I've been trying to come up with an extremely detailed martial arts system for a fantasy game I've been cooking up. The martial arts system will be extremely detailed since the game setting is part "Romance of the Three Kingdoms", part "Crouching Tiger Hidden Dragon", part "Last Samurai", and part "55 days in Peking".

One of the problems I'm having is how to classify the techniques. In my system, I want the players to be able to customize their moves based on how well they are skilled. Just like how Ars Magica allowed players to dynamically and spontaneously create magical effects, I want players to do the same with their martial art techniques. In real life, while one may often use pre-learned maneuvers, it's also very common to come up with something on the fly.

My problem is therefore to try to do something akin to what Ars Magica did by classifying martial art techniques into sub-categories. However, I've come to a roadblock because I see two ways of classifying techniques.

On the one hand, you can classify a technique by its intention, and on the other hand, you can classify it by what it does. Here's the breakdown that I came up with:

By Intention
Interception:
Offense- Strike, Grab
Defense- Block, Deflect
Manuever:
Offense- Feint, Close
Defense- Evade, Escape

By How it works
Combat Elements
Damage- to inflict either normal (bruising), nerve, or lethal damage
Takedown- to make your target prone
Neutralize- to disarm, bind, trap, pin, or takeaway
Intercept- blocking (force against force), deflection (parrying)
Evade- Full (to get out of range of combat), Partial (a dodge)
Escape- Reversal, Escape, Counter

Notice that there's some overlap. Also, the elements that define how something works aren't mutually exclusive and can have two or more elements to it. However, doing so means the maneuver is more difficult, and one would have to divide up his Skill Pool to attempt such a complicated maneuver.

If I define a character's skill by the Intention route, then it can be rated on 4 axes: Intercept Offense, Intercept Defense, and Maneuver Offense and Maneuver Defense. However, if I do this, then it doesn't have as much flexibility. If I define skill by rating a character in each of the Combat Elements, it's much more flexible, but it's also more time-consuming.

Moreover, I've decided that all techniques will also be rated on 4 other factors:

Speed- How quick is the maneuver
Accuracy- How accurate is it
Power- How hard does it hit
Defensability- How unstable or open to attack does it leave you

So not only will the player decide what Combat Elements will make up the technique on the fly, he'll also have to divy up the pool to the 4 factors above. Because my system is already extremely detailed and complex (my initiative system is based on a variable action count and goes on indefinitely much like the old Phoenix Command system, and the damage system itself is divided into 4 types of damage) I'm wondering if I should make it easier by going the first route. Personally, I'm not afraid of a complex system. I wanted this system to be as detailed and tactically crunchy as possible, so it's not for the faint of heart.

Perhaps someone else can think of another way to classify how to break down martial techniques that I havent' thought of.

How about you make them a grid of ( your intention x opponent's intention )?

Because, really, the maneuvers I will use to try to (say) Control an attacker are very different depending on whether they are trying to Attack me, Control me, Maneuver on me or Confuse me. It's not like Ars Magica, where spells are created largely in isolation. Martial Arts maneuvers assume an opposing intention as part of their very structure.

Oh, and don't elevate Defense to the level of an intention. Defense isn't an intention, it's a constant consideration in pursuance of other goals. "All out defense" is attempted Maneuvering... specifically retreat.

Whoever declares first should get the benefit of the reaction-gap (which is to say that the person who waits to see what their attacker is doing loses a few precious milliseconds as their senses and brain parse that information). Whoever declares second has whatever penalty that implies, but gets to pick their best responding move. Folks who want to both be siezing the initiative can declare secretly and reveal simultaneously.

Does that help?

