Topic: [The Shadow Conclave] (tactical RPG): Can avoidance of combat be FUN?
Started by: Ravien
Started on: 11/30/2009
Board: First Thoughts
On 11/30/2009 at 2:20am, Ravien wrote:
[The Shadow Conclave] (tactical RPG): Can avoidance of combat be FUN?
Hey all! I haven't posted here in many years (back when you helped me design Scarlet Wake, which was a tremendous learning experience for me), thanks to dramatic career changes and starting a family. But I do remember that you guys tend to have some really good ideas and collectively far more experience than I could dream of having. So I thought I'd come to you with a bit of an issue I'm having with a game I'm currently designing.
Before getting into the details of my issue, I'll note that this is being designed as an online multiplayer PC game, played by 5 players in a cooperative team (though many times, two teams (10 players) will go head to head over campaign objectives), but the important point is that as a PC game, things like specific dice mechanics, GM roles, flexibility of setting etc don't really apply.
One warning also: I tend to write long posts, which I totally understand nobody likes to read. I'll try to tone it down, but I won't lie, I do find it hard to leave out details :) I apologize in advance. If you don't like lots of reading, but are here anyway, you might want to skip down to the "My issue" section.
First some context to my issue:
Premise
The basis of the game is that you are part of an elite team of secret agents, working for an organization called the Shadow Conclave (hat tip to Raymond Feist). This organization is kinda like the CIA or FBI or whatever (I'm Australian so I don't really see the difference between those two), only almost nobody knows about it. The goal of the organization is ostensibly to protect the world and maintain order and quality of life. To those ends, the Shadow Conclave seeks out and collects every fragment of humanities ancient history, every artifact and scrap of text, to try to piece together the truth about the past. They must also seek out, kidnap, and train "gifted" children, to become agents of the Conclave, as these extraordinarily rare children are the only ones capable of using the few remaining functional ancient artifacts. When such gifted people use these artifacts, they can perform amazing and terrifying magic, but must be trained with extraordinary mental discipline, as using these artifacts has a strong tendency to send the user insane, in part due to unknowable properties of the artifacts, and in part due to the intoxicating psychological effect of wielding godlike power. However, since gifted children and ancient artifacts are so scarce, the Shadow Conclave spends most of its time and efforts "guiding" national politics from the shadows, recruiting non-gifted agents as sources of information and necessary resources like food, money, construction, education, etc., and investigating the hundreds of local superstitions and rumours in search of valid threats (which have been alluded to in fragments of ancient texts).
So players, as agents of the Shadow Conclave, will find themselves facing a huge variety of missions covering espionage, sabotage, blackmail, bribery, infiltration, investigation, and murder. One mission might involve the kidnapping of a duke's daughter, holding her ransom in exchange for him agreeing to withdraw his financial support for a private mercenary army he was secretly funding to create border conflicts to gain sympathy and support from the king, then before returning his daughter, poisoning her to ensure future leverage over the duke through controlled supply of the antidote. While another mission might involve investigating rumors in a border village of faceless demons stealing children at night, only to discover some crazy woman who went insane when her husband left her after her 3rd still birth has set up a bizarre cult deep in the nearby forest, where she has concocted a highly addictive psychotic drug which she gives to men she has seduced into her cult, sleeping with them in countless orgies during the day, and getting them to cover themselves in mud at night to sneak into the village and steal children, which she then slays and eats to absorb their life in the hope that their life will be transferred to her next child. And dozens of other missions with wildly different goals and stories, but many of containing clues and threads about the overall developing plot which I won't cover here.
In as many missions as possible, the player will be able to complete the objectives in a variety of different ways, and this is part of the issue I have, which I'll get back to later.
Setting
As alluded to above, the game is set in a medieval time period, in a world not too dissimilar to our own history. By and large, the world functions much as ours did. There are no gods, but that doesn't stop the fact that there are dozens of religions with further dozens of variants. There are no vampires or werewolves or what-have-you, but that doesn't stop the superstitions. Life is hard, magic doesn't exist (except for the extraordinarily few gifted people working for the Shadow Conclave, but even then, as the plot reveals, that's not really magic, just incredibly advanced technology beyond anything they could imagine), and bad things happen to good people. There's no "big dark bad guy" to be defeated, but that doesn't stop people from misinterpreting information and mixing reality with fancy to convince themselves that there is.
