I don't think I've ever written about how I design a game, except in general terms. I don't know of many RPG projects in general that really have running documentation from start to finish. There are a lot of decisions and personal biases that come into play when making an RPG, and even ones in development that include a detailed changelog (like Pokemon Tabletop Adventures) rarely take time to explain why every decision is made. And there are a lot of decisions to make!
As you may or may not know, I've been working pretty hard (relative to how much free time I have, which is little) on a game called Warhawks. It's a mecha combat roleplaying game, and lately I've been focusing almost entirely on the part without the robots at all. And that's the part I'm going to try and talk about today, the decisions that have gone into shaping what it looks like now.
I apologize in advance if I ramble or repeat myself. I just felt like explaining some things.
Now, the first decision I had to make was what basic system to use. There are a lot to choose from. However, I did have one big guide already - I already had a system that I was using for the mecha themselves, and as part of the same overarching system, it made sense to have something with the same basic framework. Four stats, four dice per turn, rolling against a TN.
I clearly couldn't use the same four stats, or the same rules. A human isn't a machine, and doesn't have 'maneuverability' or 'systems'. They also certainly don't take damage in the same way. So I had to pick four stats that could pretty effectively describe a person. After some thought, I chose Physique to cover pretty much everything physical, Intellect to cover their brains, Awareness for their senses and common sense, and Empathy for all their social skills. It actually made for a pretty good spread, and I haven't modified those four stats since.
As for the rules, well, a human needs to be smaller than a mecha. And the best way to represent this in a physical way is to use a different die size. Where mecha use a d10, humans can use a d6. It's another common die type, and about half the size of the d10, so it's a significant difference.
One of the primary things we want to establish at this point is the /goal/ that we're aiming for. I know it's a bit late in the process to START naming a goal, but you can see my priorities here - first, whatever we come up with needs to fit with the already-designed mecha elements of Warhawks. That's goal number one, to make the two parts of the game feel like they're part of a whole.
The second goal, as I've established, is to keep it fairly simple and light. Mecha combat can be more detailed, but there's less of a focus on people and so we don't want to have to think about it too much. A third goal here is to prototype some changes I've been considering on the mecha rules. Playtesting shows that there are some problems with initiative. At the moment, initiative looks like this:
That's essentially two initiative passes. It's easy to see it takes twice as long to resolve a combat round. Now that's not to say it's bad by itself - in a lot of ways it's quite nice and tactical. But it did require some workarounds, like being able to change your action during resolution.
One of the problems I ran into, aside from each round taking twice as long as a normal initiative method, was that you essentially had to try and remember what everyone else was doing. If you were on the top of the initiative order, where your action would be resolved immediately, you'd have to remember everything everyone had done that round while deciding on your action - before they had actually done it! It was too much of a memory game.
So, since I'm prototyping new things for the mecha game here, one of the first and most drastic changes I had to make was to eliminate the reverse declaration initiative. While great in theory, it's just too slow and cumbersome in practice.
I am very reluctant to change something in a game that distinguishes it from what I usually call the "basic game formula" - something I define as being like DnD. I lived and gamed through the d20 dark ages where every third party company was putting out shit (and I do mean shit) for d20. After seeing dozens of games that used the same system, it's hard not to think of it as being a standard. And d20 was the first system a lot of people used.
With this initiative mess, though, we need to keep things simple - remember that's our second goal - so we're just going to have to go ahead and use the default initiative system. Bigger numbers are better, go down the list and let everyone take a turn. Not original at all, but what can you do?
The following obvious question is "what do you roll, and is anything added to it?" And that's a damn good question.
The two choices here are one die, and all the dice. In this case, I have a few reasons for preferring all the dice at once - it makes sure no one rolls too many dice or forgets one, it gives way more variance than a single d6, and it gives high initiative people a better idea of who is most vulnerable (protip: the low initiative people are the most vulnerable).
And as for what to add, no single stat really covers the variation in initiative - is it reflexes, situational awareness, a quick mind, a feeling of what others will do? So we'll add none of them. 4d6 gives plenty of variation on its own.
And with all of this, we've so far managed to decide what order people go in (a major change to the system) and what stats to use.
That's a start, but far from finished. We need a way to personalize the character more. So next we need a skill system. I don't want to go with some complicated system or make people add more bonuses or anything. I want a light system here. The less thinking and less referring to the rulebook the better. I see most rolls as just being 1d6+Stat.
Instead of having skills add to that, let's make them more important. We can have them matter a lot without increasing the total range of results (as an example, since we're looking at 1d6+1-5, we have a range from 2-11) by having the skill simply skip the rolling step and give you a natural 6. But, of course, that would mean dumb luck could get you the same result. So we'll instead make it a natural /7/. It's one higher.
We can't have them do that all the time, though, so we'll limit them to using the max roll ability a number of times per session equal to the skill's rating. Giving skills a rating here will let people display more or less proficiency and also give people something else they can improve with experience.
We'll also make allowances for a long list of possible character abilities which we'll let people get a la carte. The key thing here is that none of them will affect mecha combat later - if we give people abilities that will affect that, they'll always take it over any other power. It's not even a question. So if we keep things separate, people don't have to choose. We don't need to detail them yet - and the process of making the list is basically just sitting and brainstorming a long, well, list.
Character creation we also want to keep simple. At the very start of the process, I was considering a type of lifepath system. A lifepath system has some big benefits - it gives you an idea of where your character is from, what he's done, and lets you tie it all together. But it's also somewhat constraining, and one overpowered selection can seriously warp things.
We'll cut it (for now at least, I might revisit it later) and go with something easier. Six points of stats to distribute, limit people with a max of +3 (for now). We'll make a list of skills for each stat, and let people choose a number from each stat equal to that stat's bonus (so with +3 physique, you choose 3 physique skills). And we'll let people choose, say, two Abilities from that list we came up with.
There's just one step left - gear. Almost immediately on getting to this development step I realized I really don't want to write up a huge detailed equipment list. Especially with a modern game, there are a /lot/ of possible guns and such, and that sort of goes against the quick and simple thing we're going for.
I'm going to take inspiration from another game here. Legend of the Five Rings has a fairly elegant system for starting gear. Characters start with a few things and a Travel Pack. For this travel pack they just basically list a bunch of stuff and say "Pick ten of these". The items don't need rules or even descriptions - everyone knows what a hat or perfume or playing cards are, after all.
So I decide what I'm going to do here is something similar. I'll give everyone some Requisition Points for gear and let them just take what they want, with most items just taking one point. Guns are a special case here. Since guns /do/ need stats and I don't want to stat them out, I'm going to make the players do it for me.
Next time, I'll explain what I decided to do with guns and thoughts on combat.