143. Beyond Success & Failure: Failing Forward and moves in Apocalypse World

I want to talk about the idea of failing forward as its used when people talk about Apocalypse World and why I think its misapplied.

Failing forward apparently entered public discourse in 2007 by John C. Maxwell in his book by the same title. Maxwell uses the phrase to outline ways that people can utilize their shortcomings and mistakes as a foundation for moving forward and achieving success rather than letting them function as impediments.

When applied to RPGs, the phrase generally refers to mechanics that prevent failed dice rolls (or whatever randomizer the game uses) from stalling the progress of the characters and the narrative. Apocalypse World is designed to allow forward movement of the story no matter what a player rolls with her two d6’s and is thus said to let characters fail forward.

I agree that AW propels the story forward no matter what result the dice show, but thinking of it as failing forward inaccurately describes how moves work and what they do to make misses as productive as hits.

RPGs with randomized resolution mechanics are generally structured to have some form of freeplay that takes the players to a point of uncertainty that the rules then let the dice resolve. This feature has its roots of course in RPG’s wargaming past. My army wants to fire upon your army, so we consult the rules, take the necessary factors into consideration, arrive at an appropriate probability of success, and throw the dice to see if my soldiers are having a good day. In early versions of D&D, the same type of questions were resolved in the same type of way. Do I hit? Can I find the secret door? Can I detect the trap? Safely disarm the trap? Detect the approaching wandering monster? Make the jump?

In these games of strategy and accomplishment, these questions of success or failure are the key dramatic questions. Everything important to the play and experience we are sharing hinges upon that question of whether I succeed or fail.

As RPGs have evolved, the centrality of success and failure has held fast. Even in games that want to focus on narrative and story, the assumption is that the dramatic question the dice should answer is that of success and failure remains.

For example, Hero Wars, Robin Laws’s take on Glorantha published in 2000, has all kinds of rules and features that work to make gameplay resemble cinematic and adventure narratives, but for all the ingenuity of design, the point that the fiction stops and the dice come out is when a character has attempted to do something and we need to know if they succeed in their action or not.

The core mechanic of the game is a d20 roll against a skill, trait, or quality that the character has. Rather than the roll resulting in a simple success or failure, Laws created a range of 8 possible results from from “complete success” to “complete failure,” moving through major, minor, and marginal levels at each end of the spectrum. It’s a neat feature, but even with this expanded set of possibilities, the dice still only tell us if an action is successful or not.

The limitation of the success/failure binary is not that it halts forward motion (though it certainly can if not carefully managed) but that it becomes an uninteresting question (or at least a less interesting question) when the heart of the matter isn’t a question of winning, as in the matter of a wargame, or survival, as in the matter of a dungeon crawl, for example.

For me, this is made no place more plain than in the examples throughout the Hero Wars text. Here’s a taste:

Haral (Carpentry 5W) is commissioned to build a chest. Natural resistance is 14 (in this case representing the suitability of the wood, complexity of the task, etc.) (121)

Checking to see who wakes up first is a simple ability test based on Light Sleeper, Nervous, or similar abilities. (126)

Testing to see if your hero knows the name of the king of a far-away land is an ability test using Foreign Knowledge or Well Traveled, but probably will suffer a penalty for the distance between the hero’s land and the foreign kingdom (126)

All three of these examples create a question where there is no guarantee of dramatic tension. Building a chest? Waking up first on the basis that you’re a light sleeper? Do you know the name of a king? Each of these could possibly serve as an important question in the game, but they are just as likely to be mere moments that dice are rolled and success or failure is achieved without impacting play itself. The problem isn’t that failing to make the chest stops Haral from doing something, it’s that that isn’t a very interesting question. We want to know what happens once the chest is built; does it please the noble woman for whom he built it, and does she grant him a favor? Once the characters are awake, what happens? What does the character do with the name of the king? The questions that follow the answers are where the dramatic tension lies, not in the answering of the questions. But when characters are structured around abilities and rolls are centered around success and failure in putting those abilities into practice, the game naturally creates these points of question and answer that are ho-hum.

The focus on success and failure also leads the game’s rules to care about the likelihood of success and failure, creating all kinds of bonuses and penalties to the roll that add to the math of the moment but not the drama or importance.

Rollo is in Prax, attempting to ride a bison. Kathy assigns a [minus] 5 penalty on his Ride ability, since he normally rides a horse. (123)

Rollo is climbing a tree with plenty of handholds. Kathy gives him a +4 bonus. (123)

Riding bisons and climbing trees are cool, but do you really want to roll to see if your character can climb a tree? Is that the best dramatic question your game can answer? Does the addition of bonuses and penalties make the players lean over the table to see the dice result?

