Lessons from the trad trenches, or How to structure a scenario

21 Jun, 2026

In the past few weeks I've been thinking a lot about game structure. I feel quite safe in running a dungeon crawl¹ since it's the easiest one, and also with running mysteries, but there is the "scenario" structure of games that isn't as intuitive. This is the structure you find in pretty much any non-D&D game, from your average RuneQuest session that does not involve delving a perilous place, all the way to more modern games like Apocalypse World and its family.

For that purpose, I have tested it out on my And in the Darkness, Hearts Aglow campaign to good success. Thus now I feel quite comfortable running either of these three structures, but I still have a bit of difficulty regarding the prep part of the scenario structure, and so I decided to read the DM sections of quite a few different RPG books and check what their recommendations are, and what language they use.

I won't list all of the ones I looked at, but it was around 10 of them, including pre-Forge ones like RuneQuest and Vampire, all the way up to more modern ones like Fate, Flabbergasted, and The Troubleshooters, and all of them had something to say. Interestingly, however, is that they all coalesce around a pretty specific framework. A few games refer to this as the "TV show" framework, which is unhelpful to me as someone who doesn't watch TV, but you are already familiar with it. It goes like this:

• Create a conflict, either someone wants something, or wants to prevent something, and it's down to the PCs to do something about it. Use the setting as already established to figure out what the conflict is.

• During the session, introduce the conflict to the players by using a pre-established hook of some kind - another character such as a patron, or a general disposition of a character.

• Develop the conflict using the rules of the game to play out what exactly goes on and adjudicate the actions taken by the players to solve the conflict.

• Make a big deal out of the conclusion of the conflict through a climax.

• Have an epilogue or denouement as the characters reflect on what happened with the conflict, plan for the future, and tease next session's conflict.

This is how pretty much all of them go, and you'll recognise this is the Three-Act Structure from screenwriting but adapted to RPGs. It is also very unhelpful if you're not looking for how to structure the game, but rather for inspiration on what exactly to put on the game - ideas of conflicts. At least two games, Good Society and The Troubleshoters, flat out tell you to open up your favourite book and steal its plot - more on this in a bit.

I take issue with a few things out of this structure, and in my own game I make these alterations:

1. The framing of it as "conflict" isn't necessarily limiting, but the way it is often conceived of is. I like the way FATE frames it, through story questions rather than conflict. This is implied in the structure, but I think calling it "a question" widens the net of what conflict can be. I will elaborate on this shortly, but this is similar to the way Retired Adventurer frames it.

2. I don't like the idea that the GM is the only one who can come up with the conflict or story questions. A lot of the time the preparation of the game is framed as the GM making up the questions and the players answering them through the rules of the game (still on the TV show comparison, the GM is the director and the players the actors). Sometimes the games pay lip service towards this not being the case, but very few genuinely encourage the players to be active in recognising and creating questions, often using artifice like giving them a point if their flaw gets in their way, which I consider a crude way of getting the player to not obsess with 'winning'. An interesting enough story, a character that is well defined, and a player that isn't a wiener makes these things easy to adjudicate.

3. I also don't like the framing a few games put forward that the question is made up before the game, and played out during the game. In my experience, questions arise and are responded to naturally through the game, and part of the skill of being a gamemaster is recognising what exactly is a compelling question and when one has been answered already.

4. Lastly, climax. Typically, the GM is recommended to build towards a specific climax, often being given tools of how to make it emotional and a big showdown - an arena, a boss monster, even encouragement to put on epic music. I think these are backwards: the GM shouldn't build towards the climax because this is not a screenplay, rather the resolution of the conflict, the answering of the question, is necessarily going to be the climactic moment of the game, so long as the question was correctly assessed. This is where GMs might have the issue of "I prepared this whole big fight and the players didn't even care" - this is a GM that did not recognise that the players weren't interested in the question of "Can we defeat the evil bad guy". These sorts of expectations need to be hashed out so that the GM avoids playing the dancing monkey for her players, but also so that the players come in primed for caring about something specific.

