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

[Bill Zebub]

I've been thinking more about the "can I tell if the NPC is lying?" thing. This is another one of those things where I wonder why we even use this device, and maybe the only reason players think they need to be able to detect lies is because they've gotten used to rolling for it.

I'll fall back on my position that if the resolution of something can be left to a single die roll, then it's not really that important. Let's say the character has a 70% chance of succeeding at the Insight roll. That's still a 30% chance that they will fail. And if 70/30 is good enough, then probably 50/50 is, too. So just let the players go with their hunch? The advantage of leaving them wondering is...well, it leaves them wondering! More mystery! More suspense! More "aha!" moment when they find out if the NPC was lying.

And if 50/50 is not good enough? If knowing whether or not the NPC is lying is so important that it can't be left to chance? Well, then why are they leaving it to chance with an Insight roll? They should be finding out other ways of knowing for sure.

So, yeah, I don't think lie detecting as a skill roll improves the game at all.

 

[Lanefan]

I want to argue that combat is the only part of the game in which the default should be reaching for dice and relying on mechanics. That in ALL other parts of the game we should first try to resolve things through narration/storytelling, and only roll dice as a last resort to resolve real uncertainty. Or, at least, that's how I personally would like to play. I believe that, outside of combat, rolling dice should follow this pattern:

1. The player declares an action
2. If the DM decides the outcome is uncertain, and there is a real cost to failure, the DM may call for a roll
3. The DM tells the player what the roll will be, including DC, and what the cost of failure will be
4. The player then has the option of not taking the action

And, by the way, "you can't try again" is not a cost of failure, at least by my definition. If you try to pick a lock and you fail, the door is still locked; the game state hasn't changed.

IMO your whole premise is already fatally flawed at this point. Once a player declares an action then that's it: the character is committed to taking that action. Step 4 should read "The player (or the DM, if appropriate) rolls, and the game goes on."

Litmus Test: if the DM's only tool for preventing everybody in the party from trying is by declaring (unrealistically) that only one character is allowed to try, then clearly the penalty for failure is insufficient.

Agreed, though I'm fine with failure merely maintaining the status quo.

This means:

• No passive rolls to spot things
• No rolls to see "if I know something"
• No rolling Insight to detect lies
• For those who want NPCs to "use social skills on PCs" the pattern is perfectly symmetric, which means the DM describes the action, the player decides whether outcome is uncertain, and calls for a roll, setting the DC.

And one final thing: I'm in the camp where I don't police, or even worry about, "metagaming" (using the narrow and somewhat inaccurate definition of "not separating player and character knowledge about the game world.")

Another irrecoverable flaw in the premise is that we're expected to take the bolded as the default state, which I will not do.

If there's no passive rolls to spot things then how do you handle the very frequent situation where, when the PCs aren't specifically looking for something or are looking for something else, they may or may not notice something that's not necessarily front-and-centre? Ultra-basic example: on returning to a previously-explored room they may or may not notice the desk is just slightly out of place from where it was before (in the intervening time someone's come in and quickly searched it) yet it;s highly unlikely the players are going to have their characters re-check the room in case anything has been disturbed.

But doing his is hard. Both because I got used to playing and another way, and just because sometimes it's hard. I'm still practicing DMing this way.

So to help me practice, here's the challenge: describe a scenario in which you think it would be challenging to follow these principles, and I'll see if I can figure out either how to handle the scenario, or how (and why) I would prefer to set up the scenario differently in the first place. Others are free to respond also. Maybe we'll all learn something.

Example:
"The party is exploring a maze of nearly identical passages, and there is a secret door in one otherwise unremarkable tunnel. How do you determine if the secret door is found without passive rolls or cost of failure?"

