Failing Forward

[Bedrockgames]

It depends on frequency and circumstance of the technique's use.

As an alternative example, consider the case of a solo character cautiously exploring a small dungeon. The character falls into an empty covered pit trap 20' deep that resets over him. The character tries to climb up, but the DC (20) for the wall is beyond anything the player can roll (Dex 8; skill not trained) under the best of circumstance. Even if the character somehow makes the 10-15 foot climb, he probably has no way of triggering the pit cover.

Now the table can leave the situation in stasis and wait for the character to die of thirst or starvation or even fast-forward to that point.

Alternatively, the DM can fail-forward and say the noise of the PC makes whilst failing to climb attracts a nearby denizen of the dungeon that the PC can attempt to lever to get out of the pit. Or perhaps the PC will detect one wall is damp and more cool than the rest and if the PC can breach it, a stream or water will begin to flood the pit and the PC can try to swim up at the pit fills.

I think my only issue with that example is it is a solo adventure (which is very unique circumstance). But I suppose it still works because even in a regular campaign, you could have players separated by hundreds of miles in different locations and one goes off on his or her own into a dungeon and falls into the pit.

My first thought is this might actually be a system issue, because presumably they can keep trying to climb that pit wall. I wouldn't treat that as a one roll or else we accelerate time by two months and you die. But that said, all the things you mention saving the PC, those are things the GM is always free to introduce for any reason. That has nothing to do with him failing the roll on the pit. I am fine with a GM who wants to, deciding that there are locals who hear his cries or denizens of the dungeon who take notice. Those things to me are more about the circumstances surrounding the pit than the roll to climb it or to avoid falling down. What I want from a game is there to be a possibility of falling and taking the full damage from the fall (which in some cases could be lethal) and the possibility of trying to climb a wall and not being able to do so (doesn't mean you can't try again, but for that attempt you don't succeed). Adding in plot complications due to a skill roll, just for me, feels a little odd. I can see why some people do that. I don't think it is bad on its own. It just isn't really how I like to manage these things, unless I can see a clear line from the failure to the development. This is why Pemertons posts are making a lot of sense to me on the subject, because I can really wrap my head around what he is saying Failing Forward brings to the table. It seems to be very well suited for groups that are concerned with the plot or story moving forward in an exciting direction. I can totally see how it does that, and I think it is a good tool if that if is what you want. For me, it feels like the wrong tool in my own campaign.

That said, under the right circumstances, I do want there to be a possibility of a character falling into a hole and starving if that is what it seems like ought to be the outcome. I wouldn't sit there and torture the player about it, and I would probably roll randomly to determine if someone happens by to rescue, but I'd certainly consider it a plausible outcome if the player character were sufficiently isolated and trapped.

 

[Nagol]

I think my only issue with that example is it is a solo adventure (which is very unique circumstance). But I suppose it still works because even in a regular campaign, you could have players separated by hundreds of miles in different locations and one goes off on his or her own into a dungeon and falls into the pit.

My first thought is this might actually be a system issue, because presumably they can keep trying to climb that pit wall.

Except in the example, it is impossible for the player to roll high enough to succeed -- DC 20 vs. max roll of 19; it is an auto-failure.

I wouldn't treat that as a one roll or else we accelerate time by two months and you die. But that said, all the things you mention saving the PC, those are things the GM is always free to introduce for any reason. That has nothing to do with him failing the roll on the pit. I am fine with a GM who wants to, deciding that there are locals who hear his cries or denizens of the dungeon who take notice. Those things to me are more about the circumstances surrounding the pit than the roll to climb it or to avoid falling down.

Those introductions are forms of fail-forward when used to re-energise the situation after PC failure.

What I want from a game is there to be a possibility of falling and taking the full damage from the fall (which in some cases could be lethal) and the possibility of trying to climb a wall and not being able to do so (doesn't mean you can't try again, but for that attempt you don't succeed). Adding in plot complications due to a skill roll, just for me, feels a little odd.

It doesn't need to be in response to a skill roll (or any die roll for that matter). The complication introduction takes place after an effort to change the situation fails. That could be an auto-failure like the pit situation above, a failed conversation, or any other situation where the PCs and players are flailing and the risk of stasis is rising.

Some game engines and player preferences do lean on constant change quite heavily.

