Rules do not elide

A rule can be a statement of what you can do, what you must do, or what you must not do, and it may also describe *how* you must do a thing. Rules have all kind of uses. In tabletop roleplaying games, one of those uses is to simplify arguments about how things ought to play out in the fictional situations we imagine and describe at the table. This has led to the claim that “rules elide”, and from what I can glean from the froth and din of internet discourse, some people think this is a very important claim*. But I am going to argue this is, at best, a partial representation of what rules do.

Why might we say that rules elide? Well, if you haven’t already, you can go and read the article where this idea was first publicly described (though it has now been deleted and is only available on backup, so don’t harangue the author** about it – which, come to think of it, don’t do that anyway). From reading that article, I take the essence of the “rules elide” claim to be this:

  • Without any rules, we describe the fiction in intricate detail, carefully characterising each little bit of what is happening. For example, we might describe each step, slip, swing and cut of a sword fight.
  • Rules enable us to shortcut that by instead rolling a die and saying “I win”, skipping out all the detail in between.
  • Therefore, rules elide the details of the fiction. Not necessarily all the details – and indeed, by eliding some parts, they effectively highlight what is left – but nevertheless, this is their role. Rules elide.

Whenever one describes another person’s argument there’s a risk that this leads to simplification or misrepresentation but I think the above is a fair summary of the “rules elide” argument.

The problem is that the first bullet isn’t true.

Without any rules, we might indeed describe the fiction in intricate detail. But we might not! In the linked article, an imagined player group describe the picking of a lock by working through the movements of pins and tumblers and whatnot. But that is certainly not how I would handle the picking of a lock. Here are a few ways I would think about doing it:

  • I might make a quick decision either way and describe what happens. “Nope, you can’t pick it, and your lockpick snaps.” “Cool, it opens.”
  • I might describe the lock and ask the player picking the lock what happens. “This is a top quality valyrian steel lock. How good a lockpicker are you, do you think you can open it?”
  • I might present a “yes, and” or “no, but” sort of approach. “Sure, you can open it, but it is going to take a while and you can hear the footsteps of the guards approaching, what do you do?”

You could of course argue that somewhere in there I’m using rules. My ability to “make a quick decision” rests on some rule, whether explicitly stated or implied, that I have a right to make a decision (probably because I’m the GM). But in reality if we started trying to roleplay together without discussing any rules at all, it is very very likely that at some point someone would skip over most of the fictional detail and make such a decision. Indeed, the decision to say “you find a locked door” is such a moment. The casual elision of details is actually a fundamental part of storytelling and imaginative play, regardless of whether you think there are rules involved.

So it’s not the case that roleplaying without rules is inherently very detailed.

But more than that: rules can actually supply detail where none would otherwise exist.

Consider the classic case of a game of cops and robbers. “Bang bang! You’re dead.” “No I’m not!” In this game, we skip over enormous amounts of detail. I point my two fingers at you and say bang bang, you fall over or you don’t. At no point do we consider what kind of gun or ammunition we’re using, what armour we might be wearing, how good my aim is, the potential elements affecting my aim such as distractions or (at longer ranges) wind. We are eliding the heck out of that gun fight.

I defy anyone to look at the crunchier roleplaying game manuals and tell me that these games elide compared to this simple roleplaying activity. Looking at my old copy of Shadowrun 4th edition, I see 2 and a bit pages of rules for adjudicating initiative, 6 and a bit pages of rules for different types of ranged fighting, plus 9 pages of details about different types of ranged weapons (with pictures!) This is not eliding *anything* relative to how I would normally conduct a firefight in a roleplaying game; in fact if I take it seriously as a set of rules, it enormously expands the level of detail and precision I would use in narrating such a fight.

So at the very least, I think we have to say that rules don’t always elide. Perhaps they are always eliding compared to some perfect simulation of reality, but that is certainly not the default or most commonly observed state of imaginative play when unmediated by rules.

What, then, will we say about rules? Rules are a way of guiding the conversation. We can drill into the details we find interesting, as Shadowrun players presumably find weaponry. We can skip over the details we don’t find interesting, as many other games do with the details of the same weaponry. We can remove arguments over who went bang bang first and who is now dead. We can specify how decisions should be made, and who gets to narrate the outcome of those decisions, as when Forged In The Dark tells us who gets to decide what dice to roll and who gets to decide the level of risk and effect. We can quantify things that would otherwise remain vague, as with hit points in D&D or sanity in Call Of Cthulhu. We can introduce details that don’t exist in the real world, like the characteristics of a monofilament whip. We can force the conversation at the table down particular lines, as when in Ten Candles, if the last candle goes out, the characters die regardless of what they may have had planned. We can force a player to describe something they hadn’t even thought about until a moment ago, as when in Apocalypse World a character is suddenly asked to tell us their secret pains.

Rules can sometimes elide. But more often, rules specify. Rules focus. Rule describe. Rules intrude.

Rules are a magical way to shape our conversation at the table, to direct play and to bring imagined worlds to life. Used unwisely they can be blunt instruments that get in the way of good storytelling. But the best rules help us tell stories we almost certainly wouldn’t have told otherwise. They mostly don’t do it by eliding.

If you enjoyed this blog article, please encourage me to make more by supporting the Black Armada Patreon.

*Side note: I am far from the most up-to-date person on internet discourse. I think this claim may be fairly old, at least by the frenetic standards of discussion on X and other such places. I’m sorry if I sound like an old man shouting at clouds.

**Which, to be clear, I have no idea how important the author thinks “rules elide” is as a statement or how sweepingly they intended to make that statement. Perhaps this entire article is a statement of the bleeding obvious – ironically I kind of think so. Still, discourse eh?

The rock of dramatic potential

It is a fact that some roleplaying games get into the meat of the story faster than others. One way to do this is to have a clear mission which is the focus of play, like “raid this dungeon” or “investigate this murder”. But what about the more character/relationship driven side of play? What is the difference between a game that cuts straight to interesting, meaningful drama, compared to one that takes ages to get going, whose relationships are lifeless or where the drama is just sound and fury, signifying nothing?

This isn’t just a theoretical question. I’ve lost count of the number of games that have handed me (through character/ world building at the start of the game) a bunch of interesting characters and intriguing relationships, but where I and the other players were at a loss where to start with bringing them to a head. The result, too often, is unsatisfying early scenes where we skirt the drama, or charge headlong at it, emotionally flailing at each other, but without any real sense of meaning.