Core Gameplay
Players are incredibly powerful agents, working for a secret organization. As such, the prime imperative at all times is maintaining the secrecy of the Shadow Conclave. Although the Conclave has dozens of god-like agents, they have nowhere near the numbers or capacity to deal with the threat of an entire nation's army bearing down on them, let alone several nation's armies, let alone the self-sufficiency to withstand a siege for longer than a month. In short, it is hammered into the players that they must not reveal the existence of the Conclave ever.
So this gives rise to the core mechanic which creates much of the challenge for players: Suspicion. When players enter a town, they are immediately suspicious. Not because they have magical flaming swords or anything, but merely because they are 5 strangers entering a small town in a world where gossip is the primary source of all entertainment, and the whole concept of "evidence" doesn't really exist. By virtue of their presence alone, they are interesting and suspicious. Now let's imagine that one of them decides to make lightning shoot out of their hands to strike down some random local patrolman. How would the local people react? What would the remaining constabulary do? Would that information spread like wildfire? Would the town turn on that player, condemn them as a demon or witch, then take to them with torches and pitchforks to unite against this new terrifying threat?
Suspicion is all about that. Every other character in the world has eyes and ears, and can see and hear a realistic distance. Create an explosion in a world where explosions don't really happen, and everyone within a mile is now highly alert, and everyone within a hundred meters is coming to investigate.
The challenge for the players lies primarily in exercising tactical discretion. Yes they have inordinate amounts of power, but yes, these powers do tend to be highly noticeable, and yes, will likely cause realistic and undesirable reactions from other characters. So the challenge them becomes not "how can I create maximum output of my powers?", but "how I can gain maximum effect through minimal use of my powers?"
To further this gameplay, damage dealt to and by the players is handled in precisely the same way for everyone. No hit points or any of that, just a few wound levels. Wounds range from inconsequential (skin and flesh wounds), hindering (rendering a limb functionally useless), grievous (mostly painful, slightly hindering, probably deadly in a very long period of time), mortal (incapacitating and deadly within a very short period of time), and lethal (immediate death). In addition to wound levels, there are factors like shock, adrenaline, bleeding, and pain, but I won't get into the details. Also there is a "frenzy" mechanic tied to the adrenaline factor, which allows a "reversal" element to counter the whole "spiral of death" thing where once a character takes a wound the penalties almost ensure they are going to die - frenzy allows the player a chance to turn that around. But that's not really that important either, I'm digressing.
Back to core gameplay. Players must avoid suspicion through careless or unnecessary use of their powers, and the consequences include very high chance of death (through not-particularly-forgiving damage mechanic), and mission failure (for exceeding a certain threshold of suspicion).
So in order to actually complete their missions, they must work together cooperatively, each playing a part in making the objectives possible. Combat then, clearly, is not the focus of most gameplay. Instead, players must earn the trust of the local NPCs (to lower their suspicion) by purchasing their wares, gambling with them, drinking with them, exchanging stories with them, flirting with them, etc. They must obtain information from NPCs that trust them (to get clues about how to achieve the objectives, or even where to start). And they must manipulate existing NPC relationships and factions in order to either make objectives accessible or even complete objectives for the players, by amplifying existing tensions, spreading rumours about secret affairs or whatever, agreeing to create diversions or convey threatening messages, etc. Much of the first half of any mission will involve a wide range of interactions with NPCs, to earn their trust, gain crucial information, and incite actions which result in objectives being achievable. The next half of any mission will then involve coordinated implementation by the player team to create diversions, gain access, prevent detection, actuate the primary objective, then get the hell out of dodge without anyone being the wiser (again, if they don't cover their tracks, Suspicion will bite them in the butt).
One small note: Not all missions will be like this formula. Some will actually allow/require players to use their powers to dramatic effect, some for example, even requiring that players achieve a high minimum Suspicion, such as to amplify the perception of a particular person being crazy or insane, by creating such a show that any retelling would come across as being complete lunacy.
Player Abilities
Players obviously have access to some incredible powers, each class being focused around a core tactical theme, but I won't elaborate on the classes here. Many/most of these class-specific powers are related to combat, or more specifically and accurately, particular roles within a tactical combat environment. And of course in addition to being confined to particular roles within the team, combat abilities are also balanced in a scissors/paper/rock fashion.
