Coloring Inside the Lines: Linear Adventure Design

Print Friendly, PDF & Email

December 16, 2015

Remember how we were talking about how to make your own adventures? Let’s get back to that! How’s THAT for a punchy intro, huh? I figured I should keep things simple, partly so that I don’t have a thousand dips$&%s misunderstanding me and making me clarify points that SHOULD be self-evident, and partly because we’re talking about something really simple today. At least, we’re talking about something that SEEMS simple.

Last time – whenever the last time I talked about this subject actually was – I talked about the three broad types of adventure structure. The linear structure, the branching structure, and the open structure. And the idea of structure refers to how the scenes in the adventure link up. But the truth – as will become apparent – is that the structure isn’t really about how the scenes connect. It’s actually about the choices the players can make about which scenes they move to.

Now, linear structures see simple enough. Basically, the players don’t have any choices about the scenes they deal with. The adventure begins, a scene plays out, there’s a transition, another scene plays out, there’s another transition, and another scene plays out, and on and on and on. That’s true of all adventures, of course. But in a linear adventure, the players don’t choose which scenes to transition to. They don’t choose to go left or right. They don’t decide to investigate either investigate the weird tower or visit the spooky mansion. They don’t decide whether to go to the leader of the thieves guild or the captain of the town watch for help.

The simplest example of a linear adventure is a linear dungeon obstacle course. And this structure was insanely popular in D&D 4E. Essentially, the party begins at the entrance to some location. It could be the entrance to a crypt, the mouth of a cave, the start of a jungle path, whatever. The party enters, encounters a series of obstacles, and then they eventually reach their goal. Simple, right?

Well, it’s not actually that simple. In fact, if you want to run a good game, a linear adventure can be the hardest structure to use. But if you want to run a crappy game, a linear structure makes it really easy. And that’s why so many people sneer at the idea of a linear adventure.

Does that mean that linear adventures are automatically crap? No. Absolutely not. Linear adventures are great. They can be great fun. Even sneering, snobbish thespians actually enjoy a good linear adventure. And they are easy to run; they are easy to manage. So they are great for new GMs. The trouble isn’t that linear adventures are crap. It’s that writing a linear adventure that isn’t crap takes finesse.

And we’re going to talk about why.

The Dreaded Railroad

It is almost impossible to talk about linear adventures without some shrieking idiot screaming about railroading. And that doesn’t help anyone. So, let’s talk about what railroading is. What it really is. And why most people who talk about railroading deserve to be pushed off a train.

Railroading is a verb. It refers to a GM designing or running a game in such a way that the players do not have any feeling of agency at all. What is agency? Agency is the feeling that you have some control over the outcome of the adventure. Or scene. Or action. Whatever.

For example, suppose you are walking across a meadow. Suddenly, without warning, a chunk of elemental iron from the elemental chaos falls out of the sky and kills you. In that scene, you had no agency. A chunk of metal just appeared and killed you. Done and done. You couldn’t avoid it. You couldn’t even take any precautions. There is no way you could have seen that coming. Even if the a$&hole GM who was attempting to murder your character had allowed you a saving throw, you still wouldn’t have any feeling of agency. Your life or death is decided by a coin toss. End of story.

That’s railroading. Things happen. You can’t see them coming. You can’t take actions to prevent them. You have no say in what happens except to respond. And you can only respond in limited ways.

Now, imagine you are walking down a dungeon hallway – maybe an ancient dwarven crypt – and suddenly, you hear a click from underfoot. Moments later, a giant metal sphere comes rolling down the hallway and kills you (with a saving throw). Did you have agency this time?

What’s funny – and this is what makes railroading so complicated – is that if you ask lots of gamers that question, some will say yes and some will say no. Some gamers will point out that the scene is no different. You were just walking along and got murdered. Other gamers will say that you should anticipate traps in a crypt. And that you should have been more cautious. The sound of the click indicates you sprung a trap. And that implies you could have circumvented the trap.

And THAT is the problem with agency and railroading. In the end, it is all about perception. And perception is further complicated by the fact that the players rarely have the whole story. They don’t know whether they COULD have seen the trap coming or not. Maybe the GM was just being a murderous d$&%bag. Or maybe the GM rolled a secret Perception check. Or maybe the players could have searched for traps and didn’t.

