How to handle group insight checks

[mehs]

So doing Hoard of the dragon queen with a small (3) party. Dm leveled us to three by about 3/4s way through the first episode to allow for it, etc etc context.

But the main thing, you are supposed to be sneaking around, ambushing, conserving resources as you are pretty much doing an entire night of combat with no rest, fighting dozens of people.

So that leads to the main thing, how to do deception on a group of enemies? they get to roll insight, but does each get to roll? Wouldn't that be quinquadruple advantage?

[OldTrees1]

You could use passive insight?
Or maybe the group has one person who's insight they trust (1 active check used for everyone).
Maybe a group might have advantage (+5 to passive if you are using passive).

However do some quick math to see if the chance of success/failure you are setting is close to what you think is reasonable. You don't want to have a case where you thought the PC has a 80% chance only to find out they had a 20% chance due to your choice of ruling.

[Vegan Squirrel]

In this situation, I'd have the party make a group check for charisma (deception), opposed by the passive insight scores of the enemies. A similar approach would be to just pick a difficulty and use that DC for the deception check, but the enemies' passive insight already provides the most logical DC (assuming the NPCs have stat blocks).

[Lunali]

My general rule is never do opposed checks unless they're explicitly needed, a single d20 is sufficiently random. In the case of deception vs insight this usually means rolling deception vs passive insight.

[Ettina]

In this situation, I'd have the party make a group check for charisma (deception), opposed by the passive insight scores of the enemies. A similar approach would be to just pick a difficulty and use that DC for the deception check, but the enemies' passive insight already provides the most logical DC (assuming the NPCs have stat blocks).

I'd have one PC (whoever's talking the most) take point, possibly with advantage from other PCs helping (or disadvantage if the other PCs are hindering them!).

[Tanarii]

Unless they know that you're lying for certain and are just trying to catch you out, you should use passive insight. It's a secret check (from the NOC perspective), so the dice shouldn't be rolled.

But each PC attempting to deceive should roll their own check on their side.

[CheddarChampion]

I agree that using passive insight is the way to go.

The DM sets a DC of 10+insight bonus, using the most insightful enemy's stats. The PC doing most of the talking rolls deception, maybe at advantage if another PC helped*, and if they beat the DC (a tie counts as failure) the enemies are deceived to some extent.
*Requires more than just repeating what the other character said.

[Vegan Squirrel]

I'd have one PC (whoever's talking the most) take point, possibly with advantage from other PCs helping (or disadvantage if the other PCs are hindering them!).

That works, too. I was envisioning all the PCs acting like they are cultists who belong there with the group check, but if they're talking their way through then it's probably one person doing the check, with or without help.

[MoiMagnus]

If you need a justification to why more peoples doesn't make deception check harder, you can argue that there is a "group effect" and that if you find someone dubious but all your friend behave as if their presence was normal, you might shrug it off and ask "By the way, who was that guy?" later (or forget about it).

On top of what the other said, I'd note that in general for infiltration (and in fact for any task), you want to avoid circumstances where you do a set of checks and "one failure = total failure" (or the other way around "one success = total success").

If they're gonna be a lot of rolls, some failures should be able to be compensated for by additional skill checks.

[Chronos]

Quoth Lunali:

My general rule is never do opposed checks unless they're explicitly needed, a single d20 is sufficiently random.
More than sufficient, actually. A single d20 is usually too random. Using two decreases the randomness to a more manageable level.

[BoutsofInsanity]

There are a couple of ways to handle this RAW.

First thing of of note - Page 174-176 of the PHB goes over ability checks and skills and how to handle them.

The easiest method I recommend is the following.

When trying to trick a group of enemies the following should occur.

1. The players declare how they are deceiving the enemy.
2. DM sets the DC of the task depending on how difficult it is. (You can use the stats of the creatures to help with this if so chosen. Passive Insight for example).
3. If it's a single effort - One person rolls to beat the DC.
3a. If it's a group effort, the entire group rolls and if half of the group makes it then they are successful.
4. Resolve the encounter.

Note - Per the rules, this isn't a contest. The players and enemies aren't trying to do the same thing. One is attempting to deceive the other, it's a clear case of them not going after the same thing.

