Ask Angry: Run Away, Run Away

Banner Patreon 800 x 100

Category Ask Angry 800 x 450

If you want to Ask Angry a question, see this page on how to Ask Angry!

Brendan E. asks:

How do I teach my players how strong a monster/enemy/creature is? When you look at video games, there are a lot of ways to show players the difference in power levels, but the favored method is always: “well, if they get in over their head kill them.” I’ve tried to recreate the non-lethal lessons but ultimately my party just isn’t learning, they have never run from a fight, nor do I expect them to try anytime soon. What are some solid methods for helping my party discover the gap in power level, without PKing them every time?

How do I stop these idiots from trying to fight things that are vastly more powerful than they are?

First of all, you’re all over the f$&%ing place, Brendan. You give me a one paragraph question that is “how do you communicate how powerful enemies are to players?” Fine. Easy enough. But then you include a TL;DR as if I can’t comprehend a single paragraph. And THEN, your TL;DR is a completely different question. And THEN, you fill the e-mail with a whole bunch of stuff that “didn’t work for you.” And THEN, you tell me about the campaign you’re running. And, incidentally, that little bit is important.

This is my first campaign, I’m playing 5E, we are very focused on combat and exploration. I put them on a mysterious island and tasked them with exploring it, mostly so I can avoid NPCs and junk like shopping around town. But this is where a lot of the problems come in, the Island itself has a huge variety in power levels, and it’s sometimes difficult to keep the players from getting in WAY over their heads, because I let them go where they want.

Now, I’m going to let you off the hook because at the heart of your question is something that is pretty screwed up. But that’s not your fault because (a) you’re new to this and (b) lots of GMs have this warped, screwed up thing in their head. I just don’t understand why.

This whole “how do I make my players run away from encounters that are too powerful for them” is something that comes up a lot. A LOT. And I don’t f$&%ing get it. So, let’s start there. WHY do you want encounters so powerful that the players either have to run away from them or get slaughtered by them? Seriously, I’m not being facetious here. This isn’t me being funny. Seriously. WHY? What the hell does it add to the game? How does it make the game EITHER fun OR satisfying? I mean, I’ve gone on record before as saying not every moment of the game has to be fun. That’s a terrible goal. But you want the game to ultimately be rewarding to play. So what valuable thing does a monster too strong to beat, whose only option is running away, what does that add?

Now, let me say this: it’s okay to have the occasional escape scene. I’ve done it myself. I had my players fleeing through tunnels one time from a giant, aberrant worm monster that was basically a terraforming cthuluhoid from the Far Realm. They were fleeing through a labyrinth of worm tunnels. And they knew, as they ran, they were getting themselves hopelessly lost deep in enemy territory. And they also knew they couldn’t escape unless they found ways to slow the creature down or to split up or to use the terrain. And, ultimately, they did escape. Except for one of them. Oh well.

THAT was a cool scene. But those scenes are the rare exception. You design that crap as an encounter in itself. You don’t just assume that it’s a standard consequence if the PCs wander too far into the wrong monster’s territory. Because most chases are pretty f$&%ing boring. It’s just “compare movement speeds” and occasionally “make a Constitution check to not pass out from exhaustion” and every few rounds “a thing is coming, make a saving throw to jump or dodge the thing.”

Moreover, most evasion scenes come down to this:

GM: “As you come over the hill, you see a thing you have to run away from.”
Players: “Can we fight the thing?”
GM: “If you do, you will die.”
Players: “We turn around and run away.”
GM: “The thing starts to chase, but it gives up as you run away. You ran away.”
Players: “What a heroic victory!”

First of all, there is no interesting choice involved in “if you engage this, you’ll lose no matter what.” And a role-playing game is a matter of choice. Notice the subtle use of sarcasm in my previous example to point out that the GM is basically telling the players “now, stop being heroes and run away from this thing because that is your only choice.” How is that fun? Engaging? Interesting? Satisfying? Rewarding?

I mean, it can be interesting if – as above – it leads into a chase encounter with decision points and a dramatic question and ways to win or lose and strategies the players can employ. But most of the time, it isn’t. And even if it DOES, those are only fun some of the time. After a while, the “evasive maneuvers encounter” gets stale and overdone.

And it can be interesting to have a monster that is too powerful to beat in an outright fight. One that you have to deal with in some other way. Like diplomacy, trickery, stealth, using the environment, drawing into a trap, or something else. Those are also fun encounters occasionally. And, again, they have to be carefully designed to create an alternative type of encounter.

But that isn’t what you’re talking about anyway. You’re not talking about the players stumbling onto a monster and then realizing it’s too strong and then going off and doing something else instead. And that sucks. It’s basically saying “here’s something interesting in the world, but if you interact with it, it will kill you. Now go away.”

The insidious thing is that you’re actually railroading your players when you do that. What I mean is that you’re robbing them of their agency. You’re basically putting a thing that cannot be dealt with in anyway except to turn and walk away. If you did that in any other kind of encounter, half the shrieking elitist auteur GMs on the internet would smarm at you about how terrible you are as a railroader and shun you. But when it comes to the ultra-powerful monster, they will applaud you for doing that.

“Oh look,” they will croon. “Look, how realistic your world is because it doesn’t level with the PCs. If the PCs walk in the wrong direction, they will find something they can’t deal with and then they have to run away. What a brilliant way to build a world.”

Have you considered that the reason your players don’t run away from things is that it isn’t a hell of a lot of fun to run away from things? I mean, considering the campaign you built. You basically removed everything except combat and exploration because you and your players find that s$&% fun. Good for you. You focused on what’s fun for your group. So, if the party wants to explore and fight, why would you ever want to chase them away with a too-powerful monster? “Here’s a world of things to fight and places to explore!” “Yay!” “Oops, you took a wrong turn. Stop fighting and go back the other way.” See? By putting a too-hard monster and forcing the players to run, you’re basically taking away both the fighting and the exploring.

In general, players resist running away because running away isn’t fun. And it isn’t what they are in the game to do.


Okay, so let’s say you do actually have some crazy reason for wanting to just put monsters somewhere that are too powerful for the PCs to handle. And you expect them to run away. And let’s say this reason is a better reason than “it’s realistic that way,” which is the usual bulls$&% reason GMs use to justify it. It’s entirely possible you DO have a good reason for it. How do you do it?

The first most important thing to do is JUST F$&%ING TELL YOUR PLAYERS OUTRIGHT. Stop trying to send mysterious signals. Stop trying to beat around the bush. Stop trying to be subtle. Subtlety kills players. Trust me. Explain it to your players.

“Now, players,” you say, “understand that this entire mysterious island that you are exploring is meant to provide many, many sessions of fun adventure over many months. And that means it is meant to work at many different experience levels. Some areas are more dangerous than others. And some areas are so dangerous that you will die if you are not strong enough. And you need to recognize the signs and back the f$&% out and come back when you have some more levels. Part of the game is making yourself powerful enough to deal with the most dangerous parts of the island. And I will not keep you alive if you ignore the warning signs and try to go toe-to-toe with something super powerful. But it IS possible to escape from super-powerful encounters, no matter what your Speed seems to indicate.”

Look, the thing is, it’s totally find to do the open world thing if that’s what you want to do. But you can’t expect your players to read your f$%&ing mind. Tell them that is how the game works. Tell them. Just. F%&$ing. Tell. Them. There are parts of the world that you have to be a bad enough dude or dudette to explore.

And that means you also have to be ready for their creativity. If you want to make a giant-ass island worthy of that kind of structure, that’s cool. But make sure that more powerful areas have more powerful rewards. The Desert of Terror and the Barrows of Being Torn Apart by Ghouls should have more powerful treasures than the Forest of Mostly Harmless Critters and the Plains of Just a Bunch of Wildebeasts. And that also means if the players want to try to sneak across the Barrows of Being Torn Apart by Ghouls and try to find rare and powerful treasures, you should be open to that creativity. If they cast “invisibility to undead” and sneak into a tomb and come up with some clever plans, let them have the Rod of Being Pretty Awesome in the mummy crypt.

And, see, this is where I question the whole structure of your campaign. I mean, I don’t want to be harsh. I recognize this is your first campaign and I also recognize you gave me the one paragraph elevator pitch. So don’t take this criticism the wrong way. My question is “what’s the whole point?” Is your campaign really just an island full of stuff to kill? Like Theodore Roosevelt’s attempts to kill two of every last f$&%ing thing in Africa like a psychopathic Noah? You said it’s about combat AND exploration. What, exactly, is the exploration aspect? Is the goal just to map the whole f$&%ing thing? Or are there interesting things to find?

See, by removing everything except “map the whole island and kill all the things,” you’ve robbed yourself of a lot of ways to give your game any structure. So, the players have nothing to guide them to any goal. It’s just “wander and kill.” And when you tell the players that sometimes they have to stop killing and head back, you’ve taken everything away.

If you look at those interminable open world games like Skyrim and Skyrim 40K or whatever the one that happens after the Nuclear War is called, they are just big open areas with lots of stuff to explore. But there are dungeons and quests and all sorts of odd little storylines to discover. There are little elements of structure that give you little goals here and there. Nothing too fancy or too pressing, sure. Skyrim doesn’t care whether you kill that evil dragon. Skyrum 40K doesn’t care if you kill that baby that got stolen from you at the beginning. Those are just excuses to point you in a direction. And then you discover other excuses to do other things. And all the little things are extrinsically and intrinsically rewarding.

Moreover, those little things – quests and NPCs and stuff – also provide the GM a way to say things like “the terrible Canyons of Deepening Death? Yeah, that’s way dangerous… I wouldn’t adventure there myself unless I was oh, say, 12th level.” I mean, not in those exact words. But they provided a way for the players to learn about the world OTHER THAN blundering around it getting killed by the too powerful things they had no way of knowing were around the corner.

So, here’s the thing: if you really are running an “explore the world, kill the things, take the treasure” type game, what you’re actually running is a very old school form of D&D dungeon crawler. In those, there was a big ole underground complex filled with monsters, traps, and treasure. The GM was constantly adding more rooms and monsters and traps and treasure, staying ahead of the PCs. And the PCs kept plundering deeper and deeper into the dungeon. And that’s cool. That’s a fine, fun way to play and run the game. I’m all for it.


The dungeon was kind of important to that. The walls and halls and rooms and everything? They provided structure and constraints. The party could search in any direction where there was a hall or room or whatever, but they couldn’t just start moving through walls. And the players understood – because it was part of the game assumptions – that the deeper one traveled into the dungeon, the more powerful and dangerous the monsters became. They couldn’t stumble too deep into too powerful an area without transitioning through some areas that felt “kind of tough” and “okay, things are getting really strong” first.

With a wide open landscape, you don’t have that sort of structure to signal to the players when things are getting too powerful. Which means, the players have no way of recognizing transitions from one power level to another. I mean, you can do it. You just have to be smarter about it.

So, you draw your map of your big ole island. And instead of making a big blank map with features, think of it like a patchwork of different terrains. You have the Plains of Just Starting Out. North of there, the Forest of Slightly Tougher Stuff and south of there the Hills of Haunted Tombs Perfect for Novice Adventurers. Beyond the forest, there’s the Swamp of Pretty Dangerous Stuff. And so on. Basically, make your map a patchwork of geographical regions. Then, assign each region a level of challenge. Or a range. And then make sure that no region borders any region that is more than one power level above or below it. So, the Plains are Level 1-2. The bordering Hills and Forest are level 2-3 or 3-4. Whatever. The swamp beyond the Forest is level 5-6. That way, the heroes can’t transition across too many power levels without things ramping up.

