This thread in reference to Capes. An earlier revision is available for perusal, as well as development threads here, here, here, here, here and here.

The Big Playtest has completed. We had a lot of good, chaotic, player-driven fun.

I think that the core rules are within a few short steps of being done. We only hit one minor snag by the end of the playtest.

I tested charging players (including the Editor) an action to introduce an Event. This worked like a charm: In tense, combat situations, where you'll be paying STs to get extra actions each Page, this is the equivalent of charging an ST for the Event. It discourages the proliferation of Events. In looser, non-combat scenes, where you're more likely to just call for a new Page than to spend lots of STs, it's the near equivalent of introducing the Event for free. It encourages a proliferation of lightly contested Events. Unless someone sees a deadly loophole in that, I'm probably going to stick with it.


The one thing that got really confusing in combat-level conflict was using pure Events. We wanted to be able to say "Red Queen is trying to get her armor" without declaring that she either would or would not get it. Now we could technically say that the Event is "It becomes clear whether the Red Queen will get her armor", but that's very unwieldy and counterintuitive.

I recommend that the rules allow both Events ("Mary Jane falls off the ledge") and Goals ("Green Goblin wants to beat Spiderman to a pulp"). They'd both be dealt with in pretty much exactly the same way, rules-wise. I think it's just an ease of use feature, but again I worry that there may be a loophole I'm unaware of, or that it might push people away from using unfamiliar Events, or that people would never use Goals when they can just declare that what they want is what happens. Opinions are desperately sought and will be gratefully accepted.

Tony,

We had the same problem in our last IRC session of Capes. I like the idea of a dual-structure (Events and Goals) as a solution, but I have absolutely no idea if it would be functional. I guess I will have to think on it.

Thomas

I'll see what I can do about working up a solo-playtest example this afternoon when the boys are conked out. That should give me a better sense of the interplay.

Apparent digression which has a point:

We didn't use Frames in the face-to-face playtest, and I think the lack of constraint made it harder to narrate. (Classic paradox that limits inspire creativity by giving structure to the "blank page"). When I did think of things in Frames (typically for Shell's "flashback" sequences) I found them much easier to visualize.

So perhaps the solution for Events is to think of them as Frames that will appear (and are possibly featured on the cover of this issue). In other words, the Event is a visual image (comics = visual medium, right?) with an ambiguous interpretation.

To take the seminal Complication that inspired Events, "Victoria West Friendship" is the frame of West shaking hands with one of the heroes -- we know we're going to get to that image, but not what it means.

To take the problematic question of "how to depict Clarissa Sever trying to get her Red Queen armor back," you have the Event be an image of Clarissa, still in her prison clothes, in the evidence room reaching out for a piece of her armor -- "Clariss lays hands on her armor" -- but that image doesn't say whether she gets to don it or not, or whether it's been sabotaged/decomissioned, or whatever.

Actually, much as I like the Frames mechanic, I've found that I like being able to get two to three times more action done in the same period of time much better. The Frame mechanic inevitably pulls people from the "This is happening, this is our shared imaginary space" mentality to the "This isnt' really happening, we're portraying it" mentality, which causes a stumble in game momentum.

That's why it hasn't been in recent rules editions. Consider it a victim of streamlining.

Tony,

Firstly, congrats on a successful playtest - I enjoyed reading the report in Actual Play.

Re: the Red Queen and her armour - I've been so spoilt by the IRC transcripts that I tried to find one for this Scene! I'm guessing that at some point, the Red Queen was separated form her armour, and that, later on, she got an opportunity to grab it back.

If so, I suspect that the relevant Event is the appearance of the armour.

Which leads me to think, if I were a Hero or Villain and I had been deprived of my costume/gear/whatever - is it fair play to bring it back within reach as an Event? I'd still need to Resolve the Event to actually get my stuff back, but it seems like a reasonable way to spend an Action.

Re: Frames - Even if they don't make the core rules, I think there is a lot that is good about Frames - can they be retained as an optional rule?

