Once upon a time, I summarised the difference between most theatre-sports/improvisational games/performances and RPGs thusly: “In improv, everything you say is true, but there are strict rules about who can say what, and when, whereas in gaming, everyone can talk whenever, but there are strict rules about whether anything is true.” I really like that. It’s not 100% true but the places where it isn’t are interesting enough to make it useful. I hope somebody someone quotes me on it.
It’s certainly a definition that gets to the heart of what RPGs do: mostly the game mechanics centre around answering a question of “did that happen?”. It’s been reworded in some interesting ways such as Wushu saying “it definitely happened, but it did it produce narrative scene closure”, or Hillfolk making it “do you get what you want by saying it” or Gumshoe making it “did that kind of success help or not?”. I once wrote a game where you always failed (it was a comedy game) but you rolled to see what character trait caused you to fail. It never quite worked though.
Shifting this question is hard because it’s a question so fundamental to RPGs. But on the other hand, it’s a question that is alien to the kind of narratives we want to create. Indeed, I’ve often been in the position as a writer where I’ve spent so long as a GM I have no idea whether the character I’m writing will succeed or not in the story because I’m waiting for a die roll to tell me, and when you get to that point, you have not written a good story at all, because you’re not making the right choices. Likewise, we all know at times we try too hard to make a story work and ignore successes or failures – or sometimes we accept successes and failures so much engagement or suspension of belief collapses, and it’s the fine art of GMing to negotiate those areas.
But I’m not here today to mess with failure, but to ask what failure means. It’s one thing a lot of RPGs forgot to define until recently. Does the system always determine the difference between no-and-things-get-worse and no-but-things-aren’t-so-bad, and do you have the narrative instincts to make sure you don’t accidentally set up situations where a no-and automatically happens even if you roll a no-but? For example, if the monster is reaching out to grab you and you fail to dodge, does that mean you are grabbed or you are grabbed and pulled towards its mouth? And does that happen if you fail to attack? ie does not succeeding at an attack also (as in things like Apocalypse World) mean the bad guy gets a shot in, or is that two separate narrative beats? And when you fail, do you lose a resource? Are you now tangled in the tentacle and in need of rolling to escape or is it HARDER to escape now you are tangled? If you fail to escape this time can you just keep rolling until you do, or are you bound to suffer? How much has the situation changed? How much agency remains for you?
It may read like I’m splitting hairs but these kinds of questions are the atoms of RPGs. They are the fundamental building blocks. And the reason we keep reverting back to games which measure strict simulation-esque blocks of time and actions available in that time is to answer these questions simply and easily. Because even breaking things down into no-ands vs no-buts is not simple and clear because it’s hard to get everyone at the table to see what those things are in every situation unless you end up breaking things down anyway into “if this action happens or does not, in this time frame, then this or this”.
I’m thinking about these questions because I’m finishing up an RPG which is all based around a simple mechanic of success/failure at a scene level, which means this is vague. As written, if you succeed on a random test (modified by simulated characteristics), you succeed at what you were trying to do, typically an atomic action broken down by the GM (eg escape a tentacle). If you fail, you fail, but you can acquire “damage” to succeed. Every time you take damage, though, you risk being taken out of the game. Thus, everything in this game tilts, I now realise, on what “failure” means. Is it being grabbed by the tentacle, or being pulled into the screaming chomping maw? I’ve left that hard decision to the GM.
I’m also trying to decide if I also have a hit point mechanic. Obviously, I kind of already DO have one, where you take damage to get successes. Adding a hit point mechanic allows a concrete measure of what failure looks like. In some situations, failure will mean damage. That provides the GM with a great resource. In situations where there doesn’t seem to be a logical way to make failure particularly meaningful or threatening, damage can be such a thing. But I also realise it reveals how there is no support for any other such failures. When the characters try to pick a lock, what does failure mean there? Does it mean try again immediately, but other things in other scenes become worse as time passes? Does it mean get caught by the guards and put in prison with much stronger locks? Does it mean you can never try again, find another door (or if there were fumbles, your lock picks break, never try picking locks) and is that kind of “buffet of story pathways” actually fun? Does it matter if the rogue picks the lock or the fighter smashes through the wall if the goal was to get into the room – did the rogue’s failure mean anything significant? Significant enough to take damage? Or was it just randomly choosing the style of victory?
The truth is a LOT of rpgs dodge these questions, really. Or they do like I do and provide different types of situations, ones where sometimes failures can be re-rolled (like swinging a sword at an enemy) and ones where it can’t (picking the lock), or ones where some rolls (or some results) can result in worsening situations or reduced resources and some can’t. Botches and crits fill some of this void as well, and I’ve got those in place. But without anything to lose except story, will people sacrifice themselves to avoid bad outcomes, since – and this is the lynchpin of this whole piece – MOST OF THE TIME YOU CAN JUST OUTTHINK A BAD OUTCOME?
Does failure mean something if you don’t lose hit points, since you can always roll again or find another way around? Some of it is undeniably psychological, even if every orc misses every time, players who miss five swings in a row will kill to hit the sixth time. Maybe if lockpicks fail and smashing down the door fails and digging under the door fails, then yes, tapping people out of their creative buffet of solutions is a price they would grow weary of and spend points to get success. And yes, maybe I need to just keep road-testing my mechanics. I know THAT. Don’t say that.
But DO say what do you think about the question. How do you feel about failure and what it means? How does your favourite game do it? What are some interesting examples you’ve run into, particularly ones where failure didn’t seem to slow anything down at all, or failure derailed everything, or failure was poorly defined. This is a “not sure what to do with my game” question AND a game philosophy question. Of course, the former almost always tend to be the latter, if you’re doing it right…
And if you’re interested in the game in question, you can get it be signing up to my new patreon account.