Marrying twists and conditions with the disinterested GM [Archive]

Blymurkla

2017-08-15, 04:33 AM

I love the 'failing forward' principle, where a failure of an ability check in the mechanics isn't the same as a failure to make progress in the fiction. Rather, failure indicates that there's some cost like an injury (a condition) that comes with progression or that a new threat appears which needs to be dealt with (a twist). This principle keeps the story going, PCs are never stomped to a dead halt faltering for what to do. My favourite execution of failing forward is in Mouse Guard RPG, but Apocalypse World does a great job too.

Lately, I've been toying with the idea of running a sandbox exploration game, without almost any sort of pre-made story and GMing without any intention of making cool things happen for the PCs but rather being the impartial, disinterested interpretor of the game world. And I want to marry that with twists and conditions.

But twists and conditions is all about story. A failed check in Mouse Guard or Apocalypse World means it's time for the GM to step in, dish out some sort of trouble that's interesting and makes sense in the fiction, and thus move the story along. These games don't need rules about retrying, because you can't fail in the traditional sense.

However, traditional games often suck at adjudicating failure. The baseline is that a player states what her PC does, the GM calls for a check and if it fails, the PC doesn't succeed. Seems simple? Roll Strength, fail to break down door. But sometimes an attempt to do something can 'fail' in multiple ways, or at least come with complications. Those types of failures are often risks and events that seem 'realistic', but traditional games tend to forget about them (thus making the game feel less realistic) because there aren't good at complications.

Take crossing a rapidly flowing river (a fairly common challenge in the game I'm imaging). It would probably require some sort of Swim or Athletics skill, possibly paired with an attribute like Strength or Health depending on the game system. The player says »I'm gonna cross that river by swimming« and the GM calls for a check. The check fails, thus the PC fail with what he set out to do and remains on the same side of the river. Fair enough.

But that's not really all that can go wrong when crossing a river, is it? The PC could injury himself in the rapids, or contract a disease swallowing water. His equipment could be damaged or torn away by the rapids. He could be surprised by the strength of the torrent, being swept along way longer than anticipated, ending up in a bad place. His injured-seal-like splashing could make his presence known to a foul river monster. And pretty much all of these complications could conceivably happen whether the PC succeeds to cross the river or not.

In Mouse Guard, this would be easy. Firstly, that the mouse fails to cross the river is out of the question (or at least, a twist that have him remain on the same side of it would also make crossing it irrelevant. Say that the predator the mouse was tracking appears). Then, I'd choose among these options and take the one that fits best with the story. I've already made sure that all of these options work in the fiction, they're 'realistic' outcomes. Do I want to twist and add tension with a snake, or do I want to wrap the GM turn up by choosing to make the mouse Sick and then moving on? Good stuff, but story focused.

So, what should I do in that sandbox exploration game where I'm the disinterested GM? I want to have a chance of those complicated outcomes, because I believe they make the world real.

I don't really like to make it a bunch of rolls. Okay, a Swim check to cross, a saving throw to resist disease and a random encounter roll to check for river monsters could work. Though both the disease and monster risk is in some way tied to the elegance of the swim ... And what about risk of injury and equipment loss? And being swept away. Can't keep adding checks because it bogs the game down and, in my mind, crossing a river is a discrete, singular challenge that should require one die roll.

Having degrees of success and failure could help, i.e. a 'partial success' could allow you to cross but be injured. But assigning the different complications, including whether the PC actually makes it to the other side or not, seems hard to do in an impartial manner.

Darth Ultron

2017-08-15, 06:38 AM

So, what should I do in that sandbox exploration game where I'm the disinterested GM? I want to have a chance of those complicated outcomes, because I believe they make the world real.

Most games don't nit pick so much. Doing something like crossing a river is a normally a very, very small part of a game. In general, games are about much bigger things then just 'crossing a river'.

So if you want to blow up 'just crossing a river' or even 'just walking down a dirt path' into a major game event...well, that would be a very different game play.

I'm not sure how you would do it other then lots of rolls. You can 'make saves' for bad things not to happen or maybe do percentages. But you really only have two ways to do it: random rolls or DM decides.

Chimera245

2017-08-15, 07:21 AM

If it's about "always moving forward" even at a cost, then if you're impartial with no predetermined plans, then just consider whatever the PCs were trying to do, either in the immediate action, or in the greater sense of their goal.

To continue with the crossing the river example, it's obvious that "moving forward" means getting across the river. And it could be that their greater goal is to get to an abandoned temple somewhere on the other side where some rumored treasure is said to be buried.

