What if they describe an approach to disarming the trap that obviously would work, that would take no specialized knowledge or skills? I.e., something your grandmother could do without trouble.
I think there's a difference here. Mort suggests (and plenty of others believe) a player should have his or her character act a particular way, when that is not backed by the rules of the game we're playing and is easily explained given the mutability of the fiction. Whereas the rules saying the DM should be judging the efficacy of a player's stated approach to the goal is telling the person choosing to be DM about his or her role in the game.
I follow, but actually am now a bit more puzzled (not by you - by the overall logic of the situation) because of the post of Elfcrusher's that I've posted. (The emphasis is original, though I've changed it from italics to underlining so as to maintain it in the quote format.) And maybe "intrigued" would be a better word than "puzzled" - I'm not sure, but will post on.
Judging that an approach
would work very clearly requires a robust sense of a not-too-mutable fiction. But (as you say) the player is permitted to exploit the mutability of fiction to make sense of his/her play of the character.
For this to work requires - I think - very clear boundaries around what is mutable in the hands of the player, and what the GM is permitted to rigdily establish in advance of adjudicating the "woulds" and "coulds".
I think that (eg) [MENTION=22779]Hussar[/MENTION]'s use of fortune very close to the framing, and postponing nearly all of the narration to afterwards, might be one way of trying to manage (by trying to avoid) this need for boundaries.
EDIT: I saw this just after posting:
Who's to say the character can't know these things? You, as the DM? Sure. But maybe your player disagrees. Maybe he/she says, "There was a village elder who was a great adventurer in his youth, and as a child Gord the Barbarian sat at his feet and listened to all his stories."
Now, you, as DM, may want to overrule that and say, "No, it's my game world and that didn't happen." But in that case the problem isn't metagaming (as AngryDM has done a great job of explaining) it's a problem between you and your players.
Presumably if the player disagrees, in the context of disarming a trap, about what
would work because even one's grandmother could do it without trouble, the GM is expected to have the last word.
But in the PC backstory case, and the action declaration case (
My INT 6 barbarian does such-and-such) which the PC backstory is meant to be ancilliary to, the GM having the last word is flagged as a possible source of problems.
This illustrates what I mean by the need for clear boundaries over who has what sort of authority over which bits of the shared fiction. I'm not suggesting it's going to be tricky in every case, but I think maybe it might be tricky in some cases.
Do you trust your DM to decide? Or don't you?
I don't think "trust" is the right notion, because in the context of Gord the Barbarian's backstory and action declaration you don't call on the player to trust the GM.
I think what is at issue here is the distribution of authority over establishing the fiction.