Powered by the Apocalypse, Part 7: Q&A Round 4

Using Apocalypse World to Outline and Draft Your Own RPG

Here in Part 7, I’m setting aside some time for a good old straightforward Q&A. This is round 4, the final round!

This article was made possible by the generous support of my patrons. Please consider joining them by supporting me on Patreon.


I love a PbtA game that gives a lot of advice, principles, & moves for running the game as a GM. What advice, principles, or moves do you think help a GM new to running PbtA games understand how to run them effectively?


Good question! We’re each faced with it, every time we make a game.

One of the complicating factors is that it always varies between individual games. There’s no such thing as general PbtA advice, principles, and moves — there will always be individual games that do, and should, contradict each other.

Take Apocalypse World’s principle to “look through crosshairs.” It’s great for Apocalypse World, it’s exactly what Apocalypse World requires, but you wouldn’t include it in other games without thinking it through and making careful choices. In a lot of games’ settings, individual characters can and should be safe from violence, and institutions can and should be stable, lasting and resilient.

Anyhow, in practical terms, I think the answer for most PbtA games will turn out to be a mix: some advice, principles, and moves that could apply to a lot of games generally, mixed with some that you’ve invented and formulated specifically to meet the needs of the precise game you’re writing.

The more closely the game follows PbtA conventions*, the more like other PbtA games’ its advice, principles, and moves can be, but no game follows PbtA conventions so closely that it won’t include its own unique take.

From a GM’s point of view, my recommendation would be to read each game’s GM section carefully, looking for the ways it’s different from the PbtA games you’ve played before.

* Whatever they are at the time. They’re ever-changing; every new PbtA game contributes to them and has the potential to change them radically.


On a recent episode of +1 Forward, a guest talked about the way AW has a potentially very large span of scope – from a community scale (Hardholder) to personal (Driver etc). Many later games has a more uniform scope. What are your thoughts on how PbtA handles scale?


One of our design specs from the beginning was to have the game create characters whose urgent concerns, just as you say, covered a range of scales. Immediate, short term, long term, into the past and into the future; personal, interpersonal, communal.

I think this is a genre thing more than anything. It’s not unique, but the post-apocalypse really calls for one character who’s thinking about the material needs of a community over time, contrasted with another character who’s concerned about whether there’s an ambush on the way to their next meal, with a third who’s grappling with how we got here and how we can change it. Other genres have other needs so it’s not a surprise that other games don’t duplicate Apocalypse World this way.

Anyway, if you want to make a game with characters who operate at different scales like this, you need to be thoughtful and intentional with your underlying model. You need to have ways for the characters’ contributions to build on each other, not sweep each other aside. You also need moves that switch and refocus between the scales, the way that the Hardholder’s Wealth move pulls your attention up and out, but then reading the situation draws it right back in.


I’d be very interested to hear you talk about playbook design. Like, what does a good PB include? How do you come up with a good core for a PB? How do you spin that out into a complement of moves?


You probably already read what I wrote about the theory of playbooks back in Part 4. If you haven’t, the summary is: a character in a ttrpg isn’t just a fictional character or just a playing piece, it’s both. A playbook gives you both your character’s identity and trajectory as a character, and your own position and resources as a player. It answers, for this one character, the three game questions: What do you establish in order to begin play? What do you hold constant during play and what do you change? What do you play to find out?

I also enthusiastically recommend Jay Dragon’s article about writing playbooks: Writing Playbooks: An Approach.

Your questions here are practical though, so I’ll tell you about my own process, and maybe it’ll be useful.

First: I’m constantly brainstorming lists of playbooks. It’s one of the first things I do when I start working on a game, and I keep doing it until long after the game’s finished. I find it enjoyable, so sometimes I’ll even take a break from harder design work just to brainstorm more playbooks for fun.