If they "fail" to cross the river, then have them cross the river with a cost, or remain on the same side, but catch a glimpse of the temple while they're being washed away.

You don't need to make a whole bunch of rolls if they fail. Just make a list of all the bad things that can happen, then roll to see which one occurs. Preferably one that can be generalized for more than just river-crossing.

For example:

1. Caught the attention of a monster.
2. Hurt themselves.
3. Damaged/lost equipment.
4. got swept away/lost
5. got stuck/immobilized/captured
6. disease/poison
7. triggered a trap
8. roll twice

Then whenever a failure happens, roll a d8, and see what bad thing happens. Make some DM judgement based on the result, (for example, of disease/poison comes up for crossing the river, a disease makes more sense than poison, and for choice of disease, something like filth fever makes more sense than mummy rot.) Assemble your own list if you have a better idea of dramatic bad things, use a bigger die, and assign a range to the options if you want some to be more likely than others.

That way it's either:

PC check -> Success -> continue on with the adventure
or
PC check -> Failure -> consequence roll -> continue on with the adventure.

As far as how to interpret whatever result comes up, just go with whatever makes the most sense. Since that's what you would have been doing anyway with a predetermined story. The only difference is you might have to be a little broader-minded about what constitutes the goal than in a predetermined story. But whatever the PCs are trying to do, you will have been watching it and judging it the whole time, so it shouldn't be too hard.

Blymurkla

2017-08-15, 08:17 AM

Most games don't nit pick so much. Doing something like crossing a river is a normally a very, very small part of a game. In general, games are about much bigger things then just 'crossing a river'.

So if you want to blow up 'just crossing a river' or even 'just walking down a dirt path' into a major game event...well, that would be a very different game play. This for a sandbox exploration game. I think that implies that wilderness hazards plays at least some part. Granted, I want more than just 'some part'. I love the wilds and think it has been neglected in way too many games. Crossing rivers, climbing mountain slopes, finding your way, foraging or surviving though weather is huge in actual exploring, like finding Machu Picchu or reaching the south pole. RPGs tend to gloss over that and just move to the monsters, because that's what their rules systems handle (and what most players want). There are exceptions, crossing a river in Mouse Guard matters a lot.

Anyway, what I'm asking isn't really about wilderness exploration. It's »There are multiple ways this could go wrong. How do I determine which one(s) happen?« If you want something more heroic than crossing a river, read it as this: »The PC is trying to jump onto the dragon's back to stab it in the eye. How do I determine if she clings on to the tip of the tail facing a hard climb up to the head, lands close to the head but on a poisonous spike or if she just fails?«

If it's about "always moving forward" even at a cost, then if you're impartial with no predetermined plans, then just consider whatever the PCs were trying to do, either in the immediate action, or in the greater sense of their goal.

To continue with the crossing the river example, it's obvious that "moving forward" means getting across the river. And it could be that their greater goal is to get to an abandoned temple somewhere on the other side where some rumored treasure is said to be buried.

If they "fail" to cross the river, then have them cross the river with a cost, or remain on the same side, but catch a glimpse of the temple while they're being washed away. Considering the PCs greater goal seems like considering the story. In the game I'm imaging, the PCs might be trying to get to an abandoned temple. But the rumour of the temple's existence might be false, or they've misinterpreted an old map and are actually at the wrong river, or my GM map just say that the temple is way to far from the river to be seen. If I change that because a failed roll should still be "moving forward", then I'm making story decisions and not presenting an objective world, aren't I?

I think I screwed up the beginning of my initial post. These thoughts come from the 'failing forward' principle that I've learned to love in Mouse Guard and Apocalypse World, but it's not wholly what I want here. In Mouse Guard, just failing to cross the river (a non-event) is utterly boring, so the game doesn't make that happen. In what I'm imagining, failure is still an option. Remaining on the same side means having to find a different route, something that actually matters because supplies are depleted and time is a factor. It's just that I don't want remain-on-the-same-side-type of failure to be the only option, the way I'm used to seeing traditional games work.

You don't need to make a whole bunch of rolls if they fail. Just make a list of all the bad things that can happen, then roll to see which one occurs. Preferably one that can be generalized for more than just river-crossing.

