Another Thread About Social Combat

[MGuy]

A better place to have this discussion than continuing to let shrap's thread be misused.

There are a lot of issues with Social Encounters in games as we all know so just to get the ball rolling I'll list some of mine. One of the big issues for me isn't setting a place where you can fail. That's the easy part and is as obvious as saying if X beats Y then social combat is won. It's the things that are unique to socializing that aren't the same as combat.

With combat, whether or not someone is defeated, cowed into retreating, or it is bypassed completely (barring some special situations) you still get toward some other goal. Whether it's to get to the end of the dungeon, to escape from the people or things you're engaging, etc etc. In most cases you just need to get through to the other end of the encounter. All the things you do are working toward some goal that gets you there and all the things others are doing help get you there. When another player fails that might make the end point more difficult to get to but it doesn't directly hamper you, you just lose the resource that's that player's character. With Social situations in all cases you are trying to convince a someone to do something. It doesn't matter if the someone is the person you are engaging. It doesn't matter if the something is just a general change of opinion.

What matters is that getting those goals have special roadblocks that combat, exploration, and the like don't have. For instance if a party chooses not to engage you. One of the issues that have been pointed out before is that there's no reason for a King to listen to a peasant. Worse, if you get a social combat system that can be cheesed together then the best thing to do is to not let anyone engage in social combat with you period because you can get talked into doing something you don't want to do. This happens irl of course but because this is a game it has to be treated as such and you need to have instances where your character can just get the dialogue option reliably.

Then there's the problem as Grek explains (in shrap's thread) that you need to have an endstate where you can call the conversation. The issue for me is that there are different things that need to be able to happen when you get there. Sometimes you want to convince the person you're talking to, to do/not do a thing. Sometimes you want them just to change how they feel about someone/something. Sometimes you want them to believe a thing. Sometimes you want multiple. Sometimes you want to convince someone 'else' who you aren't engaging with of these things. Sometimes you want this to be doable in one conversation. Sometimes you want this to be done over time. The differences between the desired ends will definitely matter because you are trying to effect a person and that person is going to react to the encounter in afterward instead of being just an encounter you have gotten past and can forget.

Then there's considerations for failure. There are a lot of conversations that definitely should just be failed as soon as someone says, does, or even 'is' the wrong thing when engaging. Threatening someone who is above your station can and should ruin your attempts to seek a peaceful resolution in a way that someone giving your presence away when sneaking past an encounter doesn't. Being a peasant/an unknown/a known enemy can and should gatekeep you from even being able to engage with a person in a way that being individually weaker than a monster doesn't. You could just keep the barbarian home when you go visit the king but that's sidelining a character but that's bad for the game. Yet it would be very reasonable to assume that the barbarian's presence for one reason or another works directly against your efforts. Again, this isn't an issue an issue for combat. It also can be argued that this is appropriate and something people do in the real world during negotiations. However for the game this isn't so good.

Those are failures that can just be nonstarters. You also have to consider what happens when you 'just' fail. So again, there should be a point where players lose. It should probably happen more 'here' than it does in combat since losing a conversation doesn't lose you the entire game (or at least in most cases it shouldn't lead directly to that). So this is not as much of a problem as the earlier issue with failure since you still get to play the diplo game without failing at the start. There are just as many considerations to be made about what happens when players lose, and what that means, as there are to consider when players win. How they lost should have outputs that players can gauge. Some that can be somewhat understood even before they start. There are likely going to be consequences for even asking the king to ally with the kingdom's arch rivals and players should be able to have some idea what the cost of not being able to pull it off in one speech are. Then there are people who will just 'fake' agreeing with you just to get you out of their face and then not act on the agreement/hold a grudge for the engagement that was meant to be positive. This should also be able to occur from the audience.

Lastly there are all sorts of conditions to worry about. Some I listed above but common things you might need to implement to give your system more believable are adjustments for: The personal social/material cost/gain for the person you're engaging with to agree with you (or even to be seen engaging with you), what kind of arguments work against what kind of people, faction conflicts, social status, past actions like favors/gifts/debts/trespasses, etc etc.

I list a lot of these things (that popped up off the top of my head) but there're probably more issues I could come up with if I think about it. But some aren't universal (like level differences mattering) or non issues (no the superman/lois lane thing is not a real issue). All of these also have solutions to them. Like having reputation points you can leverage to force people into conversations that they don't want to have (modified by general affability). Gifts/favors/debt/leverage/etc being able to further modify that number and provide for long term projects. Modifiers for types of arguments and how they align with a targets set of morals/philosophy/etc. Assigning player characters these things so their character's backgrounds/personality matter where this is concerned. Declared goals and degrees of success/failure determining what happens when you win/lose. Ego HP or a roll to determine how long an engagement can last to set a definitive end to any social encounter. I could go on but I haven't seen a system that gets at all these things I think I want out of a system that is going to take social encounters seriously. Also even thinking about it (and my ongoing attempts at creating one) tends to churn out something more complex than people want to deal with.

In the end I usually just settle for anything that is workable for the game and don't expect anything to be satisfying beyond that.

[Zinegata]

I'm commenting fully expecting Mguy to ban me from this thread, but what the hell - He pretty much stole this bit from me:

One of the big issues for me isn't setting a place where you can fail. That's the easy part and is as obvious as saying if X beats Y then social combat is won. It's the things that are unique to socializing that aren't the same as combat.

Because as I said in another thread:

No social combat mechanics work because social interactions don't have clear win / lose states like game-based combat. Most game-based combats are simply battles or skirmishes where one side wins while the others loses.

By contrast you don't "defeat" others in social interaction. Just because you think you bullied a poor store clerk into giving you a discount is not a "victory". Such actions have a social cost - the clerk is gonna harbor resentment that can bite you in the back later.

====

The issue with social combat isn't with the mechanics. Obsessing whether you use a D20 or a D6 to resolve "social combat" or situations doesn't add much.

The issue is that social situations in themselves are deeply complex. A social encounter is much more complicated than throwing a couple of monsters (each often with just one or two unique gimmicks) at a party.

Instead you have to track an enormous number of "interests" that each person in the conversation has - and many of those interests may not even be relevant in the present combat.

Let's say you're trying to flatter a store owner to give you a discount. You imply you'd go out with the storeowner on a date.

Most RPGs will reduce this to a single die roll, which decides whether the charm effect works or not. Then the player gets or doesn't get the discount.

In a real social situation however, the store owner will likely remember that you were hitting on them. And that can have an enormous number of branching consequences.

What if you charmed them, got a discount, but then the storeowner learns you're actually dating someone else?

What if the person you're dating then learns you hit on the store owner?

What if turns out the storeowner is actually already married, but unhappy and looking for some excitement? What happens when the husband finds out?

One very simple interaction can in fact result in a cascading branch of drama simply because each actor has different interests.

===

The issue here - and why social combat in RPGs always "fails" - is because most RPG systems just offer templates for resolving situations. In combat for instance you get your adventuring party, put them on a map, and throw in some monsters.

You can make the encounter much more intricate, complicated, and memorable - but that requires the DM to track more stats, read up on more monsters, and introduce more challenges and mechanics.

And while these more complex encounters are more memorable, they are also more prone to breaking down. A DM can totally design a 2 hour combat against a Dragon, only for the damn thing to roll a 1 on its initiative and roll a 1 against a save or die spell before it does anything.

Memorable social encounters are no different. A DM can totally make a complex chain of interlocking variables about the storeowner's strained relationship with her husband; which can be affected by the PC's attempts to charm her.

And yet it could turn out the PCs don't want to even bother charming her and pay 50% more than the standard retail price for items because they didn't want to bother haggling.

===

In short, social combat doesn't work not because of a failure of mechanics.

They instead don't work because they are mechanically complicated to plot out, and much of this plotting can just end up being unused.

It is instead always simpler to "wing" it - the DM needs to just use some empathy and common sense while the people on the table actually act out the social encounter.

