Failing Forward

[Mustrum_Ridcully]

I like the concept.

Now this is muddying things for me, because if it is just a matter of setting the stakes, I don't really see what fail forward is. If it is a matter of taking a failure and turning it into something more productive for the adventure or storyline, that I can grasp. But in any game, the GM is setting the stakes for failure. To me fail forward sounds like it is meant to sidestep the initially set stakes (i.e. stake seems to be you tumble to your death or fall down the side of the mountain, but in actuality once the failed roll occurs, it is about losing a vital piece of equipment or not----so falling down the ravine was never really a potential outcome in hindsight).

The interesting aspect of this - if your character falls down, he has no choice but to pick himself up. But if his pudding detector is falling, he can try to go without it, or continue. The fail forward aproach here has created more possible outcomes of the scenario.

 

[Maxperson]

I agree with your disagreement. I should have worded it slightly differently.

A skill check doesn't have to be so narrowly defined, and it shouldn't be presented as if that's the only option.

What I was really getting at is that a lot of people present rules like skill checks as being an either/or scenario, and claim that the rules restrict them to only that interpretation. Then I've seen newbie DMs (including a few threads ongoing right now), adhering to an interpretation of the rules that is limiting at best, and killing the game at worst. So I feel it's important for people, especially those new to the game, to understand that the DM always has the option to consider more than just an either/or scenario.

I agree with his disagreement and your agreement with it.

I'm coming at this from my perspective, not trying to say that it's the only way to do things.

'I kick in the door.' Makes a Strength check, and fails.
'The door is stronger than you think, and doesn't budge.' DM also makes a check to see if the orcs down the hall hear the attempt.

The reality is that the second check isn't necessarily needed. As a DM, you have a pretty good idea how far the sound of a strong kick on a door will travel, and you know what the orcs are doing and how attentive they are. Failure to kick in the door is enough to warrant considering whether they hear it or not, and you can use the amount of the failure to help determine that. I get that the two are only indirectly connected, and that more checks could give you potentially more possibilities.

The orcs may be alerted even upon successfully kicking in the door. That the orcs are alerted shouldn't really be a surprise to anyone, because kicking in a door isn't a very stealthy way to get in.

Right. It's not success with a cost or fail with a cost. That particular cost is there no matter what. All that remains is whether the attempt was successful or not, and neither I nor my players want to always succeed. Failure and reassessment should be an option.

If they were trying to pick a lock and enter stealthily, you could save some steps by indicating that a minor failure is a success in picking the lock, but the door squeaks loudly as they open it, alerting the orcs. A major failure indicates that picking the lock didn't work, you can continue to try (in which case I'd just use my time-based option), or you can try an alternative way to enter. Yes, this could be a series of checks (failure, success, stealth, perception), but it doesn't need to be that complicated. The action moves forward, and an interesting scene is set through the use of a single check which is determining 'did they successfully bypass the lock and enter the room without detection' rather than 'you failed, try again, OK the lock is open, what do you do? Enter stealthily, etc. Also, do you make a Stealth/Perception check on the failed check, successful check, opening the door, all three, or two of the three, etc.?

See, we would rather the failure be a failure and then perhaps decide to have Grok kick down the door. The action still moves forward with that failure, but the failure didn't equal a success.

Neither approach is inherently better or worse. But the first option streamlines it and moves the game forward without extraneous checks that as a whole don't add any more value to the game and the story than the first option.

Value is added. The ability to fail is an added value. As is the added worry that comes over having to now kick open the door, which is something the PCs KNOW makes a lot of noise. The drama shifts.

But you're writing a story, and part of a good story is, well, a good story. If you make a check for the orcs, and they fail, then there's no additional story, no additional complication. But if you go with the (aargh!) fail forward approach, then the story evolves based around that single roll. This is simpler, streamlined, and also allows the DM to go with what makes sense rather than just more die rolls.

I disagree. Later on when in the tavern, the PCs can tell the story of the deaf orcs who were so drunk (or whatever_ that they couldn't hear a door being kicked in right down the hall. It is all part of the story, not just action.

Second, proponents of the technique, and the very description of the technique, needs to retain the original possibilities and expand upon them. Most of the discussions tend to either ignore, or specifically advise against outright failure. I think that is a mistake. I think that failure is an option, and although sometimes that option isn't the most interesting option, it still must be an option - provided it serves a purpose. This speaks in part about good design in where an obstacle must serve a purpose. My opinion on what serves a purpose, though, often goes against what is usually recommended.

I think realism is interesting and fun. The most interesting option is not always going to be the best one. Which option is best depends on circumstances, the players, and realism (realism is not mirroring reality).

