Here's a game idea I've been toying with for a while, and would like some feedback on:

You play the captain of a mercenary squad in a fantasy setting. You keep 15 or so characters hired, sign contracts, fight battles, etc. One of the features is that your mercenaries have their own opinions about each other. Close friends will support each other: getting bonuses in combat when close together. Rivals will fail to cooperate well, and in the extreme even sabotage each other in battle. There'd be out of battle penalties as well: injuries from brawls, ostracized warriors quitting, etc.

The mechanism I was envisioning was that weaker mercenaries, the fodder, would gravitate into cliques around the powerful mercenaries, and slowly assume the same network. That said, if you keep them on good terms with multiple cliques, it might draw them together. I was picturing a graph representation, where you have your team spread out on a plane, with green and red lines showing friendships and animosity, with thickness representing strength. Perhaps the player can click on an edge to try to improve the relationship.

Keeping mercenaries close together in combat would tend to strengthen the bond. I was also considering using distribution of loot after combat as a mechanism here: the idea being that you give powerful equipment as a reward for good work in combat. If you show favoritism towards an individual, the weaker recruits would tend to center around them and rivals would tend to feel antagonized. The idea was also to give an interesting trade-off between giving the best equipment to the right character, and keeping your men happy by giving gear that isn't quite right to the popular hero of the last battle.

Does this sound like an interesting mechanism or a distraction? Would you enjoy having to balance keeping on a strong mercenary or firing him to promote social cohesion, or does that sound like a frustration getting in the way of fielding your awesome army? Any opinions on mechanics or interfaces that could make up or play well with this?

Does this sound like an interesting mechanism or a distraction?

One of my most cherished gaming memories occurred years ago playing Baldurs Gate 2, when I accidentally caused Minsc to entirely abandon my party in the first 20 minutes of the game, by choosing the wrong dialogue options.

Which is to say, I think managing the social and emotional wellbeing of you team seems like a great game mechanic. Watered-down variants of this exist in many modern RPGs (i.e. relationships in Dragon Age: Origins or Mass Effect), but I'd love to see a game that explores these social dynamics in more depth.

I think that sounds enjoyable - I wouldn't want it slapped into just any game, but I wouldn't mind a game series that focuses on it and is designed around it.

It kinda reminds me loosely of Fire Emblem allowing you to partner up two combatants to benefit each other.

It also gives a vehicle to explore non-romantic character relationships, like the movie Saving Private Ryan did a little, and like the TV series Band of Brothers did to an extent.

Exploring how the characters are drawn together by adversity, how real danger can break through differences like gender/race/politics/religion even while those differences are still present, and how even amidst military brotherhood relationships aren't perfect and antagonism can still exist, and how the death of a 'brother at arms' can affect the moral of the group and re-arrange the network.

And then on top of the platonic relationships you can add a few romantic relationships and play with how that affects the dynamics of the group as well.

...Which is to say, I think managing the social and emotional wellbeing of you team seems like a great game mechanic. Watered-down variants of this exist in many modern RPGs (i.e. relationships in Dragon Age: Origins or Mass Effect), but I'd love to see a game that explores these social dynamics in more depth.

I really like emotional well being as a pairing. Combat bonuses for friends are ok, but a little dry, and I worry about battles becoming babysitting affairs: ok, got to keep Jack and Jill on opposite sides of the battle... But I like the idea that in the depth of a dank cavern, hunting some unknown monstrosity, the social fractures you brought with you just snap, and the team flees for the surface or falls sobbing to the ground, while a stronger social bond allows the group to rise to the occasion.

...Exploring how the characters are drawn together by adversity, how real danger can break through differences like gender/race/politics/religion even while those differences are still present, and how even amidst military brotherhood relationships aren't perfect and antagonism can still exist, and how the death of a 'brother at arms' can affect the moral of the group and re-arrange the network...

All good ideas. I had been thinking a fantasy setting, so there's always the old 'dwarves and elves hate each other' trope to discuss racism at arm's length... Romantic relationships also sound like a great way to inject drama. Presumably a veteran player would keep the team at even keel, so a nice messy breakup sounds useful.

This also has me thinking of the bumbling but beloved soldier: including characters who can ease tensions but don't hold their own weight in combat. Or the egotistical, but genuinely talented hot shot who everybody kind of hates but who also saves their lives (and won't let them forget...)

I'd hoped to do this procedurally (In a lot of ways, I'd have to to react to the changing social dynamics), I'll have to give a lot of thought to how best to keep this dynamic without it becoming dry and mechanical: the more we discuss, the more fertile I realize this narrative ground is.

