Party creation that favors specialization

[Simons]

[J. Scott Timmerman]

I suggest you do it based on how you set up the mechanics of the abilities.  Limitations like that in Character Creation can make sense, especially if forces have limited resources for training their troops, but players might prefer having more options.

Have you thought about giving Ninja special abilities that work best when cooperating with other ninja?  Like distracting someone while your buddy blowdarts from behind?  Likewise, Fighters could have formations (like phalanx etc.), backup, and be able to make better use of arrow cover.

You could set up magic to where mages can/must cast certain targeting spells before being able to attack someone, and that fellow mages can use the same targeting.  Or you could do something like Chrono Cross, where constant casting of Fire-type spells starts to attune an area to Fire magic, and fire spells become more powerful.

Healers could just be hard to kill.  You could offer an ability where if there is more than one healer in the party, and one of the healers falls, another can heal them as a free action.  Even something as simple as a buff, casted by Healers, that only works on Healers might be simple enough here.

Another option would be to offer whole-party abilities.  Basically, the party gets to select ~2 of these abilities, and they apply to the whole group.  Having group-abilities that increase the power of one class while really not doing much for another class might encourage specialization there.

-Jason T.

[Monkeys]

Collectable card games usually favour specialisation, and they do this by having powerful cards that only certain characters/creatures can use, rewarding you for having one type of character and a deck of cards based around cards for that type. So if you used cards it'd be easy to encourage specialisation - in fact it'd be difficult not to.

[Monkeys]

You could also do what wargames often do: simply have various broad types of character, and say that each side must be made up entirely of one broad type. For example you might be able to combine Elvish magicians, knights and archers, but not Elvish archers with Giants.

Or you could have a slightly looser system, for example Paladins can't be combined with any Evil unit at all, whereas knights can be combined with anything other than Necromancers and Vampires, while human spearmen can be combined with anything.

[John Kirk]

Simon,

There aren't really all that many ways that RPG's have historically used to design characters.  These are: Class systems (including Class-Tree systems), Skill systems (including Skill-Tree systems), and Trait systems (including those using Templates).  Each technique has its advantages and disadvantages.  My personally preferred design strategy for creating themed parties is a combination of the Class-Tree and Skill-Tree patterns.  For a complete break-down of all these various strategies, you might want to take a look at an early draft of my RPG Design Patterns book.

Let me try to explain how I see these various techniques applying to your problem.  But, before I do, let me emphasize the point that these explanations assume that when you create a themed party, you actually want the game itself to be focused on that theme.  Sessions involving an all thief group would be about the group going out as a team, breaking into mansions or banks, and stealing stuff.  Sessions involving an all mage group would be about casting spells, summoning demons, enchanting magic items, etc.  If that's not true, then you really don't have a problem at all, because an all mage party can have characters who are all distinct in ways that have nothing to do with spell crafting.  But, if you create an all-mage game where the distinctions between characters are based on non-mage qualities, such as personal charm, moral beliefs, social standing, and the like, then your game wouldn't likely be about magecraft at all.  Magecraft would most likely only provide setting material.  Rather, the game would probably be more about political intrigue, infighting between various mage factions, and such.  You could probably pick any of the various character design patterns and build a good game.

But, assuming that a game supporting themed parties is really about the theme, I would state the following:

Trait-Based Systems
A trait-based game can theoretically be used to create groups that are focused in the ways you describe, assuming everyone cooperates.  The problem with doing this in a trait-based game is that it takes more cooperation than most people realize up-front.  The reason is that if you all agree to create characters with very similar backgrounds, it can be difficult to make each character distinctive.  Even if all players discuss their characters in-depth up-front and craft characters with their own niches, there is very little in a trait-based system to prevent one player from inadvertently encroaching on another player's niche over time. It's not impossible, though.  You could make this work, for example, if you didn't allow players to add new traits to characters once they were originally worked up.  That way, the originally agreed-upon niches would be set in stone.

