or 14 years in expectation.

Are my experiences representative? Absolutely not. I have no way of knowing if these estimates are high or low. I do think my surviving leagues are on the higher end for longevity (simply because many fewer people were playing dynasty in 2007 when I started). And it seems plausible to me that 50% of dynasty leagues would collapse within five years, though I wouldn't be shocked to learn the true figure was as high as 80% or as low as 10%.

On the other hand, given the amount of advertising I see for managers to take over orphaned dynasty squads, a 5% annual turnover rate seems very low. (But this might just be sampling bias based on the type of managers and writers I know and interact with.)

But whether it's right or not is probably secondary to whether it's useful, and I do think it's very useful for dynasty managers to make decisions for their team as if there's about a 10% chance there won't be a next year. It's fun to imagine the future, but it's best not to ignore the present while we do so because many of those imagined futures will never come.

And as someone who sometimes gives dynasty advice, I think it's useful to remind myself that of the pool of people who were in position to act on my recommendation on Josh Jacobs in 2020, as many as 20% might have no longer been in position to benefit by the time it paid off in 2022. Debating the difference between "early" and "wrong" can be fun, but it ignores the experience of the substantial percentage of managers for whom they're the same thing.