decoupling Reward Systems from broad-scale Story Arcs

Started by David Berg, October 16, 2011, 03:13:43 AM

Previous topic - Next topic

Dan Maruschak

Dave, I think it's hard for me to talk about game design in such an abstract way. If you were trying to actually design a particular game or achieve a particular effect it might be easier to keep the conversation grounded. At a conceptual level, I think the important thing is to keep player choices and resolution system results either orthogonal to the question of plot progress (e.g. maybe there's mechanical support for having evolving relationships but nothing that affects the character's goal: the Ringbearer is guaranteed to get to Mt. Doom, but the game is about figuring out how the Fellowship feels about each other along the way) or coordinated with it in a way that's fun (e.g. a pacing or level-of-detail mechanic rather than "reality-simulator" mechanics). You don't want the game sending mixed signals.

Obviously, I think my own game Final Hour of a Storied Age does pre-outlined plot well (although it's a GM-less game and a collaboratively created plot -- I'm skeptical that it would work if you just ripped the collaborative front-end out of my game and replaced it with a GM-decided plot, I think there wouldn't be enough buy-in). A big influence for me when writing my game was thinking about the difference between writers who work to an outline and writers who do "discovery writing", which is more like normal Story Now design where you have strongly defined characters who bang up against each other and create "plot" as they go. While a discovery writer is more prepared for surprising events, an outline writer generally knows the events in the plot but is open to being surprised by things they find out about the characters or about nuances of how particular scenes play out (at least that's my opinion). In my game, players usually have only a rough sketch of the character at the end of chargen and end up getting to know the character through play by describing their actions.

I think generalized conflict mechanics (like with negotiated stakes or whatever) could work too, but you need to figure out how to make sure that the stakes people agree to aren't going to interfere with the plot. The biggest problem I have in Mouse Guard is that the rules say players are supposed to determine their own goals in a conflict, so it's hard to put a conflict in the middle of the GM's turn because the players may not say they want the thing you think they should want. In general, I've grown kind of skeptical of negotiated stakes games (and they seem to be less popular in more modern indie games) because determining "good" stakes is something that seems to take a lot of skill or at least good storytelling instincts. I think there's potentially a way to work with slightly more mechanically discrete stakes that you could "open up" as the story progresses (e.g. killing a named NPC isn't allowed until Act II, killing a major villain isn't allowed until Act III, etc.). This could potentially give your story-builder some building blocks to work with that they could accurately determine whether they were sufficiently stable to build plot contingencies on. (I haven't really thought this through in too much depth).
my blog | my podcast | My game Final Hour of a Storied Age needs playtesters!

Callan S.

I think Filips outline is one that can possibly exist and could possibly be the case here. If it is the case yet is covered up for being called rude to talk about, that is a dire situation.

Ron: I'm not sure what you mean by preferences in regards to Filip's post? If it's along the lines of my question below, as to why roleplay when you could read aloud to a select group a text you've written? It seems a valid issue? Some desires just can't be forfilled by roleplay - roleplay can't do everything, of course.

David Berg

Gareth, I think I follow you.  So, as part of the GM's endeavor to tell the players what constraints they're operating within scene by scene, a cutaway is one more tool in the toolkit.  It may be more fun and more elegant than simply saying, "Okay, guys, don't room near the starliner's reactor core."

Planning such cutaways as part of prep allows/requires the GM to put some thought into what the relevant constraints will be before the chaos of actual play.  There are all sorts of options -- a list of planned cutaways could be a way to manifest plot points, or it could be a call for additional, supplementary plot points to fill in gaps.   

It also fits well with Todd's "make a movie" focus on an imaginary audience.
here's my blog, discussing Delve, my game in development

contracycle

Callan, there's a big difference between saying "I've had negative experiences with this sort of thing" and saying "the people who do this sort of thing are a type with some sort of malevolent personality problem.  I have never disputed that plenty of people HAVE had bad experiences, and I've also said that most of the existing GM advice as to how to do this sort of thing have played a big role in that.  What I dispute is the attribution of intentionality.

David, yes that's it.  Now the larger point is this: those cutaways are useful primarily because the help frame the at-the-table activity in the context of existing task resolution systems.  That is, to some extent - how much of an extent I'm not yet sure - they are solving a problem that arises precisely because the rest of the system is based on the world physics model. 