One of my most cherished gaming memories occurred years ago playing Baldurs Gate 2, when I accidentally caused Minsc to entirely abandon my party in the first 20 minutes of the game,

That's one HELL of a way to teach a player that dialogues matter! :P

Does this sound like an interesting mechanism or a distraction? Would you enjoy having to balance keeping on a strong mercenary or firing him to promote social cohesion, or does that sound like a frustration getting in the way of fielding your awesome army? Any opinions on mechanics or interfaces that could make up or play well with this?

It's a distraction if this isn't the core of your game. It isn't if it is the core of your game. Basically, it's the kind of feature that can make your game stand out, or confuse the hell out of your players depending on how much design space you dedicate to that as opposed to other features.

Depending on the actual mechanics for party management (how you acquire mercenaries, choose which ones are currently in the team, etc.) this could work well.

I'd be interested in seeing if you can get a team of lone wolves to work well together through clever tactics though (aka, break the game).

Would the interactions be hard or soft? Aka, would stats be directly impacted, or the impact would be subtle?

I really like the idea; I think that, as a core element of a game's design, this could make for some really interesting gameplay.

As for mechanics that might play well with this, I might suggest external stressors and temptations: monsters that horrify your troops, causing those without strong bonds to break and run; seducers that threaten romantic bonds, perhaps even turning them sour; various brands of insanity, that might varyingly affect how others see those afflicted; and so on.

I'd really like to see a prototype of this, and find out how it plays out in practice...

If you avoid the bonus mechanic (I get +5% because I am happy!) I'd have it a little more cleverly structures where the game detects the various stress levels of the combatants and modifies their possible actions accordingly.

That guy you don't like under heavy fire? Turns out you're just not going to save him. Is your character afraid? Maybe he's going to run for it.

It's a fine line to walk in game design where you get agency (you get to pick what action your characters take) and where you enforce realism (Nobody's jumping out of cover in heavy MG fire to save someone they'd shoot in the back anyway). I believe the key here is to avoid the "you lose" scenario, and simply make every scenario a situation where losses are furthering the narrative and relationships. Meaning even if you botch the mission the game goes on and you don't feel like you lost the war so much as a battle.

To top that off you need to find a way to allow the user to tell a story based on what happened. Read some good After Action Reports on emergent games - especially strategy games - to get the feel of what I am suggesting. In short, if the player can tell their friends "So we had the enemy pinned down when an enemy fires off a freakin rocket right in the middle of all my guys. Nobody takes any direct damage but they all scatter like the freakin wind because once one panicked the next one's morale dropped low enough to panic also and they fled like dominos. Now I have a whole squad of guys terrified of fire! Like literally have all gained the pyrophobia trait.

that kind of thing :D

Another interesting vein of drama: how characters react to changes in power. Susan is usually marginalized, but she was really on in that last fight and saved the day. Now she's the center of attention. Does she try to repair bridges, or does she antagonize the platoon back? How about Fred, who was the star of the team for years, but is slowing down from all his injuries. Is he bitter, or will he step back from the limelight and help the new stars achieve? Do they appreciate the help or blow him off?

It's a distraction if this isn't the core of your game. It isn't if it is the core of your game. Basically, it's the kind of feature that can make your game stand out, or confuse the hell out of your players depending on how much design space you dedicate to that as opposed to other features.

What I like is that it ties different elements together. You make decisions (hiring, rewards, conversations, etc.) outside of combat and then you suffer/benefit from the outcome during combat. And the story that unfolds during combat changes these relationships, which then changes your opportunities/costs back outside of combat. Effect loops feel like good design to me.

What I worry is that it's bandaging two systems together: that whatever mechanic is used for combat, some players will just want to do the fighting and some will just want to do the social bits and each will feel like the other piece is a detraction.

Would the interactions be hard or soft? Aka, would stats be directly impacted, or the impact would be subtle?

If you avoid the bonus mechanic (I get +5% because I am happy!) I'd have it a little more cleverly structures where the game detects the various stress levels of the combatants and modifies their possible actions accordingly.

That guy you don't like under heavy fire? Turns out you're just not going to save him. Is your character afraid? Maybe he's going to run for it.

Yeah, I feel the paired combat mechanic is going to be a very important choice. The easiest seems to be a classic strategy game with bonuses/penalties. That unit likes his platoon, so he's got a big bonus to avoid being routed. Her lover died, depending on mood she breaks down crying for a penalty or goes berzerk for a bonus. Add some flavor text "What? Big mouth can't save himself?" to convey internal state. But basically you tell them what to do and they do it.

