A few thoughts intersected the other day. Let’s see if I can get them down.

Here’s the short of it: balance is a canard. You don’t want balance, at least not in an RPG. You want interesting choices. Balance is a response to one kind of failure of interesting choices, that of a dominant strategy. But you can have a lack of interesting choices in the other direction: if every choice is equivalent, that is, if your choices are false choices, then your game (qua game) suffers just as much as if it had a dominant strategy.

So, you need to make things asymmetrical, maybe even unbalanced (gasp, gape). It’s a balance to strike, between the fire of a dominant strategy and the ice of false choices. You need to make the experience of play involve moments of “if I had only done X”, so that the rest of the choices you make are meaningful.

Now, there’s a strong version and a weak version of this claim. The weak is that, for all choices, you should have at least two meaningfully different options, where it is not transparently clear which will be the better one. The strong version is that you should also have some options that are clearly worse. I only mean to maintain the weak version, but with the caveat that what is clearly worse to you may well be compelling to someone else.

There’s another wrinkle to this, which Austin articulated very nicely the other day: when he comes to the table, he wants to see the choices other people make, and he wants those choices to be different from his own. So you need choices that ramify into other meaningful choices.

OK, theory, fine. How do you do this? Obviously, that’s context-dependent. But here are my principles on the matter:

• Make your game have multiple axes of interaction, which are not directly comparable. You can think of this as “choose the problems you want to expect”. In Apocalypse World, for example, you can focus on personal effectiveness or group effectiveness: are you a badass, or are you a badass because of and through your gang? Obviously, the latter comes with some problems—things that aren’t in your control like challenges for leadership, supply issues, communication issues, etc. But at the same time, there are things you can do with and as a gang that you simply can’t as an individual.
• Play enough to see the ruts you make when you play. If, after choosing M, you consistently go from there to Q, F, R, then consider collapsing those options. Especially if there’s no other way onto that path, and everyone else you see follows it the same way.
• Make a variety of things in your game. Like, mechanically meaningful widgets. This can be variety within a category of thing, or a variety of categories of things. This is a contentious point; a lot of games have been made, many very good, without many types of things. I acknowledge that this is a taste thing for me. I like games where you get to choose from lists of Distinctions/Moves/Gifts/whatever, and make up your own, and engage with the game outside the act of play-as-play. So, if that’s not your aim or style, at least help the players to make interesting situations where the choices they have to make in the story are valid.