D&D (2024) Auto-succeed/fail on ability checks

[UngainlyTitan]

Sorry, I didn’t intend to sound curt.

If you don’t set DCs higher than 15, then of course this rules change won’t affect you (literally—it won’t affect you, since it only exists for when a nat 20 isn’t usually good enough to succeed, which means the DC must have been above 15 even for someone with a -5 mod).

But pointing out that you use an artificially low, homebrew version of the 5e DC table—when your argument is that I shouldn’t be setting higher DCs even though RAW says DMs should do so at least occasionally—is not nitpicking.

I am trying to understand what you are trying to achieve here and explaining where I am coming from.
If you have information gated behind a DC 30 and you have level 10 PCs then a character with max stat and no proficiency is +5 on the roll. Then no chance of success even with Bless, may be with Bardic Inspiration if the bard is level 10 or better.
If they have proficiency in a relevant skill it is +9, still no chance though bless and Bardic Inspiration does make it possible. Even with expertise it is still pretty low odds even with bless or bardic inspiration.

Changing to autosuccess on a 20 dies not make a noticeable difference to the odds, not one that will be spotted in play. If the barbarian succeeds where the bard failed it can be narrated as "Hey (Bard) do you remember that song about....."

You are as a DM perfectly within your rights not to allow an attempt to succeed by someone with out a relevant proficiency.

If it is relevant to the progression of the plot, what is the point of the high DC and if it has no relevance what does it matter?

 

[OB1]

So my initial reaction to this change was that I was just going to house-rule back to the 2014 rule, but I've come up with what I think is a more elegant solution, through an easy to apply ruling.

If the DC of a task (and possibly Saving Throw) is higher than 20, only characters with training in a relevant skill (or saving throw) have a chance at success.

What I like here is that someone trained in Aracana, for example, could still succeed on a DC30 check at level 1 or fail a DC10 check at level 20, while putting a 'cap' on what someone untrained can do, thus making which skills you take more impactful.

I'm back and forth on the saving throw piece. I really like that different characters have strengths and weaknesses that push them out of the chance to succeed/fail against different things, but I also get that it's fun for players to have at least a chance.

 

[Reynard]

If the DC of a task (and possibly Saving Throw) is higher than 20, only characters with training in a relevant skill (or saving throw) have a chance at success.

Wait, so if it's a DC 21 and I have an 18 in the relevant ability I don't get to roll?

I think this underscores a common misconception in how 5E works: you don't make skill checks in 5E, you make ability checks and in some cases your proficiency bonus applies because you have an appropriate skill.

 

[Ibrandul]

If it is relevant to the progression of the plot, what is the point of the high DC and if it has no relevance what does it matter?

In my games, not every ability check is "relevant to the progression of the plot." I'd even go so far as to say that none of them are. There is no predetermined plot. There's just a world to interact with.

But I've realized that bringing up knowledge checks was a wrong turn, because (1) I'm starting to suspect most 5e DMs seldom use them; and (2) the structural problem with the new rule doesn't only apply to knowledge checks.

Again, it's clear that you run a very different kind of game than I do. You've bumped the DC tables down so that PCs failing on ability checks is much less common, and you don't ever use high DCs. That's OK! The 2014 DMG itself advises "caution" when setting high DCs because success might be impossible for low-level characters even with proficiency and a high stat mod.

But the 21–30 DC range is in the system because it's there to be used. The game is supposed to accommodate high-level play, where the difference between what one PC and another can accomplish is much greater than at low levels; and sometimes the task at hand really is harder than just "hard."

 

[OB1]

Wait, so if it's a DC 21 and I have an 18 in the relevant ability I don't get to roll?

I think this underscores a common misconception in how 5E works: you don't make skill checks in 5E, you make ability checks and in some cases your proficiency bonus applies because you have an appropriate skill.

Yes, that's correct. I'm saying that a DC over 20 implies that it takes more than natural ability to attempt it, that you must have training in a skill that could effect the outcome for it not to be impossible for you. If I believe a task can be done by someone without training, I wouldn't set the DC higher than 20.

Highest DC for a check that anyone could do - 20
Highest DC for a check requiring special training - 30

Like, I consider myself a pretty smart dude, but I have 0 chance to solve a differential equation (because I lack the specific training to do so).

