Failing Forward

[grendel111111]

The choice mattering is not synonymous with player agency, though, which was the notion I was engaging with.

Simple example: suppose the GM has decided that, in land X of the campaign world, holding out one's hand out to another upon meeting them is a grave insult. And suppose that the players (i) do not know, and (ii) inadvertently teleport the PCs to land X (say via teleport mishap, or a Well of Many Worlds, or whatever). The GM tells the players that the PCs appear not far from some NPCs. One of the players then says of his PC, "I walk up to the NPCs, arm extended as if to shake their hands in greeting".

Now the player's choice here matters to the outcome - the GM describes the NPCs as grossly offended and commencing to attack the ill-mannered outlanders - but it was not any sort of exercise of agency by the player (was it?). The relationship between the action declaration and the outcome is mere dumb luck. Like the players choosing to have the PCs go east (and therefore encounter trolls and hydras in the swamp) rather than north (where they will meet the necromancer). If the players don't know these geographical facts, then it's just dumb luck.

In a Gygaxian dungeon, there are a range of resources available to the players to eliminate dumb luck in this sense - divination magic in particular. But once you get to campaign worlds on the scale of some of those you mention (eg FR), then divination magic ceases to be very relevant. It's just luck.

Also: why do you say that in my game direction doesn't matter? If the players (in character) want to find the pyramid the orcs were heading towards, they have to head further east into the Bright Desert. If they want to recuperate in the ruined tower (which they did) then they have to head north to the foothills of the Abor-Alz.

But these choices were not made blind. The players chose to prioritise recupration over exploration, and therefore headed north.

In contrast: when the PCs were drifting through the ocean, hoping to be rescued, we didn't worry about which direction they were drifting in, as there was nothing that turned on them trying to drift one way rather than another.

I used if and it wasn't specifically your game. But in many games where you just "jump" to the next "scene" in the "narrative", direction is largely unimportant. Games such as Leverage you just jump to the targets building, where it is, beyond "somewhere in the city", isn't really important. It is playing the game in a different scale. You jump to the next interesting point.

For me I see the kind of game I like as having lots of "dumb luck" as you put it. As a player if I accept the dumb luck then it is my own stupid fault. Player agency is the players making their own agency (this means the DM has to not shut them down when they make their own agency). They make choices of how much they are going to find out about the surrounding area, what research they do, who they talk to or hire as a guide.
When people talk about player agency it so often sounds like if the DM doesn't give the players all the facts up front then the players are suddenly robbed of all agency, as if they are not capable of using their skills to gather information and make informed decisions. And there are times when it comes down to just blind luck and I am OK with that, too. If I am going to school and I can decide to go with my car on the motorway or take my motorbike on the back roads. It takes about the same amount of time. There is no agency for me because I don't know what might happen on either route. If I go on the motor way and there is a pile up of cars so I am late for school, that is dumb luck. If, before I leave for school, I check the weather conditions and listen to the traffic report then I have given myself agency, I can choose which way is best because I have more information.
Now if I just go and there is a delay I can get angry at (God/ fate/ the GM) because he didn't give me my agency. but it's my responsibility to make my own agency as well.

 

[grendel111111]

Just a quick question/comment around this... How is this any different when you are making things up on the fly. Unless you can guarantee you don't in any way think about the game, it's characters, consequences, what may happen, etc. outside of in the moment play... I'm not sure how you guarantee that your particular "vision" of the game (What you would prefer to improvise around/explore) isn't what you are pushing for, even if it's subconsciously?

As an example, in the previous mountain climbing example, there are an almost unlimited number of outcomes that could take place on a failure (especially since the relationship of skill/task resolution has no bearing on what the failed roll could lead to as an outcome, only what could "logically" arise in the fiction... and yet you as DM have a preference since you are making a specific choice out of all those possibilities... and by the fact that dying and not reaching the mountain top was taken off the table (at least for this particular failure) it seems that you are pushing towards exploration of the content that has already been established... the mountain top and the MacGuffin as opposed to say improvising around another part of the mountain the character could have discovered in his fall... So I'm not seeing how improv protects against this particular problem.