My answer:

• First, what purpose does the secret door serve in terms of making the game more fun. Is it just a random short-cut? Does it lead to a treasure room? Does it make the challenges faced by the party objectively easier?
• If it's just a random short-cut or otherwise provides a minor benefit, I might telegraph it's presence when they are near. For example, the party might intermittently notice footprints, and I'll tell them (no roll required!) that the footprints have disappeared. If they search around near where the footprints end, they find the door automatically. (Alternatively, I might eliminate the secret door as pointless.)
• If it's important, such as leading to a treasure room or making the party's objectives significantly easier, I would want to telegraph it from another location and then let them deduce the likely location. Any attempt to actively search for it in the correct location would be successful (but see next comment). The telegraph could be in a journal or map they find, a comment by a prisoner, a symmetric/geometric map in which one part is "missing", etc.
• If they are actively looking for a secret door but are under time pressure, then I might ask for a roll. The cost of failure is using up time. E.g., they are being pursued and want to use the secret door to hide from their pursuers before they are caught.

My take on the example: if they think to look for a secret door they'll have the usual odds of finding it if they look in the right place and won't find anything if they look elsewhere. I won't telegraph it, though; if it's important enough they'll eventually realize on their own that they need to find it and if it's not that important, who cares?

Put another way, it's no skin off my-as-DM nose if they never find it even if what's behind it is important to the adventure or story, or is major treasure. I've seen parties miss 75% or more of the treasure in an adventure, and generally assume they'll always miss some of it - doesn't bother me.

I'd love to have this thread NOT devolve into a debate about metagaming.

Good luck with that, given that by the bolded above you've made it a central issue to your premise.

 

[Lanefan]

The general philosophy I'm working through is to ask: is it really more fun for the players to be given information because of a successful passive perception check, rather than just being automatically handed the information? Why?

Yes it is, as the players can more easily relate to their characters who - as in real life - aren't necessarily perfect and are capable of both noticing and missing non-obvious things.

 

[Lanefan]

Yes, I agree. Which is why I never rely on telegraphing, and player deduction, for things that are necessary to success, but only for "bonuses" such as treasure rooms.

To the bolded: why not? Why can't they outright fail or blow a mission sometimes due to missing some subtle clue or hint? Yes it might come down to sheer luck (the passive roll) as to whether they notice that subtle element, but that's life.

And, really, if the players are not going to find the treasure room, I'd rather have it be because they missed or misinterpreted the clues, than because some passive check failed.

Flip side: the passive check gives them one last chance to find it even if they missed the clues.

 

[Lanefan]

Yeah, exceptions to rolling in combat are really good to keep in mind, too. I'm still haunted by an adjudication in my early days of DMing 5e when my then 11 year old son, playing a half-orc assassin, wanted to execute an unsuspecting enemy. I made him roll. Ugh.

IMO you made the right call by making him roll, as there's always the chance the assassination doesn't go as intended - the target might randomly move at the last minute, the assassin might give away his presence unintentionally, the assassin's blade might catch on a buckle or unexpected piece of armour on the target, or whatever.

It should be an easy roll, sure, but he still has to make it.

 

[Lanefan]

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....

Or it could be set up such that while bringing the portcullis down safely requires strength, anyone can just trip the ratchet to send the portcullis crashing down into the battle, potentially killing friend and-or foe outright...

 

[Gus L]

I want to argue that combat is the only part of the game in which the default should be reaching for dice and relying on mechanics. That in ALL other parts of the game we should first try to resolve things through narration/storytelling, and only roll dice as a last resort to resolve real uncertainty. Or, at least, that's how I personally would like to play. I believe that, outside of combat, rolling dice should follow this pattern:

1. The player declares an action
2. If the DM decides the outcome is uncertain, and there is a real cost to failure, the DM may call for a roll
3. The DM tells the player what the roll will be, including DC, and what the cost of failure will be
4. The player then has the option of not taking the action

I largely agree with this though I'd add some caveats:

1st Caveat. This is dependent on system and play style. If one is playing a style of game that's largely about tactical combat and character building where non-combat actions need to be pushed through quickly and are a lesser/smaller part of the game it's fine and good to have a robust system of mechanics (based on char building really) that both fill in the narrative space between set-piece tactical combat and encourage char builds that aren't entirely focused on being combat optimal.
2nd Caveat This structure is not used for EVERY obstacle or encounter outside of combat. Things like traps and surprises may simply force a roll ... a saving throw so to speak. Furthermore and related it's good to have several basic sets of mechanics that the referee can declare will be used to resolve general types of obstacles. For example I would use a "Save vs. Paralysis" (the old style save unrelated to stats) as a way to avoid getting caught in an avalanche, but might use a DEX check to leap over a chasm, potentially allowing "climbing/acrobatics" to be used by a PC with that skill.