I think of interesting RPG drama as like a rock that you must get rolling. The rock must have the heft created by meaning, but it also must have momentum so that interesting scenes can happen. To get the rock moving, you must roll it up the hill of dramatic potential, before it can roll down that hill, generating interesting drama in its wake. Games vary wildly in how big a rock they give you, but also in how much work they do to roll the rock up the hill.

Metaphors are all very well, but what does this mean in practice? Let’s take an example of a potential PC romance. A classic game like D&D wouldn’t bother to give any help making such a romance happen; it’s entirely in the player’s hands to do that. They might decide to introduce some romantic interaction but it might feel forced, or require quite a bit of work to get it going or make it feel significant.

Indie games might more typically help set up some potential, by asking you relationship questions like “which other PC do you have a crush on” or even “which other PC are you in love with”. This, then, is the rock: pre-generated emotional weight. It means weaving into your character’s backstory (even if only implied backstory) a sense that they have been interested in this other character for a while. It gives them an automatic reason to pursue a romance, and makes any resulting scenes more significant for them.

But it’s still pretty boring, as it stands. Even having established that one character (let’s call them Romeo) is in love with another (let’s call them Juliet), we don’t have a particularly dramatic relationship. One person being attracted to another, even in love with them, does not make for drama. Romeo may ask Juliet to dance with him, or suchlike, and she is left to either respond positively or negatively. Perhaps they’ll even hop into bed straight away. Which is fine, but not terribly dramatic or meaningful, because we have no real sense of their emotional context beyond “he’s in love with her”.

It’s actually pretty hard to get started on a conversation with your crush, as Peter could tell you.

For drama, there must be this emotional texture and, ideally, interesting complications. Consider these alternative starting relationships: Romeo is in love with Juliet, and Juliet is in love with Romeo, but they belong to warring factions who will never accept their love. Or: Romeo is in love with Juliet, but Juliet cannot forgive Romeo for killing her best friend. Or: Romeo is still in love with Juliet, after their marriage ended in acrimony.

These wrinkles add colour and meaning to a bland relationship, and they set up interesting stuff to happen in play. The warring factions are going to get up in Romeo and Juliet’s face and force them to work to have their romance happen. Or, Romeo is going to have to work to get Juliet to even consider him, and whether she says yes or no it will be freighted with meaning for them both. Or, their every action will be loaded with the regret and longing of their broken relationship and the question of whether it can ever be revived.

Good stuff. This has made the rock heftier, because it’s made the relationship more interesting and dramatically meaningful. Anything that happens to that relationship in future will be more significant because of the work put in to define and complicate the relationship.

But it still takes work to roll the rock up the hill. Some of the above starting relationships have more dramatic potential than others. What this amounts to is, to what extent is the relationship in a stable equilibrium where there’s no real reason to expect interesting stuff to happen, and to what extent is the relationship close to an interesting turning point or crisis that will throw it into motion. If Juliet hates Romeo’s guts because he killed her best friend, that is very interesting but seems like a brick wall in Romeo’s path. It’s hard to know how he even gets started romancing her, because the obvious answer to any move he might make is “get lost, friend-killer”. The relationship is stagnant, immobile. One of the people playing these characters is going to have to work (probably a lot of work) to get their character into a position where that can change.

Did you just kill my best friend? Get lost, friend-killer!

In contrast, if Romeo and Juliet are already in love, and there’s already a war between their factions, then that is close to crisis point. You can immediately see the possibilities for scenes that will pit their love against political reality. All we need do is have their midnight tryst witnessed by a faction member, and we are straight into crisis. Or perhaps we can have a close faction ally of one character kill a close faction ally of the other, to throw the relationship into conflict and emotional confusion. Here, the rock has been rolled nearly to the top of the hill, and it takes only a little more work to push it into action.

We can get the rock even closer to the top of the hill, very easily. Put simply, we can decide at the start of play that a crisis or turning point-inducing event has already happened. So for a starting relationship: Romeo and Juliet are in love with each other, but Romeo just killed Juliet’s best friend. Our game will start with Romeo having to decide how to break this news to Juliet or perhaps to try and conceal it from her. That’s an instant scene starter and, no matter what Romeo does, a drama-generator. The rock has practically started to roll already.

We could have started from cold (as in the D&D example above) and got into the above very dramatic situation in play, and there is a good argument that getting into stuff in play is more interesting than just defining it up front. But doing so entails a lot of work, during which no drama to speak of is happening, and with the risk that we’ll never get there. After all, good authors sometimes struggle to create engaging, meaningful drama between characters. It isn’t actually easy. Instead of taking that risk, we can kickstart the drama, propelling us towards exciting in-game decisions that lead to more drama, if work has been done to put the characters at some kind of inflection point at the start of play.

We can do the same thing but with a slightly less immediate “must address this NOW” feel by putting the crisis-inducing event further into the past. Like this: Romeo and Juliet are in love with each other, but Juliet doesn’t know that Romeo was the one who killed her best friend. Which other character knows about it? This approach puts the rock of dramatic potential at the top of the hill, but stationary. It only takes a nudge to throw it into motion: the character who knows Romeo did the deed tells Juliet. But that nudge can be held back and delivered right at the most exciting moment, when it will cause maximum emotional chaos.

Creating dramatic potential and putting things close to (or preferably at) an inflection point is particularly important for one-shot play. If you want character relationships to be front and centre in a one-shot, they simply must be made complicated and pushed to an inflection point, so that meaningful drama can happen in the session. Campaigns also benefit from this because it gets things going right away and enables the relationship to further develop rather than struggling to get going in the first place.

So, whether you’re a game designer or a GM or player, you can help to get drama going in your games early by:

  • creating emotionally charged relationships at the start;
  • complicating them; and
  • putting them at or near an inflection point as play starts.

These simple steps will virtually guarantee exciting emotional drama right from the word go, and make you wonder why you ever settled for questions like “who do you have a crush on?”

How does system matter? (part 2: house rules, mods and hacks)

This is the second in a series of articles. In the first I talked about designer intent, and mechanical procedures.

As discussed in the previous article, designer intent is important but only so far as it is successfully communicated and/or implemented through the game text. We’ve talked about how the procedures of play can shape what the game is like: now let’s talk about what happens when you modify those rules through house rules or hacking.

