D&D General Renamed Thread: "The Illusion of Agency"

[Bill Zebub]

Ok, here's a great example of my principle of "hard to find, or hard to act on, but not both." I would not require a check to spot the portcullis. I would announce its presence.

BUT....operating it would be a different matter. Maybe it's rusted, maybe it has a lock on it, maybe it just requires a lot of strength. But to use it would require taking attention (and Actions) away from the fight and dedicating them to dropping the portcullis. Maybe, for example, it would take 2 uninterrupted turns for one character to do it, but the guards would attack whoever was trying to do it, and that character would have to make what are essentially Concentration checks to continue...

So now the players have interesting decisions to make. Do they take resources away from the fight and concentrate on the portcullis? Who should do it? Should the others focus resources less on winning the fight and more on protecting that designated character? Etc.

With more thought, I might also require Str checks. That makes the decision of who should do it even more fraught, because their best front line combatant is probably also their best portcullis operator. Mwuhahahahahaha....

 

[Bill Zebub]

My point is that D&D doesn't use exploration and social rolls, their mechanics and their more associated attributes enough as is.

Reducing it more would disrupt balance too much.

It's like Intelligence before the psychic spells were added.

Intelligence was a dump stat because INT checks and saves came up so infrequently. Every party was 50% drooling moronic powerhouses as players shifted power and went out their way to avoid actively using their low Intelligence by other means.

Oh, I thought we were talking about rewarding investment, but you are talking about dump stats.

Well in that case I don't think my suggestion changes the calculus at all. I do agree with what you are saying, but whether one uses the approach I'm describing or not, most of the time a party still only needs one specialist in each of these skills.

I guess the exception might be Perception, because of the "Everybody roll Perception!" phenomenon.

 

[Bill Zebub]

It is impossible to prove you wrong that you personally prefer to play without dice rolls outside combat, but I can very easily prove you wrong that I greatly prefer the 5e Rules As Written with skillchecks to determine success outside combat.

Then I refer you to I'm A Banana's post upthread:
I believe this is how the game advises you to play it already. No changes need to accomplish this.

I'm really not arguing to change the rules, but to actually follow them as described in the PHB.

I want character builds and characters' abilities actually matter in gameplay, and not have all outcomes determined by the players' skill. This is a personal preference and I understand that others may disagree, but I would get bored very quickly if DnD worked the way you are describing.

Yes, I do, too! It's already a long thread so I can't suggest you read the whole thing, but really I think what I'm advocating is in line with what you are saying.

 

[Charlaquin]

My issue here is the zero cost of failure. If one player doesn't know anything about trolls, aren't they all going to say, "Can I roll?" And while that's not really a problem, the fact that it happens suggests to me that something isn't working well.

In combat, I might give them a choice: "I'll let you roll, and if you fail I'll still give you the information, but you'll miss your turn because you were thinking so hard."

Outside of combat....? I don't know; if they have proficiency maybe just give it to them?

I’d say the cost of failure is the opportunity cost. I know “you can’t try again” is not a cost for failure we would typically use, but I think in this case it’s appropriate - if you can’t remember any useful stories from Uncle Gustav, thinking harder won’t really change that. But, to keep this from being a complete “I see that I rolled a 2, why can’t I just try again?” situation, I like to treat failure on knowledge checks as progress with a setback. You will learn something, but what you learn may or may not be directly applicable to the situation and hand.

As for the “can I try too?” effect, well, do they all have an Uncle Gustav who used to be a troll hunter? If not, then no, they can’t try too. They might try something else, but then it’s going to be a separate action, with a separate approach. Maybe one character studied trolls in wizard school, and another is an avid reader of Volo’s works. YMMV, but I don’t mind the cascade of knowledge checks if each check has a different approach and each approach gives us a little bit of new information about the characters’ histories.

Alternatively, you could say that once two or more players make the same knowledge check, it becomes a group check. It kind of makes sense, if everyone is trying to contribute information they recall from disparate sources, and not everyone’s recollection is accurate, the real challenge is in sorting out the good information from the bad. If at least half the group passes their checks, you’re able to glean something useful from everyone’s half-remembered contributions, but if more than half fail, the signal-to-noise ratio is too poor to draw any conclusions from.

Honestly I don't actually play that way. I just let the player decide. But it's my answer to people who think that what's good for the goose is good for the gander.

