If a PC is sneaking past a bad guy, the bad guy gets a perception check against the the PCs stealth at the first plausible opportunity for the player to be noticed. If the player wins that roll and continues to sneak, how often should the bad guy get another roll?

I went looking for official rules on this and couldn't find much. From the perception skill description, it says that you may "can try to sense something you missed the first time, so long as the stimulus is still present." I am assuming that the bad guy does not know the PC is there, and has no motivation to go looking. On the other hand the sneaking PC keeps making small noises as he moves.

I have read a bunch of threads about how to set the DC of the stealth check, but no official guidance on how often to make the roll. In most circumstances, the chances of success strongly depend on both.

I am assuming that there is nothing else important going on, but if you can answer without that assumption, go ahead. I am really looking for rules or official developer commentary, but I will accept answer from experience if there are no rules.

  • \$\begingroup\$ "Stealth" is not a verb. Replacing it with an actual verb in places that it is erroneously used as a verb is often helpful when answering such questions. \$\endgroup\$ Mar 9, 2017 at 16:46
  • \$\begingroup\$ @DerekStucki “Stealth” is regularly verbed into “to stealth” in the RPG community. It's not Queen's English, but as established domain jargon, changing it doesn't meaningfully improve the readability of a post. \$\endgroup\$ Mar 9, 2017 at 22:46
  • \$\begingroup\$ @sevensideddie my argument is that it's easier for a GM to make a ruling if they know whether the PC is trying to hide vs move silently vs etc. "Stealthing" is nebulous enough (due to not being a word) that its very usage creates situations of confusion. In the context of this question, I believe it matters what the PC is actually doing that has been blanketed under "stealthing" in order to give a good answer. \$\endgroup\$ Mar 9, 2017 at 23:05
  • \$\begingroup\$ @DerekStucki Ah. Since the verb has been replaced with the equally-unspecific “sneak” though, it might be more direct to just ask for which is being done. \$\endgroup\$ Mar 10, 2017 at 0:35
  • \$\begingroup\$ @DerekStucki In Pathfinder, hiding and moving silently have been combined into an in-game skill, Stealth so the terminology is correct... \$\endgroup\$
    – Ifusaso
    Mar 10, 2017 at 13:41

3 Answers 3


There is no such time interval

If a hero successfully hides, DM is not obliged to repeat the check just because some time passes.

If there was such an interval, noticing the hidden hero was the matter of time. However, winning a Stealth/Perception contest assumes the hero can succeed (hence, stay unnoticed).

From the Core Rulebook:

Skill Checks
When your character uses a skill, he isn't guaranteed success. In order to determine success, whenever you attempt to use a skill, you must make a skill check.

So, winning the contest doesn't mean you hide "for the next N seconds", it means the whole situation was successful. However, the situation itself might change.

For example, when your hero leaves cover, you need to make another roll if you want to remain hidden:

Breaking Stealth: When you start your turn using Stealth, you can leave cover or concealment and remain unobserved as long as you succeed at a Stealth check and end your turn in cover or concealment.

That doesn't imply you need to make such roll when you don't leave cover.

DM should ask for another roll when circumstances change

If the hero tries to do something else except hiding, or the "bad guy" get a clue, or circumstances change (e.g. morning comes), the DM can ask for another check, if they thinks it is necessary - but the more checks they asks, the more risky situation becomes for the hero. How risky it should be, is completely up to the DM.

DC itself can depend on how long the hero intents to hide for. Hiding all day is harder than hiding for a few seconds. In both situations it will be a one single roll though.

  • \$\begingroup\$ If the hero continues to sneak, does he get to keep his first success, or is that a second time he is using a skill? Typically pathfinder assumes that the GM sets the riskiness of a situation through DC, not number of rolls. \$\endgroup\$ Mar 9, 2017 at 18:10
  • \$\begingroup\$ @MarcusYoder "sets the riskiness of a situation through DC, not number of rolls" - correct, that is the reason why you keep success from the first roll. You already beat the DC, no need to make another roll within the same situation. \$\endgroup\$
    – enkryptor
    Mar 9, 2017 at 18:14
  • \$\begingroup\$ You seem to be implying that continuing to sneak is the same situation. This makes sense, but can you provide rule support? \$\endgroup\$ Mar 9, 2017 at 18:48
  • \$\begingroup\$ @MarcusYoder continuing to sneak is not the same situation. It could be tho. What am I talking about is that continuing to sneak in the same situation needs no additional rolls. \$\endgroup\$
    – enkryptor
    Mar 9, 2017 at 19:22

You make Stealth checks every time you move or perform any other activity that may somehow discover you.

