Story structure in games


There are many materials for authors and film makers that teach them of story structure - Act 1, Act 2, … pitch points, decisions, returns, changing etc. The Monomyth and Dan Harmon Story Circle.

I wonder how viable it is to implement into our RP games. When someone makes a more or less railroad it might be a good pattern and advice on how to keep things moving, but what about PbtA games?

I as MC have limited influence on the story, I make my moves, make my threats, players roll misses in very unexpected moments and thus creates a plot twist at any moment.

Have you used Story Structures and in what games? What were the results?


I often find myself utterly lost in the academic writing around narrative, structure, etc. It’s all interesting and grist for the mill, but how do I integrate this vast body of advice and structure? And moreso with game design…

Still as with all things, Jason Alexander has written some good material. Probably the most relevant piece is Dont Prep Plot from 2009. This forms the cornerstone of a lot of his further work, of which there are many, many other pieces. It all draws from that original theory but he touches on a lot of interesting things in his other work, though they are mostly focused on scenario design/structure (and where do we draw the line there?).


My normal takeaway is to remember that plot is character. Plot is character! Characters with motivations will make plot and most people will shape reasonably effective narratives.

There’s a LOT of good stuff to be mined in screenwriting books, but a thing to remember is that structure is better approached as descriptive and diagnostic rather than prescriptive–that is, it’s not so much “ooh, page 80, time for the flip!” but “hrm, things look flat in the third act, maybe I need to have a bigger reversal?” or “The audience is going to expect Nice Guy to get the guy, so you should sell your anti-hero a bit stronger.”


I do use story structure but I use it a bit differently as I think the tools you’d use for written fiction don’t transfer as strongly to a story being created among a group of people in the moment.

This comes from a long time of doing long form improv and, while that’s given me a pretty strong feel for story as its being created, what I focus on during the session is less academic than breaking it into an act structure.

If I’m looking at a four hour session, I tend to want the following:

  1. Inciting incident—if it’s me kicking something off, I tend to want to have the players active in what’s going on pretty quickly. Within about 15 minutes, I want to hand things over to them so they can start to run.

  2. Building threads—it’s not unusual for me to have games where the players are separated. I bring each of those threads to a decision point or mini-cliffhanger before switching to the other story thread.

  3. At about the 60-70% mark, I start looking for how to bring everything together for the climax of the episode and leave it on a resolution or cliff-hanger. I start to move those threads towards one another.

  4. If the characters are separate, I try to bring them together in that last 45 minutes to an hour and thematically tie the whole thing off.


I am an outlier here: I don’t really care about the structure of the story, at least not to the extent that I try to drive it one way or another. I trust the players to play their characters honestly based on the situation and I do my best as GM/MC to portray the world honestly. At the end there will be one or more stories that occur - but they will be seen in the rear view mirror. I care that the narrative is good - the story (or plot if you like) will be the thing our brains do to organize what we saw. I trust in our pattern matching machines to apply story to the narrative.

That said, if my players want to steer things in a specific direction because it better fits their preferred way of understanding the narrative (ie they want some sort of closure or similar) I am happy for them to do so.

I’ve noted it in the past but I am a big fan of David Lynch. His work rarely uses anything we’d normally see as plot. But our brains will create a story out of what we see in his work. He trusts the viewer to do the heavy lifting of meaning.


I have a background in theatre so I wholeheartedly endorse the idea that characters drive plot. Transposing this to ttrpgs, on the GM-facing side, having NPCs with clearly defined motivations is key.

Setting is also important (and sometimes setting can function as a character). Having interesting set pieces that the characters can interact with is key too.

A guy named Dave Stebbins said a few years back on ENWorld that writing an RPG adventure is like writing half a story, and the players through their PCs will fill in the other half. That’s always stuck with me. The final piece then is PCs with motivations and agendas.


I also try very hard to avoid story structure when designing or running TTRPGs. I think it comes from my childhood joy at playing them and feeling the sense that as a player I drove the story and could take it anywhere I wanted.

Of course to a great extent this is restrained by the trappings of genre and the difficulties/messiness of a having multiple storytellers building off the setting, limitations and constraints imposed by the GM. There may be a lot of potential to direct genre in those constraints - view adventure design and mechanics. This though strikes me as radically different then trying to overlaying filmic or novelistic story structures (or even Structuralist story elements) on an adventure.

Of course there’s also more cooperative storytelling play where players and GM (if there is one) are engaged in playing directly with the genre and meta narrative. That is when the locus of play is less the characters and their actions and more the joy in how well a specific style or genre of fiction is cleverly emulated.

