Mechanically defining a scene as a unit of play

In the design I have been noodling around in my head for a while, I like the idea of borrowing the Dramasystem approach of having scenes as the basic unit of play, so players ( or the GM ) will set up scenes and then we play them out.

The basic elements I have for the set up are:

  • Where the scene takes place
  • When the scene takes place
  • Which characters are in the scene
  • What the establishing character hopes to get out of it.

This is mostly codifying standard stuff, I think, but unlike Dramasystem this is a game with stats, moves, playbooks and dice ( I’ve been thinking PbtA as I know that fairly well but I’m not married to it ) and I feel like there might be something fun about having an “establish scene” type dice roll at the start of a scene ( why yes, we have been playing Swords Without Master how can you tell? )

The problem is I can’t think of what would be the most interesting way to make it work - the obvious thing to me is that the roll gives a resource that you use in the scene. I have thought of these:

  • The roll giving a number of tokens you can spend on full or partial successes through the scene, replacing dice rolls altogether or the total number of dice available to be rolled through the scene.
  • The roll giving number of dice rolls you can make before you take penalties during the scene or conferring advantage/disadvantage a couple of times during the scene based on the initial result.
  • The roll a number of dice rolls that, if hit during the scene, allows you to mark XP
  • The roll affects the position of the scene in a Blades style Controlled/Risky/Desperate kind of way.
  • The roll gives a number of dice rolls before the PC loses control of the scene ( perhaps something like the way Ironsworn uses Momentum as a resource )
  • The roll defines which XP triggers are available to you ( this is along the SWM tone approach but offers a nudge to the player’s direction, in a game where players have more personal or group driven XP triggers available. )

It needs to be fun and not too complicated, but I feel like it is also an opportunity to give the players the opportunity to feel clever by making good use of it, to help create a degree of tension in scenes and to keep scenes from dragging on. It seems to me that there is something there, but I’m not sure what is the most interesting avenue to take things down and given that this is one of the places where players would most routinely interact with the games systems it also needs to be easy to pick up and play…

Are there any of these ideas that stand out to you, or is there something that you think would work better for this?


For the past few days I’ve been thinking about a framework for a game I had a dream about (monks wanting to achieve immortality through magic). I’m thinking about having a very strict flow of the game (inspired by Blades) and I want to have an unexpected twist to each scene. I’m not saying that this is good or even workable at all but my current exploration is to randomly select a complication for each scene that players have to overcome to achieve whatever it is they want the scene to result in. I’m not yet sure how would this work but I guess this could be an alternative to your list (which is mostly about rolls determining how much of certain resource players get).


For sure, I had thought about the roll dictating whether the scene is going to end in their favour or against them, but then that felt a bit like it was detracting from player agency. I guess a negative outcome isn’t necessarily “you don’t achieve your goals” and falls more into the “you achieve your goals but…” category.

1 Like

That’s not what I had in mind. Let’s say that players want to meet with the head of the merchant guild to convince him to let them use his fleet to get to the nearby island. It’s a challenge by itself but the roll indicates that there’s a complication in the form of “adversary making things harder” (the part IDK yet is what these complications should be exactly). So players go and meet with the merchant but a baroness, who believes to be wronged by the characters in the past, is present and what do you know! she happens to be his aid and makes this conversation a little bit harder.

Perhaps there is some kind of stat, say “preparedness”, that every character rolls (including GM) before the scene starts. Successes allow you to preemptively create scene tags or get tokens that can be spent to give you circumstancial bonuses during the scene.

So for example you’ve got a thief (PC) sneaking into a noble’s manor. The thief player and the GM roll scene prep, player gets 2 successes, GM gets 1. The GM uses their success to declare that the noble has become aware of the thief’s plan and has stationed additional guards. The thief spends one success to establish that it it’s raining, making it harder to detect noises and torches easier to douse. The thief also banks their second success and call it in later in the scene to say “this guard I’m trying to sneak by is especially drunk tonight, making it harder for them to detect me”


I think this is along the lines of what I’m thinking about - almost like the Engagement roll in Blades In The Dark where on a success things are going smoother than expected and on lower rolls they get worse until you’re starting in a desperate position, whatever that means for the current situation.

1 Like

I like this, allowing the players to create advantageous elements in the scene is pretty fun, kind of like creating aspects in Fate. This kind of maps onto the “advantage/disadvantage” bullet from my post above, but gives it a much nicer fictional framing.

