In Case You Missed It :: Dungeon Design, pt. 1 (Patreon)

This is a copy of the lecture I gave yesterday, the first in a series on dungeon design for 5e. Lectures are free and open to everyone, happening every Sunday at 12pm CDT (18:00 GMT -6). Content+ patrons should check out this post too, which contains a worksheet for getting started on designing your own dungeon.

Defining ‘Dungeon’

When we’re talking about dungeons, the first thing that needs to be clarified is what the definition of a dungeon is. There are a hundred or so different definitions out there, and I’m not going to try to argue what is the most correct or least correct definition. The whole purpose of defining a dungeon first is to make sure we’re all on the same page when discussing them.

For our purposes, a dungeon will be defined as a sectioned-off area of your campaign that has a limited set of entrances and/or exits, contains at least one deadly encounter, and characters are not able to take a short or long rest while within the dungeon’s area. Again, me slapping this definition down here is NOT me trying to tell you this is all a dungeon can be. We will be discussing exceptions, addendums, and all sorts of ways to modify that definition in the coming weeks. This definition is just to make sure we’re on the same page.

I also want to clarify that the point of the lectures and lessons being given this month is not to teach you how to randomly generate a dungeon or how to improv a dungeon. Each of those topics would need their own series of lectures and lessons, but, for this month, my focus will be on teaching you how to construct a dungeon ahead of time, how to carefully consider all the individual components, and how to write it all down so it’s easy to run your dungeon in sessions.

So if you have any questions about how to improv or how to randomly generate content like a roguelike, I will not be answering those this week. I want to keep the focus on premade dungeons and thoroughly teach this one subject as best I can.

How 5e Works

Before making a dungeon, you need to understand what 5e is, so you understand the tools and limitations inherent to the system you’re working with. There is a surprisingly large number of players and GMs who mistakenly think that 5e is a game of skill, but it’s not—it’s a game of chance. Everything you do is governed by a roll of some kind: attack roll, damage roll, saving throw, or ability check. You can roll bad, or you can roll good, but you can never “get good” at rolling dice. They’re dice, there is no skill involved. 5e is basically a glorified slot machine.

But if that’s the case, it begs the question, how the heck does progression in 5e work? Characters are controlled by their players, there is a limit to how effective a character can be based on the player’s individual skill. But if players can’t get good at the game, how are their characters supposed to improve over time? Because, you know, it makes sense that, regardless if your focus is on gameplay or narrative, as characters encounter new situations, they should be gaining worldly experience that they use to increase their proficiency with the skills, weapons, and tools they’ve decided to invest in. That’s narrative 101, and that’s RPG progression 101, too.

And what do you know, that’s literally how 5e’s character progression works. Your character has a number of experience points, which determines what level they’re at, which determines their proficiency bonus and all the cool features and fun actions they’ve learned over the years of adventuring they’ve done. Character progression is baked into the game itself. Players don’t have to worry about “getting good” in order for their characters to become more proficient and deadly warriors over time, it will happen naturally.

What players do need to worry about is risk assessment and resource management. That is the gameplay of 5e, those are the skills that the actual players need to improve on if they want to “get good” at playing.

Those are also the skills that you, the GM, need to be leaning into in order to actually challenge your players or their characters. I’ve done the math, I’ve ripped this game apart at the seams, I can tell you that at any time, if you’re throwing at-level content at the PCs, the PCs have a 65% chance of surviving before they start to deploy tactics on the fight. The odds are in favor of the PCs, and if you’re the kind of GM who wonders why they can throw over-leveled content at the PCs and the PCs still win most of the time, the way 5e is designed is the reason why.

But like I said, the answer to challenging players is not to simply decrease the odds of success for them. You don’t just crank up the statistical rate of failure, because even at 65% players can just roll poorly and end up losing a fight through no fault of their own. 65% is still a far cry from 100%. Plus, it feels bad. Nobody likes it when they lose just because of the dice.

What you do instead is base your gameplay around decisions characters have to make. These can be simple, mechanical decisions, such as should the barbarian rage now, or save their rage for later? Does the wizard cast a 6th level fireball now, or save the spell slot for later? You can get a little more complex with scenarios, too: does the party take the gold now and use it to upgrade their gear, or do they leave the gold where it is and use it as a trap to ambush the dragon? You can even work in moral quandaries, such as allowing the BBEG to escape in order to save the town, or stopping the BBEG right here and now so they don’t go off and kill six other towns.

How it Applies to Dungeons

When you’re making your dungeon, you need to be thinking beyond what the statistical difficulty of your encounters are. You need to be thinking about the situation you’re putting characters in, you need to be thinking about what decisions you’re asking characters to make.

This is why 5e’s core materials say to cram 6–8 encounters in an adventuring day. The more encounters you have, the greater the likelihood that players will end up in a situation where they’re rolling poorly and their characters can’t play it safe anymore, they need to expend their high-level spell slots, they need to expend the once-a-day features. In other words, more encounters means more opportunities to make decisions about resource management.

Which is a narrative tool, too, not just a mechanical one. A character may be the kind of person to spend a lot of time on risk assessment and won’t make a concrete decision until they have concrete information to work on, but a character could also be very impulsive and doesn’t think long-term about their resource management. Limited-use features and spell slots are a mechanical tool, but they are mechanical tools that are extensions of a character’s personality. A barbarian that rarely uses their rage is a very different character from a barbarian that uses their rage every single fight, regardless of how much danger they’re in!

So let’s revisit our definition of a dungeon: a dungeon is a sectioned-off area of your campaign that has a limited set of entrances and/or exits, contains at least one deadly encounter, and players are not able to take a short or long rest while within the dungeon. That last bit in there should hopefully make a bit more sense now because, as just discussed, 5e is about resource management. When you restrict the ability for PCs to take rests, you restrict the ability for PCs to regain resources, which means that you don’t need as many encounters in your dungeon to challenge players, and they don’t need to be as statistically deadly—which makes making your dungeon easier. There’s less work that you, the GM, need to do, which is pretty handy if you’re GMing as a hobby and don’t have much time to invest in pre-session preparation.

The Process for Actually Making a Dungeon

For actually making your dungeon, there’s a 4-step process that I recommend everyone do. You will, of course, create your process over time, just like how everyone has their own method of building a character sheet or how everyone has their own method of washing the dishes. It is expected and encouraged for you to adapt this process to fit your own preferences and needs, or to come up with a whole new process altogether. I recommend this process to start because it covers a lot of bases and makes things simple and easy for hobbyist GMs in particular to get content made.

Step 1: Write out the narrative of your dungeon

What is it, who made it, why was it made, why do the PCs need or want to go there, etc. All of the information you write down in this step will inform everything you make for the dungeon, and writing it down first helps ensure you make a narratively cohesive dungeon.

It also means that, even if all you have is time enough for this one step, you have armed yourself with the knowledge and information you need to improv your dungeon. No, I will not be teaching improv at this time, but yes improv is an amazing skill and worth investing in.

Step 2: Make the minimum viable dungeon

Come up with the linear path through your dungeon, the guaranteed sequence of events that will get players from start to finish. Do not write down alternate paths, do not write alternate endings, and do not concern yourself with decisions you want to present to your players just yet. I know I just gave a whole lecture on that, but:

• Your time is limited. Even me, as a professional who does this full-time, I only have a limited amount of time to make content for my players. At the very least, I need a complete dungeon to run in my next session. Not a grand dungeon, not an impressive dungeon, a complete dungeon. I need a start, I need a middle, I need a finish, so that’s the first thing I’m going to make because at the very least, I need to show up to my next session with that.
• The dice rolls themselves can force decision-making without any intervention from anyone, not from me and not from the players. A series of bad rolls can force a character’s hand and make them tap into their berserk mode in order to survive, and a series of good rolls can make players and characters feel like gods up until they encounter that final boss and realize how much shit they’re in. That is baked into the game. Trust the dice. We got a whole sticker on our server just for it.

Step 3: Use your spare time to expand the dungeon

Go back and add more content, go back and add decisions characters need to make, add alternate paths they could take, add alternate endings. Go and do all the things that make a dungeon go from good to great, and do it with the knowledge that even if you run out of time, you still have your minimum viable dungeon ready to go.

Step 4: Build skips and/or drawbacks

Skips and drawbacks are vital to the health and pacing of your dungeons. They help you make sure the dungeon doesn’t take too long to complete, or that your party actually spends the whole session in the dungeon, if that’s what you intend.

A skip is something you’ve prepared that narratively/mechanically explains why your PCs can just go straight to the end boss, or why they don’t have to do the next fight. It can be something like, the last fight they had broke down a wall, revealing a new path forward. Or it could be that an underground river is abnormally low, allowing players to follow the river deeper into the cave and not have to deal with a bunch of encounters, maybe even getting them into a surprise position against the dungeon’s final boss.

A drawback is an additional challenge you give to the PCs because they went through content too fast. It could be an additional fight, a drawbridge getting knocked down, or a rockslide blocking a path.

We’ve all been in that fight that took too long and absolutely winded the players, and we’ve all been in those fights that were over two turns into initiative. Skips and drawbacks help compensate for the variability that exists when you have an entire game run on random chance.

When building skips and drawbacks, always consider the narrative and/or mechanical reason why they are happening. You don’t have to use them if they aren’t needed, and you don’t necessarily have to foreshadow them, but they should still have some narrative cohesion. Let’s not get crazy and start using deus ex machinas all willy-nilly.

And that’s it, that’s the 4-step process of making a dungeon. I use this process all the time myself. Sometimes I get through every step, and sometimes I only make it to step 2 at most. Regardless of how many steps that get completed, you will have some degree of dungeon done and ready for your next session!

- - -

As mentioned at the beginning of this post, Content+ patrons should check out this post for the dungeon making worksheet we used in the workshop that happened right after the lecture.