How story evolves over time

[TonyLB]

Let's examine the most bog-standard story-prep imaginable:  A GM says "Okay, they'll come into town and the Mayor will tell them that his daughter has been captured.  Then they'll go out and fight the werewolves.  When they win, they'll rescue the daughter only to realize that she's been infected.  Do they leave her as she is, kill her, or sacrifice a powerful magical artifact to cure her?  Moral decision!  Yay!  I'm a great GM!"

That story has a beginning, middle and end.  It has its own timeline.  During play that timeline unspools:  events cascade from the future, into the present, and then into the past.

Before play, clearly, that timeline is dealt with as a whole.  Say the GM thinks "Hey, what if the town blacksmith is in league with the werewolves?  He'd act shifty in the beginning, and then the werewolves might be well armed (with weapons of recognizable craftsmanship) in the middle, and there's a big moral confrontation at the end."  He's just changed the story as a whole, throughout all its parts.  The story, as a whole, is evolving over time, as stories should.

So now I've laid the groundwork.  Here's my  thesis:  Stories can and should evolve, as a whole, during the time that they are being played.  The current state of the art for representing and structuring such evolution is pitiably sparse, leaving groups of players almost entirely to their own intuitive techniques (for good or ill).  Even a small contribution in this regard can make a huge difference (as shown by some games that are starting to explore the territory).

Railroading is one treatment of how a story evolves over the time it is being played:  which is to say, it doesn't.  No matter what happens in play, the structure of the story will not be altered by it.

Intuitive continuity is another treatment of how a story evolves over the time it is being played:  the past is fixed, the future is characterized by vague and deliberately unconnected elements, and the present is a mixing bowl in which the future is connected to the past and structured by player choice.  The future doesn't, generally, change in its structure until it is inserted into the present.

From what I understand of TSoY's Keys, they are a player tool for controlling the evolution of the future:  When you take a Key of "I'm gonna murder this sum-bitch!" you are creating a future event (in the unfolding timeline of the story) by which that Key will be resolved.  Probably, that event is (at the moment you create it) the murder itself.  But, of course, the future may go through several futher evolutions in the real-world time of playing up to the point where that future event spools into the story-present.  So, really, the murder is a starting point to the ongoing evolution of the story-future.  You may not end up murdering him at all, after the future events get further mucked with.

So, why not apply this sort of mechanic much more explicitly?  I mean, we all know that when a player says "Don't worry, this infiltration will be a piece of cake," it guarantees that the infiltration will go wrong.  Is there some reason there can't be an explicit mechanic to back that up?

Returning to the example of the werewolf-threatened town, a GM could simply reveal to the players "Here's the starting story: standard beleaguered town.  You'll come in, be entreated by the locals, head out to fight werewolves.  There's fighting.  There's a rescue.  There's betrayal by at least one, possibly many, of the townsfolk.  Somebody's going to get infected with lycanthropy, prompting a moral choice.  That's where we start, let's play."

Then the players (including the GM) can all vie to control the evolution of that story as it plays out.  Here are some examples of what could happen:

• One of the players says "Behind our backs, the mayors face slips into a conniving, greedy look.  I'm adding 2d6 (or whatever) to entangle him with the later betrayal plot."
• "The crazy old woman on the edge of town may know something about the werewolves.  I'm using my success at pumping townsfolk for information to remove 4d6 from the necessity of fighting.  Let's go talk to the old woman."
• "Okay, we've eliminated the fighting from the future.  Right now it's a straight rescue mission.  But, honestly, I'm liking the idea that this is a conflict between peer communities.  I want to add 3d6 to a new event, 'Negotiate with Werewolf leaders.' "
• "Hey!  The negotiations have gone really well!  But we still have an 8d6 'Rescue the mayor's daughter' thing to do.  How do we make sense of that?  Ahhhh ... that firebrand anti-human werewolf who was arguing against us.  Remember when he slipped out at the end?  Let's entangle that 3d6 'Subtle exit' event with the rescue thread ... he snuck out, overpowered the girls guards, and bolted with her toward the sacrificial altar."  "Right, so the chief takes you down to accept prisoner transfer, and there's this scene of mayhem."  (NOTE:  The significance of the past is just as open to change by these rules as the content of the future is, as just shown)
• "Okay, peace is restored.  We all know that the mayor arranged to have her daughter kidnapped in order to try to get rid of the Werewolf village.  Daughter hates father, all is well."  "Hey ... we've still got that infection plotline, to push us to a moral choice."  "Oh!  Right!  What do we do with that?"  "Daughter's pregnant from her werewolf lover."  "Oh DUDE!  That's why the young warrior came along to help us rescue her!  That explains everything!"  "But now we've got a whole 'nother moral question to unravel.  You just know the alpha wolf is never going to let his tribes bloodline be tainted that way."