I don’t know how well the two styles mix and certainly the mechanics that support the two types seem difficult to interchange.


This is a really interesting topic to consider, both as designers and as players.

I find it particularly interesting that a rift exists among writers, too - not just gamers. There are authors who believe that writing is only truly good when it’s outlined, prepared, edited, and carefully put together to hit all the right points at all the right times. Some people go really far with pre-plotting and story structure, creating stories in a nearly “paint by numbers” style. (I’ve recently run into some gamers who do so, as well!)

There is another school of thought, though, where plot is emergent and unplanned and comes about intuitively. You don’t plan ahead; you follow your instincts and the characters and you see where that takes you. Stephen King has written about this fairly extensively, and he’s certainly not alone.

I think it’s pretty interesting to see a spectrum here and to design and play everywhere along the continuum, so long as you’re honest about what you’re doing. (Roleplaying culture has long had a tradition of misleading the players about what’s actually going on, with the whole idea that 'the GM writes the story, and the players act out characters within it", which can easily lead us into pretty questionable and socially dishonest territory.)


My own style tends toward disclosure and partnership with the players. I’m very upfront about themes and narrative beats that I’m looking for, and ask my players to engage with their own thoughts and work to bend the narrative in that direction when they see opportunity to do so, but I don’t write plots. I try to go with a “writers’ room” approach, GM as show runner and players as writers with full agency of the principle characters


I’m new here and I can’t figure out how to quote Shanel but “our brains figure out the story” is so apt.

I never go in to a game with a plot in mind but I always come out of a game with a sense of the story. It always seems to resolve into something meaningful if you want it too.

I don’t think that means learning about plot structures and such aren’t useful, I think we internalize good writing when we see it a lot and that shows in the emergent story. A more “sophisticated” group will inherently create a more sophisticated story in the end without even intending to.

I also want to echo the disclosure and partnership comment. I find this works best for me to simply be direct with how I see things going, sometimes this means I realize that another player and I are on different pages but that’s a good thing. If I can see how we are thinking of the narrative differently then I can adjust mentally if adjustments are needed. And if not we can work towards a very satisfying ending.

Talking meta about the story even in the moment doesn’t ruin my emersion, though it might for others.


A key distinction is, I think, that story structure in a game is also mechanical structure. Or, at least, when it’s done right it is. You can definitely just plan some novel-ish plan out in your head and then bully your players into following it, but that’s broadly considered both bad GM’ing and bad game design.

Rather, I think it’s more useful to ensure that your gameplay matches your story; in RPGs, what you do mechanically-speaking is also what you do narratively-speaking. This sounds obvious, but it gets a little bit less so as you move further out into the abstract.

Take, for example, Blades in the Dark’s downtime. You don’t get into heavy roleplay during downtime, a lot of it is mechanically-oriented, and it’s all very high-level and undetailed. Despite that, however, it’s still critical to the story of a game of Blades in the Dark that there be designated downtime in between each of the missions—it gives the story a natural tempo.

While it’s technically true that you can’t avoid downtime in Blades, I’ve yet to meet any players or GMs that feel ‘railroaded’ by it. It’s a thing you do mechanically, and it makes sense, and so it naturally becomes a part of the story structure. It gets broken and played with and a bit gummy sometimes, sure, but it’s still a core part of the structure.

Anyway, the point is, if you define the beats of your story in mechanics instead of traditional narrative structure, your game and story will both make more sense.


I’ve been wondering if Fate’s Milestones could not be used/adapted to suit a 3-act, 5-act, hero’s journey kind of storybeats model.

The problem is in how milestones are used. I haven’t actually played in a game where we reached milestones (only played pbp and with the entire fate point bargaining and invoking and compelling, a once a week post game slows down quite a bit) so I’m not sure. It would also assume a main protagonist, which also is not quite what most people want I think.

The other thing is to have the milestones separate for each character, and have each of them have their own little arcs, which could work, but then it’s be more like a series with episodes, rather than a movie or a book.

There are movies (and certainly books too) where different characters each have their own hero’s journey or at least character growth, but most of the time these are not on the same moment. Take for example Merry and Pippin in the Movie Trilogy of Lord of the Rings. Their arc only really starts very late in the movie. One becoming brother in arms with Eowin, the other swearing fealty to the kingdom of men.

If you translate that to a game, their players would have to wait a very long time to get to be something other than comic relief sidekicks that get the protagonists in trouble (You fool of a Took!).

I guess the answer could be a game where you’re not playing a character, but playing the story and have a cast of characters that are for everyone to use. A bit like Microscope, but without the jumping forth and back in time part.