Frames or no Frames -- and I can see the argument that they can feel very artificial -- I think the key question is giving players (including the GM) the incentive and structure to create vivid images.

Obviously you can't have a rule that says "roleplay well" or "narrate well," but there should be some way to encourage "strong visuals."

Sydney: I agree with the strong visuals, and I agree that there is some connection between them and Events (possibly even with Goals). I'm not exactly sure how to describe it. It may be easiest to instill the idea through a good Example of Play, but that seems a cop-out.

Doug: The Red Queen was being sprung from prison, and Sydney (playing her) reasonably complained that most of her gadget powers were inaccessible because she didn't have her gadgets. So we decided they were in a nearby armory/evidence-room and that she wanted to go get them.

The question became this: Do we create an Event that says "She puts on her armor" and say that if the heroes win it then they've somehow booby-trapped the armor (which we considered, given the presence of her ex-partner Chessmaster and his technical skills), or otherwise changed the situation so that her donning it is to their advantage rather than hers? It looked (to me) clumsy and forceful to aim it in that direction but there wasn't a clear-cut way to simply say "Red Queen wants her armor" and leave it to the dice whether she could fulfill that goal or not. There was, essentially, no rules support for a hero standing in her way and saying "You shall not pass!"

TonyLB wrote: Doug: The Red Queen was being sprung from prison, and Sydney (playing her) reasonably complained that most of her gadget powers were inaccessible because she didn't have her gadgets. So we decided they were in a nearby armory/evidence-room and that she wanted to go get them.

But how can either side meaningfully contribute to control of that Event by way of their proxies, the heroes and villains?

Good question, I suspec this links with an earler statement you made:

TonyLB wrote: There was, essentially, no rules support for a hero standing in her way and saying "You shall not pass!"

'kay... I'll throw up a quick hypothetical. I'm not trying to be a pain in the hindquarters, just to help illustrate where the problem can arise in play.

Let's assume that a hero uses some Attitude or Power and narrates a "You shall not pass!" moment. They are now between the Red Queen and the armor. Mechanically, they Control the "Red Queen Dons Armor" Event and Resolve it. So... now the hero has to fail, right? Because the "Red Queen Dons Armor" Event is resolving, which means that the Red Queen has to don her armor, by definition. Despite the fact that he controls the Event, the player cannot choose to prevent the Red Queen from reaching her armor.

A player, knowing that "You shall not pass!" can't work on this Event, is forced to do more creative stuff. This is great when you want the more creative stuff, but it blocks out classic superhero things like just punching the other guy until he gives up what he was trying to do. For such tactics to make sense you need a situation where someone is trying something and can either succeed or fail.

If you want that effect you sometimes have to word it much more carefully, like "It becomes clear whether the Red Queen will or will not reach her armor". That's a very clunky and unintuitive way to express something by comparison with "Goal: Red Queen wants armor". If such goals will be common then it's probably worth formalizing the short-hand.

One thing to note is that anyone can veto an Event. So if not all of the players are cool with the Red Queen grabbing her armor then it just will not happen. On the other hand, this seems like it would have been a perfect place to Moralize...

Now, even with group veto power you end up with a number of Events with absolutley no suspence (not knowing what the outcome will be) or tension (not knowing how the outcome will come about). I think that loss is rather telling, which is why I like the idea of Goals. On the other hand, having two types of things that you can introduce seems kind of clunky.

Thomas

TonyLB wrote: Let's assume that a hero uses some Attitude or Power and narrates a "You shall not pass!" moment. They are now between the Red Queen and the armor. Mechanically, they Control the "Red Queen Dons Armor" Event and Resolve it. So... now the hero has to fail, right? Because the "Red Queen Dons Armor" Event is resolving, which means that the Red Queen has to don her armor, by definition. Despite the fact that he controls the Event, the player cannot choose to prevent the Red Queen from reaching her armor.

