Passive Perception, Traps and Trying Not to Be a Bad DM

[Specter]

So basically, players are/will be in a dungeon filled with traps. Those traps are of the mind flayer variant, i.e. smart and very hidden traps.

While I'm creating these multiple traps, I'm assigning DCs to each of them. The only player I'm taking into account is the Cleric, who has 20 WIS, proficiency in Perception and the Observant feat (total passive of around 24).

But as I assign these DCs, I fall into a very simple problem: either he spots the trap, or he doesn't. Considering the former makes the trapfinding process dull and largely irrelevant, and the latter makes me a bad DM ("you couldn't ever see this trap, save for DEX"), how should I deal with this situation? Should some traps be passively untraceable?

[Naanomi]

There is precident for traps having one passive DC and a lower DC for an active check... or differences in DC between perception and investigation. There are also traps you need detect magic active to detect. This is all in published modules.

Still, the character invested a feat towards this, finding a good number of traps should be a reward for that

[Tanarii]

If your traps rely on not being spotted to work, they're effectively dead weight. Any character can take 10 times as long to search, and if it's possible to find the trap, they automatically find it. DMG automatic success rule. And (as you've noticed) the game assumes they front rank of the marching order is using their Passive Perception to look for threats. That means they're very likely to notice something is wrong, even moving along at a normal exploration speed.

This is intentional. 'Gotcha' traps are terrible trap design. You're supposed to design traps that require interesting decisions even after the group notices that they're there.

Edit: Since that's mostly moralizing, I'll give you some concrete examples:

After finding the trap, they need to figure out how to disable it. (This may be as simple as an investigation check, or as complicated as the players asking 20 questions.)
After figuring out how to disable it, they need to successfully do so. (This may be as simple as a Thieves Tools check, or as complicated as locating the right resource and creatively applying it.)
... or how to bypass it.
... or it's not just a trap, but if bypassed correctly a secret door to the next area.
... or it's an in-combat trap, when each character (not just the front rank) needs to pass passive perception or use the search check.
... or it's a triggered by the enemies trap, either out of or in combat.
... or time is an issue (water filling the room trap, boulder rolling trap, bypassed the trap and you're now running back down the passway away from a nasty dragon at high speed).

Etc.

[The_Jette]

Try leaving clues that the passive perception can notice, which will encourage the party trap finder to actively search for the concealed trap. Like, a corpse at the end of a dark hallway, or a fungi covered foot left from a gnome that left it behind when the trap got him. That'll leave room for high DC traps that can't be spotted and have to be searched for (i.e. Investigation instead of Perception).

[StoicLeaf]

My advice would be to:

A) don't fill the dungeon with traps. Depending on size, 3-5 should do. Treat a trap as a fight; it appears instead of a group of monsters. Unless you're all trap fetishists.

B) d&d is all about interaction and setting interactions up. Mix the traps up a little. Have a trap that reacts with the players' passive perception. The next trap can be seen by everyone (perhaps a spear didn't retract into the floor properly), but they'll need to deduce what sets the damn thing off. Next trap is similar but magical in nature; pass. per. won't help here, detect magic and/or arcana checks. Next trap can't be seen, nor can they deduce how it works, but they know it's there because of the horribly mutilated corpses.
Pick something a pc can do and build a trap's solution around it.

[ThurlRavenscrof]

Just roll an intelligence check (with proficiency) with advantage by the mind flayer for each trap. Then you'll have a variety of good rolls that are randomly assigned

[Finieous]

Good house rule for traps and passive Perception:

* Subtract 9 from the trap's DC to get a modifier (e.g. DC 20 would be +11).
* Roll 1d20+modifier against passive Perception.

Done.

[Naanomi]

Just be sure you are not punishing the party for one of their members taking an anti-trap feat

[Specter]

Good advice all around. Naturally I won't be making the player feel like he wasted a feat: even if the party doesn't specify looking for traps, his passive check is the same as it would be if he did. Apparently what I need to figure out now are clever traps, that after spotted need a check or clever interaction to bypass. Any tips?

[StoicLeaf]

Good advice all around. Naturally I won't be making the player feel like he wasted a feat: even if the party doesn't specify looking for traps, his passive check is the same as it would be if he did. Apparently what I need to figure out now are clever traps, that after spotted need a check or clever interaction to bypass. Any tips?

what does the group consist of?

