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

[Ibrandul]

This is where our basic philosophy clashes. For lore I would call for an int check. It is ability scores that count in 5e. If they bring up arcana, history or what ever I might agree that arcana counts or history and ask if they are proficient. However, I would never gate information behind a DC 20 + or 30 check.
If they really cannot succeed then I would tell them you know nothing about it. if the Cliffs off Insanity really are impossible then I would tell them so. They would need specialist help or equipment.
I would not use the same dc either. The trained people might get a lower DC representing their greater chances of knowing something.

So to take your example: if the test is Arcana and the character is proficient and they make the Easy DC I would say that you don't know but you thing you may have seen some of those symbols relating to an ancient magical civilisation.
If they made the moderate DC then I would improve that and say that the symbols are Netherese. If they make the hard DC I would add more and more again if they make the impossible DC.
If they were not proficient I might make the information for each DC about a step or 2 steps harder.
If they have Legend Lore or some such I would give at least the Hard DC information and add where they might find out more. Or ask for a check and add in the impossible stuff as well.
I will not go around the table and let every one try. The may be able to help and give advantage or bardic inspiration or what ever,
Under the new mechanics I would reveal information on an autosuccess may be up to the Hard DC.

Thanks for explaining. It’s great to get an insight into other people’s styles. You're right—our basic philosophies are on different planets! I've never even considered the idea of setting DCs based on which character is attempting the task. I wonder how many DMs operate that way?

It seems to me that at that point you may as well dispense with stats entirely. And indeed, there are indie RPGs that do so—the players and the DM come to a mutual, qualitative understanding of the PCs’ strengths and weaknesses, and then the DM simply picks a success percentage that feels right for a given character for a given task.

 

[Ibrandul]

The playtest rule is that you can auto-succeed on a natural 20, whether or not your roll meets the DC.

If your criteria for whether or not you'll allow a player to make the attempt is that they must have at least a sufficient bonus to meet the DC on a natural 20, then you are not implementing the new rule.

Correct. This was in response to people telling me that 2014 RAW demanded that no rolls ever be made by players unless a nat 20 would succeed. My question was, if you play that way, how do you adjudicate the situation I was describing?

Yes they are. The rule is that the DM determines whether or not a roll is warranted. What criteria the DM uses in making that determination is up to them.

That’s a good point. The 2014 rules don’t provide any substantial guidance on the matter, so DMs are mostly on their own in coming up with criteria for this.

 

[UngainlyTitan]

Thanks for explaining. It’s great to get an insight into other people’s styles. You're right—our basic philosophies are on different planets! I've never even considered the idea of setting DCs based on which character is attempting the task. I wonder how many DMs operate that way?

More than you might think.

It seems to me that at that point you may as well dispense with stats entirely. And indeed, there are indie RPGs that do so—the players and the DM come to a mutual, qualitative understanding of the PCs’ strengths and weaknesses, and then the DM simply picks a success percentage that feels right for a given character for a given task.

Well no that would be too much, the stats inform the chance of success.

Imagine that you, I and Sabine Hossenfelder walk in to a room, We see a wall with arcane marking. I don't know about you but I recognise the mathematical nature of the markings.
I make an int check, if I beat the easy DC (10) I know it is about physics. If I beat the hard I might puzzle that some of it involves the general relativity equations.
I am not going to get more than that with out spending a couple of years with some very good math and physics books and I might still not get it.
Sabine looks at it and realises it is a derivation of the Alcubierre Warp Drive, she makes her int check and tell us that it a version that could be built using current tech.
30 years ago and closer to my maths degree I might be able to determine that it was a warp of spacetime description but I kind of doubt it. I would have gotten further but not by much.

I have no idea what you would make of it. My view is that the difficultly of doings something is not a flat number it is informed by who ever is trying to attempt it.
40 years ago I could struggle through a French newspaper now I could just tell you it is in French.

 

[Maxperson]

Thanks. That's a very common house rule—but it is a house rule. And it nerfs the Bard's Jack of All Trades feature, among other things.

It's not a house rule. It's a different application of RAW. RAW states that you roll when the outcome is uncertain and there's a meaningful chance for failure. Gating rolls behind being proficient is RAW, because the outcome is not uncertain to me for those who are not proficient. So far as I know, nothing says I have to give every PC equal chances.

I also don't do that for every ability check. For a lot of them anyone can roll, proficient or not. For some you would need to have gone to school(be proficient) to learn that sort of thing.

As for bards, I value bardic knowledge(now Jack of all Trades) very much and give bards much more leeway to know obscure things than I do the other classes.

You keep saying "if failure is meaningful." I know that's language from the RAW, but I confess it has always struck me as highly ambiguous in many situations.

Sure, it's easy when someone is trying to lift a gate: no success, no lift, and that's obviously meaningful. But what about with knowledge checks? Do you think it's not meaningful to learn a piece of lore (or fail to) unless it impacts the plot you have planned for the party? There are dozens of examples throughout WotC's published adventures of lore that's gated behind a roll but isn't obviously useful. (Not that WotC always follows RAW when writing their adventures, but still...)

It's not so much ambiguous as subjective, but since the DM is making the decision to call for a roll or not, it's what he thinks that matters.

With lore it depends on what is at stake. If a wizard wanted to know if the magical light that is lighting up the inn was from a permanent light spell, the blessing of an angel or whatever, I'd just tell him. What does it matter if he fails that particular knowledge roll? If on the other hand he was looking at magical runes in the Death Knight's castle and wanted to see if he could figure out what they were for, he'd need to roll. Failure in that instance has meaning and it is very much in doubt.