Players also can choose personality archetypes, which give them various social abilities and effectiveness with various social abilities. These are activated in exactly the same way as combat abilities, but obviously to different effect. And similar to combat, personality types and abilities are most effective on certain other personalities of NPCs, such that for example, a player with a highly flirtatious personality would have little luck in dealing with a character with a cold and analytical personality, much like the scissors/paper/rock balance of the combat setup.
In addition to each player choosing their class (combat) and personality (socialising), they can also choose their background. As in, what sort of family they came from. This is a more minor choice, but important nonetheless, as I'm also modelling social class. As such, and in keeping with the whole "secret agents" thing, disguises are a fundamental and very important aspect of gameplay. Your background gives you greater effectiveness with disguises which match your upbringing, and progressively significantly lesser effectiveness with disguises far removed from your upbringing. This is because disguises are more than just the clothes you wear, they are also how you carry yourself and interact with others, and even slight inconsistencies can ruin everything. And of course, this means that players wearing certain disguises for a certain social class will find it far easier to interact with NPCs of similar social class, and progressively lesser effectiveness interacting with characters of vastly different social class. I know this isn't really a player ability, but I was kinda heading into character generation territory anyway and as I warned: I ramble. :)
My Issue
Since actively avoiding combat is such a rewarded feature of the game for many of the missions (at least until the climactic penultimate missions), it has started to concern me that the rest of the game might not actually feel "fun" for a lot of players. I mean, personally, I really enjoy thinking about different ways to tackle tricky problems, and working these things out with friends. I'm designing this game for me, so I personally would find it intensely interesting and challenging. But it's a multiplayer heart at its core (nothing can be completed without a team of players), so it needs to be fun for more than just me.
I don't really know where to turn to for inspiration on how to make in-game social interaction with NPCs actually "fun" and engaging, since basically nobody has really developed a system for it, or where they have (like Fable, say), it's an undernourished beaten stepchild next to the combat system, which ends up being infinitely more enjoyable and dominating the vast majority of the gameplay. Other adventure games tend to be all focused around solving puzzles without any NPC interaction.
This concern was actually highlighted to me the other day, when I was elaborating on the competitive aspect of the game. Brief context: many missions are not only undertaken by your team, but are also undertaken by a competing team, working for a rival secret organization called the Crimson Council, who have competing and contradictory objectives. So the dynamic I was working on here, was that in addition to managing your own Suspicion by avoiding using your powers, you also had to compete with another team of players who are trying to achieve conflicting objectives to your own, and thus will almost certainly have to use your powers. And to make things interesting, I thought it'd be awesome if teams had the option of screwing with the other team's Suspicion, through things like spreading rumours and stories about them, eventually turning whole groups of NPCs against them, making it very hard for the other team to achieve their objective. So there'd be like a sort of social warfare going on during the mission build-up, followed by very tense conflict in the implementation phase of the mission. The challenge would be incredibly tactical, with players wielding enough power to be devastating very quickly, but with the very high risk that using said powers would result in the failing of their own mission.
But as much as I think this social warfare would be friggin awesome, I'm having difficulty seeing how it could be actually FUN for the players. Especially since unlike with basic combat, you can't see the direct result of your actions on the other players. In combat, you can use a power, and see your opponent explode. But in social conflict, you use your abilities to spread rumours about the other players, and the NPC might respond with something like "ooh, I don't think we'll have their sort around here then, I better warn my friends", but you can't see your opponent suffering because of this.
Questions
Btw, sorry for the really long build-up, I have a tendency to think context is more important than others think it is
So I guess I have 4 main questions:
a) Could competition with another team via indirect social actions seem as fun or engaging as direct competition via combat? How much bigger would the payoffs have to be? Like if social warfare could be so successful as to shut the other team out of being able to complete their objectives by driving the locals to mob against them to drive them out, would that be too much / still not enough?
b) If players could achieve their missions completely through clever coordinated use of the social mechanics entirely, without ever needing to resort to combat, how big would the reward have to be to make that seem worthwhile and fun, if they could achieve the same mission through more combat oriented means? In other words, should I devote time and effort into creating a range of options for players to "win", ranging from total non-use of combat, to heavy almost complete use of combat? Or should I instead limit the range of possible ways to "win" to always require a certain percentage of social actions, and a certain percentage of combat actions? If the later, would a 20:80 split inherently be more fun than an 80:20 split, or even a 50:50 split?
c) Given that there are a wide range of missions requiring a wide range of coordinated actions and objectives, would players likely find the less combat oriented mission to be more boring and undesirable than the more combat oriented missions, simply due to the presence and need for violence?
d) What sorts of ideas might I include to make social interaction as close to being as fun as combative interaction as possible? Specifically in terms of feedback from player actions, rewards and progression, challenge, etc? Or is social interaction guaranteed to always be inherently less interesting than combat, and thus I should just make it "fun enough", and devote more time to the combat side?