I don’t like story in my roleplaying games, but something I was impressed by (haven’t tried it yet) are the mechanized character arcs in Chuubo’s Marvelous Wish-Granting Engine by Jenna Moran. Playing a knight for example, my arc would look somewhat like this:

Quest 1 is about figuring out what you want to be.
Quest 2 is about the vice that gets in your way.
Quest 3 is an experience that shakes up your role.

Then, if you haven’t got enough XP to finish out the Arc, or just plain want to do more, you can go on to do:
Quest 4 is about making up for your mistakes.
Quest 5 is about becoming something better than you were.


@Julian I love this! This is what I was thinking about. I really should get to buying and reading that game sometime soon. Especially since I like to role-play with kids, and tend to want to use it in class (which I think is not viable for Chuubo, since it basically need to be resolved in about 30 minutes… Restrictions, man, sometimes they suck.)


All games necessarily have an emotional arc to them. Not just RPGs or games with stories attached, but all games from chess to Dance Dance Revolution have a changing emotional tone, where hope and fear and risk and reward change and interact as you play. Successful games have satisfying arcs, where the the stakes get higher over the game and then resolve. Story games, RPGs and narrative videogames add a layer onto this, where there are characters and plot tied into the mechanics. Which means that there is a story structure naturally built into every game.

The question isn’t “can you build a story structure into a game?” because every game has a structure built into it. The question is what to do with that: do you deliberately build a structure into the game explicitly, or let it emerge organically? In my experience it works best when the game designer consciously works to create the structure they want into the game on a fundamental level, whether or not the GM and players ever realize it. So you can have the structure built into the mechanics, but you don’t necessarily see how it influences play and you aren’t necessarily making specific decisions based on what act the game is in. (E.g., games like My Life With Master and Misspent Youth structure the narrative in a fairly overt way, whereas leveling in D&D and revealing memories in Psi*Run or epidemics in Pandemic work to structure the story in less explicit but still fundamental ways.)


@Pawel, I think it’s certainly viable to impose external narrative structures on RPGs, because most are flexible enough to accommodate such impositions. I regularly do it in 4-hour PbtA one-shots:

In the first 20 minutes we create characters and setting, in broad strokes.

The next 40 minutes comprise “Act I,” in which the status quo is established through play.

At around the 1-hour mark I offer the PCs a big choice or opportunity improvised from what’s been established (the “first threshold”). In Star Wars: A New Hope, this is when Luke find Aunt Beru and Uncle Owen dead, and chooses to leave Tatooine.

The next hour is the first half of “Act II,” where the PCs deal with the immediate consequences of their choice.

At around the 2-hour mark I try to make sure they cross a “second threshold” – enter a new emotional phase or physical location that raises the stakes. This is when Luke et al sneak onto the Death Star.

The next hour is the second half of “Act II,” where the PCs deal with further and more challenging complications.

At around the 3-hour mark I try to push them across the “third threshold” by hitting them hard where it will really hurt – someone important dies, or betrays them, etc. This is where Obi-Wan gets cut down in front of Luke.

The last hour is the PCs dealing with all of the fallout and last desperate challenges, while I look for ways to tie things together in hopes of putting a bow on it by the end.

This approach was originally inspired by Jon Aegard’s “Dragonslaying on a Timetable” idea. Over several years I’ve gotten pretty good at hitting the marks. The key things for me are not having any preconceived ideas beyond a starting situation (e.g. “You’ve crash-landed on a strange planet”) and asking the players lots of questions so I can improvise all of the important bits out of their contributions. PbtA games are ideal for this tight time frame (my go-tos are Apocalypse World, Dungeon World, Adventures on Dungeon Planet, and Spirit of '77).

I teach in a comics MFA program, and sometimes use games in the classroom to provoke discussion about character motivation, scene setting, and story structure, among other things. Traditional 3-act stories so saturate our culture that we have been trained to expect the pattern (thanks to the George Lucas/Joseph Campbell/Robert McKee feedback loop, 95% of mainstream films hit the requisite beats to the minute), and as a result can sometimes feel like something is lacking if a story doesn’t fit that pattern. (High-five to @shanel for love of David Lynch.)

After we cover the hegemony of the 3-act structure and Hero’s Journey, we talk about alternative approaches, and games are a great way to explore that. As @Nickwedig points out, every game has a built-in narrative structure, defined by its rules. I use Fiasco regularly in the classroom because it creates highly entertaining stories that pivot on a single structural axis (“The Tilt”), few problems are solved, and things generally end in disaster. (Aside: Fiasco is also great for discussing character motivation, when to cut into and out of a scene, how “color” scenes affect pacing, etc.).

