Dice Tales: Game Planning I – Arcs, Acts, and Chapters

Roman twenty-sided die(This is the thirty-fourth installment of Dice Tales, an ongoing series of posts about RPGs as storytelling.)


I like running arc-plot games. By that I mean campaigns where the GM has a central plot in mind, and the intent is for the whole thing to reach a conclusion within a reasonable span of time.

I suspect I would have always had this tendency, because my favorite type of novel series (or TV show or whatever) has the same kind of structure. I’m less satisfied with the type of ongoing tale that’s going to keep on trucking until people get bored and wander away; that approach it doesn’t deliver the kind of payoff I’m looking for. But my reasoning is also practical: I’ve been in a small number of campaigns that drew to a satisfying close, and a much larger number that just kind of . . . . stopped, on account of scheduling conflicts or the GM being too busy to run or other OOC factors. When I set out to run my own first game, I was bound and determined that I would not have one of the latter type. By god, I had a story in mind, and we were going to tell that sucker from beginning to end.

Which we succeeded in doing, not only then, but in subsequent games. My second campaign also reached an end; my third stopped after one act after a planned three-act structure, but my co-GM and I had arranged for the start for it to be sufficiently modular that you didn’t need all three acts for it to be a complete story. My current campaign is the fourth I’ve run, and although it is still ongoing, I have confidence that we’ll bring it to a proper end.

Doing something like this means you need to plan the story on a bunch of levels at once, and I thought it might be useful to look at what exactly those levels are. To illustrate, I’ll be drawing examples from my first two campaigns.

The top level is the arc plot. This is basically the concept of the game: I don’t propose a new campaign to players until I know what the arc plot will be. For Memento, my first (Changeling: The Dreaming) campaign, the arc plot was that the PCs had spent multiple lifetimes engaged in a massive alchemical working to create the Philosopher’s Stone. It started in the modern day, with them finding out they’d all known each other before; then the bulk of the campaign involved flashbacks to previous lives, filling in the details of what they’d been doing and how they got started, until it returned to the present day to finish the job. (And if you think running your story in reverse like that is a pain in the neck and not something a first-time GM should attempt, I’ll be first in line to agree. But it turned out well in the end.)

When I say I know the arc plot, I don’t mean that I’ve nailed down how it’s going to turn out. Memento was a highly structured game — the nature of its concept made that unavoidable — but in Once Upon a Time in the West, my Scion game, the arc plot was that Columbia and Uncle Sam (the gods of the new American pantheon) had become corrupted by an avatar of the Greater Titan of Hunger, a creature known as Manifest Destiny. I knew the game would involve the PCs discovering this fact and then doing something about it, but what they did? That was up to them. I was prepared for them to purify Columbia and Uncle Sam, or to kill them, or whatever. (I don’t know what I would have done if they decided they wanted to ride the Manifest Destiny gravy train and take over the world, but I also don’t run for the kinds of players who would be likely to do that.)

The arc plot tells me what the story is about, and guides my decisions at more granular levels, but it doesn’t fill out a year-plus campaign on its own. For that, I need to look at the next level, which I think of as the “acts” of the plot.

In the case of Memento, these acts were based on the number seven: seven metals in alchemy, seven sacred trees of Britain in folklore. Each act was a particular lifetime at a particular moment in English history, with a conflict themed to one metal and one tree. For example, the Elizabethan segment was the apple/gold chapter, involving a faerie queen obsessed with immortal beauty. (And if that sounds familiar, yep, this is what inspired Midnight Never Come.) Because I wanted Memento to be a relatively short campaign, each of those acts lasted for a month, i.e. four sessions. Add in the modern-day segments bookending the flashbacks, plus one distant-past bit that set up the whole thing, and the campaign played out in ten acts total.

Which is quite a high number for a campaign that only lasted about eleven months. Once Upon a Time in the West ran longer, but only had three acts: one for when the PCs were divinely-born heroes, one after they became demigods, and one after they ascended to full godhood. But it would be more accurate to say it had three top-level acts, each comprised of four sub-acts or chapters. In Scion there’s a stat called Legend, which measures your divine power. You’re supposed to buy it with XP, but not only does this create mechanical problems, it seemed to me like Legend ought to go up when a character did something truly, y’know, legendary. So I had a plot for each point of Legend, on roughly the scale of Memento’s flashback acts, but with more room for other stuff to happen around it. For example, one of the late-stage demigod plots involved the PCs freeing the spirit of fellow demigod John Henry, who had been imprisoned at Promontory Summit by the golden spike that completed the Transcontinental Railroad.

