Rules FAQ How Does Stealth Work in D&D 5E?

Stealth is a complex skill. The rules can be found in the Player’s Handbook, largely on page 177. On the surface, it seems simple: it is a Dexterity (Stealth) check opposed by a Wisdom (Perception) check. But, there is more to it than that.

---

This is the part of a weekly series of articles by a team of designers answering D&D questions for beginners. Feel free to discuss the article and add your insights or comments!

---

So let’s break it down step by step. Using stealth generally means using the Hide action. Hiding is a 4 step process:

1. Are you sufficiently obscured from the creatures you're hiding from?
2. Use Hide action; this could be a bonus action if you have certain abilities, like the rogue’s Cunning Action or the Ranger’s Vanish.
3. Compare Dexterity (Stealth) check to the passive perception scores of any creature you are hiding from and against any active Wisdom (perception) checks to search for you
4. While you remain hidden, use the same Dexterity (Stealth) result until you are detected or are no longer hiding.

While Hidden
When you are hidden (which means you have used the Hide action and a creature has not noticed you with passive or active perception):

• You have advantage on attack rolls against creatures that can’t see you.
• When you make your attack, though, you reveal your position and are no longer hidden, whether the attack hits or misses.
• If a creature tries to attack you while you are hidden (and is able to guess the space you are in), it makes its attack roll with disadvantage.

Staying Hidden
You remain hidden until you are discovered, you stop hiding, circumstances no longer allow you to hide, or you make a noise or otherwise alert others to your presence.

You do not need to continually use the Hide action every round to remain hidden, but you will need to use it again to hide once you become detected or stop hiding (this could be complex to track, as being hidden is relative to each creature).

When Can I Hide?
According to the Player’s Handbook, you “can’t hide from a creature that can see you clearly”. The complicating factor is the line "The DM decides when circumstances are appropriate for hiding”.

• The book reminds DMs that they might allow a player character to sneak up on a distracted creature, even leaving their concealment to do so, if circumstances allow it.
• It goes on to say "An invisible creature can always try to hide", noting that being unseen does not mean you are undetected.
• The Player's Handbook reminds us that the "Lightly obscured' and "heavily obscured" lighting affect what one can see. Being lightly obscured imposes a -5 penalty on Wisdom (Perception) checks that rely on sight, while being heavily obscured effectively blinds creatures to things in the obscured area and makes Wisdom (Perception) checks that rely on sight automatically fail.

We still do not have a definition for “clearly”; it is left up to DM interpretation in this context.

• We know that being invisible counts. Being invisible makes one heavily obscured "for the purposes of hiding", so heavily obscured also counts.
• Full cover is not mentioned, but since it fully blocks line of sight, it is safe to assume full cover for an opaque object would be sufficient to hide behind.

This leaves the question "Can I hide when I am only lightly obscured" or "Can I use half or 3/4ths cover to hide?" The answer seems to be left up to the DM, as there are special abilities which interact with creatures who are lightly obscured.

• The skulker feat allows you to try to hide when you are lightly obscured" implying you couldn't otherwise do this.
• Wood Elves have the mask of the wild ability that lets them use the hide action "when you are only lightly obscured by foliage, heavy rain, falling snow, mist, and other natural phenomena''.
• Lightfoot halflings have the naturally stealthy ability, which lets them hide "even when you are obscured only by a creature that is at least one size larger than you".

There are two ways to read this. The strict interpretation would be that you need these abilities in order to hide within lightly obscured areas. The loose way to interpret would be that these abilities allow you to use stealth to Hide in certain kinds of light obscurement even while being observed. As the Hide rules state you "can't hide from a creature that can't see you clearly" it depends on how the DM interprets “clearly. And, if a DM is going to allow lightly obscured areas to count as “not seen clearly”, then they may allow half cover or three-quarters cover as well.

Be sure to discuss with your DM how they intend to interpret when a creature can and cannot see you clearly.

 

[Flamestrike]

But a lot of people argue that the rogue needs advantage (typically via stealth) most of the time or they fall too far behind on DPR. I disagree with this on multiple levels, but it is a common belief.

I'd argue Tashas Aim action for Rogues is a pretty strong indicator that the Devs agree they do need advantage or else they fall behind.

 

[Oofta]

I'd argue Tashas Aim action for Rogues is a pretty strong indicator that the Devs agree they do need advantage or else they fall behind.

I don't think they need advantage, they need sneak attack.

Neither of us really know what the devs intended so it's kind of pointless. Obviously anyone can benefit from advantage and on my experience rogues work just fine.

 

[clearstream]

I don't think they need advantage, they need sneak attack.

Neither of us really know what the devs intended so it's kind of pointless. Obviously anyone can benefit from advantage and on my experience rogues work just fine.