This is something that interests me. If a fail condition is unrelated to the skill being used....you discover an uncomfortable truth for example (your father was not a man of the cloth but a con artist). Once it is used as a fail forward once, but doesn't come about (they passed the test so it is "still up for grabs"). then it is in the DM's head, it's likely to come up as a fail condition of subsequent tests until something else rules it out or it proves to be true.
We know that when you make players roll many dice they will eventually fail (thus we try to limit how many skill rolls they need to make to achieve one objective) So if this uncomfortable truth is used as a fail condition in more than one test eventually it will make it true (because eventually the characters will fail a test). How do you prevent this from happening, other than to say that a passed test also disproves the failed condition?

 

[Neonchameleon]

This is something that interests me. If a fail condition is unrelated to the skill being used....you discover an uncomfortable truth for example (your father was not a man of the cloth but a con artist). Once it is used as a fail forward once, but doesn't come about (they passed the test so it is "still up for grabs"). then it is in the DM's head, it's likely to come up as a fail condition of subsequent tests until something else rules it out or it proves to be true.
We know that when you make players roll many dice they will eventually fail (thus we try to limit how many skill rolls they need to make to achieve one objective) So if this uncomfortable truth is used as a fail condition in more than one test eventually it will make it true (because eventually the characters will fail a test). How do you prevent this from happening, other than to say that a passed test also disproves the failed condition?

Mu.

Normally failing forward and discovering an uncomfortable truth means finding something that would be thematically appropriate (or funny) based on the exact position then and there. So discovering an uncomfortable truth after a climb check might involve climbing in through the wrong window and finding your husband/wife having it off. On another check it won't be anything like as relevant.

 

[Imaro]

Mu.

Normally failing forward and discovering an uncomfortable truth means finding something that would be thematically atppropriate (or funny) based on the exact position then and there. So discovering an uncomfortable truth after a climb check might involve climbing in through the wrong window and finding your husband/wife having it off. On another check it won't be anything like as relevant.

Yes but the temptation to steer failure outcomes towards your own predisposed interests as DM is just as likely as a pre-authoring DM steering outcomes towards the creations he wants to explore... in other words how can the later be a concern when pre-authoring but in improv there is no concern around the DM steering the direction of the "story" being created towards what he is most interested in improv'ing around?

 

[Balesir]

Yeah and I think that stuff is important when the players announce a desire to try a challenge according to some specific parameters or the game operates under those assumptions.

Like "I wanna see if we can beat Tomb of Horrors rules-as-written", etc.

D'oh - here's (part of) my answer about the advantage of pre-authoring, and it's one I should have spotted long since. It's basic Forgeite Gamism - the will to take on a challenge, be it a challenge of skill or of "dumb" luck. If the GM improvises here, it's "cheating". It's not even the situation being a "puzzle" that's important; it's just that the challenge must be set beforehand and administered as written.

What experience do the players (and GM) feel they signed up for?

Yes - always and foremost. Everyone should have a clear idea of what they are expected to get, and that's why all the discussion about potential styles and vocabulary to describe them is useful.

 

[chaochou]

Yes but the temptation to steer failure outcomes towards your own predisposed interests as DM is just as likely as a pre-authoring DM steering outcomes towards the creations he wants to explore.

Prove it.

 

[Zak S]

Yes but the temptation to steer failure outcomes towards your own predisposed interests as DM is just as likely as a pre-authoring DM steering outcomes towards the creations he wants to explore.

This isn't true.

In pre-authoring world location content, you decide subjects, but the players still-in the moment--decide their reactions to those subjects (and one of the cool things about that is you can actually enable situations with more options if you pre-author rather than just go "Uhhh....there's one door!") .

So if we presume the GM wants the players not to be railroaded, (ie you CARE about giving the players their say), pre-authoring content is often the best way to do that because it means you aren't just leaning on your (often repetitive or uni-valent) instincts.

