How the crap are we supposed to group-stealth!?

[ReaderAt2046]

Well, the AngryGM proposed what I think is a very good way of handling group checks. This one's specifically for D&D 5E, but the basic idea can be adapted.

If only one member of the group needs to succeed for the group to succeed (such as with searching a room), then whoever has the highest proficiency and ability bonus rolls. If any member of the group has advantage, the check is made with advantage, and if any member of the group is benefitting from any bonuses that would apply to the check, the largest single bonus applies.

If only one member of the group needs to fail for the group to fail (as with sneaking past a guard), then whoever has the lowest proficiency and ability bonus rolls. If any member of the group has disadvantage, the check is made with disadvantage, and if any member of the group is suffering from any penalties that would apply to the check, the largest single penalty applies.

[jayem]

It's worth pointing out that most stealth jobs should not be done as a group. You can't pick a pocket as a group project. And it doesn't take thre people to sneak into the building and unlock the back door. Sneaking three people in individually should be harder than sneaking in just one.

And the things you do as a group, you should do intelligently, as a group, with mutual support.

One moves in a little bit, and finds a good spot to be a lookout. He signals to the next one that she can go further, now that somebody is watching her back. Then she looks out for the third one.

This way, they are each aiding the others' rolls.

Yes it becomes a lot harder to abstract the different components.
Technically it's probably a bit dodgy to abstract all the aspects involved in the marks vulnability into a DC, and all the different elements in the pickpockets/stealths into one roll.
But when you come to a group pickpocket or group stealth then all these parts interact much more. The distraction of the mark is ongoing and by a different person. The dummy running away... If the mark is suspicious, but not in the right way, is that a help or hindrance?
Similarly in solo-stealth, when it fails you have mook (aware of you) and you (by definition in a line with mook). If he's 50% sure you're stuck there while he investigates (while if he investigates wrong you get away, or repeat till one outcome occurs). In group-stealth none of that is true.

To some extent if it becomes officially key part of the game then if it is complex and drawn out (so sneaking past a mook is about the same as fighting one, you have the sneak map out etc...), I don't see that as bad. The whole group is after all there and involved.
And if it isn't a key part then if the approximation is terrible then there's little problem.

It's just that annoying bit where things are in the middle there's nothing interesting in the action really, but the overall outcome matters (though by that token it's only fair to demand things to be comparable for combat and battle-combat).

[Guizonde]

uh, my group kind of had that problem... dragon cleric, orc paladin, oracle, loud bard... then i came across this teamwork feat. it seems broken, but hey, at least i can sneak 800lbs of righteous violence anywhere now. gotta love inquisitors, sometimes...

https://www.d20pfsrd.com/feats/gener...ergy-teamwork/

now, unless i misunderstand it, it means that we all roll, take the best roll, and add all the modifiers, positive and negative, so long as the team has that same feat (or in the case of inquisitors, they all qualify for it).

going by this, wouldn't an easy solution be "all the team rolls, adds all the modifiers of the team, and take the worst roll"? you add the modifiers due to the team helping the least stealthy, and the least stealthy represents the lowest roll? like that, it gives a good chance of success due to cooperation, but still represents the fact that not all are good at being discreet?

[Tanarii]

IMO really boils down to what the GM is using sneaky-stuff checks for.

If it's something like AD&D's Hide in Shadows, where it's the ability to remain completely unnoticed, even while being directly looked at, effective invisibility, then it should probably be very hard. If it's sneaking right past an alert guard, again effective invisibility but this time with movement, it should also be very hard.

If it's generally being sneaky but not doing something almost magical, there's no reason it can't be the group's pointman (different from a forward scout) making most necessary stealth checks and signaling the others when it's clear and when they should hang back. With some group-style checks in the more tense moments, as they all have to scatter for cover as a patrol gets a little too close, with the stealthier members helping the less stealthy. (Keeping in mind of course that too many checks for one mission will inevitably lead to failed checks.)

[D+1]

For the sake of argument let’s say that everybody in the group is average at sneaking, and can succeed in going undetected 50% of the time. If any one of these characters attempts to creep through a sleeping dragon’s lair, they’ll get away with the Bilbo impersonation half the time. If all four of them attempt the trick, however, they’ll only avoid detection in 1 out of every 16 attempts. Even if a party designed for stealth tries to burgle Big Red, with each master ninja enjoying a full 90% success rate, at least one of them will biff the roll about 35% of the time.

First question to mind is why should this be any different when ANY two PC's are trying to do the same skill-use related job normally handed to just one? Either chance of overall failure rises (as logically it should) or the PC's coordinate their efforts and STOP trying to do everything as individuals instead of a team. A lot also depends on the rule set/edition you're using. The solution can and probably should be different for any different game.

All of the above makes a certain amount of sense from a simulation standpoint. It’s harder to get a squad of dudes behind enemy lines than a single operative. But from a narrative standpoint, I’ve always felt that this sort of thing bites. Why the crap can’t we sneak into one freaking cultist den without setting off alarm bells?

Well, which do you want your game to be? A simulation or a narrative game? If you're only looking at mechanical solutions to your problem then clearly you consider this a simulation issue. The DM in the situation can go a long way to solving the problem just by being aware that there IS a problem. Don't put PC's into situations where the multiple stealthers need to all make the same rolls, need to make repeated rolls, are duplicating their efforts in general, are facing low chances of success for those rolls, and where they cannot coordinate efforts or divide tasks to minimize rolls.

