D&D 5E (2014) Consequences of Failure

[iserith]

To clarify, I wasn't trying to suggest rolling dice is important for it's own sake.

Right, I didn't take it that way. I took it mean all of these things you go on to helpful point out that are a subset of that:

Rather, the chance introduced by the dice may useful as a tool. For example, rolling a die introduces a degree of impartiality to the knowledge check, and at some tables that emphasize the "DM as referee" style that impartiality maybe inherently valued. As another example, if the DM hadn't actually created the setting information requested, a check can be useful to determine whether the DM needs to improvise on the spot (or how much they need to improvise at tables that allow non-binary check results).

This below is the part that some people are disagreeing over, at least in terms of what does or does not qualify as a meaningful consequence:

Finally, the DM may simply think that it is uncertain whether the character would know the requested information, and want to use a check to resolve that uncertainty. (Depending on the player's intended use for the information, the DM might consider not knowing the info to be a meaningful consequence because it constrains future options, in which case the rules suggest that an ability check would indeed be appropriate.)

I submit that sometimes, as with any action, sometimes there's a meaningful consequence for failure. Others may say that you should always roll (which means it follows there is always a meaningful consequence for failure). Others say don't roll at all (which means it follows there is never a meaningful consequence for failure).

Still others base their position outside of these parameters by saying those rules don't matter, are in error, are taken out of context, or can be dismissed in certain cases.

 

[Guest 6801328]

Of course, that solution introduces the new metagame issue that if you only call for a roll when there is a potential observer, any time you call for a roll you reveal the presence of a potential observer.

Either approach creates a metagame issue, and I think people just disagree as to which one they feel is more problematic.

Or just reveal the potential observer before calling for the roll?

This is another one of those things that hinges on what you find to be interesting/exciting adventure design, including hidden knowledge. Like "random" traps. Or NPCs who may or may not be lying.

Not every methodology will be ideally suited to implementing every challenge.

 

[Celebrim]

Wait...what?!?!?! Did anybody else in this thread infer that from my first post?

"One question that might arise is whether failing a stealth check, and thus failing to hide, really counts as a consequence. Isn't that the same outcome as not rolling at all?"

Perhaps the confusion is on your understanding what I meant by my example. That is, for the purpose of my example, I am assuming that the player wants the pie, must get the pie, and that the only method of getting the pie involves getting past the orc. For that purpose, the decision to "not go in there" is the same as choosing to fail, in that the player fails in their goal, acquiring the pie.

"So I think a key feature is that the player has to actively / knowingly undertake a task with risk. If the party hears something coming and they say, "Let's all hide!" my instinct would be to say "Ok, let's have stealth checks." But in this case the failure state IS the same as not doing anything."

I've never really understood why you perceive this as a big distinction. If I change my very simple Orc and pie example such that the Orc is on patrol and part of his patrol carries him past the position that the PC is in, then it is in my opinion still isomorphic to the first scenario. If the player in this situation chooses not to hide, then they receive the same outcome as if they had failed a stealth check, just as the player approaching the sleeping orc achieves the same outcome by choosing not to be stealthy as if they had failed a stealth check. And we still end up with the same two possible states:

a) I have some advantage on the orc.

b) I do not have some advantage on the orc.

Both cases involve no change in the status quo on failure, but do have a change in the status quo on success. Is forgoing success a meaningful consequence of failure?

 

[iserith]

As a technical definition, I mean a process of play where there is a proposition filter in place that invalidates a proposition unless it is stated in a manner that gives a clear in character action that the character is performing along with a goal that they are trying to obtain.

As a less technical definition, tell the GM what your character is trying to achieve and how they are going about it.

I traditionally understand the goal of this approach is to force the player to engage with the fiction rather than the rules by removing ambiguity in process resolution that results from stating your approach in terms of the desired check or mechanic you wish to use. I imagine that in play, there are different levels of strictness applied by DMs to the proposition filter, but that they all generally would fall under the heading of "goal and approach".

Thank you, that matches my understanding of the term.

I'm afraid my answer to your previous question isn't going to be very conclusive: People will disagree on what "meaningful" means in context or whether that rule should be considered in the first place. The DM ultimately has to make the call. I would say that rule must always apply, that it's a fundamental requirement of the adjudication process. Others disagree.

An outcome is uncertain if the action is not trivially easy and not impossible. Let's say, without referencing any specific example (which is problematic on these forums), that it falls between those two standards. Then comes the requirement for a failure condition of that uncertain outcome to be meaningful. If it's not meaningful, then there's no ability check. But the DM still decides that ultimately. And some DMs aren't going to agree on that.

 

[Charlaquin]

Of course, that solution introduces the new metagame issue that if you only call for a roll when there is a potential observer, any time you call for a roll you reveal the presence of a potential observer.

Either approach creates a metagame issue, and I think people just disagree as to which one they feel is more problematic.

One approach creates an issue where players may have access to knowledge that their character’s might not have access to (commonly called “metagaming”). The other creates an issue where the player is able to retry until they get a natural 20, which must be resolved by a construct that lies outside of the standard framework (a meta-game solution). Two different uses of the combination of the word “game” and the prefix “meta.” Personally, I don’t have an problem with the former type of “metagaming,” and have a big problem with using meta-game constructs to limit a player’s ability to take action. Your mileage may vary, of course; plenty of DMs have the opposite preference.