First up, what’s the difference between house rules and hacks? I think it’s a difference of scope and formality, and can reasonably be said to exist on a sliding scale. A house rule is generally a focused change to the rules given in the official game, a hack is a much broader set of changes which, taken far enough, can become a completely new game. It’s a slightly arbitrary distinction but when you’re far enough apart on that spectrum, the distinction will matter, as we’ll see.

So starting with house rules: small, focused changes to the rules. In my experience these tend to come about when someone is playing a game, and a situation comes up that they want to handle by something other than fiat, but the official rules don’t give a mechanistic way of doing that. I emphasise mechanistic, because as we’ll see later, plenty of games give you tips or guidelines for making decisions in spaces that aren’t covered by the mechanical procedures of the game. But usually when a house rule comes about, it’s because the group wants something harder edged, that defines something you definitely can or can’t do, or that involves picking up the dice.

So we’re talking about discrete, focused, mechanical changes. They are the same as what I call “procedures”, meaning they are pretty much fixed in their operation: give them input A, they’ll give you output B (or perhaps randomly either output B or C, say). There will be interpretation about when something in the game has triggered input A, and how output B or C manifests, no doubt, but still – a fairly rigid mechanism that compels B/C to happen when A happens. And like other procedures, that means they force the game into a particular shape.

It’s useful to now contemplate the designer’s intentions again. The designer, we may hope and assume, has created a functioning whole: a collection of rules that work effectively to generate the type or types of play that the designer envisages[*]. Your intervention is hopefully giving you the type of play that you wanted, but it might push the game outside of its original parameters and into an entirely different play space. Or! It might not. Maybe your new rule is on just the same page as the designer, and it’s more of an incidental thing that they happened never to include such a rule – perhaps they didn’t have the space, or never thought of it.

Point is, by introducing a new rule, you’re changing the system. Maybe only a little bit, or maybe it’s more significant than that, either because it has a big impact on play, or because it interacts with other mechanical parts of the existing rules in unexpected ways. That’s a good thing! You wanted to change how the game worked, and that’s what you did. Just, it’s important to understand that you’re now playing a new game, and because system matters, that might make a big difference to the experience you have at the table.

An example may be useful. I was chatting to Ben Riggs on his podcast Plot Points the other day, and mid-way through a conversation about why Last Fleet isn’t a D&D hack, we came to an idea that you could allow characters in D&D to heal some hit points when they have an emotionally meaningful conversation. Ben was quite taken with this idea, and hopefully will try it out.

Here’s what I think will happen when he does: the players will start talking to each other about stuff that they never previously bothered to talk about, or at any rate will do so more frequently than they did before. Talking about feelings. Talking about things that give us feelings – happy or sad events, hopes or dreams, worries or fears.

More than that, if they’re clever they’ll start to set up stuff that they could talk about at some future juncture. Like, if you’re using the official D&D rules, the only incentive to introduce an ailing mother into your background is if you’re a drama llama who likes that sort of thing. But under the new Ben hack, doing this is an excellent idea, because when you’re low on hit points later on, you can talk about how worried you are about your mother and ker-ching! you get some hit points back.

And because it’s docked into hit points specifically, they will be having these conversations at a specific time i.e. after a fight. No point having the conversation just before a fight, in fact you’re squandering a future opportunity to heal! So you’ll get a rhythm: encounter, conversation; encounter, conversation. Now imagine if it was docked into the advantage system instead, so that you got advantage on your next roll. Or if it was docked into the XP system, so you just got some XP. That incentivises totally different types of play.

So you can see that even quite a small change to the rules could create a fairly radical change to the play experience. Now when you get into much broader changes to the rules, that’s the danger zone when it comes to system. Think of the system like a garden. If you plant a small herbaceous plant somewhere in there, it could potentially harm or benefit neighbouring plants, or those plants might harm or benefit it, but the garden as a whole won’t change that much. If you plant a gallumphing pine tree in the middle of that garden, you’re definitely going to change the look and feel of the garden, and maybe kill off some other plants. If you do that a lot, before you know it you’ve got a different garden entirely. And the key thing is, if you do it at random, the garden will be a mess, and many of the original plants may become unhappy.

Don’t get me wrong: a hotchpotch, higgledy piggledy garden could be beautiful and enjoyable. I’m not saying “you must design your game to be completely coherent and perfect”. But on the other hand if you simply throw things together randomly, you’re trusting to luck that it’s all going to work in combination. The more you venture from house rule to hack, the more you can benefit from thinking about that garden – that eco-system – as a whole, and designing stuff to fit together.

Ok, that’s enough for now. Next time I’ll start thinking about the softer stuff – venturing beyond procedures and into principles and culture.

This article is supported through the Black Armada Patreon

Become a Patron!

[*] But what about generic systems, I hear you cry! Well, so-called generic systems are generally promoting a particular type of play, too. D&D is sometimes held up as generic, but its systems aren’t designed to do just anything – for instance, they ignore completely the emotional life of the characters to focus on practical matters like whether I can kill this orc or get past this trap or seduce this guardsman. The designers of D&D thought their game would focus on orcs and traps and guardsmen, not hopes and fears; other games do the opposite. That’s the sort of thing I mean when I say the game is going to generate a specific type of experience.

How does system matter? (part 1: designer intent, and mechanical procedures)

There seems to be a mini-rash of “system matters” discussion happening at the moment. I’ve often found these discussions get lost in differing definitions – you can’t agree whether system matters if you don’t agree what system is. More importantly, different aspects of system matter in entirely different ways. So rather than debate whether it matters, I’m going to break down different aspects of “system” and consider what’s important about them. This will be a multi-part series.

Here’s a list of things I’m planning to talk about in this context. Possibly more will come up later.

  • Designer intent
  • Formal written procedures of play (“the rules”)
  • House rules, mods and hacks
  • Written principles and implicit directions
  • Unspoken rules at the table
  • Play culture

Let’s start with designer intent. You might think this is not part of system (and I pretty much agree) but it obviously shapes many of the other items on the list above, specifically the formal procedures, written principles and implicit directions.

A designer can have a greater or lesser focus on specific themes, a specific type of experience they want you to have, or particular styles of play that they favour. The important thing to say here is that designer intent only matters to the extent it’s communicated to actual or potential players. This can be done through the rules, through the background material, through guidance, through the game art, even through marketing or interventions on social media.