I don’t try to judge or guess which playbook ideas are good and which are bad. I let natural selection do it: the ones I finish, must have been the good ones. The ones I don’t finish, there must have been some incompleteness, incompatibility, or redundancy in them that kept me from finishing them. In retrospect, my analysis of my playbooks is kind of like Jay’s: one to establish the standard, one to break it, one to expand the line into a plane, one to fill out the quadrants, one to expand the plane into a 3-dimensional space…

When I sit down to work on a playbook, I can choose whichever one speaks to me from my current unfinished list. While I’m working on playbooks, I’ve also been drafting basic moves, other systems, game text, or whatever, so I’m usually not choosing blind or at random, I’m usually following some inspiration or other. I’ve been working on the system for stepping from summer to winter, for instance, and it gave me an idea for the Lostling, so the Lostling is the playbook I choose. Or I happen to have watched Romeo + Juliet with our 9th grader, I’ve been thinking about Tybalt as played by John Leguizamo, so I pick up the Cock’s-spur to work on.

Once I’ve made a playbook or two, they constitute a template that I can follow (or break) for future playbooks. It’s also crucial to get the first couple of playbooks laid out on paper as prototypes, as soon as you can, so you can see how much space you’ll need and how much space you have. That sheet of paper is a fundamental creative constraint, no less than the genre of your game or what you’re trying to accomplish with it. (Meg says: If I’m designing a playbook, and it takes up radically more or less space on the prototype page than the others, that’s an indication that either it’s very different in scope and I’m trying to load too much into it on one hand, or it’s really just a facet of another playbook and I should see how i can roll it into an existing playbook on the other.)

As far as developing the core idea of a playbook into its moves and other stuff goes, I do it the same way, actually: I brainstorm a whole list of moves by name, usually more than I’ll need, and develop the ones that speak to me. I list out all the other stuff the playbook might have, and gradually cut and develop it down to what it does have.

I don’t usually finish one playbook before starting the next. I’m usually actively working on, I dunno, 3-6 at a time. Sometimes an early playbook will hang out with an unfinished move for months, waiting for the insight that a later playbook will give me. Even the ones I’ve finished, I don’t finalize until I’ve finished them all. You never know when one playbook is going to upend your whole scheme.

So like, that’s how I approach it. It kind of boils down to “you have an idea, so you develop it and iterate it into a finished playbook, I don’t know how. However it works for you. Good luck!” But maybe there are some useful tidbits in there.


I’m wondering what you think of the power dynamics in PBTA games are and how they’re handled differently than others, like d&d. How can I as a gm help players feel in control of their characters and what options do PBTA provide for facilitating it?



When I sit down to play Apocalypse World or Under Hollow Hills with new players, I always take a few minutes to go over the basic moves. I like to have a player volunteer to read the basic moves’ names out, while I give a quick summary of when and how to use each one.

During play, I never outright tell a play what move to make. Even when I think it’s obvious, I always offer it as a choice, by name. “Hey, take a look at Go Aggro on the basic moves sheet. Is that what you’re doing? Do you want to roll it?”

The players’ moves are there specifically to give the players informed, explicit, reliable control over what their characters do and what comes of it. My goal with new players is to get them familiar with their moves and using them explicitly by name.

(This contradicts a piece of conventional PbtA wisdom, which is that the players don’t need to know their moves, they should just say what their characters do and let the GM tell them what to roll. I don’t subscribe to this idea at all.)


What’s the one thing you wish you knew before you started getting people to playtest your work?


Hm, I’m not sure. There are several things that I take pains to find out before I ask people to playtest a game. I don’t think I can narrow it down to one.

So let me lay out my playtesting timeline. Each of us here approaches it a little differently, so I can mainly speak just for myself, but playtesting time is precious. You don’t want to squander it, and you don’t want the playtesters to feel like they’ve squandered it either. You want to keep your demands small and focused.

