Fail Forward

Crew DM/Gaming Advice

Following on my Failure is always an Option post is this one about a concept I’ve seen thrown around when it comes to running an RPG. Some people classify it as a handicap of story-focused systems, others tout it as a great example of conflict resolution, and that  is ‘Fail Forward’. It has also been called other names like “Non-binary success”. The idea is basically when your players fail an action and you just say ‘You fail, nothing happens’ is basically bogging down the system with unnecessary events. There is nothing gained in the story by the character’s failure with no change in the current status quo as they just pick up the dice and roll again, trying to succeed this time. Instead, the idea should be to cause a tilt in the current player’s situation. A tilt comes from the platform/tilt model from Improv, where in every scene there is a stable platform, which someone begins to tilt and then comes to a resolution at a new balance. In other words someone sets something up, something is done to make things weird, and then something resolves the problem.

There have been some times where gaming systems will give some mechanic to help that, like when D&D introduced the idea of Taking 10 or 20 as eventually the player would roll it and there was no immediate problem with failing. However, there is so much more that can be done to make failures more than an inconvenience at the table to have to reroll. Another way to look at this advice on failing is, if there is no dramatic consequence for failure, don’t make the player roll. Trying to stay stealthy with guards all around is one thing, but in an empty building it is pointless as the most you are doing is building false tension as you cannot cash in on the paranoia of a failure meaning detection.

Fate, for example, is a roleplaying system that gives us some great building blocks to go off of with the concept of ‘Succeed at a Cost’ where the player is able to get part of the result they wanted but with a negative. Examples from the core rules are below, but the SRD goes into a lot more detail here.

A minor cost should complicate the PC’s life. Like the above suggestion, this focuses on using failure as a means to change up the situation a bit, rather than just negating whatever the PC wanted. Some suggestions:

  • Foreshadow some imminent peril. “The lock opens with a soft click, but the same can’t be said for the vault door. If they didn’t know you were here before, they sure do now.”
  • Introduce a new wrinkle. “Yes, the Guildmaster is able to put you in touch with a mage who can translate the withered tome—a guy named Berthold. You know him, actually, but the last time you saw him was years ago, when he caught you with his wife.”
  • Present the player with a tough choice. “You brace the collapsing ceiling long enough for two of the others to get through safely, but not the rest. Who’s it going to be?”
  • Place an aspect on the PC or the scene. “Somehow you manage to land on your feet, but with a Twisted Ankle as a souvenir.”
  • Give an NPC a boost. “Nikolai surprises you a bit by agreeing to your offer, but he does so with a wry smile that makes you uneasy. Clearly, Nikolai Has A Plan.”
  • Check one of the PC’s stress boxes. Careful with this one—it’s only a real cost if the PC’s likely to take more hits in the same scene. If you don’t think that’s going to happen, go with another choice.

A serious cost does more than complicate the PC’s life or promise something worse to come—it takes a serious and possibly irrevocable toll, right now.

One way you can do this is by taking a minor cost to the next level. Instead of suspecting that a guard heard them open the vault, a few guards burst in the room, weapons drawn. Instead of being merely cut off from their allies by a collapsing ceiling, one or more of those allies ends up buried in the debris. Instead of merely having to face an awkward situation with Berthold, he’s still angry and out for their blood.

Other options could include:

  • Reinforce the opposition. You might clear one of an NPC’s stress boxes, improve one of their skills by one step for the scene, or give them a new aspect with a free invocation.
  • Bring in new opposition or a new obstacle, such as additional enemies or a situation aspect that worsens the situation.
  • Delay success. The task at hand will take much longer than expected.
  • Give the PC a consequence that follows logically from the circumstances—mild if they have one available, moderate if they don’t.

If you’re stuck for just how serious a serious cost should be, you may want to use the margin of failure as a gauge. For instance, in the vault-opening example, above—the one where the guards hear the PC and burst in the room—if the player failed their Burglary roll by 1 or 2, the PCs outnumber the guards. Not a tough fight, but a fight nonetheless. If they failed it by 3 to 5, it’s an even match, one that’s likely to use up resources like fate points or consequences. But if they failed by 6 or more, they’re outnumbered and in real danger.

In essence, you are creating a scenario that keeps the story moving forward but with the player having to deal with some negative that will now make this task harder or have some sort of repercussions later. This keeps the players involved in the story while they try and overcome this new cost that occurred when they did not get a good enough success to have it go off without a hitch.

To determine options for what happens, usually it’ll be whatever best fits the story. Of course, you do want to have times where the player does fail and ends up in a new situation, such as how on the movies a criminal who gets caught and now has to make a deal with police to keep from going to jail. It becomes dramatic as the character now has to decide their loyalties, figure out how to lose the police so they don’t get caught doing anything illegal and so forth.

There are those who see fail forward is the PCs getting everything but with some possible bad thing happening. such as if the GM has them stopped by a door that the story needs the players to get by, then on a failure a plot device will have the door open such as someone coming charging through it to attack the PCs. It can be that if you run it that way. However, not all of these examples have to be focused on advancing the characters situation so much as making the scene more dramatic. A great example taken from Reddit is of Han Solo using the Fail Forward system and then even breaks down every roll Han would have made in the game. To quote that Reddit, the author states:

I think the concept of failing forward is one of the best things to happen to tabletop roleplaying in a long time. In essence, it means that when a player’s dice come up as a failure, the character is able to at least somewhat accomplish their proximate goal though they incur a substantial hardship to be dealt with. The concept was explained to me in terms of Han Solo’s choice of Lando Calrissian as someone to seek out for safety; technically, Han failed that roll, Bespin was a bad place to hide, but it led to a great and interesting story arc. Well, when you look at Han’s other (mis)adventures, you see the same pattern, and it reminds me of the Pixar rule:

“You admire a character for trying more than for their successes”

Remember, “rolling to succeed” implies “…and to avoid a consequence.”  If it’s not clear if there is a consequence, you may be thinking about the challenge in a narrow view.  Failing to climb the wall doesn’t mean you simply walk up to the wall, grab a rock, strain, slip, and shrug your shoulders.  That’s not how humans work.  They don’t give up that easily and nothing is ever that simple.  Failing to climb a wall could mean:

  • You climbed the wall, but twisted your knee, had some hard slips and falls, and cut your hand for a total of 1d6 damage.  (Succeed at a cost)
  • You tried to climb the wall, but you put too much weight on a lower handhold and broke it off when you slipped.  Now anyone trying to climb the wall has a -1 penalty.  (Game complication)
  • You tried to climb the wall, but fell noisily.  Now the guards probably know you’re here.  (Story complication)
  • You tried to climb the wall for five minutes, with no success.  Now you’re running out of time and getting nowhere.  (Raise the stakes)
  • You can’t figure out a way to get up this wall without leaving the rope and pitons behind.  (Charge for success)

Source: Run A Game’s Fail Forward Blog

These example echo the exact choices that we looked at from Fate’s “Succeed with a Cost” system. However, you can see that the PCs can lose items or be seriously delayed as a result of failure instead of it being an instant success with some challenge to deal with now or later. These will make the story that much more interesting than a pure ‘Nothing happens’ result, just like failing to hit in combat can make the combat more interesting.

Another way to look at determining the success or failure of actions is from the Yes/No scale taken from FU: The Freeform/Universal Role Playing System. Beyond just yes and no, there is And/But used as possible modifiers to mean that there was an additional bonus or penalty to whatever check needed it. You could even just number these 1 through 6 and roll them as you go, though the FU system used even numbers for the Yes, so they called their roll ‘Beating the Odds’.

Yes, and… The “and” indicates some kind of bonus.
Yes. The answer is just yes.
Yes, but… The “but” indicates some kind of penalty.
No, but… The “but” indicates that it’s not a total loss.
No. The answer is just no.
No, and… The “and” indicates it is even worse.
These give options of how the story may take an unexpected turn, as something can happen to make the scene power dynamic change.
  • Does your character manage to find a guide through unfamiliar terrain? Yes, but it is not the best guide you can find. (See the included image as an example, as this failure did manage to move the story along.)
  • Did the character find the trail they were seeking? Yes, But it disappears as it goes into the open field. So they need to figure some other way to track it.
  • Do the players manage to convince the duke of the urgency of their request for his troops to come to their aid? No, And the party offends him as coming across as telling him what to do so they are thrown from the castle grounds. They need to now find some other support as well as appease the king if they want back into the castle.
  • Does the players attack hit? Yes, And it manages to snag against the enemy’s armor throwing them off balance. Now the enemy is vulnerable until it gets a chance to right itself on its next action.
  • Do the players manage to escape from the collapsing cave? Yes, But one of the players limbs was crushed and is no longer functional, may even need to be amputated for them to be freed. (This raises a whole new series of questions about can one be a hero with a handicap like a missing limb. Every player will have their own answer to it, mine is answered by this list of 10 Heroes/Villains with artificial limbs.)

Not every failure nor even every roll needs this sort of detail. The presence of it however is quite a great option for when there is a possibility for a power shift in a scene, going back to the improv idea of tilt and balance. If someone is going to be able to gain or lose power to someone else, such as the infamous ‘I owe him a favor’ that tends to be the reason some things happen in dramas.

Some people have mentioned about the new Star Wars system of dice rolling, as seen in this YouTube video on the narrative Star Wars dice use. It is similar to these concepts, in that it does give you options of getting a full success or failure as well as having good and bad things happening outside of just the result of the roll. You could incorporate this style into your game by adding an extra die or dice to any conflict resolution, such as a 1d6 where the number determines where it falls on the success scale.

The last bit of failure mechanics that help to drive the narrative state I have today comes from Ghost/Echo which is an entire RPG system done solely on both sides of one single piece of paper and with only one real core resolution mechanic:

If you want to do something in any kind of situation of conflict, challenge, or difficulty, roll 2d6. Then assign one die to the “danger” and one die to the “goal”. High is good and low is bad. So if you get 6 and 5, you could assign the 6 to the “danger” (meaning you escape harm) and the 5 to the “goal” (meaning you succeed); if you get 3 and 4 you can assign the 3 to the danger (meaning you take some harm and the danger remains) and 4 to the “goal” (meaning you have a partial success), and so on. It gets most interesting when you get 6 and 1, obviously.

This forces you to make choices -to strategize – with your dice rolls. In some situations you might want to sacrifice success to escape harm, whereas in others you might want to succeed so much that you take whatever harm is coming your way. You have to be tactical with your luck. You can’t govern the dice results and they force you to make tough choices, but you do get a choice.

Let’s go back to our wall example from earlier. The goal is to climb the wall, while the harm need not be physical damage as we saw in those examples. Harm could be alerting the guards or damaging the wall so it is harder to climb next time or taking damage yourself.  It isn’t that you failed to climb the wall, that is a low success result. Instead, you had some negative occur as well as your positive.

You may not have every roll have any of these sort of states of partial success in your game, but occasionally it can be a great way to throw a character a lifeline to keep things moving and keep the story interesting. Let them fail, let them succeed, just make sure that your story is interesting and fun to play in.