Basically, if I pick scissors, and you *happen* to pick stone, you win, but because you could have equally lost, or drawn, you didn’t make any real decision (past the metaphorical flavour of the choice). Your win was effectively down to chance. Any choice I could have made was as good as another (though my past history of choices may affect yours… people don’t intuitively believe randomness. They tend to want to turn things into a pattern, and can veer toward predictability, too. For the same reason, a million monkeys on a type-writer would not produce the works of shakespere because they’re not likely to work purely randomly).

When the choices are uneven (positive in some ways, drawbacks in others – i.e. the low risk, low reward jabs in streetfighter vs. the high risk, high reward special moves), your decision making process is far more nuanced, and the outcome far more interesting (whether or not you win the attack). Because there’s more weight to the choices of either party, the “yomi” process can begin – you considering what I will consider etc.

You probably already know of him but David Sirlin goes into this stuff in more detail and is well worth a google.

One nitpick: he kept using Rock-Paper-Scissors to demonstrate errors in game design, despite asserting that Rock-Paper-Scissors was horribly designed to begin with. If Rock-Paper-Scissors isn’t good game design, then showing how it becomes broken in certain cases doesn’t really prove anything.

Also, the “every role has to be filled” statement just seems like nonsense to me – one can always invent new roles, so how would you know when you’ve filled them all? I think the fact that all FPS games have the same set of weapons has much more to do with genre expectations than with some fundamental law of game design.

Still, very cool talk overall!