  1. As I work out the game’s initial design, I pull people in for little proof-of-concept and mechanical tests. Just the people in the house, Meg or the kids, for like 5 minutes at a time, like “hey, would you roll these dice and make this choice? I want to see something.”
  2. Once I have a “first look” or “sneak peek” document, I release it to my patrons. I don’t ask them to playtest it, and in fact at this stage it’s probably not possible to playtest. What I’m looking for is how people react to the text.
    If I ask for feedback, it’s like this: “What questions would you need answers to, before you’d play?”
  3. Next I release a “playable draft,” but still without calling for playtesting. At this point I’ve seen people respond to the initial text, I’ve revised it thoughtfully, I have informed guesses about how it’s going to go.
    I usually say something like this: “If you feel like giving it a try, go ahead! If you have any questions first, or if any questions come up in play, ask them, I’d love to answer.”
  4. I always play the game in-house before putting out a call. Always, no matter what! Your first playtester has to be you.
  5. By the time I release a “playtesting document” and actually ask people to play, I’ve already answered a lot of questions about the game, I’ve built up a following who are eager to play, and I’ve revised the text several times. Usually, a group or three have played on their own initiative and I’ve incorporated their feedback, and I’ve certainly played it myself.
    At this point, I already know how the game plays, I know what the text is communicating, and so I have a good sense of how playtesting is going to go. There are always surprises, but I take it as my job to catch problems as early as I can, when they’ll affect the fewest possible players.

My point is that, at every stage, I try to playtest only what I need to test, and I try to know everything I can possibly know about the game and its audience first. There’s not just one thing, the more you know before you put a game through playtesting, the more you can find out beforehand, the better!

Meg says: Additionally, we’ve spent 12 years cultivating a group of players, made up mostly of our kids and their friends, who play games every week at our house and who we ask to play our games in development, as soon as there’s a “playtest document”, and then in several iterations as the game develops. We view this as a serious real-time stress-test of our work, and the kids now use it to test their own designs as well. If there are things that bog down play, or changes we can make and see the results in the following session, that’s invaluable. They do occasionally make jokes at our expense when there’s an entirely new system this week than last week, and they need to fill out a new character sheet yet again.


When do you think it’s best to use abstracted resources (Hx, coin, strings) vs descriptive mechanics (writing down traits, prompts and lists, fictional triggers).

While resources are nice and thematic, they’re oft forgotten due to their degree of separation and lack of immediacy. But descriptive mechanics can slow down play with interpretation and constant writing & rewriting. Do you consider this with your mechanics?


Oh, for sure, yes! How I think about it is, abstract resources are there to be intermediaries between descriptive mechanics. Whenever your descriptive mechanics can’t interface directly or feed directly into one another, that’s when you need an abstract resource to span the gap.

In The Wizard’s Grimoire for instance, the questions you answer during character creation are a descriptive mechanic, natural language with a direct fictional choice to make. Then during play, when you choose to exert yourself, the exertion you choose is a descriptive mechanic as well — are you exerting yourself physically, or violently, or magically, or which?

But the questions you answered about who you are, and the choices you make when you exert yourself, are removed from each other in time and scope. My solution to their disconnect is to tally your answers and hold them as an abstract resource, your qualities, to call on when you exert yourself.

Or in Apocalypse World, your experience points come from a variety of actions, both descriptive and abstract — you can think of them as gathering together the threads of play. The abstract experience system takes these disparate events and feeds them again into (sometimes) descriptive mechanics: now you have a holding of your own, now you’ve expanded the range of actions you can take by adding a new descriptive move.

Or take Read a Situation. You get +1 to your rolls when you act on the answers to your questions, an abstract resource that carries the consequence of descriptive mechanics — the answers to your questions! — forward through play.

In sum: when cause and effect are immediate, close together in time and reliable in scope, descriptive mechanics are probably the best. The bigger the gap between cause and effect, in time or in scope, the more strongly to consider bridging it with an abstract resource.


AW is characteristic (to my mind at least) by having the players neither mostly aligned or mostly opposed – in fact, it seems to be characteristic for how players are both and neither. Is there something about PbtA that is well suited for that kind of play?



