I’ve been running a 5th edition D&D game for a couple months now, playing about every other week and I feel like I’m finally starting to really grok the system. I’ve come up with a few rules for creating a good encounter. I’m probably very wrong though and I look forward to revisiting this subject when I’ve finished my campaign to see if my views have changed. For now, think of this as a first draft or a beginners attempt at encounter design.
The basic theory I subscribe to when writing an encounter is that Mechanics and Narrative need to match. The Mechanics need to exist to further the Narrative and the Narrative needs to drive towards interesting and enjoyable mechanics. I try to never introduce a mechanical element that doesn’t make narrative sense, so it fits some aspect of the story and isn’t just random for fun. There is no fighting on a rope bridge unless I wrote a story reason for the rope bridge to be there. Similarly while it might make sense for an evil undead king to be surrounded by his death knight minions and skeleton army, fighting literal hundreds of the same enemy is not fun. So instead I find a reason for him to be else where.
Everything stems from the principle that the two need to match. But what comes first? Chicken or Egg? well, for me it’s whichever one I come up with a good recipe for. Ideally I follow a 9 step process when writing up my encounter. However in execution some steps get skipped and they are often out of order based on how inspiration hits. Still the majority of the ones I write go through this process and it serves as a good checklist even for ones that didn’t.
Step 1 and 2, find your hook and define it in narrative or mechanical terms.
Step 3 and 4, Set and Fuck with narrative and mechanical expectations
Step 5, Apply narrative logic to mechanical effects and Mechanical logic to narrative situations.
Step 6, Create an additional easy mode and hard mode
Step 7, Create the enemy strategy and pre plan a few sets of actions and reactions
Step 8, Clearly define the outcome for victory and defeat.
Step 9, Apply finishing touches
Step 1: Find Your Hook
When deciding on the encounter I try to figure out the meat of it first. The one thing that sums up why this encounter should take place. If the encounter were an elevator pitch what would it be. It can something interestingly mechanical, like wanting to have an encounter on tiny platforms over an abyss or an acid rain that constantly drained everyone’s health, or just a cool or dramatic moment whether confronting the big bad or trying to save a PC’s father from the mob boss that he owes money too.
Step 2:Define it in Narrative and Mechanical terms
Once you’ve got the hook you need to figure out it’s compliment. Most likely you know what it is you want to do for either Narrative or Mechanis because that’s what the hook is. It may need alittle bit of fleshing out but really what this is about is figuring out the other side. If you created a mechanical effect where there is a giant flame thrower in the middle of a circular room that fires in random direction on a different initiative every round you need to create its raison d’etre. Whether ridiculous like a childish and evil mage that thought a trap it’s owns servants couldn’t predictably navigate or something more reasonable like an exhaust port for excess heat with a missing support screw causing it to spin, it doesn’t matter. Either one will inform the story and help design the rest of the encounter.
Alternatively if you have created the narrative moment of a cornering the thief that stole the parties stuff, you need to create mechanics that highlight that. Something like as they run the thief drops 1d4 of their goods. If they stop to grab it he gets a little farther away. If they don’t they lose that thing for good. Perhaps the thief uses their stuff as cover so any attack they make risks destroying their stuff or during the fight he might take an opportunity to swipe more of their gear rather than attack and toss it accomplices on the side. As long as you add something that makes this moment more than just another fight, that the narrative weight is matched by the mechanics so the Players are focused on that element.
Step 3: Set Expectation
Basically this is following normal encounter design from the the DM’s guide. Figure out the terrain, write the descriptions, calculate XP budget, select enemies, draw the battle map reference for yourself, decide the type of encounter(Talking, chase, combat), place traps, ect ect. This not for hammering out details but just to get a good feel of the elements that are going to be included in the encounter.
