D&D 5E Stealth Checks - How do you handle them?

[Saeviomagy]

Definitely. This is one reason why turn-based* exploration rules are a good idea, because you handle the whole turn with one check.

Turn based checks effectively mean that you will fail your infiltration unless your stealth score is so good that you don't actually need to roll. Do you have a 75% chance of remaining undetected for one roll? Well after 5 rounds, you've got a 23% chance of being undetected.

Now you might be effectively doing what I suggested: treating a failed stealth roll as a step down a track towards detection - a guard comes to investigate a funny noise - rather than full blown detection itself, which allows chances to correct for a failure.

Really passive checks should be considered separately depending on whether they are against fixed DC or an opposed roll. Against opposed roll they are actually fine (as long as the opponent is rolling normally), because the essentially change the skill contest into a single check vs DC, which is fine (the two work basically the same, unless you are using different degrees of success, since an opposed check has double swingy-ness). Against fixed DC they totally remove randomness, but if you remove randomness then why are you even wasting time thinking about what kind of check you should be using? At best you can use passive scores to "gauge" what kind of challenges you should handwave and declare an autosuccess, but then for consistency you should allow autosuccess on the same things also when the player actively tries to do them (which most DMs don't).

Agreed. Passive checks in this edition are so fraught with inconsistency they basically have no value.

Knowing where you are is not an absolute condition, everyone who gets lost believes they are headed in the right direction until they are in fact lost.

I think you'll find that absolute impossible to back up, and I have personal experience to the contrary. Any time I've gotten lost, I've been quite aware ahead of time that I'm not sure the path I have chosen is correct.

How would you know if you are spotted if you can't see the spotter or if the spotter sees you and acts as if he doesn't? You wouldn't. The perception check is to see if you think you have given up the game, either through your own knowledge or the reaction of other.

There are certain extremes which would make you aware, and certain levels of screw up where you would find a deception unbelievable and be likely to pre-emptively take countermeasures. For instance: you are walking across an open field, approaching a target from the rear. You step on a twig, which makes a large cracking sound when you are about 10ft from the target.

I imagine that an amateur at stealth would freeze and watch their target for signs they heard, then be flabbergasted when they didn't actually get a surprise round.

I imagine a pro would immediately dive 10ft forward and engage, simply assuming they had been heard.

I can't imagine a pro not noticing that they screwed up.

People who train skills believe in them, whether that belief is well founded is an awareness question.

Extensive research proves the opposite. Incompetent people are more likely to overestimate their ability: they simply don't know what they don't know. Competent tend to underestimate their own ability.

https://en.wikipedia.org/wiki/Dunning–Kruger_effect

 

[Li Shenron]

Turn based checks effectively mean that you will fail your infiltration unless your stealth score is so good that you don't actually need to roll. Do you have a 75% chance of remaining undetected for one roll? Well after 5 rounds, you've got a 23% chance of being undetected.

But no, that's not really how it works... The danger can still be just concentrated in one place, even if you are rolling multiple times, once per turn.

For example, you can decide to allow one check per turn to find traps, and to use 5-minutes turns (e.g. one turn per room or corridor). This doesn't mean that there is actually one trap to find every room. There can be anything, from zero traps at all, to one trap in the whole dungeon, to as many as you like.

Same with stealth, you can use 1-hour turns to travel through a forest and allow a Stealth check per turn, and maybe decide to use a hex map so that each turn means to move to the next hex. It doesn't mean that there is someone in each hex who will attack you if you fail the Stealth check.

And by the way, if you do put one trap every room or one guarding party per hex, then the traditional method of rolling once per danger will have pretty much the same low chances of "failing at least once" that you are attributing to turn-based checks.

 

[smbakeresq]

Turn based checks effectively mean that you will fail your infiltration unless your stealth score is so good that you don't actually need to roll. Do you have a 75% chance of remaining undetected for one roll? Well after 5 rounds, you've got a 23% chance of being undetected.

Now you might be effectively doing what I suggested: treating a failed stealth roll as a step down a track towards detection - a guard comes to investigate a funny noise - rather than full blown detection itself, which allows chances to correct for a failure.


Agreed. Passive checks in this edition are so fraught with inconsistency they basically have no value.


I think you'll find that absolute impossible to back up, and I have personal experience to the contrary. Any time I've gotten lost, I've been quite aware ahead of time that I'm not sure the path I have chosen is correct.

There are certain extremes which would make you aware, and certain levels of screw up where you would find a deception unbelievable and be likely to pre-emptively take countermeasures. For instance: you are walking across an open field, approaching a target from the rear. You step on a twig, which makes a large cracking sound when you are about 10ft from the target.

