A Long-Winded Rant About Homebrewing a Roleplaying Game System
Posted by Doug on March 16, 2009
This post is exactly what it says in the title. Writing usually helps me sort out my thoughts, and lately my thoughts have been consumed by the task of rebuilding the Mecha Monogatari roleplaying game. There is a strong possiblity that this post will make no sense whatsoever. You have been warned.
First, a little backstory.
The Mecha Monogatari began as a homebrewed traditional (“pen-and-paper”) roleplaying game, cobbled together with elements drawn from the d20 3.5 SRD (which had been distilled from the 3.0 and 3.5 versions of the Dungeons & Dragons RPG) and the Guardians of Order’s Big Eyes Small Mouth d20 system. Basically, I removed virtually all of the fantasy-themed material, leaving just the core mechanic of the combat and skill systems behind, and retasked those mechanics to simulate battles between giant military robots piloted by teenagers in a never-ending war against alien invaders.
The result worked well enough, but it really did not work well. Some rules that were central to early versions (being able to damage to different parts of a mecha separately, for example) proved too clunky and awkward when applied to massive, sprawling battles (which I love. A small band of elite soldiers literally battling against hundreds of opponents, with little or no chance to rest or recover in between, is something I always try to work into my campaigns).
Battles were horribly repetitive, typically boiling down to a simple procedure of “pick a target, use a full attack action on it, wash, rinse, repeat”—just like the fighter class in the d20 System. The wizard class had all sorts of special attacks they could perform in the form of spells, but ostentatious magic like a wizard’s spells just did not fit into the style of the Mecha Monogatari as I had envisioned it, which was a lot closer to the “real robot” genre as compared to the “super robot” genre of mecha anime (of course, there was magic of a sort in the world of the Mecha Monogatari, but it was almost always subtle).
Unfortunately, my thick skull could not make the logical leap necessary to solve this little problem, until the Fourth Edition Dungeons & Dragons came along and slapped me in the face with the solution: just make martial fighting techniques like wizard spells, and simply describe the effects as being due to physical skill rather than magic. It was such an elegant solution, it wasn’t even funny. From the moment I laid my hands on the D&D 4E Player’s Handbook, I immediately began thinking of ways to incorporate the system of powers into a revised version of the Mecha Monogatari RPG.
But, alas, there were still many elements of D&D 4E I had no use for in the Mecha Monogatari. Some could be handled simply by ignoring them (for example, alignment), but others would require more thought. Like the level system.
To sum it up quickly, in D&D 4E (and in every previous version of D&D I am aware of), the characters gain experience points through adventuring (typically by killing monsters), and when their experience point total exceeds a certain limit, they gain a level, whereupon their abilities almost always increase. This system works extremely well for combat-oriented systems like D&D, but just never seemed quite right for the Mecha Monogatari, so in the original version, the Mecha Monogatari used a point-buy system: characters gained experience points, and they could spend those experience points on improved abilities, picking and choosing what they wanted. This seemed to make better sense to me.
Consider: a character wants to learn enough of a foreign language to ask directions. In D&D, they would have to fight lots of monsters and earn enough experience to gain an entire level, at which point they would get bonuses to their attack, extra abilities, more hit points, and some skill points which they could spend to become fully fluent in two or three or more languages all in one moment. In a point-buy system, they could just spend enough experience to buy a conversational level of fluency in a single language, without necessarily becoming a better swordsman or learning a new ability that had nothing to do with language, which made better sense.
Of course, this lead to another interesting problem: exactly how much does a single, discrete skill cost in terms of experience points? I have never found a compelling answer to this question. However, my current thinking is that “how many experience points does an ability cost?” is the wrong question. The correct question is “Given that this is supposed to be a more realistic world, what would a person have to do in real life to obtain an ability like this?” If someone wanted to learn a little German, they might take classes at the local university, or buy “Learn German in 30 Minute a Day” on CD and use that product, or they might hang out with that exchange student from Düsseldorf and see if they would teach them some. And that’s the way a character would learn a skill in the Mecha Monogatari: by roleplaying the activities that would logically result in them having the ability they wanted.