This is all very heady and theoretical, so I think it's helpful to look at a one-and-a-half hour scenario I introduced for my game without all of this vocabulary, and analyse how a lot of this is already intuitive. This is what my prep for that scenario amounted to:

Right before Midterms the faculty will convene for a dinner at the monastery. Not only will the professors of the knights be there, but also members of the ecclesiastical faculty too, and the dean. This year the third years are meant to help with serving the food and such, as squires. They will get: Cora, Fiorello, Giovanna [the players], but also Bellina, Claudia, and Rina (who is meant to be “looking after” them). The rest will help with the preparation of the hall.

They will be under the “guidance” of the cook, Brother Honorius, an absent minded Carthunian, but in practice the CO is meant to be Rina. The knights are to STAY CLEAR of the kitchen itself and only take stuff from it to the hall, and vice versa.

Some ideas:

• Belina and Cora are asked to go down to the cellar to keep an eye on things or whatever. Belina starts making out and sorta going far, and it starts to become apparent this might become their first time. Can they make it through without making it awkward?

• Claudia is getting very competitive with Giovanna about general waiting, and it’s getting risky.

• Benedoro Petrani takes Fiorello by the corner and says “listen we have a situation.” He has a pet Giant Almyrian Rat which he brought in a little cage to show it to one of the monks, who’s very interested in such animals. The damn thing is sharp as a spear, though, and opened the cage. It is now on Fiorello to find the damn rat and have no one be any the wiser, otherwise it’s BOTH of their asses.

We have everything here:

• Question 1: there's a rat on the loose, and no one can know about it other than one of the players. Can Fiorello catch the rodent without jeopardizing himself and his comrades?

  • This is a safe question, as players do not typically reject stakes like these. This can be thought of as a mission rather than a story question, it is pushed onto the player.

• Question 2: This one is already framed as a story question. Cora and Belina's relationship is getting to a head. Can they navigate this without awkwardness?

  • This could have gone wrong if Cora's player was uninterested in their character's relationship to Belina. Fortunately they were, but notice how miscommunication is the bigger issue.

• Question 3: Will the competitiveness of Claudia jeopardize the mission? How will Giovanna, who has a crush on her, react?

  • Likewise, this could have gone wrong if Giovanna's player did not care about Claudia. But given that Giovanna's player is skilled, she has her own questions about her character, which I was made aware of through between-session discussions, and could provide this situation. This is less of a question in itself and more setting the stage so she can find her questions and answers.

As we played though, we discovered a few additional questions, iterating on what happened before, as I introduced conflict. The central questions were both introduced either by a patron (Fiorello's and Cora's, by "patron" I just mean "character which introduces question") or character element (Giovanna's), and then we played through it, and the moment Fiorello caught the rat felt climactic because that is naturally what the session was building towards.

Notice that the questions themselves are mostly immaterial. They're not quite conflict as classically defined in RPGs, there is no one who wants a thing - Cora's conflict is that both of them want the same thing, but we're not sure if their emotions will allow them to get it, thus my preference for FATE framing it as a dramatic question rather than conflict.

Notice too that I don't chart out where the rat will go, because I already was ideating on interesting locations the rat could go to and how it could react to other plot elements, but that "development" is the trickiest part of the entire process. During a crawl you have the procedures guiding you, and a mystery has clues which act as intermediary steps towards the potential climax (which can either be solving the mystery or not)², but a scenario has no such guidelines as a scene can be whatever you want.

A few games try to formalize this by introducing "types of scenes" you can put in your game - have a research scene followed by an investigation scene, interrupted by a fight scene that leads into a chase scene - but I think this misses the point of the scenario structure. That is: all of this stuff is up to the players and to GM facilitation. If the conflict is that there is a secret base under the Sphinx and the players are going in there, it stands to reason a fight might break out eventually, but perhaps they'll decide to sneak past, or get in touch with their contacts and go in disguise. The players should be free to answer the question with whatever tools they have available

This also gets at the heart of what Apocalypse World and the OSR both have in common when they say some variation of "play the world" and "prepare situations, not outcomes". The further refining of the "development" (or Rising Action) stage and climax stage into specific situations and scenes is a symptom of insecurity from a GM who doesn't trust her own knowledge of the world she's playing in to create meaningful development, and doesn't trust the methods of adjudication she has in her pocket (i.e. the rules) to see her through safely in case her personal adjudication is not enough.