One example is a locked door that serves as the entrance to a dungeon. Outright failure is not an option, they must get through the door. But if the rogue fails to pick the lock there are still other options. A knock spell. Kicking in the door. Finding an alternative entrance, etc. Fail forward proponents would probably recommend not having an outright fail option.

Other options = fail forward, though. Just not in the same way as failing and picking the lock anyway. I haven't seen a DM just have an outright failure with no other way to succeed since I was in high school and shortly after and we just didn't know better. That's why I say that there are two types of fail forward. The type where if you fail, the action moves forward down another path, and the type where if you fail you succeed anyway and add in a cost. I love the former and use it constantly, but rarely use the latter.

Second, current design concepts recommend placing things like locked doors unless there's a purpose for them. This also seems to be a common recommendation regarding random-based adventures. But people lock doors. It might just be their bedroom, with little of value, and doesn't further the story as well. But it does further the goal of building a believable and immersive world.

Right, and for a bunch of us, that believable and immersive world is a necessary part of the story, so it does further it well.

Your post was very good by the way.

 

[Emerikol]

What you are setting out to do in your game does effect choices that you make. Writing a story implies that the key result is the story that happens, Things should be chosen based on what would make a good story (but from whose perspective, the players, the DM, an outside observer? it could be all 3). Then story elements are key, build to a climax, a certain level of script immunity is needed, keep the story moving to a resolution, "Random" events happen not randomly, but because they add to and enhance the story.

On the other hand if your goal is to experience a world through the game, then different choices can be made. Random things happen because in life random things happen. Hitting a dead end and having to find another way to achieve you goals or even needing to change your goals, is perfectly fine if you're not tied to "story". Characters don't have to die in a story satisfying way, because it is a game and the mechanics have given the result you got. "Time for a new clone". I had a DM ask after one game where my character died if I wanted him to save my character as my story wasn't "finished". But to me his story was finished, he didn't reach all his goals, but that is part of life.

People who have a preference for one style of game over the other will see elements of the other approach as a failing and unsatisfying because it's not meeting their gaming needs.

This is a pretty good explanation. It really is a two totally different styles. Fail forward is an "of course!" idea in the story driven approach and it is not in the game approach.

When I play I want no forgone conclusions. I like sandbox worlds. I want the players to literally do whatever they want within those limitations. Life and death is determined by the dice and by good play. Victory is achieving character goals that the player as the character has set. The player never intrudes on the game except through his character. The DM spends a lot of time building a really complex and interesting sandbox that his players can explore and enjoy.

That is basically my style learned at the knee of Gygax. I'm not claiming any superiority for this style other than that it is what I enjoy. I believe it was the original roleplaying style because when coming from a world of no roleplaying the game approach is the obvious first step. I also realize the story approach has evolved out of the game approach with people wanting to play differently and in their minds better and likely for them it is better.

I don't think there is any solution to our debate. I enjoy the discussion. I just think matters of taste are hard to debate. I like X and someone else likes Y. What more is there to say?

What I do dislike is WOTC presenting ANY style as a given. I think the DMG needs to spend some time explaining to DMs how to run a great game in the style they prefer. I do think a lot of this stuff is very much a DM/player decision and maybe only needs some guidance on how to use the rules.

 

[Ilbranteloth]

You may find that some people disagree about "you're writing a story" and rather that you are experiencing a world, and that makes a difference.
What you are setting out to do in your game does effect choices that you make. Writing a story implies that the key result is the story that happens, Things should be chosen based on what would make a good story (but from whose perspective, the players, the DM, an outside observer? it could be all 3). Then story elements are key, build to a climax, a certain level of script immunity is needed, keep the story moving to a resolution, "Random" events happen not randomly, but because they add to and enhance the story.

On the other hand if your goal is to experience a world through the game, then different choices can be made. Random things happen because in life random things happen. Hitting a dead end and having to find another way to achieve you goals or even needing to change your goals, is perfectly fine if you're not tied to "story". Characters don't have to die in a story satisfying way, because it is a game and the mechanics have given the result you got. "Time for a new clone". I had a DM ask after one game where my character died if I wanted him to save my character as my story wasn't "finished". But to me his story was finished, he didn't reach all his goals, but that is part of life.

People who have a preference for one style of game over the other will see elements of the other approach as a failing and unsatisfying because it's not meeting their gaming needs.

I guess I view writing a story as a bit different. Even if you're 'experiencing the world' you're still writing a story. The whole point of a role-playing game to me is that you are writing a shared 'story', perhaps we call it a 'shared experience.'

You don't need a pre-written story arc. But even if you are simply experiencing the world, the people in the world have 'stories' of their own. There are groups of people that have shared goals.

