Archive for the ‘System Design’ Category

In my last post, I discussed the idea that in most roleplaying games, players’ time is split. Some of it is spent performing as a character, acting exactly as that character would in the fictional world, but some of it is spent on other activities. Exactly what those are depends on the game, of course, but it might be: straightforward narration of material that can’t be directly performed; engaging with some real-world mechanical tool like dice or character sheets; metatechniques like delivering an internal monologue for the benefit of the other players (while their characters remain unaware).

Despite the prevalence of these other activities, many games (and many players) do strive to be in-character and acting “realistically” (i.e. not monologuing or using other dramatic character-breaking tools) as much of the time as possible. This isn’t necessary for a good game, of course – Microscope is a great example of a powerful story-telling game which doesn’t necessarily require any in-character performance at all – but where you do want to encourage the players to stay in character, you can run into difficulties, as there are some actions necessary in the fiction that simply can’t be reproduced in the real-world performance. (The three example techniques above present some ways to get around that.) As mentioned in the last post, there are games like When the Dark is Gone that manage to stay in-character 100% of the time (at least after the setup phase), but the situations you can portray in such a game are heavily restricted.

An alternative approach is to set up your game so that the time spent out of character feels like time spent in character. Use the mechanics, rules, or game setup to foster in the players the mindset that you want your characters to have. There are plenty of great examples of this, such as:

  • Murderous Ghosts, Doomed Pilgrim, and many other games that explicitly have an adversarial relationship between one group of players playing “the world” and another playing the protagonist(s). This sets a tone where the world is not a fair place and in fact is actively working against the protagonist(s) – heightening the tension and the fear factor for their player(s).
  • The Jenga tower in Dread. A very visible and tangible indicator that everything could at any moment come crashing down on top of your character – again, pushing the tension higher and higher until it finally collapses (creating a very real sense of relief for the players not affected by the downfall!)
  • In Remodel, the various manipulations of the house (represented by a real physical set) reflect and enhance the emotional journeys the characters are taking. Physically ripping up tape to “demolish” parts of the house, for example, provides a powerful cathartic experience and prepares the player for their character’s new beginning.
  • In The Secret Lives of Serial Killers, key information is hidden from one player in order to give them a very real sense of horror and betrayal when it finally comes out.

Perhaps nothing can be more realistic than acting something out in full, but that’s a luxury we don’t often have. By choosing techniques carefully, you can “shortcut” some of the acting process and elicit feelings in your players that match those in the characters without needing a full in-character portrayal. If you know others I’ve missed (and there’ll be plenty), let me know in the comments – I’m keen to try them out!

Some semi-formed ideas have been bubbling around my head recently, about the boundary between a LARP and a tabletop roleplaying game. What differentiates the two? My friend Mo posted this write-up of a game I played in the other day – an absolutely fantastic game called When The Dark Is Gone – in which he suggested that despite its tabletop appearance, “When the Dark is Gone pretty much is a larp, apart from the fact that you’re sitting down”.

I can see where he’s coming from. The premise of When the Dark is Gone is a group therapy session, exploring the childhood memories of the (now adult) player characters, assisted by a game facilitator in the form of the group’s therapist. This premise allows play to remain completely in character at all times – players negotiate discrepancies between their memories by disagreeing in character rather than by appealing to an external arbitrator; the game’s pacing and direction is managed through the mask of the therapist; even the “I need a break” emotional safety mechanism can (very deliberately) be interpreted in-character even though it’s used for out-of-character reasons. So yes, I can see why it could be called a LARP. (Typically the players are sitting down, though, it’s true.)