In general though I think it's key in an exploration focused game, especially if one is using a higher lethality set of rules for the referee to A) question players who are about to undertake a declared action that seems odd, especially dangerous or incongruous with the situation B) Give an example of how the mechanics will work if mechanics are involved prior to the player committing to actions. This is because I assume PC competence, and acknowledge that my descriptions may be incomplete, confusing or that players may have been distracted.

Example: There's an 80' deep well and a player says "My character is just going to hop down and have a look." Obviously this is weird - but to the player they may have missed how obviously deep the well is. I would stop and confirm saying something like "You can't see the bottom of the well even with the lantern shining down it and you will take 1D6 per 10' of falling damage" ... because the PCs are going to shine a light down there before leaping. I don't need to be antagonistic and that includes concealing info. As an "old school" referee the world and narrative are entirely in my control. Lightning from the heavens can kill the PCs if I desire etc. I designed the dungeon, and it could always be harder. I have no need to hide the details of challenges - because I want the players to have all available information to try to solve them.

Another reason this is useful is because it increases trust between players and referee - meaning paradoxically that the world can actually be harsher, because the players accept that the referee is being honest and not hiding information from them. The puzzles/obstacles/challenges are about what they are intrinsically rather then a game of rules manipulation or finding the flaw in the referee's description. I actually include a lot of challenges that I don't have immediate answers to in my dungeon design (at least for my own games) because players are quite good at figuring out ways around/through them. To do this though it really helps if the players trust the referee is being fair - that if they come up with a plausible scheme the referee will first give them mechanics (replicating the characters ability to judge the intangibles that we can't describe completely) and allow success if the idea is good/or they beat the odds.

And, by the way, "you can't try again" is not a cost of failure, at least by my definition. If you try to pick a lock and you fail, the door is still locked; the game state hasn't changed.

Litmus Test: if the DM's only tool for preventing everybody in the party from trying is by declaring (unrealistically) that only one character is allowed to try, then clearly the penalty for failure is insufficient.

Fair - My penalty is almost always the passage of a turn. Turns passing means random encounter checks, which mean burned resources for no treasure (almost always). I don't think this is an absolute though, there are times when only one PC can try or there's only one chance to try. Lassoing a fleeing horse or something else momentary comes to mind. Disarming a trap also comes to mind as failure may trigger it.

I think the point about any of these procedural questions is that they are general, but not absolute. The referee exists not simply to set to clockwork of the rules in motion and enforce procedure, or even to decide when a specific rule applies -- but also to decide when a situation calls for an ad hoc adjustment to rule and procedure.

 

[Minigiant]

I've been thinking more about the "can I tell if the NPC is lying?" thing. This is another one of those things where I wonder why we even use this device, and maybe the only reason players think they need to be able to detect lies is because they've gotten used to rolling for it

Lie detection is a thing as an artificial function of making the Wisdom score matter outside of mind control spells.

This is the crux of the issue

Many of these unnecessary rolls are balancing factors.

 

[Bae'zel]

We sometimes use dice rolls for social interaction because none of my players are professional actors and/or feel comfortable acting out certain scenes (eg, seduction, torture etc).

I’m curious about those of you who omit non-combat dice rolls. How do you handle character creation? Do you tell your players NOT to invest in Persuasion, Deception, Intimidation, Perception, History etc?

In other words, do you even use skills in your game? Why are you even playing D&D post 2e?

Edit: genuinely confused and curious.

 

[TiQuinn]

We sometimes use dice rolls for social interaction because none of my players are professional actors and/or feel comfortable acting out certain scenes (eg, seduction, torture etc).

I’m curious about those of you who omit non-combat dice rolls. How do you handle character creation? Do you tell your players NOT to invest in Persuasion, Deception, Intimidation, Perception, History etc?

In other words, do you even use skills in your game? Why are you even playing D&D post 2e?

Edit: genuinely confused and curious.

Nah, if I don’t want to use skills like that, I don’t try to make 5e into a different game. I do more of that in my 2e game where skills are a lot more fine-grained.