I've come to really like the approach of looking at RPGs as a TV show. Some shows have an underlying story that drives the entire show. Others are just the day-to-day events surrounding an individual or group of individuals with no greater story tying them together other than a job, for example. Many try to do both. Most of the time I find the ones that do both to be rather annoying. For example, I enjoyed The Mentalist, but really didn't like the Red John story. It wasn't necessary, and eventually killed the show, because once they finished their story, and screwed up all of the dynamics of the characters and what they did, they didn't have a show anymore.

My home campaign right now is more of what I'd call a heroic story. Like something along the lines of Lord of the Rings. They are part of something bigger that's happening and they are among the only people that might be able to do something. Like the first couple of 5th ed D&D adventure paths, a lot of what happens is due to working with others, and organizations like the Lord's Alliance and Harpers. They want to know what's going on, so they remain engaged. They are, however, free to go whatever direction they like. And the things they stumble upon aren't all related to this epic story, and they can choose not to engage with it any time they'd like.

The campaign I'm starting at a store, however, is more of the experience the world type. I'm modeling it after the early Gygax home adventures, as well as what I envision Ed Greenwood's Shadowdale campaign was like. Each player will have multiple characters prepared, and they are all starting from the same 'home base'. Because I anticipate that there may be different players from one session to the next I'm making sure that each week we can switch up which characters are in play, while the others are involved in downtime activities until the next time they are active.

I see this as a more typical tv show where there's an ensemble cast of characters, with probably a few that are the 'main' characters in most every episode (since they are the ones that will show up every session), and the 'story' is self contained in that episode. The story is one of shared experiences, and doesn't need to go beyond that, although the options are always there to do so, since there are always things going on in the background.

Regardless, the unique thing about a role-playing game is this shared experience. The story.

That's one of the reasons why I think it's important for those promoting 'fail forward' write about how it will function well in any play style. It's another reason why I think 'success at a cost' and 'fail forward' are both bad terms since them imply you aren't failing, but that you always move 'forward'. I get the originally the idea was that the story moves forward, but it's too easily misunderstood.

I guess it's really an idea of moving beyond a binary skill check to a degree of success/failure check that also pays attention to the context and possibilities. So what, a 'possibility check?' I don't really like that either, though. But to me the concept should remind people that, unlike a video game, there are an endless number of variables and possible outcomes whenever an interesting situation arises that have a possibility for failure, or to alter the direction of the story or experience.

As I'm thinking this through, I guess there's a question of what the benefit of this technique, whatever it's called, is trying to accomplish.

So here are some thoughts to outline what I think are goals, or maybe pros and cons. Coming in part from a process design background, here's one way to look at it.

You can look at an entire game session as one process. Within that are multiple sub-processes. The most well defined sub-process in most games is combat.

So we're looking for a process to handle the non-combat events where a character's skills are tested, and a certain outcome is not guaranteed.

I'll go back to the simple scenario of a locked door leading into a room of orcs.

Depending on the PCs, the goal of this particular event is variable. I'll give a few possibilities.

They want to sneak into the room to steal something, preferably remaining undetected.
They want to burst into the room to surprise their enemies so they can eliminate them quickly, more easily, and without casualties or being injured.
They want to sneak through the room to another exit.

There are other possibilities, but these are good enough to start.

Now at it's most basic, you could have the party state their goal (pick one of the above), make a check and determine whether they succeed or fail. You can narrate a dramatic encounter based on that check, and the check can return degrees of success and failure.

But this is really boring. Yes, the success and failure may be determined by that skills of the PCs, but ultimately it's largely just the DM narrating a story.

So then we can go with the opposite. Everything is determined through the rules system with a series of checks. Going with option 1.

Stealth check for each character to get to the door, with a perception check for each of the orcs.
Listen at the door check, with another round of stealth and perception (assume there's a stealth and perception check in every round that a PC does something that might make noise).
If not successful, they spend another round of checks to listen again to make sure.
They hear that there is something in the room, now check to see if they determine: what, how many, make out any conversation, etc.
Pick locks check, with the usual stealth/perception.
If failed, repeat.
Success, stealth, perception, and check to see if the door sticks or squeaks.
Door opened, OK, stealth, perception, perception on the part of the PCs to see what they can spot through the partially opened door, etc.
Oh, I forgot to add a check to see if the dusty conditions caused one of the characters to sneeze or cough.

Anyway, I think we get the point. There's a stage where we have too many checks, and not enough story. Actually, let's call it narrative.

Combat in game systems is mechanically very mature, and there are a lot of choices for the style and approach to combat across game systems.

Non-combat action resolution is not nearly as mature. I like the idea of trying to do something better, but I think the approaches described have been partial solutions, often concepts, that are not quantified in an easy to explain way. I think before we get to that, we need to verify what we're trying to accomplish.