On 11/30/2009 at 3:24pm, Rikiji wrote:
Re: [The Shadow Conclave] (tactical RPG): Can avoidance of combat be FUN?
a) I'm not huge on competitive online play, but I would question on how well this game is targeted towards the demographic that is. You're essentially proposing a 'team puzzle game' and I think you would have to make your own audience for it.
b) No matter how big the reward is, it won't make boring gameplay worth it. Ultimately it's a virtual environment with virtual rewards. The rewards mean nothing if achieving them is no fun. As far as options are concerned, I definately think multiple solutions is the way to go. Single-option mechanics can be annoying in single-player, I can't see them working in a team game. The ratio of social vs combat is an interesting idea, but would probably be most effective if it varied from mission to mission (which is consistant with what you've posted about different mission types).
c) I don't think there's any inherant presence or need for violence. If the combat is more fun, it's because the combat system is more fun to play. If you develop a sufficiently simple and intuitve system for the social gameplay, then it shouldn't be an issue.
d) 'Social' interaction is a bit of a misnomer the way you're using it. The computer will not be able to interact socially in any meaningful way, so the only true social interaction will be between the players. Interacting with computer controlled characters is, and should be viewed as, an elaborate puzzle game or a resource-management game along the lines of the Sims or other RTS games. There are lots of options there that can be very fun for the right demographic; however it will take a lot of work to make the system fun and intuitive. Combat, by contrast, is fairly straightforward and has been done to death - it shouldn't require as much work to make the combat system effective.
Overall, the setting and concept are intriguing. I'm not sure if it would appeal to any traditional gaming demographics, but I can see a lot of potential in the game play (maybe dedicated MMO raiding guilds? It sounds like it would play very similar to a boss raid).
On 12/1/2009 at 2:16am, Ravien wrote:
RE: Re: [The Shadow Conclave] (tactical RPG): Can avoidance of combat be FUN?
Rikiji wrote: I'm not huge on competitive online play, but I would question on how well this game is targeted towards the demographic that is. You're essentially proposing a 'team puzzle game' and I think you would have to make your own audience for it.
That's a good way to think of it, though the word "puzzle" for me implies that there is one "correct" solution, and many wrong solutions. What I'm aiming more for is many ways to succeed to varying degrees, and many ways to make things more difficult to varying degrees. But you're right in concept, players will have to approach each mission as a sequence of puzzles where their success at one puzzle influences further success at another.
One of my explicit goals of the game was to create a new genre with unique gameplay. Of course, unique doesn't imply that every single molecule of the system is unlike anything else, but more that the particular arrangement of molecules and subsequent overall gestalt comes across as unique. There's only so many ways you can reinvent the wheel. So thanks!
Rikiji wrote: No matter how big the reward is, it won't make boring gameplay worth it. Ultimately it's a virtual environment with virtual rewards. The rewards mean nothing if achieving them is no fun. As far as options are concerned, I definately think multiple solutions is the way to go. Single-option mechanics can be annoying in single-player, I can't see them working in a team game. The ratio of social vs combat is an interesting idea, but would probably be most effective if it varied from mission to mission (which is consistant with what you've posted about different mission types).
I don't know if I'd entirely agree with you on your first point. There are MMORPGs built entirely around amazingly repetitive and boring gameplay (Lineage II, Final Fantasy XI, etc), where essentially players simply repeat the precise same sequence of button presses thousands of times over without requiring even the slightest amount of thought as to what to do next. Then every few thousand sequences the scenery changes. If it weren't for the repetitive button sequences they might as well be watching the most boring and repetitive TV show ever devised. Have you ever tried watching someone play a grindy MMO?