For example:

1. Caught the attention of a monster.
2. Hurt themselves.
3. Damaged/lost equipment.
4. got swept away/lost
5. got stuck/immobilized/captured
6. disease/poison
7. triggered a trap
8. roll twice

Then whenever a failure happens, roll a d8, and see what bad thing happens. Make some DM judgement based on the result, (for example, of disease/poison comes up for crossing the river, a disease makes more sense than poison, and for choice of disease, something like filth fever makes more sense than mummy rot.) Assemble your own list if you have a better idea of dramatic bad things, use a bigger die, and assign a range to the options if you want some to be more likely than others.

That way it's either:

PC check -> Success -> continue on with the adventure
or
PC check -> Failure -> consequence roll -> continue on with the adventure. Making lists is a good idea. I think I'll have a hard time designing a list that's sufficiently broad that they work in a bunch of cases, yet still helpful enough that I'm not tempted to make story decisions. But it is probably doable.

Jay R

2017-08-15, 08:45 AM

The solution is to react freeform. Don't try to build tables for every contingency like Gygax. Picture the situation in your mind and decide what seems reasonable like Arneson.

I don't really like to make it a bunch of rolls. Okay, a Swim check to cross, a saving throw to resist disease and a random encounter roll to check for river monsters could work. Though both the disease and monster risk is in some way tied to the elegance of the swim ... And what about risk of injury and equipment loss? And being swept away. Can't keep adding checks because it bogs the game down and, in my mind, crossing a river is a discrete, singular challenge that should require one die roll.

I'm guessing that you've never swum in a rapidly flowing river. It is no more a "discrete, singular challenge that should require one die roll" than a melee is.

Player: I attempt to swim across the river. <rolls low>
GM: ten feet in, your head goes under and you swallow water. Do that as many times as your CON bonus and you'll go unconscious. Roll again, with a -2 circumstance penalty.
Player: <rolls a success>.
GM: You get your head above water, and you're twenty feet downriver. Is the rest of the party staying there, running down river, or swimming after him?
Player: <rolls a failure>
GM: You feel a sharp piece of wood hit your back. One of your water bottles is now pierced and worthless.
etc.

This encounter could include river monsters (malignant or benign), losing baggage (which might be recovered), loss of hit points, saving each other, discovery of magic items (Deagol fell into the water and found the Ring), hp loss (logs in the river), water damage, etc.. Twenty minutes later, the whole party, bedraggled, soaking wet, exhausted and damaged, collects itself on the opposite shore, with the loss of some equipment including one horse, but with a shiny new item and a great, exciting encounter to talk about.

Blymurkla

2017-08-15, 09:10 AM

I'm guessing that you've never swum in a rapidly flowing river. It is no more a "discrete, singular challenge that should require one die roll" than a melee is. I have swum rapidly flowing rivers. Not 'severe risk of drowning rivers', but 'you might get injured' rivers. And I'm fine with melee being solved with a single test of the Fight skill, by the way. In certain games at least.

IPlayer: I attempt to swim across the river. <rolls low>
GM: ten feet in, your head goes under and you swallow water. Do that as many times as your CON bonus and you'll go unconscious. Roll again, with a -2 circumstance penalty.
Player: <rolls a success>.
GM: You get your head above water, and you're twenty feet downriver. Is the rest of the party staying there, running down river, or swimming after him?
Player: <rolls a failure>
GM: You feel a sharp piece of wood hit your back. One of your water bottles is now pierced and worthless.
etc.

This encounter could include river monsters (malignant or benign), losing baggage (which might be recovered), loss of hit points, saving each other, discovery of magic items (Deagol fell into the water and found the Ring), hp loss (logs in the river), water damage, etc.. Twenty minutes later, the whole party, bedraggled, soaking wet, exhausted and damaged, collects itself on the opposite shore, with the loss of some equipment including one horse, but with a shiny new item and a great, exciting encounter to talk about. The trouble with that method is that it's time consuming. I'm not entirely against that, sometimes it's great. But if a 5-day journey involves crossing a handful of rivers (which is entirely possible in heavy-rainfall mountains) then I imagine it gets a bit tedious to play out every one like that. But perhaps not doing it is leaving the old-school, combat-as-war feel entirely?

Hm. This certainly touches upon a different question. Again, a thing I love about Mouse Guard is that the scope of a skill roll is fluid. Finding your way over a network of roots under a single, huge oak tree? That's Pathfinder Ob 2, because it's interesting right now. Journeying from Copperwood to Sprucetuck? That's a full conflict, going blow-by-blow with Pathfinder, Scout and Survivalist against the Spring's Nature 6, because that's what this mission is all about. Crossing the entire territories? That's Pathfinder Ob 6, because we're in the player turn and it feels suitable.

But those are story decisions, and thus have no place in what I want right now.

