But how do you determine whether the outcome seems uncertain, if you don't even know which underlying mechanics apply? Do you just blindly guess? Do you use out-of-game knowledge?Totally disagree. I consider whether the action is possible in the fiction first, then, if it seems uncertain, then determine the mechanic.
If possessing 30hp is not criteria which proves a character is immune to being dropped from an attack for (1d8+5) damage, then what is the criteria? More importantly, how are the players supposed to know what criteria you're using, if you aren't using the criteria in the book?
Trivial examples resolve trivially at the table. I would adjudicate walking across an empty room is a Dexterity (Acrobatics) check at DC -2, so the outcome is certain unless your Dexterity modifier is worse than -3; and the only reason to roll, outside of combat, is to figure out whether you briefly stumble and look clumsy. In the vast majority of cases, it doesn't take me any longer to resolve with my method than it does with yours. The difference is that I'm actually using the rules in the book, instead of avoiding them.Take walking across a room with no hazards. According to you, I have to determine the controlling mechanic, determine the difficulty, and then determine success/uncertainty status. My way, I just say yes. But, this is easy, so let's go harder.
My way is the same as yours. I look at the approach, and make a judgment call as to the DC. If the DC is 15, then anyone with a bonus between -5 and +13 has to make a Strength (Athletics) check, with failure indicating a fall. If your bonus is worse than -5, or greater than +13, then the outcome is definitionally certain and no roll is necessary.Now the PC wants to pole vault across a dangerous stream (maybe acid, or a super-saturated mineral steam that causes immediate crystal growth, doesn't really matter) that they can't jump across. I consider if the pole is long enough then it's plausible. Seems risky, though, so both uncertain and has a consequence. Mechanics time! STR controls here, but it's not too hard, so DC 15 STR check is called for.
Your way you look at STR first, determine DC (not sure how, chart?), then look at PC stats and determine if a check is called for, for each PC? I have the one for any PC that uses that approach because it's based on the approach and not the character sheet.
If you're bad at math, or at tracking numbers, then feel free to put the "burden" back onto the players - tell them that the DC is 15, and let them figure out whether or not they need to roll.So, no, the way you see the uncertainty determination part of the loop.dies not have to work the one way. I'd arguably say that you're doing way more work than me for no appreciable gain -- you onload PC skill calculations to the DM overhead. I don't see the advantage.