Dynasty, in Theory: When to Rebuild

Photo: Bruce Kluckhohn, USA Today Sports

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.

Let's Talk About Rebuilding

One of the primary reasons people are drawn to dynasty leagues is the possibility for successes to compound. If you have a great 12-month run in redraft, the slate gets wiped clean at the end, and you have to start again from scratch. If you have a great 12-month run in dynasty, you have the opportunity to build on it and enjoy the fruits of your success for years to come. For most participants, the name is the goal-- building a fantasy football dynasty, a team that dominates the competition for years on end.

It's a lofty goal; pretty much by definition, at any given moment in a 12-team league, 10 or 11 teams will be falling short. A small handful of other teams might be within striking distance, just one or two lucky breaks away from ascending to that hallowed tier. But several other teams will find themselves far behind the league leaders with no obvious path to the top. In situations like those, it can be tempting to tear the team down and rebuild it from the ground up.

There's a lot written about how to go about rebuilding once the decision has been made (I'll likely contribute some thoughts of my own on the subject in the future), but precious little space is devoted to whether to rebuild in the first place. This is a shame because whether to rebuild at all is the more interesting (and more consequential) of the two questions.

I'm the Whether Man, not the Weather Man, for after all it's more important to know whether there will be weather than what the weather will be.

-Norton Juster, The Phantom Tollbooth

Defining Our Terms

As always, I like to begin with definitions to make sure everyone is on the same page. Some would define "rebuilding" as "reducing this year's title chances to increase your chances in future seasons", but I think this is too imprecise; I have a squad that's projected to lead its league in scoring that would qualify as "rebuilding" by this definition because I traded my #3 running back for a rookie pick.

Reducing this year's chances by some amount to increase future chances by some greater amount isn't rebuilding; it's just good team management (and more top teams should be willing to do this). Additionally, "trading current points for future points" fails as a workable definition because current points that are buried on your bench aren't helping you win, anyway.

I think the key feature that defines a rebuilder is not a willingness to reduce this year's title odds but a willingness to reduce those odds all the way to zero. Rebuilders take the current year out of the equation entirely. 2023 production becomes a literal non-factor, unworthy of consideration.

(This doesn't mean rebuilders necessarily want to avoid 2023 production either, though; a rebuilding team would love to roster Justin Jefferson or JaMarr Chase despite a strong 2023 outlook simply because they also have the strongest outlook for 2024 and beyond. The points in 2023, however, are entirely incidental.)

Now: When Should You Rebuild?

The first answer here is: whenever you want to. Fantasy football is a hobby. Hobbies are supposed to be fun. Let us never lose sight of that!

If you think rebuilding is more fun than not rebuilding, you should rebuild; "maximize fun" should be your true north in fantasy football, the direction your compass is always pointing. Similarly, if the thought of a rebuild sounds awful, you shouldn't rebuild. It doesn't matter how "good" or "bad" your team is. It doesn't matter which path is easier or more effective. Do the thing you will enjoy the most.

It's never actually necessary to rebuild. My oldest dynasty has been around for 16 seasons, and there hasn't been a single one of them where I wasn't competing until the moment I was mathematically eliminated. I've missed the playoffs, of course, and in those years, it would have been objectively "better" for my team if I'd given up on the year a bit earlier, but I opted to continually refresh my roster in lieu of ever rebuilding it, and I've been happy with the results.

On the other hand, it's also never obligatory to compete. I've taken a likely playoff team and stripped it down to studs, turning over 25 out of 27 roster spots in fourteen months and finishing 3-10 with a team that originally ranked 4th or 5th in projected points per game.

Given that every choice is strictly optional, the question as posed doesn't lend itself to an objective answer. Since "just do what you want" makes for an awfully short and unsatisfying column, let's rephrase our question to something a bit more objective. How about:

When is Rebuilding A Reasonable Strategy for Maximizing Total Championships?

Now we're cooking with gas. This turns rebuilding from a question about what you subjectively enjoy into a dispassionate cost/benefit calculation. As long as we can make a reasonable estimate of the benefits and the costs, we can arrive at an answer.

I'm actually going to flip the comparison a bit. If "rebuilding" is merely "ceasing to contend" then "Should I rebuild?" and "Should I contend?" are the same question approached from different sides. But I think "Should I contend?" lends itself to neater analysis, so that's the one I'm going to evaluate.

The benefit of choosing to contend is that you may win a championship this season. If our new goal is maximizing championships, then winning championships is a great way to go about it!

Yes, better teams have better odds, and reducing your odds this year to increase them next year can still be a net positive, but I think most managers overestimate the odds that top teams win it all and underestimate the odds a middle-of-the-pack team emerges victorious instead.

