The Alexandrian

Go to Part 1


Framing a scene in medias res is to start things in the middle of the action. It’s an effective technique because it jump-starts the scene at its most exciting or interesting part. In other mediums it also intrigues the audience by creating a mini-mystery: How did the characters find themselves in this precarious situation?

This element of mystery in the in medias res becomes problematic in an RPG, however: While it’s all right for an audience to remain ignorant of how the characters found themselves in their current circumstances, that can create significant issues when the players are supposed to be the characters and need to know what they know so that they can make coherent decisions.

Let’s lay the mini-mystery aside for a moment, therefore, and focus on a more basic form of in medias res that I highly recommend regardless of your predilections: Framing past the entrance.

Warehouse in ChicagoGMs tend to frame to the arrival: If the PCs are heading to a warehouse, the GM will cut to their car pulling up outside the warehouse. If they’re going to question a suspect, the GM will cut to them knocking on the suspect’s door. It’s a safe choice, but it’s frequently unnecessary. For example, if you know that the PCs are going to Boss Man’s office to search for clues, you don’t have to go through the laborious process of having them enter the office building, pick the lock on his door, and so forth. Instead, you can cut straight to them rifling through his filing cabinets and pulling out the incriminating documents.

This works because the assumptions you’re making in framing the scene are obvious: You’re assuming that the PCs are, in fact, going to do what they said they were going to do. And you’re assuming that they will be successful in doing it. (If the latter is in doubt, of course, you’d call for an action check and then frame accordingly. For example, if the PCs fail an infiltration check then you might frame to the moment where the flashlight of a security guard flashes across them.)


More dramatic examples of in medias res are common in other mediums, but relatively difficult to pull off in an RPG without railroading. But it can be very rewarding if you can make it work.

In my experience, that requires a high degree of trust and understanding between the GM and the players: The GM needs to know his players and their characters well enough that he can accurately predict their reactions. And then the players need to have enough trust in the GM to believe that he has made an accurate prediction.

However, there are also ways that the GM can “cheat” in order to achieve satisfactory in medias res scene openings. A basic method is, “Why are you here?” The GM opens the scene by saying something like, “You’re in a dark tunnel underneath the Parker Corporation’s headquarters. Why are you here?”

This technique basically says, “Yup, we’re railroading to this point. But I’m going to mitigate it by giving you some influence over exactly what model the locomotive is going to be.” Obviously that’s a lot heavier handed than a lot of people are going to be comfortable with. (It’s way out of my personal comfort zone.) But if it works for you and yours, the pay-off is that you can get away with much harder frames and much tighter pacing.

Another “cheat” is to use the in medias res in combination with a flashback: You show the PCs mired in a situation and then use a flashback to establish how they got there. The “future” knowledge of where they end up obviously constrains choice, but once again you’re mitigating the heavy hand you’re applying by allowing the players to influence (or even define) the path that leads to the events they’ve seen. (And there can be a fine art in portraying something compelling in your flashforward which is not particularly binding in its content.)


An epilogue scene is a specific type of color scene which immediately follows a major conflict or bang. During the epilogue scene, the characters are able to reflect and react to the things that just happened to them.

The purpose of the epilogue is both aesthetic and practical. (And it’s practical in both the game world and the metagame.) Aesthetically it provides a natural formula for effectively varying the pacing of your game: Moments of high tension and exciting action are contrasted against the relative calm of the epilogue which puts those moments into a larger context of progress or setback or revelation.

In the context of the metagame, an epilogue scene is practical because it often comes at a moment when the players and/or GM need to take stock: Damage needs to be healed. Resources need to be tallied. Notes need to be consulted. If your group can develop the ability to roleplay effectively through these moments of mechanical bookkeeping it can greatly enhance your game sessions. (Note that the roleplaying does not necessarily need to be about the mechanical bookkeeping that’s happening: It’s possible to say “we heal everybody up” and then handle the mechanics of that while actually roleplaying a later scene in which the PCs are discussing their options for raiding the palace.)

Meanwhile, in the context of the game world, it’s perfectly natural for people to kick back, relax, and take a moment to recover from a stressful situation. Or to sit down and try to talk their way through earth-shaking revelations. (“Dude, you’ve gotta bring me some beer. Betty just broke up with me. Also, I think she’s been possessed by a demon.”)