So what I think we're really trying to accomplish is:
A 'better' way to determine the outcome of a non-combat activity/event/decision point in the game.
It must take into account the skill level/abilities of the PCs to the degree that a PC with a higher skill level should be more likely to succeed.
It must be granular enough to feel like the PCs are accomplishing something, provide results on a character by character basis (one continues to listen at the door while the other picks the lock), but not generate an 'excessive' number of checks.
We need a way to account for many variables, including the environment (squeaky door), fate (sneeze), skills of others (perception), etc.
We want a method that adds to the narrative in a productive way. It can add to the narrative in a direct or indirect way (failed to pick the lock, squeaky door, sneeze), but should not preclude any possibilities - it can't guarantee success on a failed check, nor preclude total failure.
The check should take into account the outcome of other related checks or events. (a failed pick lock check might impact the stealth check, or the amount of time, such as 3 or 4 rounds to pick the lock might impact the stealth check).
It should include as many of the players as possible, but also allow the key individuals be the focus.

It should be universal (not tied to any specific rule set, skill set, etc.) but scalable, to allow different play styles to modify the granularity easily. Ultimately it should enable the DM to adjudicate the game in their style, with their goals in mind, and provide the framework for an exciting and fun game for the players.

What would you add/remove or change in this list. It's just a brainstorm essentially, and I've got a thick skin, so let's see what we can come up with. Part of that process should be to come up with a proper and appropriate name for it.

Ilbranteloth

 

[Emerikol]

I think the distinction between the two styles is this:

1. One style actively involves the DM and players teaming at the metagame level to ensure that the story outcome is interesting and exciting. So players will happily throw their characters into trouble spots via some metagame construct even if the character would never want such a thing.

2. The second style involves the DM creating a world full of challenges. The players as their characters work as a team to overcome those challenges. The primary objective is winning as a team. Overcoming the obstacles successfully. You do this by knowing your abilities and using them effectively in game. The DM's job is not to defeat the players but to accurately play the npc/monsters as if they were also part of the world. This style is very gamist. Traps exist to stop the characters from achieving their goals.


I prefer #2. Now I believe in both instances interesting stories can happen. Many of my friends and I reminisce about past adventures and the fun had. We never though at the time were seeking an interesting story. The story was presumed in the challenge. Overcoming the challenges were what focused the party goals.

In a game like I prefer selecting the right equipment for an adventure is a big deal. What you have is what you have. Preparation is part of the challenge. Choosing spells is also important. It's very much like preparing for war. Failure is absolutely a possibility. Characters die and die permanently sometimes. Because games are losable. It's a cooperative game. The DM is not going to save you from a silly death. Be prepared and play well.

 

[innerdude]

So what I think we're really trying to accomplish is:
A 'better' way to determine the outcome of a non-combat activity/event/decision point in the game.
It must take into account the skill level/abilities of the PCs to the degree that a PC with a higher skill level should be more likely to succeed.
It must be granular enough to feel like the PCs are accomplishing something, provide results on a character by character basis (one continues to listen at the door while the other picks the lock), but not generate an 'excessive' number of checks.
We need a way to account for many variables, including the environment (squeaky door), fate (sneeze), skills of others (perception), etc.
We want a method that adds to the narrative in a productive way. It can add to the narrative in a direct or indirect way (failed to pick the lock, squeaky door, sneeze), but should not preclude any possibilities - it can't guarantee success on a failed check, nor preclude total failure.
The check should take into account the outcome of other related checks or events. (a failed pick lock check might impact the stealth check, or the amount of time, such as 3 or 4 rounds to pick the lock might impact the stealth check).
It should include as many of the players as possible, but also allow the key individuals be the focus.

It should be universal (not tied to any specific rule set, skill set, etc.) but scalable, to allow different play styles to modify the granularity easily. Ultimately it should enable the DM to adjudicate the game in their style, with their goals in mind, and provide the framework for an exciting and fun game for the players.

First off, really good description of elements. I think you've identified most of the core, relevant goals for "action resolution paradigm X" (since we don't want to call it "fail forward" or "success at a cost").

Interestingly, your list sounds very much like a conversation the people at Fantasy Flight had when they made the new Star Wars/Warhammer Fantasy RPG games, because it seems like the action dice mechanic those systems use is based on this kind paradigm.