Indeed, it can be argued that trying to turn social encounters into a mechanical dice roll is very much to miss the entire point of playing games to begin with. Games are a tool of socialization. They give you an excuse to interact with others in a way you normally don't.

[MGuy]

I've said it before but I'm not interested in creating a pariah. I'm not going to ban you from anything. I've just been keeping you on ignore until you get over your persecution complex or decide to leave. As long as you don't decide to make this thread about how you are mad that people you don't like post here I will interact with you like I would anyone else.

That being said I agree that a simple dice roll isn't enough. That's what most of what I typed up was saying. The shop owner example is pretty easy to adjudicate. Without getting into a bigger framework, and just using the example mechanics I've mentioned, you can do this encounter. With some fudging of what the numbers are of course.

So you have an easy to understand goal (increase favorability to get discount). You've got the reputation you came in with (0) and you go from there. You choose to use 'charm' and your style is seduction. This happens to be a particularly pious and married man. So the seducer is going to get a significant penalty on the attempt. Using degrees of success/failure you can get the outputs: the shopkeep can become hostile and kick the seducer out, unfriendly and rebuff them sternly while increasing their prices, brush off the flattery with some discomfort but sternly reject them, be flattered by the offer and offer a minor discount, be enticed by the offer and significantly reduce prices.

I'd imagine with the penalty a critical success/best result is off the table. This would be, I think very doable with just a few modifiers and considerations.

Side note I would not bother running actual encounters like this because I find it extremely boring and would just apply a passive discount based off of attributes/skill/reputation numbers.

[Blade]

It can help to restrict the scope of your social system. In some games, there are only a few social situations that are relevant to the PC's interests. There might also be a "tone" you expect.

If you play dashing swordspersons, your social system should mostly be about using witty retorts and rousing a crowd. You don't need rules for haggling or for long negotiations with the king.

[Omegonthesane]

If there are meant to be "assist" actions that can be taken to bolster someone else's point (assuming a game where not everyone is assumed to be proficient in or capable of social combat in every context) you could have retroactive legwork as a flavour justification to let the barbarian participate anyway.

If they're a charismatic leader among their own people, the camera cuts to them coaching the party noble before the negotiation with the Count. If they're able to pass for a lowborn peasant, we cut to them fraternising with the kitchen staff or some shit earlier to learn the Count's secrets and pass them on to the party. Somewhat if a case-by-case kludge, but does mean that you do not absolutely have to ignore that someone ought not to be able to take part in the conversation to let them take part in the combat mechanically.

[Zinegata]

That being said I agree that a simple dice roll isn't enough. That's what most of what I typed up was saying.

The thing is if you'd read all of what I actually posted my conclusion is the exact reverse. That's the main reason why I responded to begin with. Rather than trying to simulate the complexity, my conclusion (and the general trend industry-wide) is that it is simply not worth simulating.

But to be very clear:

Yes, you can plot out multiple branching conversation trees. You can consider alternate definitions for success or failure. You can plan out a very well thought-out social encounter. You can make a more complex social system. And they can work out exactly as intended sometimes (and for videogames, such carefully crafted encounters are valuable and memorable).

But on the the actual tabletop much of that effort turns out to be fruitless; because the players often end up ignoring your pre-planned dialogue options in the first place.

That's why "winging it" with a few die rolls is actually what works the best in most tabletop groups. Because very rarely will a DM have pre-planned all possible social interactions with the store owner. The DM will have to "wing it" if the party starts flirting with the storeowner for instance and he didn't plan for it.

And when you're "winging it", you're unlikely to be using very complex templates. Likewise, few DMs will want to spend many minutes poring over complex social encounter rules for something that just "came up".

DMs by and large will go with the flow and simply stick to a few simple checks.

===

Note also that's why I didn't respond or object to Grek's points to begin with.

Yes, he was describing a more complex system.

The thing is that social interaction system was very clearly theme-based and replaced much of the math usually reserved for combat. That's why you have the Horror, Sanity, and other Resolve variables specifically tailored for the setting. You can't have Cthulhu without Sanity for instance.

But if you're just adjudicating regular RPG interactions - like the aforementioned store clerk discount - there's generally no need for Horror, Sanity, or Resolve.

That's the thing people keep forgetting about RPG rules. Yes, you can have a very cool system, but you need to keep asking yourself how and if the DM will actually use it. A complex social system is great if it actively pushes the core themes and ideas of the game. But if it's too complex and doesn't enhance the theme they'll just skip over it.

That's why a ton of actual social encounters simply ends up like how Sheldon Cooper gave Howard Wolowitz an arbitrary +5 to his charisma check to avoid the ogre encounter: The DM found a player's joke to be funny, so he gave the player a bonus. The thing is, I'm pretty sure that kind of ruling isn't in the rules outside of DM's fiat, and it's likewise almost impossible to adjudicate using rules. Yet it does happen all the time.

[PhoneLobster]

Dammit I knew I should have been quicker to start this thread first. That's my whole partial draft out the window.

Pure Rules Free Resolution Sucks
So you want rules. "Bang you are dead"/"Nuh ah I'm not!" is bad we know that already, it still applies to social rules.

But Players SHOULD control the decisions/actions of the characters that belong to them
A lot of panic about social rules is wildly over hyped panic about the removal of "player agency". Apparently from past experience here I think I can safely say some minority of gamers will have a serious meltdown over the idea that someone might ever succeed in a game mechanical social action of any type against their character ever in ways that they are apparently not concerned about for game mechanical physical actions succeeding against them.

It IS stupid, certainly in the degrees and ways people presented it. But I think it represents a more valid, less absolutist underlying concern (well that and grognardism+traumatic personal life experiences with bad GMs).

Social mechanics CAN have a direct control over your ability to make decisions as a player about your character that are a step further than what other rules can do. People that seriously think their character should NEVER make a decision they didn't personally approve of are petulant panic merchants who don't understand game rules as a conflict resolution mechanic.

But a poorly conceived social mechanic CAN end up depriving player of agency in too many situations. At SOME point in a good social mechanic in SOME state, ideally the default state, player agency, even total player agency over what their character thinks and does SHOULD exist. Losing player agency, even over single decision should generally ONLY be the result of being on the losing end of a rules conflict.

And, oddly some of the stupider "alternatives" people have presented me on these boards over the years provided the perfect example of this sort of fail point.

A social mechanic that basically read "On success target MUST do X... on fail target MUST do opposite of X".

Right there you have an example of a social mechanic that casually left no actually room for any control for the player that "owns" the target character. "There must be a state that player agency exists in" might seem like a frivolous and trivial premise, but if you don't keep it in mind you CAN watch as designers present a mechanic in which the moment you were targeted, not successfully "attacked" or "defeated" or whatever, JUST targeted, by a social rule, you lost all control over your character's action.

The solution is also trivial yet wildly important. Which is that failures to control your character's actions/decisions should not include compulsory instructions that control your character's actions/decisions.

A corrected mechanic could read "On success target MUST do X" and just stops there. If the target WANTS to do x, your failure to rules mechanically force them to do it, shouldn't PREVENT them from doing it. Not to mention that "the Opposite of X" can be in itself constraining even if you didn't want to do X. Or worse, the "attacker" might have actually intended to fail and force you to do the opposite of X anyway.

You can't solve everything with contextual modifiers
I mean, MAYBE if your entire system is nothing but the standard rules lite negotiation mechanic and the ONLY thing it has to work with is contextual modifiers. And I mean your ENTIRE system, not just the social bit.

But anyway. There has been, a lot, of bullshit along the lines of "eh, everything social should just be done with what amounts to a single roll skill check and then you just need to boil every single possible social action and contextual variable into a target number or modifier to look up and apply!". And then they make a bad situation worse by declaring that not only should the list of modifiers be basically infinite, but the value of circumstantial modifiers that might apply to any given action should ALSO potentially be infinite!

