Stakes and consequences in action resolution

[Ovinomancer]

Thanks - this is a useful contribution and deals with something that I've been curious about, but on which it's hard to find clear commentary, namely, the effect of the 5e DC rules.

The only thing that really caught me by surprise was the comment about the advantage/disadvantage mechanic. I would have thought this has something of a flattening effect.

I'm not sure what you mean by a flattening effect, but the way ad/disad works is that if you already have a decent chance of success, it makes that chance much better (advantage) or if you already had a small chance, it makes that chance even less likely (disadvantage). The converse might tend to "flatten" as a good chance of success is moved to a moderate chance with disadvantage and vice versa, but the prior cases are the more common (especially as it's often a player initiated resource or a team initiated one). In other words, it usually makes the likely outcome even more likely, which cuts against a resolution system that really needs reasonable chances of success or failure to do it's thing (the snowball).

I think this also touches on another corollary to your posts, that task -> conflict resolution, when well coupled, doesn't imply that a successful task resolution means a successful conflict resolution. Instead, a successful task should always be coupled to an advancement towards a successful conflict resolution, and vice versa, but not exactly one for one. The Baker quote and your example don't explicitly make this point, although it is lurking in the shadows, so to speak, especially in your play example. I wanted to bring it out more clearly.

 

[pemerton]

In addition, this approach does come with a cost - it turns a scene where the player tries to do something awesome, and ends up looking dumb. I spend a lot of resources to make sure I crack that safe, for nothing. But then, I turn around, and the paper I wanted was right there all the time! The resulting fiction is not about a great thief and safecracker - it is a fiction about a person who spends effort but only gets what they want by dumb luck.

I think you need to re-read the quote from Vincent Baker: he's putting forward that example as an example of why he doesn't like task resolution. And I'm saying that I, personally and speaking only for myself, agree with thim.

I should add - the example he gives, where the PC fails at the task to gain information but the GM feed in success anyway - is one that I have seen in more than one published adventure module. It's very common for those modules to have "backup" options for if the players fail to take, or to succeed, at the steps needed to get the requisite information.

Like Vincent Baker, I find this to be bad RPGing because it undermines the sense of stakes and consequences. And to borrow [MENTION=6696971]Manbearcat[/MENTION]'s terminology, it introduces "false action" - in this case, false failure.

There's any number of times when a thing is important because it is a critical resource of the BBEG, that ultimately the PCs want to destroy. Maybe spilling it on the floor, ruining it, is exactly what the PCs want. At this stage, they don't know. Heck, the GM didn't know. How on Earth can you claim to be honestly informing the players of the stakes if you don't know them yourself?

Well, for the reasons that I gave in my post, it was clear in the context of play that spilling the fluid was a bad thing. For the GM to turn that around would be tantamount to cheating, or at least very poor play.

I'm not a Dungeon World GM and have only a bit of play experience with that system, but I can describe what resulted from the fluid spill using some DW terminology:

* I was thinking offscreen, that is, keeping in mind the spill of the fluid for future adverse revelation;

* I took something from the characters, that is, used the spill of the fluid on the second PC's clothes as the basis, in the fiction, for his clothes being taken away (while he was treated for a broken leg) which meant that he lost the document he'd taken and hidden in his coat pocket;

* When the offscreen came back onscreen, I revealed an unwelcome truth, as the butler PC eventually found his master, having been moved to a clinic from a basement room in the house where the spill happened, with a burn on his face and head which (i) obviously to all had been caused by the spilled fluid, and (ii) confirmed (when put together with other bits of context) that he was a were-hyena.​

These all honour the player's (and PC's) failure and conform to the player's understanding of what was at stake in the situation, and what the consequences would be of mishandling the fluid.

The proper answer is that, when the player decides to help move the goo, the GM says (for example, in FATE language)), "If you fail, you are apt to spill some of the liquid on the floor, or on yourself. This will not kill you directly, but may result in a mild consequence for your character or a similarly sized aspect on a scene." Since supposedly everyone knows it is important, there's no reason to avoid explicitly saying so.