Okay, I have done my solo playtesting / example of play thing, and it has shown me a few things.

First: Despite what I feared, it seemed (to me) that there was a pretty clear place for Goals and a pretty clear place for Events. I can't really see one dominating play at the expense of the other... they both do such useful and different things. I'm open to other opinions, though. Obviously I can only stretch my own mental mindset so far.

Second: Having Goals makes it easier for players to veto Events they don't like. They are no longer saying "You cannot attempt to introduce that story element into the SIS", they are saying "You can't do it without a fight". That's a much less provocative statement. We're all used to the idea of thrashing out contributions to the SIS by way of the rules.

Third: It is hard to get an Event that directly contradicts a Goal. I aimed for it pretty specifically with the "Orphanage Collapses" Event being in play at the same time as the "Rescue the Orphans" Goal. And, frankly, I thought that the resolution implied by the "broken" order in which they resolved was actually cooler than a more obvious order (i.e. Rescue the Orphans and then the Orphanage can collapse) would have been.


Anyway, I'd gone about as far as I could go in purely theoretical terms. Here's some practical grist for the mill. Does it change anyone's opinions?

Brainstorm. Not sure if it is a really good idea, but it is an idea. Have Goals be the default option for Conflicts and allow players to spend a Story Token to make something an Event. The advantage of Events is that they will happen. Goals provide direction, but Events provide a sort of way point.

Without seeing it in play I am not sure what that would look like. But the idea of being able to impose your will (somewhat) in advance through Events is kind of cool to me. You could even encourage "recurring" Events by allowing people to state some at CharGen. These "recurring" Events could be free... so you should see them more often... Anyway, it's a thought.

Thomas

Wouldn't this discourage using Events?

Somewhat yes. At the same time it would make them more powerful (I think). The decreased occurance may in fact be a deal breaker. My thought is that it makes a clear distinction between the Goals and Events... Again, it may not be a good idea, but I thought it might be worth considering. I do not have a problem with the rules as they stand, I just have a kind of... I don't know... weird feeling about the dual Goal/Event thing...

Thomas

As it stands, I don't think there is anything too unbalancing about Events - although an Event states that something will happen, the dice will decide who benefits ffrom the Event.

However, Goals are different. Here's a quote from Tony's solo playtest:

"Tim: Your choice: You can create an Event or a Goal. The Goal would be, like you said, "Clear the street". You'll either do it or be thwarted.

Bob: And if I'm thwarted, do people get munched?

Tim: No. There's no immediate downside to losing your Goal. You're just thwarted. There's a long-term downside, because the villain might get an Inspiration out of it."

I think this is unbalanced - by declaring a Goal, the player gets a "no-risk" opportunity.

I would prefer losing a Goal to have more serious consequences. Tony, was there a reason for deciding against this? For example, why shouldn't people "get munched" if the street isn't cleared?

I want to maintain active (indeed, rabid) player interest in the Events and Goals on the table. That is my agenda whether I'm the Editor or another player. Engaging the competitive instincts of your opposition is how you gain Story Tokens, plain and simple. They stake, they spend, they win, and you profit from it.

There are two ways to help increase player interest: First, you can create Events or Goals that you have correctly deduced will interest or provoke them. Second, you can encourage them to make their own Events and Goals on topics that (one hopes) are automatically of interest to them.

One rationale for the one-sidedness of Goals is to encourage players to create them. Specifically, it is an encouragement to players who want to win. It gives them a way to achieve something without risk that their efforts will be co-opted due to bad dice rolls.

One rationale for the narrative strength of Events is to encourage players to create them. Specifically, it is an encouragement to players who want to tell a story. It gives them a way to shape the story without risk that their efforts will be wasted due to bad dice rolls.

The other rationale for the one-sidedness of Goals is to provoke player interest in Goals I create. By having the mechanical outcome radically change the future of the story, you empower players who want to achieve something tangible, increasing their interest in winning the Goal so that they achieve their aim.