In addition, the difficulty of traversing the terrain should match up with the level of challenge. The deserts and mountain peaks should be high levels. Plains, meadows, and light forests should be low levels. Hills and swamps and badlands should be somewhere in between. Dense jungle should be even higher. That way, the terrain obstacles and challenges level up along with the monsters. The Swamp of Terrible Things will have level 5-6 monsters and quicksand that can threaten level 5 characters and an ancient ruined temple with level 6 monsters and traps and it will have treasure appropriate to level 5-6 characters.

If you combine that patchwork leveled approach with a few NPC settlements, maps, clues, and other quests, you can point the PCs to quests of their level in specific regions. Or they can explore on their own and you can rest assured that the difficulty won’t ever shock them to the point where they have to die to get the point that they should turn around.

The thing is, many GMs assume that the self-paced sandbox gets them off the hook for structure. But EVERY game needs structure if it’s going to work. The self-paced sandbox just means the structure is less overt and more hidden. And that means it actually takes MORE work for the GM.

Now, I realize I haven’t really answered your actual question: how do you get your PCs to run away. Instead, I got distracted by telling you how you should have structured your whole campaign if you insist on trying to force your PCs to run away and questioning whether them running away is really worth anything anyway. Teaching the players to run away involves, first and foremost, telling them flat out you expect that they will have to run away sometimes or you will f$&%ing kill them and tell them it’s their own damned fault.

But once you do that, you need ways to help the players tell the difference between something they should run away from and something they shouldn’t. And that depends on how you want to play the whole thing. Frankly, when I give the flavor text, I like to say flat-out that the monster is f$&%ing powerful. Like, I go over the top. “The thing is huge, with rippling muscles. It’s clear it could rip any of you apart with its bare hands. And its thick scales look like they could turn aside the sharpest blades. I mean, the thing looks like it would wreck you.” I mean, hell, if you’re willing to create an encounter where the players’ only choice is to run away, you might as well tell them flat out. Anything less is just outright killing them.

See, here’s the problem with D&D. It’s pretty unforgiving when the players face something can’t beat. A monster even a few levels above the party can seriously destroy a PC in a single round. So, D&D doesn’t really lend itself to a “feeling it out” approach. And D&D also doesn’t have a lot of good ways to retreat. Especially when you are trying to drag away the corpse of a fallen comrade. So, again, if you’re willing to let the players blunder into too-strong encounters, if you don’t just tell them flat out that the encounter is too strong, you’re murdering a PC. And then the other players get to play the game of “leaving a fallen comrade to die.” Which is not a game players like to play.

The end result is this: unless you’re willing to restructure your campaign and build it around the idea of patchwork exploration with some elements that allow you to fiddle with the structure OR just cram it in a giant dungeon, the only option you really have is just to tell your players “run away or someone will die.” Flat out. That’s just how it is. And then you’re going to have to figure out how to handle the retreat because D&D doesn’t even consider Evasion and Retreat to be a thing anymore. Or just stop throwing encounters at the party they can’t beat. That’s probably the easiest fix. Just let them kill things and take their stuff. That seems to be what they want.

Just don’t get disheartened. Keep running games. At least you have me. I didn’t have me and I had to spend years figuring this s$&% out. You lucky bastard.

Banner Patreon 800 x 100

111 thoughts on “Ask Angry: Run Away, Run Away

  1. What would you consider to be good reasons to have enemies at a much higher level than the PC’s? My campaign is about the PC’s going on missions and increasing their power level, and the world is filled to the brim with characters more powerful than them. Is “any sense of immersion would be broken instantly” a valid reason for higher leveled enemies?

    • I would personally consider story to be the one good reason. Throwing a really strong enemy at your party that doesn’t kill them, but only immobilizes them and tells them to stop interfering with his plans is quite a nice motivator, basically telling the players “there’s more to come, and here’s your goal – beat THAT guy”
      But that’s about it, and you shouldn’t overdo it.

      • That’s my line of thinking as well. Aside from the story reasons, nothing serves as a better marker for power progression than being able to beat foes you couldn’t even touch before. Plus a big part of my game is teamwork and clever choices allowing the players to beat an opponent that’s stronger than them. Spent a lot of time making sure the magic system is internally consistent and allows ways to bypass abilities so that the players can do that, which would be meaningless if they could just deck the guy in the head and win with no real effort.

    • I can think of two basic reasons why you might want to include higher level creatures:

      1. World building. Your world has creatures in it, not all of which are the same level. The existence of a particular creature is an element of your world. It doesn’t mean the PCs have to encounter it, and they may well know its reputation in advance and know to avoid it. But it has a reason for being there. It may even have a role in the campaign, as a potential patron, source of information, a recurring villain (in which case the villain needs a reason to not kill the PCs when they inevitably attack it at too low a level), or the campaign boss (where confronting it is something the PCs know they need to work towards – if they decide to confront Doresain at first level, they should expect to receive a Darwin award).

      2. You want to make the decision as to whether to fight a meaningful one and/or you want to de-incentivise combat as an encounter option (to emphasize intrigue, subterfuge or the talky-talky bits, for example). Note this is a significant break from the default assumptions in third through fifth editions, which treat combat as the primary means of advancement. You may need to change the way you reward experience, and you certainly need to telegraph both the fact that you are choosing this as a playstyle, and the relative power of the monsters.

      I thought of others, but they all seemed to be variations or combinations of these.

      • The EXP for a creature in every edition of D&D is awarded when you “get past the encounter” the creature, whether you beat it to death or convince it to join your party to kill the BBEG.

        You do not need to change how you award EXP, just how you view encounters and EXP.

        • You are mistaken. In earlier editions, the largest part of experience came from the accumulation of treasure, not from combat. Combat was therefore high risk-low reward. If you could get the loot without combat, or avoid combat with monsters that didn’t have treasure, you were ahead of the game.

          With the change to combat as the primary method of obtaining XPs, combat had much less of a downside. It is much harder to encourage players not to fight if you reward them more heavily if they do.

    • As a suggestion, if you want to keep your players immersed in this world of powerful characters, have the powerful characters tag along on the mission, fulfilling objective “B”. Say they’re invading a haunted castle and fighting waves of ghouls to retrieve a map. Have a powerful character go along or show up to rescue a prisoner or some other excuse. When the players are struggling to kill the ghouls and at some point they look out the window into the courtyard and they see Hamfist SwordSwinger (or a werewolf or something if by powerful characters you mean enemies too) kill 3 ghouls with one swing, they will still be immersed in powerful characters.

      • Uh, no. No no no. Treat more powerful NPCs/enemies with restraint. I can personally count on one hand the number of times a higher-level NPC/enemy added something to the campaign in over 2 decades of RPing. The rest of the time, it’s handled poorly and I end up wondering, “Why don’t they just kill us now?” or “Why aren’t they saving the world instead of us?”

        As for a world filled to the brim with more powerful creatures, one question comes to mind: how did THEY all get that powerful without being cut down by some random more-powerful creature they crossed paths with? Did they all just have precognition about what encounters they could and couldn’t handle? Are they just the lucky few who didn’t get TPKed through bad luck or a single poor choice?

    • Usually when my players encouter things far too powerful for them, it’s because they did something very stupid.

      Often times the powerful being is not hostile, but the players managed to anger it one way or another.

      Think of it that way: The King wants to meet them, and they treat the king like garbage, when warned by said king, they dare him to do something about it. Now there is no way they can take the king’s guards, and they end up in an battle against far more powerful than them…

      But that impossible fight was not the point of the scene, they just triggered it because they pushed the scene in that direction.

      • At that point, I’d ask WHY they’re treating the King like garbage. Unless the players just like causing trouble, odds are something about the King has ticked them off. Something the GM hasn’t realized. And the GM/King demanding they show the King some respect just gets under their skin further. So they start derailing.

        • That was an example.

          The point is, forces more powerful than them exists, and i’m pretty sure that in any campaign they encounter them quite often. (Like the City Guards of a Metropolis)

          The players sometimes make bad choices, and those bad choices could have as a logical consequence to pit them against a power they can’t match with strenght alone.

  2. What do you do if you’ve designed a balanced encounter, but the battle isn’t going in the partys favour? (for any reason)

    Just flat out tell the party to run away and re-group?

    • I don’t think so. You build your encounters fair and play them to win, like the monsters would. Don’t pull punches. It’s up to the players to make choices. If the encounter is fair and they are taking a beating (due to chance or poor strategy or whatever), they must acknowledge that and consider a retreat.

      Your job, as a DM, is making sure the players know they CAN retreat. Like, make it clear, from the start of the game, that this is always an option. Sometimes the monsters will chase, sometimes they will not, and sometimes rhe retreat will be impossible (and the players should know so). But you tell them that they CAN retreat. You never tell them that they MUST retreat. That is robbing them of their choices.

      • The GM usually has a better grasp on how the party is doing overall than the party itself does. As Angry has said in other articles, it’s about the way you communicate those things. Like, “Oh, your warrior looks like he’s just barely making it right now. The enemy might want to take advantage of that.” The players are more likely to think, “right. the warrior is the worst for health right now, and he’s at a disadvantage against these flying monsters he can’t hit with his sword. Maybe we’d better protect him until we can get our healer across the field.” Etc. If it comes time that a retreat is the only way, that’s when you tell them so.
        This is especially true of players who’ve never played together. They could all be the best damned players in the world, but they have yet to find that synergy that comes from playing with each other, even just a handful of times.

        • The GM usually has a better grasp on how the party is doing overall than the party itself does… until you play final encounter for lvl30 group in DnD 4e. I still remember that long fight where players were slowly killing the boss and all his minions and then when the boss star spawn remain alone and did nothing new for 2 turns one of players said “Ït seems that is really all, maybe we can start to use daily powers and not store them for any surprise”… and that really hit me, because players succeeded to bluff me as GM that combat is well balanced, whereas it was too easy and they were just expecting more tricks in my (ok, in boss’ sleeves.

          • I’m only on board with that if you’re using monsters and such from core rule books. Angry’s method of having “boss” monsters go through phases really makes a huge difference, especially for those monsters that have all of 2 – 4 things they can do.

          • Frousteleous: I lead that few years ago. There was no single monster according MM as I recrafted all (at least decreased HP and increased attack and damage to improve action and threat).

            This boss ironically was changing through phases, although reason was different (to play with theme of unstable starspawn) and I had no idea some Angry GM exist :). And no, it did not solve it at all as it had no effect on the fact the players convinced me they lack resources for any major actions, whereas they were holding them as last reserves if really needed and were expected even bigger twist (like when you defeat final boss, you find out he was protecting the world from even greater deamon which now broke from his prison).

    • Angry actually has a great article that talks about exactly this problem. Basically, when an encounter is not going the party’s way, D&D encourages them, on a system level, to double down and fight harder. And this happens in part because there is no clear indication of a state of emergency until one of the party members is down for the count. Angry spends most of the article explaining the problem and ends with a neat rules hack meant to address it. Personally, I used his hack as inspiration to design my own that addresses the same problems, and it has worked fantastically for me. My players don’t often end up in unwinnae situations because they’re smart with their resources and I don’t usually structure my adventures in a way that they’re likely to end up in unwinnable situations. But, sometimes shit happens, and when it does, there’s a clear indication that it’s time to consider defensive tactics and maybe retreat.

