Weaving worlds: Yes but games should produce stories

In my penultimate post in this series I mentioned Ron Edwards and the Forge.  The Forge closed down in 2005, having done its job (and Ron Edwards ensuring that he was controversial  by talking about bad games giving people brain damage) – and most of the community there moved to Story-Games.  And they’ve been producing interesting enough games that they are worth the final post in this series.  Story Games tend to have seven aspects; three which are common in the Forge-ist narrative RPGs of my previous article and almost ubiquitous in the Story-Games wave, and four that are almost distinguishing features of what are often referred to as Story Games.

The three that are common in the Forge-ist RPGs are:

  • Challenge Based Resolution
  • Fail Forward
  • Everyone designing the universe

And the four that are almost distinguishing marks of this wave are:

  • “Yes-but” resolution
  • Intentional, rules-mediated inter-PC drama
  • Less, or even no role for the GM
  • Actions matter more than potential

I’m also going to mention four games in this category (or three and one hybrid toolkit game) and why they are awesome to illustrate this wave of games:

  • Leverage
  • Monsterhearts
  • Fiasco
  • WFRP 3e

Challenge Based Resolution

Challenge based (rather than task based) resolution means you resolve what people are trying to do rather than the component tasks that may or may not be relevant.  You roll to open the safe to find the document you need as that keeps the story rolling rather than rolling to open the safe when the document is actually in the skip outside.  This way the players have much more control over what they are doing, making up for the necessarily imperfect information that’s passed by the GM to the players about the world.

Fail Forward

Fail Forward is the principle that you only roll when the consequences of failure are interesting.  In old versions of D&D, if you failed your roll to pick a lock then for whatever reason you couldn’t try to pick the same lock until you went up a level.  Under a fail forward system you are assumed to be able to do anything easy – if you fail you probably still succeed the conflict with consequences (such as getting the lock open just at the time the guard appears) but the reason you were trying to open the lock just got a whole lot more complex.

Everyone designing the universe

In a classic toolbox or gamist RPG, the GM designs the setting and their territory stops short at the edge of the PCs, and the PCs have very little control over the setting, not even getting plot points/fate points/drama points.  In a Forge-influenced game, the world building is almost at the “Always say yes” levels of Improv Drama.  In games like Fate (mentioned last time) the players are expected to list ties their characters have to the setting by way of aspects.  And at the most extreme in some games, notably Smallville, the entire local environment and most of the important NPCs that aren’t villains of the week are pointed to by the players in the character generation process, collaboratively designing a relationship map (it’s easier than it looks because there are good guidelines).  And the GM has little involvement in this.

Yes-but resolution

Most rolls made by PCs in RPGs prior to this wave are simple pass/fail.  Or even “Fail/Succeed/Succeed with Style” or otherwise pass really well – as in both Fate Core and Cinematic Unisystem.  But consequences are far, far more interesting than simple pass/fail, so having a system with consequences will add much more interesting narrative hooks.  To take an example, Leverage uses a dice pool from which only the two highest rolling dice matter – but any of the dice that roll a 1 create (or intensify) a complication.  So with five dice you can succeed at what you were trying to do, but create complications that will come back to bite you.

Intentional, rules-mediated inter-PC drama

In many traditional RPGs, the PCs are all expected to be on the same side (there are exceptions like Paranoia, of course – and the perverse incentive that encouraged the rogue to steal from the party in 1e).  This works well when the play is essentially challenge derived play of the original RPGs – but when the purpose of a game is creating a good story, good stories have dramatic tension between the characters.  So RPGs that are about good stories ought to do the same – and mediate things by rules to keep everyone on the same page and the stakes where they ought to be.

Less, or even no role for the GM

There is a rule of thumb for narrative that says that collaborative stories work much better when the person responsible for creating the problems is not the same as the person responsible for solving them.  (There are exceptions in fiction; watching someone overcome their limitations can be fascinating, but that’s a story in which there’s one star).  In classic RPGs it’s the GM that responsible for creating the problems and the players solving them.  But in a drama-dependent game like Smallville, the GM’s main role is to throw in factors that set the PCs values at odds with each other rather than to run the entire world.  And with the players’ role expanding to cover large parts of the world, there is no theoretical reason you can’t remove the GM entirely – some modern RPGs do (mostly for one shot games).

Actions matter more than potential

I’ve finally reached what might be one of the fundamental design principles.  The rules should be designed to cover the sort of thing the character does on a regular basis rather than any theoretical abilities they have and are unlikely to do in the heat of the moment.  Yes, Superman can fly at almost light speed and blitz through Lex Luthor and co in theory.  But he doesn’t.  And the rules should reflect what the characters do – the Superman who turns up in the stories, not someone going by the theoretical limits Superman has (or doesn’t have).

Likewise, and more importantly (although this is hard to achieve in practice), the rules should reflect the rhythm of the story, and when the actions that make up the narrative come to an uncertain place, from which the outcomes are likely to fork.  The dice get rolled where in freeform the players would hand over to each other.

Now that sounded like a whole lot of theory – like my article on the Forge, I’m going to name some games I consider good ones that illustrate many of these points.