That's about as complicated as it needs to get. Again page 174 - 176 is your friend.

[Sharur]

Unless they know that you're lying for certain and are just trying to catch you out, you should use passive insight. It's a secret check (from the NOC perspective), so the dice shouldn't be rolled.

But each PC attempting to deceive should roll their own check on their side.

I think the requirement to "know" is too stringent (seeing as a roll is beneficial to the "actor" something like 60% of the time), I'd go with suspect, or have reason to be suspicious.

So, for example:
Overly trusting mark: Always passive.
Properly paranoid mark: Always rolls.

Being presented with an "ally" (disguised foe) in a strange circumstance, like behind enemy lines: roll (probably with advantage).
Being presented with an "ally) (disguised foe) purporting to be the reinforcements that were requested, coming from the right direction: passive.

Soldier in a camp: passive
Soldier standing sentry: roll

Guard being relieved on time (or a couple of minutes late): passive (and probably with disadvantage equivalent penalty)
Guard being relieved in unusual circumstances: roll

[Tanarii]

I think the requirement to "know" is too stringent (seeing as a roll is beneficial to the "actor" something like 60% of the time), I'd go with suspect, or have reason to be suspicious.
Not really. Like perception checks to see something you don't know for certain is there before hand, Insight checks should almost always be passive checks, due to the 'secret' requirement.

[NecessaryWeevil]

I agree with rolling vs passive Insight, but even if you give each opponent their own roll, a couple of successes doesn't mean the PCs are necessarily found out.
"I believe him."
"I don't."
Now they have to decide who's right.

[Man_Over_Game]

So that leads to the main thing, how to do deception on a group of enemies? they get to roll insight, but does each get to roll? Wouldn't that be quinquadruple advantage?

As an aside, this is a great observation of an important lesson: Don't have the players roll more than once.

Since a roll usually is for permission to do something, adding more dice to a problem usually only adds greater risk of failure.

So if the Barbarian wants to vault off a landing, grab onto a chandelier, and land on the enemy, having him make an Acrobatics check, an Athletics check, and then an attack roll will basically just doom him to failure. Resolving problems with 1 roll is best, try to reserve more for extremely difficult checks or if each check provides a separate benefit (instead of all of them needing to succeed as a whole).

When players fail, they like to fail because they knew they did something wrong, not because the procedure keeps them from having fun in the first place. Players like blaming themselves when it comes to games, surprisingly enough, but they'll blame the DM or the game if/when it makes sense to.

[Tanarii]

It's fine to call for multiple checks, the DM just needs to keep in mind the target difficulty, and how it lines up vs the reasonable result.

Take stealth or deception vs a large number of creatures for example. Passive /insight unfortunately just makes that same odds as against one creature. That's workable, but it's not necessarily reasonable. It may be more reasonable that each target of stealth or deception increases the chance of failure drastically. Or not. Depends what you think is happening with the target group. For example if any one creature succeeding doesn't necessarily alert all creatures, and they effectively operate as a group, then yeah, one TN is fine.

Compare to multiple stealth vs 1 or more creatures with identical passive scores. The larger the group trying to stealth, it gets drastically harder, because ALL creatures have to make individual checks, not one or a group check. That's entirely reasonable, it matches how things should work, if any creature screw up sneaking, they take the entire group down. But for deception, it might not be, it may be that a group check works better (although group opposed checks have no rules), because they group is working together to make their deception more (or less) effective.

[Man_Over_Game]

It's fine to call for multiple checks, the DM just needs to keep in mind the target difficulty, and how it lines up vs the reasonable result.

Take stealth or deception vs a large number of creatures for example. Passive /insight unfortunately just makes that same odds as against one creature. That's workable, but it's not necessarily reasonable. It may be more reasonable that each target of stealth or deception increases the chance of failure drastically. Or not. Depends what you think is happening with the target group. For example if any one creature succeeding doesn't necessarily alert all creatures, and they effectively operate as a group, then yeah, one TN is fine.

