Category Archives: AdventureDesign

Three Fights vs. Three Acts

A few folks brought up the three act structure in response to the three fight model, and while I get where they were coming from, there’s a distinction I want to call out that I think is REALLY important.

First, it is SUPER useful to understand the three act model for plotting your games. I would also say it’s very much worth learning other versions, most notably the 5 act model which is very common on television. Understanding these structures and tools can make you a better GM. Not because they provide strict rules to follow, but because if you get why they work on the screen and stage, then you can apply them in your play.

This is deep, valuable stuff. But I fear I consider the three fight model to be much simpler for one reason: it is much, much dumber.

That probably feels like a criticism, but it’s not. There are tools we want to be smart, and tools we want to be dumb, and the three fight model is in the latter camp. Its virtue lies in its simplicity, and I can illustrate this very simply:

  1. Imagine the cast of, say, Voltron. Any version. Pick one you know. If you don’t know Voltron, pick an action-y cartoon you like.
  2. Now, imagine that’s the game you’re running and those are the character’s played by your usual stablemates, with all their tastes and interests.
  3. Next, try to think of three cool fight scenes. Don’t go too deep, just sketch out the idea.
  4. Now set those aside, and instead think of a three act arc for them.

Was there a difference in ease between #3 and #4? If not, I envy you more than a little. See, for me #3 is super easy – I just mash up a few elements (Location, status, enemy type, maybe a gimmick) and I’m good to go. But when we get to #4, I have questions. I need to know more about the characters, the players, their interests and how to hook them in. I don’t want to tell a three act story, I want to deliver the experience the players want, which means I need a lot from them.

Heck, even if I didn’t need more information, I’ll end up struggling with the blank page problem. Building a three act structure is so profoundly open ended that I will be paralyzed just starting. In contrast, a fight scene is a constrained enough idea that I can churn it out.

That is why the three fight system is amazing to me where a lot of other structured models are merely interesting, informative or helpful. It strikes a solid balance between constrained enough for focus, but open ended enough that there are still bazillions of things to be done with it.

Now, having separated these two ideas, I should add that they do eventually grow together. Once you have the idea of three fights in your heart, it becomes possible to do more with the model (such as have the fights correspond with three act pacing, or finding other scene types that you can build as easily and well as you can fights). But I really want to call out that there’s nothing magical about structure or the number, three, the power is in how usable it is.

Three Fight Scenes Redux

Back of heads in an elevator.  Precursor moment to an amazing fight scene in Captain America: The Winter Soldier

The original Feng Shui RPG (by the ever-brilliant Robin Laws) had the absolute best adventure design guidelines of any RPG I’ve encountered. They boiled down to this:

  1. Think of 3 really awesome fight scenes.
  2. Apply some loose connective tissue to move the characters between them

That’s it. There is no 3.

Boiled down to two bullets, it’s is a little bare bones, but Feng Shui gave a LOT of guidance on what makes a good fight scene, and enough (totally wacky) background to easily justify the connective tissue. Taken as a whole, it made it staggeringly easy to write up solid, engaging adventures without undue strain on the GM. If you want to dive into it more, I’ve got some more word for you.

A lot of people who have used this model have translated it across to other games (I know I have) but it’s always interesting to me the ways in which it can be easy or hard to do so. It reveals some interesting things about assumptions of play that feel like they could unpack indefinitely.

First off, let’s consider the fight scene. This is a staple of fiction and play, but it’s worth considering what this means in play. That’s opaque, so let me approach it another way: is every scene with a fight in a a fight scene?

Technically? Yes. But in the way that Feng Shui meant it, not really. Those fights are Set Pieces1, which is entirely apt because Feng Shui is a game about Hong Kong cinematic style fight scenes. They are big, have lots of moving parts, many points of engagement on a mechanical and character level, and they’re what players are here to do2.

It’s also going to take a lot of time. Which is good. Time the budget we have to spend on things, and this approach explicitly “spends” more on the things that are most important to the game.

This is part of why it’s relatively easy to map the three fights model to other games which also have set piece fight scenes. It worked spectacularly well for Fourth Edition D&D, and there are a number of other fights games that it can do equally well with.

But beyond that, there are two difficulties with taking the model further afield.

