DC shenanigans

[Notafish]

From the thread on "D&D One" or whatever it's going to be called:


Right. An impossible DC and "DM no" are identical.


That's not the case. DCs are objective, not subjective -- what is DC 10 for one character is DC 10 for all characters. That's why getting better at something is expressed as a higher bonus to your roll, rather than a lower DC.

The question here is whether there is any DC 30 task that a novice creature should not be able to achieve, point blank. The answer is trivially yes. Meanwhile, this system has the novice succeeding at the maximally difficult task 5% of the time.

The solution for this problem would be to express getting better at a skill through lowering DCs rather than a bonus to the d20 roll. That would allow natural 20s = automatic success to make sense for all characters. But it would make skills an exception to the game's core d20+X mechanic. I think that would be more confusing than saying nat 20 = success doesn't apply to ability checks.

This is part of a longer discussion on whether having a task with a fixed DC should be able to be attempted by all players, and what the DM's roll in calling for the check should be. I wanted to share my thoughts in a separate thread - there are a bunch of directions to go down when discussing when to call for rolls and how to set DCs, beyond the question of whether DC tables are a good thing.

I fall into the camp that believes that players should only roll if a) the task is possible in the narrative and b) the consequences of a success or failure are interesting. I also think it is the usually the GM's responsibility/burden to make this call. Back when I was DMing a 5e campaign, I would often set the DC at the time of the roll, as well - using the guidelines in the PHB to set the DCs roll by roll. A few times, I changed the DCs on a task depending on who was attempting it on the logic that (for example) making a Medicine check to identify a poison would be easier for a Cleric with Medicine proficiency than an untrained Fighter just trying his best. I didn't think it through at the time, but this was effectively a way to give characters situational expertise without granting the feature, since I was reducing the DC and they were also adding their proficiency bonus. It worked well enough for my home game with friends, but this is also a somewhat sketchy houserule - I can see where this style would rankle certain players. I can also see that flexible DCs would take extra effort to incorporate into a published adventure. But not impossible.


Captain Dorcus lies slumped in her chair. Her skin is an unnatural shade of green and her breaths are shallow. She has obviously been poisoned, but to find an antidote in time the poison must first be identified. Identifying the poison and stabilizing Dorcus requires a DC15 Wisdom (Medicine) check. If a character is Proficient in Medicine, Poisoner's Kits, or Herbalism Kits they gain +5 on the check. Each attempt will take time and after 3 failed attempts Captain Dorcus will die.

In this example as written, every PC could attempt a check, but the characters with medicine or other relevant training would have a much higher chance (and their chances could be further improved due to gaining Advantage through some other means). There are clear consequences for success and failure for this challenge, and the odds are vastly different depending on what sorts of characters are attempting the check. The way a success will likely be imagined will be different, too, depending on what (if any) training applied to the check. This doesn't have to be the format for all checks, but it could add spice and guidance in a pre-written adventure.

Now, what about situations where it is unrealistic in the fiction for some (but not all) characters to have a chance of success? Is it OK to have some characters simply be denied a roll, or should DCs just be set too high for untrained/unlevelled characters to attempt? Is there a difference?

I think that sometimes it is appropriate for characters to be denied rolls, but if it is happening often it is likely indicative of a problem outside of the skill check system. I also think that there is very much a difference between having a high "objective DC" and having some checks that can be made by certain characters but not others.


There is a treasure chest with a DC10 Dexterity (Thieves' Tools) lock. The chest is DC 20 Strength (Athletics) to smash open because I can't be arsed to look up the rules for breaking materials. Should a Level 1 Druid with Dex 14, Str 8 and access to (but no proficiency with) Thieves' Tools have a chance at opening the chest with either of these checks?

I can actually imagine me ruling either way here. A lucky roll from the druid could be a cool moment, but I might not want to encourage the sort of game where players solve problems by rolling a lot and hoping for high numbers. If I tried to make the game safe from wackiness via high DCs, though, I will not be able to put this treasure chest in a starting dungeon - if my goal is to put the possibility of success out of reach for weak or unskilled characters, I'll also be pushing the chance of success lower for proficient characters.

[Skrum]

While I haven't actually put this type of rule into practice, on paper at least I would favor a rule that put some kind of a cap on players making skill checks using skills they aren't proficient in. This could be not allowing players to make rolls on stuff that has higher than a DC 15, capping their total at 15 (effectively the same), or making determinations on particular checks over whether it is "skilled" or not. An untrained person might get lucky and jump across a river that they didn't think they could make, but you can't get lucky and put on a great musical performance.