The thing is, there are good reasons why basically nothing works like that outside of pure rules lite games, or informal rules lite sections of games that only cover actions/outcomes of minor importance.

An infinite list of infinite value modifiers is whatever the GM wants it to be. It pretty much the definition of "there aren't really any rules at all" arbitrary asspulls that might favor a player, but arguably are more likely to screw over basically everyone at the GMs whim. And also, you know are a bunch of work for the GM since the GMs whim is now basically doing all the work having a rules system in the first place is meant to take over.

Now if arbitrary GM whim on a negotiated roll doesn't do particularly important things... you are pretty much fine. But social mechanics can cover outcomes that are trivial, OR wildly important.

If a GM can pull a number purely out of their ass that FORCES my character to be very slightly humiliated in a social scene in an amusing manner. Eh. If they pull a number out that makes it hard to avoid, Eh, if they pull a roll out that makes it easy to avoid, but possible in the first place, eh.

But if a GM can pull a number purely out of their ass that FORCES, makes it hard to avoid, or even makes it possible that my character's actions or decisions are compromised in an important way... well social outcomes can be VERY important, it can kill your character, it can kill another guy's character, it can cause a Total Party Kill. And with that number coming out of the GMs ass? You can go to hell.

In the end a social mechanic that can do important things and is founded in contextual modifiers of unlimited value basically IS the game mechanical independent declaration "Rocks fall you all die!"

And it's not really any better if the infinite list of infinite modifiers didn't run into it's basic existential and administrative issues. Even if the GM points to a somehow finite list of somehow finite numbers and says "yes this number here, it just screwed you, don't blame me, blame the context". Then the GM can probably still go to hell because the GM is the one who sets the god damn context.

But lets pause and consider a bigger problem with the contextual modifiers rule the social universe approach.

All Hail the King
Wow did people throw a fit over "but what if you social THE KING!" if you social "THE KING!" you win D&D!

And I mean hell. That was always really just an up-welling of subliminal obsession with penis extension NPCs. Right?

Well, probably yes, I mean mostly because it was largely raised as a counter to my proposed mechanics, which were based almost entirely on modifiers sourced from character power levels and not context which just meant "stop having level 1 kings all the time" should have put that dumb angle to bed.

But it was a massive issue for infinite contextual modifier approaches, because inevitably not a one of them was ever firmly based on character level/power. They were in the end almost always based on, or trumped by, a contextual modifier the GM just decided based on their personal feels about how much they like the situation.

When that trumps levels, well, then that means it trumps level appropriate challenges, if it trumps level appropriate challenges it trumps level appropriate rewards. Suddenly wondering whether the players can just "Win" by lying to Elvis is an actual consideration.

And for that matter, wondering whether players would even be allowed to lie to dumb goblins they've already been soundly beating up for 5 levels ALSO becomes an actual consideration that matters. And that is just kinda sad.

Not that rules lite arbitration doesn't have a place in a comprehensive social game system
Because it does. The same place that that sort of arbitration holds for everything else. The bits of trivial, or otherwise questionable, value.

When you get around to having some nice, formal, character power level based social mechanics for whatever social outcomes that matter that your game mechanics manage to represent... there ARE going to be leftovers.

ESPECIALLY left overs of social actions and outcomes that... don't really matter.

People have used the example of "Winning an argument" as a thing elaborate social combat mechanics could resolve.

And they could.

But, outside of additional largely unrelated context, they probably shouldn't.

Making someone believe a false (or even true) fact, an important one only, to ensure they now act accordingly, that is a potentially important action and should probably be treated and resolved as such.

But if the only actual outcome of the game mechanic is if your character is judged to be "winner" of a disagreement?

That's only going to matter in a game where you spend a season doing a debating tournament arc.

Lets not be obsessed with that edge case here.

What is less edge case is that a character might be mildly amused that they got to mildly frustrate some elderly wizard they were arguing with a trick of logic.

Just like they might want to wink at the busty barmaid whose affections really don't matter.

Just like they might want to haggle for a minor discount on an item the price of which is so arbitrary that they could just as easily find one lying around for free at the end of their next fight.

There is no reason to stop players doing these things. There is minimal reason to allow it. There is also no reason not to let them do it on a coin flip or a roll of any arbitrary difficulty or ease.

Even if later on by pure arbitrary GM decisions or actual random or mechanically driven game events it BECOMES important, no one could have known that bar maid would become the demon king. It shouldn't and indeed cannot be accounted for.

So you use the rule you use for everything you don't really need a rule for. "Roll versus whatever I feel like/don't bother rolling it just succeeds/fails because whatever contextual excuse"

This needs mentioning, among other things, because if you present a social mechanic that does SOME things, people will assume that you think it can, or even should do ALL things, and apparently that fairy tea party outside of that will cease to exist. I have seen this happen to me, in a really intransigent way, don't let it happen to you.

Formal rules are for Important Conflict Resolution
So the upshot of a few of these things combined leads to this simple but apparently hard to reach, and for some harder to reconcile with, conclusion.

When things don't matter and important player agency isn't impacted you use no rules or few rules.

When things matter and player agency falls into conflict with other player agency, or simply is threatened in an important way by game events, that's when your more complex, fair, and character power level based rules need to be broken out.

If I want your character to believe or do something (that is physically possible within the game rules etc...) and you are fine with that, your character should just do it with no game rules intervening.

If you aren't sure and it's pretty much a toss up either way, then it's fine to use arbitrary bullshit rules lite junk. Flip an arbitrarily weighted coin just for chuckles. Whatever.

But if you do NOT want the outcome for your character that I want. And we both do not want to budge on that. We use the most formal rules heavy options we have.

Complex formal rules in turn demand valuable important results
OK, so major conflicts over player agency require complex rules. Lets call it social combat, because it probably wants to at least broadly be about as complex and time consuming, and balanced in similar ways relating to character power levels.

And so yeah, you should never break this out if the outcomes aren't important.

You never spend an extended amount of table time resolving a large complex physical combat that at all points wasn't required for survival, doesn't give out loot, and doesn't defeat enemies. That kind of time and complexity investment should give out, well, probably all those things, but definitely some of them.

So when you do break out social mechanics of comparable investment, they should at least defeat enemies and should have the potential of being important for survival, give out loot, and otherwise be a pretty big deal relevant directly to the power of the opponents defeated.

This conveniently ties in with the concern of ending social encounters, or at least ONE of the endings social encounters should have. Because at the barest minimum just like defeating an opponent in physical combat should and does remove them from that combat as an opponent, defeating them in a social combat should remove them from social combat as an opponent.

And OH BOY did people hate that, but, equivalent time and complexity MUST equal equivalent rewards remains to me an unassailable design assumption. And frankly no only have attempts to argue the contrary failed to convince me otherwise, they have failed to convince me they are even engaging with the basic premise in the first place.

Regardless of the usual needless social mechanic panics, on another tangent there are OTHER ways social combat could and should end...

Social Combat needs transitions
A sub set of specific "gotcha" arguments presented against, well, basically having social mechanics, much less ones that achieve outcomes of any note, have somewhat puzzled me.

Because, again, a lot of them seem to think a social mechanic should exist in total isolation. And without that assumption, are basically just descriptions of things functioning as should really just be intended.

Hell one of the biggest angriest demands was that social mechanics MUST exist in total and utterly inviolate isolation to... well basically everything else. That's... a whole other thing...

But one of those gotchas was something like "If a character tries to social my character I will draw my sword and stab them!". Well. It wasn't that coherent, it was more like "I HAVE NOT OTHER CHOICE THEIR AMBIGUOUS SOCIAL INFLUENCE ITSELF IS A WAR CRIME AGAINST MY PEOPLE I WILL KILL THEM ALL EVERY TIME FOR EVER BECAUSE I HATE SOCIALS!".