Here's an alternate idea. Both your classification schemes are more or less based on qualities of the would-be maneuver: what the action itself consists of (either by intention or function). You could, howeber, change your thinking entirely, and instead create classifications based on the relationship a maneuver creates between the fighters (or a change in relationship). For instance, moves might cause one combatant to...

be closely bound to his enemy
be distanced from his enemy
deceive his enemy
reflect on the other's weaknesses/strengths
overpower his enemy with strength
enter a game of quick moves dependant on timing

Or something like that. Those are perhaps not the best examples, but I hope you see what I'm getting at. I've just been reading a book on the psychology of Eastern/Western thinking. Supposedly, people who grow up in a heavily westernized society are more likely to categorize things and focus on agency, while people brough up in the far east don't categorize so much (especialy not on the basis of traits) and instead focus on relationships. Since your game is emulating eastern sources maybe that'd be an appropriate shift to make.

Thanks for both the replies so far.

Both of you have suggested something I had an inkling of before, but the permutations just seemed too staggering to explore. I do agree that melee combat is very much akin to dancing in that what you do is based in large part on what your opponent(partner) is doing. In other words, a technique isn't created in a vacuum, but is related to how the opponent is also reacting.

In a sense, I've already gone a little bit down this path. In the beginning of the combat, the player secretly writes down two base actions that he will perform. They could both be offensive, they could both be defensive, or they could both be holds or any combination of the two. One action is declared primary, and the other secondary. Once this is done, all the primary actions are revealed by all parties to the GM, and either an offensive, defensive or neutral posture is declared by everyone (though the specifics are not revealed yet). Now in my game, initiative is determined by physical speed, skill, mental factors as well as emotional factors. From this, the GM can determine whose action starts first, though it will not necessarily reveal whose action will be completed first.

After this phase, the GM will declare whose actions are initiated first. At this point, a player may decide to switch to his secondary choice, but it will cost him some time (reflected by an increased Action Count). So for example, a player may decide to defend if he sees his opponent attack, and the opponent launches his attack first. It could very well be that the player's own attack may land sooner than his opponent (for example, a roundhouse kick takes longer to fully perform than a quick jab), but the player may hedge his bets and switch to a secondary defensive posture instead.

However, a matrix of relationships isn't out of the question, and I'll play with that idea in my head for awhile. Perhaps it might not be as complex as I worry it may be.

As regards to the defensive issue, I was thinking of having Defensability be inversely related to Power and Accuracy, and proportional to Speed. But this seemed to be somewhat unwieldy since it required yet more calculations to be made. I'm thinking of allowing only Defensive postures to buy Defensability. The elements Evade are actually of two forms, to Withdraw, and to Dodge. Withdrawing means getting out of attack range, and Dodging means keeping in combat range, but sidestepping, ducking or weaving out of an attack. Withdrawing therefore needs no Defensability trait, because if it succeeds, then the attack misses. Dodging however will still require it.

It doesn't have to be mathematically complicated, really. It doesn't have to be mathematical at all.

Say that two players are contesting. The attacker chooses the top hexagram of an I-Ching reading. The defender then chooses a bottom hexagram. The fortune is read, and colors the contest. If the attacker doesn't like that outcome, he changes his hexagram. Now fortune is different. Then the defender can change his hexagram. etc., etc.

This is a simple system, with no math (but one 64 entry lookup table) that lets each player sway the combat in response to the choices of the other.

Have you looked at The Burning Wheel? It has a system for different offensive and defensive techniques, and they are actually pre-scripted for 3 exchanges (with maneuvering as well) and carried out simultaneously. That might be giving you some inspirations for your own design.

An idea:

You already have players investing points in "defensibility" "power" "speed" and so on. Why not have moves categorised on intent, and invest points into not only "speed" et al, but also effect. This lets you create aggressive defensive moves, feints, and combinations fairly easily.

For instance:

Dragon Flaming Kick
(Offensive Strike)

Speed 2
Accuracy 2
Defense 1
Damage 5
Knockdown 3

Painful Meteor Block
(Defensive Grab)

Speed 4
Accuracy 3
Defense 2
Damage 4

Building from this, styles could give multipliers or adders to these categories.

Falling Swan style

Defense x2
Speed -2

Iron Fist of Seven Walls style

Blocking x1.5
Damage +1

Perceptive Hunter Technique

All +2 vs any Animal style.

xenopulse wrote: Have you looked at The Burning Wheel? It has a system for different offensive and defensive techniques, and they are actually pre-scripted for 3 exchanges (with maneuvering as well) and carried out simultaneously. That might be giving you some inspirations for your own design.

