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, 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. (Additionally, it serves as a vehicle for me to make jokes like "theoretically, this column will help you out".)

Happy Tanksgiving!

Given how much I see people complaining about all the turkeys in their leagues recently, I can only assume that Tanksgiving season is upon us. So this week I wanted to do a deep dive on tanking-- what it is, whether it's actually bad or has merely gotten a bad rap, why it happens, and what you can do about it.

What is Tanking?

Part of the reason discussions about tanking often don't go anywhere is that all participants are operating off of different definitions. I've seen some suggest that it's "tanking" for an eliminated team to trade Derrick Henry for future draft picks, or to cut Tom Brady for Desmond Ridder. This isn't tanking; this is just good management. Tanking is not merely making your team worse in the short run; if your short-run prospects are bleak, the best way to become competitive again is to transfer your short-run value over a longer window.

When discussing tanking, I use the following definition: tanking is taking any actions that one otherwise wouldn't take if draft order was not a consideration. There are reasons other than pure draft order to get what you can for Tom Brady or Derrick Henry today. There are not any reasons other than draft order to, say, bench Tom Brady and Derrick Henry in favor of clearly inferior options.

(As an aside: if Team A and Team B are in a dogfight for the first overall pick, and you have acquired Team A's first in a trade, then deliberately setting a bad lineup to hand Team B an extra win would also trigger this definition of "tanking". You're doing something you wouldn't do if not for draft order considerations. In my opinion, this is a feature, not a bug; anything meant to outlaw deliberately losing to improve your own draft pick should also outlaw deliberately losing to improve any other picks you might have acquired.)

I've Seen Some Suggest Tanking Is a Viable Strategy. Is Tanking Really So Bad?

Yes.

... Is That It?

Okay, fine. Some of the reasons that people oppose tanking are self-interested and, therefore bad, of the "if this high pick goes to that team it won't go to me instead" or "if this team gets into the playoffs because it was handed free wins against tanking squads, then I won't get into the playoffs instead" variety. This is a bad reason to oppose tanking because fantasy football is a zero-sum game, meaning anything that is to one team's benefit is to every other team's detriment. Managers giving into self-interested motives is why they tend to veto fair trades that make their top competitors even better. I think trade vetoes are bad because I don't think you should let self-interest scuttle trades. So how is it not hypocritical to oppose tanking despite similar arguments?

Because there are non-self-interested reasons to oppose it as well. I like to call this the "imagine a league where..." test. Imagine a league where making trades to improve your team was possible, and then imagine a league where trading was impossible (or exceedingly difficult). I'd rather play in the former than the latter, so I oppose trade vetoes.

Now imagine a league where tanking was allowed and teams would have several games a year where there was no drama or suspense because the opponent didn't even bother to set a lineup, where the top draft picks did not tend to go to the worst teams anymore and truly awful teams found it difficult sometimes to pull themselves out of the gutter. Does this league sound more fun or less fun? I think it sounds a lot less fun.

Some aspects of fantasy football are zero-sum. A typical 12-team league will produce six playoff teams, one champion, and one first-overall pick a year. The identity of those teams might change, but the quantity is fixed and anything a team does to increase its chances at one of those marks reduces everyone else's aggregate chances by the same amount.

But some aspects of fantasy football are not zero-sum. Leagues where trading is possible tend to be more fun than leagues where trading is not possible. Leagues where bad teams have an easier path to improving tend to be more fun than leagues where bad teams are stuck in the basement forever. Leagues where every game is genuinely up in the air are more fun than leagues where your opponent occasionally doesn't start anyone and you know in advance that you're guaranteed a victory.

Opposing things because they shuffle the names in the zero-sum layer is pointless. Opposing things because they reduce league quality in the non-zero-sum layer is good.

If Tanking Is Bad, Why Do It?

Because it's rewarding. Higher picks are significantly more valuable than lower picks in rookie drafts. DynastyLeagueFootball has been tracking average draft position for dynasty leagues back to 2014. This last offseason, the difference between the #1 pick (Breece Hall) and the #2 pick (Drake London) was the difference between the 13th and the 30th pick in a startup draft. Two rookie 1.01s have made it into the Top 12 overall players immediately after the draft (Ezekiel Elliott and Saquon Barkley), and a third (Clyde Edwards-Helaire) joined them after a preseason injury to his top competition. Even if you don't like the player available at 1.01, you could usually trade the top pick for the second pick plus a substantial windfall. That substantial windfall is the surplus value that tanking got you.