          • Basically, I did away with the hp pool vs. sp pool, and just said when you reach 0hp you remain conscious but become unstable (you’re rolling death saving throws) and you have the Dispirited Condition Angry wrote (disadvantage on attack rolls and treat your exhaustion level as 1 higher). The first time you fail a death saving throw (or get hit by an attack, thereby giving you a death save failure), you fall unconscious.

          • That’s a great idea! I felt the SP/HP buffer zone was a bit too strong, but making it so that you are already dying when crossing the threshold is a nice compromise. You lose some of the drama of a big hit knocking you out instantly, but escape becomes a bit easier. I feel like there could be a constitution save right at the moment of the hit to determine if you stay conscious, but then the point of the system would be partly lost. I’ll have to test this, thanks!

          • I used the sp pool Angry talked about but I heavily penalized running out of it. If my players ever run out of sp I give them a lingering injury, on top of having failed death saves add levels of exhaustion (so 6 cumulative saves kills). Since I’ve started doing this those with low health have really started getting out of the way and nobody plays Sean of the Dead and heals for as little as possible every time someone goes down. The extra health keeps them up a little longer and they now retreat when they should if they screw up a fight.

      • Based on the article I designed my own homebrew, in short:
        – if you drop to 0 HP, you determine if you go down willingly (and by that receive 1 death save success for free) or keep fighting.
        – If you keep fighting, you have disadvantage on your actions and enemies have advantage for Saves against your abilities. You continue to do death saves as normal (if you are not stabilized) at the end of each of your turns.
        – If you keep fighting and receive any damage, it counts as failed death save. The same if you fail death save on the end of turn. In both cases you go unconscious and receive no free success.
        – you die on 4 failures and stabilize on 4 successes

        It is tested and sometimes it worked as intended (it gave opportunity to retreat). Sometimes I personally encourage to go down willingly (when there are still threatening enemies acting before the character). And in two cases the players keep pushing and attack. Once it saved the party as the character hit the boss despite the disadvantage with good damage score, once it lead to utter failure…

  3. Could you do an article on Evasion and Retreat? Younguns like myself haven’t learned how that works, because we’ve only played 5e

    • To be a bit more helpful… it really depends on the monster’s motivation. If the PCs retreat will the monster follow? Why?

      If the monster is wanting to protect something then its motivation to follow is not there. If the monster is hungry or under orders to kill anything that attacks then sure – it’ll chase the PCs (and now you have a chase scene to run), but there is still a limit to how far the monster will pursue.

      Think about the motivation of your monsters when determining how easy it will be for the PCs to retreat.

  4. Do you think it’s okay for the PCs to stumble upon a big bad evil guy they’ll have to fight later on (but have no hope of defeating at the moment) as a way of showing them what they will need to prepare for?

    • Well, you want to plan that out, though, right? You don’t just have them smack into the guy and have him murder somebody to signal to the party that they should run and then have them fail to run. Instead, you want him to do what the BBEG does in countless films and games and novels, which is, show up, monologue at them, bask in his clear superiority, then set his minions on the party and go off to do something more worthy of his time. If they take a swing at him they just cut his monologue short, he laughs in their faces and leaves.

      That’s not the same as “stumbling on” him, though, and if it’s done right it doesn’t have the problems Angry outlines about why-not-to-have-too-hard-encounters.

      • It’s sort of an easy out, but if you’re looking for something more story driven, the BBEG is a great way of killing an NPC the party is fond of. BUT you can’t just go out and outright kill the NPC. The PCs do need SOME chance of saving the NPC. But it’s a great motivator.

  5. Longtime lurker, first time poster. Thanks for another great article!

    I love this. This is basically how I’ve structured my Pathfinder campaign right now, complete with challenge rating zone bubbles drawn over my version of the map. The PCs are also in charge of an organization that allows them to send groups of lower level NPCs (with PC classes) to areas to scout for them, so they have a general idea of the challenege and some tidbits about the terrain and monster types in that area once the scouting party returns (or the scouts don’t return and that says something too). They know the levels and classes of the scouts, so it’s easy for them to compare, so they’re rarely in the situation of being in over their heads, but it does happen sometimes. I gave the PCs their own version of the map too, which they’ve wisely used to label scouting reports, hazards, and other known things without also having to be cartographers.

    When they’re in over their heads, in order to allow them to feel out monsters, I’ve added two mechanics. The first is a slightly modified version of angry’s Fighting Spirit system, which allows me to hit a PC for all I’m worth without murdering them. The second mechanic I added was a retreat action. The retreat action can be declared by anyone on their turn as a move action. At that time all allies may choose to join the retreat or not, and any enemies may choose to join a pursuit or not. For everyone part of either the retreat or pursuit, combat is over and a new encounter has begun, which I settle differently based on how intense I want it to be. Running away with the retreat action does not provoke attacks of opportunity unless you’re flanked.

    That usually does the trick. If the fighter’s FS is blown out on round one, the whole party is immediately saying, “We’re in over our head, we need to get out.” And then someone calls the retreat and they’re immediately out of there and into a new encounter. Or, as in one scene, they need to stay in order to complete a task, so they switch to defensive survival stretegies and then once finished with the task retreat immediately. Great encounter that one was. I had the fighter yelling, “Hurry up with that!” and the wizard yelling back, “You can’t rush this!” and everyone made it out alive and grinning like fools.

    • You know what? This comment made me read the article about FS before my schedule (I’m trying to keep some consistence in my readings) AND decide to use it, whatever it was (because since Angry made it, it would obviously be awesome). And it IS awesome. And I will be using it.

      Also, I would love to know 2 things:

      1 – Why not just use the disengage action? There is any specific reason for this retreat action to use movement insted of the normal action? What if a player says “I retreat” and in the following turn changes his mind? I am not against your idea, I just didn’t understand very well the reasons behind it.

      2 – What was your modification to the FS system Angry presented?

      • 1- I haven’t played 5e, but in Pathfinder the withdraw action doesn’t do what I want it to do. Also it didn’t solve the problem of the PCs getting tripped up over the action economy. If there’s going to be a retreat, we have to drop initiative because the combat is now resolved and a new encounter is beginning. If the player changes his/her mind (my table is about 50/50 male/female), they have left the arena and will need to move again to re-engage. It hasn’t happened yet. Once someone calls a retreat,I give them time to discuss it at the table because they never want to leave some of them behind. I used the move action to give the retreat caller an option for a parting shot as a standard action. I had a PC drop a wall of fire after calling a retreat to cut off the pursuit one time. Only the one callig the retreat gets that parting shot, so sometimes they wait for the right player’s turn to take best advantage, in the meantime working hard to get out of flanking situations via the normal withdraw action.

        2- I will admit fully that I am not as good at the gamey math as angry is. All I really changed was how to calculate FS and HP at each level. After first (where they still get HP=FS), I have them add their Con mod (if positive) to HP and their class hit die to FS for each level. A negative Con mod applies to FS at each level. Technically they have the same amount of hit points as they normally would, plus bonus ones at first level, and split into two pools. Temp HP provides temp FS, but feats like Toughness add to HP not FS. Due to scheduling, not all my players can make it every week, so I also modified the Heal skill to provide minor FS repairs with a heal check. Like so:

        First Aid: Spend a full-round action that provokes an attack of opportunity to make a heal check on one target. For each point of the check beyond 15, heal one FS for that target. While using First Aid you threaten no squares until your next turn. A target can only benefit from first aid 1/day. A character may spend one hour to take 20 on this check.

        I love the FS system and think it has a lot of other potential. For example, I let coup de grace attacks bypass FS and go straight to HP. I feel like other bypass abilities would be cool, like sneaks or death attacks, but I haven’t tried it yet because allowing it undermines the purpose of FS, which is to alert the players to an emergency.

        • You said in your comment that it was a Pathfinder campaign and I forgot and just assumed it was 5e. Sorry about that. Now some parts of your comment make more sense. =P

          I’m thinking myself about critical hits and some poisons (not all, or I probably would need to revise a lot of game math instead just a little) bypassing FS and going directly to HP. As you said, FS alerts the players to the emergency of the situation. I believe if something suddenly affects HP whitout touching FS they will realise things can get pretty dangerous. Worst case scenario, they have full FS and start dying. Still thinking about some dangerous spells and death effects would behave, but I probably won’t be changing much about them.
          (If I do this, critical damage would not be multiplied, only bypass their FS. Maybe this work a little better on 5e, because all weapons cause x2 on a critical, there are no x3, x4 and etc weapons).

          Understood your reason for the retreat action, and I think it is a good reason. I specially like how it ends the encounter (or is supposed to) and is (mostly) independent of other combat rules.

          Thanks for the insights! I’m sure my players will approve this.

  6. What about using “unbeatable” monsters as a form of gating progression? “You can’t access this area unless/until you can defeat its guardian” kind of thing.

    • And how exactly do you expect the players to realize they aren’t going to be able to kill that guardian? By wiping? Because that’s what they’re gonna do.

      • You could do it with specific weaknesses. The problem with beefgates that are just a much-higher level than the party is that level is a vague mechanic that’s hard to describe in-game, thus the party never has any idea whether they’re equipped to tackle the beefgate until they actually try to fight the beefgate (and likely die doing it).

        But if the beefgate has a weakness to a specific type of weapon, then the requirements are crystal-clear: once you get said weapons, you can tackle that area. The Western Woods are filled with werewolves? Clearly you need to acquire silver weapons before venturing into it. An excursion to Trollville? Flaming/acid weapons and similar spells. Going into devil-infested lands where they regenerate from any attack aside from holy swords? Get some holy swords.

        You could even have the werewolves guarding the tomb with the holy swords they need to fight the devils. You could do something similar with particular spells, rituals, etc.

    • Again, this is along the same thinking as the old dungeon crawls. What’s the real purpose of putting up gates? This starts to feel a little video-game-ish, too. You’ve got an infinite world driven by your head. Game designers have to use stuff like that as a cheap trick to keep you playing longer or at least enough to “deserve” going to that area. So why have them?

      • I wouldn’t overuse it, but it would make for some decent motivation. You, as the DM, basically tell your players “Here is the world. You can do anything you want, and you can go anywhere you want. Except in that cave, where the Guardian of Super-Murder lives. The Guardian will super murder you if you go in there, so you can’t go there….yet.”

        That would really put in itch in the players to see what is in that cave. The trick though is finding a way for them to know the Guardian super murders people without super murdering them. I’d probably make it a giant stone statue that is immune to non-magical attacks “Your sword snaps in two when you hit the guardian, and you realize it will take more than mortal steel to hurt this thing.” Then give it a giant health pool and make it incredibly slow. It literally does nothing the first round, slowly begins to move the second round. Make it clear to the players that it is speeding up, so that it is getting harder and harder to dodge its barn sized fists, and make it clear that their magical missiles are doing jack.

        Then be prepared for them to come up with a clever plan of pushing it over a cliff or causing a cave in or something and getting the artifacts you placed in there even though they are only level 5. It’s only fair.