As you say, the cutaway serves to communicate the existence of constraints scene-by-scene, while the boxes-and-arrows example I gave previously was an attempt to imagine how the same sort of constraints could be communicated action-by-action.  It is that function of communication that I think will be key to properly facilitating story-before games.

The cutaways example and Todd's wagon-chase example both rely expressly on a cinematic convention that is commonly shared by players and GM's, and that provides a framework in which a working relationship can be conceived and understood.  But there are a lot of other situations in which the conventions are not so widely understood.  For example, if you project the game into an alien society where the rules of appropriate conduct are quite different to those in our world, there is always the danger that players acting with perfectly good intentions will do things that will, should, must trigger extreme responses that can derail the planned story.  Say, a player does something which breaks the law in a manner that appears trivial to us but is serious in context lese-majesty being a prime example.  The response required could have such negative consequences that it effectively hijacks the direction of play.  PC1 insults the king and gets thrown in the Tower, PC's 2 and 3 plot to free him before he gets beheaded. Even if they succeed, whatever mission or direction they had been working towards before is now redirected to a game of "being outlaws".  The GM either intervenes, or tosses their prep in the bin.

So my general thrust is this: just as story-now had to break away from the quest for system to "better model reality" and instead discover "how to preserve protagonism and agency", I think story-before will have to figure out "how to communicate scope and constraint".  Maybe not only that, but I think it's going to be a necessary part of it.  Todd does this explicitly, the cutaways do it implicitly - how can it be done systematically?
http://www.arrestblair.org/

"He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast."
- Leonardo da Vinci

Callan S.

Gareth, I don't know Filips intent in his post, but I didn't read any description by him of negative intents. The road to hellish gaming is paved with good intentions. To me, Filip is just describing the details of the paving stones. If Filip's been judgemental, just ignore that part of the post, because the rest simply describes circumstances and, by my estimate, is quite important to consider.

Filip, sorry to keep talking about you in the third person reference.

contracycle

Bait-and-switch?  Slaughtering gameplay?  Saying that "these people", roughly, shouldn't be playing RPG's at all?  I have difficulty in seeing any redeeming features in that garbage.

I don't really see what more there is that hasn't been discussed to death a thousand times.  I've already acknowledged that the hitherto existing methods applied to this end had down sides, but Filip is arguing that no effective  methods should even be discussed.
http://www.arrestblair.org/

"He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast."
- Leonardo da Vinci

Callan S.

Occasionally on RPG.net I've seen threads which, summed up, seemed to me like asking "How can I stick a fork in my eye without it hurting so much?" which to me, begs the answer "Don't stick a fork in your eye?" but it's taken as being off topic or avoiding the question. But it is a possible answer. Although there may be solutions to a situation, it may possibly be that simply not doing the activity is the solution. Maybe, maybe not. To me Filip says it even more gently in simply suggesting there may be other ways more fruitful. I don't think we enact honest consideration if we don't consider that atleast even a tiny, fringe possibility of this. If it has been considered already to some degree, then I guess my last few posts weren't needed and sorry for my thread muss up. I'll leave it there, in either case, anyway.

Ron Edwards

There's nothing we can gain from what-he-said what-I-meant what-he-meant posting.

Please talk about the stuff this thread is about. David was very, very specific about what that is a few posts ago.

Best, Ron

David Berg

I agree with Dan that it's somewhat hard to have a purposeful game design discussion without the constraints of a specific game project.  At the moment, though, rather than trying to nail the one optimal approach for a game with certain color/setting/idiom, I'd like to explore a broader realm of productive techniques. 

I think one question all Story Before designs need to answer is, "So if the players don't determine what happens, what do they determine?"  So many games' incentives seem to hinge on character success.  You do X so you can earn points that give you a better chance to succeed when you attempt Y.  To me, getting away from that paradigm is exploratory and experimental, so for now I'm interested in a wide field of what might work.

Another big question is, "How do you get the most out of Story Before?"  This starts with understanding the approach's unique strengths, like the foreshadowing, dramatic timing, and big reveals you can find in planned fiction (i.e., novels and movies), as well as the unveiling and discovery of an extant creation.

