D&D 5E 5E skills and the Perception vs Stealth imbalance

[CapnZapp]

Again how are you handling light and vision?

Sorry I should not have assumed it was obvious:

I am talking about dark conditions - caves and woods and whatnot. (Not major storms or blizzards)

(Rules that only allow parties to avoid being jumped in clearly lit calm situations isn't what I'm looking for. Sorry if that sounds snarky, I just have to say it outright)

As a player, when we're setting up camp and the others ask me to take a guard shift, I must ask myself why even post a guard, if all this can accomplish is me not being surprised.

What I want out of guard duty is a reasonable chance of seeing (hearing etc) shadows in the distance, with enough time to sneak over to each sleeping ally, wake them, allow them to get their bearings, draw weapons, or - if I have that info - quietly sneak off, effectively fleeing (leaving tents etc behind).

(This would take more than just a round or two to pull off. Still, I set my minimum threshold at two. This just illustrates that a good roll (actually a bad roll for them) should probably give me more time.)

With mediocre Wisdom and no proficiency, it's okay if I fail at this more often than not. But even with good Wisdom and proficiency, "sneaky gits" should have a chance of pulling off the ambush.

(It's one thing for a clumsy Ogre or a reckless band of Orcs or Bandits to fail. Another completely for wild animals built for the hunt, or more monstrous beasts that also "look" sneaky).

The difference between monsters are simply not wide enough if commoners have +0 and most animals and monsters have +3 or +5. That's only a difference of 15% to 25%.

Adding a racial modifier of +4 would go a long way of ensuring there is a real difference between "regular" monsters (such as untrained humanoids) and specialized hunters (such as the Owl, a wolf or big cat, or some stalking monster).

 

[eryndel]

The fact is, sneaking (Stealth) and detection (Perception) aren't born equal.

I want the following:
*) a group should not be able to rely on a single good Perception score
*) good Perception should chiefly benefit yourself and possibly a ward of yours, not an entire group
*) the rules need to favor ambushers much more, so that ambushes actually happen with some regularity.
*) the game is much more fun if dark forests and caverns actually ARE dark, foreboding and scary. Simply having a character with max Wisdom and Perception proficiency should not shortcircuit all that. Sure diligent defenders should be able to take precautions such as setting up perimeters (light torches that illuminate intruders etc), but that's an exception and not the rule.
*) monsters that are supposed to be sneaky should actually be sneaky, even level-appropriate heroes should have a hard time noticing these critters.

The basic notion of passive scores is good, since we can't have both Stealth and Perception rolled - it creates too large swings. So we'll keep that, but we'll considerably rein in what Perception can do for you.

I find this original post at odds with...

Can your players ever stop an ambush from happening?

That is, get a warning in sufficient time to regroup, flee, barricade themselves or otherwise averting the significant detrimental effects of the ambush? To start the encounter when the enemy is still not prepared, coordinated, ready?

And do all this by utilizing their watchfulness (i.e. Perception; not talking about having magical alarms go off etc). In real life, getting a 6 second warning is effectively useless. Ideally you'd want minutes to be able to prepare a meaningful response to a sudden attack. But I can live with unrealistic response and reaction times; it is after all a fantasy game where combats are over in seconds.

If we need a definition, I would put a minimal amount of time for an encounter to qualify for this at 1 complete round, but ideally you'd need two turns: one for everyone to wake up and get their bearings, the other to grab your critical gear (your weapon, a quest item or whatever) and move to a better position (less exposed, closer to your allies, away from any major threats etc).

That is, two (or one) complete turn(s) regardless of surprise or initiative. Obviously you can gain extra turns by winning initiative and/or not being surprised.

I am fully aware there really aren't any rules for how you would realistically lack awareness if you're suddenly being kicked awake in the middle of the night. Some DMs would ask players to spend their action on "active Perception" or they would have to take their action "blind" (as in before the DM reveals the nature of the threat).

If your answer isn't "yes", then my answer is "we're both playing 5E but in entirely different ways".

