Quote Originally Posted by NichG View Post
This can be part of it, but I think there are two other things as well.

One has to do with the ability to plan. The more front-loaded chance is, the more possible it is to have more involved plans. So randomness, especially of the success/failure sort rather than the 'reacting to changing situations' sort, has a consequence of shortening the planning horizon and favoring brute force or direct approaches to situations. Which in turn has knock-on effects to the importance of e.g. the character-building minigame versus the importance of decisions made during on-screen play.

The other has to do with momentum and maintaining the importance of choices and actions. The success/fail kind of resolution has a format of 'I do this', 'no you don't', which just sort of makes things get stuck or makes decisions or ideas feel wasted. Something like a dice system that says 'you succeed, but it costs you X' or 'you succeed, but in the time that takes your antagonist gets X benefit' or whatever would resolve this. But 'check if you can do a thing' seems inferior to me to 'do a thing (and)' now. This kind of goes hand in hand with a distaste for checks which the GM calls for, which strike me as a kind of 'go fish' game - in the character building minigame you're supposed to guess which checks you're likely to be asked to make, and then the GM calls for checks which basically see whether you guessed correctly. So the model I prefer now is 'the player knows well what they can and can't do, and decides from that what to do, then the GM (and GM tools like dice if you want) decide how the world responds to the fact that that happened'. But less focus on calling into question whether in fact the character can do what they think they can.
Understandable.