[Deathbird Black] The Roof, The Roof, The Roof Is On Fire

[Baxil]

So I wrote this 24-hour game about comedy, murder, backstabbing, and throwing dice across the room.  It's called Deathbird Black.  (Ronnies thread here; game here.)  Some deity must have deemed its Ronnies win a crime against nature, because my next two gaming nights were promptly snowed out ... but, finally, it's been playtested!

Overall, it was a success - everyone declared they had fun.  But it also uncovered a few glaring weak spots in the design, which I hope to promptly address.  And it has given me much fodder for idle speculation; I hope particularly to engender debate about the Big Model implications of material forfeiture of fetishized pseudorandom number genera---

CAW

AAAAAAH

CAW

I kill decapitate throw my slug it with fire my scissors molotov cocktail baseball bat ! ! ! !

CAW

(fling, fling, fling, fling)

... Oh, geez, sorry, kinda stabbed you there.  I was aiming for our Deathbird, Bob.  Really!

By the numbers

I pulled together players from two different gaming groups and ended up with me plus five (Alice, Cookie, Corey, Mike, and Q).  The rules call 5-6 players the sweet spot, and play bore that out.  We had a good variety of opportunities for player interaction and shifting allegiances, and while there were enough players that people occasionally had to step completely out of the spotlight, it didn't seem like anyone had to fight for attention. 

GM + 5 also meant that one of the dice sets being passed around was a pair of d4s, in a game with a flat target number of 8.  I got several complaints about the d4s feeling punitive.  I will have to rebalance the success number, or keep d4s from appearing except in large games.

The game ran four hours from start of setup to end of feedback - which is what I had optimistically told people to expect.  This is a great sign; learning new systems always seems to drag the first run of a new game way out, and with Deathbird, learning the rules didn't seem to get in the way.

On the other hand, implementing them did - it felt like half the game was brainstorming character goals based on the vague direction of the card suits.  This didn't stop players from having fun, but it was a real mental effort - "much more work than I expected out of such a chaotic game," one player commented.  (More on this later.)

We ended up running five Free Play scenes (and the accompanying Deathbird and Detective bits).  The game's energy was waning by the end of the fourth - because comedy games' sweet spot is only a few hours; and because the final scenes developed a weird vibe.  If #5 hadn't run us out of cards, a few players were ready to drop.  (I think once the game is aggressively streamlined, it can push out to 2x that number or more, but not in its current state.)

Scene One

So it turns out that if you put a bunch of flawed individuals with competing goals in a confined space, noir just sorta happens.  And then there's fire.  Who knew?  Bob barely had any work to do by the time he showed up!

I'll describe the scene in some detail to give the flavor of the game in action.  The setting was Gary's Smokey Bar, late in a hot summer night.  The characters:

• Retired athlete Barry Bonds (Flaw: Dirty Secret / "I used steroids for my home run record"; Goal: Diamonds / "Get the publishing rights to Marty's tell-all biography about me")
• Destitute writer Marty (Flaw: Obsession / Jason; Goal: Hearts / "Save Jenny, who I'm madly in love with, from her life of crime")
• Heroic insurance man Jason (Flaw: Addiction / Selling insurance; Goal: Diamonds / "Get Carol's lemon-scented chicken recipe")
• Ingenue Carol (Flaw: Backstabbing (!); Goal: Spades / "Destroy Barry Bonds' unfairly earned sports trophy")
• Gangster Jenny (Flaw: Backstabbing (also!); Goal: Spades / "Set fire to Gary's bar to falsely implicate Jason in an insurance fraud scheme")

"And, you're all there!" I said.  "You all know each other already, so we can skip the 'strangers meet in a tavern' cliché.  Go at it."

From that cold start, within a minute Jason was trying to sell insurance to everyone; Marty was running madly around the bar trying to both woo Jenny and get Jason to read his manuscript; Jenny was sneaking into the basement for some pyrus interruptus; and Barry, Marty, Jason, and Carol were working out a massive, intricate four-part trade, where Barry would give Carol his trophy for her recipe and give the recipe to Marty, so he could give that to Jason as incentive to read Marty's manuscript and find a buyer for it (Barry).  It was a good plan, and half the table rolled for goals at once. Then, based on the rolls, Jason ended up snatching the recipe and bailing out of the deal, leaving the others in the lurch.