      • Well, the old dungeon crawls had conventions that constituted “soft” gates, the primary one being that tougher monsters would be found on lower levels. When the players encountered stairs going down, they had to decide whether they were ready to take the greater risks (with commensurately greater rewards) associated with voluntarily entering a more dangerous area.

        However, soft gates were easier to use with earlier editions of the game, because (a) most of your XPs were earned from treasure, not combat, so that there was a disincentive to default to combat as an action, and smart players were always looking to avoid it; and (b) character generation took 10 minutes or less, so the downside of making a mistake was much lower, and games could get away with being “Nintendo hard”.

        In a game where combat is the default assumption, and is the primary method of advancement, and where character death knocks the player out for a whole session, it takes skill on the part of GMs and players to make soft gates work, and not everyone has that out of the gate.

        “Hard” gates, like doors with unpickable locks that can only be opened if you find the right McGuffin, are a form of rails that protect players from poor telegraphing by the GM, but also from their own poor decisions. I think those remove player agency.

        But a lot of the gates Angry uses in his megadungeon, for example, are in-between. A door with a lock that is just very hard to pick, or water filled passages that are very difficult to pass until certain spells or magic items are unlocked, are in a different category. The PCs can probably find a way around them if they are determined to do so, but the obstacles serve as an effective hint that “you aren’t ready for this yet, proceed at your own risk”.

        For that matter, he probably doesn’t even need to go through the exercise of figuring out how far a low level character can swim underwater, for example, in constructing a gate. When faced with the choice between taking an underwater passage, or completing the exploration of parts of the dungeon that are not underwater, players should know (or learn quickly) that taking the harder road entails greater risk, but also has greater rewards associated with it. Which goes back to telegraphing, which is what the back half of this article is all about.

        • My main point with gates, though, was “why?” A “gate” in this sense has to have a very good reason to be put up, and yes there are always GOOD ways of doing that.

          I throw lots of big, ridiculously powered opponents at my players with fair regularity. BUT, i always give them environmental factors to use. One of my first big “boss” monsters was a giant wolf creature scaling the cliff. I knew a certain player had a push ability. That ability was used, throwing the wolf off the cliff. He stumbled his way back up it, now at half health. Now, the fight was an actually reasonable one, and the players felt smart for using their surroundings against their enemies.

          Be it gate, or big bad wolf, there’s always a better way to do something.

          • Love it, had an example just play out with my group last night. Was running a WotC module for my group, starting to realize how bad an idea that was. (Horde, if you must know.)

            Anyways, they stumble upon these guards in a small room, and have a surprise round.

            Fighter is first on the order, backs off after taking to much damage last fight.

            Sorcerer was next, caught them all with burning hands, killed a bunch.

            Warlock mops up a few with Shatter, turned most everything to dust in the room, save one guard.

            Tempest cleric, bless this man, decides to use gust of wind to fan the flames from burning hands catching most the room on fire and ends up getting this last guard.

            They then realized atomized iron, wood, cloth, and people with the right air ratio basically made a bomb. After the fact. Nearly killed the party, but the point was that using the world itself to point out cool things the party can do lets them decide what challenges they can face.

            And as a real life example, would you run up to a bull elephant with a sword? Maybe. Would you run up to it after you watched it shove over a tree and gore another elephant shortly after? Maybe still, but you got a warning.

            Use your world to broadcast to the part “Hey, common sense says this may just get us killed.”

            Sorry for the ramble.

  7. Great article, Angry!

    I’d like to throw in some thoughts, if you don’t mind, on how to deal with this kind of thing:

    1. Why would you put in an encounter that is too powerful? Sometimes you have to railroad the players a bit. If they are on a quest/mission, and time is of the essence, why not have 3 Cyclops chasing your 2nd level characters? The character see them at a long enough range that even the cyclops can’t hit them with their thrown rocks. If they aren’t smart enough to get moving, and stay out of the Cyclops’s territory, they deserve what’s coming to them.

    2. What to do if the characters want to stop and fight, anyway? One option is the let the stoopid blokes die, but that isn’t fun. There are a couple other options. One is to have smart monsters that realize the low level adventures don’t have anything they want, or if they do, are willing to let the characters live in trade. “You give us your horses, we let you live. Horse meat tastes good. Elf meat is stringy and tasteless.” (or) “We’re too busy with our own business, we don’t have time to deal with the rats living in our cave. Take care of those rats for us and we’ll owe you a favor in the future.” Another is for the players to have a hidden ally. (I’ll get into this in a bit.)

    3. What if the characters are in what you thought would be a balanced fight, but because either you miscalculated, or the characters have no sense of tactics? Two options: Hidden ally again, or have some fun with the monsters. Make them into bad tacticians or, better yet, take advantage of a bad roll (behind your screen) and let them snatch defeat from the jaws of victory. “The gnoll archer puts an arrow into his ally’s back, killing him. The surviving melee gnolls turn on the archers and attack them. Now you, the players, have to decide which side to support!”

    In last Saturday’s session, I pulled the hidden ally trick. The party was facing a pack of wolves led by some dire wolves. One or the other wouldn’t have been a problem. Both at the same time was, especially since all but one in the party epically failed their perception checks. Three of the five characters were wounded in the surprise round! The battle wasn’t going well, even though I was keeping the four dire wolves back for the first few rounds. Suddenly, three (regular) wolves fell asleep! The dire wolves, seeing this, joined the battle. Then suddenly, a dragon appeared in their midst! Two of the dire wolves backed off (missing their INT check vs the illusion.) This was enough to turn the tide of battle in favor of the players.

    At the end of the battle, one of the pixies they had met and parlayed with in the previous encounter had decided to follow them, invisibly, for a few days. The pixie was responsible for helping the party, but note that she didn’t kill any of her fellow woodland creatures (corrupted though they were.) She left the actual killing to the party.

    The point is – there are a lot of tools in your arsenal as a DM to help balance fights, communicate to the players, and generally keep things fun and give those players the options they crave in the RP experience. And remember, not everything a party encounters along the way needs to be combat.

    • I’m really not a fan. By continually saving your PCs from the consequences of their actions, you teach them not to be concerned about their actions. It robs them of the tension that comes when facing a real chance to lose, and the sense of victory when they overcome that.

      Yes, even against the wolves. Bad rolls happen, and it’s the PCs’ job to be prepared for when they do.

      • I get what you’re saying, and one character did in fact get dropped to 0 hit points during the battle – only to be saved by the cleric on the next round. I don’t advocate using this as a continual thing to rescue the PCs from their bad decisions. However, as GM, I reserve the right to rescue the PCs from *MY* bad decisions.

        The other factor to consider is where we’re at in the story arc and what the purpose of the encounter is. This particular battle was not meant to be a “boss fight”, but an easy/moderate level challenge to help the PCs toward their 4th level benchmark. My mistake was miscalculating the relative power of properly-used pack tactics. On the surprise round, 6 wolves attacking 5 party members. Had the wolves gone one-on-one against the PCs, only two of the six would have had advantage on their surprise-round attack roll, the two attacking one PC. Instead, I paired up two wolves each against 3 PCs, giving all six advantage on their attacks. Wham! Dice rolls, and all 3 PCs lost half or more of their hit points in the first round, before they had the opportunity to act. Right away, this put the cleric into heal/rescue mode, effectively taking her out of the fight. So now there’s 4 fighters (3 of them severely wounded) against 10 opponents, and the 4 dire wolves hadn’t even joined the fight yet. At 3rd level, they’re still limited to 1 attack per action and only one of the PCs had a chance of one-shot killing a wolf on his attack.

        By the numbers, dire wolves are CR 1 and the the regular wolves are CR 0.25. The total CR is 3.5, against a party of five 3rd level characters. Not a pushover, but definitely not in the “boss fight” range.

        On the other hand, having the pixie help in the battle did affect the overall experience earned, because I included her when dividing the total exp gain per character.

        • I think you’re calculating the overall CR wrong here. Correct me if I have the numbers wrong, but, you have:
          6x Wolves – 1/4 CR worth 50 exp each, for 300 total exp.
          4x Dire Wolves – 1 CR worth 200 each, for 800 total exp.

          This gives us a total reward exp of 1,100. However, since we have 10 combatants, the difficulty of this is different. You have to multiply it by 2.5 (3x for number of enemies, reduced one on the scale to 2.5 for having more than 4 party members).

          That gives you an encounter difficulty of 2750 “EXP”, and an encounter challenge rating of 6. A *deadly* encounter for a party of 5 level 3 PCs tops out at 2000 “EXP”, so just be aware that this encounter was way above your PCs pay grade.

  8. I feel I have to get my 2 cents in about how to make encounters players are meant to run away from interesting.

    Because just plopping down a high level monster out of the monster manual and saying, this will 2 shot you, you should run from it, does indeed create boring as shit encounters.

    When you as a DM introduce a monster that players are supposed to run from, the hit points, attack rolls and damage of that monster are no longer useful information. Those are all stats that where created for the purpose of providing a challenge to appropriately leveled players, not for providing a fun encounter for players that are running away.

    When you as a DM decide that the players MUST run from this encounter, the actual power of the monster no longer matters. The monster is just set dressing at that point, what matters is the design of a cool “running away encounter” you should ignore the stats in the monster manual and look in the DMG instead.

    The way you handle it is no different from how you would handle the players running over a bridge that is crumbling behind them, or a tunnel that is filling up with magma behind them.
    The way to signal to players that they cannot fight this monster is to have the monster attack the floor and make it so the players are forced to move back because they are *literally losing ground*.

    Whether it is a dragon setting the forest floor on fire, or a pitchfork wielding riot chasing players through narrow city streets.
    What you basically have is not a monster, but a moving trap, each turn the trap moves and marks a new area as fucked in particular and if any players end their turn there, they get hit by the trap.
    The save for the trap is level appropriate, though the damage depends, the “do not stand here” area should be pretty fucking obvious so you can scale it up quite a bit.

    Angry described a chase sequence with a giant terraforming worm, that made for a good encounter, is pretty much that, some tunnels that set the stage and offer multiple paths, a clearly audible/visible giant mouth of death eating the ground behind the players and the expectation that the players think on their feet and come up with some clever ways to slow down the worm while they run.

    • Exactly. As Angry has mentioned elsewhere, running an encounter like that through the initiative system is guaranteed to end badly for the PCs. Run it narratively and let them have a brush with danger in the same way they’d deal with a collapsing dungeon or going over a waterfall. Monsters at that level are like a force of nature.

    • Yeah, I’ve been planning a Changeling: The Lost chronicle (got players to sign on, finally), and I’ve made sure to drill it in their heads that some monsters might have combat stats of “Ha, ha, no. Run. Just run.” It is a horror game. Fortunately, the game system does have rules for chases, and I had a similar thought for escape/evasion encounters, treating the monster like a mobile, invulnerable hazard.

      Some indicators I have in mind for such unbeatable monsters:
      1) Size. If it’s really big, it’s generally not going to be a good idea to fight it, at least not without heavy artillery and a plan.
      2) In Changeling, high Wyrd (magically powerful) creatures influence the Hedge. So if the Hedge abruptly turns into a winter wonderland, players know there’s a very powerful ice monster in the vicinity.
      3) Literally chewing the scenery: You brought it up for good reason. Destroying the surroundings is always a good indicator of a monster’s raw strength.
      4) Ignoring the PCs: I had one idea for a scene to make the players feel appropriately small: One of the villains is a giant bird monster, and I’d have him stomping through the Hedge, but he wouldn’t be after the PCs. Even if they were stupid enough to attack his toes. He’d have more important business to attend to. I won’t go into details on how exactly he can be killed, but it won’t be through combat.