Skill-Based Systems
Skill-based systems have much the same problems with party themes as trait-based systems do.  The main advantage of a skill-based system is that you, the author, have much more control over the kinds of actions that the characters will perform in your game, since skill-based systems require you to enumerate all of the various skills that a character may perform.  That is also a drawback, though, in that it reduces player flexibility in designing characters.

Class Systems
A pure class system, which generally supports only a small number of archetypes, is probably the worst choice for creating themed parties because it virtually requires all characters of the same archetype to have the same abilities.  The pattern is actually designed to require each player to select a distinct class to give his character a distinct niche.  If you only have one class, then there is no distinction possible.

Skill-Tree System
A skill-tree system is essentially a skill system where some skills can only be obtained after having gained some rank in other skills.  So, for example, a mage might be able to cast "Summon Hell Hound", but only after he attained 5th rank in "Summon Imp".  Assuming you have a large enough tree of skills, this can be a pretty good design choice for creating themed parties.  Characters will start out will similar skills, but over time their abilities will diverge into more and more specialized niches.  Of course, this assumes that characters are unable to span the entire skill tree but rather must focus on a small subset of the entire tree.

Class-Tree System
A class-tree system is essentially a class system where some classes can only be obtained after having satisfied certain conditions.  For example, a mage character might start out as an Apprentice, but can aspire to attain either the Wizard, Conjurer, or Sorcerer classes by gaining ranks in specific spells (e.g. rank 5 in Lightning Bolt to attain Wizard, rank 5 in Summon Imp to attain Conjurer, or rank 7 in Bubonic Plague to attain Sorcerer).  Class-trees also support themed parties well, because character niches will tend to diverge over time.  And, with a large enough class-tree, even beginning characters can be sufficiently distinct.  For example, rather than have a single Apprentice class, you could provide classes for Wizard's Apprentice, Conjurer's Apprentice, and Sorcerer's Apprentice, each with its own repertoire of spells.

Skill-trees and Class-trees are not a total panacea, though.  They do suffer from the problem that they can only support themes that you, as the game designer, allow for.  And, they demand a great deal of effort from the writer to craft the multitude of skills and classes required by the patterns.

I used a combination of the Skill-Tree and Class-Tree patterns in my game Legendary Quest to solve this very problem.   You might want to download it and take a look.  It just might give you some ideas. Over the years, we have played many themed parties in LQ, including all-mage, all-thief, all viking, all witch-hunter, and all sailor groups.

[John Kirk]

I should clarify my previous post concerning Class systems.  I may have made it seem like the basic Class pattern is completely inappropriate for games with themed parties.  That isn't the case.  If the theme you are promoting necessitates all characters falling into the same class, then the Class pattern is inappropriate.  However, if you design a number of specialized niche classes all having similar themes, then the pattern is appropriate.

For example, suppose you are creating a game about colonizing America and you include classes like frontiersman, sailor, pilgrim, tribesman, and preacher.  This kind of class-based game would work, as long as the theme is "colonizing America".  But, if a group tries to use this system for a different theme, such as pirating in the Caribbean, it will fall flat.  In that case, everyone will want to be a sailor and nobody will have a distinct niche.  To make a class-based system work for a pirate game, you would need classes that make finer distinctions, such as Swashbuckler, Ship's Doctor, Quartermaster, Naval Officer, Corsair, Smuggler, and the like so every character could have a unique role to fill.

I hope that makes sense.

[Simons]

[Vulpinoid]

This is along a similar track to one of the ideas that I've kept playing around with for the last few years.

You can see one of my attempts to streamline this (rather unsuccessfully) in the recent [System in a Can] challenge.

The point I'm trying to make here though is that classess probably aren't really required, and party members probably don't need to be defined by a simple set of statistics on a card. These two concepts make for a fairly static game, and you then end up relying on the interplay of those cards to get dynamic and interesting results. Of course this may be what you end up going for, because single cards are easy for new players to get their thought patterns right for the game, and combo effects allow experienced players to gain more challenges.

This is how I tried to simplify my game, and then I applied the point buy system, where each type of troop has a specific cost in an attempt to brng a level of balance between the different types of troops who make up the teams. No system is perfect, but at least this addresses some of the concerns.

