How do you set DCs?

[DrowPiratRobrts]

Just wondering if any of you have good systems for setting DCs? I've realized that I tend to ballpark something instead of setting a hard and fast number a lot of times. I'm okay with this approach as it's flexible enough to challenge players when they're going through things too easily to have fun and to give them a little break when they're rolling terribly. I already have the chart of 10, 15, 20, etc. in my head for difficulties btw, so I don't need that explained. This is more curiosity than anything.

Do you do this too? Do you write the DC down secretly before each check? Do you plan them all out in advance? Do dice rolls even matter in your games? Just curious how other DMs honestly do it.

[Millface]

I don't take it too seriously... in general I do:

DC 10 for something easy. An untrained adventurer of any kind has a 50-80% chance of success here depending on how high their ability scores are

DC 15 for something that a should be easy for someone trained, and moderately hard for someone untrained

DC 20 for something that should be moderately difficult for someone trained, and incredibly hard for someone untrained

DC 21-25 for something that should be very difficult, even with training, and next to impossible without training

As you grow in levels, get expertise, or increase your stats and proficiency, this accounts for furthering your training in a skill.

I do generally write them down in my notes before the session, but this system makes it pretty easy to create them off the cuff. If it's a medicine check and you're patching a cut, that's a 10, because it's simple. If you're trying to pop an arm back in its socket, however, someone without training could very easily make that worse, so that's somewhere in the 16-20 range.

[DrowPiratRobrts]

I don't take it too seriously... in general I do:

DC 10 for something easy. An untrained adventurer of any kind has a 50-80% chance of success here depending on how high their ability scores are

DC 15 for something that a should be easy for someone trained, and moderately hard for someone untrained

DC 20 for something that should be moderately difficult for someone trained, and incredibly hard for someone untrained

DC 21-25 for something that should be very difficult, even with training, and next to impossible without training

As you grow in levels, get expertise, or increase your stats and proficiency, this accounts for furthering your training in a skill.

I do generally write them down in my notes before the session, but this system makes it pretty easy to create them off the cuff. If it's a medicine check and you're patching a cut, that's a 10, because it's simple. If you're trying to pop an arm back in its socket, however, someone without training could very easily make that worse, so that's somewhere in the 16-20 range.

This is what I was referring to with the chart I mentioned. I'm just curious about the last bit you mentioned. So do you typically do a range like that when you have to make one up on the fly, or do you decide that it's a hard 17 or something.

[hymer]

Do you do this too? Do you write the DC down secretly before each check? Do you plan them all out in advance? Do dice rolls even matter in your games?
I'm sometimes still mulling over the exeact DC when I'm saying 'Roll a [whatever] check'. I certainly don't write the DC down every time. If someone is really fast with their roll (and some players are agonizingly slow in doing their rolls), I may never get so far as deciding the exact DC, if they roll really high or really low. Sure, I have the DCs of traps and locks and finding the clue in the closet, and suchlike. They're in my notes. But for checks I'm only just thinking of now, I just want to set a DC and get things rolling along.

Dice rolls definitely matter. If someone rolls a really high check, I'll try to give something substantive in addition to a mere success, or at least narrate it to sound particularly well done. But I try to think of ways to make proficiency and narrative sense matter, too. No reason to ask the str 18, athletic proficient guy to make a check to do every little bit of lifting or climbing. And if someone else tries later, just have them roll a DC 5 check.

[NRSASD]

I do both, by which I mean I pick a hard and fast number right before they roll. I've noticed I will have locks be in the high 20's range, but that's because our thief can reliably roll a +12+d8 bardic inspiration. Outside of that, I also give guards a high perception roll because "stealth team" has a rogue, a druid, and a bard, which frequently means they roll stealth checks in the mid 30's.

[Aaedimus]

For perception sometimes I'll set DC's to notice or realize more things or more precisely.

Infact, if they roll really low at times I've had them perceive incorrectly rather than nothing at all.

With a 10 he finds the book he was looking for, a 15 tells him something's off, higher than a 20 and he notices the glyph peaking out on the cover from behind the last page.