I feel like there might be something fun about having an “establish scene” type dice roll at the start of a scene ( why yes, we have been playing Swords Without Master how can you tell? )

What comes to mind is the Engagement Roll from Blades in the Dark.

So, focusing on the core dice roll here, the prompts depending on the result are:

  • Critical: Exceptional result. You’ve already overcome the first obstacle and you’re in a controlled position for what’s next.
  • 6: Good result. You’re in a controlled position when the action starts.
  • 4/5: Mixed result. You’re in a risky position when the action starts.
  • 1-3: Bad result. You’re in a desperate position when the action starts.

So, obviously Blades is a game about scoundrels on daring heists. It features a lot of external action, which the above mechanic reinforces.

But - what if the mechanic was interpreted or keyed on individual character states? So now if you start with a ‘success,’ you are in a dominant state in the scene. If you have a mixed result, you enter the scene tense but reserved, and if you fail your want has already left you in a vulnerable position in the scene (and a design pathway I see you’ve already explored).

That probably needs a lot of language massaging, but the basic idea is this: if your players know what should happen next, they don’t need the mechanics to codify how to frame a scene. If they don’t know what happens, the mechanics need to help generate story.

Story is keyed on conflict, as is the entire idea of “You do it, but with complication.” Complications are what drives character development, either externally or internally. So, I think whatever you come up with to frame the scene has to be broad and needs to frame the scene in terms of conflict. If there is no tension there is no scene!

For our zany romance story game The Love Balloon, we keyed in on the following six story beats:

That is plugged into two other mechanics: the character’s MOTIVATION and BACKSTORY, which are short one words or clauses that help form character.

A SALESPERSON… looking for revenge.

The resolution mechanic, KARMA, helps guide the scene by explicitly saying that one character in the scene must either succeed or fail, which the nonparticipating players agree on. That helps give an indication to the characters on when to call scene.

Again, our philosophy was to create an engine for emergent storylines. If the players know what should happen next, then the rules have done their jobs and the players should free to disregard them in lieu of the story. We hope what minimal systems we have are enough to propel the game to that point.

Anyway, that was a bit rambling. Hope some of it was helpful! Happy designing.


This is really helpful - I had come to the Engagement Roll as the general model too, and this is a good angle on it.

I like the idea of the mechanical push to help drive the story - the design I’m working on is purposely open to telling a lot of different kinds of story, but the type of character that is being played should drive that - so it could be like the engagment roll approach based on “how do you plan to achieve this goal” and/or “how important is the goal of this scene to you?” I can see this being a good way to establish the size of a clock required to achieve their goal, but I’m not a huge clock fan for reasons I have never quite understood, so I’m trying to think of something that appeals to me more.

As I think about it, this feels to me like the central mechanic for the game in play, so I really want to get it right. I guess the answer is probably to try and slam something together and playtest it to see what works or doesn’t…


I have found that if A) characters’ Ambitions and the game themes are built tight enough AND
B) The stakes of a scene and the PCs included in a scene are decided mechanically,
then determining your 1st, 3rd, and 4th bullet points is easy. Personally, I don’t have a gimmick for the start positioning of a scene. I don’t think it is really needed.

In my Old West rpg Leadtown (which is currently in draft) uses a scene as a unit of play. Mechanically, the game runs with the archaic gambling game of Faro with some built in cheats. Faro is like Roulette but with an extra suit of Spades as the wheel and with most bets pushing to the next hand rather than losing. Each hand represents the introduction and resolution to a scene.

The size of the bet controls the level of stakes. The character’s Ambition and the card hand suggest the theme, and what the establishing character hopes to get (and from whom). Then, the player chooses what their character wants in this scene using everything above plus the character’s archetype and background to fill in the details of the scene.

The way a scene starts is that any players with bets on the two cards in the hand (the winner and the loser) are included in the scene. The theme of the scene is determined by the gamblers from the two suits in the hand (hearts = relationships / influence, diamonds = investments or gambling, clubs = violence, and spades = authority). The stakes of the scene are represented by the size of the bet. If the gambler has a large wager in the second half of the game, the scene will be a critical one. (Bet maximums prevent the game from peaking in the first half).