But, vitriol aside, the strong version of the argument is, "What if circumstantially it seems like the valid response to attempted social influence is to physically attack?"

And really, what if it is? I mean sure, it could happen. You know what it SHOULD happen. OK, it shouldn't ALWAYS happen. SOMETIMES when someone tries to use friendship to influence you in a safe location where people never suddenly fight to the death and no one carries weapons SOMETIMES you shouldn't respond by killing them where they stand.

But in lawless cannibal land when someone tries to cow you into submission with social terror? I mean sure. That social encounter might reasonably end when someone gets stabbed. OR if you social system isn't for stupid reasons utterly and deliberately incompatible with even happening near let alone in physical combat, it could end once someone gets stabbed enough that they are physically defeated and therefore ALSO unable to keep trying to be scary at you.

And that's the thing, the more your social mechanics are at least compatible with your other mechanics the more end states and transitions you can represent.

If you have good compatible movement mechanics a social encounter can end when someone just walks the fuck away. Or runs away. Or sneaks away. Or, yes, resorts to stabbing fools/finishes stabbing fools.

But you can never social during physical combat because it's forbidden
I don't even want to get into the sheer level of incredible bile and hyperbole people tried on with demands that never, ever, should a game mechanic allow a social, any type of social at all with any possible modifiers at all to happen in any describable proximity at all to even an attempted stabbing.

It... it was just bizarre really the very extreme nature of the demands.

And to stop what? The prospect that someone could with any reasonable chance, even a modified reduced chance, attempt friendship, or even terror to defeat at least some opponents... instead of killing them dead with stabbings that their friends were using against other opponents, possibly not even in the same immediate location for some reason?

Worse still the idea that a game mechanic might make it viable, not even optimal, just barely, conceivably, VIABLE, for someone to personally RELENT from stabbing their opponent to death and INSTEAD try and terror them into submission OR friendly them into making up and being friends at the last minute. Wow. I mean, the idea of some sort of merciful victory, at a cost no less, well. Lets just say they called it "War crimes" and demanded everyone always be stabbed all the way to death instead because that was more ethical because that sounds LESS like a hysterical straw man than what they actually said.

So anyway. Having decided the ability to not always murder everyone every time a combat happens might be a good idea, I think you should make your important combat equivalent social options also work in combat time, and in actual combats pretty much as much as you can manage to. It creates the ability to represent far more diverse situations, parallel encounters resolving on the same time scale, characters with different ideas about physical violence still able to contribute alongside conventional ones to defeating threats, and even encounters where everyone present doesn't automatically have to be murder hobo to the stab death every time anyone draws a sword.



...and now I'm going to pretend I decided this post was over large and I'm going to come back with a few other points on this later in an organized way... and I that I DEFINITELY didn't accidentally press submit while doing a preview to see if the title sizes looked OK...

[OgreBattle]

Lets say a dragon has a middling desire to eat the party, the party doesn't want to fight, so the party tries diplomacy.

Some outcomes can be
- Party distracts dragon to flee
- Party negotiates providing a service better than being a meal
- Party convinces dragon to let them leave
- Party lulls dragon into position where they have a fighting chance perhaps

I figure a D&D social system should be able to lead to such outcomes. The 80's "roll to see what they think of you or improve stance towards you based on CHA" did that in one roll if I recall correctly.

---

Grek's Hopes & Fears

I was the one to suggest that, yes, but the description isn't quite how it works. The actual idea was you'd have a periodically rising level of Horror which is resisted by Resolve. At first, Horror grants a bonus to your rolls, as the severity of the situation galvanizes characters to take action, but once Horror exceeds Resolve, it flips to being a penalty instead as your courage fails you. Hopes and Fears modify your effective Resolve for the purposes of taking certain actions - when you get near to the breaking point, you are less able to force yourself to do things that go against your natural inclinations.

You could probably translate that into a decent enough social system - you have Resolve opposed by Pressure instead of Horror, escalating whenever the social situation gets more heated. Easy enough so far. What I'm more hesitant on is Hopes and Fears as your building blocks. That's not quite a direct match - instead I'd go with Appearances and Interests as a replacement. Personal reputations that you wish to maintain (and which could in turn interact with the foot in the door and disguise modules of your social challenges system) and personal goals which your character pursues out of personal interest, either as a hobby or as a moral concern. Beekeeping's an interest, but so is revenge.

The other concern is that unlike with horror, we'd actually like characters to be socially defeated once the Pressure gets too high - we don't want people to declare that they're going to soldier on despite the penalties they're accumulating and keep arguing - we eventually want the entire room to snub them and the butler to escort them from the building. So instead of morale penalties, the rule would be instead that if Pressure is greater than your character's effective Resolve with regards to a particular topic, your character is Overwhelmed and can no longer advance their position socially. They can still talk (to complain to their allies, to challenge someone to a duel, to cast a spell, etc.) but their words will no longer bring them any closer to getting what they want.

You should probably still give the bonuses for having Pressure (as this incentivizes the characters to talk out their problems and get involved in heated dialogue in the first place), but having people actually get knocked out of 'social combat' is important. It also means that exchanging a dramatic speech before the battle is a reasonable thing for the players to do OOCly, which should handily solve any lingering desire for them to get their narratively inappropriate mid-monologue stab on.

[Zinegata]

Lets say a dragon has a middling desire to eat the party, the party doesn't want to fight, so the party tries diplomacy.

Some outcomes can be
- Party distracts dragon to flee
- Party negotiates providing a service better than being a meal
- Party convinces dragon to let them leave
- Party lulls dragon into position where they have a fighting chance perhaps

I figure a D&D social system should be able to lead to such outcomes. The 80's "roll to see what they think of you or improve stance towards you based on CHA" did that in one roll if I recall correctly.

Limiting it to those outcomes is perfectly valid when designing a dungeon crawl, but I can't help but note that missing in this set of outcomes is the issue of motivation.

Why doesn't the party want to fight the dragon in the first place? Power disparity? Conscience? One of the PCs has dragon blood and is worried he might be killing a cousin?

I point it out because social situations in practice are guided by motivations; whereas combat mechanics have traditionally attempted to remove it altogether from the equation.

[Kaelik]

The first problem with social rules is that you want to figure out the genre you are in. The kinds of things you should be able to do and the outputs should be very different.

The second problem is that Kitchen Sink Fantasy like D&D is always a little bit trying to be all the genres, and that means you need a social system with all the outputs, or worse, multiple social systems.

There are some really different things you want to do, and it's very unlikely that one single social system is going to be able to manage all of them.

If I were hypothetically trying to make as many different social systems as I could without hating myself too much, the emblematic use case of each social system would be:

1) Meeting a stranger, deciding if you fight, team up, or pass each other. This also applies to stumbling in on the dragon in his horde.

2) Negotiating your rewards for quests and/or buying shit at the store.

3) Uh Oh, we doing politics now, convincing the king he should definitely abandon the city or stop cutting down the dryad's trees or whatever. This would also be less Kingy where you convince some not actually extremely political figure to do something they wouldn't otherwise do.

4) Still doing politics, Favor Trading, where you are on some council or working for someone on the council and you fucks are all working towards lots of shared and unshared goals and trading off shit.

5) Convincing a crowd to take your side on some issue with a rousing rhetorical speech and/or manipulating public opinion in some other way by spreading rumors.

6) Interrogating/investigating for information, where you either don't want the person to know you are looking for the info or that it is important to you, or where they do know and don't want you to find out and you have to anyway.

You almost certainly can't actually handle all of those things with the same social system without a lot of absolutely terrible mess ups where you get very unintended results or extreme simplification.

Obviously, you don't actually need 6 systems, there are ways to design systems that can plausibly cover two different things, you could imagine a way to fit Favor Trading and Reward negotiations into the same system. You could imagine Favor Trading fitting into the Politics convincing in 3) somehow as well. You could try with a lot of work to fit interrogation/investigation separately onto one or more other systems.

