Take the 2-minute tour ×
Role-playing Games Stack Exchange is a question and answer site for gamemasters and players of tabletop, paper-and-pencil role-playing games. It's 100% free, no registration required.

There are some play-styles of D&D in which the spectre of player-character death is considered a feature of the game rather than a bug. For my own reasons (which aren't the point of the question), I see them as a feature especially in exploration-focused games (e.g., sandboxes, old-school dungeon-crawling) where there are lots of ways for players to arm themselves with enough information to make informed decisions about the dangers they're willing to tackle.

Even with players that play strategically, seeking that intel and using it to pick their battles strategically, it is still on the DM's shoulders to provide good clues and tell-tales during play. How well the DM does this will significantly impact the players' ability to make informed decisions on where to go, how to prepare, what to fight, and what to avoid. If the players don't catch the clues, it should be because they weren't paying attention or didn't put 2 and 2 together, not because the DM's descriptions suck.

What techniques can the DM use to effectively telegraph to the players the existence and/or nature of the dangers they face?

These assumptions are inherent in the question:

  • The players are already playing strategically, and don't need to be "trained" in this play-style.
  • Senseless, random PC death is a real possibility, not just threat to set the tone of the game or generate fearfulness in the players.
  • Characters almost always die when players put them in dangerous situations they're not prepared to deal with.
  • Characters might live when players are observant, know what dangers to expect, are well-prepared, and don't take on dangers that are beyond their ability—and even then, the dice have to go in their favour. Knowledge is the edge they need to bend the odds in their favour.
  • The players don't necessarily have access to stat blocks or other, out-of-game information on in-game dangers.
  • Players are relying on the DM's descriptions to understand the world, including its dangers and rewards.

(This isn't an invitation to argue whether an exploration, frequent-PC-death play-style is good or not. That's just background to the question.)

share|improve this question

9 Answers 9

up vote 56 down vote accepted

The single-most overlooked, in my experience, is evidence of deaths.

It's a dragon? What's it been eating? Few creatures actually eat EVERYTHING, so what's left by the dragon? Oh, there's an owlbear's beak and claws... there's mangled bits of what used to be +5 plate... (see those runes, there, there and there?) A broken longsword. Dead mind flayers.

And don't forget the massive piles of poo.

Also overlooked: Tracks for Numbers and size! Sure, it's a dragon lair. But its a lot different when the dragon's "wheelbase" is 20x5 vs 80x20...

And don't pull any punches at low-levels. If you're going the route of the "killer GM," go ahead and use that medusa as the boss of dungeon level 1. Served by myconids... (Plants are immune to Medusae gaze). And all kinds of statues... birds fallen from the sky... small critters, big critters... "Hey, isn't that Fred the town Bully? So that's what happened to him... hey, there's my missing purse!"

share|improve this answer
Right on. I did this once with a dead, chewed-on drow discovered in an above-ground city ruin. For 1st-level characters, the presence of drow on the surface worried them, and then whatever killed this one worried them more. I'd forgotten about that! –  SevenSidedDie Oct 18 '10 at 4:37

I think it begins at the campaign level - nowadays, with many campaigns/GMs being of the "oh not character death, that would be unspeakable" ilk, you need to come out up front and tell people that "this campaign is an easy-death campaign - encounters will NOT be "scaled to your EL" and it will be up to you to determine what challenges you can pull off without snuffing it - I will TPK you without mercy if it comes to it. Don't be afraid to run." Just so the expectation is clearly set.

Then kill somebody. Might as well get it out of the way at level 1 where rerolling isn't too traumatic. Give them some kind of foreshadowing - locals in the inn saying "That guy's said to have killed thirty men" - then set it up to be likely that they'll give him static and have him frickin murder someone. Chase the rest of them out of town. "Threat of adversity" is bullcrap. Only real adversity creates behavior change.

After that, in terms of telegraphing specific threats, there's a number of ways.

  1. Word of mouth. Locals warn you away from, celebrate the deeds of, shutter their windows at night in fear of, etc. the more badass things. "Our lumberjacks fought them off but took some losses, can you go wipe them out" is manageable risk, "Our lumberjacks went into the forest and were never heard from, but a scout went in and found thirty skulls strung from the trees" is not. Eyewitness reports are a stronger version of this.

  2. Environmental. The location where something like a bad monster operates doesn't show no sign of them. Just like someone who knows the wilderness from a hole in the ground can tell if they are in bear or wolf country, a PC should be able to discover "Uh, looks like owlbear hunting grounds" from spoor, tree scratches, tracks, wildlife behavior, etc. Or, at the higher end, dead chewn armies.

  3. Skills. All those lovely knowledge skills in D&D 3e help PCs identify critters and understand their general power level. "Hey, that ghosty thing over there - I think that's way out of our league." "See this slime? Left by a grey ooze; very nasty."

  4. Seeing it. Now, it can be hard to set up a no-risk opportunity for PCs to see the enemy in action, unless it's like a civilized place and their enemy's a person type. "That guy challenged him to a duel and he turned him to stone and just laughed!" But let's say it's a big monster and the duke sends 20 men at arms with you, and the monster just tears through them, giving the PCs an opportunity to skedaddle.

