Gaming: A Philosophy of Encounter Design

It's been a while since I did a gaming post, so I'm going to do one of those, since I have a Star Wars game I'm running starting next week.

I recently changed how I write up encounters for gaming purposes.  The basic philosophy (set the scene, establish the stakes, establish the mechanics) came from 7th Sea, 2nd Edition, but since I didn't have it handy when I actually started writing them up, I'm going to claim at least some of this as my own.

The general flow is this: Purpose, Stakes, Obstacles, Complications, Set Dressing.  The actual flow is probably more PurpoStakeStaCatDressing, because each step will probably give clues to what the next step will contain and all of them will help develop the set dressing.

First thing to establish is what purpose the encounter serves in the larger scheme of things.  Some encounters exist solely to break up what might otherwise be a monotonous stretch of travel, for instance - the brief snippets we see in the background of an Indiana Jones "travel by map" montage.  The One Ring published campaign "The Darkening of Mirkwood," which I'm a player in right now, has a few of these.  Some encounters exist to introduce the characters to setting elements - the classic entry of the characters into the village of Barovia in Castle Ravenloft, for instance.  Some encounters exist specifically as a challenge which the players must overcome to advance the plot - say, the entire Mines of Moria sequence in Lord of the Rings.  Finally, at least for this list (there are doubtless further variations), there are climactic encounters, which are meant to culminate an adventure or part of an adventure - consider the Battle of Trafalgar, after Nelson had chased the French fleet from France to America and back and finally pinned them down, and no matter what happened after, it was going to be a winding-down of the campaign.

Once the purpose of the encounter has been determined, what is at stake narratively? Do the characters have another way of achieving the objective that does not rely on a narrative bottleneck? Could they bypass this encounter, and what happens if they do? The stakes are generally higher if it is an encounter that cannot be bypassed, such as climactic encounters or the Moria Underground Railroad.  These scenes, the big set-piece scenes, deserve lavish attention.  If, on the other hand, it's "search the professor's office for clues," it's entirely possible that they could achieve the same results by interviewing his students, rooting through his mail, or asking the university library about his reading habits, at which point the appropriate level of effort required for the scene design drops like a rock because there are a number of different ways that the characters could solve the problem and over-emphasis on one of them means that the others will likely be ignored.

Next, what are the obstacles? What challenge do the players have to overcome to get from A to B, and what happens if they fail to overcome them? Do they simply roll to search the professor's office for clues, or did he install a poisonous needle trap in his desk after his travels to the South Seas where he got the ancient idol? Is there a cave troll at Balin's tomb that they need to get past? Is there an ancient vampire that has to be overcome before the village can sleep soundly once more... who'll probably resurrect as soon as their backs are turned? What are the mechanics for resolving the obstacle - is it a puzzle, a trap, a social interaction, a combat encounter, or a hybrid of those basic building blocks? If there are no mechanics, is it really an obstacle, or is it actually set dressing?

Obstacles should be scaled to the narrative stakes of the scene, and the amount of effort put into designing them should be scaled appropriately.  For instance, it would be inappropriate for the search for clues in the professor's office to turn into a full-scale, party-destroying shoot-out; it certainly could happen that way, and maybe be improvised into that if the police investigate the break-in and the party is stupid, but that's not design intent.  Scenes that are meant to act basically as a speedbump should be scaled as if they are meant to act like a speedbump, not set up as if a scene isn't complete without a corpse.  Meanwhile, the passage through Moria was designed as a Wham Episode from the beginning by Tolkien - Gandalf had to be offered up as a human sacrifice to achieve the narrative impact.  Obstacles should generally be the things that characters can reasonably expect to think about as they enter the scene, and should be part of the scene setup, which is why I'm addressing them before even getting to set dressing.  This is what separates them from complications.

One of the examples above - the cave troll - is actually, to my mind, a complication rather than an obvious obstacle.  Complications are things that either give the scene a twist that changes the scene after it's been set up, or things that can be added or subtracted depending on how the characters are faring in the scene.  Classic modules often have a problem with this, where the encounter-as-written is either overloaded, or irrelevant with no actual narrative stakes.  Having played a few where parties were severely mauled because "that's how it was written," I'm not a big fan of following the scene-as-written if it wipes the party out just because that's how it was written; I once played through Tomb of Horrors (albeit as the Deadlands Mine O' Horrors) with a group where all of us rage-quit, GM included, after one more completely unavoidable "Gygaxian fuckery" trap caused because the party took a wrong turn.  It is one thing to kill the characters because of their actions or lack thereof; it is another entirely to design an encounter in such a way that it penalizes the party for existing or making an apparently no-stakes decision.  I would rather treat them as complications - additional encounters that we either run, or do not, as determined by group and narrative.  "Complications" are, in this parlance, options to be added to the encounter, rather than mandatory-fun moments, and I expect most GMs treat some entire encounters as "complications" on the adventure; this is how the first time I played Tomb of Horrors treated the silly, tedious, time-consuming door trap that serves no purpose but to frustrate players without challenging them (no, Gary, "I push on the left side of the door" is not a challenge to the player, it's just you wasting my time).