The other rationale for the narrative strength of Events is to provoke player interest in Events I create. By pre-ordaining a visually striking or conceptually important outcome, I empower players who want to have their characters in the spotlight of such events, increasing their interest in winning the Event so that they can make their character central to its narration.


I suspect that Thomas sees Events as "too powerful" because they fit with his style of play, and that Doug likewise sees Goals as "too powerful" because they fit with his style of play. Did I hit the nail right on the head that time, or did I just bang my own metaphorical thumb?

TonyLB wrote: I suspect that Thomas sees Events as "too powerful" because they fit with his style of play, and that Doug likewise sees Goals as "too powerful" because they fit with his style of play. Did I hit the nail right on the head that time, or did I just bang my own metaphorical thumb?

Out of interest, Doug, who do you think (in an ideal world) should decide success or failure?

Personally, I'm all keen to let players decide when they succeed or fail, as long as there's a way to bypass the uninteresting variation where they choose to succeed every single time. But it's a matter upon which rational people could reasonably disagree.

TonyLB wrote: I suspect that Thomas sees Events as "too powerful" because they fit with his style of play, and that Doug likewise sees Goals as "too powerful" because they fit with his style of play. Did I hit the nail right on the head that time, or did I just bang my own metaphorical thumb?

TonyLB wrote: Out of interest, Doug, who do you think (in an ideal world) should decide success or failure?

Thomas: I agree that these rules are primed for serious playtest. I hope to start up a campaign in the next month or so, to get it consistent exposure and to test the long-term game.

That having been said, I may shy away from making ground-breaking rules changes for a reason that has little to do with the aesthetics of the game: Simply put, I've assigned myself a deadline. I'd like to have a finished book in hand to sell at Dreamation, end of January. That means I really need to put my nose to the grind-stone and start putting out pages at a pretty spectacular rate, especially if I want to finish the planned supporting material (the all-comic-style example of play, the mapped out city with pre-made encounters, pages and pages of Click-and-Locks, charts for instantly naming minor characters, charts for instantly naming villainous inventions and deathtraps, and a tested Quick-start scenario for new groups and demoes).


Doug: To give credit where credit is due, I know for a fact (though my Search-Fu has failed to turn up the evidence) that Sydney has proposed that earlier versions of Capes were trending toward Editorless play. Of course I'm notoriously bad about accepting such insight until I come around to it on my own terms... it's an arrogance thing, I think.

So I'll tell you what convinced me: Over the past few days I've been writing up the One Page Rules Precis, so I was desperately trying to pull out verbiage to make all the rules fit. I found that anywhere I said "Editor" or "non-Editor-Player", it was just adding words and clouding the actual rules. I finally pulled out all mention of the Editor, and the rules turned out tighter and clearer.

I plan to do the same thing with the main rules. The final rules may retain some mention of the word "Editor", but only in discussing one way that a group could be organized. There will be no rules recognition of any privileged member of the playing group.

(Sydney chimes in from damnably expensive rental broadband in his hotel)

I was myself thinking that Goals having no downside risk was a problem -- it gives victory-minded players an incentive to do something that is less interesting dramatically. My preference would be that, as with Events, losing has a strong downside beyond generating Inspirations for the other side (since Inspirations don't always materialize due to flukes of high die vs. low die comparisons, something that also still needs a little tweaking); perhaps declaring a Goal X puts X at stake in some way that it will definitely come back to bite you in an equal-and-opposite reaction if you fail.

But this may be more a playtest issue than something to write into rules on Tony's deadline.

TonyLB wrote: Thomas: I agree that these rules are primed for serious playtest. I hope to start up a campaign in the next month or so, to get it consistent exposure and to test the long-term game.

TonyLB wrote: ... if I want to finish the planned supporting material [e.g.] the mapped out city with pre-made encounters...