I really want to make this work (at least for Misery Bubblegum, but also more generally as a tool for any developing game).  I look at it right now, in the flush of realization, and it seems easy.  Not simple, perhaps, but straightforward:  make a system where the story arc evolves and develops based on player choices.  Make sure that the mechanic for evolution doesn't forcibly turn the story into something everyone will hate playing out.  Set the players loose and watch them have fun.

What's scaring me is precisely that I don't see any traps lying in wait.  What's going to go wrong with this?  Seriously, something has to, and I'd rather have some idea of where the difficult bits will be early on.

[Troy_Costisick]

Heya,

You're talking about who has credibility to add to the story, right?  Don't Universalis, Dogs, PTA, and Sorcerer basically accomplish what you're talking about?  I do think that each of them have made a major contribution to the gaming industry and have changed the way stories in games are created over time- first off by noting the unglyness of railroading and illusionism.

Am I off base here Tony concerning what you are talking about?

Peace,

-Troy

[Sean]

So here's what I'm imagining based on your post.

1. There's a setting already agreed on. Maybe the GM made it, maybe it's built into the game, maybe the group made it by collaboration.

2. GM prep involves a series of events and a pile of dice he can assign to them. (You'll need stakes for each side of the event: e.g. 'battle with werewolves' isn't self-interpreting - do losers get slaughtered, or wounded, or what? Lots of ways to handle this but you'll need to choose one.)

3. Players have characters. Let's say the characters have two resources:

a. Trait resources. Magic dude, bulging biceps, whatever. These dictate what your character does.

b. Story resources. Dice you can contribute to future events.

4. Play goes like this. Players can use their story resource to introduce new events or weaken/remove old ones. When they actually engage with events, their traits are marshalled to battle against the event dice currently in the other pile in the usual RPG fashion.

This seems like a promising basic framework to me at least. Two questions:

Question 1: When does the story end? When people feel like it? When the biggest event is dealt with? When all events are dealt with? How do all these dice piles move towards the final payoff?

Question 2: What drives play at the mechanical level? I'd think that the obvious thing would be to build up some kind of 'hero point' type third resource, distinct from the story resource and the trait resource, that you can use to add to your trait pools.


So sort of like this. The GM comes to the session. In his mind the climax is the fight with the Big Ass Werewolf and he gives this 10d6. A couple ways this could go.

1. The players decide to work towards that very climax and accumulate enough of the Hero Dice (how do they accumulate them? by overcoming the smaller challenges?) that their Trait plus Hero dice add up to 12 or 13 d6, making success in the final conflict plausible.

2. The players decide that they want the daughter of the village hetman to marry the big ass werewolf so they drive the fighting conflict down with their story dice and introduce this new 'convince the daughter to marry the werewolf' conflict.


Or some combination thereof.


Here's another thing. Is there any graininess to the in-scene resolution? Or is it all maneuvering the conflict and rolling the dice at the end like in My Life with Master, noting how that effects your dice for future conflicts? This is a choice to make.


I think you've got a good general schema here but the details of how it's filled in matter in a lot of ways: for influencing the experience of play, for influencing which choices matter (tactical choices in-scene vs. maneuvering from scene to scene vs. deciding which scenes are going to happen). A lot, as in what choices you can make and the tactics for choosing them ARE the game.


But I do like the idea of the GM prep being a list of weighted events, managed by a GM event dice pool and given to all the players up front. That sounds like a really, really powerful potential way to run a game.

[timfire]

Ummm... I think I hear two things going on. At first, it seems like you're just asking, "How do we (the group) decide what happens next?" That's what the question "How does story evolve" seems to boils down to. The answer to that is... well, very broad. There are many ways to distribute credibility.

Then you ask about your mechanic. It sounds like it'll work. Other games have explicit mechanisms for this kind of thing. Uni jumps to mind (want to see something happen? Spend a token!) but there are others. What seems unique to my not-terribly-broad-knowledge-of-games is that players must compete via rolling dice to get events to happen, while many other games have a simple currency for that sort of thing (drama points, Uni's tokens, etc.)