But let’s face it: in most cases, people will only have take in what’s in the game book. Anything else, no matter how important, is likely going to be missed by most of your target audience.

Regardless of how it’s communicated, it doesn’t matter what you meant when you designed your game, only what the players understand. This isn’t a philosophical point about the nature of meaning, but a practical point about the nature of game design. Of course different audiences will take different meanings from what you say, and with the variety of game culture that’s out there it is very likely that someone, somewhere is going to misunderstand what you intended.

In fact it’s even worse than that, because (in my experience) many people don’t properly read the rules at all. They skim, they grab the printouts and run, they make assumptions or ignore rules they don’t understand. This is one of the things that makes playtesting important, because you don’t know how people outside your circles will read (or not) your words.

Game design is communication. Communication is messy and imperfect. No amount of playtesting can eliminate misunderstandings. You aren’t designing a car, where the systems interlock and perform in exactly the way you imagined; you’re designing practices for humans, and humans never perform the same way twice.

Anyway, that’s a long and rambly way of saying that design intent is hella important, but ultimately once you put the game out there in the world, no longer matters.

So what does matter? Well, let’s get our teeth into what most people probably think of when they hear the word “system”: the rules.

Here I like to talk about procedures. A procedure is a structured way of doing something. It takes an input, and turns it into an output, according to a fixed, mechanistic formula. Or in the case of roleplaying, it more typically takes many and complex fictional inputs, turns them into fewer and simpler mechanical components, futzes around with them (in a structured, mechanical way) to generate mechanical outputs, which in turn are translated back into fictional outputs.

Still: the distinction I’m making when I talk about procedures is that they’re mechanical. No matter how complex your rules, the procedural parts of them can be boiled down into simple if-then statements. That’s not all there is to rules – we’ll get to principles and directions, later on – but it’s an important aspect of “rules” that dominates many people’s thinking, perhaps because everyone is familiar with rules from other contexts like board games and wargames.

But it is worth pausing to note that in a roleplaying game, you cannot activate a mechanical procedure without first making a fictional interpretation. Even something simple like an attack roll requires you first to recognise that someone has attacked someone else in the game’s fiction. You have to interpret the fiction to do that. “I hit him with my sword”, cool, make an attack roll. “Did I mention my sword is made out of marshmallow”, oh, uh… I guess not then. So, as I’ll discuss in a moment, clearly rules are important, they matter, but they only function as filtered through the human and fictional medium: your brain and the brains of the other people at the table and the stuff they’re trying to imagine together. (Maybe we’ll get back to that later.)

The key thing about procedures is that they are fixed. Once you’ve decided that it’s time to make an attack roll, you must roll a d20, and if it equals or exceeds the target’s AC, you must roll damage and subtract it from the target’s hit points. If the target’s hit points reach zero, they die.

Seems pretty hard-edged, and with examples like that we can all clearly see that the rules are going to matter. If the rules say that short swords do d6 damage, and a normal human has d8 hit points, we can see that humans will typically last a lot longer than if short swords do 2d6 damage and a normal human has d4 hit points, or if we skip damage rolls altogether and just say that a successful hit roll kills the target.

All this is deciding is how quickly we go from “roll initiative” to “everyone is dead”, but it will make a huge impact on the play experience. Would you want to start a fight if one successful hit roll will kill you? That will feel a lot differently than if you and your buddies each get 50 HP while your opponents get about 10. And that’s without even getting into whether the game includes rules for fighting in the first place.

So one way in which the rules matter is that they compel you to change the fiction, and they compel you to do it in particular ways. If your game rules say that one successful hit roll = death, you’re compelled to play a game where fighting is really dangerous, and so we either won’t have very much of it, or we’ll have a lot of people dying. If your game rules say that player characters have tons of hit points but NPCs don’t, we’re compelled to play a game where fighting is pretty safe for PCs, which is very different.

Compelled? Well, yes. If you use the rules as written. We’ll be coming back to house rules, play culture and all sorts of ways in which you can ignore the rules. But enough to say here: obviously if you ignore the rules then they don’t matter. Rules only compel you if you let them. Still, something isn’t really a rule at all if you don’t obey it, right? As long as you’re following the procedures to the letter, they really fucking matter.

One other way in which the procedures of play matter is that they generally cover only specific types of thing, not everything you could possibly do. D&D Basic*, the ancient and revered forefather of the biggest fish in the roleplaying sea, didn’t have any skill system. If you wanted to, I dunno, deceive a guard, there weren’t any rules for that. You could houserule it, you could make something up on the fly (and we’ll get to the ways in which a game can actively encourage you to do that, or not). But it wasn’t in the book, and that meant that deception was only a part of play if the group decided to make it part of play. Unlike stabbing things with swords, which was explicitly and formally coded into the game.

Now obviously many people put deception into their D&D Basic game. This may have been an inevitable consequence of the narrowness of the rules in that game, the massive gaps it left, and the incentives of play: obviously someone was going to want to lie to a guard at some point. Obviously someone would need to come up with a way to do that. And so a thousand house rules were spawned, and eventually D&D 5e. But meanwhile, there wasn’t any fixed way to handle deception, and very probably many games didn’t have it in at all. And practically nobody ran D&D without fighting in it. Because, amongst other reasons, that’s what the rules were focused on: fighting, not lying.

So that’s two ways in which the procedures of play matter: they fix certain ways of doing things by making them mechanical; and they channel you towards doing certain things rather than other things. Those are pretty big impacts!

Next up, we’ll think about some things that modify the procedures of play, and some things that aren’t procedures (as I’ve defined them) at all. The next article in this series is here.

This article is supported through the Black Armada Patreon

Become a Patron!

*Ok, it wasn’t called D&D Basic** back then, and many people don’t call it that now. Probably it wasn’t even the first, because once it was chainmail or whatever. The point stands.
** I’m told I should head of pedants by saying I’m referring to the original edition of the game, circa 1974. Honestly, I’m not even sure. If that version didn’t have skills mechanics, great, it’s an example of what I’m talking about.

Push and pull mechanics

A popular mechanic which crops up in a lot of excellent games is Conditions. A Condition is a problem that’s affecting your character, like “broken leg”, “on fire” or “suspected traitor”. The idea is that the GM will hit you with trouble when your Condition would be relevant, or penalise you when it would get in the way. But this doesn’t work as well as you might think.