The thing is, I don’t actually think the LARP / tabletop distinction is all that useful. After the setup phase, play in When the Dark is Gone is entirely in-character – in other words, the players develop the situation solely through improvised acting. But it wouldn’t be true to say that 100% acting is a characteristic of LARPs. Many LARPs require players to appeal to some kind of system in order to resolve certain things that simply can’t be acted out – combat being the obvious example, but there are plenty of others, varying by game. Even the highly-immersive Nordic school of LARPing frequently uses “metatechniques” like drawing a window to introduce an internal monologue – clearly not a direct  “in character” performance, but rather a theatrical technique to allow the player to portray a perspective that simply can’t be acted out directly. To my mind this isn’t so different from a tabletop player rolling a die to get their character to climb a wall – both are character-breaking mechanical techniques that allow players to introduce particular types of content that they can’t introduce while staying in character. The types of content differ between the two styles of play, sure, but the principle of breaking character to do it applies to both.

Describing a game as LARP or tabletop doesn’t really tell me very much about it. I’d rather simply know what proportion of the game time you expect players to spend acting in character. At one end this covers games like Microscope, another brilliant game but which tells the story of a world or a setting rather than a specific set of characters, and so most of it is developed through narration rather than in-character scenes. (In fact to my mind the acted-out scene mechanics are the weakest part of the game.) At the other end of the spectrum are games like When the Dark is Gone that aim to be 100% in-character once play starts – actually a very difficult goal to achieve outside of very specific game scenarios.

And once I know how much acting I’ll be doing, the next and more important question is: What will I be doing the rest of the time? Most roleplaying games include at least some time spent acting. What the designer chooses to include for the other, character-breaking, parts of the game is possibly the clearest indication of what they intend the game to really be about. Do they include rolling dice for physical actions I can’t directly perform? Must be an action-heavy game, then. Or are they including hand gestures to indicate internal monologues? Seems like that game will be psychological and introspective. If I’m considering playing your game, I don’t care if you call it a LARP or a tabletop game – I just want to know what the gameplay is about. (And, of course, whether I’ll be sitting down.)

I’ve been thinking a lot about winning and losing in roleplaying games recently, not least because of several recent games by Vincent Baker that have explicit win (and loss) conditions. Something about these games has really inspired me, but something else has jarred a little – and it’s taken me a while to work out what both those somethings are. It’s been interesting particularly for my current work-in-progress, which pits a thief (or other ne’er-do-well) against the wits of a GM antagonist. Clearly, I could say the thief “wins” if they get away with whatever it is – but is winning-and-losing the right design for this game?

What Works

One thing winning-and-losing games do really well is match the tension out-of-character with that in-character. Pitting the GM’s wits against mine makes the threat very real, and so aligns my mental & emotional state with that of my character. In Doomed Pilgrim, my pilgrim is thinking, “Holy crap, how can I survive this using only my wits??” while I am thinking, “Holy crap, how can I get this pilgrim to survive this using only my wits??” That strong identification with my character (point #4 on this post about immersion, in fact) feels exactly right for my thief’s adventures – I want the tension high as the player uses their cunning and wits to get the thief into and out of trouble, and get away with the prize.

What Doesn’t

But fun though these games are, they lack something for me in terms of replay and longevity – and that’s because they don’t, really, tell that great a story. Sure, they make for a very cool, tense scene – but that’s all it is: one extended scene. As a story audience, I want to see the protagonist develop – if not within that scene, then at least over the course of several scenes. So they survive – phew, thank goodness! Now show me what they do next. And, more importantly, why. But these games aren’t built like that – they’re mostly built as one-shot adventures (sort of) and don’t focus much on the personality of the focus character. (Deliberately, I think! That’s not what they’re trying to do.) They’re like a longer version of a conflict scene in a more narrative-focussed game – and so they’re missing the bits before and after, that frame that conflict and give it meaning. In my thief game, I definitely want to focus on the protagonist’s character development, so a purely win/lose game doesn’t work.

What I’ll Do

So I guess I want a game that’s divided into adventures, each one tense like those winning-and-losing games, but that also strongly promotes character development within and (especially) between those adventures.

To keep the tension high, I think I do want the same “you versus the GM” model – i.e. the GM should be actively trying to thwart the protagonist, rather than (as in some narrative games) just provide an environment that reactively challenges them. The direct antagonism builds the sense of threat – and makes it more personal, too.