I think tabletop roleplaying games offer a wonderful antidote to the hegemony of the 3-act, and should be celebrated for that. While mainstream culture inundates us with formulaic narratives (which of course can be executed well or poorly), RPGs offer a multitude of alternatives. From the “simulationist” pace of old-school D&D campaigns to contemplative nature of The Quiet Year to the physics-driven tension of Dread, these alternatives allow our brains to break out of the ruts imposed on us by the past 40 years of narrative training and enjoy genuinely fresh forms. We are living in a golden age of improvised collaborative storytelling (or maybe “story remembering” to follow @shanel’s idea?), and it’s pretty awesome.


While there are many story flavors, the more mechanics get in the way, the less immersion you may get, though it may depend on gameplay and how much things the GM and players can handle in their heads without spilling math and rules into the fiction. What I’m saying is that the more you try to control the story to produce a specific emotion in the participants, the more tools you will need to hide the strings and rails.

We all know it’s possible, there’s almost a tradition of making the GM responsible for everyone’s entertainment and support railroading with rules, modules and advice from the designers. That’s old though. In the last 20 years it has been seen as being unfair as it takes choices away from the players -which totally does- yet I’ve seen posts validate it as another way to play the game. Certainly several youtube shows that feature people playing RPGs seem to follow that heavily scripted tradition.

Personally I started there, but as I learned to impro more, trust more my group of players and got a bag full of GM tricks to expose and spot my players expectations, I realized I’d rather like to have fun by not knowing how will a session end. Sure, I prep kickers, obstacles, color, but everything else from setting details to major NPCs, rough maps, fronts, motivations, etc. comes from the players. You just need 8 or less questions to get everything rolling and give the players an enhanced version of the exact experience they want to have, even if they didn’t knew from the start what was it gonna be like.

If you got your questions right you can even avoid blank page syndrome by asking players about their choices, how they envision them and come up with follow-up questions from whatever they give to you. Taking their answers seriously no matter what they are and building upon them on the spot to build the setting along with them often gives the players the confidence to keep going and input better and better material.

Having all this from the start allows you as a GM to easily search and consult players for their PCs motivations, so you can build NPCs to provide proper, meaningful conflict. Drop their names at the kicker, show the effects of their actions next and have them show up at the end and you’re guaranteed to get a story arc of some sort from that. It’s been said before; well, motivations are really that important, they make the game. The story will be perceived by the players either when they remember the session or even at some middle point in play. If the latter happens you will see them steer it towards the ending they consider the best. When that happens you just follow them, put challenges in their path and add some twists that make their worries even worse, but still keep them going, either in the same or the opposite direction (as in the person who hired you was the BBEG all along!)

You can get the same results in a more organized fashion by using a 3-4 act structure (check japanese rpgs like Double Cross or Kamigakari) though it doesn’t exactly feel completely natural-like flowing for my tastes.


Can you write a bit more about set pieces?
Is it like an action scene who? where? what? or something else?

Also how does having a set piece relates to having a good story with structure?


Some games I use a more freeform approach to the narrative, esp something like D&D where a dungeon delve is pretty much just go in, explore, kill, loot, and get out. For the more RP focused scenarios and scenes, I try to have beginnings, middles, and ends.

For more narrative focused games, esp. Star Trek, I develop the adventures in a structure more like the TV show–teaser, acts, captain’s log or personal logs, high stakes, drama, turning points, climax, etc. Varies depending on genre, game, and mood.


Have you used Story Structures and in what games? What were the results?

I just played Durance with a couple of people who had no previous experience in roleplaying. Although I don’t think any of them has seen a d20 in real life, they were experts in finding the dramatic heart of the scene, escalating the action and knowing when to cut. It took very little prodding to encourage them to use camera moves and other vocabulary from visual media in their descriptions (“We cut to the character hunched over a table…”)

This is because serialized TV is more popular than ever. People watch good shows and have become very literate in the structure and conventions of those shows. This is a huge asset for the hobby. Designers, take note.

My caveat is that If you want to get more theoretical, make sure you draw from good, applicable theory. The narratives roleplaying games produce typically don’t have much to do with those of feature-length motion pictures or commercial novels. Story theory from theater and especially improv theater is like to be more useful.

Two books I want to recommend are Impro for Storytellers by Keith Johnstone (for applying principles of improv in creating narratives) and The Scene Book by Sandra Scofield (for understanding how scenes are not about conflict).