Smart vs. Intelligence and Combatless Roleplaying Sessions

[fusangite]

That's a perfectly valid way to play a role-playing game. However, I play DnD where by RAW, a Bluff roll plus CHA bonus vs Sense Motive plus Wis allows me to succeed at convincing an NPC of my falsehood.

My understanding of the RAW is a little less absolute. Don't circumstance modifiers play a role here? If you just play that a Bluff roll convinces a person of anything if it exceeds their Sense Motive check, you're making Bluff equivalent to Suggestion and even more powerful Enchantment and Illusion spells (e.g. Mass Suggestion, etc.). So, I don't think that straight skill vs. skill is "what the framers intended" when they wrote 3.5; it is clear that circumstance modifiers are not intended only fixed-DC checks; they are a tool GMs can and should use at all times.

The type of information a player needs to give a GM in order to adjust the circumstance modifier, whether they are required to provide exact words, a general logical thrust for their statement or just a high check on Knowledge (Local), is clearly what is at issue in this particular discussion as it pertains to character social skills.

From this discussion, it's starting to look like "equality in task resolution" and task resolution through dialogue are mutually exclusive - you can't do both at the same time, though compromises are possbile.

Finally, I don't see that any part of the rules provides for absolute equality in task resolution. Otherwise playing would require no skill at all; players would just become dice-rolling machines. If two people are playing Rogues and one has a solid sense of geometry and battle tactics whereas the other one does not, there will be no equality in task resolution at all. The former player will deal double or more the amount of sneak attack damage per melee combat compared to the latter; and there is no Knowledge (Tactics) in the RAW that the latter player can roll on to compensate so as to have the GM control his character's movements in battle (assuming the GM's tactics are better than the player's in the first place). I don't see why we should legislate absolute equality in task resolution in every area except combat given this clear inequality embedded in the RAW.

 

[ThirdWizard]

My understanding of the RAW is a little less absolute. Don't circumstance modifiers play a role here? If you just play that a Bluff roll convinces a person of anything if it exceeds their Sense Motive check, you're making Bluff equivalent to Suggestion and even more powerful Enchantment and Illusion spells (e.g. Mass Suggestion, etc.).

Example Circumstances (Sense MotiveModifier)
The target wants to believe you. (-5)
The bluff is believable and doesn’t affect the target much. (+0)
The bluff is a little hard to believe or puts the target at some risk. (+5)
The bluff is hard to believe or puts the target at significant risk. (+10)
The bluff is way out there, almost too incredible to consider. (+20)

 

[DamionW]

Well for one thing, D&D isn't Monopoly. Whatever the current RAW, resolution through RP is well established in custom and practice. Just because the latest writers of the rule books decided to include these rules doesn't mean that's how I want to play.

I am aware that DnD isn't Monopoly, but I used the analogy to highlight several points:

1. DnD is a game, and all players are cooperating to enjoy themselves. One player is granted by all of the other players more authority to arbitrate, but in the end, the DM is just another individual at the table playing the game.
2. Rules exist within the framework of a game in order to ensure all players are treated fairly by the other players.
3. The choice of words to express for your character is resultant of that player's style and proficiency at developing a chain of dialogue
4. The player's proficiency with verbal exchange, including intonation, pitch, pacing, facial expressions, body gestures and all other factors that contribute to the believability of a premise or falsehood does not directly correlate with the character's proficiency in the same areas. Otherwise, there would be no need for a CHA score, a INT score or any other mental abilities because those come directly from the shared link with the player's knowledge and verbal prowess.
5. To reward that player's choice of words based on their proficiency in dialogue instead of their character's proficiency is a personal decision by the DM that they enjoyed the player's style of playing the game. It is an arbitrary, biased award because it may not be equally distributed to every player; not every player has the same proficiency.
6. Thus it circumvents the rules and is an abuse by the person granted authority by the other players.


If you don't like the social mechanics and hate being bogged down with dice rolling, here's an example of a way to still reduce the lack of fairness: Include circumstance bonuses and have one side of the discussion Take 10.