[Specter]

what does the group consist of?

- Tempest Cleric
- Vengeance Paladin
- Hunter Ranger/Thief Rogue
- Abjurer Wizard
- Lore Bard

[Armored Walrus]

I can think of two, though not very complex ones.

A pit trap in a corridor, that is as wide as the corridor. They spot that the floor is false, but the pit beneath it is filled with acid, or rot grubs, or something else nasty. So they can't just climb in, cross the gap, and climb out. They need to figure a way past it.

A section of corridor where the ceiling is propped up, and a trip wire will cause it to collapse. Studying it, they realize the tripwire (or whatever trigger makes more sense) is placed in such a way that bypassing it will be very awkward (letting them know up front that the Dex DC to get by without caving in the tunnel will be very high), but just triggering it is not an option, because that will close off the corridor. I wouldn't even come up with a solution to this in my mind; I'd put it in front of them and see how they handle it. Just make sure you don't put anything that is plot-important behind it, or make sure there's another route to get there.

[Puh Laden]

Suggested Trap: An eye set up by the mind flayers to act as a security system. I'm picturing something like the malice eyes from Breath of the Wild. The first time they encounter an eye, it's set above a door in an otherwise empty room. The eye spots anything that is visible in the room. If it spots something that isn't a mindflayer or a thrall, it causes a portcullis to drop for the door. If the room is clear, the portcullis raises. They can try to shoot the eye, but doing so will alert a modified brain in the next room to release thralls for battle. If they use invisibility or camouflage or block the eye's visiion they can get through. A DC 20 (or 22 or 25 whatever) Strength check can be used to raise the portcullis but doing so also triggers the modified brain in the next room.

[Tetrasodium]

DM: Alice, you are able to feel a slight shimmer from a web of arcane forces up ahead.
Alice: wtf does that mean? so it's a trap?
DM:you aren't quite sure, but there is a lot of arcaneenergy there
Bob: well now that Alice has pointed out the arcane web up there, I want to roll arcana with my wizard to figure out what it does.
DM: that's a good roll. Near as you can tell, it has something to do with disintegration
Chuck: oh myyy... I want t o use my thieve's tools to carefully disarm this trap. Can I get advantage on it if Bob is helping me?
DM: sure
Chuck: 18!
DM: Everyone is standing there a few feet back * holding their breath while Chuck unwinds & threads a thin wire through the gaps in that arcane web & he nails it without blowing himself up or anything. What's everyone doing while Bob & Chuck are trying not to blow everyone up?
Chuck : 16?!
everyone else: we are .....
DM: So there you are back down the hallway with chuck & bob behind you when out of the darkness clops not one, but two intellect devourers at you. Roll initiative... Chuck give me another roll for those thieve's tools & sure at advantage as long as bob can beat a dc12 arcana check guiding you. but you are going on bob's turn because you are waiting for his guidance if you want that advantage.
Chuck I rolled a 7 total so yea, wating on bob
Intellect devourer:
it does stuffDave: oh crud.. I uhh... piss myself and uhh try to be useful?
Erica: I uhh..... do stuff in horror too?
Alice I cast guidance on uhh chuck:"Give it to bob! I have expertise & rolled a one, I need that advantage" I cast it on bob
Bob: 10 plus umm... 14! we get to live!
Intellect devourer2: other stuff
Chuck: maybe get to! 13!!! we live?
DM: There chuck is holding that wire with one hand while using the other to pour a few drops of nightwater down it to the binding point of the trap to disable it sending the arcane energies unraveling harmlessly as they are grounded by the nightwater.

TL;DRL don't just make a trap as an obstacle, break it down into multiple parts for the whole group to be involved in & add external pressure.

[Laserlight]

I have my players a trap that was a pile of pots, pans and sheet metal, with a tripwire. One of them deduced (correctly) that it was an alarm for the kobold pack. Another player then deliberately knocked over the pile....
tl;dr Sometimes you can show players a trap and it will still get them.