In vengeance, Carol decided to help Jenny with her second attempt at torching the place, and the two snuck off to the shed behind the bar.  But both of their Backstabbing flaws kicked in at once.  Jenny got locked in the shed, where her blaze was happily burning, while Carol's blaze near Barry's trophy case was sabotaged by Jenny's donation of shoddy fire materials.  Jenny's fire was again interrupted by the (now life-saving) meddling of Marty, who convinced Jenny of the error of her ways ... just as the Deathbird arrived.

When the dust (and ashes) cleared, Jenny lay in the road outside the burnt bar, her head caved in by an autographed Barry Bonds baseball bat, as Marty wailed a big NOOOOOOO for their unconsummated love.  Barry defiantly Took a Stand against the police detective, lunging past him to fling Marty's manuscript in the fire as Barry died in a hail of bullets.

It was really encouraging to see this happen in the game's very first test.  When the game is firing on all cylinders, gameplay is amazing.

A note on that first Deathbird scene: amid the flinging, Jenny's player lost one of her d4s.  (She was the only one to lose dice.  She lost a second one later that night, despite being closest to the target.  We found them both before the evening was done, but I learned that I need to add a 15-Second Rule to keep dice searches from halting the game.)  And, somehow, Jason's player - farthest from the target - got both of his d8s in, gaining two points and temporary Deathbird immunity.  This would cement his lead for the rest of the game.

Later scenes

Unfortunately, things were more uneven from there.  Scene two - adding the characters Brad Pitt and Angelina Jolie, and setting the scene on a Hollywood movie set - stalled out.  Since we had made the decision to playtest using only Flaws and not Complications, it was necessary to work your Flaw in before rolling for your Goal; a combination of bad Flaw draws, player uncertainty, and conflicting plans got everyone mired in roleplaying (or simply giving up) rather than rolling for progress.  The GM made the wise decision to mercy-kill the scene via early Deathbird arrival (a feature I'm glad I put in the rules), and said she would have done so earlier if she hadn't felt constrained by the GM guideline to "allow everyone to roll for their Goal before Deathbirding if possible."  I may have to softpedal that a bit more.

My Angelina ended up murdering Carol.  Or, I should say, ended up being blamed for Carol's death after four separate fatal wounds (!!). In the detective scene, I Shifted Blame successfully to Jason (based on an informed metagame decision that his character's shtick was getting stale). 

As it turns out, I was the only person daring enough to attempt Shifting Blame.  I didn't ask the players for feedback there, but I'm worried that Rules As Written may not empower players to try it enough.  (Most of the group are also less experienced gamers than I am, though, so it could be selection bias.)

Scene Three was Brangelina, author Marty (who drew a new Flaw, "Famous", after Jason's death), a paparazzi, and a crooked cop at the Pitt estate.  Somehow we ended up with four people chasing each other in a circle A>B>C>D>A, and the GM gave up and Deathbirded. Again my Angelina got trigger-happy - ending up responsible for a double homicide.  I Took a Stand. (Again, a metagame decision, to get a fresh character.  I had already accomplished my Goal, so it was strategically pointless.)

Scene Four had interesting gameplay -- it was set at a shooting range, and bullets didn't take long to fly -- but an awkward metagame. The GM went full-on straight-faced, holding back laughter as diligently as possible.  Mike got him to crack repeatedly with a recurring out-of-character joke about Volvo 240s, but the rest of us ended up seriously scrambling for that bonus die (and mostly failing).  It drove home the discussion in the Ronnies thread about humor turning gameplay ugly once you start forcing it.  There was at least one point where I cracked up the whole table with a joke, except for the GM - and I admit I felt cheated.