Say Joe the Fighter (CHA 10, Bluff 0) is trying to convince a guard to let him pass. You RP the exchange as the guard (WIS 11, Sense Motive 0, for ex) and the in-character lie Joe's player gives seems somewhat plausible and you as DM think it could work. Pick either the player or you to roll and assume the other side is taking 10 because they're not under pressure. At that point, it becomes the equivalent of a coin flip or a d2 roll (10 Bluff vs. X Sense Motive or X Bluff vs. 10 Sense Motive. If Joe fails, you can explain it off that the character was sweating to much or stumbled on some words that the player didn't. If Joe's player wants to avoid that next time, put at least a few ranks in bluff or stop using CHA as a dump stat.

If Joe's player came up with a stupendous lie and an enjoyable table performance, you can still add a +2 to +4 circumstance bonus because you liked watching the player talk. One side can still take 10, but if the other side rolls that 1 or that 20, as appropriate, Joe's out of luck. At least it still isn't an auto-success just because you like how Joe's Player suddenly became a Level 3 abstraction of Joe.

This process removes the possibility of DM bias to a specific player and a specific style of RPing they may possess. The bottom line is if you just happen to believe the dialogue yourself based on the delivery and want the NPC to act as if they believe automatically, you're forgetting one thing. This is an area where Joe is weaker than Joe's player. Joe might be able to kill hundreds of orcs and the player couldn't hold his own in a streetfight, but the player trumps Joe in lying because the player designed Joe by definition to be less persuasive than he is.

Fair enough, play as you like. This discussion is about personal preferences for play style, as much as a few people seem to think that play styles can be determined to be objectively right or wrong. Assuming the RAW is the "correct" way to play is just as bad as the reverse; there's nothing sacred about the RAW.

All I've been saying is that in my personal experience, resolution of tasks through RP has been more fun than doing so through skill checks.

I'm not claiming one is more correct. I'm claiming one is more prone to bias because it stops examining the character's capabilities and starts relying solely on the player's capabilities. Even this is fine, everyone is entitled to play the game the way they like. I just urge if you have new players as a DM that it should be made clear, up front, that you are expecting the authority from the players to decide what's believable in game based on how believable the players are meta-game. If you wait until it arises in the course of the plot, that's too late. It is as Voadam's example of deciding you don't want ranger's to have spells, but you wait until the ranger PC is at 4th level to spring it on them.

 

[DamionW]

My understanding of the RAW is a little less absolute. Don't circumstance modifiers play a role here? If you just play that a Bluff roll convinces a person of anything if it exceeds their Sense Motive check, you're making Bluff equivalent to Suggestion and even more powerful Enchantment and Illusion spells (e.g. Mass Suggestion, etc.). So, I don't think that straight skill vs. skill is "what the framers intended" when they wrote 3.5; it is clear that circumstance modifiers are not intended only fixed-DC checks; they are a tool GMs can and should use at all times.

The type of information a player needs to give a GM in order to adjust the circumstance modifier, whether they are required to provide exact words, a general logical thrust for their statement or just a high check on Knowledge (Local), is clearly what is at issue in this particular discussion as it pertains to character social skills.Finally, I don't see that any part of the rules provides for absolute equality in task resolution. Otherwise playing would require no skill at all; players would just become dice-rolling machines. If two people are playing Rogues and one has a solid sense of geometry and battle tactics whereas the other one does not, there will be no equality in task resolution at all. The former player will deal double or more the amount of sneak attack damage per melee combat compared to the latter; and there is no Knowledge (Tactics) in the RAW that the latter player can roll on to compensate so as to have the GM control his character's movements in battle (assuming the GM's tactics are better than the player's in the first place). I don't see why we should legislate absolute equality in task resolution in every area except combat given this clear inequality embedded in the RAW.

My claim is not that "anything is believable, just make a check". My claim is that a premise of a lie described at Level 2 abstraction (see post #165) is sufficient to determine the circumstance modifier (See post #182) and the DM can determine success or failure at that point. The choice of my lie tactically, say between "I tell him his boss sent him and he wants the guard in his office now!" and another lie, say, "I tell him I'm Elminster and that I will smite him if he doesn't let me through," is the same tactical decision as flanking or not flanking as a rogue. To ask me to lie as my character would lie and he'll respond as the NPC would respond is removing the exchange to a meta-game arena where my personal skills, as a real person, to lie are brought into question. This is not done for any other aspect of the game such that my abilities are paramount to the characters, and thus is prone to bias and outside the scope of the rules.