Remember that players only see the things that actually happen in the game. They don’t know about the treasures they didn’t find, the paths they didn’t take, the monsters they didn’t encounter, and the clever plans they didn’t think of. At the end of the day, it’s impossible for the players to know how much freedom they actually had.

On top of that, most GMs are kind of stupid about railroading and agency. Which is why there’s always some shrieking idiot who will insist that a linear structure is automatically railroading. Except that it isn’t.

Agency is created when the players make choices that affect the outcome of the game. But those choices don’t always have to be structural. For example, imagine that the second scene in a linear dungeon involves a dumb but gentle ogre. If the party launches into combat with the ogre, the ogre will fight back and fight to the death, because he’s still an ogre and ogres are savage and temperamental. But if the party holds back, they can negotiate with the ogre. The ogre will let them go without a fight. And if they offer the ogre gifts or they manipulate or intimidate the ogre, the ogre will join them. And then they will have an ogre ally to help in the next room against the goblins.

It’s still a linear adventure. Step 2 is always ogre. Step 3 is always goblins. But depending on how the party handles step 2, step 3 might go very differently. And imagine if, once the players have the ogre, the goblins are afraid to fight them. Imagine the goblin chief offers to surrender.

The point is that there are different types of agency. It isn’t always about choosing left or right, this guy or that guy, the lady or the tiger. Agency is also about the freedom to engage with scenes in different ways. And to have your choices follow you through the adventure.

Now, every adventure must afford some feeling of agency. But the amount of agency players need varies from player and player and adventure to adventure. And players – it turns out – are much less sensitive to agency that most GMs assume they are. Especially advice-giving opinionated internet GMs.

Agency in Linear Adventures

In general, in a linear adventure, there are two ways to convey a sense of agency. And they are tied up in two words I’ve been harping about since I first told you how to handle a simple die roll: Approaches and Consequences.

First of all, agency arises when the players are free to decide how to engage with a particular obstacle or challenge. The ogre I mentioned above is a perfect example. The party could attack OR they could negotiate. And maybe they could also sneak past the ogre. Or trick the ogre. Who knows? The point is, when the scene opens, the players need to feel like they have the power to decide how to deal with whatever is in front of them.

Now, this is more a matter of scene and encounter design, but it’s important. Remember when I talked about how to build encounters? And one of the things I said was that the way you open the encounter – the way you present the dramatic question – is important? That it basically lays out what the whole point of the scene is and that it drives how the players respond?

Well, in order to create a sense of Agency of Approach – a feeling that the players are free to deal with an obstacle in a number of different ways – you have to set the scene in a way that creates options. If the party comes into the ogre’s den and the ogre raises his club and bellows a threat before advancing on the party, that’s going to call out for an initiative roll. Even if the ogre is just trying to scare the party away, it’s going to look like a fight is starting. And the players will assume that they have to fight their way out. Instead, if you call attention to the ogre holding his ground, hefting his club, but staring at the party unsure of what to do, you create options.

The second way agency arises is when the party carries baggage from one scene to the next. That is to say, the choices the players make in one scene affects the way other scenes play out. This Agency of Consequence arises, for example, when the party decides to either fight or befriend the ogre. If they befriend the ogre, they have an ogre friend to help them in future scenes.

But, truth be told. Agency of Consequence also arises in a very organic way in D&D and Pathfinder through combat. At least, it often does unless dumba$& GMs put a stop to it. Let me explain.

As I’ve mentioned before, a combat scene is a scene filled with decision points. The players make a lot of choices during combat. And those choices affect how the combat plays out. It isn’t just about whether the PCs win or lose. It’s about how well they win. How many resources they walk away with. If the party gets their a$&es kicked by the ogre, the handful of goblins in the next room might be exponentially more dangerous. And THAT is why action-heavy, linear adventures actually rarely feel like railroads. Combat automatically carries both Agency of Approach and Agency of Consequence.