Scene Five had an even weirder vibe.  I had thought GM-4 was an outlier on the don't-laugh scale, but GM-5 broke the scale entirely.  Dude was a solid stone wall.  Not even a chuckle the entire scene.  On the heels of #4, it seriously dampened the game.  He also said virtually nothing as GM - he's one of those strong, silent Gamists who shows up to sessions to mow down opponents with his minmaxed barbarian, and roleplays in as few words as possible. 

Fortunately, we had run out of cards at the beginning of the scene, so the apocalypse was nigh.  I elected to kill my luckless athlete after being attacked with a gun mid-scene ("wait, dude, you're dying from a shot to the foot?"), and everyone else collectively decided that they had been poisoned when they let the dog -- Q decided to play the K9 partner of the crooked cop -- brew the coffee.  So the game's ending was suitably deadly, but kind of anticlimactic.

Interestingly, those two final GMs have not read any of the game discussion here, and there is nothing in RAW to indicate that the GM's goal is to be adversarial ... it was a stance they brought to the table themselves*.  So if there's going to be a laugh-for-bonuses rule, it needs to be explicit about how to make the laughter rule work, and possibly explicitly ban that sort of stonewalling.  I retract my hypothesis that an adversarial GM stance is useful here; clearly there is some other factor in my previous comedic gameplay that has produced such good results with rules on laughter.

--
* I should note they have both played in at least one TOON game with my "GM laughter means it happens" house rule.

[Baxil]

Continued.  (Long post split into two parts)

Reflections

One of the things that distinguishes Deathbird Black's play style is a rotating, high-logistical, low-creative role for the GM. In hindsight, those duties were a hell of a lot more important role than I consciously expected from this game.  Scenes seemed to work out in proportion to the GM's prior experience with gamemastering.  Pacing, participation, and flow are major factors in awesome scenes, and those are things that good GMs constantly deal with.  The GMs that took a more passive approach to their job ended up offloading those duties onto the players, who had more than enough to deal with as they plotted their schemes.

One thing I was surprised to find is that the game was much more fun with high character churn.  (Ron's "the characters are potato chips" comment was spot-on, and I actually quoted it verbatim.)  One scene, maybe two tops, is enough to explore the shallow archetype and flaw that you've been handed -- a lot of the fun comes in jumping into brand new roles and trying to make the shooting stars shine fierce and quick.  One of the players suggested that characters should be recycled as soon as they accomplish a goal; I kinda want to keep the nobody-walks-away-alive feel, but something similar would help the game.

During the game, everyone instinctively worked to spread their goals' "Who" out evenly - which we didn't even notice until midgame, when three Goals got stacked up on one player and we talked it out.  I'm adding in the rule we worked out: everyone determines Who simultaneously, by physically pointing at the player you're targeting, and then fiddling around with the result until everyone's pointing at someone different.  Limiting your Goal target to lower-scoring players ended up as irritating and needless bookkeeping, so I'm going to drop that as well.

Speaking of score, I may want to drop virtually all the rules related to it - what seemed to work best toward my game design goal of "the fun is the point" was to have points be tracked but as close to meaningless as possible. 

Funny thing about that.  Due to an oversight, I didn't even tell the other players the game's scoring rules (points equal votes) until the final scene had ended.  Results were instructive.  There was a collective double-take, a collective "Interesting!", and then light but firm pushback.  The final score was something like 6-3-2-1-1-1, with Mike embarrassingly far in the lead due to his insane dice-throwing luck.  "So Mike doesn't win," it was pointed out, "but he's singlehandedly picking the winner.  That's no better."  The idea of being able to split your votes was floated, and quickly/mercifully shot down. ("If you've got 3 votes and 4 players, you're probably going to split them 1-1-1.  Who are you going to tell that they suck?")  Players liked the idea of calling back to the game's most memorable moments in theory, but the game was so chaotic and fast-moving that in practice nobody could come up with any "best moments" worthy of point award.

I asked, flat-out, "So, if I hadn't mentioned how scoring works, would anyone have even cared about points?" and the answer was a unanimous "No."  Ouch.  So I was right in my aim, but overthinking my method.  Goodbye, voting rule, to be replaced with a simple and profound silence about "points".  Let players assign meaning to score on their own.  I'll keep a callback-to-what-you-liked at the end of the game, but divorced from scoring or winning.