Somehow the more fitting mechanism seems to be giving the mercenaries autonomy. Let them decide whether to run over and drag the wounded comrade back. A best friend dies, maybe they just charge the enemy. And if they've got no other friends, nobody comes along to back them up. You give the tactical orders, the AI executes them to their personality. However, that mechanism scares me in terms of getting it right =). How to give them autonomy, but not feel like you're herding cats. How to leave the details up to the AI, but the interesting decisions to the player. It does open up a lot more opportunities for dynamic and non stat impacts from the network though...

To top that off you need to find a way to allow the user to tell a story based on what happened. Read some good After Action Reports on emergent games - especially strategy games - to get the feel of what I am suggesting. In short, if the player can tell their friends "So we had the enemy pinned down when an enemy fires off a freakin rocket right in the middle of all my guys. Nobody takes any direct damage but they all scatter like the freakin wind because once one panicked the next one's morale dropped low enough to panic also and they fled like dominos. Now I have a whole squad of guys terrified of fire! Like literally have all gained the pyrophobia trait.

that kind of thing

Ha, I like that a lot. And suddenly you have to start turning down lucrative contracts because fire is likely to be involved.

I gave the combat mechanic a ponder, and here's what I'm currently thinking:

My initial impulse was to do a classic turn-based strategy deal, with rock-paper-scissors advantages between unit types, terrain bonuses, attempts to flank or break through lines; that sort of thing.

But that makes the combat narrative (who needs help from who) kind of opaque, and didn't exactly mesh with the relationship network. Turn based strategy is already an analysis heavy style of game to pair with a second complex system.

So my new vision is combat as a brawl. Less focus on ranged combat, more individuals pairing off, teaming up, moving about generally fighting. Two individuals will pair off, I'm picturing some sort of tug-of-war bar to show who's got an edge. If it gets bad, friends are likely to break off combat to assist. Individual victories can snowball into tactical advantages: The flanks of your group can engulf the enemy, or you can break through the center to split them.

You'd set the formation of combatants, and assign roles. Some would try to match up 1-on-1, some would move about trying to create temporarily 2-on-1 advantages, some stay back a little to prevent the enemy from breaking through the line. During combat you'd shout orders. Switch out a tired mercenary with a reserve. Tell the rookie to stop fighting that veteran and find someone he has a chance with. Change the desired formation to meet the new realities of the battle. Choose when to retreat, or yield, or when to demand condtional or unconditional surrender.

Largely, though, the mercenaries would have autonomy. And because combat breaks up into lots of 1-on-1 pairings, the narrative should be more obvious. That guy killed my best friend. He had the opportunity to help and didn't. She knocked three opponents out of the battle herself. He immediately tired out and was no help at all.

Because I want you to get to know the characters, I want few deaths. But that works well with mercenaries anyways, why push it when you're fighting for money? So I'm thinking that characters will usually withdraw/yield after some injury, and that if one side gets a meaningful number advantage the other will give up. So if a battle can't be represented as, at worst, a group of 2-on-1 fights, it'll just end. Because these are contract fights you'll be paid something even for a loss, and there shouldn't be narrative battles that you have no choice but to win. I'd like to encourage the player to accept losses, because that introduces interesting stresses into the social group.

What I like is that it ties different elements together. You make decisions (hiring, rewards, conversations, etc.) outside of combat and then you suffer/benefit from the outcome during combat. And the story that unfolds during combat changes these relationships, which then changes your opportunities/costs back outside of combat. Effect loops feel like good design to me.

It is precisely that, it's your core design loop: the subset of repetitive actions your players will engage in. You need to make sure that this process, more than anything else, is fun.

What I worry is that it's bandaging two systems together: that whatever mechanic is used for combat, some players will just want to do the fighting and some will just want to do the social bits and each will feel like the other piece is a detraction.

Some players will not care for relationships, and for these players, there are tons of games out there that are more generic and focus on something else. For the other subset of users, which appears like your actual target audience and niche market, managing the relationships will be paramount, and fighting will be the process through which they will validate their efforts.

Think of it this way, CCGs are games where a lot of the time is spent deck building. If you ask 90% of the player base if they like to spend time designing decks, they will tell you that they do, some of them might even say that this is the part they prefer, but without actual games to play, deck building wouldn't make sense. It seems to me you're in the same scenario: you have a strong and interesting loop around managing your crew, and fighting is that which ties it together as a full-fledged experience.