The trouble is, Conditions hardly ever get used. Why? Because they require the GM to Push them into the game[*]. In other words, the GM has to remember that the Condition is in play and bring it to bear on the game’s fiction and/or mechanics. This requires the GM to divert their limited attention and make a decision. I found this out to my cost in recent playtests of Last Fleet, where a whole bunch of Condition-type mechanics just never seemed to bite.

There are various ways you can increase the salience of these GM-Push mechanics, to help avoid them disappearing into the general cacophony of demands on the GM’s attention. Taking a mechanic off the player’s character sheet and putting it onto the GM’s reference sheet where the GM can more easily see it, for instance. Or better yet, put it in big letters on a nice, visible index card that sits in plain sight right in front of the relevant player. But even so, that’s just greasing the GM’s cognitive wheels a little. You’re still putting the onus on the GM to turn those wheels.

But there’s two ways you can restructure your mechanics to make them work without needing a GM push:

  • Turn them into a Pull mechanic
  • Turn them into a Player-Push mechanic

Let’s start with Pull mechanics. A Pull mechanic is automatically activated in fixed circumstances, drastically reducing the cognitive demands of the mechanic. For example, where a Condition relating to an injury generates work for someone to apply it in the fiction, a simple Harm or Hit Points mechanic are much easier to track.

You might think that sounds kind of boring: am I really recommending Hit Points as a mechanic? Well, it needn’t be dull. The system of Marks in Night Witches is essentially a Hit Point system, but it’s one where every time you take damage, something interesting happens. It’s just that the interesting thing happens automatically. Whenever certain Moves are triggered, someone has to choose consequences which include taking a Mark, and when someone takes a Mark, the fictional or mechanical consequence is applied right away.

The other approach is to use Player-Push mechanics. The difference here is, you put the onus on the owning player to activate the relevant mechanic. A good example is the Conditions found in Masks: A New Generation. Although having the same name as the GM-Push mechanic mentioned above, these work differently. Each Condition is tied to a particular Move or Moves in the game, and creates an automatic penalty each and every time that Move is used. The penalty itself is a Pull mechanic: whenever you use those Moves, the penalty applies, so no thought or judgement required. But removing the Condition is a Player-Push mechanic: if a player wants to remove the “Angry” Condition, they have to break something important, which requires them to choose that action. Now a player character’s Anger can be applied to create interesting complications in scenes, but it’s up to the player to choose when instead of the GM. And there’s a clear incentive for them to do it, because as long as they’ve got that Condition they’re continuing to take the penalty.

Another nice thing about Player-Push mechanics, by the way, is that they hand more power and narrative control to the players. That’s usually a good thing in my view: it means they’re more engaged with the game, and it means that the trouble that is created is something they’re eager to get to grips with – after all they chose it.

You can even combine the two. An example is Pressure in my own WIP game Last Fleet. Pressure is used as a kind of Hit Point system, whenever a character takes harm but also when they take an emotional shock – a Pull mechanic that happens automatically when the rules say so. But players can also voluntarily Mark Pressure to get bonuses to rolls – a Player-Push mechanic that provides a clear incentive for the players to make trouble for themselves. Finally, when you get to 5 Pressure, you hit Breaking Point, forcing you to choose from a list of irrational or risky actions to take that will complicate your character’s life. Breaking Point is another Pull mechanic: it kicks in without any decision needing to be taken.

There’s definitely room for GM-Pull mechanics. In most games, part of the fun of being the GM is to exercise your attention and judgement to spot opportunities to make interesting stuff happen. You wouldn’t get to do that as much if only a mechanical trigger or player decision enabled you to do it. But, in the interests of lightening the burden on the GM and ensuring your mechanics actually come into play instead of sitting unused on someone’s character sheet, consider using Pull and Player-Push mechanics instead.

As a coda to this, I took three distinct GM-Push mechanics in Last Fleet and converted them over to a mix of Pull and Player-Push mechanics. I’m really happy with how they bring into sharp focus elements that were previously relegated to a minor role or just plain didn’t work.

This article is supported through the Black Armada Patreon

Become a Patron!

[*] Ok, to be fair, in a lot of games other players can activate Conditions too, using another player’s Condition against them. But the principle is the same – it requires someone else to think of the Condition and bring it into play.

Breaking down hard moves

Apocalypse World introduced the concept of Hard Moves, i.e. the individual interventions the GM makes in response to players’ actions and rolls. Subsequent PBTA games have tended to distinguish between “hard” and “soft” moves. But they vary wildly in how much time they put into explaining the distinction. Here I’m going to talk about the different ways in which a move can be “harder” or “softer”.

Before I get started I should pause to acknowledge that Magpie Games have done a pretty great job of talking about this in the past. Check out Urban Shadows and Masks: A New Generation for a particularly good treatment of the topic. What I’m trying to do here, building on that, is to break down and codify the different types of move “hardness”.

My thinking about moves is informed by my professional life, where I have some experience of risk management. Risk management is the discipline of recording and managing all the bad stuff that might happen to your business. Sometimes people use similar techniques to manage potential good stuff too (opportunities); and you can also manage “issues” which is essentially a risk or opportunity that has already come to fruition.

In risk management, we think about the probability of a risk coming to pass, and the impact if it comes to pass. You can have an absolutely terrible risk that would devastate your business, but which isn’t very likely to happen, or a fairly mild risk that would cause you problems, but is highly likely to occur. You’ll treat those two very differently, and its debatable which you ought to focus on. But for our purposes, you can apply a similar(ish) approach to thinking about moves.

Start with probability and a related concept, proximity. Let’s compare some scenarios where the GM is making a move:

  • There’s a woman with a grenade launcher, and she’s fired it at you. It’s going to explode at your feet in a few seconds’ time. What do you do?
  • There’s a woman with a grenade launcher, and she’s aiming it at you. What do you do?
  • There’s a woman with a grenade launcher, and she’s aiming it at your party. What do you do?
  • There’s a woman with a grenade launcher, a few metres away. She hasn’t seen you yet, but she’s about to. What do you do?
  • There’s a woman with a grenade launcher a few blocks away. She’s heading in your general direction. What do you do?
  • You get a message warning you about this bounty hunter, Screwball. Her favoured weapon is a grenade launcher. Apparently she’s taken a contract to take you out. What do you do?