Here's what I posted about this on the thread that this one is descended from:

I think it's important for consequences to be clear - and I don't think of this through the lens of character knowledge but player knowledge - because the player needs to have a sense of what sort of resources to throw at the check (which depending on system and circumstances could be anything from fate/inspiration points, to equipment, to spells and potions, to . . .). Luke Crane says the following about making consequences clear (BW Gold, p 32):

When a player sets out a task for his character and states his intent, it is the GM’s job to inform him of the consequences of failure before the dice are rolled.

"If you fail this…" should often be heard at the table. Let the players know the consequences of their actions. Failure is not the end of the line, but it is complication that pushes the story in another direction.

Once that is said, everyone knows what's at stake and play can continue smoothly no matter what the result of the roll is.​

However, in his subsequently-published book of GMing advice (The Adventure Burner) Crane says that, in his own game, rather than stating the consequences expressly he often relies on context - of the fiction, of the mood at the table, etc - to make them implicit. When I'm GMing, I alternate between express and implicit consequences depending on inclination and whim. But again, for me this has a different motivation from that which 5ekyu states. I'm not worried about player vs character knowledge, and so even if consequence is implicit it will be implicit to the player as well as the GM - there won't be "hidden" bits of the fiction that suddenly emerge into the action on the basis of a failure. It's about pacing and narrative continuity and not weakening emotional intensity with needless explanation.

In the example of play that I provided in the OP, the immediate stakes were clear - getting the fluid out of the house threatened by fire - and the consequences implicit.

 

[Tony Vargas]

The only thing that really caught me by surprise was the comment about the advantage/disadvantage mechanic. I would have thought this has something of a flattening effect.

I'm not sure what you mean: Adv/Dis has more of an impact the closer the original check was to 50/50.

something that I've been curious about, but on which it's hard to find clear commentary, namely, the effect of the 5e DC rules.

You mean 'Bounded Accuracy?' Or the nominal easy/hard/etc guidelines?

Based on some other threads, one effect may be to encourage the DM to rule automatic success/failure fairly often - or at least, give players an incentive to find action-declarations that yield a narrated success in preference to a check?

 

[Umbran]

I think you need to re-read the quote from Vincent Baker: he's putting forward that example as an example of why he doesn't like task resolution. And I'm saying that I, personally and speaking only for myself, agree with thim.

Okay, I see the point. And, I know what is missing (if not from the original, from your presentation): What do you do instead? That segment is titled "Task resolution vs conflict resolution," but what you gave only goes into how task resolution is bad. It doesn't give the contrast - there is no "versus". Give us the same scene of the person going into the room with the safe, done as conflict resolution, and it might be useful.

Well, for the reasons that I gave in my post, it was clear in the context of play that spilling the fluid was a bad thing. For the GM to turn that around would be tantamount to cheating, or at least very poor play.

Is "bad" a jargon word with the game? If not, "that is a bad thing" is not sufficient to understand stakes. Burning my supper is a bad thing. Coming down with leprosy is a bad thing. "Bad" does not elucidate concrete stakes. In normal use, the word is broad and vague. We are not in the context of play, so reference to it means nothing to us.

Or, is it a game in which *all* conflicts share the same possible consequences?

I'm not a Dungeon World GM and have only a bit of play experience with that system, but I can describe what resulted from the fluid spill using some DW terminology

I don't know the game, so I cannot speak to it. In this game, is there a ranking, where "bad" = I Take something from you and Reveal an unwelcome truth? Is there some clear statement of how "bad" translates to those consequences?

This is the really big thing - when you are trying to describe things to people who don't understand it, saying, "Well, my players understood it" does squat-all to get people here to understand what happened.

In the example of play that I provided in the OP, the immediate stakes were clear - getting the fluid out of the house threatened by fire - and the consequences implicit.

Why did you think an example with a large portion of the method "implicit" was a good one for trying to get others to understand? The thing that was implicit was, in fact, *central to understanding what you were doing*.

 

[Chaosmancer]

So the player who chooses not have his/her PC study the situation and instead simply to act is in many ways hostage to the GM's prior decision-making about the nature of the situation. S/he isn't making any sort of informed or deliberate contribution to the overall state of the fiction.

There is a lot here, so forgive me if I'm missing some parts. But I don't see how this making your point that it is better for the DM or GM to tell you the consequences of your actions.

Looking at this for example, if the player chooses to rush forward and smash the liquid containers, in the example you gave, bad things happen because the liquid was a retardent for the reaction.

If the DM stops the player, and tells them the consequences, there are two options for that.

1) Vague: You tell them that breaking the glass might have negative consequences because of the arcane nature of the machinery and their limited understanding of it.

This likely does not tell the player anything they did not already know. Smashing a magic machine is a tactic generally employed by characters who are limited in their ability to solve the problem in other ways. It is taken as a risk, and they know it is a risk, so they will have gained no new information from your vague consequence. They will likely take the same action with the same result.

2) Specific: You tell them the liquid is acting as a magical retardent and that smashing it will send the machine into a meltdown.

Well, now they know something they didn't know before. It has prevented them from being at the mercy of the DMs notes or whims, by instead opting to simply tell them how the machine works and gives them a fairly obvious route to shutting it down (getting more of that fluid, to slow the reaction to standstill). Since you didn't tell them this information before they took an action, it wasn't something you thought they should know immediately. However, in telling them this you have given them the information they might have obtained by analyzing the machine for the cost of no action except the declared intent to smash it, which would have had negative consequences.

So, instead of smashing it, the DM stops their character and says "No, you can see smashing it is a bad plan, and your analysis tells you why it is a bad plan". But the player didn't say their character analyzed the machine, they said their character smashed the machine.

Vincent Baker (probably best known for designing DitV and Apocalypse World, which is the progenitor of PbtA RPGs), has talked about this also (and has influenced my thinking about it)

Speaking for myself, I prefer a game which does not generate this particular sort of pressure to play an "analytic" character, or to study situations in order to ascertain what is at stake in them, and hence what potential consequences might flow from succeeding or failing at the task.

Honestly, this bit from Mr. Baker is very interesting, but I fail to see how it applies to telling someone the consequences of their actions.

Task resolution -> Do you tell the player consequences for failure to break the safe

Conflict resolution -> Do you tell the player the consequences for failure to break the safe


Now, we can talk about these two types of resolutions, but neither one requires me to tell the players more information than the other. In fact, I do quite like utilizing Conflict resolution at times. I might know that the players are looking for dirt, and that it would make sense they could find some. Maybe I know that the evidence is in the desk, but they ask about a safe. A safe is also a fine place, and if they can successfully break in they might find something useful, it is about whether their intent makes sense within the goal.

But, also, task resolution has its place. IF I have set up a puzzle with multiple types of clues, and they choose to target a red herring, they are likely going to get that red herring. The enemy is trying to throw them off the trail, and sometimes they are going to stumble into those, it adds a small scent of realism to the game if they can look and find something that isn't useful, winning leading to a failure. But, this is a very specific type of campaign and style that I would be using.

I'll give a fairly banal example from the session of Cthulhu Dark that I GMed on the weekend:

It is a fun little example, but at what point did you tell the butler character the consequences for a failed roll before they made the attempt?

That is what my posts were about, and that seems to be something you are only glancing over and makes no appearance in this example.

Yes, the player's decisions had consequences that wouldn't have existed without the character, and the player could make many assumptions based off the game, the genre, and their own observations, but when the Butler was loading the canister, before they made their roll, did you tell them that if they failed they would spill the mysterious fluid all over their character? If you had not told them that, would it have changed any of the rest of your example.

And, I think many readers will say, "Yes, then in actuality, the player did *NOT* know the real stakes." When we set stakes, we know exactly what it is we stand to lose. If I don't want to lose the stakes, I don't bet. While the player knows the fluid is significant, they don't know in what manner, or what spilling it really means - they do *not* know what there is to lose here. Will the stuff make them grow tentacles? Will the scent of it lead amorous fish men to them? Nobody knows! That's hardly a solid example of knowing what the final cost of failure will be, and does not tell them how much of their resources they should spend on success (which is a very real part of knowing stakes ahead of time).

While you are demonstrating the techniques, I am not sure you're demonstrating that this actually gives the player any more control.

I agree with this. If we are talking about the player knowing so they can make an informed decision, then they must know not have assumptions based of their knowledge. They might know that the liquid is dangerous, but they do not know what the consequences for failure were, they can only assume it was bad.