As I recall, the Fantasy Flight "resolution system" assumes that the following basic outcomes are possible for a given check:

• Overwhelming success -- Success at the described task, plus a situational boon/bonus/something cool that improves the party/player's position.
• Basic success -- You succeeded at the described task at hand, without any additional boon or hindrance.
• Success with a hindrance -- You succeeded at the task at hand, but something negative occurs which may affect future outcomes/decisions. I haven't played the game, but it's my impression that the defined hindrance should typically be applied to the immediate situation (as opposed to say, some situation that happens a week later).
• Failure with a potential benefit --- You failed at the described task, but something happened that may situationally help the player/party.
• Basic failure --- You failed at the task, with no other situational boon or hindrance.
• Catastrophic failure --- You failed at the task, and something situationally negative further hinders the player/party beyond the described task resolution.

These are hard coded into the dice mechanics and skill system. But in looking at it, the only difference between this and what I do with Savage Worlds is the middle two tiers. Savage Worlds already has overwhelming success, basic success, failure, and catastrophic failure built into its dice mechanics as well.

Thus, the middle two segments are likely the ones that are most relevant to the idea of "fail forward"/"success at cost"/Unnamed Resolution Paradigm X.

In the case of the Fantasy Flight mechanic, though, one of the key points is that by hard coding it into the dice it removes some of the judgement call from the GM. Yes, the GM and player ultimately have to agree on the nature of a boon/hindrance when it happens, but the GM isn't deciding when one should be present --- it's the dice deciding that. The GM doesn't get to arbitrarily decide BOTH that a boon/hindrance happen AND the nature of that boon/hindrance.

I think this approach also assumes some player collaboration and input to avoid abuse. If a GM's hindrances are things like, "You catastrophically failed your attack. So let's see, you drop your sword on your foot, take 20 hp damage and suffer a penalty to movement," the players are going to rebel pretty quickly.

The simplest way to apply this same kind of result without having to use the Fantasy Flight system is to first houserule a degree of success / degree of failure to dice throws. In D&D, set a range (say, +2 / -2 from the DC of the check) that is either standard success or failure, and then move up and down the scale to assign the appropriate degree thresholds.

The second thing to do would be to add a single Fudge / Fate die to your roll --- A blank means no boon or hindrance, a + means a boon, a - means a hindrance. The other way to do it would be to add a simple d6 "fate die" roll to every check. A 6 on your "fate" die means a boon; a 1 means a hindrance.

 

[Manbearcat]

I think the distinction between the two styles is this:

1. One style actively involves the DM and players teaming at the metagame level to ensure that the story outcome is interesting and exciting. So players will happily throw their characters into trouble spots via some metagame construct even if the character would never want such a thing.

2. The second style involves the DM creating a world full of challenges. The players as their characters work as a team to overcome those challenges. The primary objective is winning as a team. Overcoming the obstacles successfully. You do this by knowing your abilities and using them effectively in game. The DM's job is not to defeat the players but to accurately play the npc/monsters as if they were also part of the world. This style is very gamist. Traps exist to stop the characters from achieving their goals.

I prefer #2. Now I believe in both instances interesting stories can happen. Many of my friends and I reminisce about past adventures and the fun had. We never though at the time were seeking an interesting story. The story was presumed in the challenge. Overcoming the challenges were what focused the party goals.

In a game like I prefer selecting the right equipment for an adventure is a big deal. What you have is what you have. Preparation is part of the challenge. Choosing spells is also important. It's very much like preparing for war. Failure is absolutely a possibility. Characters die and die permanently sometimes. Because games are losable. It's a cooperative game. The DM is not going to save you from a silly death. Be prepared and play well.

Unfortunately this is a something of a false dichotomy. Since I've been using Dungeon World primarily to discuss things in this thread, I'm going to answer [MENTION=42582]pemerton[/MENTION]'s invocation of the Powered By the Apocalypse engine games below while addressing this contention directly above.

In fact there is nothing wrong with failure always producing interesting consequences which drive the action on - which is what game designers like Luke Crane, Robin Laws, Ron Edwards, Jonathan Tweet, Vincent Baker etc have in mind in advocating "fail forward" as a technique. This is completely orthogonal to whether the PCs always, frequently or only sometimes get what they want. In 4e, which is a very heroic game, the tendency is towards "frequently". In BW, which tends towards grittiness, it is more like "sometimes". Others who know the Apocalypse engine better than me can comment on the sort of frequency of success it tends to produce. But all these systems deploy "fail forward" in the sense of "the consequences of failure should be a challenging new situation that drives the action onward."

I would have to assume you place Dungeon World in your 1 above. Its system (GMing advice, player incentives, rules & resolution mechanics including its normal distribution of outcomes) pushes play coherently (and very tightly) toward an inevitable and endless snowballing of adventure and danger (which may very lead to death...legitimate death driven by player autonomy...not GM "allowing it to happen"). It achieves this snowballing of adventure and danger in no small part due to the bell curve of outcomes that the system's math was engineered to achieve. It puts the preponderance of outcomes in the 7-9 range. This is the "best" outcome because it sustains the tension/momentum of danger/action/adventure and uncertainty as it yields success but with a worse outcome than you'd like, resulting in a hard bargain where you're giving up something you'd rather not, or an ugly choice between imperfect results.