Obviously, all these scenarios vary in hardness. In all cases, there’s a potential threat – a woman with a grenade launcher. What’s different between the scenarios is how definite that threat is (how likely it is that the threat will come to pass) and/or how immediate that threat is (how close the threat is in time or space). In all cases the basic worry is “I might get blown up with a grenade”, but the hardness of the move is hugely different depending on the probability and the proximity of the threat.

A caveat to the above is that in most PBTA games, if the GM has mentioned something as a threat then we kind of know that, if we do nothing, the threat will come to pass. There’s not really any such thing as probability – the GM decides what happens. Of course other games are different; in OSR games, for instance, the GM might well roll to decide if an extant threat heads your way or not. In that case probability can be a genuine factor. Either way, a move that is currently low-probability and/or low-proximity at the very least gives me longer to react, so it’s a softer move.

Now let’s think about impact. In risk management this is broken down simply by severity, i.e. where does this sit between being an existential threat down to simply a minor inconvenience. That applies to GM moves as well. But in roleplaying we can also think about significance, i.e. how much do we care about the outcome. Some examples may help to illustrate:

  • A gang has hired an assassin to kill you. What do you do?
  • A gang his hired a local leg-breaker to rough you up. What do you do?
  • A gang has decided to burn your house down. What do you do?
  • A gang has hired a local leg-breaker to rough up your nephew. What do you do?

Now obviously in the examples above, you’re more worried about an assassin than a leg-breaker, because the severity of the threat is lower in the second case. But which is more of a threat – having your own legs broken, or those of your beloved nephew? We don’t know the answer to that unless we know how significant your nephew is. If they’re your beloved nephew, you might put their wellbeing above your own. Similarly, what about your house? Losing your house is probably less severe than losing your life, but the significance is unknown.

Moves that are significant are often much more of a body-blow than a far more severe move that’s of lower significance. Here’s some fun examples:

  • Your hated mother has hired a local assassin to kill you. What do you do?
  • Your beloved mother has hired a local assassin to kill you. What do you do?
  • Your beloved mother has tears in her eyes. Her lip trembles but she hardens her face and says “Get out. I never want to see you again.”

It seems like having your mother target you with a move is probably more significant than when it’s some gang. But it’s a big difference if your mother is beloved or not. How terrible, that your mother – who you love – has gone so far as to have you killed! But the final example is perhaps the most significant: even though the severity is lower (we aren’t talking about life and death anymore, after all) having your beloved mother cut you out of her life might be seen as a “harder” move than the other two.

In my view, significance is the key to really effective hard moves, because they hit you in your emotions. It is worth reflecting, though, that a really significant threat requires some ground-work. My nephew is going to seem more significant to me, the player, if we’ve spent some time establishing who he is and building him up as beloved, than if he’s simply introduced as “your beloved nephew, who is now under threat from a local leg-breaker”.

Related to the above is target. We tend to assume that a move that targets you is harder than one which targets someone else. But that isn’t necessarily the case – after all, people will risk their own life to save that of someone they love. Similarly, we can ask which is a harder move: your beloved mother cuts you out of her life, or your beloved mother cuts your dying father out of her life? The answer, of course, goes back to significance, but I mention target as a separate issue simply because it’s easy to forget that you have the option to target someone other than the players.

Another layer you can add on top of all this is choice, because forcing people to make decisions automatically makes a move harder, and because a choice between two bad outcomes means it might be literally impossible to avoid both. The mere fact that you had to choose the bad outcome can make the resulting badness seem more significant, too.

The final variable I want to mention is sign, as in positive or negative. That is, we can distinguish between moves that have consequences that a character perceives as bad, versus those which have consequences that the character perceives as good. I mention this mostly because – just as most risk managers focus on bad stuff over opportunities – most GMs focus on escalating bad situations over offering potential rewards. It’s a totally valid move, even in response to a bad roll, to offer an opportunity. A “positive” move might even be seen as a fairly hard move in the right circumstances, if it’s an opportunity that comes with risks.

So that’s the seven ways you can vary the hardness of a move:

  • Probability
  • Proximity (in time and space)
  • Severity
  • Significance
  • Target
  • Choice
  • Sign

What techniques do you use to keep your moves interesting? Let me know in comments!

This article is supported through our generous supporters on the Black Armada Patreon.

Become a Patron!

Is it ok to fudge rolls?

I was bodding about on Twitter recently and I came across this:

Reading through the comments I saw a lot of pushback against point 2. People saying “But it sucks if some random roll means I get killed by an orc in the first encounter.” “Some players don’t like it when they die, it isn’t fun.” “It’s ok if I got the threat level wrong, so I’m just correcting my mistake.” “You shouldn’t have rolled the dice if you weren’t ready for that outcome.” And so on. This got me thinking about why people fudge dice rolls at all.

Now as I see it, the answer is pretty straightforward. People fudge dice rolls because their chosen game isn’t giving them sufficient discretion in decision-making. They rolled the dice, and what they rolled means they are forced to either implement a fictional outcome they didn’t want, or fudge the roll.

When this happens, it is probably because the game is premised on a simple linear process:

Someone makes an attack -> roll dice -> inflict damage (or not)

See how that works? As soon as the GM picks up the dice, they’re committing to possibly inflicting damage on you. Maybe it will kill you. But that’s all that can happen. They can’t knock you out, they can’t take your stuff and leave you tied up. They can’t leave you beaten but humiliated. And that’s just thinking about possibilities relating to us fighting. We haven’t even got started on how they can’t reveal a terrible secret that will leave you crushed and sobbing, or have a totally different threat raise its head.

My point is, I think a lot of people are playing games with what you might call ballistic mechanics. You get to choose whether to pull the trigger (i.e. roll the dice) but once you’ve done that, you have no choice in where the bullet hits (i.e. what the outcome of the dice roll is). You can solve the problem by cheating – by ignoring the die roll – or by using a system that fires smart missiles instead of dumb bullets.

There’s plenty of games out there which continue to give you choices after you’ve rolled the dice. A failure doesn’t have to imply a mechanically fixed outcome. If you’re reading this and wondering what games I mean, one good avenue to google is Powered by the Apocalypse or Forged in the Dark, both of which give real, hard consequences to dice rolls – but in a way that gives you interesting choices rather than automatic pre-defined outcomes.