[MatrixGamer]

Returning to the example of the werewolf-threatened town, a GM could simply reveal to the players "Here's the starting story: standard beleaguered town.  You'll come in, be entreated by the locals, head out to fight werewolves.  There's fighting.  There's a rescue.  There's betrayal by at least one, possibly many, of the townsfolk.  Somebody's going to get infected with lycanthropy, prompting a moral choice.  That's where we start, let's play."

I see the Narrativist focus of your game play. I can't speak to that - it's not my style of play. But I do have some experience with the mechanism you've described above because I've used it in Engle Matrix Games.

I'm now calling it a "Plot Track" but it essentially describes the type of events that need to happen for a certain type of story to be told. So if it is a Sherlock Holmes game but instead of looking for clues and solving a crime, Holmes runs for public office and hires Le Strad as his dirty tricks man, then we wouldn't call it a murder mystery game. Personally I don't want to force the players to do any particular story so I view the plot track as more of a set of guide lines rather than laws (as Captain Jack Sparrow would say). I trust that if the players want to stray off the path that they will have fun doing that. It's their game so why not?

If on the other hand you want the plot track to require players to follow it, this mechanism would be solid rail roading, which I see you don't want to do. So if the plot track is open to change you may have two separate levels of play going on. One level would be the Narrativist, moral delima and dealing with the consequences of it game. The other would be a god's eye view of the game, vying over the big picture direction the players are to go in. The big picture level of play would likely not have to be visited much during the Narrativist play, but could be a pregame game. The players would then be sharing power with the game master on creating the next scenario. This would work for campaign play. Bringing it into the individual session though might be a problem. A god's eye view is pretty detached, which could undercut the dramatic effect of agonizing over "What will I do, and how will I live with that."

Matrix Games dive all the way off the cliff and make the creation game the game. It works. Play can be fun, but I don't think it's what you're after. Maybe a hybrid would work though.

In closing, I think this problem could be looked at as two levels of play.

Chris Engle
Hamster Press = Engle Matrix Games  

[TonyLB]

Don't Universalis, Dogs, PTA, and Sorcerer basically accomplish what you're talking about?

Jeez, I don't think so.  In Dogs, do players routinely say "Okay, the final confrontation of the story should be around that clocktower" and give that mechanical weight?  If so, can they do that at any time (i.e. the moment they see the clocktower in the distance, four hours of game-time before the final confrontation in question occurs) or only at the moment that their proposed story-element becomes the present-moment of the narrative?

I'm genuinely not talking about building up a set of past and present causes, in the hopes that they will have a future effect that you want.  I'm talking about rolling up your sleeves and directly impacting the future of the narrative.  If you had a book that you were reading, as a group, you'd have all the pages of the book, only one of which you're currently reading.  This would be like saying "Okay, because of what we just read in chapter 5, I'm going to rip chapters 12 and 14 out of the book and replace them with these pages.  Hand me a stapler."

Yes, in the end, it is about how you distribute authority to change the story.  But it's about providing tools that let you broaden that authority to directly change the story beyond the present moment of the narrative.  So no, I don't think Dogs and Sorceror do that.  I think Universalis could do it, but then Universalis technically can do almost anything, by building the rules in as you go.  PTA directly supports character input into the story arcs, but has no provision whatever for those story arcs to evolve as the game progresses.  You couldn't really say "Well, given what we just saw, the next episode can't be Sandy's spot-light episode.  It will have to be Mitch's spot-light nstead.  That's his third one this season!"

[TonyLB]

Sean, you have lots of specific questions that I will return to when I actually design a specific mechanic to do this for Misery Bubblegum.  Right now I'm just batting around the general concept, so I'm just going to field some of your more general issues.  I haven't forgotten the other questions though!  Pondering them!

2. GM prep involves a series of events and a pile of dice he can assign to them. (You'll need stakes for each side of the event: e.g. 'battle with werewolves' isn't self-interpreting - do losers get slaughtered, or wounded, or what? Lots of ways to handle this but you'll need to choose one.)

Well, is "Our funeral" the next event in the story?  If not then the losers clearly don't get slaughtered.  In fact, given the "rescue attempt" and "someone is infected by lycanthropy," I think the next step is pretty clearly laid out:  someone gets captured, someone gets bitten.  Losers, all 'round, and plus they're now entangled with the story elements that we knew we were going to deal with anyway!

3. Players have characters. Let's say the characters have two resources:

a. Trait resources. Magic dude, bulging biceps, whatever. These dictate what your character does.

b. Story resources. Dice you can contribute to future events.