The prospect of GM force (subordination/suspension of the action resolution mechanics/game rules in order to create outcomes that the GM is inclined toward) is as completely muted as can be in TTRPGs. This is because:

A) the "How to GM" explicitly calls for "Follow the rules" as one of its four tenets. These rules are simple, utterly clear, coherent, and transparent.

B) "Play to find out what happens" is one of the three facets of the GM's top-down agenda.

Combined, these two basically serve as the antithesis of GM Force as, effectively, the rallying cry for GM Force is White Wolf's Golden Rule (which AD&D 2e then nabbed in its rule 0): The GM may ignore or change any rule at any time for the sake of the story (which means a prescripted metaplot or backstory reveal...which you means you aren't "playing to find out what happens"...as its already happened!).

So let us examine the contention in 1 above in greater detail:

1. One style actively involves the DM and players teaming at the metagame level to ensure that the story outcome is interesting and exciting.

You may be using "teaming" here in a nuanced way. If so, I'd need to know more, but at a glance, this likely isn't accurate in the way you're meaning it. The only "teaming at the metagame level" that takes place would likely be:

- pre-play during the map-drawing process. The GM might solicit player's input for adventuring sites.

- pre-play during character building. The GM might make suggestion for interesting Bonds to resolve during play (this is another of the "adventure/danger feedback loop" drivers that sustains the momentum of play and incentivizes players, through their PCs, to do so).

- in making the map, the GMing Principles direct the GM to "Draw maps, leave blanks" and to "Ask questions and use the answers". This is to (i) help support "playing to find out what happens" while (ii) ensuring that you have (a) only what you absolutely need (this is what your prep is for...functional prep is very much detailed in the GMing section) to "Portray a Fantastic World" and "Fill the Characters’ Lives With Adventure" (the other two components of the game's GMing Agenda)...and nothing more. The rest will emerge through play and be established "on-screen" as you need it to fulfill your responsibilities as GM (a pliable backstory is essential for these types of games to work).

So players will happily throw their characters into trouble spots via some metagame construct even if the character would never want such a thing.

It seems to me that you're likely invoking games with a plot point economy here; Aspects/Fate Points in Fate here or Distinctions (et al)/Plot Points in Cortex+. I don't agree with your interpretation of the PC build components ("characters would never want such a thing") of Aspects/Dinstinctions here (they can easily be, and most often are, tied to things in life that the PC is aware of and deeply invested in...so when they come up in play, the character responding to them in the way the player declares is entirely coherent). However, I'm not using those games as an example so I won't go deeper into it.

The Dungeon World Rewards Cycle incentivizes players to play their PC based off:

* Their Alignment statement (the fulfillment of which during a session provides End of Session xp) and their Bonds (the resolution of which during a session provides End of Session xp).

* The understanding that failure and curiosity/discovery are the greatest teachers of all. Every outright failure earns you a mess that you have to get yourself out of, but also earns you xp. Discovery also earns you 1 End of Session XP.

The only aspect of the Rewards Cycle that is metagame referential is the classic D&D component of Rewards Cycle of which no one seems to have a problem. Slaying a notable monster and looting a notable treasure earns you 1 xp for either of those that you can cross off your list.

Now let us examine 2 above and see how nicely Dungeon World plays with it:

2. The second style involves the DM creating a world full of challenges.

The Dungeon World GMing advice instructs to do this in both prep and during play.

The players as their characters work as a team to overcome those challenges.

The basic conversation of play, the PC build mechanics, the GMing advice, and the focus on the fiction (which triggers "moves" and the resolution mechanics) put players in the emotional state and OODA loop of their characters. The game talks about "danger" and "adventure". Insofar as those are "challenges", the game that spills out of play definitely promotes the above.

The primary objective is winning as a team. Overcoming the obstacles successfully. You do this by knowing your abilities and using them effectively in game.

Dungeon World is a resource-intensive game that requires players make functional use of (a) HPs (which don't grow from level 1 onward - unless you raise Con by 1 when you level...which improves HPs by 1), (b) several types of broad/open-descriptor item types (including uses of Adventuring Gear, Rations, Ammo), (c) coin (for Resupplying and purchasing hirelings/services), (d) management of a comparatively paltry number of spell slots for primary casters (who also have to roll dice just to properly "Cast a Spell"!) to survive a life imperiled by endless danger and adventure.

This is definitely central to play.

The DM's job is not to defeat the players but to accurately play the npc/monsters as if they were also part of the world. <snip> Traps exist to stop the characters from achieving their goals.