Even D&D (which I assume is what we’re all thinking about here) doesn’t need to be implemented like that. I mean, come on. It isn’t like you haven’t used house rules or not-technically-RAW “roll a d20 and I’ll tell you what happens” for umpteen other things. So why be a stickler about the attack roll? Why not say that on a hit, the orc disarms you, then headbutts you into unconsciousness. You wake up in chains. Hard consequences that generate more fun, instead of snuffing out interesting possibilities.

So if I’m so keen on interesting outcomes, why not just fudge the roll and do it that way? Well, like the Tweet said, this is about social contract. If you’re playing in a group where the expectation is that successful attack rolls lead to hit point loss and hit point loss leads to death, then you’re playing with fire if you don’t enforce that. You’re essentially saying “your choices were meaningless; you thought you were risking death to achieve your goals; you thought there was a point to me rolling these dice; but you weren’t and there isn’t. You might as well stop recording your hit points and stop rolling the dice because the real decision-maker is me, the GM, and I’ll ignore the dice when it suits me.” The whole point of systems with dice rolls is to create risk and drama and make choices meaningful.

The same applies to games where hit point loss and death aren’t automatic consequences of a roll, by the way. Just because I can opt to have you KO’d and captured by the orc instead of killed, doesn’t mean its all just arbitrary GM fiat. I have to abide by the fact that the dice were rolled, so something bad happens. I can’t just shrug my shoulders and say “nothing happens”. If I do that, I’m equally guilty of denying you the fruits of your decisions as the GM who refuses to inflict those hit points because they think it wouldn’t be fun.

But I think that a lot of groups in their heart of hearts don’t want a social contract like the one we see in D&D, RAW. Dying at an arbitrary moment because of bad dice rolls is not everyone’s idea of fun. I think that’s why we see so much fudging going on – because people don’t actually like what the rules tell them they must do. The point is, you’re breaking the social contract if people thought hit point loss and death was a possibility but it actually never was; but you can have a different social contract if you want. You have to ask yourself in advance whether you want arbitrary death or not. You have to talk to your group about how you play the game, and get their consent.

“Hey, I’d like to play some D&D, but I’m not really into the whole ‘one bad roll and you die’ thing, so I’m using a different set of rules. Ok?”

If you do that, then you’ll get the outcome you wanted – you remove arbitrary “un-fun” death from your game. But you get it without having to cheat people out of meaningful choice, and waste time rolling dice when you have no intention of enforcing the results of your rolls. Everyone can engage with the encounters you present, knowing what’s on the table and what isn’t.

Indie games aren’t all about narrative

I hear a lot of people saying that indie games are more focused on narrative or story. This is said in implicit (or explicit) contrast with traditional games, which are more… I don’t know… gamey? It often flows from the said people having read and (IMO) misunderstood GNS theory. So here’s the thing: it isn’t true. Or at least, it’s an extremely partial view of the indie market.

Let’s start with those last two words – indie market. I use them advisedly: “indie” isn’t a design school or a brand. It’s just stuff that’s independently published, meaning creator-owned. So indie games includes some stuff which is part of the OSR, or “trad” gaming. Yes, there’s some definite trends visible in the indie market that are different from those in the mainstream market, and so it’s not entirely unjustified to talk about “indie design”. The point is, #notallindiegames, let’s not pretend they’re all the same.

With that said, there is definitely a chunk of the indie games market that could be justly called “games focused on narrative”. My own game (with Becky Annison) Lovecraftesque fits into this part of the market. It aims to emulate the structure of a Lovecraftian tale, and it bakes that structure directly into the rules of the game, at the level of the scene. The rules literally require you to play through a series of scenes in which clues are gradually revealed, before playing through a rapidly accelerating scary sequence, followed by a terrifying finale. Similarly Fiasco has story structure baked into it. There’s plenty more where they came from. These games have “story” in their bones, and you know when you play them, at a high level, what kind of story you’re going to tell.

But a lot of indie games that are not at all like this still get painted with the “narrative” brush. A prime example is the Himalayan range of indie game design, PBTA, and that’s what I’m going to focus on now- though I think it’s true of wider indie design.

PBTA (well, most PBTA anyway) does not at all do anything to bake narrative structure into the game, except perhaps at the very high level of the GM’s prep. In fact PBTA is, to my eye, extremely similar in structure to more traditional games: each player focuses on one character, using their abilities and skills as the main mechanical interface with the world, but also interacting with it through straightforwardly roleplaying a character, with outcomes adjudicated in a fairly freeform way by a GM. Yes, there are differences – what is traditionally regarded as GM advice is baked into the rules, each player skill/ability has its own custom rule, and so forth – but at a structural level it isn’t really that different from (say) D&D.

The point is, “the story” is something that only exists after the fact. You play the game, you play your character, the GM describes the world and plays their NPCs, and only after you’ve done that can you look back and say what the story was. Just like traditional games. There’s no “narrative” in this narrative game, or no more than any other RPG.

So why does PBTA often get described as being a narrative game? I mean, partly it’s a branding thing. It’s part of the “story games” movement, which is strongly identified with “narrativism”. It’s probably partly because the rules of a PBTA game engage directly with the “fiction”, and are typically designed to never be more than one step away from re-engaging with “the fiction”, either.

But “the fiction” isn’t story. The fiction is just the game world, from the places in that world, through to the characters, the situations they find themselves in, the minute details of their exact position as described by the words you speak at the table. A character walks into a bar, two pistols cocked, and challenges the bartender to a duel: that’s the fiction. That character is standing next to a honky-tonk piano as she does it: that’s also the fiction. The honky-tonk is playing “baby elephant walking” and its middle C is slightly flat: also the fiction. None of this is story per se. None of it would be out of place in a traditional RPG. And probably this character will next roll to taunt or manipulate or intimidate the bartender, and possibly later to shoot at him with those pistols, and that will all be mechanically fairly similar to what happens in a traditional game. And again, not story.

I think it all comes back to the decision to describe one of the three prongs of GNS as “narrativism” or “story now”. Narrativism isn’t (as I see it) about focusing on “the story”. It’s about playing to engage with human issues, in such a way that it’s more likely to make a compelling story. But “the story” per se still arises from micro decisions that you make during the game. “Story now” is so called because you get straight to the interesting bits of a story, not because you in some sense consciously try to generate a story. I’m aware that like everyone else I’m no doubt misinterpreting goddamn GNS; but no RPG theory article is complete without that, is it? The point is, narrativism isn’t about turning roleplaying into story writing, and by the way (most) indie games are mostly not especially narrativist except to the extent that they empower you to play that way if you want to.