      The general idea behind having these sorts of monsters on occasion is to encourage players to use finesse, stealth, diplomacy, and deception by default. Open combat is the risky option.

      • One of the greatest things I’ve ever read is how to properly use the Tarasque in your game. For those who are unfamiliar (look it up, there’s one in every edition), the Tarasque is this huge, unstoppable, godzilla-like beast.

        The thing is not to treat the giant monster as a combatant, but like a force of nature. Do you fight a tornado? No. But you do have to manage to get away from it, to get your family or fellow citizens to safety. This can be as fun/challenging as you;d like it to be. No different than a giant monster of insane proportions.

    • Gotta hand it to you Zwets, a top notch way of dealing with it, I assure you. And I whole heartedly agree that huge monsters are just set pieces most of the time (assuming they are for the purpose of the article.)

      Yes, a level three party of rogues can fight an ancient blue dragon.
      Theoretically, they can even win. Assuming we approach infinite numbers of different trials.
      But players are going to be players in the end.

  9. Hi Angry! Great post. Not that I want to encourage you to be wordy, but I think the section on actual retreat in combat could use expansion… in my experience DMs (and authors) don’t grasp why it doesn’t work. Whether in a reasonable combat where the dice roll poorly or in a fight they aren’t meant to defeat, the party almost always is engaged in the encounter to some extent. One or more PCs will be further in the room, there might be some goal some PCs really wanted to accomplish (grab loot/lore, investigate something interesting, recover their favorite thrown weapon they tossed last round, etc.). And, the identification of the threat as being overwhelming is usually learned through at least one PC (but usually more) either going unconscious or being within a blow of dropping.

    All of that, combined with the vagaries of initiative order, make planning a retreat a joke. One player will see the reality and start convincing the others to retreat. 1-2 players listen and start retreating. But, one might either have bad luck (they are hit before they can retreat), or they don’t immediately retreat (they go for their goal, they decide their character will tank for a round to let another escape, etc.), or they just plain disagree with the plan (“I think we can do this!”). Whatever the exact reason, 99% of the time I see a retreat there is one of these moments where a PC isn’t retreating and either they or another will then drop as a result of this. Once a PC drops during the retreat, it all falls apart. Now either the party leaves some PCs to die (almost never happens), or some members go back to help the PC. Because this is a very challenging situation, the PC going back almost always drops. The retreat fails.

    I’ve seen this tons of times in organized play. Sometimes it is in “untiered encounters” (meaning the challenge level isn’t at all part of a mathematical calculation of player-tier… it is off the charts) where the party is meant to flee, and sometimes it is in battle interactive events where the threat can escalate wildly based on group accomplishments/failures, and sometimes it just happens. Because of how poorly it goes, I almost never write or, if I am a developer, allow someone to deliberately write retreat into an adventure. I’ve made 1-2 exceptions, but only where it was a specific part of the experience with very clear player direction (the objective is to run) and with a lot of instructions to DMs. For example, in battle interactives, it became popular to allow the table at any time to “vote a retreat,” and if the majority agreed, then the party captain could cause the entire table to retreat on their initiative count. They take a failure on the mission, but they live and can go on to the next mission. However, just as the humans in Lord of the Rings forget about the rings, organized play admins and authors sometimes forget these lessons.

    My advice to DMs is to develop an understanding that fleeing is almost always failure. At the simplest level, I recommend that if the party wants to flee, you let them flee. You can assign a few conditions, such as:
    – You can flee, but these two monsters each get a parting shot. (To simulate the danger)
    – You can flee, if you can grab the unconscious party member and get them to this spot (to simulate the retreat being possible).

    Whether conditions are necessary really depends on what is fun for the party. Some players really need to feel that things are “real” and not just hand-waived. How much you through in to simulate is a tough call. You want to facilitate the retreat, not stop it, but give them the taste of reality they need.

    One thing that can work well is to abandon initiative. When a retreat is called for by a player, stop initiative and let everyone discuss it. Once the group has a consensus, let them know what is required. Give them all a turn to accomplish the retreat, and then the monsters will all go at the end of this special round. Yes, it means some PCs or monsters may actually go twice this round. But, removing initiative prevents the back-and-forth surprises (“You three retreated, I’ll just grab this thing and go… oh, I provoked an attack and I’m down.”) but still has a threat, because if they don’t pull it off in one turn, the monsters will all go and likely devastate them.

    • I think it’s worth noting that, as a rule, there is a significant disincentive for monsters to try to keep the fight going after the PCs have decided to flee.

      Generally, monsters live in dangerous areas and have access to fewer healing resources than do PCs. Protracted combat is risky for them because it they take damage, they heal slower and are vulnerable to attacks from their neighbours who might take advantage. PCs can leave the dungeon, but the injured monsters live there. An injured animal is vulnerable to predators, and possibly even to some creatures that would normally be prey. An injured humanoid has to worry about infection, or attacks from monsters or rival clans.

      Taking even a single unnecessary hit constitutes a risk. Unless they have a real reason to see the PCs dead, team monster may well prefer to let them go. And they may even tell the PCs that: take your wounded and go, and don’t come back. We’ll be ready for you if you do.

  10. You’re so full of *@#$%^$#$ on this one. And I say that because I love you. Really.

    But how long do you play with this style – calibrating the encounters to the party level and steering them into only dealing with challenges that have been designed for the party to defeat – before the players start realizing that all their successes are *your* successes? The party only ever wins because you created the winning situation and told them where the wins are. Sure, if you are good you can hide your tracks for quite a while, but eventually the players catch on.

    However much fun you have with this style – and I’m not saying that you aren’t having fun – it pales before the pleasure players get when they *bust* the DM. And this far more likely to happen when, as DM, you salt your campaign with “uncontrolled” encounters – stuff that you didn’t calibrate. Whether these come in the form of random encounters or you place them on your map in advance, providing a variety of encounter opportunities that are clearly uncalibrated both provides opportunities for the players to accomplish the unexpected and also give the illusion that your planned encounters aren’t so calibrated. To be effective, you need encounters that are both over- and under-powered. Underpowered encounters can be good role-play opportunities and offer a chance for players to break out of the murder-hobo routine. (They can also sap a party of resources they might need later.) Overpowered encounters give the players a sense of mortality and sometimes a chance to surprise the DM. Together they provide verisimilitude – the feeling of being in a natural world instead of a DM’s artificial creation.

    The bonus is that when your players feel that the world is inherently random and harsh, your calibrated encounters – and all that “story-structure” stuff that you keep blathering about – no longer seem so artificial. That structure is camouflaged.

    DM’s are control freaks by nature. Really good DM’s know their weakness and actively work against their natural grain. I love you, Angry, but you need to !@#$#@#$@ loosen the reins on your campaign. Too often it doesn’t seem like you are DMing – you seem to be designing a board game. Sadly, “” is already registered by the more talented, but lesser known, third Parker Brother.

    • You are the reason the angry GM is always so angry. You basically said, “I like your article, except for everything it stands for, and I like what it argues against doing.”

      If my players see the game as a contest between me and them, they’re not people I want to play these games with, just like I won’t play at a table where the DM’s attitude is the same. If my players derive pleasure from “busting” the DM, then they’re not people I want to play games with either, because there’s a fundamental and unreconcilable difference in how we view the purpose of the game.

      As angry says, you don’t need permission to run your game any wrong way you want, but your style is clearly different and calling for him to change his style to match yours is just pointless.

      • I’m not suggesting a Player v. DM style. I’m suggesting that when you *engineer* every encounter, you end up with a plastic, boring campaign. The players soon get bored defeating encounters that were designed for them to win. Salting the campaign with uncalibrated encounters camouflages the DM’s manipulations and gives the players a greater sense of agency and accomplishment.

        And if as the DM you don’t want to play with players that want to “bust” you, get over yourself. I have never, ever heard a player brag about an encounter in any game in which the players won doing exactly what the DM expected. Players get excited when they come up with something imaginative, unexpected, and effective. That is, when they “bust” out of the mold the DM created. That’s not DM versus Player- that’s DM PLUS Player.

    • How about because of the already spelled out advice on this website, of build to challenge fairly, play to win. The players should be steamrolling every encounter, and should be getting beat up by your “fair encounters” as well.

      CR, ER or whatever has no meaning in a game world and only exists to help guide people on how to build a fair challenge. If you want to ensure the players know there’s things that are bigger and badder than them in the game world there are a lot of better ways to do that than.

      Oh no you have to run away from a thing, despite basic knowledge of D&D movement rules telling you that probably won’t happen.

      Hey you remember that great part in a Final Fantasy game where there was a big thing and you have to hold down the run away buttons and you run away or otherwise you die? Cause I don’t… I don’t remember what the monsters name is or even which roman numeral was in the title. I’m pretty sure it happened once though… and it was nothing more than annoying.

      • Closest I can think of is one scene in FF2 or FF3 where your party encounters Bahamut very early on. I just remember seeing that, and knowing FF lore, “Oh, this is clearly an unwinnable battle. The script likely calls for me to be running, and if that doesn’t work, I’ll have a scripted defeat by Megaflare. My choice is to play along or chafe at the rails.”

        Another might be FF8, with a giant spider robot, IIRC. You could “defeat” it in every combat encounter to slow it down, but it kept getting back up and you’d run because the script said so.

        I can think of better ways to do retreat scenes with that sort of engine: Attack scenery “enemies” to create obstacles or interrupt attacks, have a message box tell you how many rounds you need to survive to escape. Or, probably better, don’t use the combat screen at all: Do it on the map, and make the escape into a puzzle or a stealth game.

        • Right, My point being… going into the battle screen and having the player hit run, isn’t particularly engaging or memorable. The complaint that the players will only ever encounter stuff they have a chance of handling can be easily mitigated by making the players be aware of and actively avoid what they can’t handle.

          If I can describe a guild’s power and connections and have my players work to avoid getting their attention/pissing them off, you can do the same thing with individual monsters or areas, and in a far more effective manner than having them go to their lair and then “run away.”

    • I would like to point out to PokeP that by definition, EVERY victory a player achieves was given by the DM. It is just how the game works. Every enemy they defeat, every fortress they conquer, every treasure they hoard. The GM is their eyes and hands in the world, and nothing ever happens that the GM can really say “well, you made it, I wasn’t responsible for this one”.

      And the players know this. Or should know. The game isn’t about beating the GM, and as NoxAeturnus said, if my players think it is they are not people I want to play with.

      RPG is a compromisse. The GM promisses not to just kill the party (and if he wants he can. He always can. He literally controls the whole world) neither to bore them. And the party promisses to abide to the world rules, try to win and not to be idiots (at least this is how it should be). And this way, everybody gets some fun.

      If you want to play something where there is no one on the steering wheel, you are looking for a board game. And that is not wrong. I like board games too.
      It is just that, if you think an RPG is a board game, you will always hate itm without even knowing why.

