Stealth in Combat

[DM_Blake]

Let's start from what we know.

• A stealth check is made as part of an action, and covers that action.
• It's possible to get ongoing stealth.
• Ongoing stealth ends if you don't have cover or concealment.
• Ongoing stealth ends if you attack or shout.
• It's possible to get make an attack stealthily. (Otherwise, the ability of some monsters to keep stealth after missing an attack would be useless.)

The logical synthesis of those tidbits is that a stealth check must do two things. First, it must hide a single action (a move, an attack, or anything else). Second, it must give you ongoing stealth in at least some cases.

The rules, unfortunately, are silent on when you get ongoing stealth. It's obvious that an attack can't give you ongoing stealth. On the other hand, not having ongoing stealth after moving would defeat the purpose of stealth.



So there are some cases that the rules cover, and some that they don't:

• Making a Stealth check as part of a move action gives you stealth for the duration of the action and afterwards.
• Making a Stealth check as part of an attack gives you stealth for the duration of the action, but not afterwards.
• Making a Stealth check as part of any other action gives you stealth for the duration of the action... and possibly afterwards, it's not clear.
• It's also not clear if ongoing stealth will last long enough to make an attack, or if you need to make a separate Stealth check as part of the attack.

Personally, I'm ruling that you only get ongoing stealth if you move, and that the stealth doesn't apply to an attack unless you make a Stealth check as part of the attack. I can see the argument for allowing attacks to keep ongoing stealth until the end, though.

The flaw here is that you would allow everyone two chances to have stealth during an attack.

Example:

I move behind a tree. Roll for stealth. I then attack, hoping to have stealth for combat advantage. I can now roll for stealth on the attack.

If you rule that my first roll for stealth gives me "ongoing stealth" that applies to my attack, then I have a decision to make: was my stealth roll really good? If it was, no need to make a stealth roll during my attack since I know I have "ongoing stealth" and will have a great chance to get combat advantage. But if my stealth roll was a bad roll, then I simply announce that I'm attempting to attack stealthily and make a new stealth roll for my attack option.

Two chances to get a good stealth roll, and either one can grant me combat advantage.

I don't think the RAI ever meant to grant everyone two chances at stealthy combat advantage.

I think the only reasonable ruling here is that if you announce you are going to attack, you automatically forego any "ongoing stealth". Because this is an attack, and because the PHB says attacks break stealth, your attack breaks your "ongoing stealth" automatically.

You can, of course, attempt to attack stealthily, following the RAW and the RAI, by including a stealth roll with your attack action.

This way, only the stealth roll during the attack action can determine whether the attack is a stealthy one deserving combat advantage.

 

[clearstream]

The flaw here is that you would allow everyone two chances to have stealth during an attack.

Not really. The flaw to my mind is making a player try to stealth twice, to get one good attack.

If you rule that my first roll for stealth gives me "ongoing stealth" that applies to my attack, then I have a decision to make: was my stealth roll really good? If it was, no need to make a stealth roll during my attack since I know I have "ongoing stealth" and will have a great chance to get combat advantage. But if my stealth roll was a bad roll, then I simply announce that I'm attempting to attack stealthily and make a new stealth roll for my attack option.

If your DM deems that appropriate then you could do that, were you a nervous sourt of rogue.

You can, of course, attempt to attack stealthily, following the RAW and the RAI, by including a stealth roll with your attack action.

This way, only the stealth roll during the attack action can determine whether the attack is a stealthy one deserving combat advantage.

By RAW an attack ends hiding. Hence you hide, then you do your charge, or deft strike, or ranged attack or whatever, to sneak attack.

Edited to add that your DM would probably not grant a stealth check unless your action is appropriate, since that remit is given her on PHB178. So polishing your cross might not do it, but hunkering down probably would.