If you were after a more free-flowing concept, where all sort of characters are possible, there are other options available.
 
You could work on a concept of each character having a base level of skill (regular, veteran, elite, hero). Where race has a stereotype fitting each level.

You may simply require each team to be made up of five regulars, two veterans and an elite (where a hero can be substituted by removing the elite and two of the regulars), or you might choose to go back to the point buy system. Once you've chosen the types of base characters to be a part of your team, you then assign them roles within the team (combatant, medic, priest, assassin, necromancer, cyborg, etc...).

To keep the game pretty open, I'd give each of these roles a single associated skill, a character trait and some kind of numbers bonus. Regulars get to pick up one role, Veterans get to pick up 2 roles, elites 3, and heroes 4.

Once you apply a basic structure like this to the game, other things start to fall into place and you can start playing with mechancis that interact within the characters and within the party.

Some character roles may require pre-requisite roles or character traits(To become a necromancer you may need a less magical role first. "Before I became a master of the dart arts I had to learn the basics"). You could think of this a little like prestige classes in 3rd ED D&D.

Some roles may have such a hatred of certain character traits, that they simply won't work with one another (eg. High Magician vs Barbarian. "I'll never work with a barbarian, they're such savages!"). 

Some character types may have a natural dislike of one another. (eg. Dwarves and Goblins might get a bonus in combat against one another due to their hatred.)

Some character types may only be available if certain other charater types are present (eg. Zombies and Skeletons may require the presence of a character who is at least Elite level and has the necromancer trait.)

Some character types might share a symbiotioc relationship with one another (eg. a Mechanic and a Cyborg, where each turn the mechanic can heal a level of damage on the cyborg by repairing his armour.)

I'm sure there are plenty of othe ways that you can interlink these character traits without adding too much complexity to the basic game.

If you have three or four classes each bearing a similar character trait (eg. 'Snipers', 'Bombardiers', 'Shock Troops' and 'Marines' all possessing the "Military" trait) then it becomes pretty easy to theme specific types of teams. In the example just given, you could specify a specific type of team that may only include members who have the military trait. They might have other traits, but as long as they possess one of these military oriented classes they fit in.

Otherwise you could follow the less restrictive path where certain party types provide some kind of bonus as long as everyone possesses a certain character trait. (An entirely elven party may gain a bonus to ranged attacks if everyone possesses the "Ranger" trait.)

I do agree somewhat with what John has written though, and would choose a number of class that fits with the theme you are trying integrate into the game. I've been developing a series of worlds using the suggestions I've described above, including a steam-punk variant using different types of fey, as well as steampunk scholars and Victorian era colonial soldiers. Another variant I've been developing has been focused on space marines and alien races.

The two games are wildly different but they use the same basic mechanics (vaguely as presented in my [System in a Can] entry), and adventuring parties from both worlds could face one another yet remain fairly balanced between one another. 

This is why I've tried to use as wide an range as possible in my examples shown through this post. This style of game has a great potential, and I'd love to see how someone else would develop the concept.

V

[Simons]

[Monkeys]

Here's a simple way to encourage specialisation:

Each character has a charisma score.

The higher a character's charisma score is, the lower other abilities have to be (because they have a certain number of total points), or the more 'expensive' that character is (because the cost of a character depends on their total score).

Characters have to have a minimum level of charisma to get on with the other characters in their party, which increases as the size of the party increases.

If the party includes characters of opposed types, eg dwarves and trolls, paladins and necromancers, this requirement for charisma is higher than it would be otherwise.

This makes mono parties 'better value', in that the members can have lower charisma.

[Monkeys]

Mr. Monkey: First, I guess what is it about cards that encourages specialization?  I mean, in magic or pokemon, there are different types of energy, and each creature usually only uses one of them, so it is impossible to generate enough diverse types of energy to have a lot of high-level creatures of each type.  Is there anything like that that one could do for a RPG party? 

Most CCG's seem to have at least two broad types of resources - for example monsters and actions.