[thereaper]

Situations like these are why passive perception has always been a dumb idea. You're better off just making perception checks every single round, just in case there's something around and you happen to roll a 20. The only reason people don't do it is because it slows down the game too much.

The better solution is to just get rid of the concept altogether and just call for perception checks whenever there's something to spot, like the old days. Then, if you're worried about your players metagaming, also occasionally call for perception checks when there's nothing to spot at all. Problem solved.

[Finieous]

The better solution is to just get rid of the concept altogether and just call for perception checks whenever there's something to spot, like the old days. Then, if you're worried about your players metagaming, also occasionally call for perception checks when there's nothing to spot at all. Problem solved.

That's not the way it worked in the old days. It wasn't always (ever) clear how it did work, especially once the thief was introduced, but it definitely didn't involve "call[ing] for perception checks whenever there's something to spot."

[Zman]

Don't forget that if they aren't moving slow and taking their time that passive perception could easily be at disadvantage, i.e. -5.

[thereaper]

That's not the way it worked in the old days. It wasn't always (ever) clear how it did work, especially once the thief was introduced, but it definitely didn't involve "call[ing] for perception checks whenever there's something to spot."

Really? I apologize for my mistake, then. I stand by the rest of my post, however.

[Pex]

If your traps rely on not being spotted to work, they're effectively dead weight. Any character can take 10 times as long to search, and if it's possible to find the trap, they automatically find it. DMG automatic success rule. And (as you've noticed) the game assumes they front rank of the marching order is using their Passive Perception to look for threats. That means they're very likely to notice something is wrong, even moving along at a normal exploration speed.

This is intentional. 'Gotcha' traps are terrible trap design. You're supposed to design traps that require interesting decisions even after the group notices that they're there.

Edit: Since that's mostly moralizing, I'll give you some concrete examples:

After finding the trap, they need to figure out how to disable it. (This may be as simple as an investigation check, or as complicated as the players asking 20 questions.)
After figuring out how to disable it, they need to successfully do so. (This may be as simple as a Thieves Tools check, or as complicated as locating the right resource and creatively applying it.)
... or how to bypass it.
... or it's not just a trap, but if bypassed correctly a secret door to the next area.
... or it's an in-combat trap, when each character (not just the front rank) needs to pass passive perception or use the search check.
... or it's a triggered by the enemies trap, either out of or in combat.
... or time is an issue (water filling the room trap, boulder rolling trap, bypassed the trap and you're now running back down the passway away from a nasty dragon at high speed).

Etc.

This.

Even though the party knows the trap is there, they still have to deal with the fact the trap is there. Don't worry about the party finding the traps. Let Mr. Observant find all the traps. The player will feel special, and that's fine. The fun is in dealing with them. For extra intrigue, when confronting the mindflayers have them be fascinated about Mr. Observant. They know they were so clever with all their traps yet this creature saw through them all and found them. They would be every interested in knowing how.

[Tanarii]

Situations like these are why passive perception has always been a dumb idea. You're better off just making perception checks every single round, just in case there's something around and you happen to roll a 20. The only reason people don't do it is because it slows down the game too much.

The better solution is to just get rid of the concept altogether and just call for perception checks whenever there's something to spot, like the old days. Then, if you're worried about your players metagaming, also occasionally call for perception checks when there's nothing to spot at all. Problem solved.
Passive perception is a decent way to deal with the question of "the PCs are always searching every round over and over again, how do I avoid rolling every single round?" As well as "how do I not give away there is or is not a secret thing, when the mere act of rolling gives that away?"

The automatic success rule is a decent way to deal with the question "the PC keep searching until they find something, how do I avoid rolling until they find it?"

What's stupid is rolling over and over again unnecessarily, or arbitrarily deciding to only roll when something is there to be found, or rolling behind a screen at random points when there isn't something there to be found.

Another not-stupid alternative is just to eschew dice completely (which the passive perception and automatic success rules come very close to doing anyway), and give hints as to what is up and coming. This is what the oldest 'old days' way of doing it was.

Unfortunately too many players complained that hints weren't enough (ie 'gotcha traps'), so there needed to be some mechanical 'find stuff' roll built into the system. (Edit: and to be more fair, it's also nice to have it built in to the formalized resolution system anyway, since looking for danger/traps is one of the most common adventuring tasks.) Which brings us full circle around to: Make traps that have interesting player decisions involved. Gotcha traps are boring.

