GM techniques (especially for non-combat challenges/resolution)

[MENTION=6881836]Josiah Stoll[/MENTION]

I saw you've been looking at some other threads discussing GMing techniques. There's a range of approaches. The default on these boards emphasises GM-driven play (or APs where the GM channels the module author). What I'm going to talk about in this post/thread is a different approach. You might find it helpful, or not - all I can say is that it's been working for me for a long time now.

RPGing is fun when the players are engaged. Combat is often engaging by default, because PC death is at stake. To make non-combat engaging, the players have to be able to see that something is at stake that they care about. This is what will get them wanting to engage. And they have to be confident that engaging the situation won't leave them hosed. The fear of being hosed leads to turtling, tedious tactically-focused play, a game that moves at the pace of treacle, etc. And also easily drifts back to GM-driven as the players look for the "correct" way to win (ie the one the GM has in mind) so that they can avoid being hosed.

Where do ge the engaging stakes from? Your players. This can be overt - some systems build this into PC gen (eg Fate aspects; Burning Wheel beliefs), but it can be done in other systems too (when I started a 4e campaign I required each player to establish a loyalty for his/her PC, and also a reason to be ready to fight goblins). Or it can be implicit - in 5e, background in particular might be an implicit signal of how a player sees his/her PC and what s/he might be ready to care about in play.

Here are four actual first sessions. None is 5e (one 4e, one Burning Wheel, one Cortex+ Heroic, one Classic Traveller). But they show various things I've done as GM, working with my players, to create situations with combat and non-combat elements to them that the players will engage with in play.

The other aspect I mentioned - avoiding a fear of hosing. "Fail forward" adjudication is designed to help with this. The key idea is that failure doesn't mean "You suck, it didn't work!" It means "Things didn't turn out how you wanted!" That could be because the PC sucks. Or because some external factor intervened that the PC didn't know about. Or maybe the GM narrates the failure by "revealing" (I use inverted commas because, at the table, the GM is making it up as part of narrating the failure result) some hidden aspect.

The BW session report I linked to gives one example: the PC tries to meet with a senior member of his cabal to get work; the check fails (in BW its a Circles check; in 5e it might be a CHA check to reach out to connections), so instead of an offer of work or a hot tip from the senior sorcerer, the sorcerer sends a thug to tell the PCs to leave town. See how, even though the player didn't get what he wanted, the stuff he cares about is still a focus of play. The sorcerous cabal is still important to the game, and the PC hasn't been shown to be a failure; but now he has to somehow win back the trust of his cabal's leader (or whatever else he wants to do in response to the situation).

Especially when it comes to getting a mediocre-skill fighter involved, these two things can work well together: if the non-combat situation invovles something the fighter PC (and the player of that PC) cares about then the player will declare actions; and even if they fail, "fail forward" adjudication means that the player didn't just get hosed or make a fool of him/herself or his/her PC - the thing s/he cares about is still there, still in play, but the situation around it has changed. So the player (and hopefully the other players too) get drawn further into the game, fiction keeps developing, instead of following the GM's trail of bread crumbs you're creating some story together.

The 5e basic set has Fail Forward embedded in its GMing section on noncombat action resolution in the same way that 13th Age does. Given that 5e draws some inspiration from 13th Age, the easiest thing to do is just follow 13th Age advice for 5e noncombat action resolution failure:

Fail Forward!

Outside of battle, when failure would tend to slow action down rather than move the action along, instead interpret it as a near-success or event that happens to carry unwanted consequences or side effects. The character probably still fails to achieve the desired goal, but that’s because something happens on the way to the goal rather than because nothing happens. In any case, the story and action still keep moving.

Click to expand...

This is a skill that must be honed by GMs but most times it begins with a Yoda lesson; "you must unlearn...what you have learned."
So to begin, this means undoing their internal and cultural-inflicted programming whereby they have been conditioned to perceive every moment of action resolution as a binary event constrained by 1st order causal logic interpretation of the micro-task at hand:

You attempt to climb a treacherous face; you climb it or you fall.

You attempt to convince a chamberlain to grant you audience with the king; he's convinced or you're rebuffed.

Action resolution (particularly noncombat action resolution) doesn't have to be governed by 1st order complications, or tight causal logic of the micro task. Failure can be interpreted as 2nd or 3rd order complications, on the grounds of genre logic, or in consideration of the macro goal.

You don't fail to climb the face. You lose something precious (perhaps a divining rod falls from your hip and clatters down the face, coming to rest in a wyvern nest which is full of eggs) that has future or immediate implications on your greater goal and now you're faced with an interesting decision-point that adds new conflict.

