D&D 5E Consequences of Failure

[Oofta]

Could you explain how what I am describing is "actively encouraging people from trying to use their abilities"?

Wouldn't players want to propose approaches to their goals that:
A) Are more likely to autosucceed because they use the strengths of their character?
B) If a roll is called for, will use one of those strengths as the basis for the roll?



I agree. Which is why I grant autosuccess (or failure) on many tasks.



I started typing an answer to this, but honestly it's just inviting thread derailment. I started the thread pretty explicitly saying "this is for people who want to play this way." I'm not going to try to persuade people they should.

It's entirely possible I'm not grokking what you're trying to accomplish. I don't see how anyone can tell you how to do something better if we don't understand what it is you feel is missing.

It is my impression that you've defined the "how" without defining the "why". That doesn't have anything to do with DMing style.

 

[Swarmkeeper]

This thread has been enlightening so far. Thanks for starting it, @Elfcrusher

At our table last week, a player asked if his character recognized the leathery winged bird-like reptilian beasts that were attacking their ship in the fog. I simply said “no, these creatures appear to be from a land that time forgot.”

After the game, I second guessed myself. Was I telling the player what his character thought? Should I have asked him in return “tell me about what in your PC’s life experience might inform him about such creatures.”? In this case, I don’t think a roll was appropriate since there is no special ability to be gleaned for said creature. It was a Yes or No call on my part, but I still felt critical of myself that I may have botched it.

After reading some responses here, I now feel confident I can choose to say No since I had already established that the creatures were not known in the civilized world at all. But, with a good backstory reason (which I should have encouraged) I might have said Yes in the moment.

Now, what if PC’s backstory somehow included possible pertinent knowledge of said creatures AND said creatures had a special trait, like their bones could be ground up to brew a Heroism potion. Meaningful cost of failure = miss out on some treats. I guess I would set a reasonably difficult DC and ask for an INT roll letting the player know if they fail, they only get partial info since these creatures are so obscure. Whaddaya think?

 

[MarkB]

For the record I love the PoA system.

Were you suggesting I "take a closer look" in order to inform how I run 5e, or because I might want to play those instead of 5e?

Mostly because some of them have a handle on things like varying the degree of consequences for failure, and for success for that matter, which may help you with the whole "trying to find meaningful consequences" issue.

I particularly like Scum & Villainy's Position and Effect system, whereby the magnitude of both the consequences of failure and the effectiveness of success can be independently dialed up and down by the combination of the characters' current situation and their proposed approach and goal - and if they revise their approach, they can see how it changes their odds.

Something like that might be useful to incorporate into your game, either overtly as a gameplay tool or simply as a thought exercise when weighing up what sort of consequences to apply in a given situation.

 

[5ekyu]

So, just as a general note, the approach you're looking for seems very reminiscent of the concept behind most Powered By The Apocalypse games, in which the DM basically never rolls dice, and the players' rolls effectively determine the success or failure of both their own actions and those of their opposition.

I'm not suggesting that you go play those games instead, but I would suggest that it may be worth your time to take a closer look at some of those systems, and see how they handle those mechanics. I'm currently running a Scum and Villainy campaign, and the system has some useful dials for setting the outcome of both success and failure of checks.


That just feels the wrong way around. I know D&D's checks are generally binary, but as a player I'd dislike being penalised more for a near-miss than a complete failure.

It also doesn't feel like fully embracing the system you're proposing. If there are meaningful consequences for failure, then there shouldn't be a "nothing gained, nothing lost" middle ground - the middle ground is choosing not to make the roll.

This may be one of those "choose which you like the least" situations.

• Meaningful consequences for failure
• Rolls in the open
• Not roleplaying known-to-be-false information

Choose any two.

"This may be one of those "choose which you like the least" situations.

• Meaningful consequences for failure
• Rolls in the open
• Not roleplaying known-to-be-false information

Choose any two."

If the player does not know the DC and you use the PHB basic "some progress with setback" then you can have all three. They rolled, they got three nsmes, and even on a bad roll some may be true or may lead to very bad outcomes.

Consider, what if one of the things not included under "some progress" was that of the three experts mentioned, that first one actively fought to suppress the discovery using lethal means? Or that contacting them is likely to start opposition to further efforts?

 

[Guest 6801328]

It's entirely possible I'm not grokking what you're trying to accomplish. I don't see how anyone can tell you how to do something better if we don't understand what it is you feel is missing.