Because of this, I think of the "scenario" type of gaming as the toughest to pull off for new GMs, as it requires trust in the players and the rules, and knowledge of the world. Thankfully these are all things built over time, and the chimeric nature of RPGs means a campaign can easily mix in multiple of these at once, as has been done since the inception of the hobby. In my own game I use a "crawl" structure as the backbone of the campaign we can return to, and scenarios as impromptu spice in the middle of that structure, even though we did switch to a pure-scenario structure for a while and it worked perfectly well.

I also believe that even to get to the scenario already requires that game-world knowledge, because you need it to create the conflict in the first place. In fact, this is why I went looking for those books at all: I had all of this internalized to a degree or another, but I was lacking ideas of what story questions to actually ask my players, what conflicts to put in the game itself, and this structure flat out does not address that.

This is because scenarios are much more individual to style of campaign and game-world - the tools and inspiration required to run a game of Good Society are way more different than the ones required to run a game of Masks - as well as the assumption by a lot of these games to rely on genre knowledge for adjudication, which I do not like at all for reasons Weird Writer expands on here.

The most helpful resource I found for this, besides literature, as stealing plot points from books is actually a perfectly good and acceptable way to run a game, or simply letting yourself be inspired by it and the world at large, was this blog post by S. John Ross of Risus fame. Not necessarily because I have used it a lot (in fact, I haven't pretty much at all), but because perusing it almost always changes the gears in my brain to what a "story" even is, and lets me see more clearly that a lot of it is quite basic and simple to adapt.

Therefore, if I were to synthetise the whole structure and wisdom I have found into a single one (and again, I often run this way by instinct rather than stopping the game to voice the structure, as the structure is a helper to organise thoughts, not the game), borrowing heavily from FATE, it would be something like this:

• The GM and players create characters suitable for the game they intend to play. This means having characters that can easily be hooked into conflicts, or bring a conflict or another to the game, and have questions the players want answered. You don't need many for a game.

• The GM introduces a story question created either by herself or by her players, using the pre-established setting, literature, and characters for inspiration on what the particulars will be.

  • This can be done through a patron or character element, even if it is as simple as "loves adventure", which is implied by all characters in adventure games.

• The players will act upon the question or conflict, and new ones will be found as the game goes. The GM will facilitate this through the conversation, and where there is risk or doubt, will utilize the method previously agreed upon by the table (reified as a book or not) to adjudicate. Individual scenes will naturally flow from this.

• At some point the question will be answered in some way. It can be answered definitively, closing down that avenue, or openly, so as to lead into further complications. A question need not be answered on the session it is introduced either; but if it is left hanging, it must be addressed in the future, perhaps mutating into something else. The nature of stories is to change.

• At the tail end of a question having been answered, if the session has already run on long enough and it is too late to introduce a new one, it is helpful to let the characters reflect on what happened. Just going around and seeing what changed in the world or the characters through the answering of that question, how they have grown or not, etc.

It is helpful to have an ongoing theme to the questions, so as to build a campaign, or a question that is open enough that it takes a long time to answer, but this isn't necessary to answer from the top either. Much like how "the story" of a videogame is "I walked to the left and fought some guys and then jumped through a window and got 4 stars on the level", "the story" of an RPG is the events that happen in it - what the OSR calls "emergent storytelling", though with less random table fetishism (I say cheekily).

This is also the part where FATE and others introduce metacurrency to reinforce the structure, sometimes even going as far as trying to enforce behaviour, but I dislike that and disagree with it on a conceptual level. So long as the table has expectations lined up, trust one another and the methods they agreed for resolution, and understand the world they're situated in (or genre, if so inclined), I feel no need to use this kind of artifice.

#advice and aids #free thoughts