Rogues are the only martial class with only one attack per turn (all others get extra attack.) The design space exploited is least-attacks/most-damage-per-hit.

The problem at higher tiers is, they need to hit else their most-damage is still zero. Advantage is a surrogate attack: a second roll.

5th Ed baseline is easy mechanical difficulty and in that sense you're right. All classes are over the line. My concerns are around overshadowing, the bad feeling in play when a few rounds go by missing, and the obstacle to offering a higher mechanical difficulty game that significant imbalance between classes presents.

 

[Lyxen]

I'd argue Tashas Aim action for Rogues is a pretty strong indicator that the Devs agree they do need advantage or else they fall behind.

7 years later, yeah, right... And first, this is just an option, so you have to put it in line with all the other options. IF you are using all options like feats and multiclassing, and IF you are using the additional options in Tasha (amongst others) which reinforce a number of classe, THEN it's a good thing that the rogue has himself an additional option that boosts his damage. However, before that, a well played rogues had many options under his belt to gain his sneak attack if played with a minimum of cleverness.

 

[Lyxen]

Agree on not necessarily predictable. And influencing: I feel like decisions influenced by consistent principles can come out different, and will still feel to players consistent. They have cohesion.

Yep.

Perhaps it is on the level of game as game that systematic abstractions drive consistency, and that consistency drives interest (of that sort.)

It's certainly the case for some players, and they are welcome to it if it's the style of game that they enjoy. It's a question of preference, they have theirs, I have mine, what I don't like is "them" trying to say that their game is superior because it's more "consistent" (as if the other ways of playing were inconsistent), as it's simply not true.

My take on this link is that it rightly says that good DM is good, without settling how we should best grasp game rules.

I beg to differ, on this, the 5e rules themselves are clear: "To play D&D, and to play it well, you don’t need to read all the rules, memorize every detail of the game, or master the fine art of rolling funny looking dice. None of those things have any bearing on what’s best about the game."

Again, it does not prevent the game from being played in the way for example of previous editions where rules were paramount, but the spirit of the game, it's intended design, has clearly changed.

Presumably best DM with best rules, grasped in the best way, is most good.

Not necessarily, this was in a sense the way of 3e, where rules were not bad inherently, but multiplied and comboed until they became unplayable.

I don't see trust as differentiating the argument: it's exactly as useful in any case. We don't want players to distrust DM, granted.

3e and 4e where editions where clearly you were told to trust the rules before the DM, they empowered the players. 5e went back to the initial paradigm, and I find this a good thing. The thing is that, in any case, nothing is absolute, we basically played the same type of game with my friends through all the editions, but some editions supported our playstyle better than others.

Rules have various jobs that they perform for RPGs. DM is defined up front in 5th as narrator and referee.

Which is different from 4e for example, where he is a referee first and foremost, with a few additional roles tagging along.

Consistency is just a tool. Written rules feed into consistency. For example, generally players can expect their paladin to use a d10 for Hit points. That consistency drives meaningful choice. A lot of consistency in application is like that: more or less invisibile.

I would argue that it's not invisible, it's just as visible as the table wants it to be. Some table will focus heavily on the rules, especially in combat, for example, and this was clearly the case forced by 4e, where combat had to happen on a grid, in squares, etc. extremely formal. Other tables will make it almost invisible within the narrative, in our last session, there was an assassination attempt on a Duke of Hell inside his Blood War command tent when the PCs where having a drink with him, Dogai (Assassin Devils) led with a heavy fog which created small islands of fighting inside and around the huge tent, all was chaos and managed through theater of the mind and mostly about the narrative, not about the consistency of the rules, but about the consistency of the game world and the stories within.

It's just that, once more, it's not absolute one way or the other at most tables, and nothing is superior in terms of gameplay, it's just best to use what creates fun at a given table.

 

[clearstream]

what I don't like is "them" trying to say that their game is superior because it's more "consistent" (as if the other ways of playing were inconsistent), as it's simply not true.

There are so many factors going into great gameplay that consistency alone cannot tell us a game is superior: an inconsistent game might be superior on sum of other factors.

Consistency has been acknowledged as valuable in convincingly presenting imaginary worlds, and also for good refereeing.

I would accept a thesis that some RP might profit from inconsistency - although that has never been my experience.

D&D 5e is rules-medium-to-heavy. Hundreds of pages of rules, most of which are applied pro forma... what I have dubbed invisibly. The damage of a longsword. The weight of a shield. The modifier for 14 Int. Spells known. Expenditure of a spell slot. Attacks per turn. The effects of a condition. Exceptions are rarer than conformances, notwithstanding that we notice them more.