 

[fusangite]

My claim is not that "anything is believable, just make a check". My claim is that a premise of a lie described at Level 2 abstraction (see post #165) is sufficient to determine the circumstance modifier (See post #182) and the DM can determine success or failure at that point. The choice of my lie tactically, say between "I tell him his boss sent him and he wants the guard in his office now!" and another lie, say, "I tell him I'm Elminster and that I will smite him if he doesn't let me through," is the same tactical decision as flanking or not flanking as a rogue. To ask me to lie as my character would lie and he'll respond as the NPC would respond is removing the exchange to a meta-game arena where my personal skills, as a real person, to lie are brought into question.

Well, I don't think anybody is proposing to simply ignore all Cha-based skills in their game. If a GM did so, I would expect him to refund my skill points in Cha-based skills so I could put them somewhere useful. So, the meta-game arena is not removed; it is just informed by a different type and quantity of data, unless the dice stop rolling and checks stop being made altogether.

Furthermore, would you not agree that the critical reasoning skills required to do your level 2 abstraction style of play are, just like talking charismatically, distributed unevenly amongst D&D players? You're not really suggesting that metagame constructs should effect complete player equality; you are suggesting that metagame constructs should allow smart people to outperform less smart people in all situations. That's not equality; it's just a more predictable form of hierarchy. Some of my players are better at speaking persuasively than they are at reasoning; others are better at reasoning than they are at speaking persuasively. What you are suggesting here is not player equality; it is privileging the latter group over the former.

What I am suggesting is that the level of abstraction/detail required from a player in setting a DC is GM-specific. Some games I've been in (I didn't like them much) would have demanded even less than the level two abstractions on which your games run. In these games, players would be making Int checks, Know (Local) and Know (Nobility) checks all the time because they didn't want to bother to remember who the NPCs were, what their agendas were and what the overarching plot was to any level of detail. In my view, these games did not violate the RAW any more than Don's does. Charisma-based skill checks, in my view, have been sensibly designed to accommodate various levels of abstraction; combat, on the other hand, is fixed to one particular level of abstraction. However, I'm not prepared to follow you down the road of arguing that the level of abstraction legislated for combat must be legislated across the board for all types of checks.

This is not done for any other aspect of the game such that my abilities are paramount to the characters, and thus is prone to bias and outside the scope of the rules.

Who said that requiring people to act our verbal interactions makes the information from the acting paramount in how the GM resolves things? 70% of communication is non-verbal anyway; the dice can still govern 100% of that.

 

[DamionW]

Well, I don't think anybody is proposing to simply ignore all Cha-based skills in their game. If a GM did so, I would expect him to refund my skill points in Cha-based skills so I could put them somewhere useful. So, the meta-game arena is not removed; it is just informed by a different type and quantity of data, unless the dice stop rolling and checks stop being made altogether.

The thing is, Fusangite, there are DMs who DO feel CHA based skills can be ignored in the game. To quote some examples in this thread:

<snip>

Even a bluff of 15+ is usually good enough to get by most bluff checks, so even an orc with aa -1 charisma and no ranks can still have a possiblity of lying his way out of encounters. Thus if I did not want to use the dice mechanic on this skill, I could very well let a charasimatic player slide with a very well played lie every now and then.

....Abstracting aspects of the game that a player is weak in into skill checks essentialy removes that aspect of play from the game. I enjoy resolving interaction with NPCs through dialogue. If you instead depend on a character's skills, it's a better simulation of the character's abilities but less fun. Basically you're evening the field between the socially adept guy and the social klutz by removing the part of the game where the socially adept guy has an advantage. For me, the cost of this approach far outweighs the benefit.

There are a number of things I think should be left to player ability because a) I enjoy doing them and b) they're difficult to model well. Social interaction is certainly one of these. It's difficult to model well with D&D because the rules are simplistic. It's difficult to model in general, because no matter how high your diplomacy skill is, you don't get to enjoy being good at a negotiation, you only get to win and enjoy the reward. You still miss out on the experience.

If I were designing the game I'd leave out skills like diplomacy and tactics, using instead something along the line of "social perception" and "tactical perception" that allow the DM to give appropriate hints but is explicitly not used as a resolution method.

<snip>

Right, the mechanics themselves are not the roleplaying.

Taking your third example giving the speech is roleplaying the character, rolling the dice is an optional task resolution mechanic. How the bandits react can be determined by mechanical skills and dice rolls or by DM judgment.

Saying "My character is good at intimidation, I'm not. I use intimidate to get them to surrender . . . 16" is simple character task resolution without roleplaying.