If he rolls a 1 he finds the wrong book... but he's sure it's the right one

[Theodoxus]

I've been using pre-written adventures mostly, so most DCs are already noted in the material. In times when the players will ask to roll something not listed (usually an Int check for information), I use basically the chart Millface posted.

I'm trying to move away from non-contested DCs though. I like the idea that if you're competent (skilled) in something, and have appropriate gear (rope for climbing; books for lore; feed and tack for animal handling, etc) and time (so, generally not in combat) then you auto succeed without rolling.

This had made probably 95% of rolls being contested - and I like that better, though I've stolen the Pathfinder 2 Training to add onto Proficiency, so someone who has Master level Stealth will more than likely escape the notice of someone just skilled in Perception.

[Pelle]

I usually base it on the percentage of success I think is appropriate for a reference hypothetical character who is either very talented, trained, or a little of both (+4). So for this character, should it be 75/50/25 % chance of success? Then DC 10/15/20.

The most important though, is figuring out what the roll is for. Is it simply to open the lock, or is it to open the lock before the guard returns? As often as possible, I try to let the players know the DC and potential consequences before rolling, it makes the roll much more significant. Example, in last session the rogue wanted to pick a door lock, and was informed he could estimate he would have no trouble to open it if given enough time, but to do it before the guard comes back would be a DC 15. The party took their chances, and rolled low. "While you are busy picking the lock, your spotter sees the guard turn the corner, what do you do?"

I usually don't plan DCs in advance, because it's so difficult to predict what exactly the players will be doing in the session, and it's so easy to make a judgement in play.

[strangebloke]

The important thing to remember is that even at level 1, DND adventurers are heroes.

So don't go overboard with calling for checks. If a guy wants to climb a normal tree, let him. In fact, for that specific case, the rules say that you have to let them climb the tree without a check.

Secondly, don't call for a check if they can succeed eventually. Yes, it might be difficult for a knight to scale a slippery cliff-face, but if its less than 15 feet and there's no time pressure he can try an infinite number of times without consequences, so all that rolling would be pointless.

Finally, don't call for a check if there's no way they'd be able to do it. Be careful with this one, but the classic example is someone trying to jump to the moon. It should be stated that ludicrous things are possible according to RAW if you can roll a 30+ check. For example, you can move immovable rods.

If you do call for a check, bear in mind the probabilities at play. A level one nobody with +0 stat an no proficiency can only make a DC 10 half the time. That level one nobody is you. If you could do it almost every time, make the DC pretty low. If you could do it, but only once in many attempts, make it pretty high.

[PhoenixPhyre]

I'll be honest. 90% of the time I don't set a detailed DC at all.

That's because one of a few things happens:
1) the fiction makes it clear what the outcome is (either positive or negative). No point in rolling.
2) It's somewhere in the middle, so a check is needed but the character has such a high modifier that any normal DC is meaningless (hello rogues with stealth expertise and reliable talent!)
3) Failure wouldn't mean anything except a trivial amount of time.

If none of those cover it, then I ask them to roll, only knowing the DC is in the 10-20 range (with 20s rare) . Usually, the roll makes it clear if they succeeded or failed. With low-level characters, a 1-4 on a roll is probably a failure, even at DC 10. And a roll of 15+ is probably a success. Only when they roll in the middle do I have to worry, and usually then it's a partial success/degrees of failure situation.

For INT skills, I'm usually doing degrees of success (you remember something even on a 5, but you remember more as the roll increases). Other cases the degree of success/failure is clear from the fiction.

[DrowPiratRobrts]

Finally, don't call for a check if there's no way they'd be able to do it. Be careful with this one, but the classic example is someone trying to jump to the moon. It should be stated that ludicrous things are possible according to RAW if you can roll a 30+ check. For example, you can move immovable rods.

This is a perspective I haven't heard before. It's interesting. I think while I approach it differently, our results end up the same. If a player at my table says they want to push over the castle wall I tell them to make a strength check. Nat 20s at my table just mean the best possible (or maybe plausible) outcome. So no matter how high they roll my response is, "You run into the wall and it's still standing. You're left wondering why you just tried to run into a wall." If they rolled a 1-5 they might take a point of damage or fall over or something. Nothing major, but I've found that my players enjoy this sort of thing. It's funny, and it helps establish boundaries for them in fiction so that they have a spectrum of things that are and are not possible.

