Failing Forward

[Bedrockgames]

I don't think the word "needs" goes anywhere near my thoughts on the subject. I think it's a useful tool with different implementations and I'n curious about folks' opinions on it. I can see both of those two suggested implementations.

I see. I do find things like this tend to trip me up, but increasingly I am becoming aware that people manage the flow of a game very differently and that it seems like a good tool if having that thread continue is important. Even though I don't use it though, I think there is an aspect to it in the way I play, in that i'm always open to solutions that ought to work, even if I didn't think of them in advance, and I assume most situations will have multiple solutions available. But I think that may be slightly different from proper fail forward.

 

[Umbran]

All right, that changes just about everything I was going to write.

As outlined by Morrus, 'fail forward' seems to be a concept to solve a problem that a well designed adventure presented by an experienced GM should never have.

"Never" is such a strong word! For example, you are probably considering an "adventure" to be pre-planned, with the really relevant monsters and encounters all planned out ahead of time, so that all possible failure points can be identified before play begins*. But, as we all should know - not everyone prepares adventures that way. That situation is not even a given expectation for all game systems, much less all GMs.

The "succeed, but at a cost" form of Fail Forward that Morrus originally posted comes (as I understand the history) from a game or two where significant parts of the adventure content and/or difficulty is determined during play, rather than in design before play. When working in a more improvisational mode, you can't predict when you'll have a die roll that grinds things to a halt. Success-at-a-cost then plays a neat double-duty of avoiding the grinding halt while also generating some of the improvisational content through those costs.



*This kind of assumes a sort of perfection in the designer that I'm a little skeptical about, to be honest, but let's accept it for now.

 

[Reinhart]

I like fail forward mechanics, but obviously you don't need special mechanics to embrace the fail forward philosophy. Any pass/fail mechanic can be interpreted as just whether or not plot progress comes at a significant cost or not.

I see a few people arguing against the Fail Forward philosophy that mistakenly believe it makes failure impossible. Fail is in the very name! The point is not that failure is impossible but that failure should still lead to interesting plot results and not cause the story to slow to a grind. Making failure engaging instead of disengaging can be challenging for a GM to manage on the fly. However, like most GM skills, you get better at it through practice and planning.

 

[Mercule]

For example - Say the PCs are exploring a tomb, hunting the BBEG, who is in his secret lair, behind a super-secret door. The players go through the dungeon, search for for secret doors, but they botch the roll, and fail to find it.

This actually sparks an example from a game several years back. I'd consider it acceptable, in terms of "failing forward", but would be interested in what others thought.

The PCs needed to retrieve some widget from a dead sage/artificer's lair. I don't remember what it was, but let's say a sword. The sword was well guarded (think Tomb of Horrors lite), but quite within the PCs ability to reach. They were under a time constraint and dangerous things occasionally spawned. They managed to get the sword but discovered that the sage also had a journal that revealed things about the campaign's BBEG and said journal was still on premises. They tried, but could not find it in a reasonable amount of time -- need to use sword to stop bad thing.

A couple months thereafter, the Warlock went back to the lair to search. Why did he go alone? 1) He's a CN Warlock. 2) He was uniquely equipped to deal with some of the dangers, so long as other PCs weren't standing in the danger zone.

Based on #2, all he needs is time and a bit of OCD. Eventually, he finds the secret door in the ceiling that led to the secret lab, gets the journal, and the group has an advantage in the next act.

I think that's within bounds because 1) the journal wasn't important to the PCs' immediate goals, 2) it made their long-term goals easier, but wasn't a "blocker" to the campaign, and 3) it was obvious to at least one player how to work around it and what the cost was. Even #3 wasn't critical to the "fail forward" concept.

 

[Celebrim]

Umbran seems to have gotten in ahead of me and said most of the things I would have wanted to say.

I generally dislike 'Fail Forward', but I believe it is appropriate and even necessary in certain situations. The basic idea of 'Fail Forward' is that after a failure, the game not only needs to go on but needs to remain exciting or interesting.

However, the method for implementing 'the game must go on' is going to very between game styles and even between specific situations. Encoding fail forward and especially encoding it mechanically tends to very much limit the range of styles and means of continuing the story that are available to the GM.

I prefer as a GM to keep my options open, and I've experienced situations both as a GM and a player were failure was earned and you just had to accept it. Those situations have helped me learn better and less brittle design techniques, which tend to mitigate against the need for a mechanically enforced solution to failure. Avoiding chokepoints and having back up plans in the event of a choke point are just necessary parts of good DMing. It also helps to have a mindset where you don't see chokepoints as chokepoints because you aren't committed to single outcomes, so that no NPC is so critical to the plot that the timing of their death has to have plot protection, or no scenario is dependent on player actions to resolve in some particular way you are committed to.

As a GM, you have to be willing to have the Rebels either lose or win the Battle of Hoth. Your idea of what the story should be shouldn't preclude that the party doesn't split up, and Han and Leia end up on Dagobah with him, or are instead destroyed by Vadar in the asteroid field, and so forth. This means you as a story teller sometimes have to relinquish what you think is the perfect literary outline for 'the story' and instead be willing to have the story be different things and go in different directions.