The scene is set in an appropriate location considering the theme from a random list of locations in the town. So, let’s say, the character being played is the gambler archetype. The Ambition of this archetype is to defeat the reigning card champion in town in a gambling game and take the thing most important to the old champion. Here are a couple examples of how a scene could go down in this system.

Just before the scene starts the player running The Gambler decides what they want to get from the scene in order to move towards accomplishing their Ambition or perhaps supporting another character in getting accomplishing theirs.

Now let’s say The Gambler is in a scene with a serious bet placed in the second half of the game and the suit chosen between the two cards is Diamonds, then The Gambler is allowed to resolve her Ambition in the scene and attempt to defeat the old champion in a card game at the saloon.

Now let’s say the timing of the hand is near the beginning of the game, the chosen suit is Hearts, and the bet is only $5 -10 (10 -20% of start $), The Gambler player should suggest something with much lower stakes, for examples:

*The player states she wants The Gambler to gain an ally in a future game against the current Champ, where she will hopefully get the advantage of free drinks or signalling when she finally plays the Champ.
*The player states she wants to ask someone in town to introduce The Gambler to the current Champ and thus be able to set up a later scene to accomplish her PCs Ambition.


This sounds complicated (as someone who has never gambled!) but it’s also really clever and super in-keeping with the theme of an old-west game, I bet if that was a paradigm of play I understood it would fit together nicely. Having it run as a hand works really well too in terms of the size of the scene and the wider context of the game. This has definitely given me some thoughts on how an initial roll could set up a dice pool.

1 Like

That was a hell of a wall of text I posted! I’m glad someone got through it! :stuck_out_tongue:
Sounds like you see the idea, if not the particulars of my game and examples.

The other nice thing about using card games that finish at the end of the deck is that it helps with pacing. Faro is always 25 hands long. If we are done with 13 hands and but we have already played for more than 90 min, I should pick up the pace. This helps when running one-shots or games at conventions.

1 Like

I’m still trying to understand what would be the benefit of dice rolls in establishing a scene. I’d probably stick to a token economy to keep things fair.
I could imagine dice roll to for random tables related to scene framing.


I agree with beepeegee. Tokens to spend or collect to determine which character has an narrative advantage seems like a good way to handle this.


What I’m interested in is having some random element - having a roll at the start would probably take the place of having dice rolls during the scene so it gives you a certain amount of currency to spend during it rather than rolling moves. I quite like the need to improvise and the sense of risk that having a random element introduces. In a way @Deckard’s card approach is really good for this- you have your cards for the scene but you get to choose how and when to use them - it’s random but restricted and you still have a lot of agency on how you use it in play.

Using it like an engagement roll at the start of the scene also gives you an opportunity to decide how complicated things are as the scene begins and have that clearly on the table, fictional positioning is going to be most of the game and that is definitely more interesting to me when there are some unexpected or unpredictable elements to it.


How do you find that stuff like this effects immersion? I’ve played more DnD and PbtA than Blades in the Dark, but I find meta-things like engagement rolls and mechanics tied to scenes position players more as editors or screenwriters than as the characters themselves.

1 Like

Well games like these with scene framing authority sure put players more in a “screenwriter” position. Some like playing like this, some players prefer staying focused on “their” character.


@CrudelyDrawnSwords, personally I’d find it important that dice don’t determine the amount of narrative authority (here, token economy seems fairer) but rather the flavor of narrative authority.
For example. cards or dice rolls could determine if you are allowed to introduce new character, location or other plot element.

Also, haave you looked at Archipelago III? The Fate cards might be interesting.


Even if we keep scene mechanics character focused, that can still be another lever by which we roleplay. If we were to make “scene preparedness” a stat, it could fit neatly into the superhero genre for “mastermind” type characters, heroes or villains.

In regards to your second comment, I think dice and tokens both have benefits. A token system ensures that all players get an equal amount of scene influence, but a random check both adds variance and lets players exert different levels of influence.

For example, in a post-apocalyptic survival game, having scene influence fluctuate emphasizes the disempowerment and need to always consider retreat. You never know when things will be stacked against you.

Alternatively, in a heist game, having “scene control” be a stat means players that enjoy setting traps or creating infiltration routes can specialize while players that prefer responding to surprises get more points to raise other stats.


It’s not the play style (or result) one is after when playing with these kind of mechanics. The “fun” should instead be about being immersed in the story (or group building).