I thought I'd talk a bit today about the process I tend to use for ideas getting whittled down.
Step 0 is I have an idea in my head.
We all have those. Depending on caffeine intake, we sometimes have a lot.
Get three gamers together and get them talking and you will have 7 game pitches half an hour later.
This stage tends to deal in broad concepts and open-ended ideas.
"A robot skirmish game where your robot might malfunction and do random things!"
90% of the ideas from Step 0 never go anywhere. At this stage we're just dreaming and dreaming is easy (and fun!)
Why does an idea fail to progress from here? Usually once you think about it a bit, there's either an existing game that already does it well enough or the idea, after a bit of time to think about it, doesn't have enough legs to stand on.
I've said before that a good game is usually a workhorse system with 1 cool idea. But if the idea isn't cool enough, then it's just one more set of rules in a world that definitely doesn't have a shortage of game rules.
Step 1 is the notebook stage.
At this stage we start writing things down. This isn't writing a full game (usually).
For me, it's a mixture of writing out concepts ("Robot Morality!"), a list of desired content ("20 robot morality modules") and fragments of rules written out.
Churn out 2-4 pages of hand-written notes at this stage. You could be a modern person and use a computer but I sometimes tend to "think better" with pencil and paper.
80-90% of the ideas that make this stage never progress beyond it.
Trying to actually put an idea down on paper has a brutal tendency to reveal that it's all gibberish.
Maybe the mechanics you write out just aren't that great and the "right" dice roll or engine keeps eluding you.
Maybe the passion just runs out at this stage.
Step 2 is the draft stage.
Turn to your computer (or a bigger, nicer notepad I guess?) and start writing an actual, playable game.
At this stage, you are writing actual rules, translating the ideas from step 1 and your head into something a group could sit down and play.
Include the basic concepts needed to play: This version needs to account for common, obvious questions ("What happens if I shoot at a tank?") but it can (and should) leave out a lot of the chrome. Campaign rules, your D1000 table for Situational Robot Morality Quirks, the rules for off-map artillery fire, that stuff can all be left to the side.
What you are hoping to end up with is a set of rules that are playable, at least with your own group. Page count will mainly depend on how complex your project is. I'd say 20-30 pages is a good middle ground for a typical game but don't sweat it exactly.
Once you are at this stage? The failure rate will depend primarily on experience.
You see, at this point you have what a lot of people are perfectly happy with: A game that broadly works, that you can tinker with and which your group can play.
Heck, for groups with a fairly open-ended play style, this may be all you ever need. The GM or group fills in any blanks when they show up.
So progressing to the next stage is not always a requirement (or even desirable).
But if you want to publish this thing, then we gotta get over the hump to Step 3 and that is where things get gnarly.
If you are inexperienced, you will run into the fact that you may not know what to look out for: The game will have rules that are worded poorly or ambiguously. An important concept may never be explained because it was obvious to everyone at the table.
Maybe you really hate writing out terrain rules and now you have to account for all of that.
Filling out a game from Step 2 to a stage where you can publish is often 50-80% writing crap you don't want to write, because someone else will need it. I hate writing terrain rules, but they have to be in there because some guy in Virginia is not going to know how I handle it at my own table.
What about that section everyone skips that talks about in-game scale and what each figure represents? Anyone enjoy writing those?
I tend to call this "The Suck" because its writing things that aren't that much fun, but are important.
So between The Suck and all the things you don't know yet, there's a high chance your game stops here. In fact, take a look at almost any project on forums and message boards and this is where they died.
When I did research before doing Renegade Scout, I found 6 or 7 attempts at reviving / cloning / redoing Rogue Trader. They all died here.
Am I amazing at writing then, since I made it?
No, but experience counts for a lot here. If you have a mental check list of obvious things people will ask about (because they asked you about them last time), you know where things should go and you have a sense of how to phrase things, you can get to a point where only 10-20% of designs fail at this stage.
How do you get those? By doing it and screwing it up. Or by doing it and doing okay. Or maybe your first game is great! Either way, there's no substitute for actually putting in time at the keyboard.
The good news is that while talent matters, a systematic sense of what to look out for is something anyone can develop with experience.
Step 3 is the Actual game
Okay, so you've tested your game, badgering someone else into testing it too, you've fixed all the stuff that wasn't explained well, tweaked the mechanic that was obviously broken, adjusted the thing that made no sense.
You've added your chrome and fluff text and found some stuff to put in there to look pretty (or go for the 80's typewriter look).
Congratulations. You have a game to inflict upon the world.
Pour one out for the 100 games that died along the way so that "Laser Robot Feels Bad" can fly to the top of the Wargame Vault best sellers list.
If you have questions, let me know.
PS: Someone should definitely write this game.