Danny: "OK, I'll put the 'Hell's Kitchen' surrogate here in Area B."
Liz: "Well, let me place the 'Blighted Industrial Park next door in C, then."
Eric: "Hmm. Too straightforward. I'll place 'Yuppie Gentrification Zone' right next door in D."
Sydney: "And 'hillside estates' in A, overlooking it all?"

TonyLB wrote: To give credit where credit is due, I know for a fact (though my Search-Fu has failed to turn up the evidence) that Sydney has proposed that earlier versions of Capes were trending toward Editorless play.

TonyLB wrote: Of course I'm notoriously bad about accepting such insight until I come around to it on my own terms... it's an arrogance thing, I think.

I feel compelled to respond quickly, given the expense Sydney's going to!

Sydney Freedberg wrote: ince Inspirations don't always materialize due to flukes of high die vs. low die comparisons, something that also still needs a little tweaking

[P]erhaps declaring a Goal X puts X at stake in some way that it will definitely come back to bite you in an equal-and-opposite reaction if you fail.

May I suggest rather than The One True City, you make up a generic map and then allow people to fill it in with click-and-lock neighborhoods? This makes for 5-minute player-driven setting generation

Hmmm, why do you need a map at all?

It's not like this is a game where you need to calculate "travel time" between different locations.

However, if you're going to have a pre-generated setting, a relationship map linking crucial people/places/objects/events would be pretty nice. It would also help players to choose Exemplars.

You're right of course. A cartographic map gives relations like "north of" and "close to" and "bigger than", which are of no use in telling comic book stories.

So I was thinking about what sort of relationships between locations would drive stories. I tried out standard emotions, and it seemed both unnuanced and uninspiring. Do Dante's Pantry and the Financial District have "antipathy" toward each other? I guess, but...

Anyway here's my thought, stop me if you've already heard it: The relationships between neighborhoods (and between people, and between people and neighborhoods) are objects and past events.

Boston's relationship with New York isn't "Antipathy". It's "The Yankees". America's relationship with France isn't "Contempt", it's "D-Day".

This would mean that relationships can often be asymmetric. Dante's Pantry labels the Financial District with "Landlords", but the Financial District labels the Pantry with "Riot of '73". That tells us that the relation between them is characterized by oppression and injustice (in the hearts of the poor) and mistrust and fear (in the hearts of the rich).

If we then add in corporate mogul Victoria West, whose relation with the Financial District is the spectacular West Tower where she lives and works, and whose relation with Dante's Pantry is the Sister Ignatius Orphanage where she grew up, things start to look pretty nuanced.

If this seems like it would work, there are two follow up questions: One, does it need any rules backing or, like Kickers in Sorceror, can it just work through unstructured Drama? Two, what should the rulebook provide? A list of possible neighborhoods, leaving the relationships to be filled in by the group? A network of relationships, leaving the neighborhoods to be filled in by the group? Something else entirely?

(don't worry, the "day" of internet access runs from 3pm PDT to 3pm PDT for some reason, so I'm still on the same $10)

TonyLB wrote: So I was thinking about what sort of relationships between locations would drive stories. ....The relationships between neighborhoods (and between people, and between people and neighborhoods) are objects and past events. Boston's relationship with New York isn't "Antipathy". It's "The Yankees". America's relationship with France isn't "Contempt", it's "D-Day". This would mean that relationships can often be asymmetric.

TonyLB wrote: If we then add in corporate mogul Victoria West, whose relation with the Financial District is the spectacular West Tower where she lives and works, and whose relation with Dante's Pantry is the Sister Ignatius Orphanage where she grew up...

TonyLB wrote: Anyway here's my thought, stop me if you've already heard it: The relationships between neighborhoods (and between people, and between people and neighborhoods) are objects and past events.

Boston's relationship with New York isn't "Antipathy". It's "The Yankees". America's relationship with France isn't "Contempt", it's "D-Day".