In practice, these scenes will often happen quite naturally if you give the PCs a bit of rope. If you’re using harder scene framing, however, you really shouldn’t neglect them.

An effective technique for this sort of thing is to establish common “set pieces” for epilogue scenes. For example, in my Ptolus campaign these scenes often happen in their rooms at the Ghostly Minstrel, during a carriage ride across the city, or while they’re looting the bodies of the dead and tending their wounded. We’ve established these as common themes in the campaign and when they show up the group has naturally developed a habit of falling into an epilogue scene.


Buffy the Vampire Slayer - In the Library

In Buffy the Vampire Slayer, epilogue scenes often happened at the school library or (in later seasons) at the Magic Box. But Buffy is also a great example of how you can use a set piece for epilogue scenes and then shift it by occasionally dropping a conflict into the location. This is a basic example of disruptive pacing.

The pacing of a book or movie often falls into a repetitive pattern. The same is true of an RPG. This regular rhythm can be very satisfying in its familiarity (particularly when it feels like the whole group is suddenly on the same page and participating in the ineffable synergy of an instinctive collaboration), but it can also easily become stupefying or stale in its predictability. To avoid that, you need to occasionally disrupt the familiar pacing by throwing in something unexpected: A scene that should be color suddenly erupts into conflict. A sequence that feels as if it’s winding down is abruptly escalated with a fresh bang.

What’s interesting about disruptive pacing is that it can easily degrade back into the expected. For example, the first time a supposedly dead monster suddenly reappeared during what was supposed to be the epilogue of the film it was wildly successful disruptive pacing. Now, however, it’s become part of the expected pacing for a horror film: You’re more shocked when the slasher villain stays dead than you are when he suddenly pops back to his feet.


I really can’t emphasize enough the degree to which we’ve barely even begun to scratch the surface of RPG pacing in this essay.

On the other hand, that may be for the best: While film, for example, has certainly benefited from a richer understanding of how pacing can be used in the filmmaker’s toolkit, the process of effectively pacing a film is still very much an art rather than a science. Similarly, it can be a little too easy to fall down the rabbit hole of one-true-wayism if one tries to develop trite maxims.

Instead, I hope that I’ve managed to convey a handful of useful conceptual tools that you can use in a myriad number of ways to better understand and control the pacing of your own campaigns. How you choose to use those tools is up to you: Experiment with them. Play with them. See what you can discover and what you can improve.

Film Banging: The Avengers
Film Banging: Alien
Film Banging: The Matrix
The Art of Pacing: Prepping Bangs
The Art of Pacing: Running Awesome Scenes
How NOT to Frame a Scene

The Art of Rulings
The Art of the Key
Gamemastery 101

Share on TumblrTweet about this on TwitterShare on StumbleUponShare on FacebookShare on RedditShare on Google+Digg this

5 Responses to “The Art of Pacing – Part 6: More Advanced Techniques”

  1. Brooser Bear says:


    You have done some brilliant writing adopting dramaturgical concepts to D&D adventure writing; better, I might add, than the similar kind of writing done in AD&D 2nd Edition and in more depth, than in Hamlet’s Hit Points.

    However, what I am wondering about, and my background in theater is zero, is how you would apply you scene development techniques in a sandbox. An open ended, two-dimensional sandbox that I design, is typically documented with a map, areas of interest described and detailed with text and random event/encounter tables, then points of interest, and maps and keys. Players can move anywhere they like and engage in pretty much any activity that they like, it is only a question of stopping the play at the right moment and fleshing out the more details as the players intentions become clearer. Conflicts typically readily emerge out of the tensions implicitly present in the setting. How would you apply your methodology to a sandbox campaign?

  2. Justin Alexander says:

    Very few of the techniques I discuss in the Art of Pacing would be inapplicable to a sandbox campaign and the ones which are not are generally called out specifically in the text (which, I suppose, shows my bias towards nonlinear and sandbox game structures).

    The elimination of empty time is entirely about identifying what the PCs intend to do, determining which obstacles might prevent them from accomplishing that (if any), and then skipping to the next meaningful choice. That’s the core of controlling pacing in an RPG and it’s as easily applied to a sandbox campaign as any other campaign.