Fair, and like I said I think it’s an interesting suggestion!

Oh that's a hard one! How devious! (I should have known I could count on you.)

Umm.....

It seems to me that as long as a critical item (e.g. a magic item) is not the object being stolen, the stakes are not so high that it's unfair to let the thief automatically succeed. So this is not well thought out, but a couple of ideas:

1. Let the players learn from an NPC that Strahd does this (maybe telling a story about it happened to this poor vampire hunter who wandered through). If they catch on and take precautions, they stop the theft. Otherwise the object is gone.
2. Same as above, but with some kind of cost for staying up night.

Good answers! I have a couple of ideas for how I might handle it as well.

1. Similar to your idea that maybe this scenario shouldn’t have an uncertain outcome. Instead of using rolls to resolve this, simply narrate the spy’s failure - “In the middle of the night, you’re awoken by a noise, and open your eyes to see an unfamiliar humanoid rummaging through your pack. They freeze for a brief moment, as their fight and flight instincts compete for dominance. What do you do?”

2. Give the player something to interact with. Narrate some indication that the character keeping watch notices, such as the sound of a twig snapping, or a shadow moving between the trees. Let the player of the character keeping watch respond as they will, and adjudicate their actions accordingly. This is basically the equivalent of telegraphing.

3. Kind of cheating here, since I said “without relying on passive checks” in my original framing of the scenario, but to my mind, keeping watch while your allies sleep is an action performed repeatedly over time, so I do think this is an appropriate case in which to have the lookout make a passive Wisdom (Perception) check.

 

[Bill Zebub]

I’d say the cost of failure is the opportunity cost. I know “you can’t try again” is not a cost for failure we would typically use, but I think in this case it’s appropriate - if you can’t remember any useful stories from Uncle Gustav, thinking harder won’t really change that. But, to keep this from being a complete “I see that I rolled a 2, why can’t I just try again?” situation, I like to treat failure on knowledge checks as progress with a setback. You will learn something, but what you learn may or may not be directly applicable to the situation and hand.

As for the “can I try too?” effect, well, do they all have an Uncle Gustav who used to be a troll hunter? If not, then no, they can’t try too. They might try something else, but then it’s going to be a separate action, with a separate approach. Maybe one character studied trolls in wizard school, and another is an avid reader of Volo’s works. YMMV, but I don’t mind the cascade of knowledge checks if each check has a different approach and each approach gives us a little bit of new information about the characters’ histories.

Alternatively, you could say that once two or more players make the same knowledge check, it becomes a group check. It kind of makes sense, if everyone is trying to contribute information they recall from disparate sources, and not everyone’s recollection is accurate, the real challenge is in sorting out the good information from the bad. If at least half the group passes their checks, you’re able to glean something useful from everyone’s half-remembered contributions, but if more than half fail, the signal-to-noise ratio is too poor to draw any conclusions from.

An idea I just thought of for knowledge checks is that on a failure you get TWO pieces of information, one correct one and one dangerously incorrect one.

That doesn't really solve for the 'cost of failure' though, since I'm sure players would still rather have that than nothing at all.

Good answers! I have a couple of ideas for how I might handle it as well.

1. Similar to your idea that maybe this scenario shouldn’t have an uncertain outcome. Instead of using rolls to resolve this, simply narrate the spy’s failure - “In the middle of the night, you’re awoken by a noise, and open your eyes to see an unfamiliar humanoid rummaging through your pack. They freeze for a brief moment, as their fight and flight instincts compete for dominance. What do you do?”

2. Give the player something to interact with. Narrate some indication that the character keeping watch notices, such as the sound of a twig snapping, or a shadow moving between the trees. Let the player of the character keeping watch respond as they will, and adjudicate their actions accordingly. This is basically the equivalent of telegraphing.

3. Kind of cheating here, since I said “without relying on passive checks” in my original framing of the scenario, but to my mind, keeping watch while your allies sleep is an action performed repeatedly over time, so I do think this is an appropriate case in which to have the lookout make a passive Wisdom (Perception) check.

Really, the more I think about it, the more I conclude that it was a huge mistake to include Perception in the game in the first place. It may be that this skill (and perhaps Investigation) just don't work in any way other than a straight-up, boardgamey "roll the dice and see what happens" way.