---

As I said, when I'm myself a player I abhor the notion that monsters just jump out of the shadows at my throat from nowhere. It might make for a good game, but to me it just feels so videogamey and artificial.

In the manner of handling Stealth v. Perception I described in my earlier post (along the same lines as Imaro), a prepared party can thwart ambushes with high situational awareness, preparation, or spells. This can be tough to do when caught completely unaware (which, in my mind, is as it should be). But, in general, it's only the high perception characters that are even remotely effective at preventing themselves from being surprised. By the time they can respond with a warning, we've already rolled initiative and evaluated who is surprised and who isn't. This seems to meet your first four requirements from the original post, although seems to run counter to the preferences you describe later.

 

[CapnZapp]

Yes, I'm talking about two things.

I dislike how guards can't actually do anything before the ambush is sprung (why post guards?).

I dislike how sneaky gits can't ambush the heroes.

The way to reconcile these two wishes must be for any Stealth rules to specifically address the outcome where the ambush is discovered "in time", and at the same time meaningfully distinguish between good and bad sneakers.

Since I completely agree all of this should be resolved by a single round of rolling, it's evident the current Stealth rules doesn't meet my expectations.

I need the Stealth rules to resolve to a set of specific outcomes:
a) the guard becomes aware of the approaching enemy well in advance. This may assume the guard is a roving ranger that's actively canvassing the local area for intruders (think Strider doubling back to camp to warn the Hobbits, rather than the unimaginative and stationary guard). Of course what the guard actually does is up to her - she might escape the area and leave her "allies" to the mercy of the attackers... or she might decide to pick off all attackers one by one by herself for what I know!
b) the guard becomes aware of the approaching enemy with at least a full round for everyone to spare
c) the guard only prevents himself to be surprised
d) the guard fails

Stealthy intruders should have a decent chance of achieving c or d.
Perceptive guards should have a decent chance of achieving a or b.

The solution is obvious. While the difference between a perceptive and non-perceptive guard is already sufficient (+0 to +10 merely taking ability + proficiency into account), the difference between a sneaky and non-sneaky monster certainly isn't.

Theoretically, monsters can too have +5 in ability and +5 in proficiency, but exceedingly few monsters have that (Invisible Stalkers and Drow Elite Warriors are two examples of Stealth +10; but in both cases I would assume this is achieved only by "Stealth Expertise").

The Stealth scores of MM statblocks doesn't explain how each bonus is calculated, but let's look at a few:

Bugbear Stealth +6. As far as I can see, this can only be achieved by giving the Bugbear "Stealth Expertise". Dexterity +2, Proficiency +2 and Proficiency again +2 =+6. While I appreciate the effort of expertise, the practical results aren't that distinctive with a proficiency bonus of only +2.

Shadow Demon +7. Even when you're an embodiment of "stealth" you don't get more than +7?!

Bullywug +3 but advantage. But why oh why qualify the "Swamp camouflage" ability to only hide checks and not move silently checks?

Owl +3

Wolf +4

I wouldn't hesitate for a second in giving all of these a +4 racial modifier each.

 

[Horwath]

The difference between monsters are simply not wide enough if commoners have +0 and most animals and monsters have +3 or +5. That's only a difference of 15% to 25%.

Adding a racial modifier of +4 would go a long way of ensuring there is a real difference between "regular" monsters (such as untrained humanoids) and specialized hunters (such as the Owl, a wolf or big cat, or some stalking monster).

Difference in +5 modifier is 25% only in absolute numbers, in relative terms the difference is quite bigger.

I.e. for passive perception DC of 10, a commoner has 55% chance for an ambush, a somewhat skilled character has +5, that is 80% chance. In relative terms he has 45% more chance to land an ambush.

Against DC 15, commoner has 30% chance to succeed, while skilled ambusher has 55% chance, that is 83% more likely to succeed.
adding -5 modifier for dim conditions gives even more advantage to the skilled character. And advantage for prepared ambush seal the deal for most ambushes.

