“Uncertainty in Games” by Greg Costikyan of Playdom
September 28th, 2013

This was a fantastic lecture to watch – the many sources of uncertainty in games isn’t something that I had really thought about in the past. To sum up the video:

“If you’re not uncertain of what’s going to happen in a game, why even bother playing?”

Every day of our lives we’re trying to reduce uncertainty – trying to keep a schedule or keep things organised, staying on top of bills, just generally trying to reduce the stress of uncertainty. Games however require some level of uncertainty to be fun, and perhaps that change in our lives is something that makes them so appealing.

Look at Tic Tac Toe – we’re all tired of it because we know exactly how to win. If both players know the old trick, then it’s a draw. Most adults are certain that a game of nought’s and crosses will result in a tie, and so it is not fun. The game relies on the naïve to be fun, and when they learn to win they’ll get bored of it too.

Costikyan suggests that there are 7 types of uncertainty used in games (I’m sure that I missed one or two though – I’ll have to rewatch it):

Performative Uncertainty
Uncertainty is based on the player’s skill. The original Super Mario Bros. is a good example of this – the entire game is made to be predictable, but uncertainty comes from your own ability to get around the difficult terrain and enemies.

Solver Uncertainty
This is largely evident in puzzle games – the uncertainty of whether or not you’re capable of solving a particular puzzle.

Player Uncertainty
Greg described Rock, Paper, Scissors as the platonic ideal of player uncertainty, as the whole game relies on you not knowing what your opponent will do.

Schedule Uncertainty
This type of uncertainty exists in games like Farmville – it relies on your time commitment (e.g. harvesting crops at a particular time), and your schedule can have an impact on how well you do in the game. Although this kind of game is brilliant as a business model – mechanics are made so that players will nag their friends into playing, and the main mechanic is to get you obsessed – it is generally considered unethical game design.

Narrative Uncertainty
Not knowing where the story will take you.

Semiotic Uncertainty
This is an interesting one. Greg talks about a Japanese game (I couldn’t quite hear the name), where at the end of a level there’s a flag pole. If you’ve ever played Super Mario, you’ll probably jump onto it to get points, but instead it kills you – it relies on your understanding of games to throw you off. Another interesting example was a game which appeared to be another boring train simulator, where you eventually come to realise that you’re delivering victims to death camps.

Feature Uncertainty
This is where the developer promises updates in future, and so players continue to play in anticipation of new features (look at Minecraft, Don’t Starve, Terraria, and just about any game following the Minecraft pre-release model).

Whilst watching his lecture, his mention of Roger Caillois’ book Man, Play and Games sent me off on a bit of a research tangent – I’d like to read it soon. What I could find was very relevant to the talk however – he talks about four play forms within a game –

Agon (competitive play, such as in chess)
Alea (chance-based play, such as slot machines)
Mimicry or Mimesis (role playing)
Ilinx  or Vertigo (altering perception)

He suggests that all games involve at least one of these – this was written in 1961 and despite the monumental development in games since then, it still appears to be true.

It has me very interested in this area.