I was once GMing for that same bestie in a 3d6-based system. I told her to roll, then realised her stats weren't high enough for her to succeed, so told her not to. She gave me a look, picked up the dice, and rolled a crit. Out of SPITE. And this is 3d6, so it's a 1 in 216 chance. She didn't need to manipulate me. Either I went along with it, or my dice would be forever cursed.

Generally speaking, to shut the player up about it

Sounds like a possibility for a really creative story moment. Maybe the comic book that character always carries around with them just so happened to use the same runes as their "secret language" and the author of that comic is some super nerd for that specific language.

Oh hey, its the funny elf face comics but from another artist!

I love it. I'm not even mad.....well....maybe a little. Lol

You could technically have negative modifiers that would make it impossible for you to succeed, where others might. You're right in that your DM likely will not even let you roll....but it's still possible.

If someone wants to jump into a cavern and use strength to flap his arms to fly, rolling a d20 can be to see how much the person fucked up. A 20 isn't an automatic success. Same when someone mixes a potion, the d20 may be to see how much it will poison the creator if they drink it. Roll to see how badly you fail.

I house rule it to anything where dumb luck might help anyway. deciphering a language you know nothing about? nah. lockpicking a simple lock despite not really having much of a skill? woah, you don't know wtf you did but things clicked. you could probably force it open with a high enough strength check too but hey.

Sadly, the elf only knows Common and Elven.

That sounds like fun!

Generally speaking it's considered bad practice for a GM to call for rolls that literally no one in the party can succeed at, but as with anything in tabletop roleplaying there is nuance. There could be a narrative reason for the player to not know just how difficult something is and you don't want to give it away by just telling the players they can't succeed. If the most capable member of the party rolls a 20 and fails then the "reward" is the narrative of the attempt and learning what you're up against. Or maybe someone in the party *could* succeed but for whatever reason the child-prodigy wizard with a strength of 8 wants to try lifting the portcullis. It wouldn't make any sense for them to actually do it.

This reminds me of it's always sunny when Charlie, who is illiterate, realises he can read Irish gaelic because his childhood penpal, who turns out to be his father, wrote to him in Irish.

I can't believe nobody has mentioned it yet. The runes clearly say "Is this Loss?" It's written plain as day

"Ok your dumbass character gets the genius idea to ask someone who knows more about runes to read them."

Bad faith and pedantry aren't the same. The comic very clearly implies that the nat 20 caused their dumbass character to be able to decipher the runes. If it didn't, the player wouldn't have announced "Nat 20", but the actual score, wirth modifiers taken into account.

it was Frodo who figured it out

Nat 20 is very, very commonly used by GMs to mean "critcal success" in or out of combat, no matter the explict rule. Same goes for nat 1 being a "critical failure." Why? Because it makes the game better for everyone to have these rare rolls rewarded or hilariously punished.

In BG3 (which mostly follows d&d 5e rules) you succeed in a skill check with DC 30 on a natural 20 even if you have less than +10 as a modifier on the roll Are you sure a 20 has no special meaning in checks in d&d (I presume you mean in d&d as it's the most popular system)

Yes he is and no it does not. That is a common house rule Larian implemented into BG3, but it is not part of the original rules of DnD 5e.

I'm glad you know them so well. In what way is such a popularly used rule not a rule?