Monday, January 11, 2010

RPG Theory: Episodic vs. serial format

Here's three models of how to run a campaign. Terminology borrowed, again, from other media. This is based on some of the discussion that came out of the game-runner's chat, again (shout out to Kirin for organizing it… you demanding attention whore, you.) This isn't exactly rocket science, but it helps to have this stuff spelled out and written down, as a take-off point to deeper discussion, hopefully.

Episodic: This is like a classic TV show. Each episode is a self-contained story arc, and events in one episode have little to no impact on what comes in subsequent episodes. Like The A-Team or Knight Rider (to use some examples from my childhood) if you missed an episode or two (or more) it didn't really matter, as long as you knew who the characters were, what the basic premise of the show was, and a few other key things like that.

Using this model for roleplaying games is a nice one, especially if 1) you're using published modules that don't necessarily have any linkages between them, and/or 2) characters (and possibly players) have a tendency to come and go, and the same cast of characters can't be counted on to be integral to each episode. I.e., if you've got someone who's attendance is spotty because of a difficult work schedule, working his character in on the nights that he's available is easy if you follow this model, as is assuming that said character simply isn't available on nights when the player can't make it.

Serial: This model, on the other hand, assumes that the campaign is one, giant narrative arc. Much more "real life" in feel, it's not broken up into discrete episodes, and stuff from the very beginning, or anywhere else throughout the campaign can be relevent throughout. This model isn't really well represented by other media, although shows like Lost or Alias probably come closest, since the "episodes" are discrete broadcasting chunks, but not necessarily episodes in the classic sense, with a discrete beginning, middle and end. This campaign model is best suited for very regular play with very regular players who really enjoy the unfolding narrative over time. As with non-episodic TV shows, if you miss an episode, you could be in trouble and quickly fall behind, becoming confused by the narrative flow that's gone on without you.

These two models are really the endpoints on a spectrum, though. I'm going to suggest a hybrid model that leans somewhat towards episodic campaign play, as the ideal model for most groups to adopt. While I'm at it, I'll add a few tips on how to mix and match a few of the good points of each opposite end of the spectrum to improve the experience a bit. Feel free to chime in on what you like and why; i.e., what strong and weak points does each model have, and how they can be improved.

Hybrid (episodic): This type of campaign is characterized by a strongly episodic nature, yet it does have some elements that tie the episodes together more tightly than merely repeating the same characters over and over again. Some examples from recent fantasy literature include Harry Potter and the Dresden Files; each book is self-contained, and includes its own major conflict, climax and denoument, but at the same time there is a thread of "metaplot" that goes through each episode; it's not recommended that you miss one or pick them up out of order, for instance, although if you did you could stumble through it OK.

This is a good model for, again, the group that doesn't always have the best attendence. For most adult gamers these days, that probably means you; real life and outside obligations tend to make the regularity of gaming sometimes spotty for folks, especially as they pick up careers, families, etc. The secret to making this work well, I believe, is to allow the episodic nature to take the forefront, but make a few recurring villains or conspiracies pop up from time to time. Don't be subtle with these references, as time and occasional missed sessions by some players may not make subtle ties evident. Have an end-game in mind for some point down the line, although that doesn't mean you need to be in a hurry to get there. As you approach this endgame, you can have threads start to come together and wrap up; in the meantime, feel free to let them dangle until you figure out what to do with them. Despite that, the "main" action of each episode needs to be under better control; I prefer to give these episodes a kind of narrative structure, with defined acts that are characterized by their place within classic story structure; i.e., a beginning, a period of rising action/tension, a climatic resolution to the main conflict of that episode, and a small denoument that closes out the episode, ties up loose threads that aren't meant to carry forward, and points a bit towards whatever action might be next on the horizon.

Adventure Paths, as published by Paizo, could be a good example of this. Each adventure is an "episode" but you're not really expected to play them by themselves; they fit into a greater framework made up of the entire adventure path together.

In theory, you could also do a hybrid (serial) which leans more towards the serial model, and which might be more appropriate for a game in which attendance is not a problem, and you play often and frequently. However, I think you still do things basically the same way in that regard, with just a decreased emphasis on "closing out" episodes. In fact, I think most of the campaigns I've run probably approximate this model more closely than hybrid (episodic), but my group has frequently been dogged by attendence issues; we don't always have exactly the same crowd every time we play, and it's not infrequent that the time between sessions stretches beyond two weeks. I think this has made running that kind of game more difficult for me, so I'm leaning more towards "closing out" each session (or two together, tops) as a kind of episode that still feeds into a greater, bigger plot, but which has a greater degree of closure at more intervals than what I'm running now.

So any tips from the peanut gallery on how to maintain that great X-files esque conspiracy model with a greater story emerging from play over time, yet with more discrete episodes, I'm interested in hearing it.


  1. This is a thoughtful post that made me realize something -- I'm a little wary of over-transferring other media models to RPG's. Maybe I've been reading too many Old-School blogs, but I see a lot of value in the pace, action, structure, and story emerging out of the player/character interactions with the world. A sandbox play would have difficulty being episodic, unless you make the party return to town from the dungeon at the end of each session. Likewise, it would be difficult for it to be a serial, as the randomness that often accompanies sandbox games makes long term metaplotting more difficult.

    What if we thought about episodes being encounters rather than game sessions?

    (Hello, btw! Sorry I haven't been around the intarwebs much lately).

  2. The problem with "episodes as encounters" is that you then have to something with sessions.

    The session is such an obvious unit that if you don't use it as part of your storytelling structure, you're just not going to be able to do very much.