  5. Historical treatises. Like word of mouth, but for moldy oldies. "Caused the end of a civilization" means "do not attempt under level 10".

And then you just have to be careful to have a way to describe the threats that can be decoded by the PCs. "Well sure whatever it was scared farmers, but they're level zero punks, it's probably just goblins." Or if you describe something that ended a civilization and it's CR5. This does affect a number of published adventures; I remember one where an ancient red dragon wanted to hire us level 1 characters to get her eggs back. "You must not want them very much!" was my thought.

share|improve this answer
+1 I like this answer the best, because it is structured. Like linux philosophy preaches, one can read the five bullet points and use them as ideas; but one can also read the whole post and get more insight. –  Vorac Jul 31 '12 at 9:07
@mxyzplk When you said, "I remember one where an ancient red dragon wanted to hire us level 1 characters to get her eggs back. "You must not want them very much!" was my thought." I actually LOL'd at that. Having a consistent campaign milieu will curb out-of-game logic from interfering with the roleplaying experience... unless of course the particular game system expects this type of play. –  user23715 May 10 '14 at 22:20

If you want to signal danger, aside from some of the excellent suggestions above (particularly those of aramis), there's a few other techniques you can use.

First, there's the "red-shirt" approach. Kill an NPC. The more horrible the danger, the closer the NPC should be to the PCs. This technique, of course, only works if you've got players who actually play out their characters caring for anything other than themselves, but if you've got such a crowd, go for it. Just don't take it to the farcical levels of Star Trek....

Upping the ante, maim the NPCs instead – particularly NPCs that the PCs care about. A death is unpleasant, but gone from sight for the most part rather quickly. A maiming is just as unpleasant and is in the face for a longer spell. (Indeed the maiming could itself be a source of quests.)

Upping the ante still further, maim a PC. This can really upset the players, though, so be careful what you do and who you do it to.

Even higher up the annoyance scale is making PCs lose prized possessions. (Yes, IME, most players are more annoyed at their PC losing possessions than limbs. I can't figure it either.)

All of the above needs to be applied in a way that signals the nature and source of the danger, of course, but doing that they're all fail-safe ways to signal danger without falling into the trap. What is that trap? Killing the PCs. PCs being killed as a "signal" is widely viewed (whether you think it "right" or not) as a "killer GM" move. Depending on the complexity involved in making characters or the attachment to characters typical for your players it may even be seen as a dick move. Killing PCs is fine in perceived-fair conflict as the result of player choices and actions (or bad luck), but doing it as a "signal" of how bad the big baddie really is can backfire spectacularly.

... Unless ...

If you can get a player to volunteer to be killed as the signal (on the sly without the other players' knowledge), this can be highly effective. It's rare to find a player willing to do this, again, IME.

share|improve this answer
Great comment on not killing PCs as a signal of danger. That would definitely defeat the purpose in this sort of game! –  SevenSidedDie Oct 19 '10 at 15:26

An underused approach is applying character expertise.

When I GM, I give different information based on who's doing the asking and what they can see/observe.

The fighters can make better judgments of how strong something is, and how well organized a group is in combat. The wizards can estimate levels of magic being thrown and whether something probably has magic resistance. The rogue always sees areas that are potentially good for ambush (against the party, or to be used BY the party). etc.

If people have taken knowledge type skills, I give a lot more raw info before the dice even get rolled.

One thing that frustrated me in a campaign I played years ago was that the GM would describe things literally - "A large wooden device, wheel shaped attached to the building" which anyone would know as a water wheel, but instead he would describe things this way. It meant that we'd often end up misinterpreting things as more or less dangerous than they were, because obvious character knowledge would have made it clear.

Even unclear things can be put through this lens: "This thing is either a trap or some kind of specialized mechanical set up. You're not sure which. Probably safer to leave it alone or get someone who really knows engineering to look at it."

The other thing is certain character types should KNOW what to look out for in certain areas:

"As a ranger, you know these hills have owlbears, displacer beasts, and giant hawks. You saw a very old set of owlbear droppings, but you should keep your eyes out for more signs. Avoid very large trees, and rocky outcroppings if you can."

These kinds of expert forewarnings allow players to make better decisions and helps reward them for the character choices they have made.

share|improve this answer

In order for players to be able to accurately assess danger, they must be able to gather intelligence. The presence or absence of players intelligence gathering activities should directly inform their chances at success, not through any mechanistic way, but through providing for planning and strategy.

The trick here is that the GM must be willing to release many of the ideas of the traditional adventure. A good party, with good intel, will perform ambushes, try to sneak past or distract danger, and otherwise confound the plots of the enemies. This can be incredibly frustrating to a GM who just spent 5 hours creating an elaborate series of encounters.

In many ways, the way to accurately telegraph danger is to provide the ability to accurately gather intelligence and to demonstrate that it's possible with NPCs.

Intelligence is comprised of trying to assess two factors: capability and intent. Capability is the form of "they are able to field 3 armies, trained to such and such specifications." In a fantasy game, it would be "This monster as a petrification attack that has destroyed X and Y notable heroes"

Intent is less relevant to assessing danger beyond the obvious "will they attack us?"