-vk

 

[komi]

(edit) Ninja'd by Mearls (via redbeard)! He seems to have close to the same interpretation, though he also has the attacker roll Stealth against passive Perception rather than what I was saying. Mathematically I'm not sure it makes much difference (I'm not thinking too hard about it). Maybe it's better aesthetically to have the stealthed attacker roll because he's the acting character.

(Because graphs are fun and this was similar to what I did for the Improved Init vs. Danger Sense discussion...)

(EDIT: I don't know why message board is distorting my image. If you view the image directly it will be much clearer.)

Mathematically there is a difference when rolling against active perception and passive perception. Here's a chart of the chance of success on stealth vs. perception for both active and passive checks. The x-axis shows the difference in skill modifiers between stealth and perception. Positive means the person attempting stealth is better than the person trying to perceive.

So, if the stealther is better than the perceiver (delta modifier is positive), it's better for the stealther to roll against passive.

We can take the change from active to passive as a modifier on an active check. Say you're stealth skill modifier is five higher than the person you're hiding from. From the chart above, you have an 80% chance of success on a passive check (e.g. 1d20+5 vs. DC 10). What modifier would you need on the active chart to achieve an 80% chance of success? Looking at the blue line on the chart above, you get about 7 (6.84 from interpolation). So if we look at chance of success, rolling with a +5 versus a passive check is about the same as rolling with a +7 versus an active check.

(1d20+5 vs. DC 10) ~= (1d20+7 vs. 1d20)

So you get an effective +2 bonus to your stealth when your DM decides you can roll against passive to save dice rolling. Here's a chart that shows this. The blue line is the actual interpolated results. The red dots show the same data rounded to the nearest integer.

The break even point where active and passive are the same is -4. At 9, you're getting a +10 to your check. You can mentally see this if you think that (1d20+9 vs. DC 10) is always a success. To get a guaranteed success for an active check you need to be +19 better than your opponent. (1d20+19 vs. 1d20) will always succeed. You have an effective +10 to your check when rolling against passive.

I'm not writing this to show anything is broken nor to state an opinion, but rather to better convey the math behind it. You can go with passive checks, but be aware that people using stealth a lot will be specialized and will very likely be in the positive x-axis domain above. In fact, once you're +9 better than you're opponent it's impossible to be seen until they attempt a minor action perception. Conversely, requiring mostly active checks means a lot of dice to be rolled and might painfully slow the game down. So when we understand both sides, we as DMs can make better decisions.

 

[deleteme123456]

So you get an effective +2 bonus to your stealth when your DM decides you can roll against passive to save dice rolling.

Does the math imply that we can simply impose a -2 penalty to the Stealth check, compare it to passive perception, and have roughly the same results as if we had rolled active perception? I'm not asking for miracles, I know it won't be exactly the same, but I'm willing to accept a little variance for a lot of ease of use.

 

[Sanzuo]

...

Okay, you know this stealth debate is getting ridiculous once the graphs come out.

 

[komi]

Does the math imply that we can simply impose a -2 penalty to the Stealth check, compare it to passive perception, and have roughly the same results as if we had rolled active perception? I'm not asking for miracles, I know it won't be exactly the same, but I'm willing to accept a little variance for a lot of ease of use.

I meant to write about that but forgot. With my +5 example a -2 penalty is about right. But if the stealther is +7 versus the opponent, it doesn't balance out. However, you might find it's close enough.

If you want to use passive checks to save time, then it would be quite impractical to look up on a chart exactly what penalty to apply since that would take longer than just rolling. But you could look at your rogue's bonus and compare them to some typical monsters. If you're rogue isn't completely tricked out in stealth, maybe a flat -2 is about right. But if he's consistently +8 or +9 higher, then you'd need a higher penalty. Just look at the graph for something appropriate.

And this is only for modeling passive checks as active. Sometimes the rules call for passive checks, in which case you shouldn't apply any penalty.

Okay, you know this stealth debate is getting ridiculous once the graphs come out.

I would say exciting!

 

[clearstream]

...until they attempt a minor action perception.