[strangebloke]

This is a perspective I haven't heard before. It's interesting. I think while I approach it differently, our results end up the same. If a player at my table says they want to push over the castle wall I tell them to make a strength check. Nat 20s at my table just mean the best possible (or maybe plausible) outcome. So no matter how high they roll my response is, "You run into the wall and it's still standing. You're left wondering why you just tried to run into a wall." If they rolled a 1-5 they might take a point of damage or fall over or something. Nothing major, but I've found that my players enjoy this sort of thing. It's funny, and it helps establish boundaries for them in fiction so that they have a spectrum of things that are and are not possible.

Well, my attitude is basically that a roll that doesn't do anything is pointless. If they were never going to push over the city wall, then don't wait for them to get out their D20, roll it, and add up their modifiers. Just say "nope!"

I'd personally be a little peeved that my nat 20 was 'wasted' on an impossible check.

[Millface]

This is what I was referring to with the chart I mentioned. I'm just curious about the last bit you mentioned. So do you typically do a range like that when you have to make one up on the fly, or do you decide that it's a hard 17 or something.

It's on the fly and based on a few factors like their surroundings or whatnot.

If something is a 15 on the chart, but you're in combat or the situation around you is stressful, that's where I use the numbers in between the chart, so 16-19. 19 would basically be something that falls under 15 happening under a really bad circumstance or distracting situation.

Or, the difference between setting a bone back in its socket in a medical tent after the battle (20), or on the field of combat during a large scale battle (24).

For fluff, I also describe how well or effortlessly they do it, or conversely, how hard it was for them, based on how much the beat the check by.

If you're trying to push a boulder out of the way, and make an athletics check and hit the DC on the nose, "You struggle and push with all you've got, just when you think you're going to run out of steam, you manage to move the boulder to the side just enough."

If they crushed the DC... "You approach the boulder and set yourself to shove, with a single, mighty heave, you roll it to the side."

[Pelle]

This is a perspective I haven't heard before.

Really? It's the basics of 5e resolution system, don't call for rolls if it is automatic or impossible.

[Joe the Rat]

Actually, a check for something that can succeed with enough attempts is a good way to gauge how long they take to do it. Sometimes time is critical.

When I'm assigning DCs, I usually go 8 / 10 / 12 / 15 / 18 / 20 / 25 / 30

8 is for Typical difficulty tasks where there is a notable consequence for failure. Otherwise they don't roll.
10 is Medium difficulty
12 is Moderate difficulty - something an average person with proficiency has about a 50/50 on
15 is Hard
18 is Challenging difficulty - the kind of thing that looks awesome or people will talk about, but is still in the range of local yokels.
20 is Heroic - even the highly trained have only modest chances
25 is Legendary - if I don't think you'll do it, but I want to give you a chance, or if you're a rogue engaged in extreme shenanigans, here you go. If there are witnesses, you will be remembered for this. It is also my "I don't have an answer to this mystery" difficulty. When I set up an idea, but don't have an answer (where did the burial mound with the Wight come from?), it took a nat 20 from the PC History buff to hit it. Which he did. Every. Single. Time. I got good at improvising lore.
30 is Gods and Punks. You're either a demigod, a top tier hero, or a damn expertise rogue to hit this.

I will also do a sliding DC: I have a target number in mind for "Pass," and the actual roll tells me something about how well or poorly you do. A 12 is enough to not make a fool of yourself performing at a festival, but the higher your checks, the better the performance. The amount you miss that DC12 determines the number of rotten vegetables thrown at you. Missing a DC by a point or 2 may result in a Success with Consequence option.

[Aaedimus]

Really? It's the basics of 5e resolution system, don't call for rolls if it is automatic or impossible.

Infact people forget this fact quite often, and you can see a real detriment to games who suffer from over-rolling. The players don't even know it's hurrying their game many times because it's such a common mistake

[Oramac]

If none of those cover it, then I ask them to roll, only knowing the DC is in the 10-20 range (with 20s rare) . Usually, the roll makes it clear if they succeeded or failed. With low-level characters, a 1-4 on a roll is probably a failure, even at DC 10. And a roll of 15+ is probably a success. Only when they roll in the middle do I have to worry, and usually then it's a partial success/degrees of failure situation.