So let's talk about the cave troll as an example of a complication.  The players have holed up in Balin's Tomb, and they are slaughtering their way through wave after wave of orcs.  They have at this point decided that the orc fight is winnable, or at least that they can do this all day, but everyone at the table is getting bored with the orc horde and they don't have any ideas what to do.  Now the cave troll bursts in, and it becomes clear that Balin's Tomb is not actually a safe place to stay, and the party has to scramble out, running and gunning the entire way, to try to escape.  It's actually a superb application of Chandler's Law ("when in doubt, have a man with a gun burst through the door").

Similarly, the party is in the professor's office.  They have bypassed the poisonous needle, and found his diary in the drawer.  At this point, the door rattles.  This is a complication - though, given the scene described, it's such a stock scene they should expect it.  It was not obvious when they entered the room, and if the GM chose, he could completely skip whatever follows, or say it was the night watchman, or whatever, rather than having the Nazis/cultists/rival thieves come in.

These are both the kind of complications that change the scene, but not particularly the dynamics of the scene.  The other kind of complication is the plot twist; the example that comes to mind is Littlefinger's betrayal of Ned Stark in A Song of Ice and Fire.  Han and Leia turning the corner to discover that Vader has beaten them to Cloud City would also count.  At that point, the entire meaning of the scene changes.  In a more gaming-friendly setting (because books and movies are railroads), the moment in Dragonlance where one of the "elves" reveals to the Heroes of the Lance that she is actually a silver dragon, and that there is a Plot Problem with the good dragons, is a good example.  The entire dynamics of the adventure change at that moment.  They can continue on their original course, or follow the Plot Problem.

In my experience, complications are what make scenes, or encounters, memorable, like seasoning in a meal.  There was a guy I gamed with when I was a teen who, whatever his other faults, was a master at pulling out twists like this - I can still remember discovering that my half-orc had a half-brother/enemy (pretty standard plot twist, but remember, we were teens, and his Reveal was spot-on).  The danger, just like seasoning, is their over-use.  When every scene has a Shocking Reveal and a Plot Twist, or every scene has a second wave of bad guys coming in halfway through the fight, the technique loses its impact.  If you think of storytelling as combat, where your goal is to register an emotional impact on the target, then reliance on one technique for telling the story means that emotional impact is easily parried or deflected, or the spot you're hitting is so deadened that it no longer registers as an impact.  George R. R. Martin is especially bad about this; by the fourth or fifth book in the series, events which receive lavish attention on his part are, to the reader, ho-hum and routine.  In contrast, the way that All Quiet on the Western Front ends is the exact opposite of this - the main character, who we've followed through his service, dies in a single sentence without any elaboration, on a day that is so routine that the daily report is "all quiet on the Western Front."  It is absolutely devastating, because Remarque has spent so long building up his struggle to survive.

So, after all of that, we reach set dressing.  Set dressing looks like it just means "everything else," but it's more like the tip of the iceberg - the part the players actually see.  This means non-player characters, with depth based on their importance to the story, this means scene descriptions, this means musical cues if you want to go that route.  You could easily call this the J. J. Abrams part; he's great at gorgeous visuals and cool scenes, and that's what this part is about.

Each scene should be designed to capture a certain response from the players, even if that response is just "well that was weird, what's next?" I find that a short paragraph of text to describe the scene is generally sufficient text setup; some older adventures might have page on page of descriptive text that serves little purpose but to test the writer's thesaurus, the GM's lung capacity, and the players' patience.

For as much as I have picked on Gary Gygax, this is actually one of the best things about Tomb of Horrors - not only did it include brief descriptions, but any room of any significance included a picture of the room meant for the players to see.  Handouts are useful, evocative, and help set the mood.  They can also be overdone, and can drift into turning game management into a nightmare of file folders and tea-staining old maps, so judgment is important.

The same is true of music - it should give an emotional cue to the scene, not be the scene.  I once made the mistake of having the Louis Armstrong version of "Cabaret" on in a scene set in a cabaret in Berlin; at least one player spent more time listening to "Cabaret" and trying to figure out what the song was than listening to the scene.  Generally, instrumentals are better than lyrical pieces because we tend to tune in on voices, and unless you are specifically running Conan, Indiana Jones, or Star Wars, sound-alikes tend to be better than the original score.  If you are, then by all means use the original score; it will help lend reality to the scene.

Finally, realize that some scenes or encounters aren't simple encounters; they are what might be called compound or complex encounters.  Let's go back to Trafalgar.  The party is at Trafalgar, and the initial volleys are done.  They are yardarm to yardarm with the French three-decker L'Indomitable and have decided they want that sweet, sweet prize money.  The likely sequence is boarding, clearing the decks, and fighting their way onto the quarterdeck to compel the captain's surrender.  Each of these is a simple encounter and can be written up that way, but it's not like you would ever run "okay, now we are going to board a ship!" as a game.  Further, even though boarding is, realistically, the moment of greatest risk, narratively it's not great if you face-plant into a pike as you're coming aboard.  Therefore, the boarding encounter needs to be brisk but not lethal; the fight across the deck should have the ability to kill or wound characters because it's potentially literally wading ankle-deep in blood; breaching the quarterdeck should either be a moment of utter collapse, without any real stakes other than having achieved the objective and taken the prisoners, or an epic clash for the fate of the ship, depending on narrative needs.  In either case, it is a compound encounter because there are three separate encounters nested within it.  Dungeon crawls are, for the most part, simply compound encounters.

That is, in a very large nutshell, how I design encounters.  I don't know that anyone will get any use out of it but me, but writing it helps crystallize some of the underlying assumptions.