An aside: However, I’ll need to increase the win rate. Vincent’s described both Murderous Ghosts and Doomed Pilgrim as having a pretty low chance of winning – somewhere around 1 in 3, I think; certainly less than half – and that doesn’t suit this game, where the protagonist is a master thief. I certainly need to adjust the survival rate – for these games, loss also means death (or doom), which doesn’t work for an extended narrative game – we can’t see character development if the character doesn’t survive for at least a few adventures! Also, their death (if it happens at all) should be a dramatic climax point, not just some mid-adventure mis-step on the part of the player.

A Model For Episodic Character Development

One game that does character development within and between episodic adventures really well is Dogs in the Vineyard (also by Vincent Baker). In Dogs, your job is to save towns from sin. You could call this a “win” condition for each town. But there’s lots of great character development as well – for three reasons I can identify:

  1. The win condition is defined by the protagonists themselves. (Strictly speaking, “save the town” is always the win condition, but what “save” actually means for a given town is decided by the protagonists.) This shows us something of who the protagonists are – what they value and what they overthrow.
  2. The definition of winning is challenged throughout the adventure. The town (or certain elements of it) will inevitably resist the protagonists’ efforts to “save” it. How hard can they push before they’re no longer saving the town, but destroying it? Again, since they decide this themselves, it reveals more about their characters.
  3. There’s an explicit character development phase in between each town. The protagonists reflect, consider whether they did really save the town, and change as a result – with game-mechanical changes to solidify that development for future play.

My game’s going to feel very different to Dogs, I think – but to promote character development despite the GM antagonism, perhaps I can learn something from it.

I don’t know if you’ve been following Vincent Baker’s posts on the object of a game (and a roleplaying game in particular). I’d recommend them. There’s a bunch, but this post and the next one are probably a good place to start.

I won’t link them all, but later there’s one about Recipes vs Games, which says this:

In a game, to get the object, you must contend with the rules.

A super easy example: in the Doomed Pilgrim game, your goal is to see me to my doom, but you can only answer my direct questions, so you may not be able to do it.

In a recipe, to get the object, you must follow the rules.

A super easy example: to make a PB&J, spread two slices of bread, one with PB, one with J, then press them together PB to J.

[Nom nom nom PB&J… But I digress.] I struggled to understand this. Mainly the phrase “contend with”; to me that suggests the rules make it less likely you’ll achieve the object – true for some games, of course, including Chess and Doomed Pilgrim, but not for games like Apocalypse World where the object is (as Vincent himself said) “to find out what the characters will make of their world”. In those games, the object’s given; the rules just make you do particular things to get it.

I asked him about this and he kindly posted his response, including the point:

It would be much, much easier for you to find out what my character Barbecue will make of his world if you could just ask me.

“Hey Vincent, what does Barbecue make of his world?”

“Oh! No sweat. He creates this oasis of normality amongst all the weirdness. He enforces it with good humor and violence only when it’s called for. Gradually his people become wealthy, and some of them set off to establish their own little oases after Barbecue’s model. It turns out that in the face of cheerful normality all the weirdness in the world breaks down, and doesn’t invade, and that creepy-ass metal-gnawing eyeless child was just a figment of his imagination after all. Ta da!”

But no. In Apocalypse World I don’t get to just tell you like that. You don’t get to just ask.

And I see his point. So in a sense the rules do make it “more difficult” for you to achieve the object – in that they make it a little more laborious. But they don’t make it less likely; you’re constantly achieving it, just not in the straightforward obvious way – which, of course, makes the outcome more interesting. You’re still gonna get from A to B, but the rules force you to take the scenic route. Even in win/lose games like Chess and Doomed Pilgrim, where the rules do make it less likely you’ll achieve the object, they do so to make the process more interesting.

So how about this restatement:

A recipe’s rules tell you the most efficient way to get the object.

A game’s rules tell you the most interesting way to get the object.


In another example of someone saying something I’ve been trying to say for ages, and doing so more simply and succinctly than I ever could, Bankuei has written an excellent post about making it clear what’s appropriate for the fiction of your game.

The bit that really struck me was the simplicity of these three questions:

What kind of conflicts make sense for this game?