But that aside, I definitely do not want any of my gameplay to be boring, so your point does stand as it relates to my game. To that end I'm trying to make social interaction fun via a few methods. In short, players have a range of primary social abilities available to them determined by their personality. All social actions are available to all players, but any action outside the scope of their personality is "secondary", and will be noticeably less effective. They perform these actions much as they would perform attacks: select target, press button. They have "buffs", such as a buff which causes them to move in a very sensual and suggestive manner, increasing the effect of any charms they use, or another example would be assuming a furious, barely contained rage manner, increasing the effect of any intimidation tactics they use, and so on. Just as a player characters' personality has strengths and weaknesses, so too do NPCs have varied resistances and weaknesses. These can be bypassed by getting the NPC into particular emotional states. In other words, if you try to, say, flatter an NPC for information, but they are highly resistant to your charms, you may try nonchalantly mentioning how people have been saying things about them behind their back, making them paranoid and dramatically lowering their resistance to your flattery as they try to win you over to cover for their anxiety. This is because every personality can be modified via different emotions. For some, anxiety makes them far more eager to please, while happiness makes them more open to reason. For another, boredom makes them more impulsive and impatient, while confidence makes them more diplomatic. So social interaction is a puzzle, as you say, of figuring out what personality the NPC has, and either matching the player personality to the NPC, or altering the NPC's emotions to make them more susceptible to one of your strengths.
The actual actions of interaction are slightly caricatured, and so would probably hold as much "fun" in and of themselves as the numerous /dance and /joke commands in almost every MMO, but that wouldn't last. Most of the fun from the actions themselves would be from the NPC reactions, which again of course would be exaggerated, and the escalation of subsequent similar actions. Maybe some NPC really annoys you, and you're most likely to get the information you want out of him if he's sufficiently scared, so you can start by insulting him, then slapping him around, then boxing his ears a bit, then building up to really laying into him and punching the shit out of him until he talks. Haggling would be slightly comedic, with the back and forths about having 7 children, then 11 and 4 of them are crippled, etc. That sort of thing. If you're swapping stories, you might be making up some complete farce in accordance with your current disguise, and they might tell you a story they heard which you recognize as actually being something your team did a few missions back, only the sketchy details have been filled in with superstition and folklore and basically sounds like it's been through a few dozen rounds of chinese whispers.
The goal of each interaction is to build a relationship with each NPC. You get feedback visually as to your progress via 4 relationship measures: how much they trust/suspect you, like/fear you, respect/disdain you, and how much you excite/disgust them. These measures of how they think and feel about you alter their behaviour towards you and to a lesser extent your team. Trust is the only measure which can't be directly manipulated, and must be manipulated via the other three. Trust is a little more complicated than that, in that you can't maximize one character's trust for you even if you maximize how much they like, respect and lust after you. The only way to maximize any one character's trust is by also earning the trust of their friends. Earning the trust of their enemies will lower their trust for you, and likewise interacting negatively with their enemies and making their enemies highly suspicious of you will increase their trust of you even further. It's a big cascading effect really.
So there are three layers. Every character has a core persistent personality (including the players), every character can feel transient surface emotions which alter their personality (including players), and every character has a social relationship network with other characters (including the players). The strategy comes in to who you build relationships with, who you make enemies with, and in the finer detail, how to build relationships with any particular NPC via manipulating their emotions according to your personality and theirs. The rewards for successfully earning the trust of a social network of NPCs are not confined to merely reducing your Suspicion for the mission and opening access to mission-specific information, but also enable social paths of progression, such as earning a Reputation, which lowers your starting Suspicion for all future missions, makes all social interactions more effective and thus trust becomes easier to build, and eventually, opens up the possibility of being Knighted and receiving your own fief. A parallel path of progression from building social networks is Investment, where certain social networks with sufficient trust will open up possibilities to invest in a variety of ventures. The Shadow Conclave encourages these activities, as having a strong Reputation makes you much less likely to be suspected in any ill-doings, and making your word carry more weight when you frame others or dismiss claims, thus safeguarding the organization's secrecy somewhat, and as per investments, this is the primary way the Conclave can make any gold at all, and you are actually investing their money (and your own, of course), and receiving a cut of their profits (in addition to all your own profits) as commission.
Anyway, it's not like it's going to be a series of dialogue trees. Talk about a yawn-fest. It's going to be much more active and with no reading (unless you have like subtitles on or something). You perform an action, they visually react according to a range of factors, usually with accompanying spoken response. The goal is not to hunt for the right line of dialogue, it's to create the right kind of relationships with the right kind of people. Once you have the right kind of relationship, any important information you need that they have is easily given. But yeah, absolutely no dialogue trees to read through and click the "right" response.