You don't get rebuffed by the chamberlain. He agrees and then casually walks you through the receiving hall while you're assailed with the horrific specter of a bloody orgy where demons are simultaneously feasting upon and mating with the king's court.

Manbearcat said:

The 5e basic set has Fail Forward embedded in its GMing section on noncombat action resolution in the same way that 13th Age does. Given that 5e draws some inspiration from 13th Age, the easiest thing to do is just follow 13th Age advice for 5e noncombat action resolution failure:



This is a skill that must be honed by GMs but most times it begins with a Yoda lesson; "you must unlearn...what you have learned."
So to begin, this means undoing their internal and cultural-inflicted programming whereby they have been conditioned to perceive every moment of action resolution as a binary event constrained by 1st order causal logic interpretation of the micro-task at hand:

You attempt to climb a treacherous face; you climb it or you fall.

You attempt to convince a chamberlain to grant you audience with the king; he's convinced or you're rebuffed.

Action resolution (particularly noncombat action resolution) doesn't have to be governed by 1st order complications, or tight causal logic of the micro task. Failure can be interpreted as 2nd or 3rd order complications, on the grounds of genre logic, or in consideration of the macro goal.

You don't fail to climb the face. You lose something precious (perhaps a divining rod falls from your hip and clatters down the face, coming to rest in a wyvern nest which is full of eggs) that has future or immediate implications on your greater goal and now you're faced with an interesting decision-point that adds new conflict.

You don't get rebuffed by the chamberlain. He agrees and then casually walks you through the receiving hall while you're assailed with the horrific specter of a bloody orgy where demons are simultaneously feasting upon and mating with the king's court.

Click to expand...

Its also standard in the PHB.

In the section on using ability checks the actual definition of what failure means is (rough quoting) you make no progress *or* you make some progress with a setback added that the gm chooses.

That being right there front and center from the beginning gives you a powerful tool useful for a lot of situations and that can help deal with some oft-cited problems.

Low roll on insight check making you sweat about player meta-gaming? If you decide its progress with setback, then maybe they did get some of the right info and later on might figure out someone else picked their pocket while focused on insight.

Everybody goes "search the room" and dice clatter everywhere? "Great now lets start with the low rolls..." and the setbacks begin.

"I keep retrying"... same.

In the DMG this is expanded in,more limited scope to cover saves and attacks with success at cost but is limited to miss by 1-2.

It's worth being aware of how Fail Forward as a technique is largely incompatible with GM-sided plot control techniques.

It should be self-evident that if the GM is the source of a) all the content including b) the characters' reasons for doing everything (everything plot-related anyway) then 'failing forward' reveals more GM content which they actually expect the players to 'earn' by having characters succeed.

This explains why some are adamant that fail forward = success with complications. Because when GM (via the 'scenario or 'plot' or 'adventure') is the actual source of the character's protoganism, it happens to be true.

But if the player is the source of their character's protaganism, then fail forward works as being discussed in the thread. The player has a stated goal and the GM disrupts reaching it (or the action succeeds, but without realizing its assumed purpose) without reference to their own ulterior motive or pre-scripted 'plot'.

To give an example, let's say the character is trying to meet the duke to talk to him about something or other.

Fail forward version 1:
Player: I talk to the chamberlain about an audience with the duke (fails charisma, persuade, oratory etc)
GM (Chamberlain): The Duke is thinking about his next move in a very important game of chess being played by correspondence with the neighbouring Dukedom. Do you play? I'll show you the current position - find a winning line and I'll grant the audience. Cheating? My dear fellow, you don't think the Dukes actually think of the moves, do you?

Fail forward version 2:
Player: I talk to the chamberlain about an audience with the duke (fails charisma, persuade, oratory etc)
GM (Chamberlain): Very well, follow me! (Leads the character through where a shocking demonic horde is controlling the Duke and his retinue in mind-wrenchingly terrible ways).

If a pre-scripted plot is underway then with either 1 or 2 all that's happened is the GM has revealed more of their material (and possibly a 'big reveal' moment) on the basis of a 'fail', when the playstyle is that the players have to earn the next bit of scripted storyline by 'winning' or 'succeeding'.

In the playstyle in which fail forward is suited, either 1 or 2 are the improvised or semi-improvised responses to player interests (chess or puzzles or the neighbouring Duke, demonic possession). So they represent a failure for the character, while also being a significant change of situation tied to their established interests.

[MENTION=99817]chaochou[/MENTION]

I'm going to paraphrase your post back to you, just to check I've followed it. (I think I have.)