In Apocalypse World, the characters start play broadly aligned, but not closely aligned. Their different concerns — especially, their concerns at different scales, different scopes — all but force them to take conflicting views of events in play. The game’s conflict model places them all in potential conflict with one another, it doesn’t make an exception for them.

But then, the game’s moves, experience system, and other systems want the players’ characters to come together. They channel the PCs toward agreement, not toward violence.

The effect is that the characters are at odds, but working together, but conflicting, but aligned. Each event in play has the potential to unsettle their alliance, but once unsettled, the game tries to draw them back to the table to ally again.

So I don’t know about PbtA generally, but yes. It’s designed on purpose into Apocalypse World, it’s not surprising that those PbtA games that follow Apocalypse World more closely, picked it up as well.


Fiction First and triggering the moves from the fiction sometimes is harder to achieve with the “questions/answers” moves, how does it work when some questions don’t make sense with what the character should be able to know?


I confess, I practically never think about what the character should be able to know. My goal when I write those moves is to give the player the information they need in order to make good decisions in play. They can choose for themselves – by choosing which questions to ask, and by interpreting for themselves how their character knows or guesses the information – whether to limit themselves to in-character knowledge, if they even care.

That said, I’ve lately been adding a bit of retroactive justification to my new versions of those moves. Here’s read a situation from The Unnamed World:

When you take a second to read a charged situation, roll your Cunning. On any hit, ask up to 3 of the following questions. The MC or the other player should answer honestly.
• What’s about to happen here? How can I tell?
• What do my instincts tell me?
• What’s my best way in, out, around, over, or through? How can I tell?
• Who here can I count on? How do I know?
• What is [x] in a position to do?
• I want [x] to happen. What might I do to make it go that way?
• A question of your own. If the MC or the other player answers it, it stands. Otherwise, go back and choose 1 of the above.

On a 10+ hit, you hold the initiative: what do you do now? On a 7–9 hit, pass the initiative to someone else.
On a miss, ask the MC what goes wrong instead.

The Unnamed World

The “how can I tell?” and “how do I know?” clauses aren’t there to limit the player to the character’s knowledge, they reinforce the idea that the character, in fact, already knows more about the situation than the player does.


How would you define the distinction between the GM’s agendas, rules, and principles?

Amelia Kurtz

Your agenda is a list of things that you’re always trying to do, across every moment of play. You shouldn’t prioritize them against each other, they’re all possible all the time.

Your rules — in Apocalypse World, these appear under “always say” — are still things you should always do, but they might not equally apply to every circumstance. If you haven’t prepped something, for instance, you can’t say what your prep demands.

Your principles are different ways you can follow the rules to achieve your agenda. They’re, like, ways to look at it, useful patterns to follow, best practices, pointers for when you need a nudge.

The agenda and principles are things that the game expects you as GM to remember and internalize, so that you can play the game correctly. The principles are different ways to express and pursue your agenda, according to the rules.

That’s what I’d say!


In your last Q&A you wrote “I think that going back and applying PbtA’s insights to play centered on the players’ imaginations, over their conversation, could produce some really breathtaking games”. And I think that’s a very interesting point to explore.

I’m starting to dig in to this idea, but I’m wondering if in the next episode you could give some example to explore better your vision about that. Thank you!


Yeah! I can give some examples from games I’ve already written that kind of point the way. I don’t have any worked examples yet, but I think these are suggestive.

Take the questions you ask yourself during character creation in The Wizard’s Grimoire:

Suppose that you face a dangerous obstacle. Will you confront it readily, or take pause to consider your possible courses? If the former, tally 1 to bold. If the latter, tally 1 to cunning.

Suppose that your chance companion departs the table in the precise moments before the proprietor arrives to settle. Will you pay your companion’s share? If so, tally 1 to virtuous. If never, tally 1 to ambitious.

The Wizard’s Grimoire

…And so on.

The game poses you these hypotheticals and has you imagine your brief way through them, but you never communicate them to another player.