I imagine that an amateur at stealth would freeze and watch their target for signs they heard, then be flabbergasted when they didn't actually get a surprise round.

I imagine a pro would immediately dive 10ft forward and engage, simply assuming they had been heard.

I can't imagine a pro not noticing that they screwed up.

Extensive research proves the opposite. Incompetent people are more likely to overestimate their ability: they simply don't know what they don't know. Competent tend to underestimate their own ability.

https://en.wikipedia.org/wiki/Dunning–Kruger_effect

Uhh that's exactly what I just said when I said this - "People who train skills believe in them, whether that belief is well founded is an awareness question." Its a confidence issue.

As far as the rest of your comment, you cant let a PC know that he failed without him trying to discover that he failed because then he and the group will change their actions based upon a meta-game event, the PC rolling a die right in open on the table. You cited Dunning- Kruger effect, that a person perception of their skill. To me that means at the table if you are trying something and are not bothering to interact with the DM to see if you failed outside of asking "did I make the roll" then you wouldn't know whether you failed or not. As a side note no DM should answer the question "Did I make the roll" with a yes or no.

Just because you are trained in a skill doesn't make you an expert. How and when are you an expert? At +3 to that skill? +10? I don't know, I don't think anyone would either. Many believe they are experts only to find out later they are not, and many experts in things fail at their expertise, they roll a 1 IRL.

The key in this is to get the players to interact and maybe use their skills to determine if they succeeded or failed, and if they failed a chance to get out it.



Really this discussion is essentially boiling down to argument over what to do when a creature or PC "cant be surprised." While I think that putting that into the game (and into the Alert feat) is probably bad for the game, you have to roll with it work with it at the table somehow to make sure players don't feel screwed over when they do get surprised by something. Stealth checks are the same way, players invest a lot into stealth and try to make it work, so if they plan it well and just get screwed on a check you have them a chance to get out of it if it is well played. Likewise if player has a high score and abuses it you have to take of that situation also.


I also use passive skills for almost all the skills, where appropriate, to avoid the problem of eventually failing a roll. As [MENTION=1465]Li Shenron[/MENTION] describes in this thread using stealth for overland trips, if a player has a high enough stealth score he can move through a forest without being spotted for long periods of time without having to make many checks.

Read more: http://www.enworld.org/forum/showth...ks-How-do-you-handle-them/page6#ixzz4a5ZLlmaV

 

[Ristamar]

Same with stealth, you can use 1-hour turns to travel through a forest and allow a Stealth check per turn, and maybe decide to use a hex map so that each turn means to move to the next hex. It doesn't mean that there is someone in each hex who will attack you if you fail the Stealth check.

I wouldn't call for a new stealth check unless the circumstances changed enough to warrant it. An arbitrary delineation, like a hex line or block of time or moving from one room to another, shouldn't be the sole impetus for another check, IMO.

 

[Li Shenron]

I wouldn't call for a new stealth check unless the circumstances changed enough to warrant it. An arbitrary delineation, like a hex line or block of time or moving from one room to another, shouldn't be the sole impetus for another check, IMO.

You are losing track of the whole point of turn-based checks.

 

[Ristamar]

You are losing track of the whole point of turn-based checks.

Fair enough. I guess my tastes are just more aligned with iserith and Bawylie. I don't have an aversion to passive scores. I don't see the need for extraneous rolling, nor do I mind making a secret roll if there is a hidden threat that could observe the sneaking PC party.

That being said, I don't think I'd ever roll Perception versus a passive Stealth check. While it wouldn't break the game, the existing 5E game elements seem to lean toward passive checks on the observer's side of a contest. It keeps things cleaner if it's used consistently, but it does require a certain level of trust between the players and the DM.

 

[Valmarius]

In a lot of games the PC will roll a 1 and know he failed and change his action.

I think this illustrates the main difference between our approaches. I only ask for checks once someone has committed to a course of action, then that roll determines the outcome. A PC cannot say, "I'm going to sneak across the courtyard, using the pillars for cover so the guard doesn't see me." Roll a 1, then say "actually I'll stay where I am."
It would be more like,
PC: "I'm going to sneak across the courtyard, using the pillars for cover so the guard doesn't see me."
DM: "Alright, its dim light and since you're sticking to cover the DC will be 14. Make a stealth check."
Rolls a 1
DM: "Unlucky, the guard happens to turn around just as you're moving between cover. He spots you and reaches for his sword..."

That said, it would be perfectly fine in my game for a PC to state they wanted to assess the guards skills before making their attempt to sneak past. A perception or insight check here might give them advantage on the following stealth roll or even reveal the guards passive perception score.

 

[Saeviomagy]

But no, that's not really how it works... The danger can still be just concentrated in one place, even if you are rolling multiple times, once per turn.

