As a DM i describe the scene to the players as best i can. What they choose to do with that information is up to them. If that leads to character deaths then that was the choice they made.
You sleep in the bed you make.
The players are about to do something awfully stupid. I take a moment to consider if this is because of a failure to communicate on my part. (Or, more properly, I've been keeping tabs on their discussions, and corrected any failures as they've come up.)
If the players are in possession of all the information they should have, and they decide to go ahead, then that's their call. If a TPK results, so be it.
It is not my job as DM to "subtly inform them that their plans are taken straight from the litter-box", nor to find "an in-game solution that could save them". They get the information, and they get to decide how to act on the information. And that includes being allowed to make stupid mistakes.
"Often incorrect?"GMs often feel they've given all the required information in a way that is obvious. GMs are often incorrect in that assessment.
"Often incorrect?"
Man, I'm so tired of "blame the referee" .
The bed is made by both the GM and the players, but only the players sleep in it, so so speak.
GMs often feel they've given all the required information in a way that is obvious. GMs are often incorrect in that assessment. Having some way to correct for this before it impacts the players is not unreasonable.
D&D gives you a decent way to deal with this: Intelligence, Wisdom, and relevant skill checks. If the plan is smart, let it lie. If the plan is bad, and the PCs should have all the information, but the players have missed something, run it through the dice. If they dice say no, you let it lie.
Make it clear to the players that this is what you are doing. So, at worst, they learn that their character stats are relevant, and *may* save them from utter stupidity.
Even on a GM-centric place like EN World, most times I see advice like the Three Clue Rule geting mentioned, someone pipes up with, "Gee, that's a good idea! I'll use that!" So, yes, I think "often" is a reasonable characterization.
And I'm really tired of advice on how to fix problems being read as a blame game, and I'm tired of GMs who get defensive any time someone suggests they could do better. But you don't see me invoking the language filter over it.
This isn't about "blame". This is about identifying issues and getting them resolved before they heavily impact results in game.