Acts and chapters help me break my arc plot into bite-sized pieces. There’s a quote from some writer (I can’t find the attribution right now) about how writing without an outline is like driving at night with the headlights on: you can only see a little bit of the road in front of you, but you can drive the whole distance that way. I don’t have to know from the start how the final confrontation is going to go down; I just need to know what I’m trying to introduce or resolve right now, and keep an eye on how that’s feeding into the larger picture. (If in fact it does feed in: sometimes you introduce a thing just because it’s cool. My Scion PCs helped Paul Bunyan out mostly because we couldn’t possibly tell a story about mythic frontier America without including him.) I might know in broad terms what the next chapter will be, and in even broader ones (like, the one-sentence level) what the next act will be, but the details don’t get nailed down until we’re closer to the that part of the story.

Of course, even at the chapter level, we’re still talking about multiple sessions’ worth of game. But there’s enough to say on this topic that I’ll save the finer-grained levels of planning for next week! In the meanwhile, I’d love to hear other GMs share how they conceptualize the bigger-scale aspects of their games!


Posted in Gaming Tagged , , permalink

About Marie Brennan

Marie Brennan is a former anthropologist and folklorist who shamelessly pillages her academic fields for inspiration. She recently misapplied her professors' hard work to the short novel Driftwood and Turning Darkness Into Light, a sequel to the Hugo Award-nominated Victorian adventure series The Memoirs of Lady Trent. She is the author of several other series, over sixty short stories, and the New Worlds series of worldbuilding guides; as half of M.A. Carrick, she has written The Mask of Mirrors, first in the Rook and Rose trilogy. For more information, visit swantower.com, Twitter @swan_tower, or her Patreon.


Dice Tales: Game Planning I – Arcs, Acts, and Chapters — 7 Comments

  1. For the Skype game I am currently running, I am taking direction from “Primetime Adventures” and Hillfolk and explicitly framing things as episode arcs. The first one for that session was “Mediocre Trouble in Modest Minneapolis.” Now that we are starting another scenario set, I have named it “Definite danger in Doomed Duluth” . If I run a third in this set, I will break it up similarly.

  2. For some weird reason, my RSS reader just today picked up on your series…

    I’m a long-time gamer and author…

    Would love to do a feature blog post, possibly with an interview with you, about this series…

    The thing is, I’d like to make a suggestion before that can be considered…

    Is it possible for you to gather these together on their own page and order their links from 1 to __?

    My experience with blogging informs me that many, if not most, readers need things easy to stay with them 🙂

  3. I know the Changeling LARP was structured as an ode to Babylon 5’s five act arc. We knew where we were headed from the start (the end of this iteration of creation), and we knew roughly what each season would center around. That gave us a structure to hang shorter-term plots and character/faction plots around (and also helped us set our pacing for plots). For me personally, how Whedon handled Buffy was also a huge influence.

    I ran games before Changeling, but I can’t even remember how I thought about them because the Changeling experience entirely reshaped how I approached GM-ing, obliterating everything that came before 😀

    But now… yeah. Metaplot, acts, short-term plots, session plot. I think of novels in the same way (and it’s probably why I struggle with short stories. It’s like running one-shot tabletops – limited scope, limited characters… there’s hardly any room to do anything _interesting!_

    • It helps when your large-scale units aren’t dependent on each other: season five of Buffy being the fight against Glory was in no way dependent on season four of Buffy having been about the Initiative and Adam (though smaller-scale elements therein were connected). I think it would be harder to do the metaplot of Supernatural as a game, because season five is dependent on what happens at the end of season four, four depends on three, three depends on two, and so forth. Or rather, you couldn’t have your five-season plan at the outset of that campaign — not unless your players are okay with a degree of railroading. Otherwise, you have to develop each new segment as you go along.