But in a game, 'defeat' - real defeat - I think needs to be an option on the table. I've seen 'fail forward' defined in ways that argue for failure to be succeeding at a cost, such that real failure is removed as an option. The cost is always turns out to be what the player can bear, so that failure always turns out to be the difference between a Marginal Victory and an Decisive Victory (at most). In my opinion, you can't really savor a character surviving if death never really was a meaningful possibility.

 

[AaronOfBarbaria]

I'd figured out other ways to avoid the situations in which the game comes to a halt if a die roll goes the wrong way before I'd ever heard the phrase "fail forward" or ever come to the conclusion on my own that a roll might be determining success without consequence vs. success with consequence, rather than success vs. consequence.

As such, by the time I actually heard the phrase and came upon to corresponding idea behind it, I didn't have nearly as much use for it as I would have years prior - so I just tucked it away in my DM "tool belt" as a safety net to use whenever I haven't had the time to re-write poorly designed sections of published adventures (you know, those ones that are overall really cool ideas and fun scenarios, but the author thought it best to put anything actually important behind a secret door with no context clues to its location nor hints that it even exists, so to fail to find it on the first try is to fail to complete the adventure).

 

[The Crimson Binome]

It doesn't create a world where failure is never a possibility. It creates one where a specific group of heroes played by some players continue forward in the narrative rather than stopping dead. Like in a movie. It doesn't apply to every task, or even to anybody but PCs.

It is a silly world where events within the game world depend on factors outside of the game. It cannot possibly matter whether a character is played by a Player or the GM.

The ability of an individual to open a lock can depend on factors inherent to the character (such as skill and familiarity), or factors inherent to the lock (such as complexity and integrity), or factors in the environment (such as lighting and tectonic stability). It can't possibly depend on whether this character is a PC or the protagonist, because those are just labels you've attached from outside of the game; they don't mean anything.

 

[Morrus]

It is a silly world where events within the game world depend on factors outside of the game. It cannot possibly matter whether a character is played by a Player or the GM.

The ability of an individual to open a lock can depend on factors inherent to the character (such as skill and familiarity), or factors inherent to the lock (such as complexity and integrity), or factors in the environment (such as lighting and tectonic stability). It can't possibly depend on whether this character is a PC or the protagonist, because those are just labels you've attached from outside of the game; they don't mean anything.

Sure they do. It's a game.

 

[Bedrockgames]

It is a silly world where events within the game world depend on factors outside of the game. It cannot possibly matter whether a character is played by a Player or the GM.

The ability of an individual to open a lock can depend on factors inherent to the character (such as skill and familiarity), or factors inherent to the lock (such as complexity and integrity), or factors in the environment (such as lighting and tectonic stability). It can't possibly depend on whether this character is a PC or the protagonist, because those are just labels you've attached from outside of the game; they don't mean anything.

I personally don't play that way, but I do think people can bring whatever 'physics' they want to the game world. For me, stuff like believability matter, with a little bit of genre physics. At the end of the day it is important to me that world have a consistent reality to it. But I don't see why someone can't decide they want the reality they are experiencing to feel more like a story or a movie. For their hero to be protected bit from stormtrooper blasters or something.

I think both sides often get caught up in this false choice, that it has to 100% represent reality, or 100% represent the content of film and literature. I doubt most groups cleanly fall on the far side of either end of that spectrum, and many probably have a mix of players who clustered in different spots.

I know what I like, so I run games that way. That doesn't mean others can't run things differently or use different rationale for making rulings.

One thing I discuss with my players when we start a game is what 'physics' are being used. We don't have stuff like protagonist protection but there are still a lot of gray areas even in a campaign that on first glance seems vaguely realistic. For example is gunpowder behaving realistically or cinematically. If you are more strict regarding gunpowder use and trying to keep it realistic, that is going to mean the players really need to pay attention to the details of their plan. If it is more cinematic and hand wavy (i.e. sure that's enough to blow up a ship and lighting the fuse is a breeze with that fork and stone) things are different. I've actually seen disconnect between people mid-game and caught it as it was occurring. It was really helpful hashing out that detail because then everyone knew how the world was meant to behave.

 

[Lanefan]

Failure is failure.

Now, depending on what they do next the PCs can turn a failure into a fail-forward, a fail-sideways, or a fail-backward...but it's down to them. The lock example, while not great, gives each option: fail-forward may be to remove the hinges (or demolish the door, whatever) and get through; fail-sideways may be to search for another access so as to get to what's behind the door from another direction, and fail-backward may be to abandon the door entirely and go elsewhere. All I need to do as DM is react, in this case; and I've seen all three outcomes many times each.

A better example is the missed secret door with the BBEG behind it. Here the failure is on both a greater scale and a less-obvious scale - the door above is an obvious fail-or-succeed situation, the missing boss is not - and the party may not even realize it has failed at all. A very plausible outcome is the PCs clean out the parts of the dungeon they can find then go back to town and report the BBEG isn't there and the information saying he was is in error; a big win for the BBEG. It's not down to me as DM to lead them by the nose to the secret door; again, all I have to do is react to the PCs, and maybe change my plot line slightly to account for the BBEG lasting a bit longer than I'd expected.

Lan-"a pleasant side effect of a missed BBEG is that I'll be able to squeeze at least one more adventure out of him later"-efan