Hm. So perhaps I need some way for things to spiral out of control, leading to the scene you describe above Jay R. But sometimes, things wrap up quickly, and you're on your way, injured or not.

Florian

2017-08-15, 09:22 AM

@Blymurkla:

Both, Splittermond and The One Ring tackle the topic of wilderness exploration. Both use a similar approach, by not going scene by scene but creating the overall "journey" first, then having the players roll the relevant skills to create the "obstacles" and "opportunities" based on margins of success (pass, but..) or margin of failure (fail, but...). You assign tasks for each "journey", like the Ranger studying the maps, the Druid trying too predict the weather, the Bard listening for rumors, and so on, with the results generating the twists and conditions that will happen later during the actual journey.

Rogue Trader is also about sandbox exploration, but the players must also come up with overall goals of what they want to achieve and present the gm with an "Endeavor".
An "Endeaver" lists the overall goal and the steps how to get there. Each step uses one or more types to classify them and is assigned a difficulty level and XP.

To use your "Machu Pichu" example, we start with a map...:
Part 1 - The Auctions (1000 XP)
1) Get an invitation for the map auction (social, mercantile, easy 100 XP)
2) Win the auction bid (social, mercantile, easy 100XP)
3) Get the map and your supplies to a ship before your rivals get you! (social, combat, medium 300 XP)
.... and so on.

The XP gauge indicates how successful the scene must be handled, or there will be consequences based on how the endeavor part is tagged (i.e. "social").

Jay R

2017-08-15, 10:32 AM

The trouble with that method is that it's time consuming. I'm not entirely against that, sometimes it's great. But if a 5-day journey involves crossing a handful of rivers (which is entirely possible in heavy-rainfall mountains) then I imagine it gets a bit tedious to play out every one like that. But perhaps not doing it is leaving the old-school, combat-as-war feel entirely?
...
Hm. So perhaps I need some way for things to spiral out of control, leading to the scene you describe above Jay R. But sometimes, things wrap up quickly, and you're on your way, injured or not.

Agreed. Don't do this with all five rivers, for the same reason you don't have five identical encounters with orcs. This is the biggest, most threatening river - in your description, the 'severe risk of drowning river'. Each river would have a threat level, and a width.

I've stepped across the Rayado River in one step, waded across the Cimarron River, and driven a mile across the Mississippi River. Rivers differ.

In fact, the surprise, and thus the suspense, will be heightened if they had no problems with a couple, and then the big one is in their path.

Airk

2017-08-15, 11:10 AM

We're kinda losing the forest for the trees here with all these bizarrely shortsighted answers focusing specifically on the idea of crossing a river. Stop it, people, you're smarter than this. (If you're not, stop trying to answer. :P)

There seems to be a bit of a fundamental misconception about what "failing forward" is here. It doesn't mean "failure moves the 'story' forward" because a lot of games that utilize absolutely argue against the idea of there being a "story" that is being moved forward. What it means is that failure moves the action/situation forward. You don't roll and fail and leave the situation unchanged - so no "You try to climb the cliff and fail, would you like to try again?" Not even if it's "You try to climb the cliff and fall, so you take 3d6 damage, would you like to try again?" The situation needs to change as a result of the failure. "You try to climb the cliff, but in the process you dislodge some stones that were more loose than they seemed. You fall back down, and now the path up the cliff looks even worse than before." or "You try to climb the cliff, but it's slow going - you've only made it about halfway up when you notice a shadow passing over you. There's a griffin circling overhead." or "You successfully climb the cliff, but as you finally reach the top, you discover that your rope has been badly frayed on the rocks and won't be safe to use again." And yes, all three of those are straight up FAILURES - no need for partial success rules here unless you want them as a guide to what level of complication to inflict, and that's nothing you can't do with a simple "How much did you fail by?" question. None of this has anything to do with "Advancing the story" because there isn't one - it's just trying to avoid "Well, okay we try until we do it." which is boring. A large part of this is to AVOID making extra rolls. Yes, adding the griffin as a complication adds more rolls, but now you're dealing with a griffin.

I don't really understand, at this point, why making a Pathfinder test in Mouse Guard is a "story decision" that you want to avoid.

So yeah. There seem to be a lot of unnecessary hangups here - there's really nothing that prevents you from applying fail forward techniques to a sandbox game in a traditional system if that is a thing you and your players want.

