D&D General [rant]The conservatism of D&D fans is exhausting.

[AbdulAlhazred]

I think the stakes are rarely clear due to downstream effects of any decision and die roll.

Less fiction, more clarity!

I get that.

Some of us view soft moves into hard moves as doing exactly this.

Hmmm, it's hard to analyze that in a general sense, but say looking at a version of the lock picking example, the attempt fails, the GM FFs the character into the house, and makes a soft move by putting a cook into the scene. This one doesn't seem like such a case, even on a DW 7-9, you are still in a position to succeed, you just need to act to avert the obvious hard move that's coming.

 

[pemerton]

I choose not to argue with @pemerton about his example. I can see the argument against the things he mentioned there being fail forward. Maybe we do use the term a bit differently, or maybe there was more to play than came across. Perhaps a bit of each. Perhaps I am simply wrong and he might tell me why, or not. But games like BW focus heavily on intent, so FF is going to often show up in the form of mixed results.

You can see more of my thoughts in the post just above.

"Fail forward" is primarily the idea that nothing happens is not a permissible GM response.

Secondarily - and this is perhaps better captured via the "no whiffing" formulation - is the idea that the narration of a failed check need not present the PC as incompetent. (See eg the example of me failing the test for Aedhros's singing.)

Burning Wheel captures these two things by instructing the GM, when narrating failure, (i) to introduce a complication and (ii) to focus on intent. (i) is what ensures that "nothing happens" is off the table. (ii) leaves open what the GM says about the attempted task, which in turn is what leaves a lot of scope to narrate failure without incompetence.

As I try to explain in that post, I think that @AlViking's response to my examples that they are merely consequence or repercussions is somewhat missing the point: they are consequences other than that nothing happens; and they are consequences of things in the fiction other than the incompetence of the PC.

 

[AbdulAlhazred]

A clear statement of personal preferences, if a little universally phrased. I don't see why you see this as a problem, since it's obviously not the way you play.

I'm simply responding to a description of a problem someone had with a certain situation and pointing out that the problem is with the situation, not the technique.

 

[Maxperson]

The comment I responded to wasn't about causation. It was about the lockpicking attempt having "nothing to do with the cook". Here is the comment:



I don't see how when picking a lock, being discovered by someone who may be on the other side has "nothing to do" with picking the lock. Of course it does!

And you say here that there should be distinct rolls for the two things. So I offered a way that there does not need to be two rolls.

I didn't say that you cannot split it up into two rolls. I didn't say there "should" only be one roll. I just offered an alternative.

We aren't talking about being discovered by someone who is already there before the lock is picked and hears the rogue. We are talking about a quantum cook that only appears in that room if the roll fails.

The cook is not connected to the roll in any way.

I don't know who you mean by "we"... because there have been several people who continue to misunderstand fail forward, or see it as only applicable to narrativist games. That people have different preferences is obvious and really doesn't need to be said anymore.

In this case, you've described two things as being "unconnected" when clearly they may be connected, so I commented to explain why and also how it could be resolved all with one roll.

You absolutely know who I mean by "we." You'd have to be ignorant of the entire thread not to. There are two significant sides in this discussion and that hasn't changed since the OP. Understanding fail forward or not doesn't have any bearing on who "we" are.

 

[pemerton]

Hmmm, it's hard to analyze that in a general sense, but say looking at a version of the lock picking example, the attempt fails, the GM FFs the character into the house, and makes a soft move by putting a cook into the scene. This one doesn't seem like such a case, even on a DW 7-9, you are still in a position to succeed, you just need to act to avert the obvious hard move that's coming.

I think we can take it that the cook screaming is a hard move. So let's reason back - what player-side move failed, such that the cook was startled and screamed? The most obvious candidate is Act Under Fire.

So what was the character doing? Maybe they're the advance scout for the assault on Dremmer's compound:

First, let's imagine the player recites their PC's knowledge - it's a bit artificial as an example of play, but provides some context.