Dice don't preclude roleplaying at the same time as game mechanical task resolution. They are just different actions.

(Emphasis added by me in editing)

So, we see three different DMs whose playstyle encourages the removal of a CHA-based die roll when RPing a dialogue between PC and NPC. The dialogue should flow naturally and using dice to determine the NPCs reactions is unnatural. They are not the only ones, I have played with several other DMs of this mindset.

That is fine with me, I can't tell them they're wrong, that's their style. However, I usually find myself blindsided by such a playstyle mid-campaign as a player when I design my character with an emphasis on CHA based skills and then don't find out I can't use them only until I'm right in front of an NPC obstacle. This usually happens because the DM finds the player-to-DM dialogue resolution mechanic as so fundamental to the game that they don't inform me up-front that that is their preferred playstyle. They simply think it's the way the game should be played and me asking for a dice roll to sanity check my character's Bluff is an alien concept

Furthermore, would you not agree that the critical reasoning skills required to do your level 2 abstraction style of play are, just like talking charismatically, distributed unevenly amongst D&D players? You're not really suggesting that metagame constructs should effect complete player equality; you are suggesting that metagame constructs should allow smart people to outperform less smart people in all situations. That's not equality; it's just a more predictable form of hierarchy. Some of my players are better at speaking persuasively than they are at reasoning; others are better at reasoning than they are at speaking persuasively. What you are suggesting here is not player equality; it is privileging the latter group over the former.

I think the skills needed to perform level 2 abstraction playing are what make roleplayers roleplayers and not Level 1 abstractionists. If you are unable to make decisions enough to abstract at Level 2 what your character should do, RPGs are probably not a good hobby for you at all. What my suggestion is is that in no one area should my ability as a real person to perform at Level 3 (to actually swing a sword in combat, to actually build a house with carpentry, to actually deliver an effective lie) dictate an automatic success or failure for my character's level 3 action in game (To fictionally swing a sword in combat with an attack roll, to fictionally build a house with Profession: carpenter, to fictionally convince an NPC with the Bluff skill). The two actions and the skills needed are separate and uncorrelated, because I am not actually my character and we don't possess identical skills and knowledge.

What I am suggesting is that the level of abstraction/detail required from a player in setting a DC is GM-specific. Some games I've been in (I didn't like them much) would have demanded even less than the level two abstractions on which your games run. In these games, players would be making Int checks, Know (Local) and Know (Nobility) checks all the time because they didn't want to bother to remember who the NPCs were, what their agendas were and what the overarching plot was to any level of detail. In my view, these games did not violate the RAW any more than Don's does. Charisma-based skill checks, in my view, have been sensibly designed to accommodate various levels of abstraction; combat, on the other hand, is fixed to one particular level of abstraction. However, I'm not prepared to follow you down the road of arguing that the level of abstraction legislated for combat must be legislated across the board for all types of checks.Who said that requiring people to act our verbal interactions makes the information from the acting paramount in how the GM resolves things? 70% of communication is non-verbal anyway; the dice can still govern 100% of that.

To answer your final question, there are DMs whose playstyle says the information I give you when verbally act out my character does trump dice rolls. I've been placed in situations where I give a perfectly plausible, well described premise for a lie in level 2 abstraction. The DM then asks me to RP out that lie to see if it works. That's asking me to abstract at Level 3 for bluffing when they never ask for it in any other mechanic, simply because their enjoyment from the game is gained by witnessing my attempts at dialogue. I cannot convince DMs like this they're wrong. My argument is that this needs to admitted up front to players at the start of a campaign where and when the player's actual abilities (to either deliver a lie, or (to get back to the OP) to solve a puzzle) will be the dictating factor in the character's success. At least then the player has the option to not design a character more charismatic or smart than they are or to not play in that DM's game. To not tell the player is to deprive them of that choice.

 

[Mishihari Lord]

If you don't like the social mechanics and hate being bogged down with dice rolling, here's an example of a way to still reduce the lack of fairness: Include circumstance bonuses and have one side of the discussion Take 10.