Blymurkla

2017-08-15, 12:22 PM

I don't really understand, at this point, why making a Pathfinder test in Mouse Guard is a "story decision" that you want to avoid. Because I'm not calling for one every time. In a sandbox with a traditional game, I'd expect going from Coppertuck to Sprucetuck to be a Pathfinder Ob 5 every time I make the journey. Sure, that difficulty level could differ with season and weather, but the assumption is that each time you try and something which entails a risk, I'd roll a check.

That's not what you do in Mouse Guard. The first time the journey is made, the GM might make an adventure of it by describing obstacles and calling for different checks - none of which need to be in the Pathfinder skill. When the patrol returns in the following player turn, that journey becomes a single Pathfinder Ob 5 check. Third time and the GM just narrates the journey briefly, calling for no checks. And the forth time the GM decides to use the conflict rules.

The rules aren't used to make a coherent representation of the game world. The rules are there for the GM to build interesting stories.

So yeah. There seem to be a lot of unnecessary hangups here - there's really nothing that prevents you from applying fail forward techniques to a sandbox game in a traditional system if that is a thing you and your players want. But there's a bit of a problem with using it in an impartial manner, if the fail forward could be more than one thing.

Airk

2017-08-15, 02:44 PM

Because I'm not calling for one every time. In a sandbox with a traditional game, I'd expect going from Coppertuck to Sprucetuck to be a Pathfinder Ob 5 every time I make the journey. Sure, that difficulty level could differ with season and weather, but the assumption is that each time you try and something which entails a risk, I'd roll a check.

That's not what you do in Mouse Guard. The first time the journey is made, the GM might make an adventure of it by describing obstacles and calling for different checks - none of which need to be in the Pathfinder skill. When the patrol returns in the following player turn, that journey becomes a single Pathfinder Ob 5 check. Third time and the GM just narrates the journey briefly, calling for no checks. And the forth time the GM decides to use the conflict rules.

The rules aren't used to make a coherent representation of the game world. The rules are there for the GM to build interesting stories.

I don't really see what this has to do with "failing forward" though? It's a completely unrelated topic in my mind. And in all honesty, applying rules consistently isn't a "story decision". Mouse Guard simply has different expectations of how the rules should be applied.

But there's a bit of a problem with using it in an impartial manner, if the fail forward could be more than one thing.

I would argue that there are already tons of problems in any game with someone trying to be completely impartial, so that's not really a big deal, but if it really bothers you for some reason, you can use random tables to disclaim responsibility (note: I'm not, overall, a big fan of this approach, because I regard as the last recourse of people who don't want to take responsibility for their own decisions, but that's more of a personal pet peeve than a problem with the solution.). Said table would need to be pretty generic, because no one is interested in creating a new table for every "encounter" but you can take a cue from the various "GM Moves" in PbtA games and create, oh, a 2d6 table like:

2 - Catastrophic mishap; Something goes very wrong and impacts the entire party. Roll again to determine the nature of the mishap, ignoring 2 and 12.
3 - Roll twice.
4 - Separation - An event occurs that splits the party
5 - Mystery - Something about this obstacle is not as it appears. The party must figure out what and deal with it.
6 - Damage: The PC takes environmental damage of some sort
7 - Cost: The party must expend some item of gear of other resources to overcome the obstacle, or seek a new path.
8 - Resources lost: The party loses some item of gear or valuable that is used or vulnerable during this action.
9 - Threat: Something dangerous appears in the distance.
10 - Encounter: A third party appear to complicate the situation
11 - Ambush: A third party takes the group by surprise
12 - Happy mischance; Roll again, and apply the inverse situation (Equipment found, friendly encounter, etc).

Roll on that and temper it with the situation. You're never going to be able to avoid that second part, whether you use fail forward or not, so you'd best get past that hangup. ;)

Blymurkla

2017-08-16, 03:39 AM

Both, Splittermond and The One Ring tackle the topic of wilderness exploration. Both use a similar approach, by not going scene by scene but creating the overall "journey" first, then having the players roll the relevant skills to create the "obstacles" and "opportunities" based on margins of success (pass, but..) or margin of failure (fail, but...). You assign tasks for each "journey", like the Ranger studying the maps, the Druid trying too predict the weather, the Bard listening for rumors, and so on, with the results generating the twists and conditions that will happen later during the actual journey. I know of The One Ring; I have read it and played it a bit. I'm not a huge Tolkien fan, but I was rather excited about the game when I first heard it would cover wilderness. I was thoroughly disappointed, there might be some good bits in The One Ring (like the party assuming roles) but it's obscured by the amount of crap that game has.

I'm not at all familiar with Splittermond, I'll try to check it out.