Why would these be different?  If your bulging biceps aren't having an impact on what story we all choose to tell, why are we even thinking about them?  Maybe your bulging biceps intimidate the werewolves, making a conflict less likely.  Maybe your confidence in your mighty might makes you arrogant and abrasive, making a conflict more likely.  But if your bulging biceps are just ... there ... then they're strictly narrative color, aren't they?

[LordSmerf]

Tony,

If I'm properly understanding your proposal (and there's no guarantee of that), then the real dangers are going to be in pacing issues.  It seems to me that in a taditional continuity sort of game the pacing arises intuitively.  That is, the conflicts build up via some sort of natural story building process (or at least via a process that's so ingrained that we don't notice it).  Once you start thinking about the future events and how they work out you have to start being highly intentional with regards to pacing.  Do we need to be pushing quickly toward the clocktower, or do we need to slow things down?

There's also the ubiquitous "What about people screwing up the parts you thought were really cool and replacing them with drek?" issue, but that's probably workable.

It also seems that inherent in this method of story structuring is a lessening of tension.  Take for example your "Nothing could go wrong" thing.  Within specific genres we know that this means something will go terribly wrong.  Assuming that we're playing in such a genre, we naturally introduce a mechanical certainty of such a thing (or so it goes in your version).  Now, let's say we spend two hours of real-time play building toward our terrible event when someone says "Hey, let's play against type!  Nothing goes wrong after all!" and spends the necessary resources to make it so.  Suddenly you have this unfulfilled expectation, this anticipation of an event that never occurs.

It seems to me that while what you're talking about here is a fascinating set of tools for story creation such that our creation of stories will probably be more interesting/more powerful, at the same time it may (I'm not sure yet) weaken things from the perspective of players as audience.  We are less able to feel the tension of things... maybe.

Wow... that's really disjointed.  Maybe some of it's useful...

Thomas

[Sean]

Hi Tony!

Cool. I think I'm getting what you're saying. I think your questions for me are more exploration of options, which is cool, because I think you're on to a neat general idea here (events in the game as something in conscious play).

One thing you could do is hand over the list of events and freeform the order. Another thing you could do is have the events in fixed time order. If it was the first (which is what I was thinking) then establishing stakes is more important. If the order's fixed (unless the players or GM spend their story resource to change it) then stakes doesn't matter as much - as you say, if the 'someone's inflicted with lycanthropy' conflict follows the 'fight with werewolves' conflict no matter what, again unless someone spends the story resource, establishing stakes is less important. (Though if someone was getting lycanthropy win or lose, what was happening in the werewolf fight? They're still there if you lose and gone if you win? There's still some need to know what the conflict decides, I think.)

The danger of the fixed order of events is that people start to feel powerless, I think. Unless everyone has a whole lot of that story resource going around. But then you run the risk of endless reshuffling. I don't know.

The trait and story resources don't have to be different. The question is, what are you resolving? There's

(a) what scenes you're going to have, and in what order

(b) how you resolve the individual scenes

Having the same resource for both might be cool, actually, depending on what you were trying to do. It could work either way.

The one thing, though, is there needs to be something you're going for in the scenes I think if you're managing the game at this level. Like how in MLwM you're trying to accumulate love, driving play towards the big scene at the end.

That could create some interesting outcome tradeoffs, like 'hmm, built up three hero points in that last encounter, do I want to use them to buy off the 'inflicted with lycanthropy' outcome or save them so I can introduce a 'the baron's daughter falls in love with me' event later?' type deal.

But this is just getting too abstract for me at this point. Thanks for the cool idea! I don't see any absolute problem with it, so maybe you should get crackin' on the system that implements it!

[Andrew Norris]

Hi Tony,

I follow what you're saying, and I definately think it's doable. So far in our group we've only had that kind of thing happen in a scene- or session-level scope, but I don't see why it couldn't be expanded to encompass events that are several sessions away.

Sorry I don't have more to add, but your idea made something "click" that's previously just been rolling around in my subconscious. I'll see if I can think of any potential pitfalls.

[MatrixGamer]

It seems to me that while what you're talking about here is a fascinating set of tools for story creation such that our creation of stories will probably be more interesting/more powerful, at the same time it may (I'm not sure yet) weaken things from the perspective of players as audience.  We are less able to feel the tension of things... maybe.

Ah, someone saying the same thing I said. There is a possibility that taking the god's eye/scenerio building perspective will ruin the tension of play.