No character levels, no experience points—each character just has a list of abilities they possess, and if they want more, they can have them, if they’re willing and able to put in the time and effort. The Mecha Monogatari system works best, I think, if it is viewed as a cooperative storytelling game with randomized task resolution. The goal of each player—including the “Game Master” or whatever you call them—is to portray the character(s) under their control in a way that makes the overall story more interesting.
An aside: in roleplaying games, the Dungeon Master or Game Master or Narrator or whatever (the person who controls the enemies and the extras and the antagonists) is also playing the game. They are a player, just like the players who control the player characters. With this note in mind, the goal of every roleplaying game is to be fun for all the players. If everyone is having fun, even if all the player characters die, everyone wins.
Back to my bloviation: so, when creating a new player character, instead of picking a class like in D&D or buying abilities with a pool of experience points like in the first versions of the Mecha Monogatari, in this new Mecha Monogatari (which I am dubbing the Second Edition or the MM 2E for short), you simply describe the character and their abilities in plain language, and then after forming a consensus with the other players, you pick game mechanics that represent those abilities. You want to play an ambidextrous youth who used to dominate the junior high kendo competitions before he was conscripted to serve in the Self-Defense Force as a titan pilot? Fine! Just note in his list of abilities ambidextrous: takes no penalty when using off-hand instead of primary hand and +2 bonus to attacks when fighting with a katana or shinai, +1 bonus to defense against melee attacks.
The players should aspire to play interesting and memorable characters without being overly concerned with the cryptic symbols and arcane numbers that encode their mechanical performance. A character with subpar abilities in combat is not and should not be treated as a liability to the players.
The question then becomes, “Given that those neat little powers from D&D 4E I like are tied to the advancement of level, and I have thrown experience points and levels out the window, and a character can learn new abilities (including powers) just by taking the time to learn them, is there any limit to the number of powers that a character has?” The answer is simply No in theory, but Yes in practice. If a player wanted to have their character learn five hundred different powers, each for a slightly different shade of situation, and they were willing to roleplay their character taking years or decades seeking different masters from the different arts to learn them, then fine—they know five hundred different techniques. In theory, no mechanical limit. Of course, it’s up to the player to keep track of all those abilities. The game should not be forced to bog down for a half hour as the player sifts through tens of pages of power descriptions, searching for the perfect one—the player can rightly be expected to select their actions within a reasonable timeframe. Further, most players will usually find five to ten really good techniques and use them almost exclusively, even if they have hundreds more available. In practice, limited by the inherent complexity and lack of real need.
But what is to stop a player from selecting a single devastating power, and using it over and over and over and over, much like the D&D 3.0/3.5 fighter spamming his full attack action (only with a little more pizazz)? This is an unresolved issue for me. The D&D 4E system of “at-will”, “per encounter”, and “daily” powers seems wrong. If a character knows how to perform the Gai Super Upper power, why should they only be able to use it once per encounter, or worse, only once per day? The D&D 4E answer is partly “because after using the power, it requires rest before it can be attempted again” but mostly “because it would unbalance the game, making all challenges trivial.”
The second answer, while more true, is answered from a purely metagame level and is unsatisfying. The first answer is an attempt at explaining the process in in-game terms: powerful techniques are fatiguing, and after becoming too fatigued, you can’t perform such powerful techniques. But it doesn’t quite ring true because if you have a number of different powers that can only be used once every so often (the norm in D&D 4E), using one of them makes you so fatigued that you cannot use it again, but you can still use any of the other powers that are equally fatiguing without penalty.
So, for the MM 2E, I thought to limit more powerful powers by how much they “telegraph“. More powerful techniques would require a “wind-up” or preparatory motion which could tip an opponent off, allowing them to easily evade the attack and/or deliver a counterattack. But this would either require doubling the number of rolls made per action (if the opponent got to make a check to see if he predicted the attack) or significantly increasing the amount of bookkeeping necessary (if the ability was keyed off of patterns in the character’s power usage), or, worse, both.