But I absolutely believe, no matter what else is going on, you aren't going to come up with a social system that addresses all 6 situations without fucking up at least two of them. So unless your genre prevents some of these things you are going to need more than one system.

The goals and players in each situation are so different than you are just not going to produce a system that addresses all of them.

Also, "Social Combat" systems where you conceptualize all social interactions as some kind of fight mechanics and attack Social HP can theoretically maybe to do 1, 3, 5, and 6, but also are extremely depressing when used for three and absolutely not what players want even a little bit for 6.

[jt]

Three thoughts:
[*] If you want to make progress, it'd be best to come up with 1-3 complex social encounters suitable for running as oneshots. Try to fit (enough of) your system into 1-2 pages. Run the oneshots, tweak the rules, try again with a different group, repeat.
[*] Status effects and battlefield control abilities might be a fruitful analogy. The players want the king to fund their crazy plan, but he has a level-headed advisor that'll shut them down. The rogue flirts with the advisor while the party makes their pitch. That is, the advisor is employing a "battlefield control" ability and the rogue is incapacitating the advisor with a "status effect."
[*] Most social encounters are the players engaging in short bad faith arguments with people they'll never see again. A single opposed roll is good enough for these. Does the system degrade to this when things are simple? Or is circumventing simple/easy combats with a single roll just a part of your system?

[Foxwarrior]

Complex formal rules in turn demand valuable important results

I feel like a lot of the "negotiating with a being that has a lot of power" social interactions would benefit from a typical success result of "getting a level-appropriate concession". Like, the level 5 party seeks a meeting with the level 12 dragon, does a decent job of conversing with the dragon, and the dragon says "sure, you can have these sweet magic items, I trust you'll go find the secret Sheep of Gul'Thrak for me to eat". Where "the guy(s) become your loyal followers" is a more typical result for the kinds of people you could easily turn into undead minions by force with no loss in quality.

Favor Trading

Quite a lot of social interactions can really be boiled down to some kind of bargaining, you can offer people things like "gold", "not having your family murdered", "a warm fuzzy feeling of helping the wood elves who you have a nostalgic fondness for", "the beatings to stop", "knowledge about your wife's killer", that sort of thing. So you could be trying to convince the person that you can and will provide the things you are offering, and trying to figure out what things the person actually wants. And yes, statting out the entire personality of every character the players might interact with is a bit like using 3.5e character building rules to uniquely build every single monster and minion in the dungeon... but that's why random tables were invented: if it turns out that the party actually wants to seduce the shopkeeper in order to save 5000 gp for some reason, you can randomly determine the shopkeeper's hopes and dreams there on the spot and maybe the ensuing conversation will be funny.

short bad faith arguments with people they'll never see again

I think the bad faith arguments actually get a lot more amusing when they take roughly four rolls with a new bad argument between each one.

[merxa]

I think the risk in formalizing, or increasingly formalizing, social 'combat', is that humans are incredibly social creatures, our social structure and social interaction is as about as complex as possible, any formalized rules will result in situations where people will want different inputs or outputs or both, inputs and outputs the system just doesn't have or cannot approximate close enough. Humans are incredibly inventive, and it seems somewhat of hopeless task trying to formalize such inventiveness.

Besides identifying your system and genre, I would also want to, at least at first, dramatically reduce the scope and function of what social combat can be used for or result in.

Also, any social combat system should encourage roleplaying, not replace it or artificially constrain it.

For any game where social combat isn't the main focus, most encounters should be able to be handled with one or two rolls.

In the examples of trying to convince the King, one thought is the King social stat block should maybe include categories or particulars of things he does or doesn't like, perhaps King NPC secretly hates the common people and disfavors any arguments made on their behalf. This is something a PC could learn in game, but going into the social encounter blindly, if during roleplaying it is brought up, the DM can check that modifier and have it apply (secretly?) to the social encounter roll. Similarly there could be categories or particulars that provide bonuses, or other effects.

Someone as socially complex as a King needs more, such as political constraints or desires, notes on rulership style, risk taking preferences. Political constraints can either be formal, ie as a constitutional monarch there are restrictions on tax changes that can be made without the consent of the nobles council, or more desirous, ie King NPC would never disband his personal guard or standing army because he fears assassination or coup attempts from Baron Evil or whatever, these could be rated as 'impossible' and DCs set to something relatively unachievable unless PCs have domination abilities, or impossible tasks can be tagged to require domination level abilities to even be achievable independent of a impossibly high social roll.

All the additional notes on an npc stat block would be to help DMs adjudicate reasonable compromises, how a given NPC would want to implement it, etc.

The way I would imagine this going on in game, is a social stat block has about 30 words, maybe 150 or more for someone like a King, that provides roleplaying prompts and modifiers for requests, redlines, and something on personality to assist in roleplaying how the NPC might change its mind, how an NPC would respond to 'failed' attempts.

Since this is a game, new stats are likely needed for any social combat system. Things like faction affinity, reputation, fame, perceived power. Even if the King isn't your friend, or even knows you, if you have a great fame, or come as the ambassador to a powerful entity or organization, the King will presumably weigh your words differently and even respond differently on failures and successes.

[Kaelik]

Favor Trading

Quite a lot of social interactions can really be boiled down to some kind of bargaining, you can offer people things like "gold", "not having your family murdered", "a warm fuzzy feeling of helping the wood elves who you have a nostalgic fondness for", "the beatings to stop", "knowledge about your wife's killer", that sort of thing. So you could be trying to convince the person that you can and will provide the things you are offering, and trying to figure out what things the person actually wants. And yes, statting out the entire personality of every character the players might interact with is a bit like using 3.5e character building rules to uniquely build every single monster and minion in the dungeon... but that's why random tables were invented: if it turns out that the party actually wants to seduce the shopkeeper in order to save 5000 gp for some reason, you can randomly determine the shopkeeper's hopes and dreams there on the spot and maybe the ensuing conversation will be funny.

I think if you build a Favor Trading Social System that tries to cover as much as possible of the social space it could cover Interrogations but not Investigation, the Favor Trading in 4, and the Convincing in 3.

I'm not sure a favor trading system could even work that produced outputs for negotiating quest rewards and sale prices, but maybe if you turned the entire system into weighing sides on a scale it could also handle the quest rewards/buying/selling outputs of 2. That might make it fail in 4 and 3 though.

But Favor Trading as a comprehensive social mechanic would completely fail to address "Run into new person, what happens" social situations, crowd/public opinion management social situations, and investigation.

[Foxwarrior]

But Favor Trading as a comprehensive social mechanic would completely fail to address "Run into new person, what happens" social situations, crowd/public opinion management social situations, and investigation.

"Run into a new person, what happens": you're trying to convince the new person to just let you go on your way without a fight, so the favor you're trading is "I won't stab you in the back as you walk past me, or pillage your village that you just let me into", and you have to convince them that you'll follow through by presenting the kind of appearance they trust.

crowd/public opinion management: if you're trying to get a crowd to riot, you have to convince them that rioting will help them accomplish their goals, and that the status quo has less to offer them. If you're trying to get them to stop rioting, isn't "can't we all just get along" a type of peace and contentment favor you're trying to offer? You have to present yourself as the sort of person that would negotiate with the king on their behalf to get them food so they don't need to overthrow the government right now after all, or whatever.

investigation: well, yeah, I guess getting someone to trip up over theirselves and reveal something they didn't mean to on accident doesn't really fit at all. Except incidentally in the way that you have to convince them that it's worthwhile to keep talking instead of just shutting up.

[Emerald]

I think the risk in formalizing, or increasingly formalizing, social 'combat', is that humans are incredibly social creatures, our social structure and social interaction is as about as complex as possible, any formalized rules will result in situations where people will want different inputs or outputs or both, inputs and outputs the system just doesn't have or cannot approximate close enough.