 

[Bill Zebub]

I’d say the cost of failure is the opportunity cost. I know “you can’t try again” is not a cost for failure we would typically use, but I think in this case it’s appropriate - if you can’t remember any useful stories from Uncle Gustav, thinking harder won’t really change that.

Actually, I want to poke at this a little more. I don't think an "opportunity cost" is really a cost, for the simple reason that if the DM says, "The DC will be 13, and if you fail you can't try again," why would anybody NOT try?

Where this takes me is: when players ask "Do I know something?" what will the impact be if the players get the information they want? And I think I have a few broad categories:

• The answer is just flavor. Give it to them.
• The answer will help them in some way, but not substantially change the game state. An example might be, "Can I read these runes?" where the runes give clues to the dungeon they are in. So I'll just give it them (or to whichever character is mostly like to 'know it').
• The answer would help them, but it will be more exciting and memorable if they find out the hard way. E.g., trolls and fire.
• The answer will have a substantial impact on the game. For this, they're going to have to work for it by tracking down the answer in a mini-adventure.

 

[bloodtide]

I mostly agree with your idea. It's close to the way I run my games. In my game it is what the players know, not the characters.

I don't use any type of knowledge or 'remember things' rolls. Characters know what the players know about the setting, lore, history and such. And yes, this very much means that if your a clueless player you can not play a character with any knowledge or skill. I require players to find information "for real". And nothing is 100% the absolute truth ever.

I allow players to "meta game". If you the player knows something about the setting, lore, history or anything else about the game world: then any character you play also knows all of that.

Where I would split your idea of just allowing the characters to automatically do endless things.....and only once in a while calling for a roll. I'm not in favor of such super powered characters that can alter game reality. And I'm not a DM who is "on the players side" and just rolls out the red carpet for whatever they want to do on a whim.

I would say I call for a roll nearly all of the time. I like the randomness of rolls, and I like the chance of triumph and defeat.

I would add that if something does not require a roll, it is not game worthy and should be skipped 100%. You touch on this.

I'm also big on the idea of making the above mechanics vague and mostly unknown to the players. Players don't get to know the DC or any way mechanically how hard something is: they only get told by description. Players know their character's 'plus' and anything in the basic rules, but not all the modifiers. And players only know what they "think" the failure will mean.

When characters find some mysterious runes, I will never in any way just "give it to them". Even if is very important. At best they will need a "mini adventure".

 

[DEFCON 1]

At my tables most communication between players is in-character, and thus I never need to worry about the "Can I try?" "Oh, can I try?" scenario. If the group is talking and someone comes up with an idea, then obviously it is now information that the entire group that is physically present has access to. So if someone mentions something that might require an Intelligence (History) check to give information on... then everyone there rolls and information will be doled out based on how well the characters rolled.

The one thing regarding ability checks that others always talk about that I really just don't follow is the idea that there has to be a "consequence for failure". I never require that, because more often than not ability checks that I call for are nothing more than basically random answer generators. The player says they wish their character to do something or know something, and rather than declare what the answer is or how well they do it... I let the die roll tell me what the answer is. Someone wants to know if they recognize a series of runes? They roll Intelligence (Arcana) and how well they do will inform me of how much information I give them. Someone wants to look for animal tracks? They roll Wisdom (Survival) and how well they roll will inform me how much info I give them regarding the tracks. If that's no information, one piece of information, two pieces, or such a good roll that I basically tell them exactly where they are headed... any of those can occur and I'm fine with giving out any and all of it.

Why? Because the way we play... the mechanics are not the important part of the game. The answers and results are. The results drive the story forward and that is what we are interested in... seeing everyone's story move forward. Mechanics are merely just a substitution for improvisation. I could improv the answers and results of everything if I really wanted to... giving out as much info out as I just chose to give in each moment... but the dice just lend a bit of credence on the player's part to know the amounts of info they can get, plus build their characters such that they get more info or action in those topics they'd prefer to specialize in.

Long story short... ability checks are basically an alternative method for determining how much I will improvise and give out rather than my own gut feeling.

 

[I'm A Banana]

Agreed.

I really don't think I'm describing anything different than exactly what it says in the PHB.
But...a lot of people don't seem to believe that. Or don't want to believe it. There seems to be a lot of assumption...based on posts here...that 5e is played the way previous editions were.