Rikiji wrote: I don't think there's any inherant presence or need for violence. If the combat is more fun, it's because the combat system is more fun to play. If you develop a sufficiently simple and intuitve system for the social gameplay, then it shouldn't be an issue.
I'd like to think that, and it's true that many games are successful and fun without any violence at all. But to me it seems like as soon as you add violence or combat to a game, that fact instantly overrides almost every other facet and becomes the core feature, while everything else is simply stuff you might do between combat encounters. I can't think of a single game of any genre where if combat is present, that is hasn't received orders of magnitude more effort and detail into the system and as a result, becomes the core of all gameplay from then on.
I'd like to design a game where there is combat and violence and yes, that part is fun and awesome. But also, there is some other factor, which isn't combat, and that part is at least as involving and fun and awesome. In my game, I'd like it to be social interaction, and both social interaction and combat are dealt with under the umbrella of the Suspicion mechanic. I'd like every single combat encounter to have a well-defined and player driven context, so that they know why they are fighting and are only doing so because it makes sense based on their interactions in the world so far. When they fight, I don't want it to be because "that's what you do", I want it to be because "I thought this was the best way to handle this situation", not the only way. There's a difference, and I'd really like to find a way to achieve that.
Rikiji wrote: 'Social' interaction is a bit of a misnomer the way you're using it. The computer will not be able to interact socially in any meaningful way, so the only true social interaction will be between the players. Interacting with computer controlled characters is, and should be viewed as, an elaborate puzzle game or a resource-management game along the lines of the Sims or other RTS games. There are lots of options there that can be very fun for the right demographic; however it will take a lot of work to make the system fun and intuitive. Combat, by contrast, is fairly straightforward and has been done to death - it shouldn't require as much work to make the combat system effective.
Granted that social interaction is definitely an elaborate puzzle game, but it needs a name, and "talking to NPCs to change their virtual relationship with you" doesn't have the same ring to it. :P
I absolutely agree that the combat is by far the easiest part to get right, which is why I'm not really concerned with what I've got. But I feel like I'm walking new territory with my social mechanics, so I'm far more concerned with the possibilities and methods to achieve my goals. Granted the Sims has a highly simplistic and comedic social system, and some modern RPGs are trying to create personalities for different NPCs, but they are still stuck on they dialogue tree method of interaction and NPC personalities are pretty much entirely static and only determine which dialogue tree option is correct. Morrowind and Oblivion tried to allow you to interact with NPCs to alter their reactions to you, but it always felt really fake and inconsequential and the abstraction into mini-games only destroyed what little immersion you might have had.
Thanks for your feedback. I agree that my idea is certainly not one that traditional gamers would latch onto, but I'm trying to forge a new niche, so that's actually a bonus for me. I'd be more worried if my ideas as a whole were too similar to existing franchises or genres.
On 12/7/2009 at 1:24am, JoyWriter wrote:
RE: Re: [The Shadow Conclave] (tactical RPG): Can avoidance of combat be FUN?
Conversation rather than combat can be much more fun! I was just playing Dragon Age and loving the dwarf politics stuff, (although I wish there was much more) and realising that I really need to get back into 4x games, or find another type of political game if I can. The thing about it was that the talking was actually more visceral to me than the fighting! It made me realise how awesome it can be to trade off political manoeuvring with the moral tension "if it were real", as well as the tactical possibilities and layers of hidden information available in that kind of thing.
Talking can be awesome if it has consequence, if appearance, standing and reputation matter, if ideologies can shift, if people can feel loyalty and betrayal, if it can be done behind other character's backs, and if it is grounded in believable lives.
On your specific implementation, I'm interested but snowed by boring real life stuff, so I'll probably have to wait to get back to you on the details, maybe even a fortnight if you can wait.
On 12/7/2009 at 3:50pm, Seamus wrote:
RE: Re: [The Shadow Conclave] (tactical RPG): Can avoidance of combat be FUN?
I don't know much about online games, but I do know avoidance of combat can be lots of fun in a regular RPG, as long as combat is still on the table. The game I developed is intended to play like an evening Crime Drama. So we made combat highly lethal. The end result in most sessions, is the players plan their actions so they don't get shot at. When combat seems like the best solution, they plan everything out in advance so they have the best chance of success. No running in with both barrels blazing. I think if you use a similar principle with combat (basically make it a very risky prospect with the ability to hedge-- calling in back up forces, making tactics matter, etc), you might avoid the issue you are concerned about. The key is to have combat on the table, but make it a less attractive option.