In a GM-driven game, "fail forward" = "success with cost/complication" = GM reveals the next bit of the story but (because the player(s) failed the check) the GM doesn't reveal all of it, or reveals it but taxes some resources, or something else that counts as a cost.

It's hard for me to judge how effective this is, but it looks kind-of like a "death spiral" - because you failed you get penalised on the way to the next bit of GM plot, and because of that penalty you're more likely to fail your next check which means you're more likely to need more "success with a cost" etc. Seems a bit sucky to me.

"Fail forward" in the sense I'm trying to explain is in lieu of GM plot. Instead of GM plot, the story is created by the checks: succesful checks and things unfold as the player playing his/her PC hoped; failed checks and the things the player cared about are in play, but it's getting worse rather than better.

pemerton said:

It's hard for me to judge how effective this is, but it looks kind-of like a "death spiral" - because you failed you get penalised on the way to the next bit of GM plot, and because of that penalty you're more likely to fail your next check which means you're more likely to need more "success with a cost" etc. Seems a bit sucky to me.

Click to expand...

I'm pretty sure this happened in 4e where failed skill checks within a skill challenge cost Hit Points, HD, other resources, penalties on the next check...etc

Sadras said:

I'm pretty sure this happened in 4e where failed skill checks within a skill challenge cost Hit Points, HD, other resources, penalties on the next check...etc

Click to expand...

Are you saying that when you run a skill challenge, the consequence of a failed check is "penalties on the way to the next bit of GM plot"? That's not how I do it. This thread, as per the OP, is focused on non-GM-driven RPGing.

pemerton said:

I'm going to paraphrase your post back to you, just to check I've followed it. (I think I have.)

In a GM-driven game, "fail forward" = "success with cost/complication" = GM reveals the next bit of the story but (because the player(s) failed the check) the GM doesn't reveal all of it, or reveals it but taxes some resources, or something else that counts as a cost.

Click to expand...

Yes. Although the 'taxes some resource' is - I think - secondary. The main thing is that if the GM has a story to tell, the only way to fail forward is by the GM telling you more story. Because in that style of game the GM is the only one authorised to tell a story.

Which leads to this kind of play: succeed - get told a story. Fail - get told a story.

The details of the story might change (or the GM opts for a simple life and the differences are just colour). Either way, since the player has no knowledge or control over it they can sit on the fairground ride and watch the show.

pemerton said:

It's hard for me to judge how effective this is...

Click to expand...

I'd say it's fairly ineffective. In the GM-sided game being posited it cheats the player(s) of the 'challenge' of finding a way to 'succeed' while simultaneously revealing that the PCs destination is pre-determined anyway.

pemerton said:

"Fail forward" in the sense I'm trying to explain is in lieu of GM plot. Instead of GM plot, the story is created by the checks: succesful checks and things unfold as the player playing his/her PC hoped; failed checks and the things the player cared about are in play, but it's getting worse rather than better.

Click to expand...

Absolutely. In this context 'fail forward' works brilliantly... and the point of my initial post was to reach the point where this distinction had been clearly drawn out.

Job done!

Alright, lots of things I want to address here. I'm not going to quote anyone's particular text here, but it will relate to each participants' (thus far) posts.

1) Fail Forward (FF) vs Success With Complications/At a Cost (SWC)

These are two different GMing techniques or components of a ruleset. I'm going to use Dungeon World and D&D 4e to untangle them.

Dungeon World

The 7-9 move is the primary mechanism through which Dungeon World sustains its "snowballing danger/narrative" effect. 7-9 is "Success With Complications/At a Cost". The player gets some of what they want/moves further toward their goal, but a decision-point which complicates things is introduced or new looming threat emerges to (potentially) interpose itself between the PC/group and their goal.

The 6- result is a hard failure. When this occurs, the immediate situation has changed dramatically and irrevocably for the worse (something bad happens immediately or some impending doom is realized with no chance for mitigation). The gamestate is changed in a way that (a) isn't desirable for the PCs but (b) is compelling to the players because something interesting has happened in relation to their expressed dramatic needs and the game's agenda. They haven't moved closer toward realizing their intent. They've removed further away or failure has been cemented (depending on the fictional positioning in context with GMing principles). This is "Fail Forward."

D&D 4e

In Skill Challenges, all success before the last are the equivalent of the DW 7-9. The last success in a SC is the equivalent of a DW 10+ (where the fictional positioning of the conflict expresses that "this move" will lock in attainment of the sought goal). All failures in a SC should be forward with the situation changing dramatically for the worse or irrevocably if its the 3rd failure of the challenge (thereby cementing defeat in the conflict).