Of course, that doesn’t mean every player LIKES endless combat. Some players like combat, some players don’t. And you can’t make an entire adventure stand on combat alone unless you have a group that really enjoys that sort of s$&% (though, there are far more players out there that DO enjoy that sort of s$&% than some GMs want to believe).

However, there’s also a way to absolute RUIN the agency that combat affords. And that is when the consequences of battle don’t follow the party. For example, let’s talk about the standard wilderness travel tradition of “no more than one encounter per day.” See, most GMs figure that wilderness travel needs something to break it up, right? And many GMs like the idea of random encounters. But most GMs also realize that dragging out wilderness travel with days of random encounters will get boring. And, as a result, you end up with these weird sort of patterns. And general, GMs follow one of two patterns. Either there will only ever be one encounter on the road, no matter how long the trip is OR there will be at most one encounter every day of travel. And both of those absolutely suck for Agency.

See, a random encounter is precisely that. It’s random. The players have no say in what they encounter or the circumstances of the encounter. It’s all down to die rolls. And most of these encounters are just random combats, so they have no say in how they deal with the encounter. And because they will never have more than one per day, any encounter that doesn’t actually kill a PC will have its consequences completely erased by the next encounter. And thus random encounters manage to erode any feeling of agency from any combat that isn’t potentially fatal.

It would be much better if the party could do things to avoid random encounters. Or take risks that increase the chance of random encounters. And the wilderness travel system in 5E SORT OF accomplishes that. A little bit. It would also be better if random wilderness encounters afforded more choices for how to approach them. That is, maybe when the PCs encounter the goblins on the road, the goblins don’t really want to fight any more than the PCs do. But they also don’t want to back down or show weakness. Or let the PCs through their territory. Whatever.

But that is one LONG digression from the basic point about agency in linear games. The point is that agency in a linear adventure has to be built into most of the scenes to make up for the lack of Agency of Structure.

Planning Scenes in a Linear Adventure

So, when you settle down to design a linear adventure and plonk down the scenes along the path, you need to constantly be asking yourself “how are the players free to engage with this scene” or “how does this scene affect the next?” And you need to be careful about accidentally erasing the consequences of scenes. If you’re designing a linear action dungeon, for example, the worst thing you can do is let the PCs freely rest and recover whenever they want. Because that’s not really a choice. Given the freedom to rest and recover without consequences, the PCs would be stupid NOT to rest. So, right off the bat, one of the things you want to ask is “what are the consequences of retreating and resting?” That choice better have an impact.

Beyond that, note that every scene doesn’t have to be drowning in agency. It’s okay to have the occasional scene that is as simple as “scale this wall, no there’s no way around it, just climb.” In fact, we’re going to talk about pacing eventually. And the trick to pacing is to vary both the speed and complexity of the scenes in your adventure. You want to mix up slow and fast scenes and complex and simple scenes.

As you’re mapping out the structure of your adventure, ask yourself where the agency is. Is there a scene with an obstacle that the party can deal with in a bunch of different ways? Is there a scene that changes the way a future scene plays out? Are there scenes that do both? Can the players easily escape the consequences of a scene? If you can point to a few spots for agency in your adventure, you’re fine and you can start planning out the individual scenes.

Except for one more issue.

Failure in Linear Adventures

Remember when we first talked about outcomes in adventures and I pointed out that a GM has to plan for multiple outcomes? Desired and undesired? And I talked about how the possibility for failure was important? Well, now that we’ve discussed agency, you understand exactly WHY the possibility of failure is important.

If you’ve reached the point where you are planning the structure of your adventure, you’ve already decided what the various planned outcomes are for your adventure (and accepted the possibility of a crazy outcome you didn’t plan because your players did some bats$&% insane thing halfway through). And you have already decided how you’re going to present that bad outcome so that the players know they failed and you don’t end up with an adventure that won’t end.

If you haven’t done those things, you’re writing a s$&% adventure. Stop.

No, here’s where things get tricky. Linear adventures are very sensitive to failure states. That is to say, because there’s only one path through the adventure, if the PCs f$&% up at any point along the way, they usually have no other way to get to the end. Now, some morons will start screaming about “fail forward,” but you already know that I won’t put up with that bulls$&%.