      • Your first point is true enough, Rafael. That’s why it is so important, as DM, to camouflage your engineering. I didn’t say “beat” the DM – I said “bust” the DM. That is, to bust the mold the DM created, to bust his expectations, to bust out of the (all-too predictable) narrative. As a DM, you obviously can’t create those moments. But you can provide a gaming environment where it seems more possible. By loosening your grip on the narrative, you give your players a greater sense of agency – and ultimately, accomplishment.

        • Your comments made me realise something. You don’t seem to know how Angry builds his encounters, do you? I mean, his mote is “build fair, play to win”. In the sense he will never put a challenge the players have no way to win, but will not give them the victory in a silver plate either. They need to work for it, and for what I have read on this site, they usually need to work smart/hard (their choice). He doesn’t build “unbreakable molds” for his challenges, where the players need to do specific things in order to win. Quite the opposit.

          Also, I feel like saying: loosening the grip on the reins can actually TAKE agency from the players. Think about it. If you put four 1st level players agains an adult red dragon they have NO WAY of fighting it. They will need to use diplomacy, lies, bribes, hide themselves… Do anything but fight. You gave them LESS options to win the challenge.
          The same goes for letting them meet a guard that won’t be intimidated or bribed, that perceives all of their lies, that has eyes everywhere… You are pretty much telling them “your can’t bypass the guard. You don’t have a choice. Just fight and kill him”.
          (This may have been a poor example. I’m thinking about something that could be a non-combat challenge but is much harder to win without combat. Like this guard that is very unlikely to listen, be bribed, be deceived and so on.)

          If your players stumble on a too hard challenge, they have less agency over it. It is much, much better (and easier, for the record) to build an encounter that can be busted in creative ways (not in a specific “creative” way, like the classic encounter in the cliff) than tell the players “surprise, you can’t fight your way out of this one! Think about something clever, and think quick!!”.

    • Nothing makes less sense than departing from the “Town of Early Beginnings’ and running into the ‘Beast of Utter Destruction”.

    • I think the thing you’re missing in your analysis is the role of dice and randomness. Even if you planned every single encounter to perfectly match the Encounter XP threshold for your party’s level, each fight still ends up feeling very different in difficulty. A few bad rolls, bad tactics and fewer resources to use could make an “average difficulty” encounter deadly. A lot of factors comes into play during combat, and there is definitely a fair bit of luck to it.

    • …Wait, have any players ACTUALLY complained that “this world doesn’t feel inherently random and harsh enough because we have a decent chance of winning every encounter”? Or is this a made-up problem? Furthermore, if you want a campaign like that, why not convince them to play another system that’s designed to be more random and harsh?

      Also, from personal experience, I’ve found that extremely-challenging encounters actually stifle creative problem-solving instead of encouraging it. The closer a fight is, the less likely players are to skip an attack to try a random idea which may or may not work. The fights that actually encourage creativity are slightly-hard encounters where the PCs feel comfortable skipping an attack to try something which might save them resources in the long-run.

  11. Thanks for the answer, there’s a lot of good info in here.

    Looking back though, I realized I had this other question, which you briefly touched on, but for some fucking reason I let it get all convoluted inside questions about my campaign.

    Really my question is about how do players know how strong a creature is, without just telling them. A group of experienced D&D players will walk up to a shambling mound at level 2 & run the fuck away. Or they might know which tactics to use so they can overcome the threat. A lot that stuff can only be learned from experience, either DMing or just playing a bunch.

    I also forgot to include the example where it goes the other way. My party walks into an encounter that will be mildly dangerous at best, and then they blow 4 max level spell slots because they just don’t know any better.

    Finally, running away was a bad example on my part. The point was, my party only have one reaction to everything, fight it as hard as they can. They never opt for anything else. Like perhaps you should try talking, or sneaking, or using the environment, or capturing the low level guys. And I’m willing to admit that’s my fault, if I do a better job setting up the scene to be a bit more obvious & to have other options perhaps I can nudge them into branching out.

    Thanks again!

    • Not sure if this will help you out or not, but one technique I use pretty commonly in open world scenarios is monster v. monster.

      As an example I did a little Jurassic Island adventure once upon a time. Low level characters sent back in time. Was fun. Anyways. They were starving so they stole some eggs from some Pachycephalosaurs (the boneheads). Had a tough fight. Barely managed it.

      Now, the island had more dangerous sections, and a little while later, as the players wandered a bit too close to a section that would have gotten them killed I had them stumble upon another group of Pachycephalosaurs… which before the parties very eyes got mauled by a T-Rex. The party immediately split.

      So in the future whenever the party would venture close to a place I knew would kill them (note not a place that just “might” kill them) I would little the ground with bones of enemies they had fought before but had had a tough time with, or actually show the players those enemies being killed by tougher foes.

      Since then I’ve used it in a lot of different open world campaigns. Anyways, maybe that will help. GL.

    • Only experience can teach your players how to manage their PCs’ resources.

      PCs, on the other hand, have a certain amount of experience baked in. In most settings, they grew up hearing folk tales involving the more common monsters. They have been trained in the art of killing men and monsters. Logically, that training would have included instruction as to what they can take right now, and what needs to be avoided. Moreover, people who are competent in their field can generally recognize other people who are competent in their field with very little information. And pretty much everyone can recognize a very competent person if they have a chance to watch them in action.

      So if the PCs know what a monster is, they probably know how tough it is, or might be. If they don’t know what a monster is, then they should approach it cautiously. They might be able to figure it out over time if they take the time to observe or research it.

      Either way, you should either tell the players how tough a fight is likely to be, or tell them that they don’t know how tough is likely to be. This is an issue of CHARACTER knowledge, so don’t rely on PLAYERS to pick up on your cues. Just tell them. Before they get to this stage, your hints can be more subtle, but once they observe the creatures, just tell them: they think this fight will be easy/moderate/hard/dangerous/foolhardy to attempt, or they just don’t know.

      Even if they don’t know, unless the creature is toying with them, they should know they are outclassed pretty much as soon as combat begins. So tell them. They might even think to retreat before it is too late.

      • I don’t know about other editions, but each monster in 4e was listed with some info and a nature check. At the beginning of a battle, someone would usually say “Have i seen this before? Have i read about it or heard about it?” etc. They’d roll, and based on that roll, i’d give up information. Better roll, better information. But i’d usually include something like “You read in a book that…blah blah.” Or, “Your uncle died of the poison these wasps sting with” etc. Adds a bit of flavor to what’s going on as well.

        • Sure, but even then it wasn’t a stat block worth of knowledge. And, players still overestimate their capabilities, or the dice roll poorly. 4E powers (especially interrupts!) that knock prone, immobilize, or apply other conditions all made running away extremely hard. Powers could also make players extremely overconfident, such as teleportation… and then they end up blinded with no line of sight and the plan falls apart. I saw countless cases in organized play where a plan to retreat went wrong, even after players knew most of what a foe could do. Honestly, if you didn’t know whether the DM would just let you retreat, it was a better strategy to never retreat. (I’m the kind of player that would try to just talk to the DM… “hey, is retreat possible? It can be tough to do it within the rules… is there a way we can work to successfully retreat beyond the normal in-initiative rules?”)

          • As I see, the system wasn’t designed to handle retreats. It just assumes the players will never run, or that the GM will never put them in a danger that requires them to run, because the purpose of the game is to overcome challenges, not to run from them (I know running away may be considered a victory, but as Angry pointed, the whole scene must be designed in another way).

            Also, I try to not handle everything in terms of combat actions, since the things a player does in combat are often a localized and conscious effort. For exemple, a monk doesn’t travel faster than other players, despite having a higher base speed. This extra speed represents a conscious effort that cannot be manteined for a long time. The same is valid for the bonus action the rogue can use to move again, for example.

            Meaning, once the retreat has started, most of the normal combat rules shouldn’t be used anymore. But I agree that actually starting the retreat is the tricky point, and may highly depend on the GM.

          • I remember several times in 4E I didn’t even bother suggesting a retreat because the enemies had obscenely higher movement than us. How do you run from a solo creature with Move 12?

          • @ Mike Lemmer, as a player you use terrain and your controller to slow it down, or failing that your defender sacrifices himself to pin it down while everyone else escapes. Plus you evaluate the situation carefully before you go in, don’t overcommit yourself, and try to keep an exit strategy in mind.

            As a GM you think about whether the monster really wants to pursue. You can also think about adding terrain to the scenario which would facilitate retreat. Assuming that is the sort of thing you want to encourage.

            And don’t forget the party will be using a double move at a run, while the monster needs to make a single move if it wants to attack, and probably doesn’t want to run because it affects its accuracy. A running party grants CA, but a running monster is -5 to hit.

      • It’s kind of like Angry’s point about skill DCs. You want a player with Climb/Athletics to know how hard it will be to climb that wall, so you tell the player the DC if they’re familiar with the system, or an idea of their chances if they’re new to the game.

        • Yes, exactly. I drew from that in other earlier comments. This is often a DM problem, just AS MUCH as a player problem. It’s truly as simple as “Guys, it looks like you’re not doing too well. You should run and regroup.” or whatever.

          Everyone keeps talking about the mechanics of the thing. Screw the mechanics. Do you NEED mechanics for running away. Decide–as the DM–whether they can run away or not. If the answer is yes, play it as so.

    • About your party only having one response (kill it), have you considered that you are posing the wrong dramatic question to them? I advise you to read this, if you have not read yet:

      Spoilling a little of the article, think about this: if you tell your players “as you enter the cave a spider drop from the ceiling and attacks”, you are pretty much telling them “kill this thing”. The “dramatic question” would be “can the players kill the thing?”. But if you tell them “as you enter you notice a spider lurking in the shadows, protecting its nest. You see the exit at the wall in the other side” the encounter changes. You are not pushing them into a fight. The question is “can the players reach the other side without dying?”.

      And remember, what matters is what the players think. If they believe their only option is to fight to the death, that is what they are going to do.

      Anyway, I strongly advise you to (re)read everything in this section of the site:

      The players only do what you tell them that is possible. If they are doing the “wrong” choices, you may be sending the wrong signals.

  12. Let’s assume you still want to include fights that the PCs should run from. Here are two easy, practical suggestions to make your PCs more likely to run.

    1. Knockback
    When the monster hits a PC (especially if it KOs them) narrate the victim getting knocked into the air by the power of the blow, and conveniently landing right at their friends’ feet, outside the monster’s attack range. In my experience, PCs are a lot quicker to retreat if they don’t need to spend 20+ feet of movement to go pick up a body, worry about and Opportunity Attack from the monster, and then move back again.

    I can’t overemphasize the effect that this slight tactical shift has on PCs. As soon as they move toward the monster, they’ll want to try to attack it. If they can spend their entire turn moving away from it without abandoning their unconscious friend, it will be a much more attractive option.

    2. Escape Route
    Include in the encounter area an obvious escape route, like a tunnel or doorway that is too small for the monster to follow the PCs through, or a ledge or rope ladder they can drop down to get out of sight. Mention it more than once, and be very clear in your description (“If you go into that cave, you think it won’t be able to follow you!”). Be generous with the mechanics of hauling KO’d friends in these situations (Sure, you can safely climb down while holding the 300 lb fighter on your back!). The PCs might start by just taking temporary cover, but when they decide retreat in necessary they’ll be that much closer.

