Why is Failing Forward so disliked on these boards? [!silva]

Lago PARANOIA · Post by **Lago PARANOIA** » Sun Dec 21, 2014 12:07 pm

Quick primer: Failing Forward is a mechanic where if you fail to accomplish an action you take escalating negative consequences equal to the distance of your failure on the RNG to accomplish what you originally set out for. For example, if you slightly fail on your Sneak, you're undetected but you raise your suspicion meter or you had to ditch some equipment. If you majorly fail forward on your Bureaucracy check you're able to implement your programme anyway but you owe a lot of political favors and you pissed some political interests off. Games may have your negative consequences related to the action you undertook; most of the ones I've seen do. A game may present Failing Forward and regular-ass failing as selectable consequences once you hear the stakes. And if a game does give you a choice, it may also not tell you the exact stakes until you commit to a course of action but then you roll to determine how bad you fail.

So my question is this: why do so many people seem to dislike Failing Forward? I mostly play D&D, a game with very discrete consequences and actions (and provides a lot of escape hatches for them), so I don't really care if it's in the game or not. However, for games that have more involved action sequences like Burning Wheel or, hell, 4E D&D's skill challenges it seems like a very good way to have attrition mechanics or non-binary consequences.[/list]

TiaC · Post by **TiaC** » Sun Dec 21, 2014 12:20 pm

Poor implementation for the most part I think.

Many systems have either no real guidelines to help the MC determine appropriate consequences or only give options worse than flat failure. It is especially bad for lower-stakes rolls without consequences. For example, I've seen this "You rolled a partial success on identifying the symbol on the motorcycle gang's helmets. You know what it is, but you had to look hard so the leader thinks you were looking at him funny, he's going to be more hostile now."

It can discourage people from trying things because the risks of their actions are ill-defined.

PhoneLobster · Post by **PhoneLobster** » Sun Dec 21, 2014 12:24 pm

The answer in one word is "Silva".

The answer at a more detailed level is that it's an informal or at best semi-formal mechanic made of fairy tea party and fail bait for dicky or even just tired GMs that cannot be designed or even adjudicated in a predictable and consistent manner.

As demonstrated by any thread involving "failing forward" and Silva.

It just doesn't work outside of rules lite systems or rules lite portions of systems. And even when people around here use those they regard them as too trivial (both in difficulty of design, and in value in play) to be worth talking about.

If you get any more formal than that and you aren't really doing recognizable "fail forward" anymore, you are basically doing Hit Points.

Koumei · Post by **Koumei** » Sun Dec 21, 2014 12:39 pm

PhoneLobster wrote:The answer in one word is "Silva".

I have to admit, when I saw the title of the thread I assumed this was a silva post. So the well of forward-failing has probably been poisoned quite badly by silva pissing in it.

Lago PARANOIA · Post by **Lago PARANOIA** » Sun Dec 21, 2014 12:41 pm

Dammit, people, stop giving the Bear World fanboys intellectual power over your fun-time orc genocide simulators. It's what they want.

TiaC wrote:Many systems have either no real guidelines to help the MC determine appropriate consequences or only give options worse than flat failure. It is especially bad for lower-stakes rolls without consequences. For example, I've seen this "You rolled a partial success on identifying the symbol on the motorcycle gang's helmets. You know what it is, but you had to look hard so the leader thinks you were looking at him funny, he's going to be more hostile now."

I don't really have an inherent problem with Fail Forward consequences being generically more painful than Fail And Fuck Off. That would still make them valuable in certain instances. For instance, if Failing Forward for kicking open a door leaves you exhausted and taking a -2 penalty on most checks for an hour, 95% of the time people won't do it. But 5% of the time people would still take that consequence if, say, they were being chased by a horde of ghouls.