"I know that Dremmer has a storeroom at the edge of the compound, with a gate for taking deliveries. There's a fancy electronic lock on it, so it's not well guarded. I reckon I can crack that lock and sneak in."

The GM nods: "OK, so you're at the gate to the storeroom. It's locked like you expected. It's not well guarded, but that doesn't mean no one ever comes by here. You haven't got all night."

"OK, I bust out my tools and work on the lock, as quickly as I can."

"That Acting Under Fire, and the fire is - you'll be spotted before you're in." The player rolls, and succeeds on a 7 to 9. The GM offers an ugly choice: "You get it open, but you can hear someone's coming. And you can't see yet what's on the other side of the gate. Do you go through into whatever's there? Or wait to see who comes?"

The player decides to go in. "There's someone in there with a torch. Looks like Dremmer's cook Pattycakes, come to grab a fresh bag of chowder powder. What do you do?"

At this point the player has a few choices, but let's suppose that, whatever they do, it fails on a 6 or less. And so the GM narrates that Pattycakes spots them and screams.

I assume that DW could play out in a pretty similar sort of way.

 

[AlViking]

You got a reply to this from @Old Fezziwig. I've also posted about this in many posts in this thread, including some in reply to you.

John Harper is assuming that the fiction that is established during play has a type of emotional and/or dramatic and/or thematic "heft". And that it has a "trajectory" or "momentum" related to that. Here are the examples that he gives:

When you make a regular MC move, all three:​

1. It follows logically from the fiction.​

2. It gives the player an opportunity to react.​

3. It sets you up for a future harder move.​

​

This means, say what happens but stop before the effect, then ask "What do you do?"​

​

- He swings the chainsaw right at your head. What do you do?​

​

• • You sneak into the garage but there's Plover right there, about to notice you any second now. What do you do?
  • She stares at you coldly. 'Leave me alone,' she says. What do you do?

​

​

When you make a hard MC move, both:​

1. It follows logically from the fiction.​

2. It's irrevocable.​

​

This means, say what happens, including the effect, then ask "What do you do?"​

​

- The chainsaw bites into your face, spraying chunks of bloody flesh all over the room. 3-harm and make the harm move!​

​

• • Plover sees you and starts yelling like mad. Intruder!
  • 'Don't come back here again.' She slams the door in your face and you hear the locks click home

.​

​

See how that works? The regular move sets up the hard move. The hard move follows through on the threat established by the regular move.​

Focus particularly on the last example. It assumes that there is a reason the player's character has come to speak to her. And that there is a reason that she has to be cold to the PC. And also that the player cares about how she responds to the PC. It is not assuming a CoC-esque scenario in which the NPC is nothing more than a possible source of clues, and the only "cost" to the player of her slamming the door is that now the clue has to be found some other way. It is assuming human relationships.

If you look at the middle example, you will see that it involves sneaking into somewhere, with the risk of being spotted. It's less intimate and more adventure-y than the last example. I don't think it's that hard to imagine a variant on that in which the risk is not being spotted by Plover, but startling a cook.

(The first example is interesting mostly because it illustrates the difference between how interpersonal violence is approached in AW and how it is approached in D&D. AW uses the interplay of GM-moves and player-moves; whereas D&D uses something much closer to a wargame style of resolution.)

I appreciate the explanation.

Well, if you know what you mean by "fail forward" then I'm not sure why you're asking about it.

Your definition of fail forward seems to be different from what other people stated which became obvious.

But this whole discussion of it began with the contrast between "fail forward" and "nothing happens". And that is how I am familiar with the term "fail forward", which also used to be called "no whiffing".

Notice how none of John Harper's examples is "nothing happens". And in my examples of play, none of them is "nothing happens".

Different games have different approaches.

How you want to read those examples is up to you. Maybe you think the people who posted them are advocating for silly RPGing.

No, just different.

But my guess is this: that with the friend being dead, they are imagining that the player knows that what is at stake is whether or not their PC's friend survives, and they are having their PC climb the cliff to try and rescue them, and - because the roll fails - the PC fails to make the climb in time.

