Running a PbtA game versus a more traditional game?

I have tried on and off to run various powered by the apocalypse games. And it’s something I’d like to do more of in the future. I mostly play OSR games and find I often approach running PbtA games with that mindset and it doesn’t quite work. Or it feels like I am doing something backwards.

I was wondering if people thought this characterization of the difference in how the games should be run is apt:

OSR: you are in this situation and you do something clever, so don’t roll the dice. Good job!
PbtA: you are in this situation (the game cares about) and you rolled well so you did something clever—what was it? Good job!

And, I guess some corollary to this would be clever play in PbtA games would be about managing when you do or don’t trigger moves in the game.

Or should I be thinking about this is a totally different way? (!!)


Also there is no OSR tag. Come on, people!


That is a really cool distinction that I think must work for some games, but it’s not exactly how I’ve run PbtA games before. I think it’s rare for a move to retroactively have you describe the past after you roll, but not unheard of; @Jeremy_Strandberg’s scouting move comes to mind (you went scouting, did you get back?), and I have to imagine there are some flashback moves out there.

A more obvious difference to me between OSR and PbtA GMing is the implied rigidity of GM processes. In D&D, we’ve always been told you should do your own thing, and what the DM says goes. In Apocalypse World, we’re told that you can make custom moves that change the rules, but you’re still operating within the universal move paradigm: “to do it, do it.” As a player, moves are triggered (whether you like it or not) by what you describe, and if you want to trigger a move, you have to describe it. And the GM’s responses are even understood as moves, every one intended to move action forward without ever saying “nothing happens.” (Though that’s typically the first thing GMs ignore, instructions be dammed. It’s hard to keep track of all those moves, and I think many PbtA GMs just think of the as a reminder of best practices.)

That might sound like a nit picky distinction, but the key takeaway to my mind is that both OSR games and PbtA games leave room for, “You had a clever idea, so you don’t even need to roll the dice. Good job!” In Apocalypse World hacks and the like, though, it’s more codified. Like, in Dungeon World, you know that if you “hack and slash” in melee range, you’re likely to take damage, so you set up a trap to trigger from afar. No specific move is triggered, so the GM responds with one of their open ended moves. If it was a cruddy plan, I’d “announce off screen badness” as they hear the trap fail. And if it was a clever plan, I’d pick “offer an opportunity, with or without a cost,” and pick “without.” The trap worked and they’re in disarray! What do you do?


All of that said, the way I run OSR games and the way I run PbtA games has converged quite a bit. Chris McDowall’s procedure for play is, to my mind, basically an OSRish reframing of the most common, useful, and universal moves: offer an opportunity, with or without a cost, and trigger the combat move (attack) or the act under fire move (save) when necessary. The rest of the PbtA move structure is basically there to help guide play for groups that want more help or a more tightly defined experience, I think.


To me, PbtA moves manage who at the table has narrative control. I think players can and should absolutely find clever ways to roleplay to get around moves, but, when a move triggers and a player gets a 10+, they’re not simply dictating what their character did, but also what is happening in the story. They get to move from an agent (affecting the story) to an author (writing the story). A 7-9 gives them some authorial control, but the GM retains some as well, whereas a 6- keeps the player as an agent — even when the GM makes a hard move on a 6-, the player character still reacts as an agent in the story. The moves are only ever there to identify dramatic points and negotiate narrative control.
In OSR, to my mind, player characters are never more than agents and the GM retains authorship.


A more obvious difference to me between OSR and PbtA GMing is the implied rigidity of GM processes.

Which is a really interesting divergence between the OSR and the original games which inspired the OSR.

If you look at the original 1974 edition of D&D, for example, it is defined by rigid procedures. Running OD&D by the book is very similar to Apocalypse World in this regard.

What’s interesting is the degree to which elements in both communities have actually moved away from this: OSR retro-clones have largely replicated the character creation and resolution mechanics of the games they mimic while failing to replicate the procedural mechanics. Any number of PbtA games have been produced which have genericized the Moves of AW to a point where it’s conceptual mush and largely defaults back to “whatever the GM says”.


I don’t find this breakdown satisfying. In fact, it doesn’t align with my understanding of either type of game. The final statement about being clever in PbtA games is much closer to my view on how these games work, and largely at odds with the previous two statements.

I presume that in most OSR games, just because you are “Clever” doesn’t mean you don’t have to roll – something can be clever while ALSO being difficult. Therefore, “clever” is not the only relevant factor here.

The description of PbtA meanwhile, is backwards – you have to explain what you are doing before you are even allowed to roll, and, much like the final statement said, there are many situations in which you can be clever and therefore avoid rolling.

Fundamentally, to me, the BIG difference between the two games is when the dice hit the table: In the OSR, it’s when you’re doing something hard or dangerous (and, presumably, have not taken enough precautions that it’s no longer hard or dangerous). In PbtA games, the dice hit the table when you are doing something that is dramatically interesting (Each game defines this differently, but provides a handy list of what kind of actions are, in fact, dramatically interesting), and where the outcome is in doubt. So in both cases, there are lots of situations where the dice don’t come out because you’ve worked your way around it.