[Lombra]

The trick is to design traps that even when noticed still pose a threat. You may notice a tripwire but perception alone isn't going to get you through the trap: a combination of investigation, sleight of hand and perception is how every trap should be resolved, so that most of the party can partecipate to overcome the obstacle, and to avoid boring traps.

[Tanarii]

The trick is to design traps that even when noticed still pose a threat. You may notice a tripwire but perception alone isn't going to get you through the trap: a combination of investigation, sleight of hand and perception is how every trap should be resolved, so that most of the party can partecipate to overcome the obstacle, and to avoid boring traps.Or clever thinking that the DM rules is an automatic success at overcoming the obstacle, bypassing the need for any checks at all.

Not saying that should always be possible. But the 5e assumption is the DM considers each action the players take, and decide if it even needs a resolution check at all, based on what they're doing.

[coolAlias]

Or clever thinking that the DM rules is an automatic success at overcoming the obstacle, bypassing the need for any checks at all.

Not saying that should always be possible. But the 5e assumption is the DM considers each action the players take, and decide if it even needs a resolution check at all, based on what they're doing.
Used to be clever thinking was the default way of trying to get past a trap. The good old days. ;)

I am very much in favor of the auto-success rule. If you as DM can describe a trap in enough detail (pending favorable Perception/Investigation checks, if warranted) that the PCs can figure out how to disable or bypass it without further rolls, that is a successful trap, in my book.

Even better if they can figure out how to use it against their enemies. :D

[Joe the Rat]

Good house rule for traps and passive Perception:

* Subtract 9 from the trap's DC to get a modifier (e.g. DC 20 would be +11).
* Roll 1d20+modifier against passive Perception.

Done.That's what I'd recommend for variable detection.

And as others have noted, spotting a trap is not necessarily the same as avoiding a trap. You know it's there, but you still need to bypass. And know when it's really safe (contemplating the resetting boulder drop we failed to deal with last weekend).

[Tanarii]

Used to be clever thinking was the default way of trying to get past a trap. The good old days. ;)Yes I know. And then in 3e it went the opposite direction of everything has a target number and roll for everything.

Now we've got a happy medium: DM decides if a roll if the outcome is uncertain or not, and roll if it is.


I am very much in favor of the auto-success rule. If you as DM can describe a trap in enough detail (pending favorable Perception/Investigation checks, if warranted) that the PCs can figure out how to disable or bypass it without further rolls, that is a successful trap, in my book.The autosuccess rule isn't for certain outcomes. It's for uncertain ones where the only uncertainty is time. ie You need to roll if you're in a hurry, but nothing actually goes wrong if you fail, you just try again until you succeed. So if you've got time, don't roll in the first place, just take the time and succeed.

Certain outcomes (ie PCs figure out how to disable it without a roll) are certain outcomes.


Even better if they can figure out how to use it against their enemies. :DMy favorite is actually when the PCs bypass a trap, then retreat from something else later and have forgotten the trap is there, and are retreating at high speed not paying attention to their surroundings, meaning they can't use passive perception at all.

I do allow Int checks to recall the trap is there though. Alternatively, I could say it's passive perception at disadvantage-5 to spot it again. But I like to give Int some play, and this is precisely what Int checks are for, recalling something. (As opposed to, as most people like to use them, state of the world 'did I ever learn something' checks.)

[coolAlias]

The autosuccess rule isn't for certain outcomes. It's for uncertain ones where the only uncertainty is time. ie You need to roll if you're in a hurry, but nothing actually goes wrong if you fail, you just try again until you succeed. So if you've got time, don't roll in the first place, just take the time and succeed.

Certain outcomes (ie PCs figure out how to disable it without a roll) are certain outcomes.
That's what I meant by auto-success, as in "don't roll if the outcome is certain", but you are correct: in the rule books, that is under action adjudication and there is an actual 'auto-success' rule for time.

Bad choice of wording on my part, there.

[Zorku]

Situations like these are why passive perception has always been a dumb idea. You're better off just making perception checks every single round, just in case there's something around and you happen to roll a 20. The only reason people don't do it is because it slows down the game too much.