That said, that kind of open-ended consequence crap is a huge problem with those systems. I think that the game should A.) give you the option between choosing a Fail Forward or Fail And Fuck Off after delineating the consequences and/or B.) spend 4-5 pages detailing the typical consequences for Failing Forward certain checks by certain amounts.

erik · Post by **erik** » Sun Dec 21, 2014 1:48 pm

Koumei wrote:
PhoneLobster wrote:The answer in one word is "Silva".
I have to admit, when I saw the title of the thread I assumed this was a silva post. So the well of forward-failing has probably been poisoned quite badly by silva pissing in it.

Indeed, if you don't want this thread shat upon then I recommend a "silva stay out" warning in the title. Maybe if it ignores enough of those then the troll gets banned?

I think in one ancient silvatrollthreads I argued that FF could be okay if done better. Say, using an expendable currency to do it. I can't be fucked to return to that shithole of a thread wherever it be now. Tis probably afflicted with radioactive bear hepatitis.

Longes · Post by **Longes** » Sun Dec 21, 2014 2:36 pm

Didn't we have this thread before?

Lago PARANOIA · Post by **Lago PARANOIA** » Sun Dec 21, 2014 2:57 pm

I probably should have used 'Succeed With Compromise', which is the term that Burning Wheel uses and is the game mechanic I had in mind when I used 'Failed Forward'. I didn't know that Bear World assholes hijacked the term to mean 'MC lulz at you and then takes control of the plot until they're bored'.

Maybe I should read more Anus World threads in the future so that I can keep up with the latest advances in Gygaxian terminology

Longes · Post by **Longes** » Sun Dec 21, 2014 2:59 pm

Well, Burning Wheel jacked up its TNs to make sure you never succeed normally.

EDIT: to clarify, in Burning Wheel you make a Resources roll to buy something. Obstacle 1 is: "Food for the day. Lodging for the night. Clothing, shoes, and other simple mundane goods. Day laborer's wages." This means that an average peasant fails to buy food at least once a week.

hogarth · Post by **hogarth** » Sun Dec 21, 2014 3:11 pm

Longes wrote:Didn't we have this thread before?

As K said in the other thread, every game "fails forward" most of the time; failure at something almost never results in saying "campaign over -- THE END".

As I noted in the other thread, most of the discussion in favour of "failing forward" was claiming that a turn where nothing happens is boring, but actually seemed to be that long processes that result in nothing happening are boring.

name_here · Post by **name_here** » Sun Dec 21, 2014 3:59 pm

I don't mind the concept, but I feel like it should be at least somewhat rare. In Bearworld, you get it on 7-9 and very well might have a +1 bonus that shifts it to 6-8. The RNG is 2d6; remember how it is curved.

silva · Post by **silva** » Sun Dec 21, 2014 7:08 pm

Lago, I think some people around here have a problem with the Apocalypse World implementation of the concept, not with the concept per se. As name_here explains above, AW purposefully makes it the more frequent occurrence as a means to promote its design goals. It's like a constant Murphy's Law at that. Both Burning Wheel, Fate and Warhammer 3 seem to use the concept in a less radical form, if I remember right.

RadiantPhoenix · Post by **RadiantPhoenix** » Sun Dec 21, 2014 7:11 pm

For those who don't know, "!X" means "not X"

In context, I'm not sure how "!silva" could mean anything except "no silva", which would almost certainly mean, "silva is not welcome in this thread"

Longes · Post by **Longes** » Sun Dec 21, 2014 7:16 pm

Fate doesn't use Failing Forward at all. In Fate you get a Fate point whenever you allow yourself to be fucked by GM, and failing a dice roll is just failing the dice roll, like in DnD.

Burning Wheel makes sure you fail all the time by jacking all the TNs really high. Even if it makes no sense (it's virtually impossible for an elf to obtain any elven equipment, since it's all TN 7 and up). You only win when rolling against other people with similar traits, or spend finite resources to give yourself extra dice.

silva · Post by **silva** » Sun Dec 21, 2014 7:17 pm

Oops. I thought the "!" meant "Hey Silva look over here!" in this context.