 

[Stalker0]

One of the most iconic moments for a dnd player, is that time you are doing the crazy stunt, you can only succeed on that 20....and you nail it. Its a moment that will live on in players minds years after the plots of the campaigns fade from memory. I know, because I've DMed for over 20 years, and I know what my players still talk about to this day.

Does it create some weirdness? Yes it does. Does it mean that as a DM you might not to just throw out checks that you see absolutely no way for a player to succeed at? Yes it does. Is it 100% totally worth it? Absolutely.


That said, one way to mitigate this a bit is instead of using high DCs for certain things....use disadvantage. This will reduce your odds to 1 in 400 for that nat 20, which might be a lot more palatable.

 

[Reynard]

Yes, that's correct. I'm saying that a DC over 20 implies that it takes more than natural ability to attempt it, that you must have training in a skill that could effect the outcome for it not to be impossible for you. If I believe a task can be done by someone without training, I wouldn't set the DC higher than 20.

Highest DC for a check that anyone could do - 20
Highest DC for a check requiring special training - 30

Like, I consider myself a pretty smart dude, but I have 0 chance to solve a differential equation (because I lack the specific training to do so).

I think trying to make the rules of D&D model "reality" is a losing proposition. On top of that, setting an arbitrary limit like that without examining how it interacts with all the related systems is just bad design.

 

[OB1]

One of the most iconic moments for a dnd player, is that time you are doing the crazy stunt, you can only succeed on that 20....and you nail it. Its a moment that will live on in players minds years after the plots of the campaigns fade from memory. I know, because I've DMed for over 20 years, and I know what my players still talk about to this day.

Does it create some weirdness? Yes it does. Does it mean that as a DM you might not to just throw out checks that you see absolutely no way for a player to succeed at? Yes it does. Is it 100% totally worth it? Absolutely.


That said, one way to mitigate this a bit is instead of using high DCs for certain things....use disadvantage. This will reduce your odds to 1 in 400 for that nat 20, which might be a lot more palatable.

I get where you are going here, but the impossible stunt that only succeeds on a Nat20 still applies with my ruling, it just puts different caps on what a nearly impossible stunt is for different PCs (ie a level 1 Rogue picking a DC30 lock, or an 8Dex Paladin sneaking past a guard with a DC20 Passive Perception).

 

[UngainlyTitan]

In my games, not every ability check is "relevant to the progression of the plot." I'd even go so far as to say that none of them are. There is no predetermined plot. There's just a world to interact with.

Ok, fair enough, but what world building is served by greater than DC 20 checks. I am curious. Could you give me an example from play?

But I've realized that bringing up knowledge checks was a wrong turn, because (1) I'm starting to suspect most 5e DMs seldom use them; and (2) the structural problem with the new rule doesn't only apply to knowledge checks.

I do use knowledge checks but generally use rolls in a non binary fashion. That is the very easy Dc might be the fail threshold, the easy DC the common knowledge, moderate for what the learned and so on.

Again, it's clear that you run a very different kind of game than I do. You've bumped the DC tables down so that PCs failing on ability checks is much less common, and you don't ever use high DCs. That's OK! The 2014 DMG itself advises "caution" when setting high DCs because success might be impossible for low-level characters even with proficiency and a high stat mod.

It is outright impossible for some of the really high numbers for low level characters.

But the 21–30 DC range is in the system because it's there to be used. The game is supposed to accommodate high-level play, where the difference between what one PC and another can accomplish is much greater than at low levels; and sometimes the task at hand really is harder than just "hard."

Out of curiosity, if a character has climbed a wall successfully once, would you ask for an ability check to allow them to climb it a second time?

 

[GMforPowergamers]

If the DC is 30 and the best a character can get to after bonus is 22, why even bother rolling?

becuse I MIGHT remember if a PC has an 8 in a stat, but by level 7 or 8 I most likely wont remember who has a +8, who has a +6 but I most likely will remember who has the +14. so the DC is 27 so player 1 and 3 can roll but player 2 I have to decide "do I let him roll with a 5% chance or not" but I don't remember... so I have to stop action to ask everyone for there skill mod...

I do like the auto-fail on a 1, though.

I dislike auto fails... if I say "Roll a dex/tinker tool check" and he says "It was a 1, but with bardic inspiration of 6 and my +9 mod that is a DC 16 or lower I beat" I like that even on that 1 he hit 16, and if the DC was 12 or 13 he made it... on a 1