"What are your true chances of winning a title" is one of my favorite questions to ask, and I've looked at it from a half-dozen different angles. My best estimate is that, as a general rule of thumb, the better team wins about 55-60% of matchups in the playoffs. If that seems low... well, breaking down the math, reasoning, and historical evidence behind that figure is a column unto itself (an entire series of columns, actually). For now, you'll simply have to trust that this result is robust and replicable.

If we accept this 55-60% win rate, then the best team in the league has a 30-36% chance of winning the championship, provided they earn a bye. Without a bye, those odds fall to 16-20%. Meanwhile, if the very worst playoff team lucks into a bye, it has a 16% shot at winning the championship, and with a playoff berth but no bye, its odds are still 6-8%.

These odds are the best estimate available at the beginning of the playoffs; measuring from the start of the season makes things more uncertain. The team projected to be best in September often isn't by December, and plenty of mediocre-on-paper teams in Week 1 look like juggernauts by Week 14. Uncertainty is bad for favorites and good for underdogs. The greater the uncertainty, the less differentiation you see, with "top contenders" seeing their odds drop and "expected also-rans" seeing their odds rise to match.

Now, if we had a perfect crystal ball, it would always be better to give up early on any season where we weren't going to make the playoffs. But I've been visiting thrift stores and antique markets and magic shops for years, and I've yet to find a crystal ball, or at least one that outperforms the old Magic 8-Ball I got when I was seven.

My first dynasty championship came on a team that was 4-7 with games left against the #1 and #3 highest-scoring teams. I opened each of the next three seasons with a 2-5 record, missing the playoffs twice and coming with a shoestring tackle of winning a second title in the third.

If there was some sort of metric or heuristic that would help me abandon my non-playoff teams earlier without giving up on any eventual playoff teams in the process, I'd gladly use it. But as things currently stand, as long as my team is alive, it's to my benefit to chase the remaining championship odds, no matter how slim.

The most extreme example I've seen of this philosophy wasn't one of my own teams. In 2010, the second-worst team in my league faced a ridiculously easy schedule and lucked into the final playoff spot with a losing record. That team averaged 137 points per game during the regular season, while every other playoff team averaged at least 148, and the playoff field as a whole averaged 158.

To this day, that terrible squad still holds the record for most points scored in a single postseason. It averaged 201.5 points per game and was the highest-scoring playoff team in all three weeks. Your title odds are never zero until your title odds are zero.

Estimating Your Championship Odds

If you want to put some numbers to your championship hopes instead of relying on handwaves and a general "so you're saying there's a chance..." mentality, there are several tools that I find useful.

My go-to for the last decade has been Footballguys' League Dominator, which is often viewed as a tool for in-season management but which contains a bounty of useful dynasty applications. My favorite page is the league power rankings, which provide projected points per game for every team going forward. From here, it's easy enough to look at the current standings and the future schedule and get a good feel for whether or not I have a reasonable chance of making the playoffs.

It's hard to directly translate "projected points per game" into "expected single-game win%", especially because the value varies from league to league. But the League Dominator also estimates odds for the current week's matchups, which acts like a Rosetta Stone. If you're favored by 10 points this week, and the League Dominator estimates your chances of winning at 60%, then in your league, 1 point translates to approximately 1% increased chances of winning a given game.

Other sites offer free resources for evaluating your team's short-term and long-term odds. For instance, if your dynasty league is hosted on either Sleeper or MFL, FantasyCalc will automatically import it and display both projected redraft and dynasty value. These are useful tools, though they're often one-size-fits-all and can't take into account your league's specific scoring and lineup settings like the League Dominator can.

Which is why the most exciting tool for this is one that was just released; in Dan Hindery's most recent monthly trade value charts, he included a new dynasty league evaluator tool that will also automatically import rosters from either Sleeper or MFL. This tool automatically adjusts for scoring and lineup settings to customize values and estimates to your league's particulars. It also includes a handy 2023 Power Ranking tab, which simulates the season 100 times and reports how often each team finished in each position.

Now, 100 is not a lot of simulations, so reported results will vary widely from one attempt to the next. (I entered the same league four times, and the tool had me winning anywhere between 12% and 37% of simulations, which is quite the range!) Simulation is a powerful tool, but it tends toward overconfidence. I wouldn't take the estimated odds as gospel, but they should give you a rough idea of how much relative championship equity you've managed to secure.

Remember, in this framework, we're considering championship odds as the benefit. If we choose to compete this season, those odds are the reward for that decision. But why have I sometimes kept contending despite a 2-5 start and other times folded a projected 4th-place finish before the season even started?

Because benefits are only half the story; to evaluate the decision to compete, we also have to look at the costs.

What Does It Cost To Compete?

A few weeks back, I argued that dynasty managers should view their teams through the prism of value. Every player has value, and the managers who assemble the most valuable rosters over the long run will be the managers who finish with the most wins and championships.