Especially at lower levels, it's relatively common for someone to be far out-performed by an untrained party member that rolls a 18 with a -1 modifier while you roll a 3 with your +5. Some skills, that makes sense, but for others, no way. And it kinda sucks to be playing a character that's invested resources into something only to get showed up by someone that's spent literally nothing on it.

[Tawmis]

I think that sometimes it is appropriate for characters to be denied rolls, but if it is happening often it is likely indicative of a problem outside of the skill check system.


Here's why I would disagree.

Ever have someone you know, who has no idea what they're doing, try to fix their own computer and make things worse?

The lack of knowledge and skill shouldn't deny someone the ability to try.

They will simply likely fail (and potentially make things worse).

[diplomancer]

I've said this on another thread: in order to make "objective DCs" work at all, where the DC is only about the objective difficulty of the task itself and not the capabilities of the character, you'd have to have a system that starts DCs in the low 20s for a lot of tasks (more specifically, those tasks that requiretraining, but can be done somewhat reliably once someone hastraining, and goes into about one hundred for the really hard ones, and abandon Bounded Accuracy completely.

Take playing a piece of music as an example; to play a relatively simple one in the piano (like a Bach's piece from his Anne Magdalene book for students) takes about at least one year of studying. A one-year student canmanage it, sometimes, but he's still struggling, and how well he will play is a bit uncertain. Meanwhile, someone who's never studied music has exactly 0% chance to succeed, and a concert pianist can play it in his sleep (and maybe does! 😜).

So, since playing the piano is mostly a Dex check, you'd have to have a DC of about 25 for it (so untrained people, no matter how dexterous, simply can't do it), and give a 1st year student about a +10, a 3rd year student about a +25, etc.

Such a system could work, but it's not 5e.

[Idkwhatmyscreen]

I think that the correct middle ground is nesting DC's together so that the party member who has no chance of success on a given roll can still contribute towards the task at hand. Basically, if you have a DC 25 Arcana Check, you will also have a DC 15 Arcana check that the player is rolling at the same time, if they clear the first DC, they can help the party in some small way with the check (Maybe this lowers the DC on the main arcana check) and passing the second check would give the party the full info

Now the Barbarian can roll an Arcana Check and if they are lucky, they can identify a single rune (which is not unreasonable, the symbol may be related to a clan crest they know, maybe they learned it from their tribe's shaman or whatever)

Of course, you can't really have nested DC's in a system that is forcing 20's to be auto successes and 1's to be auto fails

[HidesHisEyes]

I don’t think DCs are supposed to be objective in 5E. OP I run things mostly the way you describe - although I don’t reduce a DC just because a character has proficiency, since as you say that’s taken care of already by the proficiency bonus. But I might reduce it because of their background or factors that have emerged in the game’s fiction so far.

But I actually think the really important thing isn’t the DC but the stakes, ie the meaning of success and failure - which can and very often should be different for different PCs, I think. For example, performance check in a tavern to impress the townspeople and get them to like you:

Bard succeeds - an awe-inspiring performance that has the audience rapt and immersed.
Fighter succeeds - an out-of-tune but enthusiastic folk song rendition that has everyone singing along.
Bard fails - a technically perfect but somewhat uninspired performance.
Fighter fails - just really bad.

I think there’s a tendency to think failing always means “really bad”, because of the binary nature of the dice mechanic. And that leads some people to say it’s too swingy. But once you start resolving checks in context, case by case, the swinginess is really not a problem.

[Person_Man]

The rule seem fine to me, because I see D&D as a fun game and not a reality simulator. For me, its more fun for the players to always have some chance of success, unless they're asking for something that's physically impossible (in which case the DM should just tell them its physically impossible and they don't get to roll).

Yes, there are plenty of corner cases where the DM would prefer not to tell the players that something is physically impossible, and then the player tries it and rolls a 20 anyway. (Bribing a guard who is actually the store owner to let the players steal, seducing someone who is not into your gender, disarming a trap that is far beyond the players' abilities but they have no way of knowing that). But in those very rare cases when a player tries something that is impossible, that the DM doesn't want to tell them is impossible, and the player rolls a 20, the DM can just figure out a creative solution, which would hopefully be more fun then the players not trying in the first place.

Also, failed checks can have negative consequences. So having 5 players each attempt something impossible doesn't increase your chance of success to 25%. Because after the first one tries something and fails, the guard can attack, the trap can blow up, or the DM can just say "you've discovered this is impossible and no further attempts will succeed."