The better solution is to just get rid of the concept altogether and just call for perception checks whenever there's something to spot, like the old days. Then, if you're worried about your players metagaming, also occasionally call for perception checks when there's nothing to spot at all. Problem solved.

I recently read something from the angrydm blog that I like more: keep track of time in dangerous situations.

Exploring a room normally takes around 5-10 minutes, so are 1d6 to the time pool each time players enter a new room. Each time players want to perform a check of some sort add 1d6 (If anybody else wants to make a different check they can do so during this time.) If the party attracts attention to themselves roll the pool. If you roll a 1 on any of the dice there is a random encounter, or something about as bad happens. When there are six dice in the pool stop adding dice to the pool and start rolling it instead. When the pool is full and you roll a 1, clear the pool.

This represents 1 hour, but without sun dials how do you really know anyway?

-

Like this, passive perception is just what you do while you're making sure nothing in the room is going to spring an ambush on you later, and active perception is deciding you should look more thoroughly than that. There's some tension as you watch the time pool fill up, but if you know there's something important in here then go right ahead and keep searching until you find it... but there are consequences for dallying in hostile territory.

[CaptainSarathai]

Trap design for dummies (aka How Cap Does It, Because Cap is Dumb)

First:
(Wis) Perception: How it is
(Int) Insight: Why it is
Passive Perception - the minimum information you ever have, basically the lowest you can roll during a search
Rolled Perception - A 6 second look (action), to gather additional information. Also to contest Hide checks.
Take 10 Perception - they take 10 minutes, or 100 rounds, and find everything.

When they enter a room, I don't call for a search. I tell them everything they can find with a Passive Perception. If an effect beats their passive, then they'll need to either Search (fairly common) or they'll need to use Actions to look around the room (if we're rolling Initiative).
Once they spot something, I don't tell them that they've spotted a trap. I tell them that they spot a wire along the floor, or several small holes bored into the wall, or tiny brass grates in the ceiling, or a book that seems more worn that the others on the shelf. Usually that's enough for them to say,

"What the heck? What's that mean?"

And that's when I ask for Insight. Because Insight is them realizing that this wire is probably hooked up to a trap somewhere, or those holes likely fire darts, or the grates pump in gas, or the book is actually a secret lever.
Obviously, you can change the Intelligence skill being used, depending on what they've seen, or what they'd like to use. If it seems magical, or they think it might be, then they might ask to roll Arcana instead.

Second:
Stop letting everyone roll, any time.
Searching takes an Action if you just roll. That does not mean that everyone and their dad gets a roll! If they all want to roll for your same check, then all but one of them is just Assisting. If they all pass the assistance rolls, but their main spotter rolls a pair of 1's, then tough cookies; they find nothing.
If they're searching something small, like a doorway, then it's quite likely that only one of them can search at a time. So it takes even more time. This is important when the party is under stress - enemies attacking, etc. They can't just yell for a chance to see the shiny box. They have to actually move over to it and take a look, then move away and give someone else a try.

Third:
Better Hiding = Less Suck
Worse Hiding = More Suck
This is just a general rule. You don't want to give someone a DC25 Perception or die "gotcha trap." No, if there's some huge whirling blade of vorpal fire that's gonna come shooting out of the ground and instantly decapitate everyone within 2d100 feet, you put that sumb*itch in the open. The more deadly the trap, more likely it is to be an obstacle instead. They see it, it's obvious or very poorly hidden; now deal with it, because you can't get through it without taking mad damage or disabling it.

Gotcha Traps can work though. Stuff that requires a really thorough search to find, or is really hard to disable, can deal some chip damage or apply an annoying effect. The idea here is that D&D is a resource management game, and you're burning up a different kind of resource: hp/healing.
The dungeon is dangerous, and chip-traps will reinforce this. It doesn't even have to be a trap, it could be an environmental hazard that they don't notice; spiders hiding in the tube they just reached into to recover the scroll, or something.