    For example, if you were using a hexcrawl (the granddaddy of all sandboxes), a player might say: “We’re going to head west from town until we see Mount Tersius.” Following the hexcrawl structure, you’d look at your map, determine travel time, and generate random encounters. If you failed to generate any encounters, you could either transition using slow time (“on the second day out from town, a cold and bitter rain begins pelting you… etc.”) or you could just use a sharp cut (“four days later, you see the snow-cropped peak of Tersius clear the horizon”).

    What’s the agenda of this scene? Barring any other pertinent information, it’s as simple as: “What are the PCs going to do now that they can see Mount Tersius?” (The players must have something in mind, otherwise they wouldn’t have set seeing Mount Tersius as their intention.)

    The bang here is “seeing Mount Tersius”, which is not terribly exciting. But if this is really just a landmark they’re using for navigating in their journey, then the resulting scene is probably going to be about five seconds long (“okay, now we turn north”) and so it doesn’t take much of an agenda or a bang to support it.

    OTOH, if they were coming to Mount Tersius to investigate the rumors of a goblin warband gathering in the region then the agenda becomes much more interesting in context and the bang might be something more like, “You can see smoke rising from the side of the mountain.”

    (Of course, if they had chosen a different intention or if the random encounter generator had indicated an interruption, all of this would shift.)

    And so forth. As you’re filling the frame here the location is obvious; there are no supporting characters; and this is a color scene (unless we hypothesize a goblin ambush). The scene probably closes when they finish the agenda (by deciding what their next course of action is going to be), but maybe it closes on a new bang (“you hear the distant whoop of a goblin war cry”).

  3. Brooser Bear says:


    Thanks for your response. I am looking for a more detailed answer with regards to framing in a wilderness adventure. Let me explain. “Sandbox” is an open ended non linear campaign best suited for exploring the worlds. When the world itself is truly strange, and a puzzle, anywhere the players will go will bring them in contact with it. If you want to run a fantasy campaign, chances are, you have a world, and that is one of the best ways to get the players to explore the world. Mind you, that in this case, the world encompasses space delineated by geographic features. It doesn’t have to be a world world, or a wilderness as in the midst of nature. Let’s say you are running a game featuring outlaw bikers. Each one of the major clubs out there is unique – it has different organizational culture, offers different experiences and attracts different sorts of people. So, if you wanted your players to experience this, you would have sandbox of biker bars, bike runs, club parties other events and what not. As the players get drawn in, they may get involved in a story and a linear plot that is more like a traditional D&D game, but at times, the players are free to roam that world at will. Some DM’s will restrict their players artificially, but I believe that not even the player’s imagination is the limit, it is the DM’s imagination HAHAHAHAHA!!!!!!!

    When you get the players on the campaign of sandbox exploration, while outwardly the players are fighting, stalking, surviving, haunting, looking, in reality, they are immersed in your world. And in that case, COLOR becomes that, which makes it literature, it is the color that makes it a Dutch Masters painting. And hence, there is no dead time, since the DM can always focus his narrative. It is the narrative that is the artistic medium for sandbox (as I conceive of it, and for other role-playing, I suspect). This is important, because if you don’t focus on the right type of creative process, it won’t work! You know anybody who spent years writing a novel that will never be published? You seen unreadable and dreadfully boring writing filled with purple prose? You seen writing that you like that can’t quite shake the rap of the “amateur writing”? One reason might be, is that they formed their story in terms of hackneyed Star Trek episodes or retold the events of their favorite daytime soaps. If they have created something more original, the most likely reason their writing does not work, is because they have created a “mental movie” of their story and are nor telling that film with words, using the word-processor instead of the movie camera film. The story they are telling is a fantasy, pure and simple, because most likely, they don’t know enough about cinema to shoot that film or edit it into a movie, nor do they know enough about written word to know the difference between these two different types of media and take care to tell their story so that the reader will experience it as they are. The story is quite literally, lost inside their souls beyond retrieval.

    Same goes for D&D DMing. Where does D&D take place? In the maps and the map keys? In the read out-loud text boxes? In the Encounters? That is what they want you to believe, so much has been thought, developed and done, to make the encounters more manageable! No, the D&D game resides in DM telling the players what is happening, i.e., the DMs NARRATIVE. Keep that in mind, we will come back to it.