For example, you can decide to allow one check per turn to find traps, and to use 5-minutes turns (e.g. one turn per room or corridor). This doesn't mean that there is actually one trap to find every room. There can be anything, from zero traps at all, to one trap in the whole dungeon, to as many as you like.

In which case you're doing a lot of pointless rolling. If there's one thing to spot, you can make one roll to spot it, no matter how many turns there are. If you're really heavy into obfuscating things, then make one roll each time the players go anywhere.

The only scenario you have to avoid is splitting things up into small chunks and then only rolling for a chunk when there is something to find.

Same with stealth, you can use 1-hour turns to travel through a forest and allow a Stealth check per turn, and maybe decide to use a hex map so that each turn means to move to the next hex. It doesn't mean that there is someone in each hex who will attack you if you fail the Stealth check.

Right, this is a different scenario to what I thought you meant: you aren't modelling a single encounter here, you've got lots of disjointed encounters with nothingness in between. There is no single roll failure, so feel free to do it this way. It just seems like there's going to be a lot of pointless rolling going on.

And by the way, if you do put one trap every room or one guarding party per hex, then the traditional method of rolling once per danger will have pretty much the same low chances of "failing at least once" that you are attributing to turn-based checks.

Which is irrelevant in the situations you're talking about for the reasons I expressed above. You're describing independent events here.

Uhh that's exactly what I just said when I said this - "People who train skills believe in them, whether that belief is well founded is an awareness question." Its a confidence issue.

No. People who don't train skills believe that they are good at things. People who actually receive training have a much more accurate picture of their own ability. That's the entire point. It's not confidence, it's competence.

As for how much training is enough? Based off the rules for training tool proficiencies, someone who is proficient has 250 days of exclusive 8-hour a day training with a competent teacher and a small class size (unless teachers are freaking rich). It's apparently enough training to learn a whole language and not need to roll to understand things or express yourself.

As far as the rest of your comment, you cant let a PC know that he failed without him trying to discover that he failed because then he and the group will change their actions based upon a meta-game event, the PC rolling a die right in open on the table.

There's no roll without the action and no action without the roll. You can't roll and then do something else: the roll was you doing the thing that you were rolling for. There isn't any metagaming happening: the roll dictates in-game events.

You can't say "I sneak past the guard", roll stealth, get a 1 and then say "oh, actually I stay here in cover". That would be a problematic scenario.

You CAN say "I sneak past the guard", get a 1, then jump the guard. Your stealth roll still failed - your character still tried to sneak, stepped on a passing squirrel and alerted the guard (or maybe not! the guard might have a really terrible perception check!), he just didn't wait for the DM to describe the guard drawing his sword and charging. Personally I'd be willing to give a benefit for that (not full surprise, but maybe advantage on initiative): after all, you're losing out on the chance for the guard to screw up and your stealth to be ok.

You cited Dunning- Kruger effect, that a person perception of their skill. To me that means at the table if you are trying something and are not bothering to interact with the DM to see if you failed outside of asking "did I make the roll" then you wouldn't know whether you failed or not. As a side note no DM should answer the question "Did I make the roll" with a yes or no.

In games I play, the DM asks you to make a roll, you roll, then the DM describes the outcome. He never answers the question "did I succeed at the roll?" except by describing what happens.

Now sometimes the actual pass-fail of a roll will be unclear because the fallout from the roll is not immediate. But I think the player knowing his roll and being able to plan future actions based on that is not a bad thing: there's still uncertainty there after all, since he doesn't know the DC.

 

[iserith]

Assuming they don't roll the second I say "Let's see a Stealth check..." (which happens sometimes), I do go on to say what the DC is and what happens if the PC succeeds or fails before the roll is made. I want the player aware of what the stakes are and to make sure we're on the same page with regard to the difficulty of the achieving the goal given the approach and the circumstances. It's also an opportunity for the player to consider whether it's a good time to spend Inspiration or other resources to improve the odds of success. Finally, I find knowing the chances of success actually increases tension at the table when the odds are not in the PC's favor as opposed to not knowing.

 

[Li Shenron]

In which case you're doing a lot of pointless rolling.

Depends what you think is pointless

One reason for this variant rules is to handle paranoid players who check for traps or hidden things every step. By allowing a "blanket check " every turn, you can avoid even more pointless rolls. There are other ways to handle the same, but I like this because it doesn't work as a punishment.

Compared to checking only when it matters, I like thatit doesn't give away hints that it actually matters.

I only don't think I would use it for a stocked dungeon, where normally the fun is for the players to guess where or what to check. It works best for repetitive or featureless environments such as caves, forests and long term travels.