Thanks for being a good sport!

Lindybeige once put dice rolls into a different perspective. Rather than the dice describing how well the action was performed, his suggesting was that the dice would describe the environment. In this case, that would describe how complicated the code is. One of his examples were for athletics, where he thought of it as describing how tall a wall is. Your athletics was 14, this wall turned out to be 15, sorry, you just barely didn't make it over.

The bottom bit looks like Loss; ~~:.|:;~~

"Umm...'mellon'?"

I think you answered your rethorical question yourself: If it is not in the official books, it is not an official rule. And I would not say that they leave it vague. To quote the PHB: "To make an ability check, roll a d20 and add the relevant ability modifier. As with other d20 rolls, apply bonuses and penalties, and compare the total to the De. If the total equals or exceeds the DC, the ability check is a success [...]. Otherwise, it's a failure, which means the character or monster makes no progress toward the objective[...]." That does not leave much room for interpretation. It plainly say that if the exceed, then they succeed and if they don't, than they fail. Yes they don't make an explicit remark about critical results, but they don't need to, because such a rule was never meant to exist in 5e aside attack rolls and death saves. Not to say that you can't make it a rule at your table (same as with everything else), but there is still not much room for missunderstanding the official print.

Also a Nat 20 / Critical Skill check doesn't guarantee success, just the best possible result.

Also a reminder that Pathfinder 2E has a significantly better system for criticals, in a way that makes sense for ability checks. It has degrees of success and failure, and a crit only moves it one degree higher or lower, so a crit can potentially still be a failure if your really bad at something or it's very hard. Really, P2e is better at almost everything, especially making it so you don't need to remember tons of exceptions like D&D5e. You also aren't supporting Hasbro, which is always a good thing.

To expand on the other comment, to include in the rules everything that *doesn't* happen would be insane. If it isn't in the rules it isn't in the rules. You don't have to list every possible thing that a player may say applies for it to not be included. If a player falls out of their chair, does that change the result? It isn't included in either of these rulesets...

Depending on what skill they are rolling, a nat 20 doesn't necessarily mean they just instantly have all knowledge of the thing they are deciphering. Let's say it's an ancient form of Elvish, and the character speaks Elvish. They'd know *modern* Elvish, and might be able to use that to discern enough to get the general gist of the writing, but not a perfect translation. Which could come into play later on to hilarious effect. If they used uh... I forgot the skill name but the generic "adventurer knowledge" one, they might not know what it *says* at all, but they may be able to know what language it is, who wrote it, and what they might expect based on knowing about the script in other dungeons. If it's an arcana check they could understand it is magical, and what it does; maybe activate/deactivate it but not how to recreate it or *translate* it, per se.

i think the real answer is using whatever makes sense in context: if your character has some experience with the language they could have a brainwave where they see a connection with their existing knowledge, whereas if your character has no way of actually figuring it out they might for example look at the number of characters and blurt out some sounds that fit and that turns out to be correct (or just close enough).

Personally I'd rather have a character who has *approximate* knowledge of all things. Like, correct enough that progress can occur, but with enough wrong that they're hopelessly misguided or just generally not getting things. Bonus points if they're really arrogant about their intelligence.

The trouble with doing that is that you end up in the stupid situation described by this comic!

one of my players once created character so dumb, he can barely speak. at one ocassion during an fight he asked to roll for intelligence (he traded most of it for strenght) and got nat20. Nobody could stand the dumb mountain of flesh, suddenly telling everyone strategy tips.

A nat20 isn't an automatic success on skill checks, but there shouldn't even be a roll for something that is outright impossible. Unless there's some plot reason why this otherwise mentally challenged character would suddenly become fluent and a top notch strategist out of the blue?

So just playing as the average Lemmy user? Seems boring and like you would still get in lots of fights.

My observation is that when push comes to shove, the one good at a thing absolutely cocks it, only for the worst at the thing to catch it in the best way possible. DM: Decipher this thing. Roll for int. WIZARD: 3.... that makes 7... BARBARIAN: 19, that makes 17... WIZARD: [lenny face] DM: [lenny face] Perhaps a good DM or table could create good narrative spins for this....

One thing we often do is we gate the ability to roll on a check through whether or not the skill is trained - for example in our games with lockpicking, you can only attempt to pick a lock if you have proficiency. This prevents the situation where the character with years of training and practice in a specific niche skill beefs it, but someone with no idea what they're doing then tries and succeeds - we say some things are only possible to try if you know how. --- Don't apply this house rule to everything, but it's worth considering, especially in games where your party can casually drop a +10 or a +15 onto any skill check through the right magic to force a success anyway.

if you're going to let dice determine probability, then use the right probability dice. d10's can be chained to make a number between _ and infinity if you have enough dice. need a number between 1 and 1000? use 3d10. first die is 1000's place, second die is 100's place, third die is 1's place.

Or just luck. Like with lockpicking, it's possible to bang a keyed lock and pull on it and get lucky with the pins all lining up and it opens. There's vibration devices that work by doing this constantly until it works. The odds are much lower than 5% (without a vibrator), but dnd is supposed to be fun and exaggerated odds can be fun. I've played a homebrew game that was based on the devil may cry world, where the cut scenes were constantly full of crazy shit, so the GM decided that the crazier the idea you came up with, the more likely it was to succeed, and it was *fun*. It did really help that he was very good at thinking on his feet and let the game flow more naturally instead of getting stuck in situations where a player succeeding at some random thing they want to do breaks the whole campaign. For the scenario shown in the OP, a character could get lucky and guess what runes mean. The context could give clues, or maybe one rune looks kinda like something else, is a red herring on its own, but just happens to lead to the correct conclusion in that particular case. Though it would be fair for the next (unsuccessful) roll to give actual useless red herrings. It's probably better for the GM to do the rolls for the player so they don't know if it's a nat 20. I like this for any kind of information gathering rolls, like spot checks, because it allows players to second guess roll results without it being meta gaming. Also pre-rolling some of those can help, because "roll a spot check" tells the players that there's something to spot, even if they all fail. And not asking can imply there's nothing there.