There is absolutely nothing wrong with the success or failure of your adventure hinging on a single scene. A linear adventure is, by definition, a bottleneck. It’s fine. It’s fine. As long as you are okay with the adventure failing one scene in. The thing is, though, most GMs are not okay with that. And I understand that. It does sort of suck to end the night’s session after scene one because the players f$&%ed up and couldn’t recover.

But, again, there is NOTHING inherently wrong with that happening. This is all about what you PREFER.

So, as you start dropping the scenes along the path of your adventure, you also want to consider what happens if a given scene is a failure. Does all forward progress stop? Is the adventure over? Or do the players come out of it dragging a consequence into another scene? And if the adventure IS over, how do the players know it’s over? What happens next?

Now, obviously, the most obvious failure state is that everyone dies. And that’s a possibility in most D&D adventures. And that usually brings the game to a screaming halt because everyone needs to make new characters. And again, that is fine. It’s okay if the adventure is simply a gauntlet and failure means death. It is really okay. Don’t let anyone tell you otherwise.

But, if your adventure isn’t just a life-or-death gauntlet, you need to decide which scenes end the adventure and which scenes offer a way to keep moving forward. And, in the end, if you can’t wrap your head around that, you usually end up building a branching adventure.

Usually, your adventure failure state will help you figure out how to deal with encounters that have bad outcomes. For example, if the party is tracking down a foe and they take too long dealing with an obstacle or can’t get past an obstacle, the quarry will get away and the trail will go cold. If there’s a time limit, the time limit will expire. The princess will be sacrificed. The temple will sink back into the desert sands for another century. Whatever. If the party is evading an enemy, the enemy catches up.

The point is, as you’re planning the structure and the scenes in your adventure, you need to have one eye on how failure comes up in your adventure.

In the end though, if you wrap your head around the failure issue and can build agency into your scenes, a linear adventure can be a hell of a lot of fun. They are tricky to write, but once you get those issues down, they are easy to run and a great way to build your first homebrew adventure.


Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

18 thoughts on “Coloring Inside the Lines: Linear Adventure Design

  1. This article is great! This is my favorite series of articles. After all building custom monsters and stuff is great but if I don’t build an adventure to put them into it all goes to waste.

    This was perfectly timed because my cousin and I are building a Starwars adventure to celebrate the new film. Hes never DMed before So this was kismet!

  2. This is a fantastic article. I can feel my brain brimming with ideas and questions after reading this. Looking forward to the next in the series.

  3. I liked this article a lot, but one thing I was hoping you’d cover in it was linear adventures in a non-dungeon environment. It seems to me that it is much more difficult to pull off a linear structure outside of a classic dungeon without running roughshod over player agency. In a dungeon you can simply say a room has only two doors: the one you came in and the one leading to the next challenge. Outside of the dungeon, though, it seems like there would always be the opportunity for the players to come to an obstacle and say “we backtrack and go around it.” And maybe the answer really is that if the players choose to avoid it in that context that a failure state has been reached. What are your thoughts on non-dungeon linear games?

    • Asked and answered 😉
      https://theangrygm.com/every-adventures-a-dungeon/

      As long as you structure your scenes linearly you end up with a linear adventure. Think of it as a murder mystery in which each scene gives exactly one clue leading to the next scene.

      BTW, backtracking and going around is just player agency. Ways to overcome said obstacle (a cliff for instance) not a real branch in the adventure.

    • Good point. Non-dungeon adventures can certainly be linear but they will likely feel more forced. Say the players want to get passage on a ship, and they must have permission of the king. But maybe they decide it would be better to impersonate sailors, or stowaway in the cargo, or buy their own ship or whatever. You’re prepared for a “Bargain with the King” encounter, and they want to go a different route. Maybe that sort of challenge only belongs in a branching structure. Or maybe the way to look at it is to scale back and see “Get on a ship” as an encounter and the different ways as approaches.

      • This seems like the correct approach to me. In a more explorable area, each encounter might consist of many approaches. Just because the PCs can skip the king’ permission scene doesn’t mean they don’t have to cross the ocean. It might be a sort of linear superstructure with a branching/open encounter within it.