Or, take this page from Murderous Ghosts:

What do you most hope it doesn’t do? Choose by vote:
• Look at me.
• Reach out to me.
• Speak.
• Stop still.
Any of you can ask the MC questions before you choose.
13–20: It does none of them. Tell the MC to turn to 13 in her book.
6–12: it does one of the others. Choose by vote and tell the MC which. Tell the MC to turn to 15 in her book and to take Branch (b).
Bust: It does the one you hoped it wouldn’t. Tell the MC which. Tell the MC to turn to 15 in her book and to take Branch (b).

Murderous Ghosts

The game has you briefly imagine each of the things it might do, so that you can judge which one you hope it doesn’t do. None of them come true until you’ve decided and made your draw.

Or even this play of the Crooked Wand’s in Under Hollow Hills:

When you lay someone bare, roll. On any hit, you reveal something to them, about them. On a 10+ hit, choose 2. On a 7–9 hit, choose 1.
• You reveal to them their secret heart.
• You reveal to them the hearts and natures of those nearest to them.
• You reveal to them their best way forward.
• You reveal to them their lost past.

Whichever you choose, you don’t know what they learn, but ask them how they react to the revelation. However, if you have a glass, a basin, or an oracle to read, you can catch a shadow or a mirror image of their revelation. Ask them what you see.
On a miss, choose 1 anyway, but in the act, you lay yourself bare as well. The MC chooses 1 against you, and asks you how you react in turn.

Under Hollow Hills: the Crooked Wand

When you make this play, you don’t get direct information yourself. What happens is, your subject imagines what you reveal to them, and tells you how they react. It treats what they imagine as consequential, as real, without first having them communicate it so that everyone can agree to its reality.

These games overall, they don’t generally treat what one player’s imagining as the reality of the game, they generally treat what the players have communicated with each other and collectively agreed to as the reality of the game. Still, here are little moments in each when what a player imagines, privately, makes all the difference.


So what is the design legacy that we have today thanks to the Pbta, what are its heirs and what lessons have we learned from it?


I would absolutely love to punt this question to my rpg historian friends. I have my take, but it’s really not my field.

My take is: people have been predicting the end of the PbtA boom, looking for “the next PbtA,” anticipating the passage of PbtA from currency into history, since 2011. It hasn’t happened so far. I don’t even think we’re particularly close to it. The ideas keep spreading further into the mainstream, keep reaching new audiences, and keep provoking new designs.

PbtA is still actively creating its legacy. I think it’s too soon to say what it’ll be.

The End of the Q&A

But if you have any more questions, please ask! It’ll be my pleasure to answer.

Thanks, everyone!

Past Installments:

  • Back in Part 1, I laid out Apocalypse World’s philosophy and foundation, described the fit and purpose of its systems, and talked about which features are central to its workings and which aren’t.
  • Then in Part 2, I walked through the beginnings of taking Apocalypse World’s parts and using them as the basis for a whole new game.
  • In Part 3, I dived back into Apocalypse World’s basic moves. I went through them one by one to talk about how and why they work the way they do.
  • In Part 4, I talked about playbooks, by request. What are they, do you want them in your game, and what are the alternatives?
  • In Part 5, I took a quick aside to talk about some different ways that moves can fit into the conversation of play.
  • In Part 6, I used an Ursula K. LeGuin quote — you probably already know the one! — as an outline for alternative models to Apocalypse World’s model of conflict.
  • Part 7: Q&A Round 1. Q&A Round 2. Q&A Round 3: the Lightning Round!

Next Installments:

  • In Part 8, upcoming, I think let’s take a nice close look at our second-best selling PbtA game, MF0: Firebrands.

Reminder: The Goal is to Create a Playable Outline

Goal: Create a Playable Outline
Cycle: Outline, Play & Revise, Full Draft
* Game design means iteration.
* The first step is to create something you can try.
* Is PbtA your final goal? Could be, could be not.


He / him.

Leave a Reply

Your email address will not be published. Required fields are marked *