    Let’s look at a style of play called the Dungeon Crawl – Each Encounter is FRAMED by a room, with several paths leading to and from it. Now in the supposed wilderness adventure called the Hex Crawl, each Encounter is FRAMED by a hex, that is so and so miles long, has such and such features, and always has six pathways in and out of it. Each hex may have some unique features and a possibility for an adventure encounter. The only remaining question is as to the realism of framing the game in terms of adventure areas about one days ride miles across. In reality Hex crawl is a same as Dungeon crawl applied to a different type of a map. But it has nothing to do with the feel of wilderness exploration or wilderness adventure. How do you frame the natural world? And how do you handle a million acre forest? Do you draw a bunch of little trees on a hex map and each hex is 40 miles across? 10 miles across? 1 mile across? But it is still too abstract, and it is still the hex crawl. There are different types of forested terrain, each offers different cover and visibility, how do we present a realistic forest for exploration? or even a city, of say 150 thousand inhabitants without having to draw out each building?

    I will continue tomorrow. Any thoughts, Justin, anyone?

  4. grammar_guy says:

    Thanks for the very interesting article(s). Just throwing in a terribly nitpicky bit: It’s in medias res, not in media res.

  5. Brooser Bear says:


    To make the long post short, there is discrete and there is continuous space (as per Rene Descartes). Human minds perceives the world as a Discrete phenomenon – i.e. existing as a series of frames. Nature is actually a Continuous phenomena, i.e. rivers flow and erode the river bank all the time and not from one frame to the next. Once you get into subatomic physics, the space actually becomes Discrete once again. By the same token, there is a discrete and continuous narrative. If you go from room to room, or from hex to hex, or from scene to scene, you are conducting a discrete narrative.

    The key feature of the wilderness adventure, and hence the Sandbox, for me, is that my narrative as a DM, is continuous. How does a continuous narrative differ from discrete. Reading a room description or a location description from the Key is discrete narrative. A party is moving along a road, from point A to point B, and DM is narrating the party’s progress, describing as they go along, what they see, rolling on encounter and event tables, is a continuous narrative.

    A DM’s narrative is always influenced by the game mechanics, players input, text input, map location and die rolls, but the DM always has the narration and assimilates those influences into his or her story. In the Sandbox, players may decide where to go an what to do, twist8ing and turning the river bed, so to speak, but it is always the DM’s narrative and DM’s river following those river bends. And because the DM is the story teller, he has the power and responsibility for shifting the focus of the narrative, thereby minimizing the dead space. DM has the discretion for not taking the player to where their characters are routinely eating, fornicating, and defecating, instead the DM can refocus his narrative on the meaningful, sure, it may be all or mostly local color, but, he can tell the players about whom the fighter brought to the dance last night, how and how it seemed to have went, get into the NPC’s character and tell the Magic User the latest in guild politics and juiciest rumors, etc, etc, maybe not so much color, but exposition and signposting. And so, the DM’s narrative flows on like a mystic river.

    A more complex problem than party walking along a road like the pilgrims in the Canterbury Tales, is the party moving across a forest, or a plain, or a swamp. True, historically the humanity moved along the trails and roads and rivers, that’s why we have Rangers and their tracking ability! But suppose the party decides to move cross country. How do you do it? If you do as the D&D hacks, you handle it as discrete space, or the hex crawl, if you handle it more gracefully, you start looking for a more natural framing. The mainstream approach is to draw out a detailed map, with the thickets and clearings and points of interest as “rooms” on your map key. But what if the forest is a week’s hike in each direction? The solution is to handle it topologically and topographically. You have to realize that there will be many, many terrain features that will serve as “rooms” in a large ad unexplored forest. You key the areas of interest on the map, and then write detailed, loving descriptions of the various types of primary and secondary terrain that can be found in your forest. Say, giant old growth Oak forests, red clay terrain and shrub and dwarf pine wooded terrain, thickets of aspen and cottonwood trees. Think of who and what can be encountered there and tabulate it, all kinds of fun events, encounters, trails etc. And then build your continuous narrative based on those random and semi-random rolls until the players get to where they are going or get hopelessly lost. Keep track of the party’s progress on your maps, and fill in the details of what they explored after the game.

    So Justin, my question boils down to this: Let’s say that the players are moving from Campaign Adventure Event Node A to Campaign Adventure Event Node B, as they are going through that fantasy forest. Would you be running your technique of scene framing, setting and banging in a continuous narrative example given above?



Recent Posts

Recent Comments