Say Joe the Fighter (CHA 10, Bluff 0) is trying to convince a guard to let him pass. You RP the exchange as the guard (WIS 11, Sense Motive 0, for ex) and the in-character lie Joe's player gives seems somewhat plausible and you as DM think it could work. Pick either the player or you to roll and assume the other side is taking 10 because they're not under pressure. At that point, it becomes the equivalent of a coin flip or a d2 roll (10 Bluff vs. X Sense Motive or X Bluff vs. 10 Sense Motive. If Joe fails, you can explain it off that the character was sweating to much or stumbled on some words that the player didn't. If Joe's player wants to avoid that next time, put at least a few ranks in bluff or stop using CHA as a dump stat.

If Joe's player came up with a stupendous lie and an enjoyable table performance, you can still add a +2 to +4 circumstance bonus because you liked watching the player talk. One side can still take 10, but if the other side rolls that 1 or that 20, as appropriate, Joe's out of luck. At least it still isn't an auto-success just because you like how Joe's Player suddenly became a Level 3 abstraction of Joe.

This process removes the possibility of DM bias to a specific player and a specific style of RPing they may possess. The bottom line is if you just happen to believe the dialogue yourself based on the delivery and want the NPC to act as if they believe automatically, you're forgetting one thing. This is an area where Joe is weaker than Joe's player. Joe might be able to kill hundreds of orcs and the player couldn't hold his own in a streetfight, but the player trumps Joe in lying because the player designed Joe by definition to be less persuasive than he is.

I like the method you outlined, and may give it a try. I'd probably add a few things, though.

1) To encourage newbies, I'd put in a "handicap." i.e. If they're just not that good at in-character dialogue yet, then I'd judge their peformance more leniently than someone who's been playing for years. The only way to get better at it is to do it, and I would like to get them up to speed as quickly as possible.

2) I'd like the circumstance modifier for performance stronger, give it a +10 to -10 range.

3) Insist that the dialogue be appropriate to the character played in order to get a positive circumstance modifier. A typical bard and a typical barbarian might have the same level of skill at diplomacy, but how they persuade will be very different. As an example, the bard might use flattery and extravagant words, while the barbarian would use a guileless look, a pledge on his honor, and an oath on his father's soul.

The only issue I had with your statement was where you said this addresses "lack of fairness" If persuasion through in-character RP is part of play, then one player being more persuasuve than another is no more unfair than one player being better than another in a one-on-one basketball game.

 

[ThirdWizard]

The only issue I had with your statement was where you said this addresses "lack of fairness" If persuasion through in-character RP is part of play, then one player being more persuasuve than another is no more unfair than one player being better than another in a one-on-one basketball game.

The "lack of fairness" here would probably be the 8 Charisma no ranks in Bluff character whose Player is very persuasive being a better lyer than the 20 Charisma character with 15 ranks in Bluff.

 

[Lonely Tylenol]

Well, I don't think anybody is proposing to simply ignore all Cha-based skills in their game. If a GM did so, I would expect him to refund my skill points in Cha-based skills so I could put them somewhere useful. So, the meta-game arena is not removed; it is just informed by a different type and quantity of data, unless the dice stop rolling and checks stop being made altogether.

I think if you reread some of the posts in this thread, you'll find that certain posters are suggesting exactly that the dice stop rolling and Cha-based skills be ignored. Whether they'd advise players of this style and allow them to avoid spending skill points in these skills has not been established.

Charisma-based skill checks, in my view, have been sensibly designed to accommodate various levels of abstraction; combat, on the other hand, is fixed to one particular level of abstraction. However, I'm not prepared to follow you down the road of arguing that the level of abstraction legislated for combat must be legislated across the board for all types of checks.[/QUOTE]

I'm not sure he's arguing that. Certainly I'm not. But I do feel that, in a game that includes Cha-based skill checks, using player skill to sidestep the mechanic for character skill is essentially cheating. It's not cheating only if checks are enforced (with or without circumstance bonuses for good RP) or if those skills are removed entirely. In the latter case, it's still an unfair system. The players with better skill at negotiation have an unfair advantage over those with poor skills in that area. This is not balanced by anything, and so if you're not a "face man" in real life you can't be in game. Not to mention that since all social challenge resolution is determined by the DM's whim, you might as well just buy him a pizza and hope he decides he likes you for the rest of the session. I thought we had left arbitrary and unfair DM rulings behind with the old editions. I certainly don't miss them.

No matter what level of abstraction is present in the role-playing side of the game, the mechanical side is what determines what is fair and balanced for play.

