David M. asks:
How do you marry scenes (framing, agenda, bangs, etc.) with scenarios/game structures? Do you prepare scene ideas within your scenario “tool bag”, or do they typically pop up in your games organically through your scenario prep?
I ask because scenarios, which I view as more freeform, amorphous, and/or a “bag of tools”, don’t seem to lead directly into scene-framing; however, prepping scenes seem anathema to the flexibility of nodes, scenarios, etc.
The answer to this question is going to make a lot more sense if you’re familiar with The Art of Pacing, so if you haven’t read that series, you may want to pop over and take a peek. In brief, though, a scene has:
- An agenda, which is what the scene is about. It can be thought of as the question the scene is asking. (For example: Can the PCs escape the kobolds? Can Baron von Stauffen trick the PCs into revealing the identity of their patron? Can they steal the Ruby of Omarrat?)
- A bang, which is the inciting incident that kicks off the scene.
- A location and characters.
- An ending, which may or may not resolve the original agenda of the scene.
The quick answer to your question is that I generally don’t prep full scenes. There are exceptions, but in situation-based scenarios you’re far more likely to be prepping parts of scenes — agendas, bangs, locations, characters, etc. — and then combining and framing those parts into fully realized scenes in reaction to what the players are doing.
For example, my campaign status document is often stocked with a timeline of bangs, many of which are generated by things that have previously happened during play. For example, maybe the PCs have angered the Domingo cartel, so the cartel sends an assassin to kill them.
You could, in fact, prep this as a full scene: The assassin will attack the PCs when they go to a specific place and in a specific way. (For example, when the PCs go to the 1029 Bar, their favorite hangout, the assassin will try to poison their drinks.) And there are many cases when that’s exactly what you should do.
In practice, though, I’m far more likely to just put the assassin in my campaign status document. The actual scene I frame will be the result of combining that scene fragment with the events of actual play. For example, maybe the PCs decide to hole up in a motel outside of town. What might happen next? Well, I can look at my timeline of bangs, pull the assassin, combine it with the given circumstances, and frame up a scene where the assassin attacks them at the motel.
Similarly, if I’ve running a mystery scenario, my adventure notes will likely be filled with locations and characters for the PCs to investigate. But those aren’t necessarily scenes. They’re just parts of scenes, and the actual scenes that get framed up will depend on where the PCs go, what they do, and how events play out.
To take a simple example, the PCs might identify a suspect. When they decide to investigate the suspect, what scene(s) will you frame up? Well, that depends. Are they going to interrogate them? Put them under surveillance? Hack their phone? Try to seduce them under false pretenses? Each of those would be completely different scenes, often playing out at different locations, with different bangs, and with very different agendas.
As this suggests, scenario structures are going to give you guidance on what scene elements to prep and how to use them. In many cases, the structure tell you how to frame your scenes: What scenes to frame, what questions are important to answer, and what to fill those scenes with. It’ll also often tell you a lot about empty time — the unimportant stuff you can and should be framing past to the next scene.
For example, consider a dungeon scenario: You prep individual rooms filled with threats, secrets, and treasures. The structure naturally leads you to frame each room as a scene, with common agendas focused on the content keyed to the room like:
- Can the PCs defeat the monsters?
- Can the PCs find the treasure?
- Can the PCs solve the puzzle?
Similarly, in a node-based scenario one of your nodes might be an NPC with key information (that will lead the PCs to other nodes and/or reveal deeper truths about the conspiracy). The structure here is going to naturally lead you to frame scenes that have some specific variation of, “Can the PCs find the leads they need to continue their investigation?” as the agenda.
Of course, you’re not limited to these basic, structurally suggested scenes. Your scenarios will be enriched if you can find — or follow your players’ lead in finding — agendas that are more unique, personal, and tailored to the context of the campaign. But the fundamental guidance of the structure will nevertheless be helpful.
Wow, something just clicked for me. I never understood how to use nodes, but I think I was treating them like scenes. Important distinction.
To be honest, the first time I red about the node based scenario, I was treating each node as a scene. Like “ah yes, so the first node is the scene with the body, and it leads to either the scene with the father of the victim in his home, or the scene at the victim’s work place with coworkers”. And it didn’t work very well because it felt very artificial. It took me a while and several rereadings to understand that a node is not a scene but a component of a scene. And with this post it’s crystal clear.
However it is hard to warp my head around pulling several scene components in real time. It seems way easier to prep the assassin poisoning the drinks at the bar (all those components are supposed to go together), than to prep the assassin and the bar separately, come up with the poisoning idea and integrate it seamlessly at the moment.
@Quintino: The last part to think about is that you DON’T have to do this all by yourself. You don’t, for example, have to think to yourself, “Okay, I have an assassin. What can I combine the assassin with to make a scene?”
Instead, the players will say something like, “Okay, let’s spend the evening back at the bar.”
You aren’t sure what might happen at the bar, so you look to your list of bangs. Maybe you’re just resolving those in order and the assassin is next. Or maybe it’s more like a menu with a half dozen options or so and you can just pick whichever one inspires you in the moment.
But you can see that it’s the PLAYERS who provided the bar. Then you grab the assassin and have a much simpler question: “How could this assassin try to kill them in the bar?”
And maybe that’s poisoning them. Maybe it’s shooting them through a window. Maybe it’s donning a disguise and sneaking up behind them.
There’s no right option and so many possibilities. And if you can’t think of a way for the assassin to attack them in the bar (or the library or their house or the abandoned warehouse or wherever it is they go)… well, that’s OK. You don’t have to use that bang right now.
Off-topic, but where do “Ask The Alexandrian” questions actually come from? Twitter/Bluesky, email, a forum?
Quintino@2: I tend to think you’re actually right, it seems easier to prep the whole scene than to prep parts and components and assemble them in the moment, because it *is* easier. At least, it’s easier to *prep*. But the problem is that if you prep whole scenes, now you’re committed to forcing the players into them, and *that* is either crazy difficult or clunky and not enjoyable. Ie, you end up having to do railroading, and the downsides of that are covered all over this blog.
You level up your GM skill by learning to choose the recipe and mix the parts on the fly in response to your players. And the upside is that the skill is not really *that* hard to learn — it’s not nothing, but it’s not *hard* — and it gets you really noticeable return on investment.
@WhyLater: This one was e-mail. But they’ve also come from the Alexandrian Discord, social media, Youtube comments, etc.