Josh_Kablack · Post by **Josh_Kablack** » Sun Dec 21, 2014 7:18 pm

Well you can formalize "success at a cost" pretty easily by doing something like making every check cost fatigue points and using something like the 3.x rules for taking 10 and taking 20 but that immediately runs into 2 problems.

Firstly, if your system is anything like 3.x D&D players would need to track hundreds of fatigue points to have characters capable of making as many checks per adventuring day as the game assumes.

And secondly it requires that skill uses all be at the same degree of abstraction (or at least highly formalized for each possible use of the skill) Thus a Spot check to notice a sneaking enemy, a Ride Check to calm your warhourse, a Forgery check to fake a letter from the Duke and an alchemy check to set up your entire lab are all looking at similar time and fatigue constraints in such a system.

Kaelik · Post by **Kaelik** » Sun Dec 21, 2014 7:25 pm

RadiantPhoenix wrote:For those who don't know, "!X" means "not X"

In context, I'm not sure how "!silva" could mean anything except "no silva", which would almost certainly mean, "silva is not welcome in this thread"

I thought he might be Portugese or something? And at least in spanish they begin and end exclamations with a (upside down) exclamation point.

Pedantic · Post by **Pedantic** » Sun Dec 21, 2014 7:54 pm

As I understand it, the "problem" that those success at a cost mechanics are trying to solve is "sometimes rolling a failure results in no change to the game state." In that case, I can't see a real benefit to the mechanic that can't just be covered with clearer rules for timing, and consistent reference to take 10/20 (and other default success) mechanics when appropriate.

The place such a mechanic might have more interesting applications is when players are under pressure from outside circumstances and care more about success than usual. If the players can opt to take some set of negative consequences in exchange for success on rolls that would otherwise fail, that's more interesting. You could codify appropriate penalties, perhaps by the retroactive amount of bonus they'd be worth on a d20 roll.

I think you'd want to give players an idea of what sort of penalty they're signing up for, but I'd leave the specifics up to the DM or down to a roll on a table.

Lago PARANOIA · Post by **Lago PARANOIA** » Sun Dec 21, 2014 8:29 pm

Well, like I said I don't see the value of Fail Forward Succeed With Cost in games that heavily salami slice time units and game effects like 3E D&D. That shit is just an unnecessary complication the vast majority of the time. If the DM offers a Fail Forward option in 3E D&D as she is played then most of the time the players will just tell the game mechanic to fuck off and pull out their Take d20/magic scroll/summoned critter and you just have a waste of page space on your hands.

For games or subsystems with more abstract resolution mechanics or phases such as FATE or Shadowrun hacking or 4E D&D-style skill challenges I think the idea might have some merit.

Whipstitch · Post by **Whipstitch** » Sun Dec 21, 2014 8:53 pm

My issue with fail forward is that about the only times where it seems truly appropriate are the usually the times where such events are such an obvious outcome that the game has almost certainly already accounted for it. For example, falling down the 50 foot chasm is already an emergent consequence of rolling shitty on your Jump check in D&D--you roll the dice, you go X distance and when you check your final destination you realize that you're going about 50 more feet the hard way.

Mask_De_H · Post by **Mask_De_H** » Sun Dec 21, 2014 8:56 pm

Kaelik wrote:
RadiantPhoenix wrote:For those who don't know, "!X" means "not X"

In context, I'm not sure how "!silva" could mean anything except "no silva", which would almost certainly mean, "silva is not welcome in this thread"
I thought he might be Portugese or something? And at least in spanish they begin and end exclamations with a (upside down) exclamation point.

He's Brazilian and dense.