The resources are limited in how they can combine with other types of resources: for example, you might have actions that only certain types of monsters can take.

If you have only one type of action and only one type of monster (eg), and the type of monster you have can use the type of actions you have, then you'll always be able to use the action in your hand with the monster in your hand. Thus the more specialised decks are, the better.

Obviously the rules make it more complicated than that, but that's the relevant answer to why cards, actually CCGs in particular, encourage specialisation.

You could obviously make your game a card game...

[Monkeys]

	Logged

[Vulpinoid]

Here's a simple way to encourage specialisation:

Each character has a charisma score.

The higher a character's charisma score is, the lower other abilities have to be (because they have a certain number of total points), or the more 'expensive' that character is (because the cost of a character depends on their total score).

...etc.

The only problem I would have with this sort of system is that it doesn't really address the concept where certain character types are more inclide to work with certain other character types. Once you add this type of concept into the picture, then you leave behind the need for the charisma system. Maybe I'm missing something in it, and I can see a kernel of a good idea in there, but it seems to be heading in a slightly different direction.

Maybe a "Charisma" stat could be allocated to leaders, and this would determine the maximum party size, but that's about it. Monsters might count as 2 party members for this type of party set up, while zombies might only count as half a party member for a leader who happens to be a necromancer. Leaders with lower charisma stats would be forced to take more focused teams and might have to load up their teams with extra equipment, spells or tactical advantages to make up for their lack of troops.

This following comment is only my opinion based on what I've played and encountered.

Games that tend to be successful are the ones that are inclusive rather than exclusive. If you generate a game where half the rules say, this can't work with that, or this only works with that you'll get a bunch of players who say, but why can't I {insert rules breakage here}. For example; if Goblins can't work with Dwarves and that's a specific ruling in the game, you will always end up with a player who says, "But I want to play a faction of outcastes who have banded together to fight both the goblins and the dwarves".

On the other hand, if you used the "charisma" method described above, you could say that Goblins and Dwarves really don't like working with one another because they are opposed character types. A band of warriors lead by a goblin might be able to recruit a dwarf, but this dwarf would count as two team members for the purposes of maximum party size, or in other team generation methods, the dwarf might cost an extra point to recruit (either way the dwarf would instantly gain the "outcast" trait). Vice versa if a Goblin tried to join a team lead by a Dwarf.


Here are some suggestions, take them or leave them...

If the theme of you game is battle and you've mentioned magic you'll probably want at least five types of characters.

a. Expendable Grunts - weak guys who are only good at distracting opponents (unless they gang up).
b. Heavy Hitters - Individuals who can seriously punch holes in enemy ranks.
c. Defensive Walls - Individuals who can block all but the most devastating strikes.
d. Tactical Veterans - Individuals who offer some kind of unique ability that modifies the way combat is played out.
      This can include medics, cavalry, siege engineers, sharpshooters, etc.
e. Battle Mages - Specialist Tactical veterans who bring magic to the field of play.

Thieves and scholars could still be useful to a party, especially for players who like to think outside the square. A thief could be a character who only counts as an expendable grunt during the course of play, but before the game begins they might have a chance of stealing a vital piece of the opponent's equipment (which could entirely break their strategy for the game). A scholar may have knowledge about certain types of monsters and could give all of their party a bonus against specific character types ("Use cold Iron, they're fey!")

The number of character options you choose to incorporate into the game is your own decision. Low numbers keep it simple for new players being introduced to your game. High numbers keep the options open for players looking for a bit more depth. If you want the best of both worlds you can always have a basic game and then add some advanced options for expanded play.

V

[Simons]

I kind of like the idea for character charisma, but I don't think it would work in my game, because I need to keep character creation simple.  Basically, before my first playtest I tried to create a "sample" party, and each character took, maybe 2/3rds as long to make as a D&D character (after finishing the first, I said "screw it!" and uber-simplified it).

And Vulpinoid, I do kind of agree.  As I think I said somewhere, you can get a behavior by encouraging it, or by outlawing any other behavior, and I always preferred the former to the later. 