One interesting document is here (google cache) looking at the US Army's Humint practices.

Beyond humint, and in order to have an accurate idea of the threat of a monster, players must be given sufficient information to reconstruct recent battles between the threat and opposition. With sufficient tracking, they'll be able to create a hierarchy of threats that, eventually, will correlate with something that they fought.

Example Gratia: Players are investigating a complex swamp, before entering into the temple. They've fought the Swamp Goblins on the outskirts and defeated a large group easily, having a little bit of trouble with the swamp goblin shaman.

Investigating suggests that a huge python has been preying on the swamp goblins and that, from the actions of the sentry, they never heard it approach. (This suggests danger by showing that the python one-hit-killed a goblin and the means by which it did so.)

Another battlefield shows that the shaman easily dispatched a group of swamp bugs.

A third area shows that the shaman physically kneeled (and maybe trembled in fear) before a pair of large scaly feet, monster type unknown.)

And so on and so forth. By having players able to look at the history of an area through tracking or word of mouth, they'll be able to discover the correlation of forces and create a threat hierarchy.

share|improve this answer

Something that may need to be considered by the GM is how the group approaches the role-playing aspect. The brawny and low-intelligent fighter-type may run headlong into danger no matter what while the rest of the party wants to have some sort of "plan of attack". While I am personally not of the opinion that a GM should hold back anything if the party does something really stupid, there are times when it is necessary to create an idea of the danger involved by having them learn by experience (ie. near-death experiences rather than character death).

I have always been of the belief that the PCs should not only learn in their specific areas of specialized knowledge, but they should learn by experience how to handle certain situations. That fighter-type that runs headlong into danger and comes face-to-face with an enemy that can easily destroy him may be so badly hurt that the party has to negotiate his release from the enemy in the hopes that he has learned something so that they can come back at a later time to defeat said enemy.

It has been mentioned above about characters not having access to stat blocks and such, which is highly reasonable unless they have specifically done research about a particular enemy. Even in that scenario, there is always the possibilty of a "exceptional" enemy that doesn't fit the mold of their studies. Again, another learning experience. That seems to be a problem with many low-level groups and encounters..."Oh, a group of goblins. We can handle that." It should get trickier as the party gains levels, even if they're?running into another group of goblins..."Oh, a group of goblins? We can handle that...wait, did that goblin just cast a fireba...?" [searing pain]

Word of mouth, local legends (about the same thing), presonal experience, and ancient tomes are my favorite ways of telegraphing danger, but there are others that are just as effective, if not more.

share|improve this answer
Thanks. I'll give a +1 for the last paragraph, but I do have to say that the rest isn't a good answer to this question: ways to teach players that the world is dangerous is a different question than what techniques I can use to convey the specific dangers of their current situation or region. (In my current game, the teaching technique is straightforward and is working fine: no dice fudging, and let them experience character death if that's what their choices lead to.) –  SevenSidedDie May 2 '12 at 21:07

If the players have any intelligence 2 stuff around, this can be an excellent way to communicate danger, whilst keeping some of the mystery. Force a ride check on anyone on a horse as they try to bolt. Have riding dogs sniff the air and whimper, or animal companions growl/snarl/trumpet a warning. The success of this sort of low level information can depend on your players "well, the DM wouldn't put us in a fight we can't handle, so I say we walk into the ambush with a readied action to kill anything that moves", but a cautious group may send in their rogue/animal companion/familiar/summon monster I/whatever to gain infromation and form a plan.

Pretty much any of the int, wis or cha based skills, knowledge or survival as the most detailed option or just a spot/search/listen otherwise, gather information etc.

If you want to introduce something that the answer should definitely be flee (and maybe fight another day), simply have it appear a long way away and do something particularly impressive. Use a high level spell or 1 hit KO something they know is powerful.

share|improve this answer

One trick I've used to make them more stressed about danger is give them time limits in the real world, similar to using a Game Clock in chess.

So for example:

  • Give them only 60 seconds to resolve their turn in combat
    • all of a sudden when you pass through an opponents' space it's a big deal; can you afford to lose 10 of your seconds resolving an attack of opportunity?
  • Tell them that in 10 minutes of play time - irrelevant of combat rounds - some major game-changer happens. (the dragon wakes up; the sorcerer finishes summoning the biggest nastiest demon)

This is great for players who tend to suffer from Analysis Paralysis.

Bear in mind, a player can still think through what they want to do while other player's have their turn, but once it's your turn the heat's on...

share|improve this answer
This is less helpful when you consider the edit, but depending on the group you could still end up with solid strategic play. –  LeguRi Oct 17 '10 at 23:39

Don't be afraid to mess with equipment; sometimes it can scare players more to break their swords and shields than to kill their comrades ;)

Seriously; I've known players who would charge Red Dragons and run from Rust Monsters.

share|improve this answer
More to the point, how would you hint that a Rust Monster is around? I'm not into scaring them—just furnishing clues and warning signs of what dangers they face if they choose to go in a particular direction. –  SevenSidedDie Oct 18 '10 at 0:44

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.