Fourth:
Make smarter, more intuitive traps!
Don't just randomly trap every door and hallway in the dungeon. The party doesn't know that these traps exist, and so they start searching every door, and every hallway. It slows the game down, and it's stupid and frustrating because it's quite likely that only the guy with a wild high Perception is gonna be rolling.
Rather, establish early on that only likely areas will be trapped.
"Oh, a secret door! Hold on, maybe there's a trap?"
"Ooh, the boss's loot chest! Hang on, we don't have the key - it's probably trapped."

This goes hand in hand with the well-hidden traps idea. If the party should reasonably suspect a trap, you can expect them to search for one, and can therefore make that trap more dangerous.
If they hear a dragon waking from it's slumber deeper in the cave, and they're looking at a fancy treasure chest that they suspect might be trapped, then they're faced with a risk-reward scenario; Passive Perception says no trap, but there's likely a trap here. You could spend 10 minutes and know for sure, but maybe end up dragon chow, OR, you could have the keenest Legolas in your party waste an action to try and make a better spot, and if he doesn't find anything new, then you gotta just wrench that sucker open and hope for the best.

Fifth:
Roll your trap DCs instead!
If you don't like the idea of Passive Perception meaning that a Wis16 Cleric will find DC13 traps all of the time, then roll for your traps. Take the trap's usual DC, -10 and then add a d20. So the DC13 becomes d20+3, yielding a DC between 4 and 23.

DC20+ is insanely difficult. Better still is to use a d10, and add half the trap DC, rounding up. So dc13 becomes d10+7, for a total between 8 and 17, and an average of 12.5

I do this because presenting a DC14 trap to a party with 13 Passive Perception just feels really dirty, as a DM. At least now there's a chance they'll catch it on a passive.

Lastly:
"Click!"
I love this, because it can be downright hilarious.
If the party misses a trap trigger, because they didn't search, or whatever, give them one last chance to avoid their fate.
This is much better than just "reflex or suck", and it can actually save your Paladin's life (because their Dex sucks, get it?)
Let them know they've hit a trigger, and ask what they're doing. Remember, they only get a fraction of a second to do this, so it's as much about the players' quick thinking as it is their characters'.
Have these reactions then inform the saving throws of the trap. So for example, if a pit trap is opening up right under the Rogue and her response to "Click" is
"I leap to the side!"
Then she gets to make a Reflex to save. Maybe with advantage.
But if she is expecting a thrusting spear, she might instead say,
"Oh no! I dive to the floor!"
Well now she's just screwed. If she gets a save at all, it will be with disadvantage, but she might not even get that. The pit opens and she throws herself to a floor that's not there!

This is fun for two reasons. First, it gives players one last way to try and "outsmart" the trap, just like they could have if they had seen it coming. Secondly, it usually gives everyone a good laugh, because inevitably, someone always tries to do the worst thing possible in reaction.
------

Anyway, I hope this helps you out with trap and dungeon design.

[Vogonjeltz]

So basically, players are/will be in a dungeon filled with traps. Those traps are of the mind flayer variant, i.e. smart and very hidden traps.

While I'm creating these multiple traps, I'm assigning DCs to each of them. The only player I'm taking into account is the Cleric, who has 20 WIS, proficiency in Perception and the Observant feat (total passive of around 24).

But as I assign these DCs, I fall into a very simple problem: either he spots the trap, or he doesn't. Considering the former makes the trapfinding process dull and largely irrelevant, and the latter makes me a bad DM ("you couldn't ever see this trap, save for DEX"), how should I deal with this situation? Should some traps be passively untraceable?

It's a very rare trap that would be of the 25-30 difficulty range. Take a look at the example traps in the DMG and consider how you could justify a trap in that range and what it might be.

Tomb of Horrors has a plethora of examples of traps that, although fairly easy to spot, are extremely devious and dangerous even to a character that doesn't blunder into them. Consider using a couple of those for variety.

[thereaper]

Passive perception is a decent way to deal with the question of "the PCs are always searching every round over and over again, how do I avoid rolling every single round?" As well as "how do I not give away there is or is not a secret thing, when the mere act of rolling gives that away?"

The automatic success rule is a decent way to deal with the question "the PC keep searching until they find something, how do I avoid rolling until they find it?"

What's stupid is rolling over and over again unnecessarily, or arbitrarily deciding to only roll when something is there to be found, or rolling behind a screen at random points when there isn't something there to be found.