And now the big issue

Goals.  Yikes, Goals.  Character generation on paper is "two cards for character, one card for Goal."  Character generation in practice is "two cards for character, endless discussion for Goal."  Prep for the first scene was longer than the first scene, and it didn't significantly improve by the end of the game.

And that was without Complications.

The thing is, lack of Complications hurt the game too.  At least one scene stalled as everyone tried to work in their Flaws.  Doubling character generation time is not an acceptable tradeoff for fixing this, but the problem that Complications are there to fix needs to be addressed.  Perhaps with a simple shift to "you have to involve one other player in your plot to make a roll"?  But if I do that, then Flaws have no effect in the currency/reward system, and I'm leery of game elements with no reward effect.

Anyway, the most critical work for the next revision is going to be chargen streamlining.  Here's what I'm thinking.

The "characters are potato chips" theory worked out really well.  "Draw a card, here's your archetype.  Draw a card, here's your flaw.  Done."  Virtually every character was playable and fun and cool out of the box, and the two combinations that didn't work were fixed with a redraw.

So I think I need to frontload the Goal work onto me, the designer, rather than pushing all the Goal details into collective player discussion.  Make a new table for Goals; come up with 13 generic noir Things To Want, or 13 variations of Lust/Greed/Vengeance/Destruction.  Print it on the back of the character sheet.  If I can tack down another few elements of the goal rather than leaving it so ambiguous, it will be easier for players to quickly fill in the details.

Anyway!  Game was a great experience and is going to have huge effects on Deathbird Black v.Alpha-3.  Thoughts?  Questions on the things I'm not mentioning?

[Paolo D.]

Hi Baxil,

I'm happy that you were able to playtest it, finally! I'd like to try it too, but first I'll have to find out some friends that are accustomed to this kind of dark-noir humor :-)

About your biggest issue:

The "characters are potato chips" theory worked out really well.  "Draw a card, here's your archetype.  Draw a card, here's your flaw.  Done."  Virtually every character was playable and fun and cool out of the box, and the two combinations that didn't work were fixed with a redraw.

So I think I need to frontload the Goal work onto me, the designer, rather than pushing all the Goal details into collective player discussion.  Make a new table for Goals; come up with 13 generic noir Things To Want, or 13 variations of Lust/Greed/Vengeance/Destruction.  Print it on the back of the character sheet.  If I can tack down another few elements of the goal rather than leaving it so ambiguous, it will be easier for players to quickly fill in the details.

I think that this could work. To master the power of colorful tables is very useful, indeed (most of all at chargen) - they are working great with my Wings of Blood too. :-)

[Ron Edwards]

I love it. I think the most basic question, whether the baseline idea of this game is actually fun and not, for instance, utterly ridiculous, is answered.

I don't think I ever articulated exactly why I find the whole concept so charming. It's the Deathbird, specifically that this entity has no actual fictional presence in the game prior to its arrival, and then, even though characters are in fact stated to take action against it, there is no mention of it in the aftermath and consequences. In other words, conceivably, the murder is in fact committed by the accused character upon the victim, just as the detective claims, and there was no Deathbird at all - just a collective spasm of murderous intent cresting at the same time.

The ambiguity between this interpretation and the more literal notion that a horrendous ... um, bird appeared to attack the characters is fucking genius. It's hilarious and yet somehow makes so much sense as long as the primary concept in mind is that we are spoofing noir and celebrating it at the same time.

1. Let's start with the biggest issue, which I think you may have resolved partly or entirely, but I'll toss my dice at that container anyway:

How about ... the GM sets a single goal as part of the scenario, which everyone wants, announcing it exactly one second before play proper begins?

So every character has an archetype and a flaw, two-card draw, done. And then wham, they all want this goal. Go! It might be especially fun if the goal is something that can be shared, so that alliances might actually be honest, you know, except when they're not.

This may seem simplistic but it also seems practical and fun. If you want a little more flexibility per character, then maybe provide two, but I'd say no more than that.