Like in the moment you might go (as a GM) "There's a Frog Temple!"--if you have time to prep you can go "Ok there are eight temples and the order and number you choose to visit them will determine your experience in ways I can't predict".

 

[Manbearcat]

Just a quick question/comment around this... How is this any different when you are making things up on the fly. Unless you can guarantee you don't in any way think about the game, it's characters, consequences, what may happen, etc. outside of in the moment play... I'm not sure how you guarantee that your particular "vision" of the game (What you would prefer to improvise around/explore) isn't what you are pushing for, even if it's subconsciously?

There is a brief, less involved answer to this and a much more complex (and correspondingly less brief) answer to this. Let us start at the brief, less involved one and move to the more complex as we need to.

First, a caveat:

There can be no guarantee that a game is 100 % safeguarded against GM Force. Therefore, I certainly don't offer it. There are only means to mitigate against it. These means come in the form of GM incentives, streamlined/intuitive resolution mechanics (which works to ensure that unwanted mental overhead which distracts a GM from excelling at free-form/improv is minimized), and a clear/transparent/coherent top-down agenda of play and principles for the GM to follow. Games that are meant to be improv/freeform-friendly have the guidance, incentives, and system means in play to support it (we can do examples here if need be).

On with the brief, less involved answer:

The premise I'm working from is that humans are inclined to want to share a creation in proportion to their own investment in it. This investment might be emotional. It might be blood, sweat, and tears. It might be pride/vanity.

The investment (any or all of the above) of pre-game prepped setting and situation (specifically of the granular, high resolution variety) is going to be significant when compared to a game where GM prep is considerably less (typically setting and situation are of much lower granularity/resolution and are firmed up during play with improved content being generated as required; "just in time" or "story now"). Consequently, the seductive forces of "human investment" are less potent. Less potent means less likely to be given sway as the primary motivation for future content introduced into the shared imaginary space during play (rather than content introduction via player decision-points, the output of the resolution mechanics, and following the game's agenda and GMing principles).

As an example, in the previous mountain climbing example, there are an almost unlimited number of outcomes that could take place on a failure (especially since the relationship of skill/task resolution has no bearing on what the failed roll could lead to as an outcome, only what could "logically" arise in the fiction... and yet you as DM have a preference since you are making a specific choice out of all those possibilities... and by the fact that dying and not reaching the mountain top was taken off the table (at least for this particular failure) it seems that you are pushing towards exploration of the content that has already been established... the mountain top and the MacGuffin as opposed to say improvising around another part of the mountain the character could have discovered in his fall... So I'm not seeing how improv protects against this particular problem.

I'm going to forgo getting into this for the time being. However, a polite request. Because Bob (I think I named him?) and his pudding quest were meant for one very narrow purpose (to exhibit a system-neutral depiction of vanilla Fail Forward in action), could we maybe use the play example from my Dungeon World game that I posted upthread (where someone actually does fall down a glacial crevasse). That is an extended example with both play and system context so it would be many times more productive when attempting to use play examples to further conversation.

For now, lets focus on the above (and find common ground or further divergence).

 

[pemerton]

As a player playing a PC that goes to a strange land, I (and my PC) am aware that they may have strange customs that I don't know and I can use that knowledge to try and avoid insult by learning the local customs.

<snip>

Divination magic deals in revealing the unknown. How does it cease to be relevant in a world that contains the unknown? It has more relevance in a pre-authored world that is rich in things to divine.

When people talk about player agency it so often sounds like if the DM doesn't give the players all the facts up front then the players are suddenly robbed of all agency, as if they are not capable of using their skills to gather information and make informed decisions. And there are times when it comes down to just blind luck and I am OK with that, too.

AD&D has no magic for learning local customs, nor any skill system for that either.

