You seem to be making the mistake of thinking that since my position is not in conflict with yours, it isn’t useful.
Not at all.
But in my experience, there are many DMs who do not consistently insure that a meaningful outcome directly results from a player rolling under the target number. If that advice isn’t useful to you, great, but that doesn’t mean it’s without value to others.
And I'm saying that they don't actually need the advice people generally give them, they need
very different advice, and that "don't roll unless there's a meaningful consequence for failure", which is generally the wording of the advice generally given, is
very bad advice. It also happens, apperently, to be nearly the opposite of the advice you, at least, are meaning to give. Why talk about consequence for failure if that isn't actually what you mean?
Yes, I agree. What I’m saying is that it’s not worth rolling a check if there are not multiple possible success states (and/or failure states) depending on the number rolled.
That isn't consequence for failure, though. This whole rabbit hole we just went down would not have happened without "don't ask for a check unless there are (meaningful is often put here, but not always) consequences for failure".
I genuinely don't think anyone is intentionally asking for rolls when the roll cannot lead to multiple outcomes.
Better advice would be, "Make sure that the results of checks you ask for are something you've thought of already, or that you're thinking about them when you ask for the check, and make sure that there are at least two distinct outcomes. They can both be positive, both be negative, or be a mix, or even several along a spectrum, but they should be distinct."
I've always liked rolling. If I feel that the difficulty is low, then you'd succeed no matter how low you roll, but you might succeed with a bonus if you roll high. Or succeed, but have an interesting negative side-effect occur as well if you roll really low. I like the dice to tell us all, even ME (as GM) what's happening in the world.
Agreed. The check tells us how well the character performed, not what happened around them while they did the thing.
It's totally reasonable to say that a master of stealth simply will not fail simple stealth tasks. But it is also completely reasonable to introduce randomly selected variance to how "on" the master is while performing a task, and narrate the scene differently based on the result,
even if the difference doesn't "matter".
IME, "you get to the top of the cliff, more winded than you feel like you should be, more than you know you would normally be. Why was that harder than it should have been?" leads to further characterization from the player. It doesn't matter that no mechanical difference occurs, it doesn't matter what the stakes are, players who are engaged in their character's mindset and care about the character and the fiction will care.
Punishing players for wanting to make a check, is, IMO, bad DMing. If there is no chance of failure if they describe a certain approach, great. If they want to roll anyway to see how well they do the thing, that is not a good reason to invent negative consequences for a low roll. The nature of the task shouldn't change because a player likes to use the d20 as part of the fiction-building that is play.