I can see why some people do that. I don't think it is bad on its own. It just isn't really how I like to manage these things, unless I can see a clear line from the failure to the development. This is why Pemertons posts are making a lot of sense to me on the subject, because I can really wrap my head around what he is saying Failing Forward brings to the table. It seems to be very well suited for groups that are concerned with the plot or story moving forward in an exciting direction. I can totally see how it does that, and I think it is a good tool if that if is what you want. For me, it feels like the wrong tool in my own campaign.

I also appreciate a clear line from failure to complication in the case of skill checks; it minimises the discordance of the player thought process and the character thought process to which arbitrary complications can lead.

 

[Bedrockgames]

Except in the example, it is impossible for the player to roll high enough to succeed -- DC 20 vs. max roll of 19; it is an auto-failure.

I don't play 5E and haven't played 3E in ages, so not sure how much that would align with the situation as described. But if it is completely impossible to succeed and that doesn't seem to match what should be the case in the situation, I'd call it a problem with the system or something the GM needs to step in an rule on.

Those introductions are forms of fail-forward when used to re-energise the situation after PC failure.

I don't think the GM introducing something dungeon like that is fail forward. That is a technique GMs have been using long before Fail forward was a thing and it has almost no direct relationship to the roll itself. I feel like there are two kinds of fail forward being discussed here. On, the one pemerton is describing, is very easy to grasp and clear. This other one is something I am having a lot of trouble distinguishing between other common GM tactics.

It doesn't need to be in response to a skill roll (or any die roll for that matter). The complication introduction takes place after an effort to change the situation fails. That could be an auto-failure like the pit situation above, a failed conversation, or any other situation where the PCs and players are flailing and the risk of stasis is rising.

But then that isn't anything all that unusual. GMs do that all the time without it being called fail forward. By that logic, any time I introduce any element to the setting that helps the players in any way, I potentially using fail forward. To me, a GM who takes pity on a PC trapped in a pit and has someone walk by so they don't die, isn't really doing anything to advance the plot or adventure, he is just saving the PC.

 

[Nagol]

<snip>

But then that isn't anything all that unusual. GMs do that all the time without it being called fail forward. By that logic, any time I introduce any element to the setting that helps the players in any way, I potentially using fail forward. To me, a GM who takes pity on a PC trapped in a pit and has someone walk by so they don't die, isn't really doing anything to advance the plot or adventure, he is just saving the PC.

Well, yes GMs have been doing it for a long time -- it is a common enough trope in the source fiction. You are using fail-forward any time you introduce an element that (1) the PCs can or are forced to react to that (2) helps destabilise the current status quo that was (3) introduced as a (usually unexpected) consequence of failure of a PC gambit.

Once upon a time, a PC enters dungeon, falls in pit, and starves to death. The end.

vs.

Once upon a time, a PC enters dungeon, falls in pit, manages to get out with lateral thinking/unexpected allies/tricking enemies/?. The PC goes on to ????. The end?

Which plot is more advanced?

 

[Bedrockgames]

Well, yes GMs have been doing it for a long time -- it is a common enough trope in the source fiction. You are using fail-forward any time you introduce an element that (1) the PCs can or are forced to react to that (2) helps destabilise the current status quo.

Once upon a time, a PC enters dungeon, falls in pit, and starves to death. The end.

vs.

Once upon a time, a PC enters dungeon, falls in pit, manages to get out with lateral thinking/unexpected allies/tricking enemies/?. The PC goes on to ????. The end?

Which plot is more advanced?

Again, it is about the utility of the concept. If Fail Forward is just making sure stuff happens, I am frankly not sure how it is any different from things like Situational Adventures or Bangs. As a concept, I don't find it terribly clear if all it is is the GM stepping in to keep things moving at a given moment (that could also apply to railroading). If on the other hand, it is a reaction to a failure when a player attempts something, that I can get. That is clear. But I feel like people are both saying it avoids complete failure by making story advancement occur while at the same time also keeps complete failure on the table as a possibility. This is the part that is really giving me trouble. I also feel like it is being confused with gradient success systems (where you have Success, Failure, Partial Failure, etc). It just seems to be getting overly broad usage in this discussion.

Your above definition could apply to any number of techniques. I introduce elements all the time that force players to respond in some way. I would not at all consider that fail forward. I would probably call it situational GMing. Again, random encounter table introduce elements that 1) force players to react and 2) help destabilize the status quo. I don't think anyone would honestly regard them as an example of failing forward.

 

[Ilbranteloth]

Well, yes GMs have been doing it for a long time -- it is a common enough trope in the source fiction. You are using fail-forward any time you introduce an element that (1) the PCs can or are forced to react to that (2) helps destabilise the current status quo that was (3) introduced as a (usually unexpected) consequence of failure of a PC gambit.