The Leverage TV series is a heist/con show, and the Leverage RPG is far and away the best con-based RPG I’ve ever played.  Because it’s a con centred around a competent team, the play is episodic and automatically story focussed (the story being how our heroes conned that scumbag) and it was my example of a “Yes-but” resolution above.  Flashback scenes, where the players get to retroactively change an aspect of the universe that hasn’t been revealed are expected.  And the character “skills” are the broad areas of expertise (Grifter, Hitter, Hacker, Mastermind, Thief) with what the characters choose to do trumping exact skills.  It also has a short and simple character sheet.  What it lacks from the general properties is inter-personal conflict (John Rogers, the executive producer claims that the lack of tension between the con-artist protagonists is the biggest fantasy in the show; they’d have ripped each other off before long).


The fan based tagline for Monsterhearts is “The bloodiest, sexiest teen supernatural drama ever” – and it lives up to that billing.  The premise of the game is that teenagers are messed up – and monstrous teenagers take normal teenage issues and turn them up to 11, with the monstrous nature reflecting the issues of the character.  Also that the teenagers will screw up and will screw each other up, and the primary driver of the story comes from the teenagers messing themselves and each other up, and growing up, and getting over themselves.  Interpersonal favours owed and getting under each others skin is represented by “strings” that can be cashed, and the system is very light and simple with a basic resolution system that involves rolling two six sided dice, adding your stat (based on things teenagers do), and on a 10+ picking a few outcomes from a list of good results – on a 7-9 using picks to say what good outcomes happen and what bad outcomes do or don’t happen.  (So a 7-9 violence check succeeds, but you need to pick one consequence out of being hurt yourself, giving them a string as they understand you better, or turning into your darkest self).  It’s simpler than it sounds, creates interesting outcomes, and creates them at all the right decision points for a good story.  Sounds complex?  I’ve just given about half the rules for the entire game.

I should in theory have talked about Apocalypse World here – Monsterhearts being a hack of Apocalypse World – but I find Monsterhearts more immediate and more polished, and the Apocalypse World webserver is currently down.  I’d also be neglectful to not mention that there’s a kickstarter for extra Monsterhearts guidance and skins going on right now.


In the spirit of Show, Don’t Tell, Wil Wheaton’s Tabletop did a playthrough.  Excellent improvised story?  It normally comes out almost that well.

Fiasco is a game about creating the story of a heist gone wrong or a Coen Brothers movie.  Or it’s effectively a superb real-time creative writing exercise generator.  In it every player takes one character, and they all have a relationship with each adjacent character.  They also share either a need, a location, or an object with each adjacent characters to create a setting.  The needs are the important part, driving the story to its spectacular, horrible conclusion.  And then once you’ve semi-randomly established your characters starting positions, you take turns to either create scenes (with the rest of the table deciding whether they end well or badly), or letting the rest of the table create a scene for your character, and you decide whether it ends well or badly – with half the scenes having to go each way.  Half way through you get a tilt – the wheels coming off (if they were on on the first place).  And there’s a fast table to show what happened in the end which you then turn into a scene each (and there’s no sequel, something the characters are normally very glad of).

An entire funny, black comedy created in under 2 hours between five people.  With you having known nothing going in.  And the game does this reliably.  What more could you want from a storytelling game.

Warhammer Fantasy Roleplay 3e

This is essentially a hybrid between this wave and a large toolbox game; most of this wave of games come as thin paperbacks, with most of the pages being guidance rather than rules.  It is, however inspiring with its resolution mechanic – using multiple types of custom dice so the outcome is rated for both success and luck (and with reckless and conservative play both being more likely to be successful than average, but coming with drawbacks).  Warhammer Fantasy Roleplay has always been a game about things going wrong and quite intentionally with PCs facing insanity or worse. The characters are all supposed to be working together (note: supposed) to an end so they take a party sheet based on their style of character, whether brash young fools, ruthless scoundrels, or even a secret coven on the run – this gives both bonuses and negative consequences for party relations going bad.  Characters are largely made up out of their skills and their actions – i.e. what they specialise in and are better at than anyone else.  It’s probably the most mechanically evocative and inspiring toolbox RPG I’ve played – but more than slightly fiddly especially compared with the other games in this category.

So if these games sound like fun?  Other than the hybrid that is WFRP 3e they tend to be thin, cheap, and easy to read.  I’ve named and linked three good ones – but they unashamedly focus on what they focus on and if you want a different story, find one that fits (or ask although my knowledge is far from encyclopaedic).  There’s probably one out there.  For that matter there’s probably an Apocalypse World hack (like Monsterhearts, Dungeon World, or Monster of the Week) out there.


3 thoughts on “Weaving worlds: Yes but games should produce stories

  1. Conflict resolution rather than task resolution exists at least in Maelstrom Storytelling, and arguably goes back to D&D’s combat mechanics and its hit point system.

    In the 20th anniversary edition of Over the Edge, Jonathan Tweet attributes fail forward particularly to Luke Crane (and his game Burning Wheel), though related ideas – like not asking for a check unless there are interesting consequences for failure as well as success go back to Robin Law HeroWars, and Maelstrom Storytelling.
    The idea of player-player conflict is also pretty important in Burning Wheel, and especially its Duel of Wits mechanic.

    • If I ever implied that any of those absolutely didn’t exist in older games, I apologise. Them being hallmarks of Storygames mean that almost all such games have those mechanics rather than that it is a distinguishing feature of the game to find them there. There’s very little new under the sun, but there are many new combinations.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s