Mike,

... not sure what you mean regarding axes, in an applied sort of way. Do you mean that, instead of different singular actions, you'd want different aspects of actions that you can combine in various ways?

I don't have TRoS, so you might well be right in that regard.

Well, TROS isn't really the greatest application of this principle. It shows some other good principles, however.

But, basically, let's say I have four stances, four angles, and four limbs that I can designate for an attack. That gives me 64 combinations that I can make. Now, let's say that I have three stances, three angles, three limbs, and three animal styles. This method has 81 combinations. Note that both models have 12 options available, so I've created the same amount of options in either case. But by breaking them into more axes, I get more combinations.

Mike

I was thinking of requiring a specification of which limb(s) would be required as well as an attack angle (hi, mid, low, and a 12 o'clock angle). But this might be overload too.

I had envisioned the game essentially requiring miniatures to be used because you have to have a firm grasp of distance between the oppoents as well as facing. For example, even if your back is to an opponent, if you are skilled enough, you may not face too much of a penalty to defend and none to attack.

I also need the techniques to be based off of more than the 4 secondary elements of Speed, Power, Accuracy and Defensability. The techniques have to be defined in terms of not just what the move intends to do, but how it's actually done. The trick is, do I go by intention or do I go by implementation?

I'm starting to lean in the direction of having the Skill Pool defined by its Intention rather than by it's Implementation. The reason is that ultimately, when a fighter chooses a move, he doesn't think about what it does, he thinks about what he wants the effect to be. As a matter of fact, in Aikido, the main principle is that the form of the attack is irrelevant, what's important is that you stop the intention of the attack and neutralize it. The specific implementation of the attack is not what's important, and it's not the essence of the attack. In other words, Intention = function, Implementation = form.

However, merely stating your intention, say for example an Offensive Strike, doesn't really tell you how it works. For example, is it a kick, a head butt or a punch? Is it coming from the left side or right side? How quick is it? How accurate is it? If I define the Skill Pool based on Intention and not Implementation, it's less granular and means that one's skill level is more coarsely defined. For example, there may be two fighters with equal Skill Pools in Offensive Strike, but one may be really good with kicks, while the other is really good with punches. Or maybe one stylist mainly only knows quick but not very damaging techniques, while another is slower but delivers mostly very powerful techniques.

If I go the tabular Matrix route, I'm also worried about the game turning into a Rocks Papers Scissors-fest. Of course it'll be more complicated than a simple 3x3 matrix, but it'll still wind up with a "this move beats this move" mentality which I find not only somewhat unrealistic, but also leading to a style of fighting that I don't want. However, there's still some merit in the idea. A Matrix representation can be further defined by the secondary and tertiary elements (tertiary elements are requirements, limitations or further definitions of a technique...for example, a sacrifice takedown in which the attacker also winds up prone, or Requires, in which one technique can only be attempted once another technique has been successfully completed).

Can you clarify exactly what you want here?

I mean, yeah... you want a "detailed" system for martial arts. But why? Just for the sake of having it detailed? Or are you in pursuit of the elusive realism? Or do you want certain types of tactical challenges? Or are you simulating a certain martial-artsy feel? Or what?

I think a bit more focussed a subject would help everyone to offer you more helpful advice.

I wanted this for a couple of reasons. I really liked the Ars Magica magical system in which you could spontaneously create magic on the fly. I basically wanted to do the same thing to the martial arts genre what Ars did for magic. I also think it's a much more realistic way of modeling melee combat, since you don't always rely on pre-wired autonomous combat moves that you practice over and over again in your katas. The essence or function of a technique is ultimately more important than its form. But virtually all games to date rely on fixed combat moves which have static rather than dynamic effects.