Rogue Trader is also about sandbox exploration, but the players must also come up with overall goals of what they want to achieve and present the gm with an "Endeavor".
An "Endeaver" lists the overall goal and the steps how to get there. Each step uses one or more types to classify them and is assigned a difficulty level and XP.

To use your "Machu Pichu" example, we start with a map...:
Part 1 - The Auctions (1000 XP)
1) Get an invitation for the map auction (social, mercantile, easy 100 XP)
2) Win the auction bid (social, mercantile, easy 100XP)
3) Get the map and your supplies to a ship before your rivals get you! (social, combat, medium 300 XP)
.... and so on.

The XP gauge indicates how successful the scene must be handled, or there will be consequences based on how the endeavor part is tagged (i.e. "social"). Hm. Is an endeavour like a skill challenge (in say D&D 4e or The One Ring) or how does it play out? Are all steps decided upon at the beginning? Do only the GM know the exact steps (including their difficulty and xp) or is it all done in the open?

Max_Killjoy

2017-08-16, 09:24 AM

It can be be as simple as deciding that the characters make it across the river, but a failed check will result in damage proportional to the degree of failure.

Koo Rehtorb

2017-08-16, 09:46 AM

I've struggled with this myself in Torchbearer. I want rolls to not be a binary pass/fail state because it's boring (as well as the game demanding it). But I also want whether the PCs succeed or fail in general to be up to them, and the dice, not me.

The short answer is, you can't do it. You can't list out everything that could possibly happen on any given roll, especially since you can't foresee every roll that's going to be made in the session, and without doing that this sort of gameplay is always going to come down to GM fiat. However, you can make a reasonable approximation if you put a bunch of effort into understanding the setting. If you prepare the location the PCs are likely to be in very well then you can have a pretty clear idea of what failure result makes the most sense for the current situation. That's the best you're going to get.

Tinkerer

2017-08-16, 10:54 AM

I've run three different mechanics for this in my time and have had both of them work out fairly well.

In the first case it's simply using the players roll to determine the degree of mishap. The problem inherent in this can come from the system that you are using. In a D&D example if a player only needs a 3 or higher to succeed that gives you very few possible outcomes. It also disincentives taking the long shot as that's a whole lot of negative spectrum if you fail. I normally use this method for flavour consequences, things which can be entertaining but which aren't really negative... except for the number of nicknames they spawn.

Second I've used a side effect table as Airk mentioned (I use a slightly different one but theirs is nice). I usually use this as my crit failure/success table but I have tried it for normal failures. It really depends on how common success is. I've found it's only a good choice where the players have a really low chance of failure (Savage Worlds spring to mind). Otherwise rather than your players looking for a chance to use their skills they wind up looking for a chance to avoid using them and that's just not fun. "Oh we're at a cliff! I have a good climbing skill but... since really bad stuff will happen if we fail lets spend a day looking for a path." I do like that it can result in some results which you wouldn't normally think of as a side effect of a failed skill roll, such as an ambush.

And third there is the common sense as Koo Rehtorb pointed out. This tends to override the other two methods. One thing that I found rather odd in your initial message was:

Take crossing a rapidly flowing river (a fairly common challenge in the game I'm imaging). It would probably require some sort of Swim or Athletics skill, possibly paired with an attribute like Strength or Health depending on the game system. The player says »I'm gonna cross that river by swimming« and the GM calls for a check. The check fails, thus the PC fail with what he set out to do and remains on the same side of the river. Fair enough.
I have never played a game with that rule on failure to swim. All of them have been if you fail swimming, you start drowning! You fail climbing, you start falling! Quite a few skills have a built in punishment for failure and these are generally enough.

Jay R

2017-08-16, 01:22 PM

We're kinda losing the forest for the trees here with all these bizarrely shortsighted answers focusing specifically on the idea of crossing a river. Stop it, people, you're smarter than this. (If you're not, stop trying to answer. :P)

In fact, the river was merely an example. My main point, which I opened with, was this:

"The solution is to react freeform. Don't try to build tables for every contingency like Gygax. Picture the situation in your mind and decide what seems reasonable like Arneson."

Since you don't like using the river as an example, here's another one:
The game was Flashing Blades, a musketeer game. The rogue in the party had decided to learn the Etiquette skill, which takes three months. He'd spent two weeks on it. To make a successful role, you have to roll your Charm or less on a d20.