Yeah, this is basically the Realistic Fighter Problem writ large. Most players have a vague mental image of what a martial character can "realistically" do and how combat "realistically" takes place (more Hollywood than reality, of course, but they call it realistic), so a bit of abstraction and exaggeration in the combat rules is fine but a lot of people will balk if the rules exceed an arbitrary realism threshold.

But all players have a quite detailed mental image of how social interaction works because they do it themselves all the time, so to reach the same plausibility threshold as the combat rules you have to either get much more complex and detailed or leave things incredibly vague to be filled in by players, both of which approaches have a lot of issues.

In the examples of trying to convince the King, one thought is the King social stat block should maybe include categories or particulars of things he does or doesn't like, perhaps King NPC secretly hates the common people and disfavors any arguments made on their behalf. This is something a PC could learn in game, but going into the social encounter blindly, if during roleplaying it is brought up, the DM can check that modifier and have it apply (secretly?) to the social encounter roll. Similarly there could be categories or particulars that provide bonuses, or other effects.
[...]
The way I would imagine this going on in game, is a social stat block has about 30 words, maybe 150 or more for someone like a King, that provides roleplaying prompts and modifiers for requests, redlines, and something on personality to assist in roleplaying how the NPC might change its mind, how an NPC would respond to 'failed' attempts.

Since this is a game, new stats are likely needed for any social combat system. Things like faction affinity, reputation, fame, perceived power. Even if the King isn't your friend, or even knows you, if you have a great fame, or come as the ambassador to a powerful entity or organization, the King will presumably weigh your words differently and even respond differently on failures and successes.

This basically sounds like Fate. Every character in Fate has three things that define them: Aspects (various aspects of their personality), Skills (things you roll to do things), and Stunts (situational modifiers to skills or ways to use skills differently).

Aspects can be invoked (used for their own benefit, e.g. "I'm known as a Righteous Paladin so you can be sure I wouldn't lie to you") or compelled (used against them, e.g. "The king is Unreasonably Paranoid, so he'll definitely believe his advisor is plotting against him"), and you can take actions to discover what Aspects another character has. Characters can have any number of Aspects, each of which is a word or phrase serving as a roleplaying prompt as well, and the more Aspects you can tap (invoking your own or compelling a target's) the higher DCs you can reach.

Skills can be capabilities like "Jump +2" or "Spot +3" like in D&D, but they can also be anything else that can be rated numerically, whatever the GM feels is appropriate for the setting. Fate Accelerated uses "Approaches" like "Sneakily +3" or "Cunningly -1", I've seen some Fate hacks use "Affiliations" or "Backgrounds" like "Guild Member +2" or "Cultist +1", and so on.

Stunts are usually of the form "do this thing X/session" or "do X instead of Y when Z," and can be used to represent the more fuzzy social mechanics like calling in favors, being able to demand an audience, and similar.

Put all that together, and you can basically layer Fate on top of another game as a quick-'n'-dirty social system to figure out what scenarios work well and what you want to emphasize, which wouldn't be a terrible idea as a first draft of whatever system you want to work on.

investigation: well, yeah, I guess getting someone to trip up over theirselves and reveal something they didn't mean to on accident doesn't really fit at all. Except incidentally in the way that you have to convince them that it's worthwhile to keep talking instead of just shutting up.

I think you can actually make Favor Trading work for Investigation if you reframe things right. Favor trading is basically a matter of two sides setting the stakes of what they want, determining what they're willing to offer/expend/etc. to get that, and then coming up with a bargain that works for both sides, right?

Well, investigation kind of works the same way, except the opposing party is "the mystery" instead of "the dude you want to convince" and the resources you're working with are things like time/contacts/investigative skill instead of reputation/gold/negotiating skill.

Imagine a scenario where the party is in a haunted house and they're trying to figure out how to put the ghost in it to rest before it kills someone else. You can frame that in terms of an investigation between the party and the house where the party is wagering "time until the ghost attacks again" and "some scrolls of divination spells" while the house is wagering "clues about the ghost" and "not getting discovered." Then it comes down to the "bargaining," where the party debates whether they want to take longer or use more resources--or go for broke and spend both time and scrolls--and can force the house to "concede" clues or stealth to them as the investigation progresses.

It's obviously not a perfect one-to-one mapping, but I think the basic framework is pretty flexible so long as you're not too specific about how resources and modifiers are defined.

[PhoneLobster]

OK so there were a few more bits I wanted to cover.

The power of social influence
A very, very, important aspect of a social mechanic is understanding the potential power and impact of social influence.

If you can make someone believe a thing is true, if you can make them like and trust you as a friend, if they fear you more than death, what can that let you achieve? What VALUE does that have in game terms?

We need to know, the impact and value will decide which sort of rules will apply. When it's nothing much we can get away with informal or semi-formal rules. When it's big deal life and death and personal possessions stuff it's going to be combat equivalent formal rules.

Well. The problem is it's going to require some judgements in play, but in the end, social influence is wildly powerful. The right lie KILLS YOU. Trust in the wrong friend leaves you immensely vulnerable to assassination or robbery.

You need to admit this and understand it. Because people will otherwise leverage your system in bad exploitative ways.

"It's just ONE LITTLE LIE" they will say "It should be quick and EASY to convince them to do this ONE MINOR THING!" they will say... as they carefully plot the one little lie, the one minor thing that gives them everything they ever wanted from the target character including the shirt off their back and their life itself.

The exact divide where you need to demand that no, it's not JUST one little lie, time to break out the formal rules based on character power levels and NOT give out a "just a minor lie" semi-formal/informal easy ride is always going to be a bit fuzzy but it's OK the divide itself can and should be arbitrary. In the end it's a divide that happens when one player says "but I want your character to believe this" and you say "but I don't want them to" hard enough that you want to use the rules that matter to make arbitrate the conflict in a fair way. The actual thing being fought over doesn't matter, it's the fact that you decide to fight over it that makes it officially worth fighting over.

So, since there is no hard requirement to trigger the important rules for important results, be on the careful side, if it looks like any attempt to gain significant leverage, demand a formal conflict, it's what the rules are for and if you instead let them do an easy semi formal single roll to get everything they ever wanted from the Emperor of All Money you've essentially rendered your real social rules redundant and possibly undermined the rest of your game in the process. Play it safe instead.

Players get to trigger the important rules too
It's not fair for only the GM to decide when the real rules kick in, and when it's just a little lie about nothing that uses the fake easy rules.

The players need a say too. If their characters can be a target of social, and they damn well should then sure they CAN be convinced with no rules at all... and that WILL require convincing the player themselves and that is fine.

And sure the player might also be fine with a semi-formal single roll with bullshit modifiers based on the feels. They might even inform you of the feels. "I think my character would be reluctant to believe that" and the GM says "OK, how about we make it a pretty difficult roll" and they say "sure whatever". That's OK.

But it's also OK for a player to just say "No I don't want to, no rules free convincing me, no bullshit roll you decide the DC of, if you want to influence my character this way defeat them in social combat or get the fuck out".

That is fine. That's what it's for. Giving the player a fair chance based on their character power level to say no to ANY social influence they arbitrarily declare as important.

Because the "it's just a minor lie" thing still applies, the GM might be getting everything they ever wanted from that character, they might be compromising them in deeply dangerous ways.

But more than that, the GM can't know and can't measure what the player deeply personally cares about. I mean you should TRY and in some cases shouldn't even resort to the formal rules to do something the player utterly despises. But in the mean time the player should AT LEAST have the power to say "no I don't like this, a demand trial by combat" as a bare minimum protection from things they don't want to happen.

And another thing!
Your social system cannot cover EVERYTHING. It can cover a damn lot of things. It can especially do so if you use a few different methods like my informal/semi formal/formal structure. But it can't do EVERYTHING.

But, you are also going to see some really bullshit demands leveled towards it, and I think a some good examples are listed here already.