The detail is there for the flavor it gives, but I want the detail more for a simulationist way of doing things rather than a gamist way of doing things. In fact, I want the detail there to help with the description of what goes on. For me, it's not enough to just describe what the move is, I want that description to have concrete and tangible (and non-arbitrary) effects. The detail will hopefully create a more immersive environment. In some ways, you can think of all the detail as a reward both in the descriptive sense, and also in a tactical sense.

EDIT

Oh I almost forgot. The system is there for a pan-pseudo-asiatic world set in a semi fantasy world which has technology bordering on our late 1800's. Ultimately, it will be about science vs. mysticism, eastern vs. western viewpoints, and will focus a great deal on many brotherhoods and other orders (spiritual, fighting, monastic, occult and others). Because of the asiatic and philosophical bent, I really want a detailed martial arts system that would do justice to not just the physical aspect of combat, but its mental component as well. I'm also therefore working on the effects of discipline, willpower/concentration, morale and passions/emotions in their play on combat effectiveness.

Wow, you sort of want everything. That's a hard row to hoe. I wish you all the best of luck with that.

I think one of your greatest obstacles is that martial arts really doesn't translate to words very well. It's pretty visceral, even when viscera aren't explicitly involved. For magic (as in Ars Magica) you have lots of useful symbols to hand: "And then fire rains down from the cloud, striking the ground like a flight of arrows and smoldering there like marshlight." Martial arts is more sort of "the-thing-itself", in defiance of description: "And then he sort of... you know... grabbed the guys arm and pulled it over his shoulder, like this... no, the other shoulder..."

Maybe some good practice for you would be to memorize some Jackie Chan fight sequences (I particularly recommend Drunken Master and Drunken Master 2) and then try to describe them in words to someone who hasn't seen them. I haven't tried that, but I suspect that it will force you to develop some verbal techniques that will lead you in fruitful directions for your design.

Again, best of luck! I look forward to seeing what you create.

TonyLB wrote:
I think one of your greatest obstacles is that martial arts really doesn't translate to words very well. It's pretty visceral, even when viscera aren't explicitly involved. For magic (as in Ars Magica) you have lots of useful symbols to hand: "And then fire rains down from the cloud, striking the ground like a flight of arrows and smoldering there like marshlight." Martial arts is more sort of "the-thing-itself", in defiance of description: "And then he sort of... you know... grabbed the guys arm and pulled it over his shoulder, like this... no, the other shoulder..."

Dauntless wrote: hopefully I can capture the sense of flow in combat that will make people realize that combat IS dynamic and not static, where the line between offense and defense blurs, and where ultimately the mind should be nowhere in particular.

I think that last suggestion is brilliant :-D

Pick up and read the game Swashbuckler! by Jolly Roger Games. While addressing an entirely different genre, it captures a lot of what it feels like you want, in a mechanism which is tight and smooth. (It's also essentially incomplete, really it's just a combat system pretending to be an RPG, but hey - that's what you want it for anyway.)

In short the way it works is that they've broken down swordfighting into a set of twentyish maneuvers - things like "fast slash" or "feint". Both players select a maneuver, reveal, and cross-index the two; the result is a modifier to their rolls on 1d20. Winner's intention is successful, doing damage as listed under the maneuver. The cute trick is that each maneuver also has a list of things you can do next which is quite constrained. For instance, Lunge is statistically fairly powerful... but the list of things you can do next is basically "Recover stance" and maybe a weak parry, and that's it.

The interesting thing is the way it brings sequences of maneuvers to the forefront of tactical thinking. The bonus on an appropriate defence is high enough that the defender will almost always win that roll, so it's all about timing and positioning, waiting for the other guy to trap himself into a situation where he can't properly deal with what you throw at him. All it would take for this to be wuxia-compatible is basically to give names and visualizations to these sequences, emergent from a set of maneuvers designed for martial arts.

Anyway. Just a pointer to a source that has done something kind of like what you're talking about...

- Eric

The last comment had a really important bit - your action determines what you can do next. The character will be affected both by the response of his opponent, and the position he has put himself in. If the last move was "headfirst dive at the other guy's feet," then there's no way that the next move could ever be "roundhouse" - an intermediate step (such as "get back to feet") is required.