I.E. passive perception DC 15, now it is 10 for dim light, ambusher is commoner with +0 bonus, but has advantage for preparing the ambush point(good terrain, higher ground, matching clothes, or whatever),
+0 vs DC 10 with advantage has 80% chance to pull it off. 80%! For someone with no training just with a good spot chosen.


Also you should roll one stealth roll for the NPCs and add their respective modifiers to each. That way you prevent the d20 to work against you.

PCs send their stealth guy alone usually so they avoid this problem.

 

[Imaro]

Sorry I should not have assumed it was obvious:

I am talking about dark conditions - caves and woods and whatnot. (Not major storms or blizzards)

Ok cool... so Darkvision, and lighting ranges (i.e. torches, lanterns, etc.) are in effect. Correct?

(Rules that only allow parties to avoid being jumped in clearly lit calm situations isn't what I'm looking for. Sorry if that sounds snarky, I just have to say it outright)

Not snarky but you do seem to be shifting your position somewhat. Originally you asserted ambushes and stealth attacks weren't effective enough... now you seem to be veering into the opposite direction.

As a player, when we're setting up camp and the others ask me to take a guard shift, I must ask myself why even post a guard, if all this can accomplish is me not being surprised.

Well if you patrol on guard duty far enough out from the camp... you have more than one round to act and could alert the party before they are ambushed, same thing with scouting ahead... of course as I said earlier, then you risk being cut off and attacked without the party. I am confused here though.. I thought the argument was that stealth and ambushing was too weak. Now you seem to be arguing that when the rules are followed it's too powerful... which is it?

What I want out of guard duty is a reasonable chance of seeing (hearing etc) shadows in the distance, with enough time to sneak over to each sleeping ally, wake them, allow them to get their bearings, draw weapons, or - if I have that info - quietly sneak off, effectively fleeing (leaving tents etc behind).

(This would take more than just a round or two to pull off. Still, I set my minimum threshold at two. This just illustrates that a good roll (actually a bad roll for them) should probably give me more time.)

So patrol further out from the camp... say 150ft The further from camp you set up your perimeter the more time you will give those back at camp, but again there is risk involved.

With mediocre Wisdom and no proficiency, it's okay if I fail at this more often than not. But even with good Wisdom and proficiency, "sneaky gits" should have a chance of pulling off the ambush.

I think they do especially if you use lighting and Darkvision rules... most will have a -5 to detect an ambush or stealthy creatures who stay at the perimeter of their vision.

(It's one thing for a clumsy Ogre or a reckless band of Orcs or Bandits to fail. Another completely for wild animals built for the hunt, or more monstrous beasts that also "look" sneaky).

The difference between monsters are simply not wide enough if commoners have +0 and most animals and monsters have +3 or +5. That's only a difference of 15% to 25%.

Why does the difference between monsters matter here? It's an opposed check against the characters... The difference that matters is between characters perception scores and creatures stealth scores.

Adding a racial modifier of +4 would go a long way of ensuring there is a real difference between "regular" monsters (such as untrained humanoids) and specialized hunters (such as the Owl, a wolf or big cat, or some stalking monster).

Again let's keep the vision and lighting rules in perspective here and use a wolf as our example.

Wolf CR 1/4 (+4 stealth) so let's say levels 1-3 are optimal for usage of this creature as an ambusher.
Let's use a max wisdom/perception character at 3rd level so +5 and a passive of 15. Let's give him Darkvision.

Now in Darkness his chances to spot the wolf before it is within closing distance for an attack are at disadvantage since you'd see as if it was dim light...
so passive now drops to 10 (roll of 7 or higher for the wolf to beat or 65%) or if actively searching for wolves it roughly negates your bonus... giving the Wolf a +20% chance of you failing to spot him... and this is best case scenario. I'm failing to see why he needs a +4 on top of that... auto succeeds against best perception/wis combo and now has a +40% chance to beat you if you are actively searching for him.

 

[Imaro]

Since I completely agree all of this should be resolved by a single round of rolling, it's evident the current Stealth rules doesn't meet my expectations.

