On RPGs and failure
2015-Jul-15, Wednesday 17:25![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
dorchadas
This post is prompted by this post about how to be a better player, which I have several disagreements with mostly due to having a different gaming philosophy--nowadays I don't think of RPGs as sitting down to tell a story, I think of them as sitting down to play a game and the story is how how you describe the game to people after it's over--but I'm not going to talk about those because they're mostly a matter of taste! I'm going to talk about failure from the GM's side of the table.
The article says:
The first is the humble stealth roll, staple of RPGs the world over. One thing I used to do is have multiple rolls for sneaking in--one per guard, one per area, what have you--but this is setting the PCs up for failure. Even with one PC with a 95% chance to succeed, if they have to roll for each of ten guards, that's only a 60% chance to make it past all of them. Add in penalties for circumstances or two or more PCs who each have to make all the rolls and you're setting the PCs up for failure.
Rule One: Don't Set the PCs up for Failure
I ran headlong into this problem in the past, but nowadays I tend to deal with these situations in two ways. The first is by treating "sneaking into the area" as a single task, so it's not just a situation where I call for rolls until someone fails, and the second is by using the cooperative rules so that a single low-skill compatriot makes the task harder but doesn't mean that it's futile to even try. I also tend to stick to a Three Strikes rule. Much like in computer stealth games, as long as you don't run out and shout at the guards, a failed roll isn't going to immediately lead to a full-auto gunfight, it will lead to a guard saying, "Huh, what's that noise?" and looking around a bit. Further failures result in more alert guards and eventually the giant ❗️ appears over the guards' heads and they attack, but successes get things back on track.
It's like combat. Generally, the GM won't take the player aside and inform them that a sniper shot them from five kilometers away and now they're dead, so sad, better roll up a new character, because while that's the kind of thing that can happen and it is indeed realistic, it's not fun at all. That kind of all-or-nothing scope to the results just sets players up to do literally everything to avoid failure, which leads me to my next point:
Rule Two: Don't Let It All Come Down to a Single Roll
There's a reason RPG combat is almost never just an opposed roll where the loser dies. Partially because it's no fun and the G does stand for game, but partially because those kind of consequences are widely viewed as unacceptable for a single roll. GMs should really take that attitude and apply it to other areas of gamemastery as well. For example, the investigation roll. Call of Cthulhu is so notorious for adventures getting completely derailed due to a failed Library Use roll that an entire RPG (Trail of Cthulhu) was written to solve this problem. If there is a roll where the consequences of failure are "Everyone does nothing because there is nothing to do," don't require that roll.
A brief aside here. I tend to roll my eyes at articles that talk about improv theatre and not saying no in RPGs because these are different media with different aims. For example, if the players fail a lock picking roll, there's no need to make the alarm go off or the guards show up or please anything happen just because it stops the action. Unlike improv theatre, generally the players have plenty of pre-existing knowledge of the world to draw on. If they can't pick the lock, maybe they can try getting the captain of the guard drunk and stealing her keys. Maybe they can consult with a witch in the town and find out a secret passage. Maybe they can disguise themselves as servants and sneak in. And the flip side is that a GM has to be open to these additional solutions. It's okay to say no to individual courses of action, or to let the RNG shut them down, but not to say no to everything except one course of action.
And now, failure in combat:
Rule Three: Not All Fights Are Fights to the Death
Despite what video games have told all of us, animals do not generally attack on sight and fight until slain. Intelligent enemies will fight until wounded before either surrendering or throwing down arms and running. Fighting to the death is not common and generally requires instinctual overrides like parents defending children or lots of training. In real life, fights generally go until one person is moderately hurt and then stop abruptly.
As a GM, it's incredibly important to realize this. Have enemies run. Have enemies surrender. Show through NPC attitudes that murdering prisoners and killing literally everyone who ever dares raise a hand against you is the mark of tyrants and is going to raise eyebrows at the very least. This can be hard to overcome, both because video games have trained us that all combats are to the death of one side or the other and because a lot of players have the attitude that killing all enemies prevents them from returning to haunt the PCs later. Which is fair enough, but the solution here is probably to talk to the players about it. If they know that losing a single battle is not automatically a TPK, they won't be afraid to be taken prisoner.
In some ways, this fits under Rule One, too. Even if the party has a 95% chance to win every battle, given enough battles...
And, the final rule I have for dealing with failure:
Rule Four: Set Appropriate Stakes
If the party is fighting to SAVE THE WORLD, of course they're going to be deathly afraid of failing because the world is where they keep all of their stuff. If they're fighting to SAVE THIS TOWN, then that still adds plenty of space for things to go wrong. This also ties in with the other rules and is in some way the ur-rule. Don't make the stakes on a stealth roll "You sneak past this one guard" vs "You fail the mission." Don't make the stakes in every fight "TPK."
I ran a Delta Green game for three years, and I'd say the PCs probably failed half the operas they went on. But in some cases, they pulled out a partial success--sure the whole town of Groversville is dead, but at least we eliminated the alien experiments on the water supply--and in some cases the interesting part became how they explained their actions as a success to their superiors even when they considered it a failure. And by letting them do that, I encourage player creativity while also enforcing the theme that too much secrecy is morally corrosive even if you're doing it for the right reasons. Win win.
Failure is what you make of it, but it's important that the GM give the PCs enough that they can make something of it.