Step 4: Fuck with Expectations
Now that you now the basics, figure out elements you’re going to change. There should be no such thing as a “standard” encounter in your campaign. It’s boring and pointless. The group is here to do amazing things. Give them amazing things to do. Killing a bunch of orcs in a mountain pass is passe. Even with a hook(orcs are rolling boulder done to crush the party or the orcs are actually undead) it’s still basically the same thing, just now you’ve managed to make it better than a random encounter. It can still be better than that. Spice it up a notch. As they say in the Dresden files RPG, don’t just make it hot, make it Fuego.
I like to change at least 1 standard expectation and try not to do more than 3. It keeps the players from ever feeling fully set in knowing how things work but does give them enough of a framework to hang on that they still can proceed without worrying about turning into whales or potted plants. No improbability drive levels of randomness in my games.
But what expectations should you change? Well if you break down encounters there are really 3 top level elements. The place, the time, and the players. The setting, the scene, and the actors. Each of these can break down into both mechanical and narrative elements. Mechanically Time becomes Initiative, actions per round, and time per round. Narratively Time is about having deadlines(either getting something done before something happens or keeping something from being done until after something happens) and scale(is the encounter over minutes or over days?). Place breaks down mechanically into terrain, cover, and environment(light, smoke, heat, acid rain) while it’s narrative elements are mostly the same but also include things like what the players can see, hear, smell, but don’t interact with and also the emotional history(a castle where your best friend died is different from the bedchamber of a drow merchant). Lastly the players and actors. Mechanically you can change literally anything, HP, AC, Immunity’s, ect ect, even getting into odd things like saying each enemy is actually representative if 10 if the hero’s are fighting an army or changing the way they attack, requiring contests of skill and dealing damage or status effect off of that(instead of using saves though I don’t know why you make that change) or having any damage dealt to them instead effect the person who dealt the damage. I however find the narrative elements to be more fun, making odd enemy objectives, motivations, and behaviors.
As long as some of the expectation on these are inverted or just subverted you should keep every encounter feeling fresh and new. Just make sure you don’t pick the same ones each time. 🙂
Step 5:Apply narrative logic to mechanical effects and Mechanical logic to narrative situations.
Now that you have all the base mechanical elements and narrative ones, you need to go back over and make sure every narrative element has a mechanical counterweight and vice verse. There should be a story of the place the fighting is happening and the people involved. You don’t have to know everything, just how they got here and why they are involved. You should also have mechanical notations for every thing, whether that’s just, this square is difficult vs normal, or detailed rules for a save vs effect that happens on a specific imitative count when two people occupy this region, or whatever. There should be a complete story and that story should be represented mechanically. Look to see where you may have left gaps and fill them in.
Step 6, Create an Easy and A hard mode.
While I love the bounded accuracy of 5e, the math can be a bit loose. I almost killed my entire party with a trap using their quick damage table. I’ve also had “deadly” encounters be swept with no issue and almost party wiped on medium. In general it’s pretty good but still too random for my liking. As such if my party is getting their ass kicked by a group of mook guards who were just meant to delay them letting the big bad start prepping an escape, I prefer to have an option to drop down the difficulty rather than starting to fudge dice. Alternatively if they’ve reached the final boss and they look like they’ll kill her in the first round an hour into the session, it’s nice to have a built in buff system.
Now everyone should know the add/subtract HP trick, but I like to make it a bit less fiat if I use it. I roll HP for all my monsters, and if the HP is less than the average I use it if the party is getting destroyed vs if it’s higher I use it if the encounter seems too easy. However modifying stats seems like a cheesy way to adjust encounters. I like adding a set of more flavorful options.
For making the encounter harder I tend to use Lair actions. In general I only want big dramatic boss encounters to go longer, so I don’t worry about having these outside of confronting the boss in their chamber of power. Lair actions to use include calling minions, finding health potions, applying status effects. This is a really easy option that’s just built into the game. I like to add one or two that are more powerful than the others that I only pull to make it particularly harder.