    I usually see players try to use this terrain to kite the monster (pop out, shoot an arrow, pop back in), so be ready to shut that down with restrictive terrain, limited ammo (cantrips are a problem), or a Ready Action from the monster. (Note: if they come up with a legitimately clever way to use terrain, let them do it).

    A big problem comes when PCs think “I’ll try just one more attack, and if it misses I can run next round.” They will keep thinking this every round until they die. Try to use these suggestions to create (ugh) FOMO. Make it clear that your chance to run is NOW, and if you spent your action futilely attacking you won’t have the same chance again.

    • AlphaStream has once more great and easy-to-implement suggestion above, which is simply to abandon Initiative when one side starts to run.

      Actually, this is a piece of prehistoric AngryGM advice. Angry wrote back in 2014:
      “the D&D combat rules really only work when two forces of roughly equal power level are going at each other… On the rare occasions when the party tries to get away from something, especially if they go for a fighting withdrawal, they end up tripping over the action economy and the turn order. It becomes an impediment. And DMs rarely seem to want to drop the initiative order once the party has agreed to flee and run things more narratively.”

      I can’t believe Angry didn’t link this himself in the article, because it totally applies to the situation. Check it out:

  13. I found the advice given here to be odd. Not so much because of what was said, but because it was said from someone writing a mega-dungeon. A mega-dungeon is pretty much the purest form of D&D sandbox, and one of its major features is the idea that it serves as your setting for low to high levels, meaning there’s areas you can go to where you’ll be over your head, and areas where you may encounter things under your level. As PCs you have a lot more control than you would in a traditional dungeon and to many this is a main appeal of a mega-dungeon. It’s a world in a bottle, and that world exists to be explored. So it surprises me that the overall stance of the article is designed with the idea that all encounters should be scaled for the PCs CRs. That practically contradicts the entire point of a megadungeon.

    As far as the actual issue of overpowered encounters, I think the issue comes down to this.

    Having varying encounters is really about one thing: Exploration matters. There’s a bunch of things you do in RPGs. Combat, exploration and interaction are the main ones. In many games, only combat matters. In other words, bad decisions in combat can get you killed, but if you interact poorly or you explore poorly it probably won’t matter much to you. The PCs can insult the quest-giver or blindly choose paths in a dungeon and it won’t really impact their performance. Yes, maybe going right instead of left might mean you face a different combat encounter, but since all combat encounters are balanced to your level, it really doesn’t matter one way or the other. Maybe by not searching for secret doors, you’ll miss some treasure, but that’s about it. You won’t lose the game in the exploration phase no matter what you do.

    Now some people are okay with that, but it also does lead to the sitgma that D&D is “all combat”. Exploration abilities are largely flavor. Using survival to see there’s dragon tracks isn’t a valuable clue most of the time since the dragon is designed to be beatable anyway, so really who cares if you know that ahead of time? I’m sure there’s some argument about conserving resources and all that crap, but truth be told most groups don’t care about that. The PCs are after the loot anyways. So for all the time I’ve seen people try to be clever and listen at the door or scout with clairvoyance, they end up opening the thing 100% of the time anyway. Because ultimately the PCs know the encounter is beatable regardless of what it is. So at the end of the day, it doesn’t matter what they detect. So kick down the door and get on with the next fight.

    If on the other hand, you include some very difficult encounters then it now means the guy that heard loud dragon snoring on the other side of the door or cast an augury spell can end up saving the day. By opting to avoid that room, it’s a huge benefit. Exploration decisions and information now matter much more heavily in determining the adventurer’s fate.

    Largely though, it’s a matter of opinion whether that’s a good thing. Some groups just hate exploration in general and simply want a transition from one battle to the next as 95% of all 4E D&D games worked, but I wouldn’t say it’s an automatic mistake to include tougher monsters as it can specifically reward your PCs that decided to take information gathering exploration skills. If you’ve got a ranger PC that actually wants to feel special, allowing him to use expert survival knowledge to avoid a brutal battle is one way to do it.

    • Two or three things.

      Number one, Angry has designed his megadungeon to teach players as they go, using baby steps. No such railing for players in an island free-roam with a new GM more often than not.

      Number two, as was pointed out in the question itself and reiterated by Angry later on, the campaign boils down to ‘Explore, Fight’
      Angry also pointed out that removing the ‘Fight’ part of that basically waters this down further.

      Number three, You won’t lose the game in the exploration phase no matter what you do. I take issue with this sentence for a variety of reasons but lets get the big one out of the way. As you said in your post, exploration matters, to which I am in agreement. I do believe that exploration can be failed however, because I consider all sorts of things to be included under the large umbrella of exploration. Climbing down rock walls, traversing trap laden treasure rooms, riding a horse to outrun an avalanche of dead bodies pouring from a demon gate in a mausoleum. (funny story about that one.) Point is that each of those can end in disaster, up to and including the dreaded TPK.

      Now, I can not profess to know your groups, but if they are similar to the situations you have described, you have my pity. I guess I was just blessed with a group who can reason things out on a character survival and growth level, without the “Herp Derp, MUNCHKIN HO!” bull.

  14. I see two… not issues, but situations that this doesn’t work for, for me.

    The first is introducing powerful villains before they can be defeated. ‘I’m way to powerful, I’ll let you go’ doesn’t work for me, especially when they have the upper hand. I squash bugs, so do my NPCs. Maybe I need to all but wipe the party a few times.

    The second is the above mentioned force of nature encounter. There’s a thing that you could engage in combat with, or you could talk to, but if you fight you’ll lose.. The zero level of this is that there’s a dragon that serves as an aspect of the dungeon rather than an encounter, but the players keep insisting on attacking it (or almost as bad, avoiding the whole dungeon because they know they can’t kill it.)

    It’s something I’ve been fighting with for a while: How to create a sense of overwhelming danger for overarching plot reasons without killing PCs? I can create the overwhelming danger… it’s the sense I’m missing. How do you establish the rules of the combat? How easy is it to run away? What is likely to happen if you surrender? Should you attack, or let the bad thing happen as a character motivation moment?

    • As someone mentioned earlier – have the monster be doing something terrible to behold. Don’t just have it sitting docilely waiting for something to attack. Have it be the super-predator/awesome beast it is.

    • The way you say “there’s a dragon that serves as an aspect of the dungeon rather than an encounter” would imply that the dragon IS the encounter, and/or that a dungeon feature can’t be one. Both statements are incorrect.

      An encouter can revolve around a trap, that is basically an aspect of the dungeon. And the dragon itself isn’t the encounter. If the party has to stop the dragon from eating the villagers the encounter is about that. It isn’t “dragon”, neither “killing the dragon”, neither “parlaying with the dragon”. I mean, technically the characters acomplish their mission if they kill and burn all the villagers (since their goal is not to ensure the villagers safety, but to not let the dragon eat them).

      That said, if the players keep trying to kill the dragon, or they don’t realise there are other options or they really want to kill a dragon. The same way that, if they are trying to break a wall to reach a room, they probably don’t realise there is a door (and you made a mistake when designing the door/wall/room) or they’d rather break the wall, whatever the reason.

      • Try: “The party must be quiet in the dungeon or else the dragon wakes up and eats them” as an aspect of the dungeon, and “The party must deal with an owlbear”, “The party must cross a broken bridge”, and “The party must deal with a group of goblins” as encounters. The design is such that the players never accidentally trigger the dragon, they just can’t use loud (sonic) attacks or metal weapons. Instead of not using sonic attacks, the players instead ring a gong and try to fight the dragon, who promptly eats them.

        More realistically, the players realize they can’t kill the dragon, so avoid the dungeon altogether, and go look for something else to do, because they don’t know how to deal with something that’s more powerful than them and not a friendly NPC

        • It is actually a cool idea. Just seems to lack 2 things:

          0 – Well, it was supposed to be 2 things. But there is one assumption: the players know that the dragon is too strong for them to fight. If they don’t know, it is a vital information that you left out of their reach, thus, bad design.
          1 – a reason for the players to be in the dungeon, given they can just choose to not enter. Maybe they need some item from the dragon hoard and can’t wait to get it when they are stronger. If you want them to enter the dungeon but don’t have a reason for it, it’s bad design.
          2 – a way out if they wake up the dragon. They may convince it to let them live, they may summon a creature to be eaten while they run, they may have prepared a scape route earlier… Or you may do this things. Just, leave them a way to run, probably with some drawback. Maybe they couldn’t get the item and next time it will be harder to retrieve it. If you don’t want to kill them but didn’t toke precautions agains players being players… Bad design.

          That said, if the players purposely awoke a dragon they knew was too strong for them, I se no reason to let them live.

    • As for creating a sense of danger, I believe the best way is to make them fear goblins.

      In my experience, no one trully fears a goblin. Or an army of goblins, for the matter. “They are just goblins”. But if your players feared the goblins, imagine what would be the dread of facing a dragon?

      I mean, if usually an enconter poses no challenge, the players have no reason to think that the next one will be really dangerous, even if it looks (and is) dangerous.Just don’t overdo it, or the players will become too cautious to ever face anything again, even if it’s just goblins.

      As for the specific details about how much difficult is running away, the outcome of a capture and etc, I believe this highly depends on the enemies and their motivations. A beast may be happy if the characters leave its territory. A assassin trying to recover a stolen map may stop for nothing, not even death, and return to haunt the characters even if they slay him.

      If I believe running away is a strongly possible outcome I would design the encounter with that in mind. If I don’t want to, then I would not make an encounter from what the players will need to retract. If they are strong enougth to win they are probably strong enought to run (even if they don’t need to).

  15. According my experience the major problem of retreat is INVESTMENT

    There is no problem when group finds powerful enemy which shows in one turn that is impossible to beat. Ideally it will let players play first, most of their attacks will not beat the AC or saves and those few which will succeed with have clear description that although they should cripple normal man, on this boss they have negligible effect. And then he unleash some “mass shame”, like a spell which will put all in sleep, unconsciousness or throw them back and will do serious damage. And then he simply leaves as he has no intention to waste time with characters. Players will hate such enemy and will look for revenge once they get stronger.

    There is also no problem if there is some obvious way to retreat and later turn retreat to strategic advantage to finish the enemy.

    But once the combat lasts some turns, players and characters invested in it. They do not understand sunken costs. It is something abstract which does not work naturally They wasted time as players by spending time to play encounter, and as characters they used potions, spells, inspiration and other resources. They also see that enemy is damaged. And there is that spark of hope they will manage to have two crits in row and they will turn this so far failure to victory and will give all this expenses some meaning.

    Can DM do something about it? Hardly. Not directly at least. It needs to be changed as Angry suggested to prevent doing investments and turning them to sunken costs at all.


    And what to do with BAD LUCK on dice?

    1) Fights with unintelligent beasts should not be designed as challenging. If the moderate challenge is combined with bad luck, it turns to tough challenge with ease, but still is beatable.

    2) Fights which are challenging or deadly from beginning should have some safety aspect. Examples…

    2.1) Poison used by enemies stabilize whoever is unconscious, as they want them as food for later or slaves or sacrifice – still it means that game continues even after the defeat.

    2.2) Enemy is intelligent and tries to subdue enemies. It can turn to twist, that he will try to convince characters to turn sides, saying that they fought bravely, but only with his leadership they will flourish. And that gives possibility of major change in story or for players to look for further confrontation in more suitable conditions.