As for Complications, I think that your "involve someone else" rule is all you need. The color and direction it provides will support the currency in non-quantitative but highly practical ways. It's actually much better for a game with concrete currency mechanics to include some features which merely help the cycles without being pieces of them.

2. I think your conclusions about the voting are sound, and the player who asked, "Who are you going to tell that they suck?", gets my personal high-five. See my comments and links in [Veterans] Ronnies feedback if you're interested.

I have no real view yet about what to do with points at the end of the game. You may be onto something with the idea that they do, well, nothing.

3. Little things ...

i) I'm thinkin' that once your character achieves a goal, and if that character survives the scene, then in later scenes, he or she is restricted to only one die, period. This might be over-correcting what isn't even a problem, but who knows, keep it in mind if the "done" character issue crops up again.

ii) I not likin' the idea of naming/using known characters or real people as characters in the game. It seems lazy to me at best. Apparently you played Angelina. Was this in response to another player naming Brad, in that scene? And I'm really gonna guess here, and ask, is that player also the same one who played Barry Bonds in the previous scene? Annnnnnd ... is that player one of the two final GMs who stonewalled?

That's about it. I am psyched. I also found more stuffed crows with a little searching around, but haven't bought one yet. Soon, when I find the right combo of crow + cute + disturbing.

Best, Ron

[David Shockley]

If the points don't matter, why have them at all?

When I read the rules I had assumed that Shifting Blame would be the default. I think it would be for me, and at least a couple people I've played with in the past.

[Ron Edwards]

Hi David,

There are a number of games for which explicit points and win conditions turn out to be secondary features, reinforcing a more important focus for fun without ultimately being goals by themselves. The Shab al-Hiri Roach is one of them, and so was the brilliant The Great Ork Gods, which is unfortunately not available at present.* In games of this sort, the pseudo-winning is a perfectly useful component of play but it is not the point.

Whether Deathbird Black turns out to be like that, or whether the points and winning do turn out to be important in some way (and thus need some redesigning to be less locked down in the last round), is still an open question, at least for me.

Best, Ron

* Another example which is possibly problematic to discuss on-line is the distinction between certain functional versions of D&D play, one of which strongly emphasizes leveling-up and character survival as winning criteria, and one of which uses these mechanics only as framing devices for other priorities. However, since dysfunctional versions of both ways to play exist and tend to take over conversations about it, I'll keep this as a footnote.

[David Shockley]

In case its not clear, I wasn't asking rhetorically. I'm assuming they serve some sort of purpose here, and I just don't understand it. (I probably made this more confusing by asserting that the 'points don't matter' in my question. But thats because I have Drew Carey from 'Whose Line Is It Anyway?' stuck in my head.)

I think I see your point with D&D, both about the multiple functions of levels, and about the potential to derail the thread

[Moreno R.]

and so was the brilliant The Great Ork Gods, which is unfortunately not available at present.

The latest Beta, as far as I know, can be found att this page:
http://www.greatorkgods.co.uk/

[Baxil]

First of all, a shout out to the Mom's Basement podcast - I'd like to thank them for also playtesting Deathbird Black, and discussing it in Episode 24 (along with some thoughts on playtesting in general, and the controversy over Secret Lives of Serial Killers).

Paolo, you'll be interested that they recommended expanding the character tables (to 26 or 52, instead of 13).  This seems like a good idea to me.

David:

If the points don't matter, why have them at all?

Well ... points do matter.  I said I want them to be meaningless.  I hope this doesn't sound like semantic quibbling.  Let me take a stab at explaining it.

Essentially, what I'm trying to do is to pull a fast one on Deathbird Black players to guide them into playing the game in a fun way.  Everyone's familiar with the idea of a "score", and I'm using "points" in an intuitive sense: you earn them by succeeding at things (Goals, dice-flinging, lying to the detective).  The gamer brain is hard-wired to look at a mechanic like that and assume that's the game's big reward cycle.  So what are you going to do?  Accomplish Goals, fling dice, and lie to the detective.  As you do this, there are other mechanics that push you toward the "have fun during gameplay" reward cycle: make the GM laugh for bonus dice, etc.  But your focus is going to be on things that move the game forward.

