Systems and bloggers encourage GMs and players to beware frivolous die rolls. We are warned to avoid die rolls when the outcome would be obvious, or when one outcome would be uninteresting.
Okay, so when are die rolls actually interesting?
Simple failure is usually uninteresting (in RPGs, at least). Players want to jump across a chasm…and fail. So, what, they fall in and die? Not very heroic, and usually not very interesting.
Before we go any further: Yes, a great GM and great players can make that interesting. But it’s a high-level skill, not common in groups.
What if, instead of success/failure, you rolled to determine whether the PC faces a negative consequence?
In other words, if a PC wants to bash open a broken door, they don’t roll to see if they do it; they roll to see if that attracts the attention of nearby monsters. If you fail the roll to bribe the guard, he doesn’t just politely turn you down; the bribe is a crime and he tries to arrest you.
I’m sure there are systems that do this (I’m writing one). But imagine it accepted broadly. What if popular systems were built around the idea of dice rolls providing more than a binary answer of success or failure for the action itself?
Maybe give it a try!