2) How does system (in particular player-facing, codified resolution and transparent principles/agenda which constrain GM decision-making) play into subverting or enabling GM-sided plot control techniques (Force) when deploying SWC or FF?

Simple.

- The more opaque the system is, the more GM Force is enabled.

- The less player-facing a system is, the more GM Force is enabled.

- The more abstract/zoomed out/vanilla a system is with respect to its goals of play, the more GM Force is enabled.

- The more latitude a GM is granted to ignore rules or change things in order to achieve their vision of what should make a good game, the more GM Force is enabled.

- The less a system bakes in an overt premise to be addressed during play, the more GM Force is enabled.

- The less systematized dramatic needs that express PC protagonism are, the more GM Force is enabled.

So, plainly, if there was a continuum of GM Force like so...

<Force Subverted ------------------------------------ Force Enabled>

...you would find Dungeon World, Burning Wheel, Torchbearer on the far left. On the far right, you would find AD&D 2e. Because of the above, SWC and FF works in the service of an emergent narrative for those games on the left. If you grafted SWC and FF onto AD&D 2e's action resolution system, its default would be to work in the service of GM-side plot control.

Manbearcat said:

On the far right, you would find AD&D 2e.

<snip>

If you grafted SWC and FF onto AD&D 2e's action resolution system, its default would be to work in the service of GM-side plot control.

Click to expand...

A complicating factor here is that AD&D doesn't really have an action resolution system! There's combat which has its own fairly detailed system; there are thief abiliites, which at least as presented are purely task resolution except perhaps hide in shadows; and there are some rules for dealing with doors and traps.

Fail forward depends on there being a player intent behind the action declaration, which the GM then draws upon to establish the failure. In AD&D, you might be able to use ability checks or similar to resolve actions and adjudicate them in a fail forward fashion - but this would be complicated by the pretty ad hoc gating of certain capabilities behind non-weapon proficiencies.

And a comment on GM-force: if the GM is providing the intent of the action (eg deciding what will be gained by opening a door, listening at a door, searching for a trap, etc; this goes back to [MENTION=99817]chaochou[/MENTION]'s post upthread) then success with complications tends to mean that the intent is realised but something bad accompanies it; and the "badness" is also a reflection of the GM's intent as to where the "plot" should go. You find the clue, but break your thieve's tools in doing so or You open the door, but make a loud noise.

That is quite different from "fail forward" as I'm putting it forward in this thread, which is about player-established intent.

And on 4e skill challenges - because there is a defined structure for resolution, the way "fail forward" works is a bit different. Within a challenge, success can't be total or else there would be no need to go on - it can either be partial success, or DW 7-9 style "one step forward, one step back", or any other sort of change to the fiction that both (i) respects the success and (ii) maintains the pressure on the PCs (and thus the players) to act. And within a challenge, a failure can likewise be (partial) success with a cost, or just making the situation worse (like DW 6-) - the change to the fiction should (i) give voice to the failure, and thereby (ii) amp up the pressure on the PCs (and thus the players) to act, while (iii) leaving overall success a possibility. Whether a failure inevitably changes the stakes of the challenge (eg take some damage on a failure; make a lasting enemy on a failure) is up for grabs from challenge to challenge, I think, though it probably adds to the sense of "heft" if failures carry weight beyond the resolution of the challenge rather than simply affecting the colour and framing within it.

To elaborate on that last comment about framing within a challenge: a success within a skill challenge should make it more feasible for the players to declare the sorts of actions they want to - it reflects the PCs taking control of the situation; while a failure in the challenge should make it harder for the players to do this, and increase the pressure on them to declare actions that they don't want to (eg the wizard finds him-/herseld needing to take some physical action). And as per the previous paragraph, this is something that is independent of the question of whether failure generates enduring consequences.

(To elaborate further with reference to Burning Wheel: in BW both the melee combat subsystem (Fight!) and the social conflict subsytem (Duel of Wits) operate a little bit like a skill challenge. Failures in Fight! generate lingering consequences - injuries - as well as impacting the framing of downstream declarations within the combat. By default, though, while failures in DoW impact the framing of downstream declarations within the conflict, they tend not to generate lingering consequences - eg affects on reputation or relationships - unless the GM introduces some sort of consequence outside the formal resolution framework.)

In a 4e skill challenge, the ultimate outcome should respect fail forward: if the PCs succeed then their intent is realised; if they fail, then the GM needs to narrate a situation in which they don't get what they want, but stuff that they (and their players) care about is still able to be engaged.

In a fail forward game, if the upshot of resolution is that the GM establishes fiction in which nothing the players/PCs care about is present, then the campaign is over! The PCs have resolved everything that mattered to them in the context of the game.