Once upon a time, a PC enters dungeon, falls in pit, and starves to death. The end.

vs.

Once upon a time, a PC enters dungeon, falls in pit, manages to get out with lateral thinking/unexpected allies/tricking enemies/?. The PC goes on to ????. The end?

Which plot is more advanced?

Well, one might argue that the first is more advanced, or more mature anyway, in light of the current trend to avoid character death at all costs (of which I am often guilty of as well). Wander into a dangerous, unexplored location alone and (apparently) unprepared? As my daughter learned in kindergarten - you get what you get and you don't get upset.

It might be a very, very interesting prelude to an adventure where the next group finds the body of that individual. It's also along the lines of the sports star that falls out of his boat and drowns while fishing which has happened.

Having said that, at least in D&D, a natural 20 is generally assumed to be success. It's not RAW, but it's a lot like Free Parking in Monopoly. In addition, unless the DM intended for that to be a possible consequence (even if it was by accident), I suspect that most would work for a way out. Perhaps they didn't realize that the character would have a -1 to their check. Of course, I've seen creative players figure out far more complicated situations than this as well. If it was not intended, you simply come up with a way out. Sure, you can call it failing forward. And this is what I think a lot of people consider that term to be (as I did). I just see it as fixing a mistake. It's a limited, or (hopefully) one-time occurrence rather than a DM/gameplay philosophy.

And now to contradict myself - in my campaign, traps exist for a reason. It's not uncommon for those traps to exist to kill. Particularly if it's in a sealed tomb that the PC has just violated. There aren't patrols. There aren't wandering monsters. The trap exists solely to keep anybody entering to go any further. Period. I generally hint (through lore) that entering something like a sealed tomb or crypt is among the most dangerous things around. You are isolated and potentially entering a death trap. I'm the DM and you violated my tomb. Prepare to die.

If it's a pit with a disarm mechanism or a way to avoid it as protection to a lair for something, then yes, there will be patrols or wandering monsters. Maybe it's built by something long ago, and has been forgotten, but the location is occupied by something else now.

As for what I think you were actually asking - which is more interesting? Well, the second if the character gets out with lateral thinking. Absolutely the most interesting.

Unexpected allies/enemies that suddenly appear because I screwed up? If the possibility existed (it was a living dungeon, not an abandoned crypt), then it's better. Interesting for the players, but less interesting for me. Essentially what it comes down to in terms of what's interesting is partially dependent on how much of the solution I have to write. Not that writing an interesting solution can't be fun too. But part of the fun of an RPG is not knowing what's going to happen, on both sides of the table.

Ilbranteloth

 

[grendel111111]

Well, one might argue that the first is more advanced, or more mature anyway, in light of the current trend to avoid character death at all costs (of which I am often guilty of as well). Wander into a dangerous, unexplored location alone and (apparently) unprepared? As my daughter learned in kindergarten - you get what you get and you don't get upset.

Here is a very clear example of story first vs. gamist approach to RPG's. The characters death due to being trapped at the bottom of a pit is less satifying or heroic than being killed in a dramatic stand off over a lava pit. So if your focus is primarily on having a good story then this kind of ending to the story is not OK.
But if it OK for you that a story ends like this then it is just how it ends. You lost this time around, lets start a new game.
In terms of time at the table saying "you die from the large sword that the sneaky goblin thrust into your side." takes no more time than "you slip into unconsciousness and slowly die from dehydration." But it feels less heroic. So in a heroic action focused game this would not be desirable, in a "realism" game then dying at the bottom of a pit because you feel and broke your leg is an acceptable end to the story.

It might be a very, very interesting prelude to an adventure where the next group finds the body of that individual. It's also along the lines of the sports star that falls out of his boat and drowns while fishing which has happened.

There are so many different ways to get out of the situation.
One of my favorites would be "as you slip into unconsciousness you think back to the day you left the town and the note you had written to your young apprentice (or insert other NPC the player knows) saying you will return in 4 days, but that if you didn't he was to organize a search party and come looking for you.......... then hand over the NPC who becomes their new PC till they have completed the rescue.
They are then on a timer to find their original character before he or she dies at the bottom of the pit. (Players get bonus points if they end up with the rescuer trapped in the same pit as the person they try to rescue and die together).