It’s also true that unlike OSR play, PbtA play is generally not concerned with “challenging the players”. But that’s less of a process “how do I run this game?” issue and more of a higher level “Why might I choose this game?” issue.


I think my including “what was it” muddies what I was trying to say. I just wanted to highlight that the decision about clever or not seems like some that happens post hoc. (In a reply on twitter @Justin_Alexander pointed out this happens with D&D as well with things like natural 20s and the like.)

I have a bad habit with PbtA games where I will ignore triggers because in my head I am all, “good job you do the thing.” And later I’ll realise I should have called for a roll. I am trying to figure out what a simple way to think about the difference in GMing approaches might be, because it does feel different. (Though my framing it as PbtA vs. Trad might be wrong.)


“Good job, you do the thing” is often a completely appropriate response. Moves are only rolled when there is uncertainty, generally – all the usual advice that gets handed out about “when to call for a roll” is, IMHO, still correct in PbtA. You don’t call for rolls when the outcome isn’t in doubt, or when some outcomes are “unacceptable.” As a result, I don’t think you can only “do something clever” in a PbtA game after the dice are rolled.

Which PbtA games are you running? I often think it’s sortof a mistake to discuss PbtA games “in general” as opposed to individually, because there are often substantial differences in how things should be approached.


Zombie World has been the one I’ve been playing the most. But I wanted to try and play Cartel now that there is more material out for it to backers.

1 Like

It’s also worth noting (with an awkward shrug) that PbtA move triggers can sometimes be extremely vague, which gives you a lot of leeway in calling whether a move gets triggered or not. The original Apocalypse World has a move with the trigger, “When you go aggro…”, but also has an example in the text in which a PC flat-out kills an NPC without rolling because they were surprised and defenseless. Wait, wasn’t that going aggro? Or seizing [the secure location that poor bastard previously occupied] by force? Apparently not! I think there’s a lot of wiggle room.


Anyone with experience with Zombie World want to weigh in? It’s outside my bailiwick in more ways than one.

I think there are obviously a lot of different things that PbtA is trying to accomplish through its conflict resolution system (aka rolling) depending on the system. As someone who runs a lot of PbtA, the thing that has the most impact on how I run games is that in PbtA, miss, success at a cost, full success isn’t a measure of whether the character succeeds at their action, but whether the player succeeds at their goal.

This is distinct from what I sometimes see in other systems, where for example:

  • a player tries try to break the lock on a chest with their sword.
  • they make an attack roll, and get a critical hit.
  • the GM tells them that they hit the lock so hard that they smash the chest and destroy its contents.

That scenario would be very confusing in PbtA, because the player’s goal isn’t to break the lock per se, their goal is access the contents of the chest. Therefore, smashing the chest’s contents would be an appropriate outcome for a miss, or even a success at a cost (“you smash the chest open, but you broke something. Here’s what’s in the chest, choose what you broke”), but definitely not a 10+.


I mean, I also emailed Mark from Magpie. I thought this would also make for a good topic of discussion, though.

I think @Airk pointing at uncertainty is really key here. Forgive me, but… all of this makes me think of some pretty 50,000 foot RPG theory! Let me see if I can be succinct about it.

One of the big questions for me as a designer is “why do we use dice/cards/randomizers to tell these stories?” In my experience, we invoke these randomizers to declaim responsibility about outcomes. If your paladin tries to kill an orc, we can say:

  • sure, you do it.
  • no, you don’t do it
  • let’s roll some dice!

The outcome is predicated on the group’s understanding of the uncertainty. You don’t roll dice to tie your shoes (sure, you do it) or to travel through time (no, you don’t do it), but there are plenty of situations in which the game is stronger because we don’t already have a clear sense of the outcome when the action is taken. That’s the secret sauce of RPGs imo: we are telling a story we do not yet know.

So the real question for me in any system I’m working in/running is “What things require me (the GM) to declaim responsibility for the outcome? When can I say ‘yes/no’ and when do I have to invoke the mechanics of the game to see how things play out?”

In OSR games, the task-based nature of the game tends to create situations of uncertainty around accomplishment, i.e. we roll dice to see whether or not you can do the thing. But the GM has to have some limits on that declamation. If a PC tries to throw a rock into space, then the GM has the authority to say “I’m not declaiming responsibility for this outcome; you can’t do it.” Thus, if the PCs find a way in the fiction to kill a monster or overcome a trap in a way that satisfies the fictional position of the situation, then there is no roll because there is no uncertainty.

I think that all scans for most folks even if they are new to the concept of uncertainty and declaiming responsibility for outcomes. What’s strange for most players is how PbtA games handle uncertainty, a concept made more opaque by many published games that fail to use it wisely.

In PbtA games, we still only roll when the outcome is uncertain. If you shoot an incapacitated dude in the head at point blank range in Apocalypse World (1e), then that dude is dead. There’s no roll, no seizing by force, no _going aggro. If I’m the GM, I look at those outcomes (harm, impress/frighten/etc) and I think “All of this is known. This guy dies. The cost is a bullet.” But if that guy is standing up and shooting at you, then all those stakes suddenly spring to life and the outcome is uncertain enough to merit a roll.