I could just institute a “technique point” system, where each character has a limited (but replenishable) pool of technique points (TP), and every power has a technique point cost to use. For example, a character will start each encounter with 15 TP, and in order to use a powers it will cost at least 3 TP. Basic attacks cost 0 TP to use. TP can be restored by spending actions for that purpose, allowing a character to rest for a moment, collect themselves, and then use another power. This would still not address the issue of characters spamming their best attacks—there would be nothing stopping them from selecting their best power, using it over and over again until they ran out of TP, then resting until they could use it again, then using the same power again, then resting, and so forth.
This is an unresolved problem.
Hit points, the staple of D&D and the d20 System since forever, need to go. It does not make sense to me that, with all other factors being equal, a character with 1 hp left can fight as effectively as a character with 1000 hp left. The first character has been beat on pretty harsh, but it doesn’t slow them down in the least, but if they are injured any more, even by the mere scratch of a cat’s claws, suddenly they are completely incapacitated. I’m thinking on using a different system, and it goes a little like this:
If you’re hit by an attack, this actually means you were hit by the attack—none of this abstract “hit points are just plot armor” jazz. Of course, not all hits are equal. When an attack roll beats your defense, the attack’s damage is rolled, but instead of being subtracted from a hit point total, the damage is compared to a “Damage Threshold” which represents the minimum amount of damage an attack must do to be significant. There is also a Health Level, which represents how many injuries you can sustain before being completely incapacitated. If the attack dealt less damage than the damage threshold, you were merely nicked—put a band-aid on it later, if you bother to do anything. If the damage roll was equal to or greater than the damage threshold, the injury was significant, but not necessarily critical—it might have just clipped you, or it might be enough to incapacitate you.
Each time you are hit and the damage roll is equal to or greater than your Damage Threshold, you lose a Health Level. If your Health Level drops to 5 or less, you’ve actually been injured enough that the injuries or the pain or a combination thereof starts to make it difficult for you to function, and you take a −1 penalty to all actions. At Health Level 0, you are seriously injured: the penalty increases to −3, and your actions are sharply limited (you cannot run, for example). Further, you must succeed on a check at the beginning of each of your turns or lose another Health Level automatically. At Health Level −1, the penalty increases, you can perform fewer actions, and the difficulty of the checks to keep from deteriorating more begins to escalate. At Health Level −2, you are mostly incapacitated: the penalty to your checks increases to the point to where all actions are virtually impossible, and your actions are limited to attempting to crawl to safety.
At Health Level −3, you are critically injured, and from this point on, your fate is determined cinematically, governed by the needs of the storyline; rather than mechanically at the hands of the dice. You do not deteriorate further, but you can do virtually nothing. Character death is a real possibility in the Mecha Monogatari, but main characters will only die if their deaths will advance or enrich the story (for example, by setting up a revenge subplot, or spurring another character to snap out of their funk, or to add have a heroic sacrifice remembered for generations).
Armor would be represented mechanically as a combination of increasing defense and by increasing the damage threshold.
Or something like that.
Really, I guess I am exchanging one variety of hit points for another, but I feel that making combat more concrete (getting hit in combat actually means you get hit) lends to the more realistic/gritty vibe I am trying to attain with the Mecha Monogatari.
I am aware of the ridiculousness of using the word realistic in reference to a game where teenagers pilot twenty-meter-tall walking war machines against alien invaders. Still, the use of the word is valid, because my goal is not to make the setting as realistic as reality, but simply to make the game feel more like a “Real Robot” anime world than a “Super Robot” anime world. It is in this sense that I use the word “realistic.”
Another element I would like to look into is replacing the statistically “flat” d20 roll with a “bell-curved” dice roll for task resolution. With the d20 roll, it is just as likely that your performance will be great (roll a 20) as it would be more-or-less average (roll a 10). This type of roll makes skill worth less and luck worth more. The bell curve rolls I am thinking on using I first saw in the JAGS roleplaying system: roll four six-sided dice, and add them together, counting sixes as zero. This results in a number between 0 and 20, with an average roll equal to exactly 10. But it is far more likely to get a middling result (52% chance of rolling an 8 to a 12) than it is to roll extremely high or extremely low (only a little over 5% chance of rolling a 16 to a 20; the same for a 0 to a 4). Using this sort of bell curve roll means that a character’s skill (as measured by the modifiers they add to the roll) means more and the roll itself means less.