This is a problem I became aware of playing 1E D&D with just ONE stealth character in a party. Their chances for success at silent movement, finding traps, removing them, climbing a wall... ALL of the standard things they were expected to be doing on a constant basis had a low chance of success in the first place, but then DM's would, without thinking about it, call for constant, repeated checks for all their skills, not realizing they were only ensuring failure rather than allowing a PC to ever actually demonstrate any competence whatever at any of them. The solution I applied is to increase their base chances of success to at least about 50/50, to simply allow them to succeed automatically if circumstances already favored them, to ensure that one roll would indicate success/failure over as much of a physical area or span of time as possible to avoid the phenomenon of "seeking failure" rather than "seeking success", and to ensure that in as many circumstances as possible even if the stealthers failed that success could still be achieved - that if a trap is found but can't be removed it can still be prevented from operating, or its effects nonetheless avoided; that if a character can't accomplish something by stealth that it can still be accomplished by some other swift, coordinated actions (can't sneak up and take out guards one at a time from behind, but one can be distracted or lured to where they can be taken out and then the remainder shot from a distance or rushed.)

Do any of you guys have a good system or house rule to fix this one? Because this issue has honestly stumped me for years.

If you had intended to create an ALL stealth-character party then the game would be run very differently. Why should the game not adapt at least partly to a party that is heavily constituted of sneaks? Rather than adjust the game to better suit the party, the party - despite its character composition - would seem to be expected to adapt to an inflexible game. Those PC's should not have to compete with each other for a single spotlight, but should be trying to work more as a stealth team. If not given more opportunity to operate as a team they should at least not find themselves required to step on each other for time in the TOO-SMALL spotlight for their class' niche.

[Quertus]

Well, the AngryGM proposed what I think is a very good way of handling group checks. This one's specifically for D&D 5E, but the basic idea can be adapted.

If only one member of the group needs to succeed for the group to succeed (such as with searching a room), then whoever has the highest proficiency and ability bonus rolls. If any member of the group has advantage, the check is made with advantage, and if any member of the group is benefitting from any bonuses that would apply to the check, the largest single bonus applies.

If only one member of the group needs to fail for the group to fail (as with sneaking past a guard), then whoever has the lowest proficiency and ability bonus rolls. If any member of the group has disadvantage, the check is made with disadvantage, and if any member of the group is suffering from any penalties that would apply to the check, the largest single penalty applies.

Angry's method is nice by dent of being simple, and making everyone's the biggest contributors' contributions matter. It's fine if the point of the roll is to feel like there are somewhat believable rules in place, let's hurry up and make a roll so that we can get back to the good stuff. Beyond that, it's ****. It's a less accurate simulation than HP, and it's neither an engaging minigame not narratively interesting.

In other words, I fully support using it.

First question to mind is why should this be any different when ANY two PC's are trying to do the same skill-use related job normally handed to just one? Either chance of overall failure rises (as logically it should) or the PC's coordinate their efforts and STOP trying to do everything as individuals instead of a team.

Now, that's an interesting sentiment. What would you suggest?

A lot also depends on the rule set/edition you're using. The solution can and probably should be different for any different game.

?

I'd imagine that, if there were some logic as to how the simulation should run, that it would hold true for all simulation-based games.

But, yes, one certainly wouldn't expect that logic to work in non-simulation-based games.

This is a problem I became aware of playing 1E D&D with just ONE stealth character in a party. Their chances for success at silent movement, finding traps, removing them, climbing a wall... ALL of the standard things they were expected to be doing on a constant basis had a low chance of success in the first place, but then DM's would, without thinking about it, call for constant, repeated checks for all their skills, not realizing they were only ensuring failure rather than allowing a PC to ever actually demonstrate any competence whatever at any of them.

Whenever I see this sentiment, I've just got to add my standard comment that I had no issue with this behavior. When you try to sneak past 50 patrols, it's not so much a question of if, but when you'll be noticed. It involves using player skills, and makes for a much more engaging minigame than just making rolls. It makes the rolls quite tense, and makes for a good risk / reward minigame.

As to the demonstrating competence bit... Hmmm... that's certainly something that's important to me. I am strongly opposed to systems (and GMs!) that enforce pants on head stupidity. But that's not the vibe I got from, well, 2e, at least. By 5th level, a thief could have - even without bonuses from dex / race / items / lack of armor - a 90% chance of success on their unopposed roll. By happy coincidence, a 5th level 3e rogue with no bonuses besides max ranks who takes a 10 would have an 18 on their opposed roll, which is a 90% chance of success against an average peasant unskilled observer.

That doesn't feel "unskilled" / completely lacking competence to me.

[Jay R]

It's worth pointing out that three warriors in the same party don't try to swing the same sword.

Similarly, three rogues should have specific, and separate, goals. One has snuck into camp to listen to the general's plans, one is investigating the perimeter, looking for a way for the party to attack unexpectedly, while the third is supposed to be sabotaging the enemy's equipment, but is really stealing their magic items.

[Tanarii]

This is a problem I became aware of playing 1E D&D with just ONE stealth character in a party. Their chances for success at silent movement, finding traps, removing them, climbing a wall... ALL of the standard things they were expected to be doing on a constant basis had a low chance of success in the first place, but then DM's would, without thinking about it, call for constant, repeated checks for all their skills, not realizing they were only ensuring failure rather than allowing a PC to ever actually demonstrate any competence whatever at any of them.

Yeah, many DMs had no idea how to use Thief skills properly. They were for doing near magical things, or short cutting the system. Like Hiding invisibly in the Shadows or Moving totally Silently. Not hiding behind something or moving quietly. For Climbing Sheer Surfaces, not for climbing something skilled climber with some modern climbing shoes could climb. For finding and removing traps with a simple % roll, not having to use your brains to carefully investigate and figure out where they were and how to disable them.