The first is logistical. Feng Shui has the benefit that it’s 100% cool for all of its set pieces to be fight scenes, but if you take it to another game that may not be true. Other games may want social, investigation or exploration set pieces. In that case the underlying idea (three set pieces and some connective tissue) still absolutely works, but that is definitely a less simple model. It will take more energy to think of the variety of set pieces, they’re less modular and interchangeable, and they’re harder to write up3. In short, it’s more work.

That doesn’t mean it can’t be done – it can, and quite well – but it’s hard to learn to the point where it’s comfortable. If you learn it with fight scenes and then expand your repertoire, that’s more practical than doing it all at once.

The second barrier is philosophical. “Loose connective tissue” isn’t going to fly in every game. The reason I say this works very well for 4e D&D, but don’t call out other editions is because dungeon crawling is all about very explicit connective tissue. Playing through and engaging the transitions and choices along the way is what brings some people to the table. Tables like that want to spend their time budget differently, and you need to plan for that.

On the other end of the spectrum, there will always be players who will treat any level of GM direction as railroading. They expect to be making the decisions about what’s important and taking the game in those directions. Again, this calls for a different set of “budget” priorities, so the GM is better armed for flexible response.

Now, for all that, there is a lot of space available between these objections. As noted, the model can work beyond fight scenes, and while it isn’t necessarily suited to full player authorship, it can work with even very empowered players.

The key is that the set pieces are what the players are these for. That is what justifies their budget in time an energy. If they are what everyone wants, then it will all click. If they aren’t, the model will fail, and in doing so will probably reveal a deeper disconnect.

Anyway, if you are a GM and feeling skeptical about this, I encourage you to just try it. Whatever your game is – even if it’s not normally very fighty. Just think of three awesome fight scenes that would be super fun to run4, then loosely sketch out how they might be connected. It will take you less time than you think (especially if you just sketch out the fights – don’t go full stat block until you’re happy with the structure) and I am willing to bet you will be surprised at how well it hangs together.

  1. In film, a Set Piece is a scene requiring a lot of work to plan and shoot, usually at great cost. Since money matters, you only really want to do this when there’s a real payoff. While this is a necessity of budget, it is also a fruitful constraint, since it works naturally with an upswing in focus with an upswing in importance. ↩︎
  2. Critically, that does not mean it’s the only thing they can do. There is still space for other awesome things, but if none of those materialize organically, then the game is still going to deliver these fight scenes. ↩︎
  3. Howso? As a hobby, we have honed the technology for writing up a fight scene to a fine art. Stat blocks. Maps and Zones. Tactics and roles. We know how to put information on paper that someone else can read and know how to run that fight effectively. We do not have anything like that kind of polish or sophistication for communication and running other scene types. ↩︎
  4. Tip: The more wildly diverse they are, the better. It serves as a kind f inspirational fuel when you think about connecting things. ↩︎

When the Campaign is the Game

One thing that Blades in the Dark has really made me think about is fruitfully constraining the setting of a game.  The Blades setting is actually pretty big and I could think of a LOT of games I could fit within it (up to an including Exalted) but as written it ignores all of those options in favor of a very specific focus within a very specific setting.

The focus is not as small as it could be.  The variety of crew types and the size of the city both leave a lot of flexibility, but it’s still a fairly narrow slice, especially when compared to most setting driven RPGs. 

Now, I admit, this is counter to my instincts. I’m a kitchen sink guy.  I want to offer readers as many tools and options as possible, and that can be great for certain things, but it definitely comes at the cost of focus. If I do zoom in, it’s usually in an attempt at brevity, but it’s worth noting that Blades is not a concise book.  I kind of want to deliberately subvert my own instincts and see what designing at a verbose but tight zoom would produce.

My current thinking is that its skeleton would need to be something akin to a certain style of published adventure, specifically a certain style of campaign book which evolves a particular location.  I mean, I guess I could make a game that is explicitly designed to do the Slave Lords arc, but i genuinely don’t see how that would work very well, since it’s just a series of dungeons and dick moves. But more specifically I’m thinking about things like Pool of Radiance or Ruins of Intrigue, where there’s a specific place to play in, with content that unlocks over time.   I could very easily see narrowing a game down and saying “Here are the rules for doing this well.  There is more in the world, and maybe there are other conversations about that, but right now? We’re doing just this thing”. 

So now I’m thinking about other adventures that might work for this.  I mean, there’s probably a whole game to be distilled out of Keep on the Borderlands, but that game might be called “Basic Dungeons and Dragons”.  Dragon Heist is awesome, but I already have Blades and Dusk City Outlaws, so I’m kind of covered there. 