Kaeliks list and the one about influencing the dragon include some entries that are vague enough that I guarantee you someone wanting to pour bullshit on well, ANY social system WILL no-true-scottsman into not fitting into any of your mechanics.

I'm going to specifically target "Demanding the king change just one policy" and "Convincing the dragon to let you leave".

Because these are things which someone could, if they were determined to make trouble, try to jam through the gaps in my methodology.

Because in the end both are "It's just one MINOR thing!" while simultaneously being the one and only thing that the characters, apparently, want from their targets.

You COULD deal with both as either a semi-formal single roll, or by demanding full social combat and defeat of the dragon/king itself. "It's not that big a deal" is not unreasonable to put it into the semi-formal category. The "well actually come to think of it it's a pretty big deal and dragon-king says no" is also pretty reasonable to put it into the formal category.

So it isn't hard for a stubborn bastard, and when discussing design of a rules set the entire genre of which some people oppose with a life long hatred and the rest of which fall into endless factional bloodshed, to leverage that and demand that no, actually "the one minor but also important thing" doesn't fit into either category thanks, they demand ANOTHER ONE! ... or more likely that they just demand that social mechanics go away entirely.

Well. Too bad. Your social mechanics can't cover absolutely everything perfrectly. You try to reduce gaps, you try to get close enough, and sometimes you may need to nudge some things into categorys where they just kinda fit. That's fine. In fact the dragon-king and the just one thing scenario, it's a challenge at the design discussion, but at the table just picking a category and wedging it in is easy as all hell. You decide, you tell the players what mechanic is applying today and you start rolling, it works fine, disputes seem to be minimal, close enough IS in fact good enough.

I mean, OR you could have made another mechanic to cover another category of thing that fits better. But... something that fits better than resorting to the semi-formal mechanic? Something more flexible than the fully informal mechanic? Something more fair than the formal combat mechanic?

How edge case is the scenario? How much rules complexity is it going to cost. In SOME cases, absolutely yes, make a dedicated mechanic, especially if it is a highly definable single thing to achieve that is broadly applicable and can be somehow represented by a single action, but if it needs it's own additional minigame, and entire new substructure of mechanics? Maybe, even probably not a good idea.

Maybe don't try and do favors and trading
A major just one thing... category people seem to want is social favor currency and "bargaining".

They are basically giant pitfalls to cover some edge case categories you can wedge into other more generic mechanics with better results than creating entire custom sub systems.

Both basically are just variants of the "infinite lists of infinite modifiers" thing with all the problems inherit and more besides. Representing the values of favors or bargains is a never ending ever growing list of bullshit modifiers you probably got wrong that can never be subjectively or objectively correct and almost inherently divorce game play results from character power level progression and result in all sorts of unfair outcomes for everyone.

But even aside from that... do you NEED a bargaining system?

I mean you can just bargain with the player controlling the character without rules interference. "I will give you this sword for this hammer" and they say OK or not. The flexibility and fairness is... hard to beat. And if they don't agree or think it's unfair and you really want to make them do it, you can escalate all the way up to the formal rules and make them do the one thing you wanted from them that was so important you demanded to over-ride their player agency.

Alternatively do you really think you can make a FAIR evaluation of "the dragon lets you leave" such that it's a hard formal modifier/target number/whatever in a hard formal system for hard form bargaining that has non-crazy outcomes for game play? Letting you leave is such a minor thing, it avoids a fight you avoid a fight right? Or is it the most important thing in the universe because you know where it sleeps now? What if it doesn't realize the second thing? What if YOU don't but it does? and even if you do I mean, both things are correct, making it super hard because its a big deal to them, but only letting you achieve the equivalent of no-encounter on success is... not really great and I DO hope you aren't spending any real time or rules complexity on literally achieving nothing...

And as for favors. I mean, lets just start with this. If you want to do a formal favor currency and you think you can do better than completely informal rules free "yeah seems like this guy would feel like he owes you one". Lets ask, before you try to create the worlds most complex currency system that needs to also account for information spread, deception, and infinite subjective view points and attitudes... do you even have a particularly function REGULAR currency system?

When you think about it is you regular currency system kinda just a bit of a kludge that doesn't account for a lot of things real currency does, and actually your players kinda just find free items sometimes and it's all a bit of a mess?

Because if so... maybe... maybe you aren't going to make another, much more complex currency system...

Maybe, if you can't say for sure how much regular currency an item actually cost someone then determining how much favor currency it will cost, a significantly more complex task, is a fantasy.


Anyway this time I'm not pressing submit by accident but I do need to come back with at least one more major point I really want to mention.

[The Adventurer's Almanac]

Jesus, every time I write something up, somebody else makes a post.

And to stop what? The prospect that someone could with any reasonable chance, even a modified reduced chance, attempt friendship, or even terror to defeat at least some opponents... instead of killing them dead with stabbings that their friends were using against other opponents, possibly not even in the same immediate location for some reason?

Worse still the idea that a game mechanic might make it viable, not even optimal, just barely, conceivably, VIABLE, for someone to personally RELENT from stabbing their opponent to death and INSTEAD try and terror them into submission OR friendly them into making up and being friends at the last minute. Wow. I mean, the idea of some sort of merciful victory, at a cost no less, well. Lets just say they called it "War crimes" and demanded everyone always be stabbed all the way to death instead because that was more ethical because that sounds LESS like a hysterical straw man than what they actually said.

You know, one of my preferred tactics in OSR games is to start off with a massive, bloody display of force after telling my enemies they shouldn't fuck with me, and then after beating the first few to a bloody pulp with a warhammer I tell them that if the rest don't leave, I will slaughter them all. This genuinely results in less blood and death than the alternative, and it's quicker than fighting them, too. With no XP loss, either!
I'm a great Paladin.

Anyway, I remember some of those arguments getting hilariously vitriolic back in the day. Lots of rape accusations and paragraphs of shitflinging. What a time to be alive. I'm pretty much on the same boat as you are when it comes to... "social combat". I don't see anything wrong in trying to make a quick Charm or Intimidate check to get some of the weaker enemies you're fighting to chill out or run away.

You almost certainly can't actually handle all of those things with the same social system without a lot of absolutely terrible mess ups where you get very unintended results or extreme simplification.

Obviously, you don't actually need 6 systems, there are ways to design systems that can plausibly cover two different things, you could imagine a way to fit Favor Trading and Reward negotiations into the same system. You could imagine Favor Trading fitting into the Politics convincing in 3) somehow as well. You could try with a lot of work to fit interrogation/investigation separately onto one or more other systems.

But I absolutely believe, no matter what else is going on, you aren't going to come up with a social system that addresses all 6 situations without fucking up at least two of them. So unless your genre prevents some of these things you are going to need more than one system.

1 sounds to me like it could be handled with some sort of short-term attitude system that's been around for like... 40 years. Quick and dirty, but it gets the job done for determine how much each side wants to stab the other. 2, 4, and 5 sound like they could be combined into the same thing, since they all boil down to "convince the other person/people to do or think something you want them to". That's basically what 3 is too, just at a higher level with more complexity. That just leaves 6, which is totally something that we want adventurers to be able to do and interact with other aspects of... "social combat".
Perhaps I'm overlooking something, but it sounds like all of that can be done with a minimum of 3 systems, which doesn't sound too horrible. Maybe I'm expecting something more simple than other people, but I'm thinking about this from the context of a game where people sit around, drink beer, and bend the world to their wills through pulp-esque adventures.

[Foxwarrior]

PhoneLobster, I think you're taking the "One Minor Thing is possibly everything the other party wants" realization the wrong way... I think long ago the argument was something like: "'here, hold this briefcase (which contains a bomb you don't know about)' is equivalent to 'let me kill you' because it has the same result, therefore if you win social combat hard enough to get them to hold your briefcase you should be allowed to get them to commit suicide instead." And I think that's fair but not good... I genuinely want social conflicts to result in more "here hold this briefcase" endings than "let me kill you" endings, because that feels more social conflict-y.