The party went to a high-status hunting party, and at one point, the rogue decided that he was going to go talk to the duke's daughter, who is surrounded by noble suitors. They tried to tell him that he cannot go introduce himself to her; he needs a proper introduction. But he decided that since he was learning Etiquette, he could do it anyway.

So he barged through a collection of high-level nobles and introduced himself to her, and said, "I want to make an Etiquette roll to impress her."

So, he is attempting to use a cross-class skill he has not in fact learned, in front of several masters of the skill, having already misbehaved, in a high-stress environment, and had to roll an 8 or less (if he had the skill at all).

He rolled a 20. Critical fumble.

I said, "You compliment her beauty, look soulfully into her eyes, take her hand gently, bend over it, raise it to your lips, ... and f*rt."

jayem

2017-08-16, 06:13 PM

If somethings is plot wise as significant in fighting then it seems right that it's about as complex and a similar number of dice roles. Lord of the rings, the going down a cliff scene pretty much is the same as some battles.

A little stream I crossed the other day had to climb down head height, walk over some stepping stones and then up a similar distance (I'm kind of surprised it was a footpath). A roll for each step doesn't seem too bad (if they can be kept simple).

General Outcomes:
Safe paused progress failure (I don't think this is a bad thing, if it keeps up suspense).
Safe blocked progress failure.
[Safe stuck half way through progress failure (where you have to decide do you go back or on). If you have multiple stages then this happens naturally]
Safe passage
And then more or less the same with injury/kit damage

But it probably does require more improvisation, as everything is different.
I don't know what the dice pattern should be (perhaps double dice, one for damage one for 'success'). There might be a way to adapt into a generic challenge. [At that point you then just need a list of steps and skills, which is a bit smaller, which cuts the improvisation a bit]

kyoryu

2017-08-16, 08:25 PM

Just to chime in - "fail forward" doesn't have to mean that you succeed at a cost. It can very well mean that you fail utterly, but in such a way that your situation has changed.

So fail to swim across the river? Cool! The river drags you down to the next beach, where you struggle ashore only to see some very unhappy locals pointing spears at you.

The story has moved forward - even if it's not in the direction you anticipated.

jayem

2017-08-17, 02:18 AM

Challenges are drafted into complex, intermediate and simple.

In a complex challenge, the challenge is split into a number of steps along an ideal path.
Each step has a "Ease Of Success", "Risk Of Minor Harm" "Risk Of Severe Harm" and "Risk Of Mild Failure" "Risk Of Severe Failure" and skills
Role three D20's. Red, Blue, Green.

If the Green+Skill level is greater than the EOS then the step is completed and the player is on the next step
If the Green+Skill level is less than the EOS then the step is not completed and the player remains on the previous step
(note at the GM/PC's discretion in the event of harm or failure on the other dice, this can be reversed)

If the Red+Skill level is greater than the ROMH then the activity is carried out safely
If the Red+Skill level is between the ROMH and ROMH then appropriate 'minor damage' is done (in vague proportion, perhaps effectively costing a skill level for immediate future tasks)
If the Red+Skill level is less than ROMH then appropriate 'major damage' is done (at the extremes this is damage that cripples the player, and needs serious action)

If the Blue+Skill level is greater than the ROMF then the activity is carried out in a repeatable/reversible way
If the Blue+Skill level is between the ROMF and ROSF then something happens that makes it harder to repeat/reverse the step
If the Blue+Skill level is below ROSF then something happens that makes it impossible to repeat/reverse the step

For an intermediate challenge, the difficulties are combined. The blue dice specifies the position in the task where failure occured, the red dice the intensity of effect.
For a simple challenge...

Martin Greywolf

2017-08-17, 02:44 AM

Disinterested vs impartial DM

Okay, let's get our terminology straight. Disinterested DM is universally a bad thing, because it means that DM is not having fun and doesn't care about the game. There's no argument to make this a desirable thing.

Impartial DM is a lot different, though I'd argue that truly impartial DM is also not good. As a DM, your primary goal should be fun for both player and yourself, and everything else should go after that. That means that, should it come to a decision your world needs to make, you should err on the side of excitement, provided it doesn't violate verisimilitude too much. A better way to say that is that you should weight branches on your decision trees proportionately to fun.

What you seem to be going for is a world where the PCs are not special by themselves, only by their actions, and where the plot mostly doesn't exist and certainly doesn't make any accommodations to the players. That doesn't mean that you should stop making plot-like decisions, it's just that the plots will be about someone else as often as about PCs, and started by the PC's actions in the latter case.