Knowing the liquid is important is plenty sufficient to knowing you don't want to spill it. In the style of play [MENTION=42582]pemerton[/MENTION] is advocating, the effects of the liquid really depend on more play establishing those effects. Having it spilled on you just opens up future play possibilities involving that aspect. Recall that the established state of the liquid is that it us important to the master and that it may be sinister. Whether it makes you sprout tentacles later will be up to later play establishing this -- it's not currently determined. As such, there's no final consequence to be revealed, so pemerton's play has exactly as much horse as it needs to pull its cart. Calls for more are misassuming the needs in play.

This style is very different from traditional play as evinced by D&D (4e notwithstanding). It's easy to make incorrect assumptions based on prior experience where the GM has preplanned things like the nature and effect of the liquid. This isn't yet established in play, though, so the exact nature of the liquid will be established in later play according to the mechanics of play.

Then the player does not know the real stakes.

The entire point I've been told is that we have to tell the player the consequences of their actions, or otherwise they cannot make an informed decision about the fate of their character.

"This is important, failing is bad" is not what I have taken that to mean. Of course failing is bad. Of course the liquid is important. But there is a big difference between it being important because it was a clue to the monster involved in the plot, and it being important and melting the player to bone killing them instantly.

The player is not making an informed decision unless they know what the consequences are(according to the position I have been told pemerton and I assume you are taking), but you are saying that even the DM doesn't know the consequences... so nobody is informed enough by the standards my quotes were responding to in the original thread.





Or reading the next few posts, is knowing the consequeces for an action are simply bad, and that failure will have an effect on the game, all you were going for? Because if that is the case, then I see no reason to tell the players this before every single roll, which was what I was arguing against in the original thread. Telling the player the consequences for their actions is meaningless, the roll can continue either way, because they are not being told anything that would change their actions.

And if they are being told something that is changing their actions, why?

 

[Ratskinner]

I've brought it up before, but there's a game called Capes that focuses solely on Conflict Resolution, and it taught me more about it than all the discussions ever could.

Firstly, just to get it out of the way, I hate Vincent's oft-quoted example. Because, asking "Why?" makes no sense outside of the context of the Conflicts and how they determined. Don't ask "Why?", instead ask "Which Conflict are you trying to resolve?"

So, were I to try and implement a Universal Conflict Resolution system in D&D/5e, I would put the following on my "to do" list:
0) A "scene-framing" moment of play, for the GM to introduce a new scene and the Basic Conflicts to be resolved in the scene.
1) A countdown or tracking mechanism for Conflicts (ICRPG does something like this for a lightish D&D system), that determines when a Conflict is resolved.
a) This includes the "Not Yet" rule: if a Conflict has not mechanically been resolved (won?), you cannot narrate an event that would effectively resolve it.
2) A mechanism for players to introduce new Conflicts.
3) Rules/advice for what makes for a good Conflict Definition.

Big Problems:
1) This doesn't play super nice with the quasi-simulation skill system.
2) D&D combat is already (in some ways) a very complicated Conflict Resolution system (its only resolving the one event over and over again...but...) and making the two systems mesh might be problematic. For example, in a Capes fight scene, it would be totally legit to put out an Event(type of Conflict) named "Somebody puts out an eye." The players would then compete through play for control of that event, and eventually someone would have the privilege of narrating just who and how somebody lost their eye. I'm not sure how you work something like that alongside the D&D combat system. If you drop the combat system....are you still playing D&D?

 

[pemerton]

I'm not sure what you mean: Adv/Dis has more of an impact the closer the original check was to 50/50.

I'm not sure what you mean by a flattening effect, but the way ad/disad works is that if you already have a decent chance of success, it makes that chance much better (advantage) or if you already had a small chance, it makes that chance even less likely (disadvantage). The converse might tend to "flatten" as a good chance of success is moved to a moderate chance with disadvantage and vice versa, but the prior cases are the more common (especially as it's often a player initiated resource or a team initiated one). In other words, it usually makes the likely outcome even more likely, which cuts against a resolution system that really needs reasonable chances of success or failure to do it's thing (the snowball).

I was thinking more about advantage than disadvantage, so I'll stick to that.