[owlassociate]

I'm wondering if something like the escalation mechanic from dogs in the vineyard would be applicable to a social combat system for pulp fantasy adventures

[PhoneLobster]

"'here, hold this briefcase (which contains a bomb you don't know about)' is equivalent to 'let me kill you' because it has the same result, therefore if you win social combat hard enough to get them to hold your briefcase you should be allowed to get them to commit suicide instead.

It's not a "should be" it's an "you already are".

It isn't a demand that you should be allowed to tell them "kill yourself" and they do.

It is an acknowledgement that telling them "hold this bomb for me" IS telling them to kill themselves and then they do.

The only difference between the two is flavor text. And while you can make them distinctly mechanically separate abilities based on differing flavor text alone, so that the ability that lets you make people hold suspicious ticking packages isn't the same one that lets you demand they die for no reason...

...the "should be" conclusion you should be coming to here is that the two abilities need to be of basically the same cost and difficulty if they can have the same outcome.

And again, having the same outcome is the pre-existing fact you are acknowledging. Not a thing I'm making up or deciding, not a rule I'm proposing a simple fact you need to acknowledge before you start making rules that need to deal with that being a fact.

Forcing a character to believe a lie can kill them. So forcing them to believe a lie that the player owning them doesn't want them to believe needs to be costed accordingly as a potentially fatal ability.

It's not a complex line of reasoning and it's founded in indisputable fact.

[Orca]

I sort of think it should be easier to convince someone to hold a briefcase for a minute than to hold Wile E. Coyote's bundled sticks of dynamite with fizzing fuses for the same time. Because one is an obvious threat and one isn't. Consequences aren't the only metric, ease of getting there is another.

OK, no one other than PL needs to be told that, but I think it needs to be said here.

[Foxwarrior]

...the "should be" conclusion you should be coming to here is that the two abilities need to be of basically the same cost and difficulty if they can have the same outcome.

I think this is where we diverge. I agree that convincing someone to hold your briefcase should be in the ballpark of as difficult as stabbing them to death, but I don't think it follows that convincing someone to hold your fizzing Wile E. Coyote dynamite should be only as hard as convincing them to hold your briefcase. Strategies that are "in theme" for the setting and characters should be more effective than strategies that are out of theme, or your game will become a disjointed mess.

I guess I don't have the same objection to a guy who's specialized in being depressing being able to convince people that commit suicide as easily as a guy who's equally specialized in lying convinces people to hold his briefcase; it's really just the unspecialized comparison where I want lying about the briefcase to be the better strategy. Maybe this is more a disagreement about general RPG design than Social Combat specifically...

[PhoneLobster]

Consequences aren't the only metric, ease of getting there is another.

Well goes to show how that went right over your head then didn't it.

The entire point is that "consequences" ARE the only metric that you can actually hope to work with.

"The ease of getting there" (you were looking for "subjective context" or rather your subjective valuation of the also subjective context) is barely functional rarely and exclusively in the fairy tea party heavily informal part of your rules system. You know, the bit that basically isn't rules at all.

"I think it feels easier" is not an excuse to break your entire game, kill the player characters without allowing them a fair chance and destroy the entire concept of level appropriate challenges existing at all within your system. I don't care if you put the word "social" on the front of it. That principle, does not change.

It's a basic design principle thing. You have to put the function of the game first, ahead of your silly and fickle feels about the realurzumzs.

[MGuy]

I agree with a lot of what's being said here. Especially the idea that you have to make decisions about what you're going to shoot for with your social encounter system. I also agree that different scenarios might require different mechanics to handle. In my mind and notes I use multiple schemes to help simulate different social encounters and I think that's the right way to go.

As I outlined above I don't really want to cover low impact bargaining scenes. I think it's better just to allow people who are interested in getting a bit more out of their monetary interactions just to get a modifier for it and move on. I also am leaning toward using abstract wealth so there wouldn't be as much need to worry about trading swords for food as you'd just turn your treasure into pocket change. That way, by using another subsystem, I can avoid having to make new rules to cover that social situation.

There are other issues that I think are easier to handle by just letting the GMs and players decide on things. This isn't me saying that we should throw out frameworks arbitrarily but there should be room in a framework to just let people decide on things. My go to example is bluff because by the nature of what it is and what it's supposed to do context HAS to be king in implementing it. When you're attempting to lie to someone there are a lot of factors that have to be accounted for between you and the person you're lying to. Any system that has this in it will have to essentially let the GM, on the fly or not, decide on a lot of things about any given NPC who is lying or being lied to. What they know, how willing they are to believe the lie, how much they trust the person, etc. Sure you should probably have guidelines and modifiers to make certain considerations important but you're largely depending on the GM to make up a lot of stuff in order to get an idea of how difficult this is to pull off. When designing around it you're going to be dictating how much you personally think every consideration you write in should be weighed.

I think this is fine. You can be more or less heavy handed in this part as you want to be as a designer and the less you decide to weigh in the more you're just leaving it to the GM to decide what they think. Since we are social creatures who socialize every day I think that it is just fine to allow this to happen. I also think this can be applied broadly to make non issues out of things people are worried about.

For instance the first thing on kaelik's list is basically what happens when the party meets X. My answer to this with what little outline I spoke about before would be "I don't know. Depends on what the GM and/or players want to happen". If players stumble in on a dragon guarding their hoard I am fine with letting the GM figure out if the dragon wants to be hostile and if so what that means to them. If the dragon decides anything it doesn't like that nears its hoard should be killed on sight I'm fine with that being adjudicated without a die roll. If the players want to bargain with it then I think options to parley after hostilities have started should be available. Not just with a dragon like this but in general because I can think of a lot of situations where such a thing would be appropriate.

If instead the players meet some unscrupulous seeming people on the road that aren't immediately hostile then the players get to decide what their approach is going to be. I don't think it's a big ask to allow the GM to determine some general characteristics about random people players meet on the fly. I agree with

but that's why random tables were invented: if it turns out that the party actually wants to seduce the shopkeeper in order to save 5000 gp for some reason, you can randomly determine the shopkeeper's hopes and dreams there on the spot and maybe the ensuing conversation will be funny.

and I am pretty sure I said it myself multiple times during multiple diplo arguments that have raged here. I am ok with letting the people running and enjoying the game decide these kind of things and I have still not heard an argument that's convinced me that rolling for first meetings is good/useful.

I agree with a lot of what PL has said in regards to this but I do not think that special rules need to be made for this brief case scenario. I 'get' the reasoning. That the actual results of tricking someone with a thing that can be fatal is significant and I get the desire to make obstacles to tricking someone into such a scenario. This 'is' a game after all and the results matter. However I think there are less heavy handed ways to do it. I'm speaking assuming that this is an issue specifically with higher level or equal level challenges. I also am going to point out that another issue not being pointed out is keeping things 'like' the briefcase bomb' from being an easy lever to pull in any/and all relevant scenarios. After all if exploding runes 'just work' you're encouraging your players to spam it.

So in this bomb scenario you are probably looking to kill someone specifically or otherwise sneak a dangerous weapon into an area in an inconspicuous way. In this case, via brief case. This is a scenario that has been both in fictional media and a huge worry in real life. I don't think that putting a bomb in a suit case to sneak past guards should warrant a bigger check than sneaking contraband food past those same guards. At least not inherently. I think the decider should be what the defensive measures the guards are taking in that area. If bombs are a known threat for anyone important then they have likely taken measures to secure places against such a thing and getting your bomb past the safeguards is likely more complicated than just putting it in a briefcase. This is assuming a situation where the bomb is a threat so probably a system with very few differences between levels or a level-less system. In a leveled system the bomb is probably not more threatening than getting an early fireball off in the surprise round and so wouldn't be an issue I'd be all that up in arms about. If abilities are staying level appropriate you are likely not one-shotting anyone who is at your level or higher with that stunt.