Stealth check is a part of a move action, so every time you want to move stelthily you have to apply Stealth. You may hide in place, also as a move action, and once hidden you do not need any rolls in subsequent rounds untill new activity is performed.
It isn't mentioned directly in skill's description, but it is obvious enough that if you want to, for example, draw a hidden weapon silently while sitting in the ambush, you should roll Stealth in case you want no one hear you doing it.
For actions of that kind (non-movement ones) it may be questionable though if you roll Stealth or the action itself has fixed Pereption DC. Opening a door may, for example, be noisy no matter how good you are in doing things silently, and it is probably impossible to open it under observation without observer noticing. It most likely is up to DM.

Your opponent makes Perception checks as a reaction to every stimulus. Or she can actively look for something as a move action.

Every your movement or any other activity mentioned above is a stimulus and results in a Perception check reaction. While every 10 feet of distance affect your opponent's rolls, it is still once per your activity and probably takes into account worst distance for you (or medial maybe, the rules do not clarify).
If you freeze and wait there is no obvious stimulus for your opponent to make additional checks untill she is actively looking for something. But it is only true untill situation changes in some more or less significant way. If, for example, your opponent moves toward you, it is reasonable for DM to assign additional rolls at least at some points, when difference in expected Perception check outcomes due to change in distance modifier is significant enough. Same may be true if circumstances change. As an example immagine wagon train passing the road you successfully hiding nearby. It is moonless night and lights from wagon lamps are casting shadows, your among others. Someone may notice that.


The perception checks are based on the stealth checks, which are based on the PCs actions.

From experience with other table-tops (and I believe Pathfinder behaves in the same way),

When a player has stated their intention and approach, ask yourself the following questions:

  • Is the action possible?
  • Do I need to use the game mechanics to figure out what happens?
  • How do I determine random success?

So, when your player describes how to move stealthily in some situation, you decide when the check is necessary, which is usually when the situation has changed. A simple example is moving from cover to cover. The player can be seen several times, hence several Stealth checks are necessary. Stealth Checks is defined as:

You are skilled at avoiding detection, allowing you to slip past foes or strike from an unseen position. This skill covers hiding and moving silently.

and Perception Checks as:

Perception has a number of uses, the most common of which is an opposed check versus an opponent’s Stealth check to notice the opponent and avoid being surprised.

So, when the PCs roll for Stealth, the opponents can notice (actively or passively) the movement.

Consider you have some orc seating near a campfire. Two boulders are nearby and your player wants to move through the shadows to the first boulder, then to the second, and then jump on the orc. The other option would be to just charge straight ahead at the orc, but that would not be possible to do stealthily, so the PC wouldn't even bother rolling.

You would tell him to make an easy Stealth check to approach the first bolder, then another to crawl to the second boulder, and maybe a final third to jump the orc. Therefore, the orc would get 3 Perception checks against the Stealth checks if the orc was facing (or had a chance to see) the move.

In other words, you define what actions require an additional Stealth check and the orc can make Perception checks in response to those.

Another easier situation would be if you could actually move right to the second bolder (and closer to the orc) but without ever being in the line of sight of the orc. The DM could rule that it would be impossible for the orc to see you, and have him roll Perception with a negative modifier, like the one shown in the Perception modifier table:

  • Through a wall || +10/foot

which makes the difficulty for the Orc to notice you much harder.

The difficulty and amount of rolls you have to make are ultimately dependent on the actions the PCs execute and the different situations they are facing.

  • 2
    \$\begingroup\$ Could you clarify what part of the book this comes from, or what experience of yours caused you to reach this conclusion? \$\endgroup\$
    – user31942
    Mar 9, 2017 at 17:28
  • \$\begingroup\$ Don't have the books atm, so I can't. My campaign works this way, and it is how we do it in D&D as well. If anyone can find good references, I'll add them in. \$\endgroup\$
    – BlueMoon93
    Mar 9, 2017 at 17:30
  • 1
    \$\begingroup\$ -1: This answer indicates that the more reasonably sneaky a stealth plan is the less effective it is, and provides no rules to indicate that that is actually the case, nor any acknowledgement of the consequences of this houserule. Also since the OP was kind enough to put explicit requirements as to answer quality, I'm flagging this as not an answer \$\endgroup\$ Mar 9, 2017 at 17:33
  • \$\begingroup\$ I have to agree with Mactrent, and dark wanderer. Your interpretation is reasonable, but the rules you cite don't actually help. \$\endgroup\$ Mar 9, 2017 at 17:35
  • \$\begingroup\$ Updated. This is my best answer, imho. Let me know what you think \$\endgroup\$
    – BlueMoon93
    Mar 9, 2017 at 18:01

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .