The Shape of Adventure

Banner Patreon 800 x 100

04-22-15 Feature Art 800 x 450

Remember how I was trying to teach you yutzes how to write an adventure before it seemed like a really good idea to blow a month on custom monster building? Well, I was. And I still am. Let’s get back to that.

First of all, let’s recall that an adventure is basically a container for a whole bunch of scenes. An adventure begins with a motivation, a reason for BOTH the characters AND the players to want to complete the adventure. And an adventure ends with a resolution, which includes the various successful and unsuccessful ways the adventure can turn out. And in between the motivation and the resolution, a whole bunch of scenes play out. And those scenes are connected by a structure, a way of connecting it all together.

It might seem obvious, but I’m going to say it anyway, because a lot of things seem obvious to me but then I have someone shrieking in my comment section that I forgot to cover a thing and that, as a result of that oversight, I’m clearly a moron with my head lodged up my own rectum and do not deserve to be listened to by anyone ever and that’s what I get for giving my audience credit and f$&% the Internet…

But I digress.

It might seem obvious to say it, but once you’ve worked out the resolution and the motivation, adventure building comes down to building scenes and joining them up. And that’s pretty much the meat work of adventure building. That’s what most people think of as adventure building. The reason it took so long to get here, though, is because if you forget the resolution and motivation and skip right to building scenes and joining them up, your adventure sucks. That and the fact that there’s a lot of fine detail in both the resolution and the motivation that people don’t even realize need to be there. Like: “why would your players want to play your stupid ripoff of your favorite Twilight Zone episode, but starring Dr. Who” and “what happens when your idiot players fail to locate the sonic screwdriver so they can drill open the plane window, kill the gremlin, and take the cookbook?”

But once you’ve figured out how the adventure will probably end (the good ending and the bad ending and any other endings it might have) and how the adventure will begin (and why the PCs and their stupid players want to get the good ending and avoid the bad ending), it’s time to start thinking about the actual s$&% that will actually happen in the adventure.


Remember how I told you how an adventure is not just a pile of scenes and encounters. And, in fact, the scenes and encounters aren’t really the adventure at all? It’s the structure (along with the motivations and the resolution) that are really important. Well, that’s still true. And while you can, in theory, just start drawing a map or a flowchart and interconnecting scene ideas willy-nilly, it’s always better to have some sort of a plan.

It’s better to decide what your overall adventure is shaped like.

The Shape of the Adventure

If you think of your adventure like a flowchart or a dungeon map – and you SHOULD, even if it’s not a dungeon – if you think of your adventure like a map, it’s easier to understand what I mean when I say that every adventure has a particular shape. And that shape guides your structure. In some sense, it constrains it, but it also helps you design it.

For example, let’s imagine we’re designing an actual dungeon adventure. Like, a real dungeon. Not just a dungeon in the sense that all adventures are dungeons. A dungeon dungeon. And let’s say we’ve decided our dungeon is a big-ole tower. The fact that it’s a tower tells us about how to draw the map, right? Basically, a tower is a vertical structure and each floor more-or-less matches the shape of the lower floors. The tenth floor of the tower can’t be drastically bigger than the ninth floor because of things like gravity and architecture and common f$&%ing sense. The dungeon is a tower because it’s shaped like a tower.

Now, that can be a pain in the a$&. It means that every floor of the tower must be roughly the same shape and size as every other floor and that stairs have to line up and things like that. But that also helps you. It’s easier to map each floor because you know the overall shape and you know where the entrance to each floor is because it has to line up with the exit from the previous floor because that’s how stairs work.

Well, just like all adventures are dungeons, all adventures have shapes. And, fortunately, there’s really only three basic shapes for an adventure. So, it’s easy to keep track of. For simplicity, I’m going to call those shapes: linear, branching, and open. Often, you’ll hear people refer to railroads and sandboxes. But the reason I don’t use those terms is that they have a lot of emotional baggage because dumba$& GMs never stop arguing about what structures are OBJECTIVELTY the best. Many GMs sneer at the idea of a “railroad” as a crime against role-playing. Lots of GMs despise the “sandbox” as a game of “go find the fun I’ve hidden somewhere in the boring.” And no one – NO ONE – seems to realize there’s a third structure. They all tend to think of a spectrum with “railroad” on one end and “sandbox” on the other. But the truth is, a branching structure is not merely a hybrid structure somewhere between linear and open structures. It’s actually a completely different creature.

In fact, the three structures are so different and require such different levels of understanding and planning that they need to be addressed separately. A good linear adventure is put together completely differently than a good open adventure and neither is the same as a good branching adventure. So, we’re going to look at each of the three different structures in their own turn. But first, we’re going to look at the overall point of structure and the issues that affect all structures?

What am I saying? I’m saying: settle in, we’re going to be talking about this crap for at least a few articles. But when we’re done, you pretty much have everything you need to build adventures. Because, after that, it becomes a matter of putting the pieces together.

The BEST Structure

Let’s take just a moment to talk about which structure is the best structure.

There isn’t one dumba$&.

So shut up.

Different Types of Freedom

All right, let me address the point a little more seriously. The push and pull of “linear” vs. “sandbox” comes down to the idea of freedom. At least, it supposedly does. The problem is, freedom isn’t really the issue that people think it is.

See, the conventional wisdom is that a linear adventure – one in which the heroes overcome a sequence of obstacles in an attempt to accomplish a goal – is inherently constrained. And, because a role-playing game is ALL ABOUT TEH FREEDUMZ, linear games are evil. Seriously, Adolf Hitler only ran linear games according to some people.

The problem, the reason people think of linear games as inherently limited, comes down to two things. First of all, it comes down to execution. Linear games are hard to design properly, but they SEEM easy to design. Just set the heroes along a road, drop three to five obstacles along that road, and then turn them loose. Done and done. And if that’s your mentality, you’re a sucky adventure builder and it has nothing to do with your chosen structure.

Second, most people think of freedom in only one way. Or rather, on only one scale. If the heroes don’t choose their path through the adventure, they don’t really have freedom. If they aren’t free to deviate from the path, they are prisoners of the adventure. But the thing is, that’s stupid. People are stupid.

Recall that an RPG has several different levels of play. Several different scales. There’s the campaign level, the overall storyline that connects all of the adventures. And there’s the adventure level, where individual stories play out. And there’s the scene or encounter level, where the heroes deal with individual conflicts. And there’s the action level where a single hero chooses a single course of action.

Freedom and constraint can exist at any or all of those levels. For example, when the hero has a giant rolling boulder coming for him, the GM can either say “roll a saving throw” or can say “what do you?” The first is constrained, the second is free. At the encounter level, the GM can either say “roll initiative” or “how do you deal with the orc.” Constrained vs. free. At the adventure level, the GM can say “the dragon must be slain” or “the dragon is going to destroy the village, can you help?” And so on.

Now, any RPG is a combination of constrained actions and free choices. Sometimes, there really is no choice. Sometimes, all you can do is roll a saving throw. Sometimes, all you can do is defend yourself against the orc trying to kill you. And so on. There’s nothing inherently wrong with constrained actions, encounters, adventures, or even campaigns.

The key, though, is that the players’ choices somehow have an impact on what happens in the game. Otherwise, the game isn’t really about choice at all. And it has to be about choice. That’s what a role-playing game is. And when that isn’t happening, that’s when the game starts approaching a “railroad” in the parlance of the shrieking mouth-breathing Internet elitist GM.

And that is why an adventure never has just a single resolution. At the very least, an adventure must have two possibilities in the resolutions. If you don’t have at least two possible resolutions, you don’t have any freedom in your adventure. Normally, those resolutions are “success” and “failure,” but they don’t have to be. And there don’t only have to be just two. You could have three different “success” outcomes and no failure outcome.

How is that possible? Well, it comes down the difference between outcomes and consequences.

Outcomes vs. Consequences

A long, LONG time ago, I talked about how to adjudicate actions. Remember? And, as part of that whole, long spiel, I described two different things that come out of every action. The outcome arises from what the player was trying to accomplish. The player either succeeds in what they were trying to do or fails. Consequences, though, grow out of HOW the player handled the action. Consequences ripple forward, affecting other actions, other encounters, other adventures, or the campaign as a whole. I think, way back then, I might have used the example of someone trying to convince a king to give them aid. Either they end up with the soldiers they need or they don’t. That’s the outcome. But if they are rude or if they threaten the king, the fact that the king will remember their behavior and react accordingly is a consequence. The truth of the matter is that it’s consequences more than outcomes that truly reflect the free choices that players make. And we’re not just talking about outcomes vs. consequences at the action level.

For example, imagine the PCs encounter a lone orc on the road. The orc is desperately hungry. He’s been exiled from his tribe, he’s been wandering for days, and he’s starving. In addition, he’s an orc. He knows the pink-skins will just kill him given half a chance. So, the orc is inclined to fight with the PCs.

Now, by all accounts, this is a simple encounter. A desperate orc willing to do battle with the PCs. Either the PCs will survive or they won’t. And depending on the PCs’ levels, the outcome might not even be very uncertain. The orc might have little chance of killing them all.

Now, if the PCs kill the orc, they take some damage, maybe spend a few resources, whatever. All of that crap – the damage, the spent resources, everything that comes out of the fight – those are consequences that carry over into the next scene. The wizard might have spent a magic missile and doesn’t have spell available later. The fighter might have gotten hit hard enough to be low on hit points. Whatever. Combat scenes automatically carry consequences like that. The trouble is most of those consequences don’t matter except in other combat scenes unless someone actually dies.

But suppose the heroes don’t want to murder the orc. Suppose instead they realize he’s desperate and starving and offer him some food. Maybe they win him over. Maybe, now, they have an ally to bring into the next scene. Or the orc gives them useful information. Or maybe later, when they run out of food, it’s because they fed the orc. See? Consequences.

Suppose the heroes won’t kill the orc but can’t calm him down. So, they leave him unconscious. And the orc, humiliated, now hunts them, trying to pick them off one at a time and steal enough food to survive. Or the orc returns to his tribe, offering to lead the tribe to the PCs in return for forgiveness for whatever he did. Consequences.

The point is, when we talk about freedom in an RPG, it’s not just about the freedom to pick a path. In fact, it isn’t really about freedom at all. It’s actually about consequences. It’s about the fact that the story that eventually emerges, the game that eventually plays out, could only happen because the characters did what they did. A different group of characters would experience a completely different game. A completely different story. See?

And this is true at every level – every scope – of play. For example, imagine an adventure in which the heroes need to protect a village from a dragon. Imagine group one slays the dragon, group two cuts a deal, and group three drives the dragon off. The outcome is the same, the village is safe. But how do those choices reverberate across the campaign. The village might enjoy great prosperity once the dragon is dead, but the deal with the dragon might starve it out. Or, what if the villagers and the dragon expand on the deal in the future and the dragon helps the villagers conquer other villages to increase their mutual wealth? That could be an interesting surprise six months down the road for the heroes. And if the heroes merely drive off that dragon, the dragon is going to become someone else’s problem. Again, that could be a very awkward problem later on.

Even if you decide failure isn’t a possibility, even if you design an adventure at which the heroes can only possibly be successful, through the power of reverberating consequences, you still have multiple possibilities built into the resolution.

Now, it’s also important to note that sometimes the OUTCOME is the CONSEQUENCE. Or rather, outcomes themselves carry consequences. For example, if the orc manages to kill all of the heroes? Well, the consequence is making new characters and starting a new game. The adventure ends there. Done and done. And if the heroes fail to get the king to pledge soldiers to their cause, one of the consequences is that they have to defend the mountain pass without soldiers to aid them.

But remember, outcomes are just successes and failures. Do the heroes get what they want or do they fail? Consequences can carry forward from outcomes, but those are often the least interesting consequences. Most outcomes are determined as much by random chance as by choice. Consequences that arise directly from player choices are much better ways to demonstrate player “freedom.”

Every Adventure is Constrained

When you get down to it, every adventure has a shape and that shape necessarily imposes some constraints. The game itself imposes some constraints. There is no total freedom. Hell, the whole concept of adventure design only works on the assumption that there are constraints.

You invent a resolution and a motivation, basically a finish line and a starting line. If the players accept the premise of the game – and they will if you design your motivation right – they will walk a path from beginning to end, from scene to scene, choosing actions in each scene that deal with the dramatic question in each scene. Even the most open structure in the universe still has that basic idea at its heart. You set the PCs in motion and point them toward a resolution and then they walk the path. And there’s nothing wrong with that. That’s exactly how it’s supposed to work. And the structure, the shape, doesn’t matter. As long as you design it well.

The fact is, the constraints are VITAL to the game. You, as a GM, need them. The players actually need them. And there is nothing inherently wrong with constraints.

The Three Structures

As a way to wrap this up, let’s briefly introduce the three structures in a little bit of detail. We’ll be discussing them more later on. And, we’ll be designing a few. But it’s important to understand the concepts behind the structures first.

Linear Adventures

A linear adventure is one in which the heroes move from the motivation to the resolution through a sequence of scenes. The players generally do not get to choose which scenes they experience or what path to take from beginning to end. For that reason, I call these Obstacle Course Adventures. If you picture a linear adventure as a dungeon, it’s a sequence of rooms that the heroes have to go through in order. Another example of a linear adventure is a timeline driven adventure. This is an adventure where certain events happen at certain specific times.

Branching Adventures

A branching adventure is one in which the heroes move from the motivation to the resolution by choosing a path through interconnected scenes. As the players move from scene to scene, they generally have a choice of which scenes to move to. I call these Labyrinth Adventures because they can play out like trying to navigate a maze. Many tradition dungeon-based exploration games are branching adventures. And believe it or not, many investigation and mystery adventures tend to be branching adventures.

Open Adventures

An open adventure is one in which the heroes can move freely from any scene to any other scene, with no paths or connections between the scenes. Instead, the scenes are all just there and the players can choose to engage with any of them at any time. I just call these Open Adventures because I don’t have a cleverer name. I mean, you COULD call them sandboxes, but I hate that name. Some mysteries and investigations fall into this category. And free-exploration wilderness adventures (usually called hex-crawls) fall into this category as well. If you picture an open adventure like a map, it looks kind of like one of those gross Jell-O desserts – a morass filled with little floating bits of fruit and stuff – or like a gelatinous cube with bits of bone and coin and gem floating in them.

Now, to the untrained non-genius who isn’t me, branching adventures and open adventures can look very similar. Don’t worry though, I’ll teach you the differences. That way, you’ll be a trained non-genius who isn’t me.

… And the Rest

Even though there are only three basic structures, that’s only half the story. In fact, due to the weird mathematics that happen when you divide by infinity, that’s barely one zeroth of the story. Because most adventures don’t conform to just one of the three structures. In point of fact, most adventures are hybrids.

So, for example, imagine the heroes have to gather three mystical keys from three mystical dungeons on the mystical map to unlock the mystical treasure of Mt. Mystical. They can visit the three dungeons in any order. That’s an open structure. But one of the dungeons is a linear dungeon. And one is branching. And one actually requires the heroes to figure out how to get in by going through other scenes in the open structure and then involves a branching dungeon. And Mt. Mystical is a linear series of five challenges. Or whatever.

The point is, very few adventures actually have just one shape. They are often blobby, half-formed mutant hybrids of multiple shapes all Frankensteined together. And this can make things much more complicated, but it can also make things much easier.


Well, simply put, each of the three basic structures presents its own unique design challenges. Open structures have very different design problems from linear structures and branching structures. And often, GMs tend to have preferences. Not just in what they like, but what they are good at. For example, I kind of hate Open Adventures. I hate writing them. I hate running them. I can do it. But they don’t come as naturally for me as branching and linear structures.

But when you combine structures, the different bits and pieces can cover for each other. The open bits of my Mt. Mystical Misadventure might not work so great, but the linear and branching parts will be good enough to cover for them. Someone else might suck at designing good linear adventures, but by sticking them like spokes coming off an open structure, they might build something great.

That said, you can’t build a good Frankenstein’s monster without at least a rudimentary understanding of where all the different bits go. Franky needs BOTH a digestive AND a nervous system. You don’t have to be a neurosurgeon, but you have to be smart enough not to route his spinal column through his pooper. In short, you can’t cover for your strengths and weaknesses until you know what they actually are.

Yes, I know about vagal nerve. But are we really going to argue that being able to poop yourself into a heart attack is an example of GOOD design?

I didn’t think so.

Anyway, that’s why we’re going to be looking at each of the three basic structures over the course of several articles. But we had to lay the groundwork first. And don’t forget our running examples. We’re going to be putting them to good use across those articles.

Banner Patreon 800 x 100

10 thoughts on “The Shape of Adventure

  1. Woo! Adventure building! Love the article. Between this and the Megadungeon, I’m really starting to feel more confident in putting together good sessions for my players.

    And thank you for addressing the point that linear adventures with different consequences have freedom. Even if it’s a straight path to the end, the challenge is getting there with your resources and lives intact. Don’t be the dumba$$ who blows all his 3rd-level spells on the rat fight in the celler. That’s how you lose at D&D.

  2. I would be very interested in an article dissecting the 5e Starter Adventure (which is generally considered to be good) and revealing the underlying patterns/systems/approaches that were taken with it as discussed in this series of articles. In fact I might even do this myself as an “exercise for the reader”… 🙂

    And thanks for coming back to this topic. I enjoyed the monster building but that seems like icing on top of adventure design.

  3. I’ve been “trying” to GM for the last five years. I didn’t feel like anything I created prior to finding Angry’s website was worth it’s weight in spit.

    Angry! I remember somewhere you mentioned that you we’re planning to release a “HOW TO GM” book for morons. I am anxiously awaiting THAT PRODUCT. PLEASE make it happen!

    I love your stuff on building monsters and MEGA DUNGEONS. Know this though, what brings me back to this website everyday. Reading and re-reading article after article? It’s your articles on GAME MASTERING – The actual act of playing the game. I thank you, my players thank me, so extending their thanks to you for showing me HOW to build the games I’ve always wanted to play.

    Keep up the amazing advice!

  4. I second (tenth?) the hope for a book. There is so VERY much good stuff that each article is both immersive and overwhelming. I feel like my only hope to integrate more of these ideas is to have them in a nice linear fashion that builds, rather than fits and starts as is the nature of web articles. In other words I think I need Angry to hold my hand and keep me on track (and out of traffic!).

Leave a Reply