You'd said something like that on a failure where there was no time pressure they'd just succeed; to which my point in reply was that succeeding isn't a failure.
Ok, I see. What I’m saying is, failure when there is no pressure, such as time or other risk involved in the attempt, does not preclude the possibility of eventual success. Unless there is something stopping you from trying again, you will eventually be successful as long as you stick with it. I just narrate over that process and skip to the part where you succeed.
There's a big disconnect somewhere here. I think it's because you're stuck on allowing unlimited retries, which to me are an absolute non-starter.
I'm not deciding when a character (or player) gives up, I'm merely narrating the results of their best attempt (as set by the roll) to do whatever they're doing. Up to the player/character whether she gives up or not, but simply trying again will accomplish nothing as your roll has already determined the best you'll do without some significant change.
Ok, let me try to walk you through my thought process cause I agree we seem to be talking past each other somewhere.
I allow players to re-try failed actions. I do this for a variety of reasons. One of these reasons, as we discussed earlier, is that assuming a single roll represents the culmination of a character’s best efforts, possibly over many attempts, to be too abstract and disconnected from the fiction for my taste. I prefer each roll to represent a single action.
Now, most of the time, retries are largely self-limiting. Picking the lock on a door takes time, and every attempt you make to unlock it, the risk of encountering wandering monsters increases, or the evil ritual creeps a little closer to completion. Bashing a door down is noisy, and every time you attempt it, you risk attracting the attention of wandering monsters, or alerting the troll in the room you’re trying to break into of your presence. Casting a spell like knock to open the door costs a limited resource, usually spell slots and/or or expensive material components. This is something I really like about allowing retries - each attempt is a calculated risk/reward, which is a good thing because decision points are what make the game fun for me.
of course, there are some times when attempting a task doesn’t carry any risk. Maybe there are no wandering monsters and there’s no ticking clock. No pressure preventing you from taking all the time you need to pick that lock, making all the noise you need breaking that door down, or taking a long rest after casting a single knock spell. These situations don’t come up terribly often in my games, but they do come up, and in such situations, retries aren’t interesting. If there’s no risk or cost, then making the check isn’t decision point, it’s just “press A to continue” only you also have a certain percent chance that pressing A to continue won’t work, so you have to press it again.
Now, saying “you can only try once” is one way to resolve the situation in the above paragraph, but it’s not a satisfying option for me. It doesn’t make sense within the fiction for a roll of less than 20 to represent the character’s best attempt. If you roll a 7, that attempt was objectively only 35% of your maximum potential on that task. Now, you can justify that in the fiction by assuming there are other factors preventing the character from producing their best results, but this is again too abstract for me. I want to be able to say what factors are causing that issue. You could say that, for whatever reason, the task confounded the character and they gave up on the task, even though it was theoretically within their power to do. But I don’t like to tell players what their characters think. I leave it up to the players at what point their characters get too frustrated to continue, which in my experience means characters basically never get frustrated and give up. Understandably because we’re here to have fun and that’s not a fun outcome. I could say that this one roll represents a whole lot of attempts, and at some unspecified point, they get frustrated and give up, but that doesn’t really resolve the issue of me telling my players what their characters think, it just hides it behind an abstract number of attempts, and it runs into the afformentioned problem of abstracting multiple attempts into a single roll.
Another way to handle it might be to model the mounting frustration with a cumulative penalty for each successive roll. Each try gets a bit harder until eventually you can’t hit the DC because the penalty is too much to overcome even with a 20. That would circumvent my problem with the one-roll approaches abstractly representing a whole series of attempts with a single roll, but still it doesn’t actually fix the problem of telling the players their characters are getting frustrated; it just changes the mechanical representation of that DM-imposed frustration. On top of that, it doesn’t resolve the problem of retries being uninteresting without a risk or cost. There’s still nothing actually at stake and no decision point. You just try until you succeed or the penalty becomes too great.
So, in a no-stakes situation like the locked door with no monsters on the other side, in a place with no wandering monsters, and no ticking clock, I just let the action succeed, under the assumption that if there’s nothing in-universe preventing them from continuing to try until they get it, they do just that. Frankly, I’m not sure why I would even have put that locked door there in the first place if there wasn’t going to be any risk or cost to trying to get it open.
Which again gives them more information than they should be getting. Why not go through the motions of rolling, to keep them guessing?
Because I don’t want my games to be “going through the motions” and I don’t value “keeping them guessing” over keeping their decisions meaningful.
So in effect you don't allow your players to invent and then pursue their own red herrings?
Not so. If the players have it in their heads that there’s a secret door in this room, they can and probably will try all sorts of ways to look for it. As 5ekyu and I went on about for several pages earlier in the thread, knowing that there is no chance of finding a seam in the wall by dragging your blade across the mortar doesn’t really tell you there’s no secret door. All it tells you is that if there is a secret door, it either has no seam, or the seam cannot be found by dragging your knife across the mortar. Players who are convinced there’s a secret door to be found can still spend time pursuing that red herring. The difference is, that pursuit looks like describing actions and having the world respond to those actions in a believable way, instead of going down the Skill list trying to figure out what check that they haven’t already blown their one try with will give them the best chance of finding it. As well, when they walk away without finding a secret door, it’s because they are out of ideas for ways to look, not because they ran out of tries the DM allows them to make.
Sorry, I just can't get behind that.
That’s fine. I’m not here to convince you to switch to my way of doing things, just to discuss why I prefer it.