The GMs job, as noted above is to:

General:

* Describe the situation
* Follow the rules
* Make moves
* (Prep and) Exploit your prep

Agenda:

* Portray a fantastic world
* Fill the characters’ lives with adventure
* Play to find out what happens

Principles:

* Begin and end with the fiction
* Draw maps, leave blanks
* Address the characters, not the players
* Embrace the fantastic
* Make a move that follows (from the fiction that immediately preceded it)
* Never speak the name of your move
* Give every monster life
* Name every person
* Ask questions and use the answers
* Be a fan of the characters (in your disposition with your players...not in your GMing of conflicts)
* Think dangerous
* Think offscreen, too

As I posted upthread, This and this is an example of play. It exhibits Failing Forward in a way that can be entirely lethal at the end. At the end, Otthor's strength fails him and he falls into the darkness of the glacial crevasse, splashing down into the hypothermia-inducing underground river. That river spills him out at the PC's ultimate destination!

That fall could have killed Otthor. The 2nd order effect of hypothermia could have killed him (it was a danger in the following scenes that had to be dealt with). The Roper and the Darkmantle that were the Hobgoblin Dragon Sorcerer King's pets and he had put in the basement of his stronghold to (a) serve as garbage removal and (b) protect the underground access?...they definitely could have killed him! The fact that he was forced into a position to have to "deal with them" (he slew them, thus wrong-footing their future Parlay with the King to petition him and his Dragon overlord to assist them against the Aboleth intrusion from the Far Realm?) could have killed him (and his companions, by-proxy!).

That is what Fail Forward does. It doesn't mean that danger and death are removed from the equation of play. Not at all. It means player goals are compromised (sometimes into a death spiral!) and they have to rebound/rally, be resilient to setbacks, and figure out a way forward. It means that narrative momentum/tension & excitement at the table is maintained and play is always propelled by "something interesting/dangerous" happens.

 

[Ilbranteloth]

I think the distinction between the two styles is this:

1. One style actively involves the DM and players teaming at the metagame level to ensure that the story outcome is interesting and exciting. So players will happily throw their characters into trouble spots via some metagame construct even if the character would never want such a thing.

2. The second style involves the DM creating a world full of challenges. The players as their characters work as a team to overcome those challenges. The primary objective is winning as a team. Overcoming the obstacles successfully. You do this by knowing your abilities and using them effectively in game. The DM's job is not to defeat the players but to accurately play the npc/monsters as if they were also part of the world. This style is very gamist. Traps exist to stop the characters from achieving their goals.


I prefer #2. Now I believe in both instances interesting stories can happen. Many of my friends and I reminisce about past adventures and the fun had. We never though at the time were seeking an interesting story. The story was presumed in the challenge. Overcoming the challenges were what focused the party goals.

In a game like I prefer selecting the right equipment for an adventure is a big deal. What you have is what you have. Preparation is part of the challenge. Choosing spells is also important. It's very much like preparing for war. Failure is absolutely a possibility. Characters die and die permanently sometimes. Because games are losable. It's a cooperative game. The DM is not going to save you from a silly death. Be prepared and play well.

I get #2, but #1 doesn't fit any type of game I've run, although I think you're trying to describe what others would consider a story-driven campaign.

I have two basic types of campaigns I run, and what seems to differentiate them at one level is the amount of commitment the players have toward their characters. This is not definitive, and doesn't always happen this way, but it seems to be more common than not.

Campaign type #1, which my home campaigns tend to be, have a story arc running through them that's typically on a larger scale. I don't typically have 'save the world' type stories, but something similar and on a smaller scale, such as a region. Sometimes it's not related to a geographical conflict, but it has the same sense of scale.

In these campaigns, the 'main' story is sometimes the focus, sometimes not. I play off the backgrounds of the characters, as well as things the players say, to flesh out the story, so the characters have direct and indirect links to the actual story. There are a great many other events going on, because the world as a whole is fleshed out. Some groups, mysteries, and plots are related to the main story, some not. Those that have activities related to the ongoing main plot also have other things that aren't related going on too. The campaigns typically take us years to get through the major stories, and could be compared to the published adventure paths in that it follows a group of people as they increase in power (levels) from novice to hero (1st to 10th+).

I would characterize these as 'heroic' stories, and while there will be characters, and players that come and go, the core group typically remains consistent and the characters are integral to the story. So they tend not to die as much, or if they do it's notable, and they may be raised.

Campaign type #2, which my public campaigns tend to be, are what I characterize as Gygaxian, or Greenwood Shadowdale types. The players have multiple characters, and the group of players may change from week to week. Most adventurers are short, single-session scenarios in which the PCs leave and return to their home base (usually a town or city) at the end of the session. During the next session it's probably a different group from the pool of characters. The non-active characters are involved in downtime activities and such.