Again, it does not prevent the game from being played in the way for example of previous editions where rules were paramount, but the spirit of the game, it's intended design, has clearly changed.

3e and 4e where editions where clearly you were told to trust the rules before the DM, they empowered the players. 5e went back to the initial paradigm, and I find this a good thing. The thing is that, in any case, nothing is absolute, we basically played the same type of game with my friends through all the editions, but some editions supported our playstyle better than others.

For us trust in the DM has been foundational in every edition. It's not new, and there was no lacuna. 5e intentionally de-risks player commitments - more so than other editions - in response to backlash to 4e's opinionated design: that's true.

Which is different from 4e for example, where he is a referee first and foremost, with a few additional roles tagging along.

I don't see that as tied crucially to rulset, although designs can certainly be opinionated in that way.

Contemporary cultures of play aim to empower players in different ways. None exclude DM trust, unless by trust you mean power? One way to empower players is via narrative fiats. Another is via mechanical fiats. As for me the most exciting contribution of mechanics to RPG is in their production of narrative fiats, the two for me share sympathies.

 

[Lyxen]

There are so many factors going into great gameplay that consistency alone cannot tell us a game is superior: an inconsistent game might be superior on sum of other factors.

I agree, still there are those who argue that, since it's a quality, they have more...

Consistency has been acknowledged as valuable in convincingly presenting imaginary worlds, and also for good refereeing.

See above, what I'm aguing is that there are at least two types of consistency, that of the rules and that of the world, and that they can actually be at odds.

D&D 5e is rules-medium-to-heavy. Hundreds of pages of rules, most of which are applied pro forma... what I have dubbed invisibly. The damage of a longsword. The weight of a shield. The modifier for 14 Int. Spells known. Expenditure of a spell slot. Attacks per turn. The effects of a condition. Exceptions are rarer than conformances, notwithstanding that we notice them more.

5e is actually rules-light in terms of D&D, for example the combat section is 30% of the combat section of 4e.

For us trust in the DM has been foundational in every edition. It's not new, and there was no lacuna.

So it has been for us, still we had our share of ruleslawyers in 3e, as it encouraged it for example.

5e intentionally de-risks player commitments - more so than other editions - in response to backlash to 4e's opinionated design: that's true.

We agree there.

I don't see that as tied crucially to rulset, although designs can certainly be opinionated in that way.

For me, it's the other way around, you get a design intent first and it influences the ruleset, quite heavily. For example, in 5e, the decision not to create a game jargon so as not to scare new players (which I think actually worked quite well) certainly influenced the way the rules were written.

Contemporary cultures of play aim to empower players in different ways. None exclude DM trust, unless by trust you mean power? One way to empower players is via narrative fiats. Another is via mechanical fiats. As for me the most exciting contribution of mechanics to RPG is in their production of narrative fiats, the two for me share sympathies.

For me, there is a massive difference between telling a DM "you are the lead storyteller" and "you are like the referee in a competitive sport". Both have authority and power, but if applied to the letter, it results in very different games.

 

[clearstream]

For me, there is a massive difference between telling a DM "you are the lead storyteller" and "you are like the referee in a competitive sport". Both have authority and power, but if applied to the letter, it results in very different games.

5e hedges nicely with "One player, however, takes on the role of the Dungeon Master (DM). the game's lead storyteller and referee."

My ideal is where rules deliver story. Especially stochastically, so that we cannot know what will happen next. The G abets the RP. In that context storyteller and referee are not either/or, but both.

 

[Lyxen]

5e hedges nicely with "One player, however, takes on the role of the Dungeon Master (DM). the game's lead storyteller and referee."

My ideal is where rules deliver story. Especially stochastically, so that we cannot know what will happen next. The G abets the RP. In that context storyteller and referee are not either/or, but both.

It's all a question of degree and of the balance that you want at your table, and I understand your perspective, but in the end (and the answer does not have to be the same one even depending on circumstances at the same table), it all comes down to one question, if you have to choose between story and rule, which one will prevail ?

Because, there will be cases like that, and I can definitely tell you that, at our table, story will in general prevail.

Which, coming back to the subject of the thread, is why stealth rules are not that important, the guidelines that we have are good enough for storytelling intrigue and even interesting combat.

 

[Bill Zebub]

I don't think they need advantage, they need sneak attack.

Neither of us really know what the devs intended so it's kind of pointless. Obviously anyone can benefit from advantage and on my experience rogues work just fine.

I can think of four possible design goals behind Aim:
1) Rogues are supposed to have Advantage
2) Rogues are supposed to have two chances to hit per turn
3) Rogues are supposed to have Sneak Attack
4) Rogues are supposed to have cool things to do with their bonus action