Oh... yeah, I hadn't even noticed that now you can have "Free" Events and Goals, since they now have a cost. Which makes a meaningful niche for Recurring Conflicts attached to Exemplars, because after all they're free! I'll gorge myself on samples I don't even like when a store is giving them away for free. And the system doesn't suffer for a glut of Events and Goals (in fact it thrives on them).

As for how and when to add more... this system's got a conspicuous lack of character advancement. So maybe at the end of a session a player or players can pay ten Story Tokens to create a new Recurring Conflict associated with someone or something... your Exemplar, your archnemesis, a neighborhood, the mayor... whatever.

Tony,

Perhaps I haven't mentioned it recently, but you have really good ideas. I love the idea of character advancement being tied to producing more "free" Events...

Thomas

Cool. Glad you like it.

Do you think it should be a straight cost (e.g. Ten Story Tokens), or should it scale with the number of recurring Conflicts already associated with a given person or place (e.g. Five plus two per Recurring Conflict already on that element)?

i.e. Should you be encouraged to set Lois Lane up with "Falls from a High Place", "Denigrates Kansas Origins", "Goal: Romance", "Gets the Scoop", etc., etc., etc.? Or should there be an incentive to spread the Conflict goodness around, adding some recurring conflicts to Jimmy Olsen, Lex Luthor, the Daily Planet and so on?

One thought that struck me when I woke up early this morning (and tried to go back to sleep, and couldn't -- shouldn't 3 hours of jet lag from going FROM California TO the East Coast work the other way?) is that Exemplars could be, instead of simply Things On Your Sheet, a fourth category of Abilities, i.e. things that let you roll dice. (Rather like Relationships being treated as skills in HeroQuest, Dogs in the Vineyard, etc.) So alongside Powers, Attitudes, and Tropes, you could have a fourth category, Relationships -- covering both individuals (Exemplars) and neighborhoods -- whose strength and in-game power is rated on a scale of 1-5 like other Abilities.

Though I think simply allowing people to introduce a free Event/Goal (i.e. one that doesn't count as your turn) when their relationship is involved somehow is probably more elegant.

But this does bring me to something that's been nagging me about the system for some time -- I've raised it a while back, when someone mentioned Miller's With Great Power.... having everything about the character covered by one mechanic, Aspects:

The division of Powers vs. Attitudes vs. Tropes (aka Reactions) still feels somewhat artificial to me. The immediate pragmatic problem is that, counting 5 Drives, 3 Exemplars, a set of 5 Abilities, a set of 4, and a set of 3, a player must define 5+3+5+4+3= 20 (twenty ) distinct things to create a starting character -- of which 5+4+3 = 12, the Abilities, are completely freeform (whereas most systems with this many mandatory elements just have you assigning points among a fixed list of Skills Everyone Has, e.g. Paranoia). Click & Lock goes a long way to fixing this, though not all the way yet.

But the bigger issue I have -- and this is probably as much my personal game-design aesthetic as much as anything else -- is elegance: It just seems that we're making more distinctions than we need. Someone brought this up a while ago on an earlier thread, so I won't take credit, but I'll restate it, with the added insight of four sessions of playtest:

The various Abilities are really defined by three things. One is level (a 4, a 5, a 2, whatever). But the other two, the really crucial ones, are binary:
- Is it an action or a reaction, i.e. can be it used only on your turn or only on someone else's turn?
- Is it mundane, i.e. can be used once free of Debt but is then blocked, or heroic, i.e. can be used infinitely but costs Debt each time? Currently Powers are always heroic, Attitudes always Mundane, but Tropes/Reactions can be either -- which seems awkward.

Frankly, since chain-reactions of "no, I trope that" was such a big part of the fun in playtest -- and such a powerful form of teamwork -- that I'd be tempted to let everything be used as a trope/reaction:

Andy: "I blast by you using my Super-Speed -- I roll a 4!"
Bob: "Well, I use my level 4 Commanding Attitude to proclaim 'You shall not pass! I get a -- uh -- damn, two. I don't accept that result..."
Claire: "No, take it, I can trope that."
Bob: "Okay..."
Claire: "I use my level 2 'Unsettling Stare' to back up your Commanding Attitude and force him to pay attention -- I get a six! Ha! I am the champion!"