Actually, here's an idea that might work better than "charisma."  Let me know what you think.  In my game, each turn a general can only issue a limited number of orders (determined by your general's "leadership").  Some orders can be given as group orders (like 3 people move or 2 shoot), some cannot (like cast a spell or teleport) (everything as a possible group action doesn't work, or at least it's been broken in early versions of the game).  How about this: If you have very diverse groups, it lowers your general's leadership.  This way, there's nothing to specifically forbid you from taking a holy, mystic, evil, & nature group, but if you do you might only get to make 2 or 3 commands per turn.  Might this work, or does this fall into the category of "exclusion"?

And, actually, so far in my game everyone starts with the same basic set of statistics, and then gains abilities based on their template.  Thus, a wizard is just as good as a fighter, or at least he was until he took +1 casting level and the fighter took power attack.  There has been a real balancing force to this, because it has kept the game from being dominated by fighters, and has stopped one strong character from being unbeatable.  Although, I have been debating lately having characters with different ability levels (although, part of the charm of everyone having the same stats is that it's easy to keep track of).  My question was more how many divisions are needed, how many templates in each division, etc?

That does make a lot of sense about CCGs and resources.  I'm not sure of a way of incorperating that into my game (and I doubt that lowering overall leadership values, would work, since I think it I would need to lower it to levels so low that it would upset players).

And of course I'll have a way of simplifying it for new players.  Something to the effect of premade (or mostly premade) party rosters, at least until the learn the ropes well enough to go on their own. 

Oh, and perhaps my example about a theif in the party was a bad one.  I guess, well, I guess if I do have a "theif" template, it just means that the abilities need to be oriented around combat, or about improving a party's position pre-combat. 

Simon

[Vulpinoid]

I think the "leadership" score concept is a good idea. I've seen many wargames use a similar tactic.

If I remember correctly, Rag'Narok 1st Ed. from the French company Rackham uses a similar system where each turn a General basically gets to issue a set number of orders to their troops (depending on the level of their tactical knowledge) + half the roll of a d6. They can basically transmit these orders to groups of miniatures within a certain distance of them, and if they have sergeants and "lesser leaders" they can transfer these orders further.

I think Heroclix uses a similar type of effect where you can only allocate a certain number of actions to your team each turn, and this means that some characters will always miss out on acting. It forces a more strategic level of play.

If you were using a system like this it could simply cost 2 orders to issue a command to someone with an opposed character trait. The commander basically has to do a bit more convincing to get this particular team member to follow instructions.

If you've decided to stick to basic uniform character stats with a template build to personalise the characters, it might be an idea to consider the skill tree system of advancement. I'd give every troop two levels to progress along the skill tree. First level basically determines the overall character type (These two guys can be "Fighters", that guy can be a "Mage", and this other one can be a "Medic".). The second level brings a specialisation, or can be used to purchase an alternative path progression. (Let's make the first fighter a specialist martial artist, the second fighter can pick up a secondary character type of "Scout", the mage can now be refined as an "Elementalist", and we'll give the medic a secondary character type of "Scout" as well so that he's a bit harder to hit and can move around more easily). One character can be given a "Leader" level. If you wanted to really theme things you could give everyone a "Race Level" (Lets make them all "Orcs"), and that race level could open up new character options.

If I were going to present this in an easy manner for new players, I'd cap the level progression tree to two or three steps. No more than a half dozen basic occupation templates, the same for races. Each basic template would open up access to two or three higher level templates to allow players a good chance at team specialisation. These higher level templates might offer the chance to access "elite" templates, for players who really want to focus their teams.

For a starting game I'd restrict players to characters with a single template, while offering the chance for characters with two or three upgrade levels to players wanting a more advanced type of game.

Just to confuse the issue further...

Another option could be considered by looking at products like the Pirates collectable constructible ship game, where you could have specific types of teams, and then populate those teams with specialists who have specific traits/templates. A team of sabotage specialists who get a specific range of combat stats, you then apply a warrior template to some of them, or a medic template to another, etc.   

I'm just trying to present some possible ideas you might not have considered.

V