The point I was making is that it fails at that goal. Mathematically speaking, you are better off rolling every round, because passive perception might never see the hidden thing, while rolling might. The alternative I proposed, while not necessarily ideal, is better than passive perception for that reason alone.

[mephnick]

My main issue with traps is that they never seem to kill people.

All traps should be strong enough to like...one shot a troll or something. What good is a trap that can't kill anything remotely tough in a single hit?

[Thrudd]

I agree with everyone above who has suggested only describing what the character actually sees.

The passive perception is check determines what they see and hear. So only tell them exactly that - don't interpret it for them or tell them what it is or what it's purpose is.
They see a seam in the floor or the wall, they see a wire across the floor, they see strange markings in a pattern on the ceiling, they hear a ticking sound from behind the door, etc.

After that, they may do whatever they want - make investigations, use detect magic or arcana, poke at it, or just say "that's weird, I wonder what that is" and keep walking (that's when you get them!). high enough investigation rolls tell them what it is or how it works (if it isn't obvious), then it might require high dexterity rolls with the tools to disarm it. Maybe some can't be disarmed, only avoided.

That cleric is going to see everything, pretty much, unless it's an invisible trip wire that's also the width of a hair. So make the traps hard to disarm or require creativity to move past, because it is unlikely there will be any way to surprise them.

I suggest never rolling or asking for perception rolls. Possibly use the passive as a minimum for rolls if the characters are being specifically still and quiet and trying to listen carefully for noises (and even then always roll in secret). Use the passive scores just as a way to decide what you should tell your players. Perception should never tell the player the function or purpose of the thing they are seeing, just exactly what they see or hear.

[Tanarii]

My main issue with traps is that they never seem to kill people.PCs are basically super heroes past about level 2. Even then really nasty traps ca theoretically kill one.


All traps should be strong enough to like...one shot a troll or something. What good is a trap that can't kill anything remotely tough in a single hit?Trolls are basically super-villains. Or something. I'm not this post is coherent at this point ...

[Zorku]

My main issue with traps is that they never seem to kill people.

All traps should be strong enough to like...one shot a troll or something. What good is a trap that can't kill anything remotely tough in a single hit?
Generally if they're not death traps then they alert some other people to the presence of intruders. Tombs and things where nobody is supposed to go will have some kind of crappy means of killing people that enter anyway, but usually you're looking at a bandit fort or some wizard's tower, where you've got people (or constructs) that can come and scare off or murder whoever dared to sneak into the place.



Fifth:
Roll your trap DCs instead!
If you don't like the idea of Passive Perception meaning that a Wis16 Cleric will find DC13 traps all of the time, then roll for your traps. Take the trap's usual DC, -10 and then add a d20. So the DC13 becomes d20+3, yielding a DC between 4 and 23.

DC20+ is insanely difficult. Better still is to use a d10, and add half the trap DC, rounding up. So dc13 becomes d10+7, for a total between 8 and 17, and an average of 12.5Honestly I think I want these sorts of things on more of a bell curve.
I'll take the DC-10 but roll 3d6 instead. Pretty high chance of it being the same number it always was, but with some notable room for deviation. The range is 2 points smaller, but if I only roll a 19 on a d20 then they probably weren't going to spot that either.

[BeefGood]

Good house rule for traps and passive Perception:

* Subtract 9 from the trap's DC to get a modifier (e.g. DC 20 would be +11).
* Roll 1d20+modifier against passive Perception.

Done.

I understand the proposed mechanic but I guess I don't know what it "means". The trap is different each time?
Also, it seems to come full circle in that a "normal" check requires the character to roll against a fixed DC. But rolling is sometimes not convenient so then you do a characters fixed score (usually passive perception) against a fixed DC. But that's not always satisfactory so then you use a characters fixed score against a rolled/variable DC. Why not just stick with the original variety of rolled vs fixed?

[CaptainSarathai]

I understand the proposed mechanic but I guess I don't know what it "means". The trap is different each time?
Also, it seems to come full circle in that a "normal" check requires the character to roll against a fixed DC. But rolling is sometimes not convenient so then you do a characters fixed score (usually passive perception) against a fixed DC. But that's not always satisfactory so then you use a characters fixed score against a rolled/variable DC. Why not just stick with the original variety of rolled vs fixed?