Having said that, at least in D&D, a natural 20 is generally assumed to be success. It's not RAW, but it's a lot like Free Parking in Monopoly. In addition, unless the DM intended for that to be a possible consequence (even if it was by accident), I suspect that most would work for a way out. Perhaps they didn't realize that the character would have a -1 to their check. Of course, I've seen creative players figure out far more complicated situations than this as well. If it was not intended, you simply come up with a way out. Sure, you can call it failing forward. And this is what I think a lot of people consider that term to be (as I did). I just see it as fixing a mistake. It's a limited, or (hopefully) one-time occurrence rather than a DM/gameplay philosophy.

And now to contradict myself - in my campaign, traps exist for a reason. It's not uncommon for those traps to exist to kill. Particularly if it's in a sealed tomb that the PC has just violated. There aren't patrols. There aren't wandering monsters. The trap exists solely to keep anybody entering to go any further. Period. I generally hint (through lore) that entering something like a sealed tomb or crypt is among the most dangerous things around. You are isolated and potentially entering a death trap. I'm the DM and you violated my tomb. Prepare to die.

If it's a pit with a disarm mechanism or a way to avoid it as protection to a lair for something, then yes, there will be patrols or wandering monsters. Maybe it's built by something long ago, and has been forgotten, but the location is occupied by something else now.

As for what I think you were actually asking - which is more interesting? Well, the second if the character gets out with lateral thinking. Absolutely the most interesting.

Unexpected allies/enemies that suddenly appear because I screwed up? If the possibility existed (it was a living dungeon, not an abandoned crypt), then it's better. Interesting for the players, but less interesting for me. Essentially what it comes down to in terms of what's interesting is partially dependent on how much of the solution I have to write. Not that writing an interesting solution can't be fun too. But part of the fun of an RPG is not knowing what's going to happen, on both sides of the table.

Ilbranteloth

It almost feels like "fail forward" is trying to take credit for anything a DM does. Which makes it a fairly meaningless idea to discuss. I suspect that not everyone would view it as all encompassing as this, but it is clearly more than "succeed at a cost". I'm just unclear where the line gets drawn. Is levels of success and failure part of fail forward? As far as I know it has been around long before this term was used (even critical success and critical failure are a form of this).

 

[GMMichael]

Once upon a time, a PC enters dungeon, falls in pit, and starves to death. The end.

This has a nice ring to it. But it's also missing:

1) the PC learns that he should have brought allies.
2) the other PCs learn that there are real (enough) stakes involved. Excitement increases.

Now, a problem with this plot is that it's horribly written - it's super linear, and offers all the action-options of a 1980s video game. Maybe that's why it sounds so much like a 1980s RPG plot?

 

[Balesir]

This has a nice ring to it. But it's also missing:

1) the PC learns that he should have brought allies.
2) the other PCs learn that there are real (enough) stakes involved. Excitement increases.

Now, a problem with this plot is that it's horribly written - it's super linear, and offers all the action-options of a 1980s video game. Maybe that's why it sounds so much like a 1980s RPG plot?

I think you mean "the player(s)" rather than "the PC(s)". The PC is dead, so anything (s)he might have learned is pretty much moot...

The effects thus described are pretty meta, though, and I wonder how that fits with the "character immersion" agendas?

 

[Maxperson]

Sometimes, sure. Other times not so much.

Untrue. The situation can develop during play -- see my lone PC falling into a pit, above. Additionally, there could be a dozen ways out of the situation, but the PCs do not have the capability for half, consumed their resources for another third, and can't think of the 2 remaining methods.

The DM knows the capabilities of the party (and players). If he created a situation where the PCs only have one way in or out of something, that's his fault.

Which is an equivalent to failing-forward. You are introducing new information to the players that the PCs can take advantage of. The only difference is the change is internal to the PC as opposed to coming from the environment surrounding them. It is another technique I use.

It's similar, but different. There is no disconnect with my method.

Failing-forward is also connected to what is going on -- just it is connected to what is going on in the environment the PCs find themselves within. A well-done fail-forward also can provide additional implications about the situation from which the players can draw inferences.

If you've read the thread, fail forward can have nothing directly to do with the roll at hand. Like the rod falling out of the backpack as the result of a failed climb check.

As for what makes players happy, that is too varied to generalise. I do make an effort to develop situations where the PCs can utilise all purchased abilities, but in some systems that also means I get to use their contacts, favours, and other environmental perks as a failing forward lever.

Whereas unless it is directly related to the task at hand, those things will not be used as part of a roll. I can and do use them elsewhere, though.