Compare to multiple stealth vs 1 or more creatures with identical passive scores. The larger the group trying to stealth, it gets drastically harder, because ALL creatures have to make individual checks, not one or a group check. That's entirely reasonable, it matches how things should work, if any creature screw up sneaking, they take the entire group down. But for deception, it might not be, it may be that a group check works better (although group opposed checks have no rules), because they group is working together to make their deception more (or less) effective.

You make a great point; I do disagree on the opinion, though.

Most of my players would prefer that I was more lenient and flexible to allow them to play the game how they want, but I can definitely recognize that a lot of other players (even myself, sometimes) would prefer something more simulation-y or challenging. If they wanted to use stealth as a group, I try to come up with a compromise that lets them try with a reasonable amount of success.


Tho I really miss playing with nerds that took it seriously, tbh. Felt a lot easier.

[Sigreid]

For checks of this nature I treat it as a help action to the individual with the best chance. So basically best chance rolls with advantage.

[Man_Over_Game]

For checks of this nature I treat it as a help action to the individual with the best chance. So basically best chance rolls with advantage.

If you want to get really crazy with it, you can do the best score with Disadvantage, the worst score with Advantage, and then take the highest of those two scores.

That way, you're slightly biased to succeed the more players you add but not enough that it's too much.

[Angelalex242]

I wouldn't really do group insight.

Instead...

"You think he's lying."

"You think he's more full of fecal material than the local stable."

"He sounds honest to you."

"He's telling the truth of the Gods to you."

"You have no idea if it's true or not."

[DwarfFighter]

I don't have the context of OP's group of monsters making Insight checks, but this bit should cover it. From The Player Handbook:


Group Checks
When a number of individuals are trying to accomplish something as a group, the DM might ask for a group ability check. In such a situation, the characters who are skilled at a particular task help cover those who aren't.

To make a group ability check, everyone in the group makes the ability check. If at least half the group succeeds, the whole group succeeds. Otherwise, the group fails.

Group checks don't come up very often, and they're most useful when all the characters succeed or fail as a group. (...examples...)

This presume there is an established DC each individual will check against. However, if you are making "opposed group contests", how do you select the check results from each group? This is pretty much up to the GM to decide! The simplest way is to have all roll and select the best result on each side to be compared.

However, if you have the time to consider the context of the situation, you may want to 1) change who gets to (or needs to) make their rolls, and 2) how you select the "winning" roll of each side.

1. Who rolls?
As mentioned, "everyone rolls" is a simple solution that costs no time to prepare and more time execute. If you have lots of individuals on one side, this may be cumbersome. Of course, you could have the PCs make their rolls first and start rolling for the NPCs until one of them beats the PCs result.

For very large groups of NPCs you may want to stop rolling once you've made one or two rolls per PC - there is such a thing as advantage in numbers, but it may not be a good play experience to make that advantage too overwhelming.


2. Which rolls?
When you've made a large number of rolls, two things become apparent: You have several high rolls, and the average of the d20 is pretty spot on. i.e. 10-11. Some of the options here are:


Select the highest result. Simple.
Break out your calculator and figure out the average result.
Starting with the lowest result, drop the lowest and highest results among the rolls until you have one remaining result. Select that.


I feel the last option is the one that most closely matches the group ability check rules.

---


If you want to get really crazy with it, you can do the best score with Disadvantage, the worst score with Advantage, and then take the highest of those two scores.

That way, you're slightly biased to succeed the more players you add but not enough that it's too much.

I have no idea what you mean since you seem to be using "score" to mean different things, the most reasonable of which seems to be "check result", but how do you apply advantage or disadvantage to a roll after the fact?

If you are thinking of "best modifiers and stuff", that's pretty vague since there are a host of rules that come into play to modify the result after the roll.

I think you have an idea of how this would work, but I don't think you've thought it through.

---

GM: Ok, everyone roll Insight, the worst of you have advantage and the best of you have disadvantage.

Player 1: I have a +6, and that UA class feature that lets me roll three dice when I have advantage on Insight.

Player 2: I have a +4, but I also have Guidance from before for another +1d4. Is that better or worse than +6?