For INT skills, I'm usually doing degrees of success (you remember something even on a 5, but you remember more as the roll increases). Other cases the degree of success/failure is clear from the fiction.

This is more or less what I do too. I'd say 90% (ish) of the time I use more of a DC Range than a hard number. If I do set a specific number, it's usually 15.

I tend to focus a lot on the narrative rather than the mechanics, so I'll use degrees of success for any given roll. It does take a bit of improvising, but I've found that my players like it, and it makes the story more fun for them.

For example (off the top of my head): I might have a player say they want to jump across this 15 foot wide chasm. So I ask for an Athletics check. I know already that this is a feat of strength that would be difficult for most people, so I'll use that as a gauge for their result.

Nat 20: "You run full speed towards the edge of the chasm, taking your final leap at the last possible second. Flying through the air, you're certain of your jump and as you land you perform a flawless shoulder roll, standing up and taking a bow before your friends"

Modified 20: "You run full speed towards the edge of the chasm, taking your final leap at the last possible second. Your footing slips slightly, but you're confident in your jump. Reaching the other side, you land rather gracefully, but not so much as to be proud of your jump"

15-19: "You run full speed towards the edge of the chasm, taking your final leap a foot back from the edge. Your jump is well done, but certainly not flawless. Landing on the other side, you slightly tweak your ankle (not enough to take damage)."

12-14: "You run full speed towards the edge of the chasm, taking your final leap a couple feet from the edge. You're not so confident in your jump, and you can tell you're coming up a bit short. You land on the other side with your waist right at the edge of the chasm, knocking the wind out of you. You manage to climb fully to the other side, but know your jump looked rather hilarious" (might take 1d4 damage)

10-11: "You run full speed towards the edge of the chasm, taking your final leap a couple feet from the edge. Your jump is clearly not great, and in the air you begin looking for something to grab as you land short of the chasm face. Luckily you are able to catch a strong vine (or whatever) and keep yourself from falling into the chasm" (takes 1d6 damage and make another very easy Athletics check to climb up to the top of the chasm)

1-9: you fall into the chasm.

They're not all like this, but it's a good example of something I might do.

[Sigreid]

/badjoke: I set the DC for climbing small trees to 25 and using that as a reference go up from there.

[strangebloke]

Really? It's the basics of 5e resolution system, don't call for rolls if it is automatic or impossible.

Players want to roll for everything in my experience. To the point that I had players get annoyed when I said: "you can't roll low enough and they can't roll high enough. You succeed on your stealth check."

So players pressure DMs, and DMs give in because its easier for them to say 'you fail' if the roll is bad. It works great until someone rolls a nat 20 trying to lift castle greyskull.

[Keravath]

I do both, by which I mean I pick a hard and fast number right before they roll. I've noticed I will have locks be in the high 20's range, but that's because our thief can reliably roll a +12+d8 bardic inspiration. Outside of that, I also give guards a high perception roll because "stealth team" has a rogue, a druid, and a bard, which frequently means they roll stealth checks in the mid 30's.

Interesting. So in your game, what is the point of characters working to specialize and be good at specific skills? The thief could have just ignored taking expertise in thieves tools for example ... or expertise in anything for that matter ... since you just make the DCs harder because they have high skills. Doesn't this make it pointless from the player perspective to try to be good at anything since you just respond by making their choices meaningless by making everything harder?

That thief, once they reach level 11 will have a minimum 22+d8 to open a lock due to reliable talent. It will get higher as proficiency goes up ... +17 at 17th level with 20 dex and expertise in thieves tools. It gives a minumum of 27. By this point you will be creating locks with DC of 35 to 40 ... beyond impossible (30) ... just because the thief specialized in opening locks.

Similar story with perception and stealth. The creature that can see the hidden rogue will trivially spot everyone else. Why bother with stealth? Pass without trace from the druid combined with the rogue's abilities and expertise in stealth will give the rogue a base stealth of 37 (+27+10) ... so a creature will have a perception of 40 so they can notice the rogue about 1/2 the time?

Anyway, my comment would be that the different classes have different strengths and weaknesses and making DCs more challenging just because it is something the characters are good at basically says to the characters ... don't bother being good at anything since I will make it hard no matter what your scores are.