PHB186 deems active perception a standard action. I know somewhere along the community became informed it was a minor action. Can anyone let me know where/when that was?

-vk

 

[komi]

PHB186 deems active perception a standard action. I know somewhere along the community became informed it was a minor action. Can anyone let me know where/when that was?

PHB p. 281 at the bottom of the page. It's poor editing that this is not mentioned in the skill description. Since this is in the context of searching for unseen opponents, other uses of active perception require a standard action.

 

[Xorn]

I don't know if I should be happy, sad, or embarrassed that I started this thread.

I'm going to quit resisting the urge to post in it, and dive right back in. Combat advantage is granted for reasons related to stealth either because the target can't see you, or isn't aware of you.

The attack at the start of a combat is obviously because they aren't aware of you, even if you aren't totally out of sight/invisible. What happens in combat after you have been spotted was the original intent of this thread.

The question originally posed is--can I regain combat advantage through stealth in combat? That means that through stealth, I must:

1. Make the target unable to see me.
OR
2. Make the target unaware of me.

There are no clear rules for this, at all. What we KNOW:

A Bluff check makes them lose sight of you for a moment.
Total Concealment makes them lose sight of you.
Superior Cover makes them lose sight of you.
Regular cover/concealment allows you to make stealth checks.

But there is no indication that stealth can make someone no longer see you (or no longer be aware of you) with a stealth check. There is also no indication that stealth will not allow this--so that's the line that divides the two camps, really.

I think we all agree that with Bluff or Total Concealment or Superior Cover, the target can't see you, and you get combat advantage. What's left is these two splits:

1. You can make a stealth check in combat to make yourself unseen by a target if you have cover/concealment, as a (insert) action. (Most agree on Move action, as kissing your good luck charm does not hide you.)

2. You can't make a stealth check in combat to make yourself unseen by a target if you have cover/concealment, you must get out of sight. You may of course use stealth to move back into regular cover/concealment undetected, though.

I'm in camp #2. The split seems to be pretty even between the two. I'm not saying either camp is wrong, only that neither camp can legitimately claim to be right. My latest discoveries in running Keep on the Shadowfell and reading the Monster Manual on the toilet (you know you do, too) is that the Tactics never once mention making a stealth check in combat to gain combat advantage, even though it's often mentioned to make a stealth check before combat to gain combat advantage.

The only mention of gaining combat advantage during a fight is a goblin fight where they are completely out of sight, then can make a stealth check to attack from hiding. I'm still not saying I'm right--I'm just saying that reaffirms the choice of camp I made.

 

[deleteme123456]

I meant to write about that but forgot. With my +5 example a -2 penalty is about right. But if the stealther is +7 versus the opponent, it doesn't balance out. However, you might find it's close enough.

If you want to use passive checks to save time, then it would be quite impractical to look up on a chart exactly what penalty to apply since that would take longer than just rolling. But you could look at your rogue's bonus and compare them to some typical monsters. If you're rogue isn't completely tricked out in stealth, maybe a flat -2 is about right. But if he's consistently +8 or +9 higher, then you'd need a higher penalty. Just look at the graph for something appropriate.

And this is only for modeling passive checks as active. Sometimes the rules call for passive checks, in which case you shouldn't apply any penalty.



I would say exciting!

Unless of course we can model it with a simple formula
looks like -5 to +1 is about even or close enough.
+2 to +4 is about 1 off
+5 to +7 about 2 off
+8 to +10 ... 3 off is close enough probably
+11 to +13 ... 2 off
+14 to +16 ... again, 1 off isn't perfect but honestly your odds are so good at +16 that soaking up a -1 penalty isn't going to kill you.

So you just have to remember that it starts at +2 with a +1 and goes up every +3 more you have until 10 and then it goes back down.

I wouldn't even worry about the times when stealth is lower than perception... it's probably not going to be more than -5 or you're really not going to try, but if a player really insisted I would give the bonus ... I suppose.