I think the answers to these questions ought to connect.  Like, the players need to determine something that is complementary to foreshadowing, big reveals, etc.  Two possibilities that come to mind are details and impact.

Details:
The GM writes the plot, but the entire group makes it come to life.  Standards are applied for what is valuable in that endeavor, whether it be genre emulation, a vision of a specific setting, certain themes, etc.  Rewards ought to motivate the right sorts of contributions as per those standards.  Indirectness may be necessary, as I've found "Good job, that was very Lovecraftian, here's a token" to be underwhelming.

Impact:
The GM writes events.  The players maneuver their characters into positions where those events will matter greatly to them when they occur.  The payoff is that, when the GM reveals that the killer was brilliantly manipulated by the inept-seeming villain from earlier, the players respond intensely in character with shock, disbelief, and outrage.  "We let him go and now find out he's killed our best friend!  He's been laughing at us the whole time, that sadistic fuck!"  This provides the ultimate "well plotted, sir!" to the GM, as well as a huge adrenaline rush for the players. 

In trying to facilitate this, there's a potential conflict between the complicity of awareness and the genuine response of surprise.  How many specifics can the GM really plan and still be on target for these players and these characters?  If the players know too little of the plan, they may stray off into territory where the plan won't matter to them; if they know too much, the plan may come off flat.  Rewards here ought to motivate the strengthening of bonds between character and situation.  I, uh, have no idea what that would look like.  Drop flags for next session, strive to buy in right now?

Anyway, I hope all that babble throws some more fuel into the pot.  I'm happy if this thread goes down plenty of paths of different techniques for Story Before (as long as none of them are "Consider not bothering").  Dan and Gareth, I'll reply to your thoughts shortly.
here's my blog, discussing Delve, my game in development

David Berg

This was intended to address Dan's points but then got far broader.  So, anyone who's reading, consider this addressed to you as well.

Hi Dan,

That Storied Age mechanic sounds like it might apply.  The idea that there is a Chapter 2 that you will get to, but what you do in Chapter 1 determines when and in what position -- that strikes me as very appropriate.

My first guess is that "when" is the less important of those two in Story Before.  As a matter of fact, I'd like to leave the GM free to generally dictate timing of important fictional developments.  Letting a die roll delay a Turning Point for a few minutes sounds fine, but letting a whole series of resolution outcomes possibly delay it to next session sounds less suited to Story Before's potential strengths.

The more meaningfully the characters' positions evolve the better, I would think.  Whether that's effectiveness scores rocketing up and down the scale, resources flooding and bottoming out, or relationships and connections being drawn and crossed out, I'd guess that you want bang for your buck here.

At the same time, all these important changes need to not change the PCs' relationship to the Story Arc.

Hmm.  I wonder if GM and players could agree on character limits as part of character creation.  Brainstorm:

1) Each character gets a Dynamic.  The Dynamic is the type of character change the player is most interested in (game comes with list of genre-suitable Dynamics, GM refines further, then players pick?)

2) The GM and player discuss the bounds of each Dynamic.  How high and low can the character go?  Example: Courage.  The GM sees problems only if the character gets utterly fearless, but finds total terror compatible with the intended plot, so the limits are set at Very Courageous and Ruled By Terror.  The game system then moved the character around within that range.

Does that say anything about what the resolution system should resolve? 

If it's about success of character actions, then it needs to also (directly or indirectly, immediately or eventually) produce changes to the character. 

But maybe it's not about success of character actions at all.  Maybe participationist play is that rare situation when the Play Pretend model of "someone decides" is usable.  Maybe the GM decides, maybe the player decides, or maybe the player with the most points decides (with points earned for contributions to the stated aesthetic goals of play, perhaps).  If the GM decides the result of every attempt, and the mechanics resolve only how that effects the character, that is at least a pretty clear statement about why we're playing.

Note: in the above example, given goals of dynamics and limits, we probably want a negative feedback  mechanism, so anyone who gets Very Courageous doesn't simply stay there.

Quote from: Dan Maruschak on October 20, 2011, 09:29:01 PMAre more structural mechanics something you're open to, or do you also want to preserve the classic paradigm of players playing characters with capabilities described in terms what they can do within a fictional world?

At this point I'm open to anything that seems like it'd work. 