Sorry I can't explain it in a way that works for you! But thanks for trying to contribute.

 

[Oofta]

This thread has been enlightening so far. Thanks for starting it, @Elfcrusher

At our table last week, a player asked if his character recognized the leathery winged bird-like reptilian beasts that were attacking their ship in the fog. I simply said “no, these creatures appear to be from a land that time forgot.”

After the game, I second guessed myself. Was I telling the player what his character thought? Should I have asked him in return “tell me about what in your PC’s life experience might inform him about such creatures.”? In this case, I don’t think a roll was appropriate since there is no special ability to be gleaned for said creature. It was a Yes or No call on my part, but I still felt critical of myself that I may have botched it.

After reading some responses here, I now feel confident I can choose to say No since I had already established that the creatures were not known in the civilized world at all. But, with a good backstory reason (which I should have encouraged) I might have said Yes in the moment.

Now, what if PC’s backstory somehow included possible pertinent knowledge of said creatures AND said creatures had a special trait, like their bones could be ground up to brew a Heroism potion. Meaningful cost of failure = miss out on some treats. I guess I would set a reasonably difficult DC and ask for an INT roll letting the player know if they fail, they only get partial info since these creatures are so obscure. Whaddaya think?

I would have run it the same way in my campaign unless someone had previously established an unusual background.

Let's say someone has told me they don't remember where they came from and had just found themselves on the beach one day. That player has left open room for knowing a creature that is otherwise unknown.

But otherwise? If it hasn't been something they at least mentioned in their backstory, I don't see why some players wouldn't try to justify anything. Your flying creature? Yeah, that was in a weird book I found. That symbol to an unknown religion? An old priest in my hometown used to wear that and he told us all about it, I guess he wasn't crazy after all.

 

[Celebrim]

At our table last week, a player asked if his character recognized the leathery winged bird-like reptilian beasts that were attacking their ship in the fog. I simply said “no, these creatures appear to be from a land that time forgot.” ...After reading some responses here, I now feel confident I can choose to say No since I had already established that the creatures were not known in the civilized world at all.

Whaddaya think?

I think what you did counts to me as saying "Yes", not saying "No.", since you in fact imparted to the character a pertinent piece of information - that the creatures are not in fact known to the civilized world. That to me counts as the same sort of minimum success that I might grant to a player had they chosen to roll (or I had chosen to allow them to roll), as "This creature is not known to science" requires a pretty high knowledge of natural science and tells you something that is potentially important right there.

From my perspective what you decided was the DC to recognize that the creature was unknown to science was trivial, but the DC to know anything specific about a creature unknown to science was so high as to make it pointless to roll. So you didn't bother rolling.

I've been meaning to write up my own monster manual that systemizes what a player can learn about creatures from a background knowledge check, but the fact that it is hard to be very systematic about that - your decision that learning specifics about a creature no one has encountered before and lived to tell the tale is a valid one and an example of why being systematic about these things is hard - makes me hesitate to do that. That and the enormous amount of work for comparatively little gain. Deciding on a semi-arbitrary DC and providing one fact to the player with the highest roll for each point they beat the DC by has worked pretty well as an informal system.

 

[TheCosmicKid]

I've been meaning to write up my own monster manual that systemizes what a player can learn about creatures from a background knowledge check, but the fact that it is hard to be very systematic about that - your decision that learning specifics about a creature no one has encountered before and lived to tell the tale is a valid one and an example of why being systematic about these things is hard - makes me hesitate to do that. That and the enormous amount of work for comparatively little gain. Deciding on a semi-arbitrary DC and providing one fact to the player with the highest roll for each point they beat the DC by has worked pretty well as an informal system.

Ugh. I tried to write a detailed "Monster Lore" feature for a ranger rework. Ran into I suspect a lot of the same obstacles for systematizing knowledge checks. In the end I just punted: gave 'em advantage on checks to recall information about creatures, and left it to the DM to figure out what that meant.

 

[Lanefan]

A few thoughts on reading through this, in no particular order:

1. If you're looking for more-than-binary roll outcomes, maybe take a glance at PF2 and their critical success-failure ideas. There, it seems any check has 4 possible outcomes; and while over-codifed (of course it is, it's PF after all!) the general idea is worth a look.