 

[fusangite]

I think if you reread some of the posts in this thread, you'll find that certain posters are suggesting exactly that the dice stop rolling and Cha-based skills be ignored. Whether they'd advise players of this style and allow them to avoid spending skill points in these skills has not been established.

Thanks. I did re-read the thread more carefully and I do find myself rather less sympathetic to the more extreme positions on my "side." And I am with you in wanting to hear from Don et al what happens to points in these skills.

But I do feel that, in a game that includes Cha-based skill checks, using player skill to sidestep the mechanic for character skill is essentially cheating. It's not cheating only if checks are enforced (with or without circumstance bonuses for good RP) or if those skills are removed entirely.

Agreed. Player persuasiveness may inform these checks; it may heavily inform them but I agree with you that it should absolutely not wholly replace them.

In the latter case, it's still an unfair system. The players with better skill at negotiation have an unfair advantage over those with poor skills in that area.

So what? People with better geometry skills get more flanking bonuses because they know where to position their characters. People with better logic can make more effective use of suggestions spells. Etc. Etc. Not all players are as smart as eachother; not all players are as observant as one another. Does this make D&D unfair? Of course not. I do not understand what is so special about articulateness that it is not treated as just another talent a player can bring to bear in the game.

This is not balanced by anything, and so if you're not a "face man" in real life you can't be in game.

Richard Nixon became President, for God's sake. The idea that one cannot succeed at diplomacy or politics without personal charisma is just bunk. It's not impossible, just harder. It just takes more work.

I'm a pretty funny, charismatic guy in many situations. But I suck at comic book dialogue; I have no intuition for it. So, in one of Teflon Billy's superhero games, I used to write down heroic things for my character to say in advance because I lacked the talent to make them up on the fly.

It sounds like many people are frustrated by the idea that the personal attributes that make them good at some of D&D don't make them good at all of D&D. In my view, if one wants a good gaming dynamic, one's game should reward the widest rather than narrowest range of real-life skills. The more different kinds of real world skills allow your players to shine, the more diverse and interesting the group you can assemble.

Not to mention that since all social challenge resolution is determined by the DM's whim, you might as well just buy him a pizza and hope he decides he likes you for the rest of the session.

DM "whim" determines the stats of every monster you face and the DC of every skill check you make; why would social skills be a special area of corruption?

No matter what level of abstraction is present in the role-playing side of the game, the mechanical side is what determines what is fair and balanced for play.

Fair and balanced for whom? How is that "fair and balanced" for people who are bad at game mechanics? Obviously, your idea of fairness is privileging people good at game mechanics over people with all other types of skills. To me that's not especially fair; so I try to make my games a hybrid of the two things that you are placing in opposition.

And frankly, on its face, your concluding statement is completely ridiculous. Basically, you are saying that it is unfair to reward role playing in a role playing game. Why are they called "role playing games," then?

The thing is, Fusangite, there are DMs who DO feel CHA based skills can be ignored in the game. To quote some examples in this thread:

I stand corrected.

What my suggestion is is that in no one area should my ability as a real person to perform at Level 3 (to actually swing a sword in combat, to actually build a house with carpentry, to actually deliver an effective lie) dictate an automatic success or failure for my character's level 3 action in game (To fictionally swing a sword in combat with an attack roll, to fictionally build a house with Profession: carpenter, to fictionally convince an NPC with the Bluff skill). The two actions and the skills needed are separate and uncorrelated, because I am not actually my character and we don't possess identical skills and knowledge.

But combat is a special case. To argue that because combat can't be informed by real world skills, neither can any other aspect of the game is fallacious. It does not logically follow. As you, yourself point out, figuring out the plot of the game and remembering crucial details is just the opposite: there is such total overlap between character skill and player skill that these things are barely represented in the mechanics at all. RPGs contain a continuum of player vs. character skill overlap with combat on one extreme and memory on the opposite.

To answer your final question, there are DMs whose playstyle says the information I give you when verbally act out my character does trump dice rolls. I've been placed in situations where I give a perfectly plausible, well described premise for a lie in level 2 abstraction. The DM then asks me to RP out that lie to see if it works. That's asking me to abstract at Level 3 for bluffing when they never ask for it in any other mechanic, simply because their enjoyment from the game is gained by witnessing my attempts at dialogue. I cannot convince DMs like this they're wrong.

Well, they are wrong to go to that extreme. But because people carry acting out interactions too far doesn't mean that this kind of play has no place in D&D.