The divination magic in AD&D is overwhelming short-range (ie for dungeon use) and aimed at geographical exploration (Secret Door Detection, Clairvoynace etc), or finding treasure (Locate Object, Treasure Finding, etc) or detecting hostile beings (Enemy Detection, ESP, etc). This reflects the game's origin in dungeon exploration and looting.

In my post I talked about instances where players have the resources to learn the relevant backstory. But the example I gave is one where this is not the case. It's an example involving dumb/blind luck.

Many people enjoy a world where their choices matter. While it may be dumb luck if nobody is a sailor or the like, picking a direction in the Realms gives choice meaning.

<snip>

Choice has no real meaning when no matter which direction you choose, it's all going to end up the same.

The last sentence is not in dispute. That's why, in my BW game, when nothing turned on which way the PCs were drifting on the ocean we didn't both worrying about which way they were drifting.

But I prefer a game in which choices matter and the players know that and why they matter. Choosing blind - sticking your hand into the GM's bag and seeing what colour ball you pull out - doesn't give the choice very much meaning, in my view.

 

[pemerton]

Just a quick question/comment around this... How is this any different when you are making things up on the fly. Unless you can guarantee you don't in any way think about the game, it's characters, consequences, what may happen, etc. outside of in the moment play... I'm not sure how you guarantee that your particular "vision" of the game (What you would prefer to improvise around/explore) isn't what you are pushing for, even if it's subconsciously?

the temptation to steer failure outcomes towards your own predisposed interests as DM is just as likely as a pre-authoring DM steering outcomes towards the creations he wants to explore

What's your threshold for, or measure of, GM force?

In my own case, I can tell you absolutely that I spend a lot of time thinking about the game, the characters and what may happen in the game, for both my 4e and my BW campaign. I come up with ideas for antagonists, possible locations, situations etc.

To give a concrete example from my BW game:

After the session in which the PCs escaped the orcs in the desert and took shelter at an oasis with a friendly naga guardian, I thought that I wanted to use a dark elf in the game. (Looking at the chapter on Maeglin in the Silmarillion even led to me re-reading the whole of that book!)

In the next session, when the PCs travelled to the ruined tower, I used the dark elf as a failure result for two checks: a failed orientation check led the PCs to a waterhole which had been recently fouled. Investigation of the excrement suggested that it was elvish. Pursuing the perpetrator led to a brief encounter with the dark elf, who escaped (but lost his knife when he through it and the PCs kept it). Another failed check (I can't remember what, now - maybe the failed attempt to track the dark elf) meant that when they got to the tower the dark elf had also got there first, and hence had had a chance to fill the well with rubble.

In the session when the PCs left the tower, I also used the elf - he dropped a deadfall on the PCs as they were walking through a defile in the Abor-Alz. And this was when it was revealed that he was wielding the nickel-silver mace.

That the dark elf would have the mace had already been anticipated, though, by the player of the mage: in an email following the session where the mace wasn't found, that player conjectured that the mace would be in the hands of the dark elf.

To me, this seems to be a GM doing his/her job. It's my job to manage the fiction and, of particular relevance to this particular matter, to come up with consequences for failure and to frame the PCs into challenging situations.

With this concrete example in mind, can you explain the relationship you see to pre-authoring? To me, it makes a difference that the initial encounters with the dark elf and his filthy deeds were triggered by failed checks; that the mace only came into play at all because a player wrote it into his PC's backstory and Beliefs; that the ruined tower came into play at all only because the same player had it written into his PC's backstory; that the dark elf was intended to (and, in play, did) provide material for the elven ronin PC to play off his Belief that he "will always keep the Elven ways"; etc.

I'm not sure how this would all have been pre-authored, though. If the checks to move through the desert had succeeded, for instance, then there would have been no occasion to introduce the dark elf into play in the way that I did. I might still have used the dark elf as an antagonist with the deadfall - but that itself was triggered by the players deciding to have their PCs leave the tower.

I'm very unclear in what respect you are suggesting that pre-authorship vs "just in time"/"story now" made no difference here, in respect of the balance of GM and player agency.