But PbtA is built on a framework of moves instead of tasks. Those moves don’t describe your ability to do or not do a thing—although some designers make the mistake of thinking that’s what moves do—as much as they highlight a point of categorical uncertainty. Cartel, for example, says that the stakes of shooting someone invoke harm, sure, but the moves around violence also invoke collateral damage and resisting stress. In other words, the violence in Cartel is always uncertain because the stakes for violence are about what happens to everyone around the violence when violence is done.

So in many PbtA games, broad triggers (@JasonT) require judgement calls. Are you fleeing from the dead if you get in your car and drive away from a swarm of zombies? Or does it only trigger when you try to get away from them on foot? That’s up to the GM (and group) to adjudicate, but as long as the conversation relies on uncertainty—“Yeah, I think there’s no roll here because all of the stakes are about failing to flee the zombies and I think you’re fine in the car”—then you’re running the game correctly.

So in my mind:

OSR: you are in this situation and you do something clever, so don’t roll the dice. Good job!
PbtA: you are in a situation that the game has hard-coded as “uncertain,” so we will resolve the outcome of that uncertainty in the way the move has described. If you somehow removed the uncertainty, you wouldn’t have to roll!

So in essence, I think you’re right that “cleverness in PbtA is a function of triggering or not triggering moves” because that’s where the tactical play lies! If you don’t want to get in someone’s face in ZW because you’ve got a low Savagery, then… don’t trigger the move! Find a different move to trigger to get what you want with a stronger roll… or find a way to avoid moves entirely and resolve it without any uncertainty.


There are some great answers here. I agree that lumping all PbtA games together is likely dangerous or confusing (because there are so many now!), and I’m not familiar with Zombie World, but I’ll say the following:

I can speak for Apocalypse World and its cousins by designers close enough to the source (e.g. Monsterhearts, Sagas of the Icelanders, Masks). Those are intended to be played in an absolutely fundamentally different way from OSR adventure games. They have a different creative goal.

(This is muddled a great detail by Dungeon World and its descendants, which do not have a clear creative direction or goal in the same way.)

There are enough common procedures between OSR games and PbtA games that it’s easy to get confused, but I wouldn’t want to play in either style with the mechanics and procedures of the other.

@funkaoshi, let’s get together and chat over coffee or beer, or play one together. I’ve got a lot more I could say in person, in interactive dialogue, and we don’t live too far apart!

In my opinion:

Many of the rules and procedures in OSR style “adventure games” are designed to create repeatable and reliable procedures for the fair adjudication of particular actions and events. They are usually symmetrical, which helps with impartial resolution. Impartial resolution can be used for creating and addressing challenge, and works really well when you use it right. A turn sequence and wandering monster checks, for example, make spending time in a dungeon a risky and interesting game proposition, requiring players to make tough choices. This is enabled by following the procedures strictly and consistently.

In PbtA designs, almost all the procedures are fundamentally asymmetrical: they are not designed to allow for impartial resolution, but rather to move around spotlight and to focus on certain characters, moments, and themes. They are about shining a light on a particular character, a particular moment, and to highlight a certain theme. That’s what good players/groups do in order to make them work.

There is almost no guarantee (or need!) that two similar situations will be resolved the same way, and that’s by design: the system allows us to express ourselves and the themes of the game in play in a flexible way.

There are some games that reliably get at their themes simply by iterating the procedures (e.g. Dogs in the Vineyard), but, in my opinion, PbtA games are not like that. They are similar enough to “traditional” RPGs that it’s really easy to play them without getting the desired playstyle.

@funkaoshi - seriously, send me a PM! Let’s get together.


If it’s not too much trouble, could you expand upon this? I think it’s an interesting way to frame the issue, because at least as I’ve understood your framing, when a PbtA game is giving each participant more narrative control, it’s going to reduce occurrences of “non-optimal” uncertainty. In a non-PbtA game, if a violent and dangerous monster is charging at a combat-poor character, that character has somewhat limited options on how to respond; but choosing between three different kinds of hell is exciting and tense, and makes success against crazy odds hugely dramatic. If I’m understanding what you’ve wrote correctly, in a PbtA game where the GM does not have as much narrative control they will be forced to spin the story more reactively and thus can’t squeeze a player in the quite the same way. Doing so to an unreasonable degree is bullying, but at the same time I think the most dramatic beats of a story are challenging characters with their weaknesses rather than their strengths.


I actually don’t think it’s that clear cut. In a lot of OSR style play, the characters go forth to encounter danger and trouble. They seek it out and choose what challenges to confront.

In a PbtA game, the MC is expected to mess around with the characters and throw them trouble they can’t deal with easily.

Most games have some kind of XP trigger complex which incentivizes players to make moves they’re “bad at”, as well.

So it can go both ways.


I’d love to say more, but I don’t think I really understand what you mean here. Can you give me an example of a PbtA game that “gives each participant more narrative control”?


Most PbtA games, and specifically the culture around MCing them, communal world building and “provocative questions”, certainly do end up giving the players much more influence over the contents of the game.

However, the focus is different. In fun PbtA play, the players use that freedom to create problems for themselves.