What I had in mind is that if my chance of success is 1 in 20, then advantage nearly doubles that (39/400 is near enough to 1 in 10); while if my chance is already good, then advantage doesn't increase it as much (eg if its 50/50 it goes to 3 in 4, which is only 50% more likely; if its 4 in 5 then it goes to 24/25, which is only 20% more likely).

But if I'm following properly, the general experience is that doubling a small chance doesn't, in practice, make much difference (eg because those checks don't come up often enough for the doubling to show through) while the more modest increase in to big chances does make a difference (eg because those checks come up a fair bit and already weren't too likely to fail and now are even less likely).

You mean 'Bounded Accuracy?' Or the nominal easy/hard/etc guidelines?

The two in combination, I think, because its the relationship between bonuses and DCs that determines the prospects of success, which matter to the viability of conflict resolution for the reasons [MENTION=16814]Ovinomancer[/MENTION] has given.

Thinking through some more maths:

Suppose a DC of 15 and a bonus of +1. Then the chance of success is 7 in 20, but with advantage is 231/400, or about 11.5 in 20. The latter sort of odds is enough to support conflict resolution in Burning Wheel, but the result is that the players (and their PCs) do fail a lot and hence the play experience can be pretty demanding on them. And demanding on the GM too, because it puts a lot of pressure on the GM to effectively narrate failures.

I think D&D (and I include 4e here) has never provided a lot of support to the GM in narrating failure effectively. I don't have a good sense of how much better 5e might be in this respect, but if the general tendency in play is to incline towards making checks with significantly better than 50/50 odds then maybe it doesn't come up too much?

I think this also touches on another corollary to your posts, that task -> conflict resolution, when well coupled, doesn't imply that a successful task resolution means a successful conflict resolution. Instead, a successful task should always be coupled to an advancement towards a successful conflict resolution, and vice versa, but not exactly one for one. The Baker quote and your example don't explicitly make this point, although it is lurking in the shadows, so to speak, especially in your play example. I wanted to bring it out more clearly.

Can you elaborate? You may have noticed something in my play example that I missed.

 

[pemerton]

Just picking up on a few bits of your interesting post:

a) This includes the "Not Yet" rule: if a Conflict has not mechanically been resolved (won?), you cannot narrate an event that would effectively resolve it.

<snip>

1) This doesn't play super nice with the quasi-simulation skill system.
2) D&D combat is already (in some ways) a very complicated Conflict Resolution system (its only resolving the one event over and over again...but...) and making the two systems mesh might be problematic.

The most developed non-combat resolution system for D&D that I'm aware of is the skill challenge in 4e. It needs your (a) but no rulebook directly states it. A GM needs either to bring that from outside (normally by experience with another game with better-stated rules), or intuit it, or else complain that skill challenges are broken because we have to keep rolling the dice even though the conflict is resolved!

I'm not sure about your (1). In 4e the standard solution is to just ignore all the quasi-simulationist stuff in the PHB skills chapter (which is mostly dropped in Essentials, for good reason). I'm not sure that 5e skills are really even quasi-simulatoinist, though I'm not the best qualified to comment. I think the issue is less about quasi-simulation and more about setting appropriate expectations for players and GMs: eg having a good Investigation skill means (something like) when a conflict involves investigating stuff, than I'm more likely to succeed at that conflict than others. This will cause a lot of players to go ballistic but for culture/expectation reasons rather than narrowly mechanical reasons.

Your (2) is a recurring issue in 4e play, although most of us who care have developed various sorts of workarounds/coping mechanisms. It's an issue even in a system that one might expect to be tighter than 4e, like BW: eg the rules for BW set out both an action economy and a correlation of that action economy to ingame fictional passage of time for it's ranged skirmish conflict resolution system and its melee combat conflict resolution system. The ingame time is different for each (a roughly 10:1 ratio), but the GM is encouraged, when adjudicating the two side-by-side, to ignore this and run them more-or-less back-and-forth. It's a smaller thing than the 4e issues, but is still a blemish rather than a virtue.

 

[pemerton]

Why did you think an example with a large portion of the method "implicit" was a good one for trying to get others to understand? The thing that was implicit was, in fact, *central to understanding what you were doing*.