Picking locks, disarming traps and noticing things are not supposed to be hard for a character that is specialized in these areas. On the other hand, the average paladin doesn't have a prayer of noticing the trap, seeing the hidden ambush or sneaking up on anyone (unless of course those same traps and enemies have a much lower DC when the paladin is looking than the rogue ...?)

[Keravath]

Actually, a check for something that can succeed with enough attempts is a good way to gauge how long they take to do it. Sometimes time is critical.

...



Yep ... this is where passive skills can play a role. If they have time to make a few attempts and failure doesn't have significant consequences then check the passive skill first and if that is good enough just narrate the skill check.

e.g. After the player says they are searching a room. check their passive perception ... if it is high enough ... just narrate "You notice faint traces of scratchs on the floor near the wall. Closer examination shows that there may be a secret door there" .. if not, ask for a roll.

Similarly, when the character tries to look for a mechanism to open it ... check passive investigation .. if it is high enough ... just narrate ... "after searching the wall you find a small indentation about two feet to the left of the opening ... it looks like you could press it" ... if the passive investigation isn't enough, ask for a roll.

If the character says they will then search the door mechanism for traps ... check their passive perception or investigation first ... ask for a roll if it isn't sufficient. Also, if there is a chance that investigating will set off the trap then ask for the roll anyway ... reveal the information if the passive is high enough but the roll sets off the trap.

Passive skills exist for situations where you don't have a time constraint. It also lets the characters with high passive skills make good use of them ... they will be able to automatically succeed on the lower DC checks if they have sufficient time ... which is exactly as it should be for a skilled character.

[Pelle]

Players want to roll for everything in my experience. To the point that I had players get annoyed when I said: "you can't roll low enough and they can't roll high enough. You succeed on your stealth check."

So players pressure DMs, and DMs give in because its easier for them to say 'you fail' if the roll is bad. It works great until someone rolls a nat 20 trying to lift castle greyskull.

True that. One of my players feels it is a bit patronizing when the DM just 'allows' him to succeed, and want to roll for Stealth checks he can't fail to demonstrate his investment in the skill and succeed by his own volition. It is a reflex from 3.5, most of my players are used to and aboard with the 5e approach now. Great improvement IMO, it runs much smoother at the table and there's less 'roll-playing' also.

[strangebloke]

True that. One of my players feels it is a bit patronizing when the DM just 'allows' him to succeed, and want to roll for Stealth checks he can't fail to demonstrate his investment in the skill and succeed by his own volition. It is a reflex from 3.5, most of my players are used to and aboard with the 5e approach now. Great improvement IMO, it runs much smoother at the table and there's less 'roll-playing' also.

In many cases the 'you have to roll' ethos of 3.5 was downright insulting.

"Roll a will-save, Mordecai" *snicker*

You could succeed on a nat 20 in many cases, but that was also the only way you were succeeding on a lot of checks you had to make.

[DrowPiratRobrts]

Really? It's the basics of 5e resolution system, don't call for rolls if it is automatic or impossible.

I agree in general and often defer to that. I took what he said differently I think, so maybe I misunderstood his meaning. He said that RAW someone could roll 30+ to move an immovable rod or something similar. It seemed to me that he was saying he doesn't allow players to roll for things like that are questionably plausible, because he realizes that if they make the DC that he sets then he is obligated to allow them to "do the impossible." Not sure if that makes sense, but I probably just misunderstood what he said.

So I get why people just say, no you can't push the wall over. But rolling too much isn't typically part of any of my games. We often go 20-30 minutes with only a handful of rolls. So when somebody wants to try something stupid I enjoy making them play it out with consequences (whether that's spending their time rolling for something they know they'll fail at or giving their character a point of damage). It seems to cut down on the, "I do this stupid thing...hahaha, no I'm just joking. I never did that." I could say more on my reasoning behind why I approach it this way, but it's specific to me and my groups and won't apply to everyone.