Going to have to go through the bookshelf for ideas, so with that in mind, suggestions are welcome.  Bear in mind, it doesn’t really matter if they’re good adventures (The Pool of Radiance module is…not) but rather that they’re structured in a way that seems like it could contain a whole game. 

Atomic Adventures

Fate AtomThe atomic model of action in RPGs has implications outside of when to roll the dice. Most notably, it has a profound impact on adventure design. Skills make adventure design much easier, because they provide a pre-existing list of challenges to draw upon. Yes, there is still art in arranging them in an entertaining fashion, but if you were to break down the average published adventure, you could easily produce an itemized list suggesting that this adventure is composed of 7 lock pick rolls, 5 find trap rolls, 18 Athletics checks and so on. In this context, combats as discrete events make all the sense in the world, since they are really just microcosms of the same (a certain number of attack rolls and so on).

Consider our ever-popular locked door. We put it in an adventure because we want to give the thief something to do, not because its particularly interesting. In fact, we put in so many such things that it takes only the tiniest bit of plot and motivation to make an adventure work, so long as it just moves from atom to atom.

But if the game removes that roadmap (as more abstract games like FAE or certain Cortex Plus builds do) then that whole infrastructure goes away. With skills (and other calls to mechanics) you could add interest to things which were not intrinsically interesting by engaging them with mechanics. In the absence of that, you are forced to look long and hard at what characters are actually doing in your game.[1]

The first time you strip away this veneer, it can seem daunting. SO MUCH of what classically defines an adventure is removed that you wonder how you can fill the time. If you no longer accepts that fights are disproportionately interesting, then what are you supposed to do? But fear not, the basics of adventure design will still come to your aid.

At its most basic, and adventure revolves around the players/characters wanting something and facing opposition to attempt to get it. When crafting an atomic adventure, it was easy for those to be fluffy things like treasure and adventure, but without the atomic structure to rely on, you realize that a weak motivation opens the floor to all manner of questions, including “is there a better way to get what we need than murder and home invasion?” With that in mind, motive needs to be a bit more rigorous.

Thankfully, this one is not hard to address, because it’s a questions your players should have already answered. If there is something on their character sheet which tells you what the want (like an aspect) then you have a clear signpost. And in the absence of that, you can also just ask them.[2]

Assuming you have aspects (or similar) to draw on, then here is one key piece of advice – you do not want your characters to use their aspects in lame or uninspired ways, and you need to hold yourself to the same standard. Do not look at the sheet, see “Knight of the Red Dragon” on someone’s sheet and just say “The knighthood of the Red Dragon has asked you to investigate those mysterious ruins”. That is lazy and wasteful. If you’re going to use the knighthood as a hook, then the knighthood needs to really matter in this regard, and if that means changing the adventure to reflect that, then that is absolutely what you should do. And then make it transparent – mysterious orders from on high make for crappy motivation.

Once you have a compelling motivation, the hard part is done, because it’s a simple matter of asking “Why don’t they already have it?”. Look over the answers to that, and if they aren’t interesting enough, add some more. Do not think of them as mandatory hurdles to be cleared, but merely things which are true and which must me accounted for.

That’s all there is to it. Simple. Sort of.

Notice that there’s no reference to mechanics anywhere in this, and that’s why I would actually endorse creating adventures without a system in mind. You might go in with certain rules for how the world works (like magic) in much the same way a novelist might, but don’t stress over the specific names of things (though if you want to use an adventure design system, like 5b5 or random tables, go nuts). If you’re used to thinking about adventures in game terms, this may be a hard transition. It feels absolutely creaky when contrasted with your highly refined skills of thinking about atomic adventures, but once you have done it a few times you may find it’s very liberating. Once you can think of adventures that are interesting without the system, then adding system back in will only improve the experience.

Or such is the hope.


  1. This comes in many nuanced flavors. It is entirely possible to have no skill system to speak of but just have characters narratively move from encounter to encounter.  ↩
  2. You also can create motive, by initiating something bad which the players will want to address. This can work very well, but needs to be handled with care. The frequency, intensity and nature of the GM pushes can vary based on table taste and genre (it’s a staple for supers, for example) but if you lean too heavily on them, you risk ignoring the players interests in favor of your own.  ↩