Something I do, very informally and always situation-dependent, is let the die roll not only set the succeed-fail outcome but the degree of such. Let's use the sneak-past-the-dragon example, and say for these purposes that the outright pass-fail point is a natural 10 on the die. Well, if someone rolls 11 I'm going to narrate their sneaking as being barely good enough, and maybe even that the dragon noticed something amiss* but didn't specifically notice you. Meanwhile, rolling a 19 means you sail through, no problem. Ditto the other way - rolling 8 means your sneaking wasn't up to scratch and the dragon's now alert* to something, while rolling a 2 means you blew it and now there's a dragon in your face.

* - if I'm giving the dragon a perception check these middling rolls will inform how difficult/easy said perception roll will be.

2. For obscure knowledge checks I almost always give a roll (even if the odds are very poor), if only because we haven't roleplayed every minute of these characters' lives and who knows what obscure knowledge or trivia one or more of them might have picked up sometime in the past; be it during class training, or at grandma's knee, or while faffing around in the pub one night.

How I do this is that whoever asks gets a roll, then characters of class(es) that might know e.g. Clerics get a roll for anything religion-based, then I'll give one group roll for everyone else combined. And depending how obscure the sought information is, even a perfect roll might only get partial info.

3. The biggest problem with "make every roll meaningful" is this: players very quickly learn in the metagame that no roll means nothing to see here, even though the PCs have no way of knowing this, and this metagame aspect negatively affects play. Thus, I'm constantly throwing in 'null' rolls - an example of such might be searching a room for something that I-as-DM know isn't there - a good roll means they're also now sure it's not there, but a bad roll means they aren't sure if a) it's not there at all or b) it's there and they missed it.

4. Regarding "try until you succeed" situations: I always interpret a roll as being the absolute best shot you can give it, given the situation at hand. Thus, there's no such thing as 'take 20' or equivalent - if the DC of a lock, for example, is 17 and you've got all day to try and open it, if you roll a 15 you ain't opening that lock no matter how long you spend at it - this time, the lock wins. You don't get to keep rolling until you succeed.

Yes I know this goes against 3e-and-forward design. Tough.

 

[Bawylie]

A few thoughts on reading through this, in no particular order:

1. If you're looking for more-than-binary roll outcomes, maybe take a glance at PF2 and their critical success-failure ideas. There, it seems any check has 4 possible outcomes; and while over-codifed (of course it is, it's PF after all!) the general idea is worth a look.

Something I do, very informally and always situation-dependent, is let the die roll not only set the succeed-fail outcome but the degree of such. Let's use the sneak-past-the-dragon example, and say for these purposes that the outright pass-fail point is a natural 10 on the die. Well, if someone rolls 11 I'm going to narrate their sneaking as being barely good enough, and maybe even that the dragon noticed something amiss* but didn't specifically notice you. Meanwhile, rolling a 19 means you sail through, no problem. Ditto the other way - rolling 8 means your sneaking wasn't up to scratch and the dragon's now alert* to something, while rolling a 2 means you blew it and now there's a dragon in your face.

* - if I'm giving the dragon a perception check these middling rolls will inform how difficult/easy said perception roll will be.

2. For obscure knowledge checks I almost always give a roll (even if the odds are very poor), if only because we haven't roleplayed every minute of these characters' lives and who knows what obscure knowledge or trivia one or more of them might have picked up sometime in the past; be it during class training, or at grandma's knee, or while faffing around in the pub one night.

How I do this is that whoever asks gets a roll, then characters of class(es) that might know e.g. Clerics get a roll for anything religion-based, then I'll give one group roll for everyone else combined. And depending how obscure the sought information is, even a perfect roll might only get partial info.

3. The biggest problem with "make every roll meaningful" is this: players very quickly learn in the metagame that no roll means nothing to see here, even though the PCs have no way of knowing this, and this metagame aspect negatively affects play. Thus, I'm constantly throwing in 'null' rolls - an example of such might be searching a room for something that I-as-DM know isn't there - a good roll means they're also now sure it's not there, but a bad roll means they aren't sure if a) it's not there at all or b) it's there and they missed it.

4. Regarding "try until you succeed" situations: I always interpret a roll as being the absolute best shot you can give it, given the situation at hand. Thus, there's no such thing as 'take 20' or equivalent - if the DC of a lock, for example, is 17 and you've got all day to try and open it, if you roll a 15 you ain't opening that lock no matter how long you spend at it - this time, the lock wins. You don't get to keep rolling until you succeed.

Yes I know this goes against 3e-and-forward design. Tough.

I think point 3 makes assumptions of facts not in evidence.