As for characters defined by in-fiction capabilities, I have two opposite thoughts:

1) Screw that!  The mindset of, "Here I am, here's what I want, what are my options, what would work?" totally butts heads with a planned story.

2) Yes, keep it!  The mindset of, "Think in character, try stuff, and discover what happens," is perfect for resonance and intense appreciation of the developments the GM unveils.

Personally, I agree with Eero that transparent and repetitive regurgitation of tropes is unappealing.  And I think that's a risk if you stick players in author/director stance with limited author/director powers.  The contribution channel may get a little too narrow. 

Interestingly, Todd's solution in Unknown Armies is to briefly hand out director reins to a given player at a moment that spotlights their character.  So that's an option.

My personal sweet spot would be if I could build a character who's a machine well-suited to the game's agenda and then just play them like a real person.  I think designing such a game might be more work than the alternatives, but it'd be super cool.

Quote from: Dan Maruschak on October 22, 2011, 01:11:49 AMI think the important thing is to keep player choices and resolution system results either orthogonal to the question of plot progress (e.g. maybe there's mechanical support for having evolving relationships but nothing that affects the character's goal: the Ringbearer is guaranteed to get to Mt. Doom, but the game is about figuring out how the Fellowship feels about each other along the way) or coordinated with it in a way that's fun (e.g. a pacing or level-of-detail mechanic . . . )

Agreed!

I think my brainstorms above fall under "orthogonal". 

As for "coordinated":

Maybe if the character changes being produced dictate which plot point hits when?  Like, if there are 4 PCs with 4 Dynamics and each Dynamic has two endpoints for a total of 8, then the GM devises 4 or 8 plot developments that will be triggered by hitting those endpoints.  Ehn, kinda cool, but kinda not proper Story Before.

Maybe if the character changes and the plot are both pulling on the same aesthetic rope in some structured way.  Something beyond just "we're all doing cinema horror and we know good cinema horror when we see it".  Like, I dunno, there are Potentials within both story and character that can be Unlocked by meeting certain conditions.  Like, if the group decides that Courage 3 / Loyalty 5 / Sanity 0 would be a great place to wind up, then achieving that rewards everyone involved with... uh.. crap, I don't know.  With a reminder to do whatever you were stoked about, that caused you to declare that a great place to wind up, I guess.  Or there could be an audience rating, some critic meter of how good the movie is.  A bad rating shouldn't mean a not-fun play experience, but a great rating could be something to shoot for.  The rating would have to be based on something that neither GM nor players can do alone.  This idea needs more thought!  Help would be most welcome!

Maybe the players can write wish lists of scenes they want to have, and helping the GM pull off the GM's vision earns them such scenes.  Though "trade" and "taking turns" is no good; there needs to be synergy.

Quote from: Dan Maruschak on October 22, 2011, 01:11:49 AMan outline writer generally knows the events in the plot but is open to being surprised by things they find out about the characters or about nuances of how particular scenes play out

That reminds me!  I can't believe I haven't touched on the play -> prep cycle.  It's common in my experience to have the GM's prep for session 3 react very strongly to what the players did in session 2 even if the GM is trying to tell his/her own story.  I've never seen this structured in such a way that the players know this is happening and feel rewarded by it, though.  Maybe at the end of session 2, the players could have earned a certain number of input points, which they can then spend on scenes or events or NPCs or objects or locations, which the GM must then include in the prep for session 3.

More generally, reminder to self: tell GMs in big letters on page one, "You get to be a control freak about these certain things over here, but not about these other things over there.  If you can't enjoy being surprised, just do a reading, don't play a multi-player game."

Quote from: Dan Maruschak on October 22, 2011, 01:11:49 AMI think generalized conflict mechanics (like with negotiated stakes or whatever) could work too

I'm actually not having any inspirations in that direction, but I'd be happy to hear suggestions.

Ps,
-David
here's my blog, discussing Delve, my game in development

contracycle

Some points based on the above.

A planned story doesn't necessarily need to be total.  So it would be quite possible to constrain the whole story of "how the ring was brought to Mt Doom" and then leave the decision to throw it in or not to players or system or a mix of both.  What would happen then is essentially a bunch of special effects that play out as the credits roll.

