Photo: 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 Video Games

This will probably come as a shock to anyone who has ever played a video game before, but occasionally games will ship with glitches. Some of these glitches even impact the gameplay! For example, say you have the ability to drop an item that you're carrying, but if you open the inventory at a precise moment, you also hold onto a copy, duplicating the item in question. (Item duplication glitches are some of the most famous-- and most common-- glitches in games.)

When I was a kid, if a game shipped with a glitch there was nothing to be done, that glitch was now part of the game (for better or for worse). But nowadays, since everything has a hard drive and an internet connection, game developers can create patches for video games even after they've shipped to fix some of these glitches if they're so inclined.

For multiplayer games, whether to patch or not patch is a fairly straightforward question. Anything that gives someone an unintended advantage gets patched so it doesn't ruin the gameplay experience for everyone else. But for single-player games, patches are much more controversial. Many players will argue that how they play doesn't impact anyone so developers should leave the glitches lying around and allow players to choose whether to use them or not.

But developers have found that glitches are like Chekhov's gun -- an unpatched exploit above the mantel place in the first act will go off in the third act. While players can choose to use a glitch or not, developers have consistently found that it feels bad for players to knowingly play sub-optimally. If they discover a helpful glitch, they'll either use it and, if the glitch is severe enough, ruin the intended experience, or else they'll knowingly avoid it and often feel bad (or at least conflicted) about that decision.

Because of this, standard practice for developers is to patch any glitch that has a sufficient impact on the intended experience to protect players from themselves. And tests consistently find that (regardless of gamers' stated preferences) this results in a more enjoyable gaming experience.

Fantasy football leagues are not video games, but many of the lessons of good game design carry over. People crave a challenge that's tough but feels fair. They want to strive and win or lose on their own merits. They want to feel some sense of progress. And sometimes, just sometimes, you have to design the game in a way that protects managers from themselves.

With that in mind, and since it's Tanksgiving season (I warned you last week that pun was coming), let's talk about tanking.

Tanking: Is It Good?

No.

Tanking is not fun for the league as a whole for obvious reasons. But even setting aside all the prosocial arguments against tanking, it's worth realizing that it's not fun for the individual teams that decide to tank, either!

(Sure, getting the #1 pick and dreaming about what to do with it can be fun, but given two managers that each landed the #1 pick, the manager who tried to play spoiler all season probably had a more entertaining year than the manager who just set a lineup in September and checked back in January.)

So why do teams tank? Because an unpatched exploit above the mantel place in the first act will go off in the third act. Tanking is sometimes optimal play, and for many, it feels bad to knowingly play sub-optimally. I understand the temptation. Many leagues try to stop tanking by banning tanking, but the problem with this is you're still relying on managers to knowingly play sub-optimally, and this often results in a conflict around the edges of the definition, with managers testing the boundaries to see how much they can get away with.

It's better to take a page from video game design and just patch the exploit so it's not sitting there tempting players. The problem is that tanking is sometimes optimal, so the solution is to make it so tanking is never optimal. How do we do this? We decouple draft position from losing.

Now, there's a good reason we award draft position based on losses; we want the worst teams to get the best picks so everyone has a chance to compete in the future. But losses are only a proxy for badness, and all proxies can be gamed.

What we need is a proxy that does a better job of identifying actual badness, one which doesn't suck all the fun out of a league once managers start trying to game it. To that end, here are a few potential solutions, starting simple and getting more exotic as we go:

Award Picks Via Lottery

From a game-design standpoint, this solution fails both goals. It makes it less likely that the worst teams get the best picks, which reduces parity. Additionally, assuming the lottery is weighted so that the worst teams have the best odds, it doesn't even remove the motivation to lose. I'm not a fan.

Use Potential Points

Potential points is a measure of how many points your team would have scored if you started your best possible lineup every week (such as if the league were a Best Ball league). Potential points are a much better measure of team quality; potential points in Year N correlates much more strongly to wins in Year N+1 than wins or even total points in Year N does. (In large part, this is because potential points is one of the few measures of team quality that considers bench strength, as well.)

If draft position among non-playoff teams is decided by potential points, there's no benefit to starting a weaker lineup, and there's no incentive for non-playoff teams to lose an extra game or two at the end of the season, so they can feel free to play spoiler without worrying about hurting their team. Additionally, it creates much more parity than awarding based on wins.

Potential points can be gamed. (All proxies can be gamed.) But the methods of gaming potential points aren't really unfun for the rest of the league. The best way to reduce your potential points is to just cut your best players, but that is counterproductive in a rebuild. Trading away aging stars will greatly assist with the task of rebuilding. Buying injured players and future picks helps rebuild and improves potential points tiebreakers. Only carrying one defense to reduce depth helps, too.

All of these are things rebuilding teams should be considering anyway. This is the solution I use-- it's easy to understand, easy to administer, and probably achieves the best balance of both objectives.