Rolling dice, playing cards, betting tokens, comparing scores, dancing, RPS, whatever. Resolution in general.
Open a random rpg. There will be a section of some length on how the dice are rolled, what the result means, how character traits affect it, and so on. All absolutely essential and useful material. Sometimes there is elegence, rarely true innovation, sometimes cumbersome chart-look-ups (but I don’t like charts, so maybe there are elegant uses of them somewhere).
At least for me, when the dice are rolled is far more significant when actual gameplay is considered. If there is little to nothing in the book, I default to style mostly stolen from Burning Wheel. What follows is a listing of some”when”s of rolling the dice.
- When the situation is dramatic or meaningful. In my mind, this is a fundamental requirement. No boring rolls, please. Boring scenes are generally not very good idea, either, but rolling dice there is adding insult to injury.
- When there are consequences for failing. There is little point in rolling if failure means simple retries ad infinitum. This should be pretty obvious, too.
- When two players want a different thing to happen in play. If GM want’s Bob character to escape and all players want it too, it can be argued that there is no point in rolling, and Bob’s character simply escapes (or maybe a style roll determines how impressively Bob or the GM will describe the run-away, but that is not relevant). This is far from obvious principle. I don’t use this one, for example. In my opinion, the next is better for the flow of game because it introduces more trouble.
- When there is more than one possible outcome and all of them move the story forward. That is, if someone (usually the GM) can come up with an interesting complication in case of failure, the dice are rolled. There is no point in rolling to discover the secret door which is integral to the story, if failure means not finding it. But if failure means that before finding the door, the party is tracked down by a ferocious minotaur (because finding the door took time), suddenly the roll has no chance of screwing the game and also rewards players for being good at finding secret doors.
- When two fictional characters are in conflict. Character may be interpreted liberally (allowing the rockslide to be a character trying to bury the hapless mountaineer) or literally. I think this is a pretty good rule of thumb, because there often are interesting consequences when two characters conflict.
- When GM (or nobody) knows or has a stake in what will happen next. Basically, as a means to avoid unfun decisions or to move blame. I think this is a useful tool in moderation (and great way to use random encounters), but widespread use is a sign of trouble. I might be wrong. Rationale: The game may be too slow (“what do you do next?”), GM uninspired (watch for burnout), GM unprepared (learn to improvise or take a break), GM not listening to players (they are bound to have some ideas for what their characters will do next), or the game simply boring (take a longer break, start a new game, get someone else to GM).
I prefer number four. Other people have other preferences. Do know that this is a fairly imporant choice and think for yourself.