The article says:
We need to view failures as setbacks and explain why our character didn’t achieve their goal, and we need to understand that failure is not the end of the worldwhich is good advice. If you're playing a game where randomization is part of the means of determining actions, sometimes the RNG won't go your way and you have to deal with that. But it's also important for the GM not to make too many rolls a make-or-break situation either. Here's a couple common examples I've seen and what I did with them.
The first is the humble stealth roll, staple of RPGs the world over. One thing I used to do is have multiple rolls for sneaking in--one per guard, one per area, what have you--but this is setting the PCs up for failure. Even with one PC with a 95% chance to succeed, if they have to roll for each of ten guards, that's only a 60% chance to make it past all of them. Add in penalties for circumstances or two or more PCs who each have to make all the rolls and you're setting the PCs up for failure.
Rule One: Don't Set the PCs up for Failure
I ran headlong into this problem in the past, but nowadays I tend to deal with these situations in two ways. The first is by treating "sneaking into the area" as a single task, so it's not just a situation where I call for rolls until someone fails, and the second is by using the cooperative rules so that a single low-skill compatriot makes the task harder but doesn't mean that it's futile to even try. I also tend to stick to a Three Strikes rule. Much like in computer stealth games, as long as you don't run out and shout at the guards, a failed roll isn't going to immediately lead to a full-auto gunfight, it will lead to a guard saying, "Huh, what's that noise?" and looking around a bit. Further failures result in more alert guards and eventually the giant ❗️ appears over the guards' heads and they attack, but successes get things back on track.
It's like combat. Generally, the GM won't take the player aside and inform them that a sniper shot them from five kilometers away and now they're dead, so sad, better roll up a new character, because while that's the kind of thing that can happen and it is indeed realistic, it's not fun at all. That kind of all-or-nothing scope to the results just sets players up to do literally everything to avoid failure, which leads me to my next point:
Rule Two: Don't Let It All Come Down to a Single Roll
There's a reason RPG combat is almost never just an opposed roll where the loser dies. Partially because it's no fun and the G does stand for game, but partially because those kind of consequences are widely viewed as unacceptable for a single roll. GMs should really take that attitude and apply it to other areas of gamemastery as well. For example, the investigation roll. Call of Cthulhu is so notorious for adventures getting completely derailed due to a failed Library Use roll that an entire RPG (Trail of Cthulhu) was written to solve this problem. If there is a roll where the consequences of failure are "Everyone does nothing because there is nothing to do," don't require that roll.
A brief aside here. I tend to roll my eyes at articles that talk about improv theatre and not saying no in RPGs because these are different media with different aims. For example, if the players fail a lock picking roll, there's no need to make the alarm go off or the guards show up or please anything happen just because it stops the action. Unlike improv theatre, generally the players have plenty of pre-existing knowledge of the world to draw on. If they can't pick the lock, maybe they can try getting the captain of the guard drunk and stealing her keys. Maybe they can consult with a witch in the town and find out a secret passage. Maybe they can disguise themselves as servants and sneak in. And the flip side is that a GM has to be open to these additional solutions. It's okay to say no to individual courses of action, or to let the RNG shut them down, but not to say no to everything except one course of action.
And now, failure in combat:
Rule Three: Not All Fights Are Fights to the Death
Despite what video games have told all of us, animals do not generally attack on sight and fight until slain. Intelligent enemies will fight until wounded before either surrendering or throwing down arms and running. Fighting to the death is not common and generally requires instinctual overrides like parents defending children or lots of training. In real life, fights generally go until one person is moderately hurt and then stop abruptly.
As a GM, it's incredibly important to realize this. Have enemies run. Have enemies surrender. Show through NPC attitudes that murdering prisoners and killing literally everyone who ever dares raise a hand against you is the mark of tyrants and is going to raise eyebrows at the very least. This can be hard to overcome, both because video games have trained us that all combats are to the death of one side or the other and because a lot of players have the attitude that killing all enemies prevents them from returning to haunt the PCs later. Which is fair enough, but the solution here is probably to talk to the players about it. If they know that losing a single battle is not automatically a TPK, they won't be afraid to be taken prisoner.
In some ways, this fits under Rule One, too. Even if the party has a 95% chance to win every battle, given enough battles...
And, the final rule I have for dealing with failure:
Rule Four: Set Appropriate Stakes
If the party is fighting to SAVE THE WORLD, of course they're going to be deathly afraid of failing because the world is where they keep all of their stuff. If they're fighting to SAVE THIS TOWN, then that still adds plenty of space for things to go wrong. This also ties in with the other rules and is in some way the ur-rule. Don't make the stakes on a stealth roll "You sneak past this one guard" vs "You fail the mission." Don't make the stakes in every fight "TPK."
I ran a Delta Green game for three years, and I'd say the PCs probably failed half the operas they went on. But in some cases, they pulled out a partial success--sure the whole town of Groversville is dead, but at least we eliminated the alien experiments on the water supply--and in some cases the interesting part became how they explained their actions as a success to their superiors even when they considered it a failure. And by letting them do that, I encourage player creativity while also enforcing the theme that too much secrecy is morally corrosive even if you're doing it for the right reasons. Win win.
Failure is what you make of it, but it's important that the GM give the PCs enough that they can make something of it.