      • The players deciding how to get on the ship doesn’t make the adventure non linear. As Angry said a linear adventure means that scene A will always go to scene B. What really matters is how the players decide to get from A to B. Then you as a Gm weighing consequences from there actions. I do have to say if you see one of the consequences of failure being thrown in prison then the adventure becomes branching because it is no longer scene A to B. But scene A to either B or C.

        • It could still be considered linear if being thrown into Jail is a failure state. Either it makes a later scene harder or makes the rest of the linear adventure not possible at all.

        • But in my example, the players aren’t going to B (“Bargain with the King”) at all, but bypassing it to go to C, perhaps even by adding a new scene.

          Is it still branching if the branches come together at the end, that is, only one possible start and climax but many possible routes to get there? I think so.

          • The problem, Randy M, isn’t that B is “bargain with the king”. B is “get across the ocean to the plot over there”.

            The bargaining with the king is the same as stowing away or stealing their own ship and sailing it, options (Approaches!) to getting over the ocean.

    • An escort mission for example. The escorted person will go from A, to B, to C regardless of the choices of the players since they are just along for the ride, not calling the shots about the travel.

      An adventure where their objective is to find the kidnapped kid of the count. The first scene would be when they are hired, then they are given a clue that takes them to B(say it’s a witness), the witness says the kidnappers were goblins who where seen leaving to the ruins(scene C), the ruins themselves are a dungeon so you get a number of extra scenes.

      Basically, every adventure that goes from one scene to another without the players calling the shots as to which is the next scene.

  4. Nice article! I’ve been playing with the plot of a short, linear adventure I want to run with 5th edition rules. This is great food for thought. One note about random encounters in the wilderness though: I don’t like falling into the same old random encounter patterns in the wilderness at all. So, for each region the party travels through, I make a chart of things that actually live there. One or two big monsters with lairs, and a variety of monsters that are just part of the local ecosystem. When the party is traveling in the wilds, I then pre-roll the random encounters for the ammount of days their traveling through any given area, including an extra two or three, in case the trip takes longer than expected. Then, if any encounters come up (sometimes none do, sometimes lots), I pre-make the “random” encounter, including rolling stats, creating various signs like fresh kills or dug up bushes, or even little totems for intelligent monsters, and this way the players get little clues, and you have investigation scenes and the players get tense along the way. Eventually they usually decide if they want to travel around said area (taking longer), thus avoiding whatever is leaving the signs, or they can even just lay low hoping whatever is out there moves on, those kinda of things. I also roll for the time of day the encounter will occur; to keep them guessing in that respect as well. This all takes a bit of work but I find it creates continuity in your world (ecosystems, territories etc), and has the benefit of players really thinking hard about their next jaunt through the woods or over that next mountain. Plus, your random encounters end up coming across as being just as “professional” as your set piece encounters. Admittedly, it’s work, and no matter what you do, sometimes the players will do things you just had no chance to plan for. But following these lines I’ve had players say they like the way I run wilderness and travel. It’s a good feeling to know they enjoy it, making it well worth the extra prep. And deep down, as a DM, you can still revel in the fun of rolling your random encounters.

  5. Funny enough, I once gave my group a moment to rest between battles. Instead they decided to push on with their injuries. I don’t remember giving a great sense of urgency, they were just trailing someone.

  6. “Remember that players only see the things that actually happen in the game. They don’t know about the treasures they didn’t find, the paths they didn’t take, the monsters they didn’t encounter, and the clever plans they didn’t think of. At the end of the day, it’s impossible for the players to know how much freedom they actually had.”

    This is important. My first great game as a player was a linear adventure, to escort a low level wizard to “see a friend”. Turns out he was there to release a demon. We didn’t realize at the time, I did saw something fishy but the other players cut me off because “we have to follow the adventure”. Long story short, it ended in a TPK. We still had a great time as it was a really good linear adventure. But then the GM showed me the adventure and it turns out the good ending was to stop the wizard, go back and return the object that trapped the demon to his wizard academy, but since the players didn’t know going back was a game option, no one stopped the wizard.

Leave a Reply to Mike RannieCancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.