This seems more realistic to me: if a character has a certain level of skill, they should be able to rely on using that skill at that level more often than not. Furthermore, I am strongly considering removing automatic successes and failures. Some things should simply be outside the reach of a person’s skill, and thus spur them to improve themselves. It does not make sense to me that the most abysmal marksman wielding the most inaccurate weapon would always have a 5% chance of hitting and injuring the world’s greatest soldier, who is wearing full body armor and using his legendary skills to avoid getting shot at the weapon’s absolute maximum range. It also doesn’t make sense that the world’s greatest soldier with the most tricked-out weapon has a 5% chance of flat-out missing a sleeping, fully-grown elephant at point blank range. Those sorts of things can stand in a more fantastic setting, but I just don’t feel they belong in the Mecha Monogatari.
The non-combat task resolution system will likely end up looking much like the combat system, just with different skills providing modifiers. For example, if you’re trying to write an elaborate computer program, the Game Master simply picks an appropriate difficulty number to represent how hard it is to gain a minor success (much like an enemy’s defense in combat), and how many minor successes are necessary to complete the program (think health levels); the character uses the modifier from their Computer Programming skill instead of their modifier from one of their Weapon Use skills, each “round” is decided to last an hour, and voila! Writing a computer program plays out much like a battle. Add to it the fact that the task “attacks” their “patience” every hour, and if the character runs out of patience, they must rest before continuing…
Of course, some skill usages will require only a single check (do you see the tree that’s in front of the inept ninja, or not?).
Another thing I would like to weave into the rules of the MM 2E is the theme of teamwork. Working together with other people should be better than working alone at least as often as working alone is better than working in a team. It should usually be easier for a team of four to fight four enemies than it is for each one of that team to fight one enemy separately. Loners should still be able to exist—nothing quite says cool like the person who strikes out on their own, with no backup—but the coolness should be in the fact that it would be easier to work together in a team, and they’re just so damn good they can still complete their objectives without any assistance.
Central to this theme of teamwork is the role of a leader. A leader in the MM 2E will not just be someone who can buff their allies, but a character who actually leads the other characters. This is not simply someone who barks orders—any idiot can do that, and many do. In a military setting, it is not necessarily the highest-ranking individual present, either. A leader is someone who can organize their teammates to better accomplish their common goal, whatever that goal may be, and accomplish this without annoying their teammates so much that they are too distracted being angry at the “leader” to actually do anything worthwhile. A leader’s catchphrase should be “Follow me!” rather than “Do what I say.”
Naturally, the role of a leader should be established by consensus of the players, and by roleplaying the necessary interactions to result in the characters recognizing that one as their leader. A group of individuals with a common goal struggling to get past their differences and work together as a team makes for an classic subplot.
Furthermore, the leader does not necessarily have to be the “best” character: consider the titular characters from Kurosawa Akira’s Seven Samurai (or Shichinin no Samurai, in the original Japanese. It’s fifty-four years old, and it’s still my favorite movie. You should watch it). Kambei was clearly the leader, but he wasn’t the best swordsman—that was Kyuzo. Kambei was not as good an archer as Gorobei, either. No, Kambei did not get the girl—Katsushiro did (well, Katsushiro didn’t get the girl in the end, either, but the romance subplot was all his). And more than any of them, the most memorable character was Kikuchiyo, the farmer-turned-samurai with the BFS. But still, Kambei was their leader—the cool, calm, charismatic man who used his skills to lead all the other samurai and the villagers, as well, not for his own self-aggrandizement, but for the sake of their common goal. When I contemplate what it means to be a leader, I invariably think of Kambei.
Still, the work on the rules that will ultimately be packaged together and named the Mecha Monogatari Second Edition continues; there is still a long, long ways to go before it can be considered even a “beta” version. My goal is to make a nice neat PDF with all the rules clearly explained just so that it can sit out here on the ‘net and gather virtual dust anyone who is interested can take a look. Hopefully, I will be able to complete it before the year is out.