Rollo is sneaking over the wall into town late one night. The wall is rated at 20 resistance: While it is tall, it has numerous rough handholds. Rollo’s player declares that he is using Rollo’s Climb Walls ability (by now a respectable 15W), Rollo’s player throws a 14, a success, which is bumped up by his mastery to a critical success. Kathy rolls for the wall; a 12 – a success. Comparing Rollo’s critical success to the wall’s success, we see that Rollo wins a minor victory, easily climbing the wall. (128)

Rollo tries to sweet-talk his way into the heart (or at least the arms) of a pretty girl. His Fast Talk is 12W; her Chaste is 16W. . . . They both roll and both fail. Rollo’s roll of 15 is lower than her 19, so his line is good enough to interest her, though not totally (121)

These scenarios are the natural offshoot of basing a game’s dramatic question on the issue of success and failure. Rollo’s scaling the wall into town takes up a huge chunk of play (and math!), all to find out that he “easily climb[s] the wall.” Let Rollo climb that wall easily anyway, because the exciting shit is what happens after he gets over that wall. If he fails the roll, then the exciting moment isn’t heightened, just pointlessly delayed. Similarly, we care more about the interplay between Rollo and his love interest than the math of his sweet-talking. The roll will help the GM and the player figure out how to play the scene somewhat, but the dice delay the exciting moment, not improve it.

These examples and their question of success or failure can, of course, impact the story significantly, but it’s not reliably the case. And these examples are from a game that openly cares more about cinematic storytelling than other more traditional RPGs.

Moreover, when a game focuses on “story” rather than a mere sequence of events, like a classic dungeon crawl, the issue of failure becomes stickier, because failure could derail the story. We see this often in investigative RPGs, but the problem is certainly not limited to those kinds of games. Letting the dice decide success and failure (and only success and failure) means that sooner or later the rules will back everyone into a corner. The two major workarounds to this problem are the automatic success and bennies. The first lets the GM declare that the players don’t need to roll at this moment (even if they’ve rolled for similar moments), and the latter lets the player ignore, modify, or replace the die roll to ensure success when failure will seize up the game. Both solutions are designed to undercut the dice rather than support them.

Now, back to Apocalypse World.

When people use “failing forward” to describe how a 6- result works in AW, or “success with a complication” to describe the 7-9 result, we are seeing the prominence of the success/failure binary in the minds of roleplayers. The dramatic question posed by a move in Apocalypse World is not about success or failure. In all the barter moves, for example, (see posts 136-138), your character is going to get what she seeks. The dice don’t decide success or failure; they decide how much you have to offer, risk, or endure to get what you seek. That’s the dramatic question posed by the move and resolved by the dice, and it is what makes us lean over the table to see what the dice say, knowing that whatever the answer, it’s going to be exciting and rewarding. We saw this in our discussion of going aggro, that the dice do not determine if you hit or not, but how ignorable you are to the target of the move. Reading a sitch will let you ask one question no matter what you roll. The question is how observant are you and how much will you learn.

Moves are narrative pivot points, those moments of dramatic action that can send the narrative spinning off in a number of structured directions. The designers decide what these pivotal moments are when they create the triggers for the moves. They decide what directions the narrative can go when they create the results of the move. It’s then up the MC to take that general direction and turn it into specific fiction that carries the story forward.

Apocalypse World doesn’t need or want rules for “automatic successes” because the trigger for the move is the trigger for the move: if you do it, you do it. There is no way to side-step a dramatic moment if your character does what the game considers a dramatic thing. Nor does the game want or need bennies because the move is going to take you some place exciting no matter what you roll.

In a 2008 post on anyway called “Rules Vs. Vigorous Creative Agreement,” Vincent says: “As far as I'm concerned, the purpose of an rpg's rules is to create the unwelcome and the unwanted in the game's fiction. The reason to play by rules is because you want the unwelcome and the unwanted - you want things that no vigorous creative agreement would ever create. And it's not that you want one person's wanted, welcome vision to win out over another's - that's weak sauce. . . . No, what you want are outcomes that upset every single person at the table. You want things that if you hadn't agreed to abide by the rules' results, you would reject.” (http://www.lumpley.com/index.php/anyway/thread/360)

Yeah, that’s what moves in Apocalypse World give you. You have to abide by the roll of the dice and make your moves as players and MCs when the rules tell you to. And to make you abide by the roll, the game has to make it rewarding and fun to abide by them, which is what moves are designed to do.

Of course, the solution that moves provide are not for every game. Apocalypse World uses moves to create a chain of charged situations to keep things moving and evolving without requiring the MC to plan out anything on a grand scale; in fact, the game’s rules mandate that the MC not plan out anything because the moves will make those decisions for her.

AW wants narrative pivot points built into the moves of the game. In classic D&D, pivot points are built into the dungeon itself, since each room or encounter provides a situation to which the characters react. In such a game, success and failure are the exact thing that the dice need to decide to make that game exciting. In Hero Wars, as becomes clear when you read the published campaign material, the pivot points are supplied by the metaplot of the narrative, and in turn the GM, who has to translate the successes and failures of rolls into narrative thrust and to provide fresh pressure on the characters.

So, it’s not failing forward, and it’s not success with a complication, that happens during a move, because moves don’t ask if an action is a success or a failure; it asks what can happen within the narrative when a character takes such an action. Moves, more than any other innovation in RPGs that I can think of, break the primacy of success and failure as the central question addressed by dice when a character takes action. But of course, I have a limited understanding, so if you can think of other mechanics and other games that create interesting questions for the dice to answer, I’d love to hear about them.