I think the other problem is that in official WotC adventures there are examples that contradict the PHB.

People are coming to D&D with outdated assumptions? The heck you say?

Totally agree about the effect of having that list of skills. But, really, I was trying to discuss how to DM this way, not how to change D&D to better support it.

I think I see the problem as one of design, as the intent of the rules is pretty clearly spelled out. If play isn't lining up with intent, that's a design issue in my mind.

I do think it's different. The key thing missing is the DM's declaration of the stakes, and the fact that there were no consequence for failure from player A. Player A didn't really have to make any kind of hard decision before declaring (and following through on) that action.

Here's the test: if player B, with +4 Intimidate, also fails, does player C, with a measly +2 Intimidate, also want to try? Then player D?

If so, I think something is not working well. Without meaningful consequences for failure, the only way to prevent the whole party from trying is for the DM to rule that no more attempts can be made, which I think is pretty common practice, and is accepted by players because it is common practice, but doesn't really have any in-story justification. (I personally hate it.)

I think there's broad variance. I'd call for a group check once the second person spoke up and then adjudicate based on the pass/fail ratio, possibly revising my previous declaration. But I don't know that this is "better" than everyone doing their own check.

Maybe a way to reframe the conversation: what are we looking to encourage? Because if we're hoping for more structure and bigger consequences in our skill checks, we probably need to give DM's a system that's as easy to plug into as combat is. 4e skill challenges and 3e complex checks were a run at this idea, though both had flaws. What's the new version of it?

 

[Charlaquin]

An idea I just thought of for knowledge checks is that on a failure you get TWO pieces of information, one correct one and one dangerously incorrect one.

That’s kind of fun, though I’m always hesitant with giving players false information on a knowledge check. I think there’s a lot of value in being able to maintain a clear line that you can always trust information given directly by the DM to be accurate, whereas information given via NPCs may or may not be.

Really, the more I think about it, the more I conclude that it was a huge mistake to include Perception in the game in the first place. It may be that this skill (and perhaps Investigation) just don't work in any way other than a straight-up, boardgamey "roll the dice and see what happens" way.

I disagree. I definitely think because it’s so much more useful than other skills it probably shouldn’t compete with them for character building resources, but I do think the game benefits from having something you can roll when the outcome of an attempt to discern sensory information is uncertain, and the game also benefits from the consistency of the “roll a d20, and an ability modifier, and potentially add proficiency bonus” universal resolution mechanic.

One useful way to think about it is to essentially treat all ability checks as saving throws. What you’re “saving” against is the consequences of your action failing.

Actually, I want to poke at this a little more. I don't think an "opportunity cost" is really a cost, for the simple reason that if the DM says, "The DC will be 13, and if you fail you can't try again," why would anybody NOT try?

Well, it depends. If the attempt requires a certain buy-in, such as the knowledge check requiring a declaration of where you might have learned this information you’re trying to recall, some players just won’t think of any. Moreover though, if you go with the “a knowledge check two or more characters attempt is a group check,” then the answer is that more characters contributing actually makes success less likely.

Mathematically, the ideal number of characters to have working on a group check is exactly two. Two characters making the same check and collectively succeeding if either of them succeeds is very similar to the character with the higher bonus rolling with advantage - slightly worse, but pretty similar. Once you add a third character, you now need at least two of them to succeed, which is always going to be less likely, regardless of the DC and the characters’ bonuses. Adding a fourth character makes your chances slightly better, but it’s always still worse than two characters, and that pattern repeats: an even number of contributors is always better than an odd number of contributors, and a smaller even number of contributors will always have a better chance than a bigger even number.

Where this takes me is: when players ask "Do I know something?" what will the impact be if the players get the information they want? And I think I have a few broad categories:

• The answer is just flavor. Give it to them.
• The answer will help them in some way, but not substantially change the game state. An example might be, "Can I read these runes?" where the runes give clues to the dungeon they are in. So I'll just give it them (or to whichever character is mostly like to 'know it').
• The answer would help them, but it will be more exciting and memorable if they find out the hard way. E.g., trolls and fire.
• The answer will have a substantial impact on the game. For this, they're going to have to work for it by tracking down the answer in a mini-adventure.

Fair enough, I suppose. I prefer a possibility of failure on types 1 and 2, and I’m a bit skeptical that type 3 really exists independently of type 4.