Failure is Critical

Feared in myth and legend!

D&D is a game of wonderful absurdity. I mean, somebody looked at a blank sheet of paper and thought, "You know what would be cool? A floating eye. But like, with more eyes around it. And a mouth. Heck, a whole body! But no legs. Or arms. Or torsos."

However, our conception of reality is deeply tied to this fantasy universe. We expect wizards to be wise, barbarians to be strong, fire to burn, etc. And sometimes the game fails to meet our expectations.

This usually happens when a character makes a skill check they should be good at, but fails miserably. A fighter kicks at a wooden door but doesn't open it. A wizard examines a glowing glyph but doesn't pick up what it is, despite it being a common spell. A rogue fails to pick the lock on a child's toy chest.

Most times, this can be easily mitigated by the DM properly adjudicating actions. If it's a weak wooden door, there's not really a possibility for failure, so no roll should be needed. But in some cases (such as if the party is in a hurry or there are enemies on the other side of the door who would be notified if the door isn't broken down in one fell kick), the roll is still needed.

So how do we handle failure when our concept of the game's reality gets in the way? Well, there's a few ways.

Failing

First, though, let's talk about where these rules don't apply.

In combat, a miss is a miss. Combat is chaotic and fast, and even the most seasoned veteran will occasionally fail to harm their target. Also, the consequences are already baked into the system: the enemy didn't get hurt. They will survive to attack you again.

This can also help us see there are two parts to every failure.

1. The OUTCOME: the PC's attack misses
2. The CONSEQUENCES: the enemy isn't harmed

This seems to be so logical and obvious that is isn't worth thinking about. But it is, and it can help us answer our problem of discrepancy between failure and reality.

First, let's look at the Outcome.

See, that just... no. That wouldn't happen.

Most rolls determine something about the character rolling the die. The die determines information about the action.

• Does my character see X?
• Does my character know X?
• Is my character suave enough to change the mind of X person?
• Is my character skilled enough to perform X task?
• Is my character skilled enough to hit X with their attack?

However, these rolls don't have to be all about the character. Instead, they can change information in the world.

• How well hidden is X?
• How obscure is the knowledge of X?
• How stubborn or insightful is X person?
• Are there complicating factors that make X more difficult?
• How good at dodging attacks is X?

This is a subtle distinction, but if we use it carefully, we can make the characters and the world more realistic. Let's look at a couple examples.

BARBARIAN: I kick down the door!

ACTION DM: Whoops, you failed. Your puny kick fails to open the door. Better try again!

WORLD DM: You failed! Despite your mighty kick, the door stands. And only now do you think to check the frame - of course! Adamantine! It's been hardily reinforced. 

Here, the DM chooses to make the roll determine information about the world, not the character. The barbarian still gets to feel strong, it's just a particularly solid door. Additionally, this solves the problem of everybody in the party giving it a try (and the STR 8 wizard invariably doing the task) . In its place, we now have a new scene: How do you get past an indestructible door? Here's an opportunity for some creative role-playing.

How about another example?

DM: The wolf snarls at your party, baring its fangs. If you venture too close, it looks like it will strike.

RANGER: I'd like to try and calm the wolf, so it can let us pass.

ACTION DM: Ooh, bad luck. You accidentally pull out some rotten fish instead of fresh! Now the wolf is even more angry.

WORLD DM: You cautiously approach the wolf, but as you do, you notice a litter of pups just behind a nearby tree. And at this distance, the origin of the scars on its muzzle are now clear: these are sword wounds. You doubt this mother wolf will back down.

Now, the DM is creating new information about an NPC (the wolf), instead of assuming the Ranger did something wrong. What Ranger would have never encountered a Wolf before? They would most certainly have great empathy for the animal, and revealing more about the creature's character helps build the Ranger's character. Now it's not just a mechanical choice - it's a moral one. Perhaps the Ranger will have a talk with the party's Fighter later about hunting wild game. Perhaps the Ranger will decide to attack anyway and take a wolf pup to raise as their own.

Notice that again, the DM is shutting down the "I try again" route - in the first DM's scenario, the player would just pull fresh fish out and give it another shot. In the second, it's clear that failure has a powerful outcome: this wolf won't ever trust humans.

Now, how does this help us with the problem of failure vs reality?

PHYSICS DOES NOT WORK THAT WAY

Well, let's talk about consequences. Every roll has consequences, which are defined as the story changes based on the roll. If a group of players chooses to cross a pit trap by building a rope bridge, then the monsters pursuing them will be able to use the bridge as well. That's consequences.