I need the Stealth rules to resolve to a set of specific outcomes:
a) the guard becomes aware of the approaching enemy well in advance. This may assume the guard is a roving ranger that's actively canvassing the local area for intruders (think Strider doubling back to camp to warn the Hobbits, rather than the unimaginative and stationary guard). Of course what the guard actually does is up to her - she might escape the area and leave her "allies" to the mercy of the attackers... or she might decide to pick off all attackers one by one by herself for what I know!
b) the guard becomes aware of the approaching enemy with at least a full round for everyone to spare
c) the guard only prevents himself to be surprised
d) the guard fails

Stealthy intruders should have a decent chance of achieving c or d.
Perceptive guards should have a decent chance of achieving a or b.

I'll say it again 5e's rules account for everything you want... See my previous post.

 

[Tony Vargas]

Well, obviously I'm one of those DMs that aren't content with having rules you need to ignore for things to make sense

Doesn't mean it's a bad approach. What it means you're way less sensitive to "sloppy" rules than I.

OH, I sensed 'em.

One suggestion has been "have ambushers make a group check". This might not be my personal favorite solution, but it does address the math. By allowing the group to succeed even if 2 out of 5 fail, you have shown a basic awareness of the statistics.

IMHO, between avoiding contested checks using 'passive' calculations on one side as DCs on the other, and using group checks, you can get a less swingy play experience out of D&D. 'Passive' has been a thing since 3e, and 4e introduced group checks - 5e has done away with neither, so the 'solution' is already there. It's just up to the DM to call for that type of check instead of individual ones when it'd do the most good. And that, too, is 'RAW' in the sense that it is the DM's role, by the book, to set DCs and call for checks.

Can your players ever stop an ambush from happening?

Sure. They can stay home.

I am fully aware there really aren't any rules for how you would realistically lack awareness if you're suddenly being kicked awake in the middle of the night.

The more 'realistic' the rule, the less fun it brings to the game, and the worse a job it does at modeling the fantasy genre.

RL is all the realistism it's fair to ask us to handle, and then some.

What does this tell you about the rules as written?

"Use at own risk?"

 

[ThePolarBear]

Yes, I'm talking about two things.

I dislike how guards can't actually do anything before the ambush is sprung (why post guards?).

Oh, i see... you do not mean an ambush but a night raid! Ok, now i understand a little bit better some comments you made. Btw, the only meaning i know for "ambush" is "waiting someone while hiding to attack them by surprise", and you instead were indicating the opposite.

Bullywug +3 but advantage. But why oh why qualify the "Swamp camouflage" ability to only hide checks and not move silently checks?

There are no move silently checks that are not checks for hiding. Moving silently uses the same rules as hiding. Unless an ability requires you to be stationary, you can assume that hiding also means "moving trying to not make sound".

By the way, you could use, as someone else suggested, the degrees of failure version of "failing" ability rolls. That way you reach a middle ground most of the times (failing a roll on 2/3 points) even on high end (well, not "optimized high end") perception scores if you combine advantage and disadvantage according to situations.

And you should really add some personal touch to monsters if your players go for the optimized route. Otherwise you should be fine for most of the raids you can think of i think.

...so passive now drops to 10 (roll of 7 or higher for the wolf to beat or 65%)...

It's actually 6+. Skill contests on a tie resolve with the previous situation being unchanged. It means in this case that the guard remains unaware of the wolves approaching.

 

[Imaro]

It's actually 6+. Skill contests on a tie resolve with the previous situation being unchanged. It means in this case that the guard remains unaware of the wolves approaching.

Yep you are correct. That brings it up to 70%. Thanks.

 

[DEFCON 1]

What does this tell you about the rules as written?

That they were written a certain way for a particular reason.

But what's your point? They aren't getting changed. They aren't getting errata'd. The rules are the rules. We've known and been told this since the game was first released. So all you are doing is looking to create your own rulings on how they work... which is exactly what we've already seen over and over and over and over and over and over and over here on ENWorld.

If you haven't come up with your own ruling yet, I don't know what to tell you.