On a similar note, it is possible to do branching plans, but this is usually sub-optimal because it means that some amount of prep will not be used.  As such it's quite an inefficient method for someone writing for themselves and their group, but I should mention that this problem goes away when the writer is a third party, providing material to multiple groups.  So maybe game1 has a branch decision at the end, and you follow it on with game A or game B depending on which branch your group decided to take.  If the person writing these is in the same relationship as a module writer of old, selling their work to various groups depending on what choice they made, then the branches won't be wasted after all.
http://www.arrestblair.org/

"He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast."
- Leonardo da Vinci

Filip Luszczyk

Quote from: David Berg on October 21, 2011, 07:32:40 PMEarly in my roleplaying career, I tried to use my GM position to tell my stories, and though there were many times when that didn't go so well, there were some times where it went really well for everyone involved.  I know I got something unique out of it, and I think the players did too.  So I'm convinced that functional possibilities do exist in this direction.  How hard they are to design for is another question.

Everyone, I'd prefer that we keep the focus on the "designing for" part here in this thread, please.

As for designing. Do you know that even today people are using this time-proven medium to tell stories to small immediate audiences? Seems like you are designing with this specific activity as the goal. Perhaps it would be more fruitful to start your design from the activity itself rather than refit the conventions of another "game-like" activity into a rough shape of the former.

You can learn more about traditional storytelling via sites like this one. I think it shouldn't be difficult to apply its methods to not so traditional topics, like pulp genre or other nerdy matters, should that be needed based on the target audience.

Next, I believe you might find reading this classic article useful. I also think it's particularly useful to notice another phenomenon existing within the hobby, one roughly equivalent to the one described, but related to games rather than stories. Specifically, the linked article gains a new meaning and mostly still makes perfect sense when all instances of the word "story" are substituted with the word "game".

I also believe homosexual people should engage in homosexual sex, while heterosexual people should engage in heterosexual sex. Not necessarily in each other's vicinity though, unless bisexual, when in turn they don't benefit from searching for a vagina on a men's body. But hey, I never said I'm not a bigot, did I?

David Berg

Quote from: contracycle on October 24, 2011, 10:22:33 AMit would be quite possible to constrain the whole story of "how the ring was brought to Mt Doom" and then leave the decision to throw it in or not to players or system or a mix of both.

That appears to be what Todd does.  I must confess, it confuses me a bit. 

I mean, if he told me up front, "you get to decide the ending," then I'd probably be looking forward to that the whole game, evaluating momentary changes in position with an eye toward the finale.  That seems like a bad thing, as opposed to having things matter immediately in their own right.

And if he didn't tell me that, I'd probably be a little disoriented when it hit.  Like, "wait, I get no control over the plot, but now I do get control over how it all turns out?"

But maybe I'm failing to imagine the experience accurately.

Quote from: contracycle on October 24, 2011, 10:22:33 AMOn a similar note, it is possible to do branching plans, but this is usually sub-optimal because it means that some amount of prep will not be used.  As such it's quite an inefficient method for someone writing for themselves and their group, but I should mention that this problem goes away when the writer is a third party, providing material to multiple groups.

It's an interesting question.  Some of the GM tastes and techniques that apply here probably apply fairly well to writing modules.  That said, for now I'd like to focus on a proper interactive RPG, for which case I'd have to agree that doing prep that won't get used is generally a bad thing.  I guess it depends on what sort of prep, though. 

Some GMs have no trouble at all prepping and then implementing NPC stuff that may affect the players in different ways depending on what they've gotten up to.  "The Vampires declare war on the CIA" is going to change your espionage game one way or the other, but it plays out differently depending on if your character is standing there for the declaration.  I think that as long as the prepped event does matter to the players, how they experience it is often a good thing to leave up to chance and individual decisions.
here's my blog, discussing Delve, my game in development

contracycle

Well, on the topic of a final decision, let me draw on the example of CRPG's.  In the second Knights of the Old Republic game, for example,  at the end the Sith Lord gives you a lecture and tries to convert you to the Dark Side.  Depending on how you choose, you get a different final boss fight and different end cinematics.  In  Deus Ex: Invisible War you get arguments from three different factions, and deconstructions of each faction's position from the point of view of the others, and have to choose which one to throw in with.  In Mass Effect 2, you have a whole bunch of subsidiary goals to complete before triggering the final act, and depending on which of those you achieved, various ally characters get killed or not.