You might be thinking that successful rolls deserve good consequences, and failed rolls deserve bad consequences. Well, yes, for the most part. But sometimes you can mix and match outcome and consequence to create the right sense of reality for your game.

Note that this can help you define the amount of "gritty" or "fluffy"-ness in your game world. A world where there are often bad consequences for good outcomes is gritty. A world with good consequences for bad outcomes is "fluffy". This is important to keep in mind, because leaning too far in either direction can change the feel of the reality of your game world. It doesn't matter what reality you want to use, but it should be consistent.

So, in a situation where a character "shouldn't" fail but does, the best way to handle it is to allow them to succeed (good outcome) but add a penalty or complication (bad consequences).

Here are some ways to accomplish this:

• You succeeded, but you took a long time doing it. Random Encounter while you wait.
• You succeeded, but you made a lot more noise than expected. Random Encounter draws near!
• You succeeded, but you pulled a muscle or pricked your finger in the process. take damage!
• You succeeded for now, but the enemy is aware something strange is happening and are now on alert!
• You succeeded, but did so in a way that made an important person upset with you.

There is no limit to the amount of different consequences you can add to a success to make it a failure. For added DM fun, only make the consequences apparent much later. Carefully explain to the players that yes, they failed, but they did indeed complete the task. The dread in the air can be palpable.

Untrustworthy townsfolk, or EVERYONE IS VAMPIRES?

On a side note, this is why I am less partial to "Dungeon World"-esque systems. The dice you roll force the DM to add consequences even when they aren't needed, or prevent you from adding consequences if the dice show a "complete success" or "complete failure". Having the flexibility to add your own consequences gives the d20 system a powerful edge over those games.

This is also why I avoid "critical failure" and "critical success" tables. Let me decide the consequences, dang it!

This can also apply to combat scenarios. Not individual attack rolls: I discussed that above. But the overall combat. If the players fail to defeat their foes, it doesn't have to be a Total Party Kill (bad outcome). Instead, they could survive, but face consequences:

• The group realizes the monster won't move past a certain distance because its eggs/babies are nearby
• The monsters are looking for living hostages for their master or their own purposes
• The monsters are hungry and will simply snap up an unconscious NPC, retreating with the PC's ally in their jaws.

This all happens when the OUTCOME is bad (the PCs lost) but you don't want to make the CONSEQUENCES bad as well. The best part of this is that it often leads to a new goal for the party: escape captivity, or rescue their allies.

This is also a good way to make the game more difficult without the players fearing for their lives in every combat. Give the bad guys prisoners, so the players know for sure that they will become prisoners if they lose a fight. Give the monsters motivations, and let the players work with them. This also has the added effect of making fights with monsters who simply want the PCs dead even more frightening.

This also answers one of the common problems that people have with "railroading" campaigns: If we know the players are going to win, then what's the point?

The point is what they have to give up along the way. How the world changes based on their actions. Make failure hurt, but let them keep pushing forward. Give everything consequences, good and bad. Make the world a different place by the time their quest is finished.

I want to talk about one last thing: Critical failures.

Actually reasonable. Well done.

Like any failure, these abysmal rolls don't always have to have a bad outcome. But if you do deign to grant a good outcome to a critical failure, you need to increase the consequences by a significant factor.

BARBARIAN: I kick the door open.

DM: Ouch, critical failure! Well, the door is a weak wooden door, so you kick it open... but standing on the other side is General Vicious, the terrible leader of the Lich Lord's Army!

I actually like these situations. DMs who want to plan the exact nature of the world before they sit down at the table are cutting off a major part of tabletop role-playing: the element of randomness. Obviously, General Vicious isn't going to be standing behind every door, but the Barbarian spraining their ankle would have been an equally good consequence.

I think the best example of this, though is the simple RPG "All Outta Bubblegum". It's so simple, I'm not even going to explain the rules. You can read it at that link.

Essentially, though, "Failing" to do a normal task in AOBG is just like any other RPG - you get a bad outcome and some consequences, good or bad. But in our group, we play it with a home rule: Catastrophic Failure. That means any failure isn't just a bad outcome, it's also a SUCCESS on kicking ass. Being polite to someone means you risk punching their head off.

That's actually pretty analogous to a critical failure. It's not just a bad outcome for the player, it's a good outcome for the forces working against them. Whether that's monsters, traps, the universe, luck, fate, whatever - that's up to the DM.

Thanks for reading!