For encounter that don’t have “lairs” or when the lair actions seem to tame there are a set of good options. Changing the weather is a great one, making every square difficult terrain, causing disadvantage, making Con saves or take cold damage. Changing terrain, a tree, pillar, building falls. A pit opens up. Changing the players. Add a third side to the encounter a group with objectives counter to both the party and their enemy. If you don’t want to lengthen the encounter just make it more costly, final form type transformations are great. Have a set of actions that only can be used if the party seems to being to well. Maybe a second attack on another initiative, a large damaging attack as a reaction, a bonus action that lets them target the whole party.
If you need to make encounters easier, you can do all those things in reverse. Alternatively, have some weakness that only get exposed under certain “conditions.” A player misses an attack but you narrate out how they were able to see that the skeleton is vulnerable to bludgeoning damage. I also like having a “retreat” trigger. This can happen both if the party does something extremely well, like drop a bad guy in one hit(despite mostly losing) or if they do extremely poorly. Individual NPC’s may decide they don’t need to help with this and will wander off, or decide, you know, even though the group might win, they really don’t want to be the one to die getting that victory. It’s important to be careful here because you don’t want it to seem like you are Dues Ex Machina.
Step 7, Create the enemy strategy
Now you have you’re encounter designed in detail from a perspective of where everything starts, but you don’t have the details of where it will go. Preplanning how the enemy will act means that during the encounter things can move much smoother and also takes a medium difficulty encounter and can make it almost a TPK. Tuckers Kobolds style strategy is thing it’s best to pre-plan, so you know whether this encounter should be scaled up or down.
I like to determine exactly what the enemy would prefer to do if the party were obliging. Then I decide if they have back up plans for if they lose initiative, if the party tries to fight at range, at melee, or any particularly obvious tricks. Depending on the enemies intelligence I’ll give them one or two other plans based on when their HP drops or when the party begins to drop. Things to think of, what targets will they select first, what range do they want to fight at, how will they use the environment, and how do they synergise with their allies.
Step 8, Clearly define the outcome for victory and defeat.
You now have the beginning and middle states of the encounter planned but not the finale. It’s good to have the immediate scene after the fight as close to planned as possible. It’s always an anti-climatic moment when the last monster drops and you pause for a second to think of what happens. It’s even worse if you just say, well battles over, clean up the mat and pick up mini’s while I look over my notes for the next scene. You don’t want to lose the momentum victory or defeat and what ever that emotional tone is you should build on it.
Obviously the parties actions will determine some aspect of it, but outlining whether they get killed, captured, stripped of weapons and left if they lose is a great step. If they win, having someone ready on hand to come congratulate them, yell at them, or supplicate to them, is a great way to keep the action moving. Other things are simply narrating the silence that falls as the last warrior dies as everyone has cleared the area or how a blast of magic blows across them since they have now upset or restored some balance of nature thing.
Step 9, Apply finishing touches
At this point you are essentially done, but it’s worth it to go back through from beginning to end and polish everything up. Maybe now that you know how it might end you want to change a few of the enemies you added to give a bit more narrative punch. Or perhaps based on the strategy it’s a bit too powerful and you might want to scale it back and take the archers from 15 down to 10. Regardless this is where you finalize everything.
So anyway, that’s the process. Tomorrow I’ll run through the steps to create two encounters so the theory can be seen in applied scenario.
NOTE, this is not how to run encounters, but just how to build them. The two subjects are distinct and while complementary the skills for one are very different from the skills for the other.
NOTE the second
If you subscribe to GNS theory as I kinda do, you’ll be wondering why I don’t have an option for the hook being based on a theme, tone, internal consistency, ect. So might set up an encounter as a natural result of player actions or as a moment to highlight a major convention of the “genre.” Examples include having the MegaCorp they pissed off nuke their base in ShadowRun or having an encounter that investigators can’t win and they have to run from in Call of Cthulhu. I don’t included these, a) because I’m not great at highlighting tone and theme b) I’m not sure D&D has a set of genre conventions that are worth highlighting c) I haven’t really figured out how to work all 3 together in a fun and enjoyable way. Narrative(as in plot, not theme) and Mechanical I have learned how to tie together. I really just haven’t mastered S in the model. So, it may be a thing you want to try and highlight but it’s not one I will focus on