Scenario building - intergames - would be a good thread on its own. That and looking at time lines (for instance how does a game handle flashbacks in play?)

Chris Engle
Hamster Press = Engle Matrix Games

[Doug Ruff]

Hi Tony,

For what it's worth, what you're talking about is something I've been thinking about for a while... which was inspired by some of our chats about Capes and Misery Bubblegum, so I'm certainly not claiming prior creativity.

I think this is mainly about giving player input into story element a 'mechanical' basis, by which I mean players have resources to devote to story elements, which are then 'rolled' in some way as a conflict.

If so, Universalis does come very close to this - player impact on story is mediated in a 'diceless' way, using Token bids. It's at the character-level of conflict that the dice get rolled.

To make sure I'm on the right track - if we played a game of Uni, and the characters weren't made-up people (or aliens, or furries) but were story elements instead, would this be close to what you are trying to achieve?

[TonyLB]

It also seems that inherent in this method of story structuring is a lessening of tension.  Take for example your "Nothing could go wrong" thing.  Within specific genres we know that this means something will go terribly wrong.  Assuming that we're playing in such a genre, we naturally introduce a mechanical certainty of such a thing (or so it goes in your version).  Now, let's say we spend two hours of real-time play building toward our terrible event when someone says "Hey, let's play against type!  Nothing goes wrong after all!" and spends the necessary resources to make it so.  Suddenly you have this unfulfilled expectation, this anticipation of an event that never occurs.

I think I get what you're saying, Thomas.  Either it is hard to defy expectations or it's easy.  If it's easy then nobody is encouraged to build up their expectations, because the mechanics don't support them.  But if it's hard then how can you suddenly reverse people's expectations?  And if you can't suddenly reverse people's expectations, why would they ever feel surprised at the outcome of anything?  It's all just their expectations coming to pass, and those expectations have probably been in place for some time.  They might have been surprised that the future changed an hour ago, but they've had an hour of game-time to adjust.  Is that right?

[pells]

Returning to the example of the werewolf-threatened town, a GM could simply reveal to the players "Here's the starting story: standard beleaguered town.  You'll come in, be entreated by the locals, head out to fight werewolves.  There's fighting.  There's a rescue.  There's betrayal by at least one, possibly many, of the townsfolk.  Somebody's going to get infected with lycanthropy, prompting a moral choice.  That's where we start, let's play."

A question :
Instead of telling your players this premise, let's say you keep your story line to this basic, no more preparation from your part. You've got characters, groups, each one having their own set of motivations. You have a timeline, werewolfs terrorise town, mayor offers his daughter...
You still don't know what your players will do, the game isn't played yet.
Let's take back your exemples :

Then the players (including the GM) can all vie to control the evolution of that story as it plays out.  Here are some examples of what could happen:

• you tell your players "the mayors face slips into a conniving, greedy look". They choose to keep an eye on him. Following him in the night, they see him meet strange characters in the forest. Who knows, maybe they'll prevent the daugther from being captive but unleashed the fury of the werewolves.
• your players decide "The crazy old woman on the edge of town may know something about the werewolves". Uhm, good question... you tell them of an old story about a curse. Maybe there's a way to remove it.

I think you can achieve the same goal this way. By preparing, not a fixed set of linear events but a context (I doubt it's the right word, though). And furthermore, you can throw in battles or investigations or moral dilemma, depending on your players needs.
The question finally : Why are you looking for a mechanic to do so ?

Mechanics or not, I believe, you, and not only your players, will have more fun.

Last thing :

I really want to make this work (at least for Misery Bubblegum, but also more generally as a tool for any developing game).  I look at it right now, in the flush of realization, and it seems easy.

I would agree, but add a lot of work. As you'll not use all the material you'll be writing.

[TonyLB]

I think you can achieve the same goal this way. By preparing, not a fixed set of linear events but a context (I doubt it's the right word, though). And furthermore, you can throw in battles or investigations or moral dilemma, depending on your players needs.
The question finally : Why are you looking for a mechanic to do so ?

Because I want to give the players power, not just influence.

Influence:  The players make recommendations, and the GM takes them really seriously, but always has the final word on what happens.  If the players strongly disagree with the GM then they are simply out of luck, because the only way they can achieve something is through his approval.

Power:  The players don't make recommendations, they take action on their own.  If the GM doesn't like it then he has tools with which to oppose them, which may succeed or fail.  Nobody has to ask the GMs permission to exercise their authority within the rules.