Playing to Win with our Democracy

Haven’t done one of these in a while haha. Grad school man, it does eat up one’s time and mental energy. Relatedly, I kind of meant to write this story a week ago. Oh well.

So when Joe Biden unveiled his “framework” for the Build Back Better Act some number of days ago (what is time anyway), one strain of criticism that emerged, notably from Jonathan Chait, was that the bill made a grave error by funding a number of programs for a few years, rather than making a smaller number of programs permanent. Chait’s point is that any program not made permanent will predictably expire in (I believe the timeline is) six years, because the Republican Party will in all likelihood have taken back partial if not total control of the federal government by that point and will not agree to extend the programs, no matter how popular they have become in the interim.

I tweeted at the time that I was a mild dissent from this position, logical though it seems:

And this is actually a point that illustrates a lot about how I think about our politics these days, so I thought I’d flesh it out in post form.

When I was starting to learn the game Magic: the Gathering a few years back, one of the best things I did was watch a series of instructional videos by Paulo Vito Damo Da Rosa, possibly one of the top five or so best players of all time. Obviously most of the stuff in those videos was specific to the strategy of this one specific card game. But some of it was more general, things like how to think about a strategic context that involves multiple scarce resources that can be traded off for one another (independent of what those resources are exactly in the MTG context).

In this second vein was a discussion of the difference between “playing to win” and “playing not to lose.” Now, as someone who has watched a fair amount of sports over the years, when I saw these items listed in the video descriptions I assumed that “playing to win” was good, and “playing not to lose” was bad: that’s always how they get discussed in the sporting context, where playing not to lose is taken to connote a kind of timidity that impedes good performance. Not so! In fact if anything playing not to lose is if anything the less intuitive and more important-to-learn side.

The basic idea behind these two expressions is that your gameplay should change depending on the strategic posture: are you ahead or are you behind? Gameplay in Magic involves making a lot of decisions under conditions of uncertainty: you don’t know what cards are in your opponent’s hand, and you don’t know what either player is going to draw in future turns. And your attitude toward that uncertainty should not always be the same.

So the classic example of playing not to lose goes something like this: suppose you are so far ahead in the game that you are virtually certain to win on your next turn. You control several creatures, the opponent has none, and their life total is very low, low enough that any one of your creatures can finish them off. You have another creature in your hand that you could play; should you? If we are only thinking tactically and not strategically, you might as well: adding another creature to the board improves your position. But in fact this is the wrong play. The only way you can lose this game is if your opponent has a “board sweeper,” a card that can destroy all of your creatures at once. And if they have a sweeper, then playing out your last creature is a disaster: you would be left with nothing, whereas if you don’t play the creature this turn you can play it after they blow up the board and continue applying pressure. Crucially, in this position you should not bother to ask whether they have the sweeper. Perhaps this is a “limited” format where high-rarity cards are, well, rare, and there is only one sweeper in the entire format and it’s a rare. (If you don’t know what any of that means it’s not important haha, the point is just that it’s very unlikely that they have it.) Doesn’t matter! If they don’t have it you win. So you might as well assume that they have it, no matter how unlikely that is, and make the play that is correct in that scenario. You focus on the worlds in which you could lose, and adjust your tactics accordingly.

Playing to win is just the reverse: if you are behind, particularly if you are very far behind, you should picture what it would look like for you to win the game, no matter how unlikely it is, and play toward that. When your opponent is going to win the game eventually if something doesn’t happen to shake things up (a concept Magic players call “inevitability”) you throw prediction out the window and act like the things you need in order to win will happen. The goal is to make sure that, if those things do happen, you’ve set yourself up to take advantage and pull off the comeback. You still lose most of the time, playing like that. But you make sure that you win the small fraction of the games that you could have won from those sorts of dire positions, because you stay receptive to the kind of good luck that can turn things in your favor.

So what do I mean by “playing to win with our democracy”? I mean we’re losing. It may not look like that exactly, since we won the last election and noted non-fascist Joe Biden is president. But the game has more than one turn, and we are not well positioned over the next handful of turns. Indeed it seems that Republicans have a fairly strong form of inevitability: they have made clear their intention to sabotage our democracy, and after four years with Trump in power it’s pretty clear that American democracy cannot withstand very many more blows. All the Republicans need to “win the game,” in other words, is to get back into power once. Indeed it seems as though even winning one House of Congress is probably enough to ensure their victory, as that perch would allow them to “win” a presidential election by fiat with the help of allied (gerrymandered) state legislatures.