[DrowPiratRobrts]

Yep ... this is where passive skills can play a role. If they have time to make a few attempts and failure doesn't have significant consequences then check the passive skill first and if that is good enough just narrate the skill check.

e.g. After the player says they are searching a room. check their passive perception ... if it is high enough ... just narrate "You notice faint traces of scratchs on the floor near the wall. Closer examination shows that there may be a secret door there" .. if not, ask for a roll.

Similarly, when the character tries to look for a mechanism to open it ... check passive investigation .. if it is high enough ... just narrate ... "after searching the wall you find a small indentation about two feet to the left of the opening ... it looks like you could press it" ... if the passive investigation isn't enough, ask for a roll.

If the character says they will then search the door mechanism for traps ... check their passive perception or investigation first ... ask for a roll if it isn't sufficient. Also, if there is a chance that investigating will set off the trap then ask for the roll anyway ... reveal the information if the passive is high enough but the roll sets off the trap.

Passive skills exist for situations where you don't have a time constraint. It also lets the characters with high passive skills make good use of them ... they will be able to automatically succeed on the lower DC checks if they have sufficient time ... which is exactly as it should be for a skilled character.

My only contention would be that this doesn't allow for small details to be missed by players. That might be what you're going for, but I like the element of missing information occasionally. When rolling perception I'm typically accounting for thoroughness but also for luck. You might be able to easily notice the bloodstain under the bed from most angles, but if you roll a 2 it just so happens that you look under the bed and believe you've taken in all the information, not noticing anything with a careful look. In reality, if you moved your head to the right about a foot you'd have noticed that the leg of the bed (or a shadow or something) was blocking the small stain from view.

[Pex]

If you would like specific examples to help set your own DCs you can borrow the DC skill tables from 3E/Pathfinder. I would suggest subtracting 5 from all the DCs to reflect the bounded accuracy math of 5E. Condition modifiers on the tables that would add or subtract 5 would instead give advantage/disadvantage.

[NRSASD]

Interesting. So in your game, what is the point of characters working to specialize and be good at specific skills? The thief could have just ignored taking expertise in thieves tools for example ... or expertise in anything for that matter ... since you just make the DCs harder because they have high skills. Doesn't this make it pointless from the player perspective to try to be good at anything since you just respond by making their choices meaningless by making everything harder?

That thief, once they reach level 11 will have a minimum 22+d8 to open a lock due to reliable talent. It will get higher as proficiency goes up ... +17 at 17th level with 20 dex and expertise in thieves tools. It gives a minumum of 27. By this point you will be creating locks with DC of 35 to 40 ... beyond impossible (30) ... just because the thief specialized in opening locks.

Similar story with perception and stealth. The creature that can see the hidden rogue will trivially spot everyone else. Why bother with stealth? Pass without trace from the druid combined with the rogue's abilities and expertise in stealth will give the rogue a base stealth of 37 (+27+10) ... so a creature will have a perception of 40 so they can notice the rogue about 1/2 the time?

Anyway, my comment would be that the different classes have different strengths and weaknesses and making DCs more challenging just because it is something the characters are good at basically says to the characters ... don't bother being good at anything since I will make it hard no matter what your scores are.

Picking locks, disarming traps and noticing things are not supposed to be hard for a character that is specialized in these areas. On the other hand, the average paladin doesn't have a prayer of noticing the trap, seeing the hidden ambush or sneaking up on anyone (unless of course those same traps and enemies have a much lower DC when the paladin is looking than the rogue ...?)

This is very definitely a case of YMMV, what-works-in-my-games-doesn't-necessarily-work-in-yours, but here goes.

I should also mention I use a house rule that makes NPC passive skills a lot more useful. 4 characters try to sneak past a sleeping guard with a passive perception of 8. Rather than give the PC fighter a 40% chance of waking him outright, I prefer to multiply the passive perception of 8 by the number of characters attempting the stealth roll, so 4 in this case. That gives us a new DC of 32, which the sum of player stealth rolls try to beat. So if the rogue rolls a 29 and the fighter rolls a 5, the guard stays asleep. While not painstakingly realistic, it does let skilled characters compensate for their clueless colleagues.