Player 3: I have a +3, but I also have a Bardic Inspiration for +1d8 that I maybe want to use if I get a low roll, but I'm only using it if it's got a good chance of giving me the best result in the group.

Player 4: I have a Flash of Genius to add my Intelligence modifier to any of one of us after the roll!

GM: ...So, who gets advantage and disadvantage?

-DF

[Tanarii]

If you're doing group check charisma (deception) for the PCs because they can work together to succeed or fail, and it's not against passive insight that's a single value passive insight for all enemies, it should either be against the highest insight for the enemies or the highest rolled insight for the enemies. Assuming get caught out by any one enemy lets that one tell them all.

[Man_Over_Game]

I don't have the context of OP's group of monsters making Insight checks, but this bit should cover it. From The Player Handbook:


Group Checks
When a number of individuals are trying to accomplish something as a group, the DM might ask for a group ability check. In such a situation, the characters who are skilled at a particular task help cover those who aren't.

To make a group ability check, everyone in the group makes the ability check. If at least half the group succeeds, the whole group succeeds. Otherwise, the group fails.

Group checks don't come up very often, and they're most useful when all the characters succeed or fail as a group. (...examples...)

This presume there is an established DC each individual will check against. However, if you are making "opposed group contests", how do you select the check results from each group? This is pretty much up to the GM to decide! The simplest way is to have all roll and select the best result on each side to be compared.

However, if you have the time to consider the context of the situation, you may want to 1) change who gets to (or needs to) make their rolls, and 2) how you select the "winning" roll of each side.

1. Who rolls?
As mentioned, "everyone rolls" is a simple solution that costs no time to prepare and more time execute. If you have lots of individuals on one side, this may be cumbersome. Of course, you could have the PCs make their rolls first and start rolling for the NPCs until one of them beats the PCs result.

For very large groups of NPCs you may want to stop rolling once you've made one or two rolls per PC - there is such a thing as advantage in numbers, but it may not be a good play experience to make that advantage too overwhelming.


2. Which rolls?
When you've made a large number of rolls, two things become apparent: You have several high rolls, and the average of the d20 is pretty spot on. i.e. 10-11. Some of the options here are:


Select the highest result. Simple.
Break out your calculator and figure out the average result.
Starting with the lowest result, drop the lowest and highest results among the rolls until you have one remaining result. Select that.


I feel the last option is the one that most closely matches the group ability check rules.

---



I have no idea what you mean since you seem to be using "score" to mean different things, the most reasonable of which seems to be "check result", but how do you apply advantage or disadvantage to a roll after the fact?

If you are thinking of "best modifiers and stuff", that's pretty vague since there are a host of rules that come into play to modify the result after the roll.

I think you have an idea of how this would work, but I don't think you've thought it through.

---

GM: Ok, everyone roll Insight, the worst of you have advantage and the best of you have disadvantage.

Player 1: I have a +6, and that UA class feature that lets me roll three dice when I have advantage on Insight.

Player 2: I have a +4, but I also have Guidance from before for another +1d4. Is that better or worse than +6?

Player 3: I have a +3, but I also have a Bardic Inspiration for +1d8 that I maybe want to use if I get a low roll, but I'm only using it if it's got a good chance of giving me the best result in the group.

Player 4: I have a Flash of Genius to add my Intelligence modifier to any of one of us after the roll!

GM: ...So, who gets advantage and disadvantage?

-DF

Yeah, you're right. Thank you.

[Demonslayer666]

For checks of this nature I treat it as a help action to the individual with the best chance. So basically best chance rolls with advantage.
This would work to save time, but I would handle it differently.


I wouldn't really do group insight.

Instead...

"You think he's lying."

"You think he's more full of fecal material than the local stable."

"He sounds honest to you."

"He's telling the truth of the Gods to you."

"You have no idea if it's true or not."

Agreed.

Dogpiling this roll doesn't make sense. You can't help someone intuit other's deception. You can decide together afterwards the course of action taken and share your thoughts.

Only those that want to scrutinize the intentions should get to roll. This should be roleplayed out too. I discourage my players from dogpiling rolls. Not everyone would intently paying attention to every word.