In that sense my scoring is the reverse of Marks from Diary of a Skull Soldier, I think.  Rather than assigning values to gameplay actions to get people to reflect on the numbers, I want to use people's intuitive assumptions of the numbers to push gameplay actions.

But why have points at all if the real reward is the comedy?  Because I don't think the mechanics can directly push toward that reward.   In both this thread and the Ronnies thread, it has repeatedly come up that forcing comedy is awkward and unfunny.  Comedy is what happens along the way toward something else, and the score cycle is the "something else" I've chosen.  (This is also why I don't want laughter to directly award points.  That makes it harder for point whoring to foul up the real reward cycle.)

... That's my theory, anyway.

When I read the rules I had assumed that Shifting Blame would be the default. I think it would be for me, and at least a couple people I've played with in the past.

I was rather surprised that Shifting Blame was not more popular.  I think it was due to the relative inexperience of my playtest group; I'll keep an eye on it as other people and groups play.

[Baxil]

Ron,

Thanks for the comments on the Deathbird - it's interesting how a force of nature like that lends itself to interpretation.    I think it helps the cathartic nature of the scenes to have the Deathbird be so narratively vague.

1. Let's start with the biggest issue, which I think you may have resolved partly or entirely, but I'll toss my dice at that container anyway:

How about ... the GM sets a single goal as part of the scenario, which everyone wants, announcing it exactly one second before play proper begins?

Honestly, I'm not seeing it ... a noir game where everyone's cooperating is falling down somewhere, and turning up the competition with a shared goal means getting truly vicious about points, which I don't think works here (see previous post).  The chaotic collision of individual goals is working thematically, and is okay (if shaky) on the practical level; let me see how the revisions work out before even considering anything this drastic.

I think I see what you're saying with Complications and Flaws being able to help the reward cycles.  If the rule is "involve someone else," you have one or two spiffy character elements that suggest a way to do that, but don't constrain.  I'll give that a shot.

3. Little things ...

i) I'm thinkin' that once your character achieves a goal, and if that character survives the scene, then in later scenes, he or she is restricted to only one die, period. This might be over-correcting what isn't even a problem, but who knows, keep it in mind if the "done" character issue crops up again.

ii) I not likin' the idea of naming/using known characters or real people as characters in the game. It seems lazy to me at best. Apparently you played Angelina. Was this in response to another player naming Brad, in that scene? And I'm really gonna guess here, and ask, is that player also the same one who played Barry Bonds in the previous scene? Annnnnnd ... is that player one of the two final GMs who stonewalled?

i) Noted for later, I'm not at the point where I'm tuning that yet.

ii) I did play Angelina in response to Brad, after drawing the "Spouse of another PC" card.  Low-hanging fruit.  And, man, your guesses were 3 for 3 ... gold star. (I'm going to guess, for my own part, you've dealt with that sort of neo-grognard in game demos and such?)

In this case I agree with you about the laziness ... but, in the general case, I think there's a strong case to be made for that sort of theft in the comedy genre.  In our last game of Toon, the Minecraft creeper and the Old Spice Guy stole the show; in the past, one of my PCs has had some crazy memorable moments playing the character of Fighter from the webcomic 8-Bit Theater.

I think what makes it laziness here is that you already have a character ... Deathbird Black specifically hands you an archetype, and by grounding that in a specific known person, you're not letting yourself explore it and own it.  Toon doesn't suffer that grounding problem because the game is so nonsensical in the first place.

[Ron Edwards]

To clarify my suggestion about the shared Goal - the idea was not that everyone can cooperate to get it, but rather than everyone is stepping on everyone else to get it. The idea that it could be shared makes alliances possible, but the notion that everyone (or anyone) would share it did not occur to me, and seems very unlikely. Although if you wanted, tontine logic (its value is higher the fewer people have it) might be applied.

I'm not posting to defend or push the suggestion, merely to clarify it.

Best, Ron