Thursday, January 17, 2013

Certainty and Uncertainty in RPGs

What part does uncertainty play in an RPG?

We’ll start with defining what a game is. My definition of a game is “certainty intersected with uncertainty, and by navigating that, achieve a goal.” Whether this is the best definition of a game can be debated by the peanut gallery ad nausea. However, it is useful enough for me, and I’m the one writing this article.

What can we extrapolate from this definition?


  • Certainty is needed for structure, or at least a definition of “win”.
  • Certainty should be emphasized at many disconnected points.
  • Uncertainty should be emphasized at those disconnected points.

Using Monopoly as an example, we have certainties and uncertainties. The rules, the property costs, the board, and even the players involved are all certainties. The dice rolls and chance cards introduce uncertainty, but there’s more uncertainties than that. We also have uncertainties introduced when other players take their turns, and uncertainties introduced when players trade properties. You can not predict the outcome of these actions.

In modern D&D, your character is certain as the player fully controls the character creation choices, but the future of that character is uncertain. In no possible way can you guess the character’s future adventures. Meanwhile, each player runs the same character almost every game, but you can not predict the actions that the player will decide. You can predict that battles will happen, but you can not predict when, where, and against what.

Does uncertainty have a quality? I think that it does. So what makes GOOD uncertainty and what makes BAD uncertainty? For that matter, what makes GOOD certainty and BAD certainty? I have to believe that if you pick correctly, you wind up with something compelling, but if you pick wrong, you wind up with something capricious or mechanical.

I liken certainty to chains and uncertainty to broadness.

Certainty is a chain. The longer that chain, the more powerful that chain. Once a player has a chain of certainty, he will use that certainty over and over to gain victory. The more effective that certainty, the more often that certainty will be used. Applied to often, and that leads to an increasingly rigid game where the most optimal paths lead to victory and all other paths, regardless of their other merits, get thrown aside.

On the other hand, make the chain of decisions too short and you never create meaningful decisions. Hair color, eye color, and favorite band add flavor to a character, but as these choices do not actually affect other mechanics, they wind up being meaningless. Only when decisions actually affect some other aspect of the game is the chain long enough.

So chains of certainty must be long enough to be meaningful and interact with other components of the game, but not so long as to be de regeur.

I measure uncertainty in broadness. The narrower the uncertainty, the easier it is to predict what will happen, making many game mechanics irrelevant. The broader the uncertainty, the less that the player understands the outcome of his actions, until finally, the player has no idea how to proceed as they can no longer imagine outcomes.

Uncertainty is best when the player understands the certainties on either side of the uncertainty, tempting the player an outcome that is worth risk.