These are what I would characterize as more sandbox oriented. I still have all sorts of plots and activities going on, and I still tie the characters into the campaign. But most of them don't have the same investment in their characters. If somebody dies, they aren't often as concerned about raising them.

These aren't textbook cases of 'story' vs 'sandbox' play, but I think they are pretty close. And the biggest differentiator in the groups I've played is the investment they make in their characters. Since they might play a different one of their characters each week, and they aren't the protagonists in some heroic or epic quest, they are much less concerned about the death of their characters.

We do keep track of stuff like equipment, ammunition, encumbrance, resource management, etc. People who have played in both don't see a meaningful difference in how I handle things. The world is the same, the events happening in the background are the same (in fact they are usually both running at during the same 'game time' and their characters may even be in the same place at the same time. In fact, I currently have three separate ongoing campaign stories going on right now.

But the group in the heroic campaign handle encounters a bit differently. They are more cautious, and have the tendency to skip things that they don't think are important to their overall goals. They don't go through a dungeon to clean it out, the look for the fastest route in and out to get what they are looking for.

I guess the major difference is the players themselves. In the heroic campaign they've picked up on an interesting hidden agenda by a group or groups and decided that's what they wanted to follow up on. The public campaigns tend not to follow up on the hidden agendas, although it's not uncommon for some of them to be interested in that, in which case they often get spun off into a home campaign.

So I don't know, I guess it's a sandbox with elaborate subplots waiting to be discovered and followed?

Ilbranteloth

 

[Emerikol]

Well it's clear I'm not good at describing these styles. I definitely am not denigrating either approach. I obviously like one style more but that is something I recognize as taste.

When I say the DM and players team at the metagame level, I mean that interaction occurs outside of what the character knows. For example, in Numenera, the GM can offer an XP in exchange for having something happen to the PC. The PC can accept the XP and if he does then the thing happens. It's usually bad in some way but it furthers the storyline. This is the GM and the Player having a transaction at a level outside the characters mind.

A popular example is falling through a trap door. No character is going to want to fall unexpectedly through a trap door. No sane character anyway. The player though is willing to accept an XP in exchange for letting that happen. XP though is not something real in game. Characters don't know about XP.

A PC will later use an XP to make some roll more favorable based on something about that PC. Kind of like an aspect in Fate. This again is bending the reality of the game. The player is intruding where his character has no knowledge. XP, Fate Points, etc.. are not real things IN GAME.

Now. I don't like that style of play for myself. There is an objective chance that something can happen based upon situation and established probabilities in my games. Players can't influence anything except through their character. If they get some bonus for background knowledge they get it everytime the given situation IN GAME exists.

I hope that clears up what I mean by Player control versus Character control. I like the latter. There is nothing wrong with the former. It is a really different sort of game. Obviously you can use Player Control to any degree you like from mildly to all the time.

Failing forward that was instituted in advance using well defined charts would likely become the physics of the world. I'd be okay with a critical failure role forcing a role on another table. Any of those sorts of things can be made to work without leaving behind the character focused style. You just have to handle it right. I do though think that Fail Forward was promoted originally as a story driven style element. Many at WOTC seem to lack any understanding about how different these styles are and why people prefer one or the other.

 

[Emerikol]

I guess the major difference is the players themselves. In the heroic campaign they've picked up on an interesting hidden agenda by a group or groups and decided that's what they wanted to follow up on. The public campaigns tend not to follow up on the hidden agendas, although it's not uncommon for some of them to be interested in that, in which case they often get spun off into a home campaign.

So I don't know, I guess it's a sandbox with elaborate subplots waiting to be discovered and followed?
Ilbranteloth

I don't agree. I think the level of heroism in the game is independent of playstyle. I definitely think that Dungeonworld though is a game devoted to the storyteller style more than it is to the hard gamist style.

I'm not saying that DM's don't create adventures and worlds in both cases. It really is about player vs character intrusion into the game. When I say character, of course I mean the player acting solely as the character.

And I'm not saying with some work that Dungeonworld couldn't be more to my style either. Just saying it doesn't go out of it's way to support my style. Just like Fate, which is definitely a player intrusion game, I could hack it and remove a lot of the player intrusion elements but it wouldn't really be the same game.

In fact, I haven't played Dungeon world. I only attended a seminar once. So perhaps I don't understand all of angles on that game. I don't really want to debate a particular game. I'm just saying that there is one very big axis that gamers fall along. That axis is how much they want to intrude on the game outside of their characters. Some like me will say "not at all!" whereas others will say "All the time!" It is a matter of taste.