 

[Guest 6801328]

"One question that might arise is whether failing a stealth check, and thus failing to hide, really counts as a consequence. Isn't that the same outcome as not rolling at all?"

Oh. Apparently my rhetorical techniques are poorly presented.

What I meant by this is "One objection some of you may have is..." Not that I myself believe it.

Perhaps the confusion is on your understanding what I meant by my example. That is, for the purpose of my example, I am assuming that the player wants the pie, must get the pie, and that the only method of getting the pie involves getting past the orc. For that purpose, the decision to "not go in there" is the same as choosing to fail, in that the player fails in their goal, acquiring the pie.

I think that's a very different failure, in the sense that the game state hasn't changed, the player can still choose to stealth, and they also still have other options because the orc has not yet been alerted.

"So I think a key feature is that the player has to actively / knowingly undertake a task with risk. If the party hears something coming and they say, "Let's all hide!" my instinct would be to say "Ok, let's have stealth checks." But in this case the failure state IS the same as not doing anything."

I've never really understood why you perceive this as a big distinction. If I change my very simple Orc and pie example such that the Orc is on patrol and part of his patrol carries him past the position that the PC is in, then it is in my opinion still isomorphic to the first scenario. If the player in this situation chooses not to hide, then they receive the same outcome as if they had failed a stealth check, just as the player approaching the sleeping orc achieves the same outcome by choosing not to be stealthy as if they had failed a stealth check. And we still end up with the same two possible states:

a) I have some advantage on the orc.

b) I do not have some advantage on the orc.

Both cases involve no change in the status quo on failure, but do have a change in the status quo on success. Is forgoing success a meaningful consequence of failure?

In this variant the game state is about to change unless the PCs take action.

The distinction is clear to me. It's interesting that it's not to you. I'm guessing there's some other fundamental difference in our POVs, and this difference is a symptom of that.

 

[Ratskinner]

Wow, sorry I missed to much of the discussion on this already.

Going back to the OP, and not having read all the posts in between....just tossing in my $.02

Other games have "solved" some of what you're talking about.

The important thing is that everyone knows what is at stake with a roll.

One game that seems to be designed almost entirely along the lines you are talking about is Schema. When a player is about to take an action, the GM lays out the stakes. (The game actually has little cards you can print out to place on the table with a good selection of general risks.) These include both risks and rewards. The player rolls an appropriate dice pool and can use the results to buy rewards or buy off risks (each die will come up either useless, good for buying a reward, or good for buying off a risk). You can even specify your attitude toward the action where acting boldly will let you buy more rewards, but being cautious lets you buy off more risks. (Each makes it less likely that you can get the other results.)

Thinking about: "Waking the Dragon". Games like Blades in the Dark, and even ICRPG use "clocks" or "countdowns" to make impending consequences mechanically evident and clear. In fact, the ur-example in the BitD book is "Alerting the Guards".

As ICRPG might indicate, adding clocks to D&D is relatively easy: Good rolls either mark a box on a positive clock (get help from the witch) and failures mark boxes on negative clocks (the guards have found us). When the clock has counted down, that thing happens.

I hope that's helpful.

 

[Xetheral]

Or just reveal the potential observer before calling for the roll?

This is another one of those things that hinges on what you find to be interesting/exciting adventure design, including hidden knowledge. Like "random" traps. Or NPCs who may or may not be lying.

Not every methodology will be ideally suited to implementing every challenge.

Very true. I was thinking particularly of situations where the potential observer is itself hidden (or otherwise unobserved), but I didn't emphasize that in my post.

As an example, if a PC is trying to move through a crowded riot without being identified by anyone who is looking for her, she will likely need to make a stealth check. If you wait until there is someone looking for her nearby to call for the stealth check, you reveal such a person is nearby. Whether that's better or worse than the player knowing in advance whether the character is doing a good or a bad job is definitely a matter of opinion.

 

[iserith]

Of course, that solution introduces the new metagame issue that if you only call for a roll when there is a potential observer, any time you call for a roll you reveal the presence of a potential observer.

Which only matters in the context of games where rolls don't result in a consequence that changes the situation. My solution: Ask for rolls when there's an uncertain outcome and a meaningful consequence for failure.

Edit: Reworded that to be a little less confusing.

 

[iserith]

Very true. I was thinking particularly of situations where the potential observer is itself hidden (or otherwise unobserved), but I didn't emphasize that in my post.

As an example, if a PC is trying to move through a crowded riot without being identified by anyone who is looking for her, she will likely need to make a stealth check. If you wait until there is someone looking for her nearby to call for the stealth check, you reveal such a person is nearby. Whether that's better or worse than the player knowing in advance whether the character is doing a good or a bad job is definitely a matter of opinion.

In that case, the DM describes the observer as acting in a manner that he or she has spotted the PC, which changes the situation, and leads to the next decision point: "What do you do about that, PC?"