    At the beginning of each session, players need to recall what happened last time, and as time moves forward and it gets later in the night, everyone gets more intense and focused because they know things are wrapping up soon.

    Of course it's an arbitrary constraint placed on the story; but then so is the 42-minute time limit placed on TV drama. I think the fact that it's an arbitrary time limit is perhaps part of its structural power. The metre of a sonnet is arbitrary, but is the key to its expressive potency.

    I think other realities of gaming, as Josh outlines, make it basically impossible for a game to be fully serial, but then, the concept of experience makes true episodic play impossible, too -- the A-Team can't get BETTER from one episode to the next -- they have to start each episode in exactly the same place they start every episode, so the audience can pick up the story without any background required. A primary conceit in gaming is that characters develop and increase in effectiveness over time -- that's a serial concept right there.

    Interesting thoughts!

  3. Like Corey, I've come to see the session as the building block of any gaming campaign (with the exception of odd formats, like play-by-post games, or somesuch). I'll agree with you on the value of the "action, structure, and story emerging out of the player/character interactions with the world" with the caveat that you take out pacing. Pacing is almost always the purvue of the GM. Players driving pacing could happen theoretically... but I've never really seen it.

    So utilizing the session as the building block of the game means utilizing the pacing to make each session sorta resemble an episode.

    Corey's also right; no game (or at least very few) would ever find themselves literally on the endpoints of the spectrum. Those are more theoretical places to have your game, not practical ones. However, certainly you can decide which endpoint you more closely want to approach, and utilize the strengths of each, mitigate the weaknesses, and, if you give it some thought, hopefully improve your game and your GMing technique while you're at it.

  4. As usual, you guys are wrong! ;)

    Actually, I agree with most of what you're saying. The session does make sense as a unit to use. There are some games that make that explicit, with sessions called episodes -- my friend Tim is a big Buffy gamer. I agree that the session is an arbitrary marker, but have no problem with arbitrariness. I think my resistance comes from the fact that the session is more than arbitrary, it's also very fuzzy. It's not always 42 minutes. Sometimes it's 4 hours, sometimes it's 2, and sometimes you don't know which it's going to be until you're halfway done. Focus may build toward a climatic encounter at the end of the session, or it may wane even as that encounter takes place, as it becomes late on a Friday night when you've worked all week. (This just happened to me last week. Climactic encounter with a vampire and I just wanted to get it over with so I could go to bed). Or the climatic encounter could happen at the beginning, with the remainder of the session spent in denouement, which is a fancy word for dividing up the loot.

    I am not saying thinking about one's game as episodic or serial isn't useful. I guess I am just focusing on the differences between gaming and (other?) forms of storytelling. Lindhoff and Cruise don't have to worry about their audience being tired on Tuesday nights and that affecting what happens to Sawyer.

  5. I hear my ears tingling. I'm the Tim that Nakia refers to. I've run Buffy the Vampire Slayer: RPG for many years now and we are just entering our season 6. For someone like me, it is just plain obvious that the episodic structure works (though I guess it's more of a Hybrid since we do have a seasonal story arc to maintain).

    A session isn't all that arbitrary to us. We're pretty good about hitting the 4 hour mark. We share GM duties, though, so timing can be off... but only a little. We've never run more than an hour or so late.

    We *always* build towards an episode climax (as well as the occasional cliffhanger). BTVS:RPG is a much more narrative game, so timing is a lot easier than if I were running D&D. Combats just don't take that long and its easy to sort of wave your hand and push things forward or drag them out.

    In Buffy, timing is everything. We start with a "Previously on..." segment to remind everyone what happened last session and earlier in the season. We only play every few weeks and memories can be fuzzy. Then, we do a nice Teaser sequence. When possible, the rest of the episode is broken up into a standard three act structure.

    I'm *not* saying that this structure should be used for all games. But when your game is explicitly trying to model a TV show you sure as heck better use a TV sow structure.

    PS See for our campaign website.

  6. To be honest, I think that an ability to manage exactly this sort of pacing thing is a key GM skill. Like most key GM skills, it's very difficult to quantify or teach, and so I don't know that it even gets noticed all that much, but all great GMs pace their sessions so that the exciting stuff happens appropriately, ending either with satisfying resolution or horrifying cliffhanger (I once famously ended a session by telling my players, "And then 500 dinosaurs come charging over the hill at you. We'll pick this up next week.").

    I don't get it right EVERY time, by any means, but it's awfully satisfying when it all comes together and your players sit back at the end of a session saying, "That was AWESOME!"

    Or, "I hate you so much." Either way.

    And I do think that such a practice is valuable for ANY sort of game. How you tie one episode to another is perhaps something that depends to a degree on your ruleset or other game suppositions, but the need for a single episode to function as a dramatically coherent unit is necessary in any game.

  7. Oh, somehow I didn't realize that there were newer comments on this post.

    Once again, I find myself in agreement with barsoomcore. We do, after all, share most of the same brain.

    It's true that sometimes the session is a bit fuzzy. You don't know for sure when you start it, how long you're going to play, sometimes. You don't know how long it's going to take for the PCs to add up all the clues and move on, or if they're going to get sidetracked by something that suddenly becomes super important to them, unexpectedly. None of that is untrue.

    However, that doesn't mean that a good GM throws up his hands and gives up about monitoring, and to a great extent, controlling the pacing of the game. It just means that it's more difficult to do.

    I've certainly noticed that when it works, it greatly improves the quality of each individual session. I've also noticed that attempting to do it, even when I don't quite hit it, still improves the quality of the session even so.