What kind of protagonists make sense for this game?

What kind of outcomes make sense for this game?

These are exactly the things that a well-designed story game not only enforces and guides during play, but also gets across quickly and clearly to anyone considering playing it.

I’d better make sure my games do that, then :-)

A few times recently I have used the word/phrase “micro-playtesting” and people have given me blank looks, confused scowls and/or outright snorts of dismissal.

In the absence of a proper post, let this stand here for all time as the definition of what I meant – and will continue to mean – from this comment:

At this stage of development, an internal playtest just means grabbing a couple of friends for half an hour and trying a thing out. I “go to playtesting” when I need confirmation that what I’m designing will, in fact, work, before I move on to creating dependent systems.

It starts very small, just testing tiny system interactions: “hey Meg, pretend you’re trying to push me off a wall. Roll these dice. Say I dunno, something about pushing me off the wall, okay? … Huh. Okay, thanks!”

Or “hey Meg, you’re creating a character, a dragon slayer. I’m going to ask you a couple of questions, choose the answers from these lists here, okay? … Huh. Okay, thanks!”

When I have a complete subsystem I need to test, that’s when I ask friends to sit down with me for half an hour and give it a try. This was the case with The Dragon – I needed to test character creation, to (a) make sure that it made interesting characters, (b) see whether it gave me everything I needed as GM to launch into play, and (c) if not (which I expected), clarify what it was missing, so I could go about creating it with some experience to build on.

Eventually there’s no way to see what works and what doesn’t without just sitting down to play the game, so that’s what you do.

I’ve been re-reading the clouds and dice series on It’s a series of posts about how rules refer to fictional material and to real-world cues (like dice and character sheets). It’s a lot of material but I’d strongly recommend reading it if you’re interested in writing game rules that encourage good fiction.

I was reading part 6 – “Now where WAS I…”, especially the comment thread, when a couple of exciting questions hit me.

To summarise the context:

  • Mechanical effects that emerge from the fiction (“if you have the high ground [fiction], you get a +2 bonus [mechanics]”) help encourage good fiction because they mean the fiction matters – players have to create detailed fiction in order to use the mechanics effectively.
  • But being able to invent your own fictional bonus-producers whenever you want (“err, so, there’s probably a table, right? I jump on it for +2 high ground bonus”) can result in fictional details that are trivial and transient – only there for the bonus, not for the lasting effect – so that in fact the fiction no longer matters after all. (This is an effect I see – and dislike – in Dogs in the Vineyard, sometimes, although I think it is deliberate there in order to encourage escalation for bonuses. But I digress.)

So, ideally, we’d prevent players from creating fictional details on-the-fly just to get a bonus. My questions are:

  1. Can we enforce a system where players have to establish something a few “turns” before they use it?
  2. More interestingly, can we enforce that bonuses can only be taken for statements other people have made?

Note that both these questions are pre-empted by Vincent’s suggested mechanics in comment #43 on that thread (which I read after having the above light-bulb, but you’ll have to take my word for it). What I’m really wondering is can we make better mechanics than those somewhat simplistic teaching examples. In comment #46, Moreno suggests that point 2 is in fact mandated in Spione – I’d better check that out. I wonder if any other games do anything similar?

I think there’s a risk that unless your fictional detail is specifically angled toward someone else getting a bonus for it (e.g. “I provide covering fire”), then it may not be that usable for mechanical effects in most cases. Perhaps I’m being narrow-minded, though – I need to think about it some more. I think a bigger risk is that if such details can be used to get mechanical bonuses against you, then you might be encouraged not to provide them, which would be totally counter-productive. Hmm.

However, I’m particularly hot for Jesse Burneko’s comment #55 – it’s all about communication. By using such mechanics, players are encouraged to communicate effectively – both providing fictional input that has meaningful implications, and considering the implications of the fictional input provided by others.

I think we can rely on players to pay attention to others’ input in order to trigger mechanics (e.g. bonuses). My concern, though, is that it may not reliably get players to give that input in the first place. What encourages that, beside the hope that others will reciprocate for you?