I'm not sure how this would be handled - the first thing I thought of was trees of action, where each potential move has a limited number of next moves listed, to be modified by opponent's actions. This is where a more nebulous system, like the intention based ideas up above, would simplify things quite a bit, simply because your lists could be much shorter!

I also am not afraid of complicated combat systems, but the thing to be careful of is that your system doesn't turn into a tabletop game, or that the drama isn't utterly wiped out because it takes ten minutes to do one round.

Cheers! :)

hmmm, very interesting suggestions so far.

There was a nebulous thought in my brain that it was the relationships between postures and moves that was important...which was how the idea of a matrix of moves first came into my head, but I couldn't seem to figure out how to flesh out the idea. I had a vague notion of postures before, but I think Shreyas hit it on the head that it is the blending of intention and implementation within the movement of actions. It hadn't really occurred me to combine movement and technique together (I had even gone so far as to seperate the two into Intercept and Maneuver categories which would imply that one could either focus on striking/grabbing, or moving in/out of combat range).

The suggestion that the sequence of moves is important also hadn't occurred to me. Obviously if a fighter winds up on the ground, it's going to be hard to do some techniques. Trying to account for the gazillion possibilites might be too difficult to officially categorize, so this may have to be left up to GM judgment on the issue. Still, it'd be nice to at least give rough guidelines. In some ways I had accounted for this with tertiary combat elements. For example, some moves have a Follow element, which means that the move can only be attempted after the completion of another successful technique (often after grabs). I think if I want to include, it will be absolutely necessary to indicate which limbs are involved in the technique, and what angle of attack they are used in.

Dauntless wrote: Trying to account for the gazillion possibilites might be too difficult to officially categorize, so this may have to be left up to GM judgment on the issue.

My main suggestion is this:

Make a set of cards for each combatant. Each card represents a given choice. On the card are the possible results of that choice, based upon what you picked and what your opponent picked, plus a list of valid next choices.

I've been trying not to say, Multiverser does that, because I'm not sure it does everything you want the way you want it to do it; but I'm posting this because it does several of the things you want, and could easily do several more.

The primary point I see is that it permits players to create moves on the fly and attempt them during combat.

It divides moves into categories based loosely on intent--moves that do intensified damage, moves that provide tactical advantage, moves that overcome tactical disadvantage, moves that block or parry attacks, and similar sorts of things. It provides baselines for these and ways to modify them to match whatever the player describes. Thus if the player says, "I'm going to do a partial split so that my torso falls straight down, and then make a straight jab for his groin" (a maneuver someone actually used in play), the tools exist to determine how likely this is to succeed and what it will cost the player to attempt it.

Once a character has created a move, it becomes part of that character's abilities, and he can use it again whenever he wishes. Repeated use or practice of the same move improves the character's ability with that move, and so makes it more effective. He can still create a new move at any time. There is a degree to which relying on moves you know well is more dependable than creating new ones, but new situations can call for new moves.

Defensive moves can work in any of several ways. The most common of these is to act as a penalty against attacks/damage (the single roll attack and damage system means that any penalty to the chance to hit is inherently a penalty against the maximum possible damage that hit can do). Although the system does not specialize these in the main, that would not be difficult to do.

You were interested in distinguishing the guy who punches well from the guy who kicks well. That's not inherent in the system, but can easily be included. As written, the system easily supports a character being better at maneuvers he already knows which are based on a particular body part, and it supports setting up styles that favor particular body parts (in numbers of attacks permitted with that body part). It would be simple to provide individual characters with bonuses on a specific kind of attack which would apply whether the attack was known or newly created; it's just not specified as a normal variable in the rules.

It doesn't have the move-versus-move matricing of something like Burning Wheel or Riddle of Steel, but such complexities are easily added if desired. It's just a matter of more narrowly defining the maneuvers.

I hope that helps. As always, I'm glad to answer questions about it here or by e-mail.

--M. J. Young