(This is where I add a standard disclaimer that every draft class is different and tiers matter more than picks. The 2021 class, for instance, had a clear Top 3-- JaMarr Chase, Najee Harris, and Kyle Pitts-- followed by a 2-person second tier-- Travis Etienne and Javonte Williams. The biggest gains were therefore from moving from pick #6 to pick #5 or from pick #4 to pick #3.)

From the perspective of an individual team, tanking is perfectly rational. It's one of the easiest and most effective ways to improve your dynasty squad if you're not competing for a title. Ultimately, this is a case of misaligned incentives: the thing that is good for individual teams is also bad for the league.

What Can Be Done About It?

The most common path is for dynasty leagues to merely outlaw tanking, but this often devolves into a game of Whac-A-Mole. Benching Derrick Henry for James Robinson is obviously tanking. But what about benching Antonio Gibson for DOnta Foreman in PPR scoring? Footballguys' projections currently have Gibson 15th for the week and Foreman 26th, a projected difference of 4.2 points. But it's easy to tell a story of why Foreman would be better; he's had three strong games in the last five weeks, while Gibson hasn't even been the starter for Washington since Brian Robinson Jr returned from his injury.

What about when a team sets its lineup on Thursday morning and doesn't update it later when a player who was questionable is downgraded to out? Is this an example of tanking? Or did the manager perhaps just go out of town for the weekend and forget to check back in?

Policing edge cases like this is a nightmare for a commissioner, and the more tightly they are policed, the smaller the edges unscrupulous managers will seek out. As I mentioned above, the entire problem of tanking results from misaligned incentives. Any time it's to a team's benefit to lose, that team will be tempted to find creative ways to accomplish that goal.

As a result, the best solution is to align incentives, to make it so that doing what's in the league's best interest (competing to the best of one's ability) remains in the individual manager's best interest, too. And ultimately that means awarding draft picks based on something other than won/loss record.

There are lots of different systems out there, but my preferred one uses something called "potential points". Potential points are what your team would have scored if it had started the best possible lineup every week. At the end of the year, the team with the fewest potential points among all non-playoff teams drafts first, the team with the second-fewest drafts second, and so on. Then the playoff teams draft based on when they were eliminated, with potential points again serving as a tiebreaker.

This has plenty of advantages. Most obviously, whether Derrick Henry is in your lineup or on your bench ceases to matter, he impacts your draft position just as much either way, so there's no reason not to start him and try to win. An ancillary benefit: potential points in year N correlates much better with actual points in year N+1 than wins in Year N does, which means it's a much better measure of team quality. Potential points does a better job than raw win/loss record at giving the best picks to the worst teams and the worst picks to the best teams.

This isn't to say that potential points is perfect. Like any system, it can be gamed. Bad teams might only carry a single quarterback and defense to reduce the chances of a backup having a big game. Trading productive players for future picks or currently injured players makes your team "look" worse to potential points even though it hasn't gotten any worse overall.

But the ways that potential points can be gamed aren't as damaging to the overall integrity of the league. It's good, actually, if bad teams are trading for more future picks and injured stars.

There are other potential issues with potential points, depending on how you structure them. If the team that is dead last in potential points finds itself on the cusp of the playoffs, the manager might decide to lose on purpose again (knowing that if they miss the playoffs they're guaranteed the #1 rookie pick). These issues can be mitigated, perhaps by allowing playoff teams that don't make the championship game to rise as much as three spots in the draft if their potential point total warrants it (so if a team ranks last in the league in potential points, makes the playoffs, and is eliminated in the wildcard round, it's still drafting 4th overall instead of 7th).

Or you could even eliminate the playoff / non-playoff distinction entirely and just reserve the final two draft picks for the championship participants. If the team with the lowest potential point total reaches the final four only to get eliminated in the semifinals, they can still draft 1st overall. In this case, there's absolutely no incentive to lose (except, I suppose, if a team decided to throw a game in the semifinals to lock in the #1 pick, but who on earth would do that with a 50/50 shot at the championship on the line?)

I don't mean to suggest that potential points is the only possible solution to the tanking problem. There are lots of other systems in use out there with their own proponents. But it is best to clearly identify the source of the problem (misaligned incentives) and settle on a solution that addresses that as directly as possible.

After all, fantasy football is more fun for everyone when everyone's playing.