Failing forward and arbitrary nature thereof

What probably trips you up here is that you want to have full transparency on what can happen a la tables of results, and failing forward doesn't have that. Well, that i sby design, or rather, it can't be any other way - failing forward relies on changing the situation the PCs are involved in depending on circumstances, and that just can't fit in any reasonable table. You therefore have to make arbitrary decision what the failed check will result in - will that failed river crossing sweep the PCs towards a waterfall, will it just loose them supplies etc etc.

This is one of the reasons why a lot of games that use this have either a meta currency that allows the players to make their own changes (e.g. FATE compels), or put a clause in the rules about DM talking the results out with the players.

Die roll scope

There is no reason why you couldn't apply variable check scopes and variable round lengths to d20. The journey can be like a combat round, Survival is your attack, supplies your HP. You;ll have to play with this system a bit, but that's what happens when a system doesn't have mechanics that can handle this universally.

What scope you pick should, again, depend on the situation. If the players want to go explore a never before seen land, make it several checks, if they travel a well-travelled road, make it just one. Your idea with having the same checks for the same routes will get really tiresome if the PCs travel around a lot - making the same set of skill checks with no changes for the fifth time on a row is not fun.

The golden rule of die checks

You should only roll for something if the failure is interesting and relevant. For example, if someone is trying to pick a lock on a box in their own home, don't roll. They will succeed eventually - d20 tries to simulate it with taking 20 and succeeds to a degree, just explains it poorly and perhaps makes it more complicated than it should be.

If the contents of the box can be destroyed by fiddling with the lock, THEN you roll, because now there's a relevant consequence for failure, and the roll has some tension to it.

This applies to any die roll you do. If the PCs are well stocked with supplies and are travelling across a good road, or one that they know well, then don't roll unless you're tracking their finances, and even then, make it just one check, because failure is pretty uninteresting - whoop de doo, we lost 10 gold worth of supplies that we can replace in the next town.

spinningdice

2017-08-17, 06:24 AM

I don't see why it's an issue. Games with this involved, tend to just use it and not codify it. You just set the pass and fail effects to what you want, or use staggered difficulty. There's no rule that says a climb check has to result in climbing or not climbing (running with D&D 5e at least, 3.5/PF tend to quantify stuff a bit more).

If you want something divorced from success or failure (such as star wars' funny dice or ORE's length and breadth success system) then you're in a bit more difficulty with a D&D type setting. You'd have to institute a rule such as if you pass by the exact number you need then you pass but with a complication, or if you roll a number devisable by 5 you get a complication regardless of pass or fail, but even then, you may be better with a graded system of pass/fail.

daniel_ream

2017-08-18, 04:58 AM

The party went to a high-status hunting party, and at one point, the rogue decided that he was going to go talk to the duke's daughter, who is surrounded by noble suitors. They tried to tell him that he cannot go introduce himself to her; he needs a proper introduction. But he decided that since he was learning Etiquette, he could do it anyway.

The movie My Fair Lady has pretty much the iconic example of how this goes down (the scene at the Ascot races (https://www.youtube.com/watch?v=8uozGujfdS0)). Eliza gets the form right - the diction and poses, but fails miserably at understanding that some topics are inappropriate in that context.

A note on binary pass/fail systems - RuneQuest 6/"Mythras" subtly changes things up a bit by defining a successful roll as a success, a critical failure as a failure, and a failed roll as simply "not-success". For instance, a failed climbing roll means you just don't make any progress up the wall because it's too hard. You only fall on a critical failure. For the Swimming example, on a failed Swimming roll you might make it partway across and then have to hang on to a branch or portruding rock because the current's too strong or you're too exhausted to continue. Only on a fumble would you drown or be swept away.

kyoryu

2017-08-18, 10:40 AM

For instance, a failed climbing roll means you just don't make any progress up the wall because it's too hard. You only fall on a critical failure.

I'm generally not a fan of this. Because any player will then just say "okay, I try again." You're just rolling more to get to the end.

The "fail forward" model mostly just presumes this, and so assumes that the pass/fail modes are what happens after all the retrying. Also, in most systems that promote the use of "fail forward", you're asked to think about what's actually happening, and decide on failure based on that. Why are you climbing the cliff, anyway? Is it to get away from someone? If so, failure might just mean they catch you. Or you could make it up too late for another reason. Or you could disturb some of the critters living on/near the cliff (if it's been established that there are such critters). Or, you could just fall.

"Nothing happens" works in the context of original D&D games because of the presumption of being in a dungeon. So the cost of failure is a chance of a random monster. This provides a natural "backpressure" against just trying things over and over.