Success with cost mechanics are useful in time/state sensitive checks. They're also useful for degrees of success checks, when you don't have fine divisions of outcomes want to give a diminished, but still positive outcome. They're not well liked here because a) our main game has the necessary fine divisions of outcomes (DCs, taking 10/20, failing/succeeding by five or more) and b) the examples of the system we are shown are non-functional (Bear World). In some games, you want to keep the narrative moving, so a null state failure is a waste of time. There, you would want to offer the success with a cost in addition to or instead of the null state failure setting. You just have to clearly define what the cost is and make it actually count as a success, instead of the "you succeed at sneaking except you don't" sort of thing. Using FATE as an example, a success with a cost would give you a negative Aspect as well as whatever the basic success metric is.

I actually remember certain tests in Asymmetric Threat having something of a success with cost mechanic; you needed X successes before you got Y failures, but you could get a failure and a success notch on the same check.

Orion · Post by **Orion** » Sun Dec 21, 2014 10:33 pm

Failing forward is really appealing to me because players are lazy and uncreative gits. Most groups will seriously only come up with 2 ideas to handle a problem, and if they both fail then the game grinds to a halt while they stare at you hoping you'll bail them out.

ETortoise · Post by **ETortoise** » Mon Dec 22, 2014 2:04 am

I've definitely been in games where the action grinds to a halt because we failed a few rolls. In Burning Wheel revised they use the example of a locked door, if the players fail to pick the lock or knock down the door they are stymied but with a fail forward mechanic the door can be bypassed with the side effect of alerting nearby guards. Now, I can think of a couple ways to deal with the locked door now, both as a player and a GM, but I have literally been in games where this has screwed us up (granted, we were 13 or 14.)

In many ways a failing forward mechanic works as a formal reminder to the GM to keep things moving forward. This can also be done with techniques like the Three Clues or more interesting encounter design but failing forward is one size fits all.

I do like Lago's suggestion of having the players' choose whether to fail forward or simply fail.

Longes: In BW there are lots of ways to scrounge up extra dice for a roll, and the peasant only has to make his resource test to feed himself once a year at harvest time. The difficulties are hard because they want you to spend artha to increase your chances so you'll want to earn artha by playing towards your beliefs.

TheFlatline · Post by **TheFlatline** » Mon Dec 22, 2014 3:03 am

From my experience in Star Wars, it really increases the burden on the GM by $TEXAS because now you have to come up with a result that's interesting and novel for *every* die roll, instead of for crits or whatever.

Because we always say "oh yeah if you fail a sneak roll you still get through but you set the guards on alert" and that's great for *one* roll. But if you roll 10-15 sneak rolls over the course of a session (and that works out to 3-4 party attempts at sneaking, which isn't unreasonable in an infiltration situation) then you can only repeat the same "fail forward" events so many times before your players get board and glaze over.

In one combat, you miss 50%-66% of the time on average (up to 75% of the time I guess depending on the system) and can you imagine coming up with something interesting for 2/3 of the fucking die rolls?

Or worse yet, you fail your sneaking roll and the asshole GM decides "LOL you leave your artifact sword behind to sneak past the nose-picking goober guard".

Player involvement isn't even a solution to any of this because as was mentioned before, players get lazy and also, what they think is a justifiable penalty to pay for blowing a roll but still getting to proceed is probably not exactly in line with what you have in mind as the GM.

And that's not even touching the idea of Schrodinger's Bears (I like it more than quantum bears, but it's the same idea) and accepting that shit that normally wouldn't exist is going to pop into existence on a fail roll.

silva · Post by **silva** » Mon Dec 22, 2014 3:43 am

TheFlatline wrote:Because we always say "oh yeah if you fail a sneak roll you still get through but you set the guards on alert" and that's great for *one* roll. But if you roll 10-15 sneak rolls over the course of a session (and that works out to 3-4 party attempts at sneaking, which isn't unreasonable in an infiltration situation) then you can only repeat the same "fail forward" events so many times before your players get board and glaze over.

Agreed. Thats why I find the concept works better with intent/conflict resolution systems like the one in AW, where the amount of rolls you do is much smaller.