There was no indication that the stakes of the climb were established. In D&D parlance I would call it a ticking clock - if you don't get to the top of the cliff something bad happens. But I would consider it bad judgment on part of a GM to A) not let the players know there was a ticking clock and B) not give them multiple options of approach.

In my games if there is no negative consequence of failing a climb check and you will eventually make the climb but it doesn't matter how long it would take then I wouldn't ask for a check.

And with the screaming cook, I imagine that whoever posted it has something in mind very similar to John Harper's example with Plover.

Notice how, in none of the examples, does "nothing happen".

That's your choice on how to run your game, I don't happen to agree it's best practice for my game.

Aedhros doesn't simply fail to find a kidnap victim - rather, a potential victim escapes, causing word to spread of a knife-wielding assailant. Aedhros doesn't simply fail to bolster his sense of self by singing - rather, he is harassed by a guard. Aedhros doesn't simply fail to have a helpful Elven Etharch turn up - a second guard turns up to join the first.

Notice also that the failure narrations don't have to entail that Aedhros is no good at what he is trying to do. The first tends towards that implication - he can't kidnap someone. But the second doesn't - there's nothing wrong with his singing, but it attracts a guard. And the third is ambiguous - it's true that the guard arrives and no Elves do, but is that because Aedrhos is hopelessly alienated from his fellow Elves, or simply because guards like to travel in pairs?

(Notice also that these guard NPCs have never been mentioned before at the table. The GM nevertheless brings them onto the stage because they are implicit in our shared idea of a pseudo-mediaeval city. Maybe the screaming cook is similarly implicit in an attempt to break into a rich house by sneaking through a kitchen entrance.)

Anyway, these examples are what "fail forward" looks like. I'm sure that everyone's game has some of it. But I'm equally sure that there are some games that don't use it consistently - eg they would narrate the failed Sing test as singing poorly, perhaps for laughs at the table. Or they would resolve the kidnap attempt by first calling for an encounter roll or a Streetwise test to find a victim, and if that failed the narration would be that Aedhros is wandering the city streets but not finding anyone - Aedhros goes out looking for a victim, but nothing happens.

It's what fail forward looks like to you. It is not how I've seen fail forward explained anywhere else.

 

[Maxperson]

Are you an expert on satellite imaging? Or do you rely on authorities to assure you that Google street view is an accurate representation?

What authority do you have standing over your shoulder telling you that the Google street view is accurate or not? That's just weird.

 

[AlViking]

You can see more of my thoughts in the post just above.

"Fail forward" is primarily the idea that nothing happens is not a permissible GM response.

Secondarily - and this is perhaps better captured via the "no whiffing" formulation - is the idea that the narration of a failed check need not present the PC as incompetent. (See eg the example of me failing the test for Aedhros's singing.)

Burning Wheel captures these two things by instructing the GM, when narrating failure, (i) to introduce a complication and (ii) to focus on intent. (i) is what ensures that "nothing happens" is off the table. (ii) leaves open what the GM says about the attempted task, which in turn is what leaves a lot of scope to narrate failure without incompetence.

As I try to explain in that post, I think that @AlViking's response to my examples that they are merely consequence or repercussions is somewhat missing the point: they are consequences other than that nothing happens; and they are consequences of things in the fiction other than the incompetence of the PC.

They are logical consequences to failure that you've implemented because you're playing a style of game where something must happen. But they are logical and connected directly to the failure in a cause-and-effect way.

I don't have a problem with nothing happening because there will always be other ways for the players to drive the action forward, as GM I provide opportunities that the characters can pursue I don't drive the flow of play.

 

[Maxperson]

To be honest, I'm just taking other people's word for it that Australia is real.

If it isn't, some other place owes us an apology for Paul Hogan.

 

[Maxperson]

The real question is did they account for the fact that the people are driving upside down because they're in Australia?

Psh! Everyone knows you dig a hole down to China, not Australia.