So there are quite a lot of ways to have the overall course of play focus on a climactic moment in which some sort of choice is made.  I don't think that comes across as particularly confusing. In the ring scenario, maybe you have a tracker which is modified by decisions over the course of play to determine whether you have enough control to dispose of the ring or not.  One of the purposes of constraint up to that point will be to project the characters into situations where such choices may be made - the wraiths are hunting you, do you try to go invisible?, etc.

The thing is that whether a choice matters to the players and whether it matters to the GM are two different things.  They only matter to the GM to the extent that they obviate future planning, and so where a choice can be made without doing so - by putting it at the end, for example - then its perfectly feasible to include that sort of thing.  It isn't necessarily the case that the GM is imposing an interpretation on events.  So it isn't really necessary for the idea that the GM has control over "the plot" to include in that the final decisive point of play.  What they're really controlling is setting up that decisive moment, contextualising it, adding dramatic flourishes and so on.
http://www.arrestblair.org/

"He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast."
- Leonardo da Vinci

Dan Maruschak

Quote from: David Berg on October 24, 2011, 08:46:37 AMMy first guess is that "when" is the less important of those two in Story Before.  As a matter of fact, I'd like to leave the GM free to generally dictate timing of important fictional developments.
I think pacing is an interesting question. Generally, I think of it as a pretty low-level thing, on the same sort of level as things like the amount of description. It can have an impact on the feel and tone of a story, but it isn't necessarily core to a story. I was thinking about HBO's Game of Thrones, where most of the battles are sort of skimmed over (presumably for production reasons) in contrast to the books where there's usually a bit of detail to them: the story is the same, but the way it's told is slightly different, with different pacing. But like you say, the timing of important events is in some ways what "plot" is all about. I wonder if there's an easy way to differentiate the two, or if it's more subtle, like the difference between "important fictional details" and "just color".

QuoteHmm.  I wonder if GM and players could agree on character limits as part of character creation.  Brainstorm:

1) Each character gets a Dynamic.  The Dynamic is the type of character change the player is most interested in (game comes with list of genre-suitable Dynamics, GM refines further, then players pick?)

2) The GM and player discuss the bounds of each Dynamic.  How high and low can the character go?  Example: Courage.  The GM sees problems only if the character gets utterly fearless, but finds total terror compatible with the intended plot, so the limits are set at Very Courageous and Ruled By Terror.  The game system then moved the character around within that range.
Do you think there's a danger of making things over-determined? If the GM is going to be bringing a strongly pre-planned plot, I'd worry about players also trying to pre-plan or pre-explore characters. But maybe I'm just having a personal taste reaction, since I don't like to overintellectualize what I want to explore with a character before I play them because it keeps me from engaging with them emotionally.

QuoteDoes that say anything about what the resolution system should resolve? 

If it's about success of character actions, then it needs to also (directly or indirectly, immediately or eventually) produce changes to the character. 

But maybe it's not about success of character actions at all.
I think I might quibble with the idea that it needs to lead to change. I think it could also be about revelation. There have been a few instances in my Storied Age playtests when we were essentially asked by the game: "You failed. Why?" Having to come up with good narration that answered that question helped us develop the characters. But I suppose that some people might classify "learning something about the character that you didn't know before" as a kind of change.

QuoteMy personal sweet spot would be if I could build a character who's a machine well-suited to the game's agenda and then just play them like a real person.  I think designing such a game might be more work than the alternatives, but it'd be super cool.
I think games like DITV and Apocalypse World do a good job of framing player decisions about character actions in an outward facing way that lets you think about what you want to do without worrying about knowing if what you do will succeed. Personally, I think I have an easier time "just playing my character" in a game like DITV than I do in FATE, which tells me explicitly what I'm good at (or average, fair, great, or superb at) in terms of interacting with the world. I'm thinking that the big problem is the reliable expectations you can build up: if you can know as a player that you should have a 30% chance of successfully shooting out the truck's tires then you can't mesh comfortably in a story that requires that the tires not be shot out. But what mechanics are or aren't "anti-immersive" is a highly debated topic.
my blog | my podcast | My game Final Hour of a Storied Age needs playtesters!