 

[Reynard]

If a 20 always succeeds then the argument that there is "no chance of success" is severely weakened. I don't want to have that argument all the time with my players.

But, I already committed to the change for my Iron Gods 5E campaign so I will get to actively test it (alongside no monster crits) so we'll see.

 

[Maxperson]

If a 20 always succeeds then the argument that there is "no chance of success" is severely weakened. I don't want to have that argument all the time with my players.

But, I already committed to the change for my Iron Gods 5E campaign so I will get to actively test it (alongside no monster crits) so we'll see.

The rule isn't that a 20 always succeeds. The rule is that "If you roll a 20 on the d20, the d20 test automatically succeeds..." it also says the following, "Rolling a 20 doesn’t bypass limitations on the test, such as range and line of sight." Not getting a roll due to impossibility is a limitation on the test, and the 20 won't bypass that.

Another thing is that the new rule is...

"The DM determines whether a d20 Test is warranted in any given circumstance. To be warranted, a d20 Test must have a target number no less than 5 and no greater than 30."

The bolded right there says that the d20 roll isn't automatic, even with the "20 always succeeds rule." The other part of that is that for the test to be warranted, the DC has to be 5 to 30. Why? Because anything less is automatically a yes and anything more is automatic failure.

So we have two different ways for the DM to deny a player a roll. First, the DM determines if a test is warranted. That includes gating the roll behind proficiency or any other limitation the DM wants. Second, if the DC is so high or so low as to be automatically successful or impossible, no roll is given.

There is no "A roll of 20 can succeed at anything attempted" as some are arguing.

 

[GMforPowergamers]

As for bards, I value bardic knowledge(now Jack of all Trades) very much and give bards much more leeway to know obscure things than I do the other classes

we have (mostly) settled on letting the bard count as prof in everything... so they get the rolls of 'only prof' (mostly)

we do ALSO however have a background thing were sometimes it's not "anyone can roll this" or "anyone prof" but it is "This one character can roll to know this" but we ALWAYS let bards aid/help action on those rolls if they want. (and if they have guidance they can stack that with bard inspiration and now advantage)

 

[Maxperson]

we have (mostly) settled on letting the bard count as prof in everything... so they get the rolls of 'only prof' (mostly)

Yeah. I let bards roll(mostly) as well.

we do ALSO however have a background thing were sometimes it's not "anyone can roll this" or "anyone prof" but it is "This one character can roll to know this" but we ALWAYS let bards aid/help action on those rolls if they want. (and if they have guidance they can stack that with bard inspiration and now advantage)

Yeah. Backgrounds matter in my game as well. Sometimes what a player has written in his background will give him a roll, and sometimes it will just be successful based on what he has seen and done.

In my current campaign one of the PCs rolled(we use Central Casting for backgrounds) that he had gotten access to a secret book on great evils of the planes. He gets a roll for anything CR 10 or higher that is native to a lower plane or an outsider, even if someone proficient wouldn't be able to roll, and if those proficient can roll, he gets advantage.

 

[Reynard]

The rule isn't that a 20 always succeeds. The rule is that "If you roll a 20 on the d20, the d20 test automatically succeeds..." it also says the following, "Rolling a 20 doesn’t bypass limitations on the test, such as range and line of sight." Not getting a roll due to impossibility is a limitation on the test, and the 20 won't bypass that.

Another thing is that the new rule is...

"The DM determines whether a d20 Test is warranted in any given circumstance. To be warranted, a d20 Test must have a target number no less than 5 and no greater than 30."

The bolded right there says that the d20 roll isn't automatic, even with the "20 always succeeds rule." The other part of that is that for the test to be warranted, the DC has to be 5 to 30. Why? Because anything less is automatically a yes and anything more is automatic failure.

So we have two different ways for the DM to deny a player a roll. First, the DM determines if a test is warranted. That includes gating the roll behind proficiency or any other limitation the DM wants. Second, if the DC is so high or so low as to be automatically successful or impossible, no roll is given.

There is no "A roll of 20 can succeed at anything attempted" as some are arguing.

Sure, but if the DC is 30 or less a player is justified in getting a roll.

 

[Ibrandul]

Sure, but if the DC is 30 or less a player is justified in getting a roll.

I think the fundamental wellspring of disagreement here is that some DMs (myself, for one) think of ability check DCs as the difficulty that an indeterminate character would have performing the task. And it turns out that other DMs (Maxperson and UngainlyTitan, for two, if I understand them correctly) think of ability check DCs as determined relative to the particular character who wishes to perform the task.

That's an enormous difference in DMing procedure, one that I hadn't realized existed until tonight. And it will affect everything that has to do with ability checks.

With that procedure, it's not a question of whether "a player" gets to roll; it's always a question of whether this player gets to roll. And moreover, it's not even a question of how hard the task "is," it's always a question of how hard the task is for this PC. Five party members could all climb the same rope, and they could have five different DCs for the task.

I can't imagine DMing that way, or playing that way (though if the DM were quiet enough about it, I guess I might never even suspect it, until the day when PC 2's lower roll succeeds on a task where PC 1's higher roll has failed!).

But I have to admit I can find nothing whatsoever in the 2014 rules that would prohibit it—not only because the 2014 rules give DMs practically infinite latitude for making rulings, but also because they give rather little guidance on how to set a DC for an ability check in the first place.