I can only work with what I've got!

Is "bad" a jargon word with the game?

I was hoping that people's general familiarity with the Cthulhu genre would do some work in parsing the example. So whether fluid spilled down a grate is going to awaken Deep Ones in the sewers, or send the inhabitants of the house insane by contaminating their water supply, or something else appropriately Cthulhu-esque is an open question, but we can all see - I hope - that these are the sorts of bad things that happen in a Cthulhu RPG.

 

[pemerton]

There is a lot here, so forgive me if I'm missing some parts. But I don't see how this making your point that it is better for the DM or GM to tell you the consequences of your actions.

<snip>

This likely does not tell the player anything they did not already know.

<snip>

The entire point I've been told is that we have to tell the player the consequences of their actions, or otherwise they cannot make an informed decision about the fate of their character.

"This is important, failing is bad" is not what I have taken that to mean.

<snip>

Or reading the next few posts, is knowing the consequences for an action are simply bad, and that failure will have an effect on the game, all you were going for? Because if that is the case, then I see no reason to tell the players this before every single roll, which was what I was arguing against in the original thread.

I've honed in on these bits of your post because I think they might be the best place to start.

I personally think the issue of telling - if that means explicitly stating as a precursor to the roll - is a bit of a red herring, because in RPG play, especially among participants who are used to playing together, there are many ways to convey information and establish expectations other than explicit telling.

But I think reducing what is conveyed to [/I]consequences for an action are simply bad[/i] is not correct. And that's really what I see as the focus of the discussion. It's not irrelevant - [MENTION=177]Umbran[/MENTION] canvassed, upthread, that a consequence of spilling might be good (eg if it stops the BBEG getting the fluid) and that's a possibility that is excluded in the context of my play example - but it's too simplistic. What's the nature of the badness? Who is implicated? What sorts of things might be required to effectively respond to it?

I've played games in which the answers to those questions are known only to the GM, and the players can't act on them except by way of either (i) guesswork, or (ii) declaring actions that will get the GM to release the answers, or bits of them (which is the analysis/study I referred to in my OP). Whereas in my example, the player knows the parameters of the answers to those questions without having to guess and without having to engage in further action declarations. The player knows that the badness will pertain to something that the player has put forward as significant in the game (in virtue of his PC build) - the master/servant relationship, the propriety of butlering, or something in that neighbourhood. Given the master/servant relationships established in play - the PC's own, and that between the NPC butler and the NPC master, the latter of whom is ultimately in charge of the mysterious fluid - the player knows what elements of the shared fiction have to be focused on in order to uncover and possibly resolve or at least respond to the bad thing.

One way to describe the approach, at a fairly high level of generalisation, is that by choosing to play a butler and then by following that up with a choice to help his butler friend do his duty by looking after the fluid, the player has placed a constraint on what I, as GM, am entitled to establish as consequences of failing in the context of that second choice. So the player knows the general parameters for consequences because he set them. So they can't be something known only to the GM!

Once we step down from that high level of generalisation we can add another gloss: the GM is allowed to push the player on what it means to be a butler, and what counts as doing the right thing as a butler. And one way of doing that is through the narration of consequences. I regard this as just about the highest-risk aspect of the approach I'm describing: because if you don't push at all then the player may never feel challenged and things can be too "pat"; but if you push too hard or misjudge what the player will regard as fair provocation (as opposed to just overriding their concept) then the whole edifice can fall over, with unhappy consequences for trust and other feelings. That's why I regard the back-and-forth between player and GM as super-important, even among people who know one another (because we might know one another well, but have we ever discussed butlering before?). And that back-and-forth itself helps warm the player up for what might be coming downstream, as well as providing the GM with the stepping stones for more dramatic or hard-hitting consequences downstream, and so it both telegraphs and lays the ground for surprise - which might seem paradoxical, but hopefully I've succeeded in explaining why it's not.

For some RPGers all of the above might seem obvious, and straightforward or even oversimplified.

But my own experiences, both of play and of posting, make it clear to me that it's quite a different approach from that which many RPGers use. Just to give one example: I've read many posters saying that they design adventures before the players generate their PCs; whereas the approach I'm describing makes that literally impossible.