The Spot DC is different each time, yes.
This does two things:

The first, and main reason to do this is to avoid exactly the situation the OP felt hesitant about; namely, characters will either see all DC__ traps, but no DC__+1 traps, so putting in a DC+1 trap feels like a "screw you" move to that character. Rolling for the trap means that it's a surprise to the DM (potentially) and also takes away the yucky feeling of willfully boosting a trap beyond their passive score.

Secondly, it can give the variance of a roll, without necessarily calling for the player to make a roll. You can, as a DM, generate random DCs for every trap in your dungeon using this method, and then not have to roll at all at the table (unless the player asks for a roll).
Asking the player to make a roll does slow the game down, but also tips off the player that something is out there, and it takes a very skilled player indeed, to not metagame this information.

[thereaper]

Rolling perception every turn would still be statistically slightly better than passive perception, though (average of 10.5+Perception vs 10+Perception).

It does make me think, though. Why not have the traps (or hidden things) roll, but get rid of perception rolls altogether?

In other words, the players are always using their passive perception, and whatever is hiding out there in the darkness rolls to see if it can beat their passive perception. It completely inverts the way it currently functions, but it would solve all the issues.

[Zorku]

The trap is different each time?
The Spot DC is different each time, yes.

so putting in a DC+1 trap feels like a "screw you" move to that character.
I was gonna chime in that you just roll during dungeon generation and stick with a DC once it is established, but I'm kind of put off by this response.

You can do things for that reason of keeping yourself surprised, but it seems like DM masturbation to me. I also want my dice rolls to be establishing permanent facts about the world, even trifling ones, so maybe I'm just biased for other reasons, but it seems like your job is to present a story. If you don't want to just screw the players with traps they never would have seen then don't screw the players with traps they never would have seen, but what you really need to do is not make them feel like you're screwing them with traps they never would have seen.

I'm decent at not taking offense at things the party does in the world, and just giving a straight faced presentation of how the world reacts to whatever they do, so I'm not especially worried about creating a string of traps designed to punish them without any current input on their parts. Not everyone's emotions work in quite the same way, but if you start placing traps that they couldn't have seen in areas that you specifically think will screw them over, how do you not realize you're doing that?

I do still like to randomize things with dice just so that I don't have to convince myself that I'm a better random number generator than I really am. As we've gone over, totally random traps aren't very believable, but I'll roll the dice to choose a location, and then ask myself if I can come up with a good reason for there to be a trap in that location, and then maybe I roll some more dice to determine the nature of the trap. That seems like random input a plenty, and even if I decide this one's got a DC17 check when I know that the party's best passive perception is a 16, that's fine. They could have a higher passive perception (or the ever important, advantage on the check,) and that's all they're gonna think was involved when the wall piston knocks a crumbling statue into one of them for 2d8 bludgeoning damage. They don't have any idea how long ago I decided what the DC on this trap was, and if I'm being a good narrator then none of them even considers that I was writing this last night.

[CaptainSarathai]

snip

Right, you arbitrarily decide to give it a DC17.
How is that much different than what I do? If you want to skip the roll, go ahead. When I roll, I use the normal
10, 15, 20, 25 DCs from the DMG and then to give a random variance, do the roll. So for your DC17, how is that different than me saying
15-10 = 5
roll 3d6 = 11
11+5 = dc16
??

Often, I don't roll my trap DCs. When in doubt, when something is a DC that they party could probably find, but I'm not 100% sure that I do/don't want them to, I'll roll it out. The option to roll is there, but it's not discussed in the DMG. It's just another tool for DMs to use. They don't have to if they'd rather have more control and set the DC themselves

[BeefGood]

Everything made sense to me except this first sentence:


The Spot DC is different each time, yes.
.

Are you saying that if the same trap is encountered more than once, it could have a different DC each time? That would be hard to put into story terms. It would be as if the trap makes an attack on each character, making a new attack roll each time.

Maybe if the traps mechanism is really complicated it would make sense that sometimes it "hits" and sometimes it "misses." Like in the old board game Mousetrap. Sometimes the contraption worked correctly and caught the mouse but sometimes it didn't.