This means that Democrats have to do something to change the game. I’m not convinced that we can achieve a final “victory” in this “game” without some sort of new moment of constitutional foundation. But we can imagine shorter-term “win conditions” that would at least push back the inevitable defeat by a meaningful amount of time: say, passing meaningful federal election reform and admitting several new states (and probably doing something about the Supreme Court that would otherwise invalidate most of the above). And it should be reasonably plain that in order to win this game, we need to never ever let the Republicans take back either House of Congress.

Of course that’s very unlikely! Our whole political culture is thermostatic; it’s been twenty years since a president’s party didn’t get crushed in a midterm. But this is where the strategic point comes in: it doesn’t matter. If this doesn’t happen, we lose the game. Of course this doesn’t mean we should “assume” that we will win every election indefinitely in the sense of not bothering to work toward that very goal. To the contrary, literally every move we make needs to be attuned to the goal of passing major election laws before the Republicans next win an election, and that includes trying as hard as we possibly can (both through campaigning and through governance) to win elections! If as seems likely this Congress has no appetite for democratic reforms, then we need to gain seats in 2022; there are no other choices.

What we should not do is try to optimize for the scenarios in which we lose the midterms, or (even worse) lose the 2024 presidential election. Circling back to Chait, yes, it’s assuredly true that if Republicans control the government when these programs expire, they will not extend them! And that would be bad! I don’t care though! Because if Republicans take control of the federal government ever again, we have much much bigger problems than the expiration of some social programs: in all likelihood that would mean that the Republic has fallen. We will have lost the game! Now that wouldn’t be the end of history, of course, but it would mean that we had to start playing a very different game, the “resist a fascist regime” game, which is a much less pleasant game than the one we’re in now.

Of course this doesn’t mean that the funding structure of BBB is necessarily good. But it does suggest to me that we shouldn’t object to it specifically because it is not robust against the Republicans taking back power. Of course it isn’t! Literally nothing we care about is robust against that! It doesn’t matter, strategically, how likely they are to win power. We should design BBB to optimize its contribution to our political fortunes in 2022, and, insofar as that leaves further choices, we should do good things because they are good to do. That’s it.

Coda: The idea of playing to win has other deeper implications for Democratic (and democratic) strategy; in a way the application to BBB is unusually surface-level (and I just used it as an illustration). The most radical recommendation of playing-to-win strategy is that you should be willing to risk losing “worse”/more quickly. Say the opponent is threatening to defeat you on their next turn, and you have a play you could make that would prevent this. But it would only mean that you lose in two turns, and in the process you’ve spent a resource that you might have needed to actually win the game. You still lose, it just takes a little longer: so what? The suggestion for politics would be something like this: to win we need a Congress that will pass really quite radical elections laws. Therefore, we need to work toward such a Congress, by running candidates who support those kinds of reforms, even if that means taking a higher risk of losing elections! After all, the strategy of “elect Joe Manchin and then hope that Joe Manchin will cooperate with the drastic measures” does not seem to be working. And if we don’t get a Congress that will pass those measures, it doesn’t matter how many elections we do manage to win by some miracle of happenstance: eventually we will lose another election, and at that point we lose the game.

I’m not at all sure that strategic conclusion is sound! For one thing, the game of politics is different in important regards from a game of Magic. In the latter it really doesn’t matter how quickly you lose, whereas in the former it very plausibly does: every year that we can forestall a fascist regime in America is that much harm averted. And there are also plausible accounts of why running less “electable” candidates is not actually the best path toward our win condition: perhaps instead we should simply try to win as many seats as possible, so that we can afford a few defections, even if the individual candidates are still quite moderate.

I’m less interested, for this post at least, in those sorts of specific conclusions and more in the strategic mindset. We are losing. If things keep up as they are, we are going to lose, probably quite soon but inevitably in any case. The singular focus of our politics should be on finding a path to victory, however remote or narrow, and pursuing that path with all our might.