[Psyren]

Here's why I would disagree.

Ever have someone you know, who has no idea what they're doing, try to fix their own computer and make things worse?

The lack of knowledge and skill shouldn't deny someone the ability to try.

They will simply likely fail (and potentially make things worse).

You are equating "try" with "roll." D&D does not do this.

Rolls mean one thing , that there is uncertainty in the outcome. Certain failure is not uncertainty.

[Tanarii]

DC is set by the task difficulty. So is automatic success or failure.
Bonus is set by the PCs ability score and proficiency, which represents their capabilities at related checks.

Task difficulty may be impacted by something physiological or character history related aspects, because they're actually different tasks.

Examples of possible check rulings related to the latter:
- if a creature has four feet (ie centaur or wolf), climbing a tree is DC 30. Otherwise it's automatic success.
- if your PC doesn't know something, they automatically fail checks to recall the information at a critical moment. Otherwise it's DC 10.
- if your PC lived in Candlekeep, the DC to recall the details of {widely known Candlekeep history} are DC 10, otherwise it's DC 20.

The latter may seem like it's a character specific varying DC, but it's actually two separate but closely related tasks with different DCs. Same for the others, they're two different tasks each.

--------

But under no circumstances should that be considered an endorsement of basing DC on race, class, background, or skill proficiency. IMO that's not what 5e intends at all. If you're setting a different DC or automatic/failure success for two different characters, do it base on established facts at the table that make them two different tasks. Not archetype assumptions.

[RazorChain]

I often just allow PCs to roll that have the relevant knowledge. A fighter trying to find out how a murder victim was killed with a weapon, sure roll a relevant skill. A fighter with no poisoners kit proficiency or medicine skill trying to identify poison why would he roll?

I'm running a game that happens in Baldur's gate atm and the PC's that have come from Baldur's Gate can roll on things relevant to their expertise because they are locals. The others who are playing outsiders don't get the same rolls on local lore, society, laws etc.

Sometimes I just ask the players if their characters should know anything about this or that. If they come up with a reasonable explanation they get to roll.

Some skills of course are always functional in many circumstances, like perception, athletics, insight, stealth but only those who are proficient have the extra knowledge to assess a situation properly, like how hard a task is.

[HidesHisEyes]

I often just allow PCs to roll that have the relevant knowledge. A fighter trying to find out how a murder victim was killed with a weapon, sure roll a relevant skill. A fighter with no poisoners kit proficiency or medicine skill trying to identify poison why would he roll?

I'm running a game that happens in Baldur's gate atm and the PC's that have come from Baldur's Gate can roll on things relevant to their expertise because they are locals. The others who are playing outsiders don't get the same rolls on local lore, society, laws etc.

Sometimes I just ask the players if their characters should know anything about this or that. If they come up with a reasonable explanation they get to roll.

Some skills of course are always functional in many circumstances, like perception, athletics, insight, stealth but only those who are proficient have the extra knowledge to assess a situation properly, like how hard a task is.

I like this kind of thing.

The other thing I think people overlook in these discussions is it’s not just about whether there’s a roll or what the DC is, but also the stakes of a roll. That is, what success and failure mean to different characters and in different situations. A fighter might be able to make a check to suss out some arcane runes, and if they succeed maybe they happen to recognise a couple of the runes and know their literal meaning, but don’t know much more than that. Meanwhile the wizard might make the same check (at the same DC even), fail, and still be able to hazard a bit of a guess as to what the runes are.

The secret is that it’s not the same check at all, because ability checks aren’t immutable entities stapled to objects in the game world. They are a tool we use to resolve moments of uncertainty on a case by case basis.

[Psyren]

I like this kind of thing.

The other thing I think people overlook in these discussions is it’s not just about whether there’s a roll or what the DC is, but also the stakes of a roll. That is, what success and failure mean to different characters and in different situations. A fighter might be able to make a check to suss out some arcane runes, and if they succeed maybe they happen to recognise a couple of the runes and know their literal meaning, but don’t know much more than that. Meanwhile the wizard might make the same check (at the same DC even), fail, and still be able to hazard a bit of a guess as to what the runes are.

The secret is that it’s not the same check at all, because ability checks aren’t immutable entities stapled to objects in the game world. They are a tool we use to resolve moments of uncertainty on a case by case basis.

Indeed. And DMG 242 even tells you you have the mandate to do this. You get to decide, not just the DC, but what success and failure mean.