P.S.: And if someone can find the original post where the two binary oppositions of "heroic-debt/mundane-blocked" and "action/reaction" were originally proposed, please post the link, because I can't find it, and I want to make clear this wasn't originally my idea -- credit where credit is due.

Also, the other thought I had based on playtest experience -- mostly playing villains -- is that it is possible to get stuck so deep in Debt that you don't care: yeah, you accept being Overdrawn (with consequent die roll penalties), and never ever able to get out again, so you don't try. I'm wondering if (a) there should be more levels of hurt beyond just "you're overdrawn" and (b) if there should be a way to buy your way out of extreme debt with Story Tokens.

And hopefully this thought is more in line with the theme of "polishing" (vice "taking apart and putting back togethe) than the previous post!

Okay folks, big post time. Apologies for the length. I've tried to tighten it so I'm not rambling.

I understand the appeal of unified system. It's not to my personal taste. I think a little differentiation is one of those (to quote Dinh, who Sydney pointed me to) "wise restraints that set us free". So far I'm seeing this as a matter of personal preference, but there may well be an objective argument one way or the other that I've missed.

I'm feeling sort of smug, however, because I can field the other two points with a single answer... even though they look pretty distant. "Why aren't Exemplars a character trait?" and "Why would villains ever get out of deep debt?"

My answer to both of those questions is this: "The current rules are in place to encourage other players to grub for Story Tokens." For context on what I'm saying, go to the most recent solo playtest, search for the word "opportunity" and start reading from about three or four lines before that, forward. It points out my new hypothesis about how the game works.

Hypothesis: Debt on another character will be viewed as Story Tokens not yet realized. Much of the competition of the game will be to earn those Story Tokens.

Evidence: The last session of the playtest started with every player holding about five Story Tokens and me holding slightly less than that. This was in large part because the ST-distribution rules had been slightly broken in previous sessions (particularly that STs were rewarded for teamwork more often than for opposition, and they weren't spent quickly enough). The session ended with most players holding one or two STs. I was holding fourteen. Why? Because after the prison break I looked around the table and saw a huge pile of debt sitting on everybody's sheets, and I said "I should create a scene for them to burn that off and prove themselves."

This wasn't premeditated, it was serendipity. I was mostly saying "God, Liz is really unhappy about how deeply overdrawn her character is... if we end the final session on that note she's going to be bummed". I was, in short, thinking about it as a classic GM, who sets out to lose challenges from a selfless goal of abetting the other players' fun.

But if I'd had the rules system better explained to me (as it will be better explained in the rulebook) I would have done the same thing from purely selfish motives, to gain Story Tokens. I absolutely cashed in on that scene. I gave you the ability to Stake and win, and you did so with a vengeance, to my advantage.

So here's why the villains ended up with so much Debt... you guys had too many Story Tokens from previous sessions. There was little motivation for you to lose Conflicts, and absolutely no motivation for you to create Conflicts custom-made to be lost. In short, you weren't motivated to take the Debt away from them.

But imagine a different situation... imagine many (if not all) of you had no STs to your name. Imagine further that I'd adequately described the rules. It's not much of a leap to think of Eric introducing a "Goal: Shame Atomaton" for Clarice, and then deliberately losing it. That's three tokens of Pride less on her Debt (because she'd have instantly staked) and three STs more in his pocket. And as long as the villain has Debt and somebody wants Story Tokens there's motivation to do it again and again. I think that's the feedback loop that should prevent massive Overdraw... other players won't let the villain get that overdrawn.

That's also why Exemplars aren't a character ability. Exemplars are not a resource for the player of that character. They're a resource for every other player around the table. You play an Exemplar when the hero is holding a bunch of your potential Story Tokens and you want to get them before anyone else does.