So what? Isn’t this just a tedious argument about semantics? Probably at some level it is. But I think probably some of the holy wars of the roleplaying community – the imagined conflict between OSR and indie and trad, for instance – are rooted in these wrongheaded ideas about indie. And I think the idea that indie games are “about story” makes them seem kind of intimidating and highfalutin to players who are used to D&D. The truth is, they are different from traditional RPGs. They do play differently – otherwise they would be the same game. But they aren’t a fundamentally different “story-focused” thing, so if you’ve heard tales of scary story games, those tales are wide of the mark. Maybe it’s worth giving them a try.

PBTA – Moves overload

Part of the job of a game designer is to consider the level of cognitive burden and handling time required to run a game, and pitch that at a level which works for the intended audience. I have been noticing recently how that counts in spades for PBTA.

The player-facing side of most PBTA games consists of a series of Moves, each of which has some sort of fictional trigger, and then mechanical steps you execute whenever that trigger occurs, which in turn feed back into the fiction. It’s the first of these – the fictional trigger – that makes cognitive burden a particular challenge for the PBTA designer.

The reason is that, because each Move has a fictional trigger, and because that fictional trigger is (typically) a fairly specific circumstance occurring, you have to constantly scan the fiction as it develops to check if that trigger has happened. Most PBTA games take as fundamental “to do it, you have to do it” and “if you do it, you do it” which is PBTA-speak for “if the fictional thing happens, the Move is triggered” (and vice versa, though I’m less interested in that here). This means you can’t just wait for someone to decide they want to use a particular Move and call it out – their actions in the fiction may mandate that Move.

Take a typical example from Night Witches “when you act up (by acting like a hooligan, by acting like a lady, by acting like a natural born soviet airwoman). This Move requires you to notice when someone is acting outside their normal social boundaries, and then decide whether that acting out fits with either of the three categories (if not, it doesn’t trigger the Move). So it’s a fairly complex, nuanced decision you have to take. You could miss a moment where the Move should have triggered, if you don’t pay attention.

All of this is therefore inherently cognitively burdensome, and the designer must therefore consider in each case whether that level of burden is worth the benefits delivered. Sometimes a complex, nuanced Move is worth it, sometimes it’s better to go for something simpler than might not be as precise.

Of course, it’s not just the complexity of individual triggers (though that is a factor), it’s the sheer combined weight of all the triggers that have to be considered. You could write a PBTA game with 100 basic Moves, but nobody could play it; they’d constantly fail to notice when they were triggering Moves, even if they were fairly simple.

This leads me to some critique of PBTA games I’ve played recently. They’re games I like and have enjoyed, and which are pretty popular, so hopefully it’s clear that I’m not going for a take-down of anyone’s game here. I’m using them as examples of an issue which needn’t be fatal to a game, but which I find difficult in play.

The first example is Sagas of the Icelanders, a game which attempts to emulate the eponymous stories of the early Icelandic settlers. In doing so, it imposes some quite restrictive roles on the characters, specifically relating to their gender. If you play a man, your Moves are focused on physical feats and (rather more complex) defending your honour and attacking the honour and dignity of other men. If you play a woman, your Moves are focused on influencing other characters (particularly, but not exclusively, men) through reason, emotion and sexual attraction. (The way this is set up creates a focus on male characters that is interesting but not the focus of my critique.)

Whenever a character takes action in Sagas, like every PBTA game, you have to mentally compare what they did to the Moves to see if one was triggered. Of course, you first have to focus on the correct set of gendered Moves, since actions that would trigger a Move for a man won’t necessarily do so for a woman, and vice versa. So gender introduces an element of complexity up-front. But – particularly for women – the specific triggers are quite nuanced. You can raise your voice and talk sense, but only to player characters. You can goad to action, but only aimed at a man. And because of the very specific and culturally relevant triggers, many things that intuitively feel like Moves aren’t. All of which is fine, but kind of hard work to parse in play.

Next example is Urban Shadows. Urban Shadows has Moves which are somewhat more intuitive (for me) than the likes of Sagas, which is a plus. Many of the Moves only trigger in fairly well-defined contexts (more on this in a moment). However, it has a lot of Moves. The Urban Shadows Basic Moves sheet has 15 Moves on it, not counting the rules on advancement. That is a lot of mental checks to go through every time someone does something! It’s a lot of possible mechanical triggers to remember, full stop. [Edit: discussion elsewhere has reminded me that the theoretical limit of human short-term memory is seven items, plus or minus two.] On top of that, there are four different triggers for advancement, and a trigger for corruption, that you have to keep an eye on. Again, this is fine, but kind of hard work.

So what can you do about this, if you’re designing a game? Well, the obvious stuff is:

– Keep your Moves simple, with straightforward, intuitive triggers that don’t require a lot of thought to judge.

– Try to keep your Moves list as short as you can. This in turn means…

– …focus on the Moves that really matter for the game you’re trying to write. Don’t waste your players’ mental space with Moves that aren’t all that important – PBTA lets the players and MC negotiate the fiction pretty well even if a Move isn’t triggered.

Another trick is to try and group your Moves by the context in which they occur. Apocalypse World, for instance, has Battle Moves. Including them roughly doubles the number of Moves in the game, taking it from a pretty simple half-a-dozen Moves up to more like a dozen. But you only have to think about the Battle Moves if you’re in battle. If nobody has any weapons out, you needn’t waste any brain space thinking about them. Similarly, Night Witches divides its Moves into day and night Moves, and in almost all cases you therefore only have to think about half the potential Moves at any given moment.

You might think that the Sagas approach is kind of like grouping Moves by context. Well… I don’t know what to tell you. That’s not how I experience it. Two characters are having a conversation, one of them starts to talk sense, and then I realise that person is a man, so the “talk sense” Move doesn’t apply. That small mental effort, repeated several times a session, is burdensome in a way that “am I in a battle right now” isn’t.

None of this is to say that you can’t write a game with complex, nuanced Moves. With lots of Moves. With Moves that only apply in specific circumstances that require a bit of thought to judge. All of this is permissible, and can be good design. But it is a cost that you are making your players pay to play your game. Make sure you’ve chosen punchy Moves that deliver something worthwhile, so that it’s a cost they’ll be glad they paid.