    2.3) Enemy is slow and throws targets away as part of attack.

    2.4) I have currently used hydra on lvl1 party, which did not want to kill. It simply chewed off one leg instead of killing outright (who said that all enemies want to kill characters, some might willingly do targeted attack with disadvantage) and heads started to fight for the leg, giving group obvious sign that they should retreat and take unconscious onelegged buddy with them… and in next village was druid able to regrow the leg, although it will be plant-leg and the group will have to return the favor… (note that hydra ignored characters and just was catching fish, the characters were the aggressors).

    2.5) There should be some sure and obvious way to retreat (narrow tunnel when beaing beaten by huge monster).

  16. There are a lot of people asking what qualifies as a “good reason” for making a must-run encounter, and a lot of other people responding with terrible reasons like “realism”. Sorry, but there is nothing realistic about huge murderbeasts living anywhere near civilized areas, because sensible civilizations wouldn’t live there in the first place. I’ve actually got some good reasons – but each of these scenarios has a *very specific gameplay purpose* for the giant monster involved.

    There are also two design considerations you need for building always-retreat monsters. NOTE: VIRTUALLY ZERO OF THE STOCK MONSTERS IN D&D 5E ARE BUILT TO ACCOMMODATE EITHER POINT. The most important is that they need to be *easy to retreat from*. Retreating is the whole freaking point. Don’t make it hard or unintuitive. Second only to that is that you generally have to heavily focus on defenses over offense: Instead of high-damage attacks, give it regeneration, self-healing, humongous hp, untouchable AC and saves, that kind of thing. If the players engage, you want to be clear that the monster will win, but you don’t want it to win *fast* – otherwise the party won’t even get a chance to retreat until half of them are already dead.

    So, five actually good reasons to include unkillable uberbeasts:

    1) The encounter is quick foreshadowing of some later challenge for the party to rise to. They see the dragon make a flyby and return to its lair. Generally speaking, in this scenario the monster shouldn’t even try to fight the PCs, or something should stop it from hurting them, but it can be useful to have it attack something else. The whole point here is basically just to explicitly say, “There’s a giant monster in that place over there, and fighting it conventionally will obviously get you all killed.” Don’t be subtle about that last part. Those exact words are pretty good.

    2) You want a guardian to block off an area that the party is not meant to get into until they are much much more badass. This is basically just dungeon gating made of meat. If you use this kind of uberbeast, come up with a really good reason why it can’t leave the room. Leaving whatever thing it’s guarding defeats the entire purpose of its existence.

    3) You are using the uberbeast as a threat of failure in a skill-based scenario – usually a stealth segment. Be real careful about those, because players screw up stealth segments more often than they succeed. Have a backup plan for the beastie being alerted other than “TPK”. It’s okay to make that a mission fail – the PCs don’t necessarily need a way to accomplish their goal if the monster wakes up – but the failure can’t destroy the planet, stonewall the adventure, or otherwise screw up the world to the point where you can’t run it anymore.

    4) As three, but instead of what happens when you fail a skill-based scenario, the monster is the penalty for failing a resource management challenge. Maybe the uberbeast is easily distracted by thrown raw meat or bags of gold. Maybe the six holy charms the Church gave you will each drive off the ghost army once. Maybe it is very dark and you may be eaten by a Grue. Running out of resources in the dungeon can safely be game-ending in ways that a failed skill challenge can’t. The party has a lot more control over resource management, and they can generally turn around and leave if they are running low. But you do need to make sure that the world doesn’t end if they are forced to leave.

    5) You are running a dungeon and want a monster that forces the party to move. This monster basically is a trap, like a one-way door, pitfall chute, or teleporter, but slightly less railroady in that the PCs can decide where they run to. I’m personally quite fond of sticking an uberbeast as a rare result on the wandering monster list. It keeps the players on their toes, forces them to be choosier about campsites, and when it shows up blocking their escape route, it incentivizes exploration risks that smart PCs would otherwise never even consider.

    HOWEVER, here it’s very important that some places in the dungeon are obviously safe from the uberbeast. The Magma Dragon’s too big for the narrow corridors, the Vampire Queen can’t cross running water, the Moonlight Demon can only exist outside at night, that sort of thing. The monster flat out can’t do its job if the party has no safe space to escape to. So don’t try this gimmick on overland maps or other open-world scenarios with no barriers – it won’t work.

  17. I personally also love the idea of areas that become dangerous because of the players’ actions. The Plains of Not Much But Rubble turn into Wights A’Plenty After Dark after the characters pilfered the sacred relic that was keeping them asleep. Neighborhood of Suspiciously Friendly People is revealed as Doppelgangersburg after Pastor Poor-Impulse-Control punches the snooping waiter at the tavern. This also can incorporate a fun escape from the hazard scene that maybe introduces some of the new danger, gives players a notion of the scope of it, and reinforces that their actions have consequences. They will also feel especially cool when they once and for all slay every last Wight or negotiate peace with the Doppleganger Doge.

  18. I have to state again that I feel like too many people are looking for mechanics when it comes to running away from enemies. But this isn’t the “RUN” button in a JRPG. It’s a fantasy tabeltop game taht occurs in your head. Ask yourself, as the GM, would my players be able to get away from this beast? ould it even chase any of them down int the first place? Does it have any kind of reason to do so? If the answer is no, allow the creature(s) to become distracted by something. Make that decision as DM. If the creature(s) WOULD give chase, and there’s good reason for it, would the beast ever catch up? At what point does it give up? What would be distracting enough fo it to stop giving chase? If a large beast is territorial, maybe it won’t leave the confines of a certain area. Maybe it’s a afraid of water. Throw a stream in the party didn’t “notice” before. It’s up to you as the GM to GIVE those exits.

    This is one of those things where you sort of have to think on the spot, or plan out in advance. If you KNOW you’re going to throw a big enemy at the party, make sure you DO have outs for them to use. A cliff to push it off, a pool of lava or acid to use as a barrier so it’s harder to chase the players, a river or similar piece of land, a cave the beast won’t fit into so that the party can hide and actually ends up starting the quest of get-the-heck-away-from-that-thing! and finds an exit at the other end of the cave, or something within to help defeat that beast.

    Per all of Angry’s previous advice, encounters NEED to have a modicum of planning. It’s cool to try and “sandbox” the game, but even sandboxes contain linear play, which includes PLANNING. Period.

    • Mechanics are the only tool the PCs have to determine when an action is worth it or not. They use them to help them choose their actions.

      When a player says “I hit him” He knows how good he is at hitting things. When a player says “I run away”, he SHOULD have a good way of knowing what his chances are. Good “escape” mechanics are required if “escape” is to be a choice as valid as any other.

      Without mechanics, you end up with the GMs arbitrary decisions, and running away no longer becomes a “real” choice. And asking the GM everytime “would I be able to run away?” seems boring to me.

      What happens if the PCs are unlucky, and want to run away from a monster you didn’t think would need an escape route? You choose on the spot? You will probably ROLL for it, and if you do, then it’s managed by mechanics.

      • I see what you’re getting at, and you’re right. You do need a basis to make decisions on. But I feel like this is one of those things that should be happening fairly rarely. Adding in a full on mechanic, especially if it’s determined by randomness via dice rolls, may add too much to the game for something you’re not bound to use often.

        I suppose realistically, my mechanic is as simple as the yes-no-how much questions i asked above. It’s why i feel like this is so situational and deterministic that it’s hard for me to think that it needs a fully-brushed, number-crunched mechanic to begin with.

        • Interestingly enough, my players never go in without an exit strategy. For some reason they think they need one most of the time. Might be the fact that I encourage them to have one. Especially when their objective is to sneak aboard an airship in the middle of a heavy conflict between the dwarves who built it (400 or so) and the humans who captured them (800 or so).

          Numbers prompt people near as well as sheer power of a thing, and it teaches them that they are not invincible as well, without crushing them underfoot. They can go toe-to-toe with a few guys, but the more you add, the more likely they are to be captured or killed, and the more likely they are to be more cautious in the future.

          In relevance, though I agree that adding a retreat mechanic would see little use, I feel that so many mechanics in the game are already a little niche anyways. Why, or why not, add another?

          If your players present a well thought out exit strategy (which can be very easy, given good world building and narrative) then why bother with a mechanic? Let them act on THEIR stuff since, after all, choice is king. If they lack for that, then it is time for the GM to step in, either narrating a thrilling chase seen laden with choices that could end in horrific death or triumphant retreat… Or by taking the easy way out with “You are dead” or “You have escaped”

          RamblyTops is Rambly.

        • Yeah, oftentimes you don’t need to change much to do a big difference.

          I agree that a whole set of rules could be too big and complicated, running away should be a (kinda) niche option.

          What I did myself, was give a penalty to attack a target running away from you and not doing anything else and randomized movement speed when sprinting. It was simple enough to add some uncertainty, but making an escape a good defensive option. (And even added some layer of strategy, when a target could run away to get defensive bonus, to come back later)

  19. D&D isn’t a video game, with every encounter built and immutable. The players have no idea what’s going to happen next Monday night, and to be honest, neither do you, generic DM! Although you may have a campaign flow chart you’re following, you have total control whether Smaug is actually home when the players stumble into the Lonely Mountain, or whether he’s out eating sheep or something. Encounters don’t have to go the way they’ve been written, if as written they’ll wipe out the party. The pace of the campaign isn’t something that can be written in stone before Session 0, it emerges from play session to play session.

  20. I kind of have the opposite problem, with players (occasionally including myself) seeing a monster and thinking “This guy seems really powerful, we shouldn’t anger him”, only to find out later that he probably would have gone down in a round or two.

    If someone says that, is it alright for the GM to say “Nah, you can totally take this guy”, or is there a better solution?
    An obvious alternative would be to roll a Nature check or similar, but then you run the risk of failing it and being explicitly unaware of the difficulty.

    It does remind me of a past article about telling players the DC beforehand because an experienced rock climber should know how difficult a cliff will be to climb just by looking at it.
    Would the same apply to creatures’ Challenge Ratings?

    • Where appropriate 🙂

      I regularly tell players the CR of various mobs they should be familiar with, and spitball for them mobs they aren’t.

      Of course, this leaves me able to mislead them when necessary. (i.e. telling them the CR of a beholder, while not mentioning that the thing they are looking at is a spectator).

      Also note that several classes have information acquisition abilities (fighter and mastermind rogues at least) that allow them to gauge an opponent’s stats after a brief examination. RaW it’s a minute, I believe, but there’s no reason to make it take even that long.

      • Yeah, I do try not to make a class feature obsolete when I think about these things, but I also want people who aren’t Fighters or Rogues to not run away in fear from a Goblin-equivalent mook. 😛

        Also, technically that class feature says nothing about Challenge Rating, only various specific stats (including Class Levels, which is the closest you’ll get but by RAW doesn’t work on anything in the Monster Manual).

        • The way I read that is that it starts with the assumption that the PCs know what a thing is and then they can see how you modified it. In particular, I’d say ‘Class Levels’ lets you tell the difference between a goblin mook and a goblin boss, or whatever they are called in the MM.

          Regardless, I’d start with that as a model if you want something more specific than just telling them flat out. Just give the lesser form (Relative CR?) to everyone.

Leave a Reply