I've noticed my players really appreciate feeling like the resources and extra effort they expend on making a roll is well spent. If a guard has a passive perception of 13, and our rogue has a base +6 stealth at 1st level, he's going to remain undetected pretty easily. Fair enough you might say, he's a rogue; it's his job to be stealthy. But when the rogue and druid team up at level 3 or 4, and the rogue starts rocking a base +9+10 pass w/o trace+d6 bardic inspiration, that just gets silly. If level 5 characters are the high end of commonly seen NPC adventurers, why would anyone ever hire guards?

So I try to compensate for it in 4 different ways, listed in order of how I like to deploy them.

A. Have better guards. Not statline padding, but higher quality guards/monsters. A night watchman may have a passive perception of 13, but what about the passive perception of an assassin/temple guardsman over-watching a very important secret meeting? I don't want to drive his WIS score through the roof just to challenge the PCs stealth-wise, cause a high NPC WIS score limits their options unnecessarily. So I'll give him a +1 WIS bonus, a +3 proficiency with expertise (for +6), and a +6 circumstance bonus of being in their own temple patrolling hallways where they themselves know how to hide while guarding a meeting that could have dire, personal consequences for them if it was overheard.

Having guards with a base +13 to perception means the players are more likely to weigh the merits of using that potion of invisibility or stone form, of splitting themselves into Stealth team and Charisma team, of debating whether bringing the bard along is worth it for the inspiration. In other words, it forces the players to make interesting choices, which is ultimately why my group plays D&D.

B. Make the objectives themselves more difficult. Sneaking past a night watchman is easy. Stealing his coffee mug off the table in front of him without him noticing is not. In that case I'd give the base DC of 13 a hefty +15 circumstance bonus, cause he's literally looking at it right there and hey where'd it go?

The PC still has a chance for failure, and feels even more epic for pulling it off. The PC wouldn't have dared to try it when he was a 1st level character, but now that's he's a 5th level it's a piece of cake.

C. Change the base difficulty DCs based in the beginning of the campaign based on what the party is strong at. If a party is heavily geared towards social and stealth prowess in the beginning, the average DC for such checks might be 13 while athletics checks might be 10. The party is going to get a lot better at stealth and social stuff way faster than they will with athletics, so it'll keep the difficulty more relevant longer. If they autowin every challenge, why are we rolling dice at all?

D. Pad the base guard statline. I don't like doing this, so I rarely do. But sometimes things get very out of line due to unforeseen treasure consequences or interesting and hilarious PC shenanigans, and a little bit of editing is needed to keep the world's verisimilitude if not intact, then at least functional. If you're playing to "win" or "break" the D&D game, you probably won't fit in at our table. This is in no way a judgement on those who do play for those reasons, just why I DM the way I do.

[Theodoxus]

If you would like specific examples to help set your own DCs you can borrow the DC skill tables from 3E/Pathfinder. I would suggest subtracting 5 from all the DCs to reflect the bounded accuracy math of 5E. Condition modifiers on the tables that would add or subtract 5 would instead give advantage/disadvantage.

The other thing I do, is use 2d10 instead of a d20 for ability checks. This gets away from the idea of a critical fail, though my players will still call out a "nat 20" for rolling double oughts... /sigh.

Lowering DCs becomes mandatory when you're rolling lower than average due to the bell curve. But it also enhances the importance of being skilled and trained over the luck of the die.

[ErHo]

Well, whatever you do, do not go to this post and use it as a rough guide because it has almost every situation example, albeit not too consistent with cross referencing DCs vs certain materials, etc.

http://www.giantitp.com/forums/showthread.php?569662-WIP-Ability-Check-DC-Masterlist-All-the-ability-check-DCs-from-official-adventures :smallcool:

[Aaedimus]

I should also mention I use a house rule that makes NPC passive skills a lot more useful. 4 characters try to sneak past a sleeping guard with a passive perception of 8. Rather than give the PC fighter a 40% chance of waking him outright, I prefer to multiply the passive perception of 8 by the number of characters attempting the stealth roll, so 4 in this case. That gives us a new DC of 32, which the sum of player stealth rolls try to beat. So if the rogue rolls a 29 and the fighter rolls a 5, the guard stays asleep. While not painstakingly realistic, it does let skilled characters compensate for their clueless colleagues.

This houserule is mathematically identical to averaging the group's checks, which is a RAW tool for many situations. Instead of dividing the combined total by the group, you're multiplying the target by the same number.