Tunnel of Thoughts is where we mull over ideas and concepts from our RPG games. This time we’ll be looking at failure in RPGs. This comes on the back of writing a product featuring a good bit of investigation and reaching a section in our campaign where the party is seeking information over conflict. We’ll be considering how failure works in most D20-based systems, and how to handle it the table, and during preparation through quest design.
So, in most D20-based systems a player picks up their D20 and rolls it, and failure is a very real possibility. You’re most likely rolling against a static number that dictates whether you succeed or not. In some cases, you might be making an opposed roll where the GM rolls on behalf of what you’re tackling. Generally, things are pass/fail in a binary fashion. Some systems/GMs create degrees of failure to checks, if you fail by more than 5, or succeed by more than 5 etc. This should all be familiar territory.
Let’s do a quick example, using a barkeep who has some key information to solve the quest. The party want to know what some thugs who came into the tavern last week looked like. I won’t be commenting on designing a quest in this way until later. The party choose the fighter because he has a +7 on this kind of skill check. The fighter walks up, rolls diplomacy/persuasion and rolls a 2, with his +6 bonus that gives him and 8. The GM has set the DC of the check at 12 because they wanted it to be easy. The barkeep doesn’t give the fighter the information he needs and the party is stuck.
This is a terrible outcome. The failure served no purpose except to stop the story dead in its tracks. This is a totally non-dramatic failure. The story is not advanced, there is no real consequence of import for anyone.
There are two issues at work creating this non-dramatic failure. The first is that the GM took the die result as an indicator of failure of outcome, rather than failure of method. The other is that the adventure was designed with a gated check to allow the players to get essential information.
Die rolls as cues
We’ll deal with the die roll first. Whenever someone rolls and a number comes up that represents failure or success, this is a cue. It doesn’t have to be a hard and fast dictation of failure. The ultimate fate of the roll lies with the description of outcome given by the GM or perhaps the player.
Let’s consider our previous example. The fighter walks up to the bar and rolls his 2.
Player: “I accidentally knock over a drink on the bar and spray the barman with beer. I start apologising, and offer to buy a round of drinks.”
Barkeep(GM): “That’s ok, you mean well and you’re nothing compared to these three louts who in here last week.”
Player: “I slide over a couple of silver pieces to pay for the spilt drinks, and ask what was so bad about these thugs.”
Barkeep: “Three of there were, looked like they’d been sleeping rough for a while. Big lads, one of them had a big scar on his face, and another had his arm in a sling. Came in with a bag of cash and tried to drink us dry. Started a few fights and smashed some stuff up”
This outcome gives us a valid outcome for the low roll. The PC embarrassed themselves, had to spend a few coins on dealing with their mistake, and it was amusing. The interaction was important and they needed the information to allow the party to progress, so they got what they needed. It was just a small moment of roleplay where the die roll acted as a cue.
How often you use this type of cue to allow a success of sorts even when failure occurs is up to you in your game. The players may or may not know whether each roll is a real pass/fail roll or a succeed/succeed with caveats roll.
Gated difficulty in design
This business about cues is all well and good, and can result in a great opportunity for people to do things that are in character, grow some amusing anecdotes about what happened in their game, and keep the story moving.
Another question in the example given is “why the hell was it designed like that anyway?”. Here, the answer is of course, as an example. This doesn’t stop many adventures or GMs having skill checks where the whole adventure can end in non-dramatic failure.
Why did the party need to ask what the thugs looked like? If they need a piece of information, it should probably find its way to them almost regardless of what they do so the story can move forward. When they enter the tavern, the barkeep could be moaning loudly about the thugs who smashed up tow of his tables last week. He could come over and ask the party if they’ve seen these thugs. He could offer free beer in exchange for more information about who they are. In this instance, there is simply no need to hide the information behind a skill check gate if the party need it.
This doesn’t hold true in every case, of course. Let’s say our party find a scroll belonging to a secret society they’re trying to unmask. Secret societies generally want to be, well, secret so it is written in some sort of code the party can’t readily read. The scroll gets passed to the wizard, who’s a bit handy with linguistics and ancient writings, but he fails his skill check. The gated difficulty seems necessary to the story, but failure also results in a dead end.
The solution to this is once again, to add caveats. Here’s two examples that could work in this case.
One, the wizard can’t break the code, he’s totally stumped. He does however, remember an old friend from his wizard college days who was excellent at writing and breaking codes, and he believes he could do it. He knows that he’s currently in the employ of mad king whatshisname in the thingummabob forest. Off the players go on a side quest. Instead of just opening the skill check gate and moving along the main story, they go around the gate.
Two, the wizard decodes it, but he gets some vital detail wrong. Let’s say the scroll contains details of a meeting, where, when and who. The party believe they’re going to show up and rumble the meeting, but they find it has already finished, and only one member of the society, a junior acolyte is there, cleaning up. Instead of getting the drop on the whole society, they get a new opportunity to learn more about their inner workings. The story movies forward, but there is a consequence to the failure that the players may not even be able to connect back to that skill roll.
Why roll at all?
At this point, some people are no doubt asking this question. Rightly so. When the fighter asks the barkeep for pivotal information, don’t roll, just give it to them. When the wizard decodes a key scroll, just let him do it. The rogue opens the safe when it really matters and the cleric is able to heal that NPC who matters. This is fine, if you want to do that, go with it.
The issue with this is it totally gives away what is being gifted to the players. It can come across as the GM removing the players’ agency. They don’t have a choice in how things progress, they don’t get to roleplay as much, they don’t get to feel there is a risk to what they do. A huge part of roleplaying is that the random dice rolls shape things and create outcomes we wouldn’t arrive at without them.
This has been about D20 systems, but many other systems handle failure in very different ways. There’s a lot of stuff out there, and we’re going to make two recommendations if you want to read more.
Cthulhu Dark has a free preview PDF which gives you all the rules for the system, and some starter materials. Failure barely exists, and it’s a really interesting take on how to do skill checks etc that encourages roleplaying and storytelling. Look out for a follow-up post describing a house rule for D20 systems based on how Cthulhu Dark handles failure.
Evil Hat’s implementation of the Fate system is pay what you want on DriveThru, and is well-known for doing things differently in terms of failure. Well-worth checking out if you don’t know it already.
We’d love to know your thoughts on this topic in the comments. You can follow us on Twitter or Facebook if you’re keen for more. If you’d like to support us to keep writing, head on over to DriveThruRPG and throw some money at us.