Victory Conditions

How do I see myself succeeding at this game design game? I'd like to create something playable; something fun. I see three main ways this might be achieved, three kinds of games, or more accurately three kinds of game designs, that might forge that path.

1. The real game. The traditional-sized, fairly complex game that doesn't reinvent the hobby, but that I just happen to make better than average through some effective designing. This is the hardest to accomplish, the hardest to prototype, and the most daunting to playtest.
My Examples: Pirate coop, nearly every early game I designed
Real-World Successes: Puerto Rico, Arkham Horror, most games.

2. The simple game. Sure its not the best game you ever played, but its an illustration of a clever mechanic or interaction, and it was obviously designed as a "small" game. Pet games might use cheap components, or existing components like traditional dice or cards, piecepacks, or minimal custom decks. Can seem unexciting, but are far more feasible to prototype and play.
My Examples: Several designs, tellingly though, few that get talked about here.
Real-World Successes: Many public domain games, many cheapass games, card games such as 6-nimmt and rage, For Sale, many smaller Knizia games.

3. The innovative game. Take a simple idea and make it fly, providing something really new, without necessarily recombining things that you'd seen in the past. This best explained through:
Real World Successes: Apples to Apples was able to go with nothing but a decko f words and like 3 rules to create gameplay. Tales of the Arabian Nights brought the book of tales concept to its full height (and most of the rules they added afterwards really weren't necessary). Finstere Flure built a simple set of rules around the idea of an autonomous monster, Roborally added some (too many, perhaps) rules around the basic notion of prealigning moves with a card each. Magic obviously blew things up with its collectible notion (though the game itself is one of the most complex around if you really get down into it).

The real game is the hardest to create, the hardest to work with, but its easy to slip into. I often start with something simple or innovative, and I really like it. But it doesn't quite work, and I can't quite get it to come together, but I liked the original idea enough that I can't really let go of it. Soon I've painted myself into a corner, and don't know it, or cling onto the idea anyway. Making a real game work requires, as I've described recently, resources and/or bravery that I currently don't have, and rather than fight it, I'd like to satisfy myself with smaller designs for a while. So, how to proceed.

1) Accept the growth and go ahead and try to make the "real" designs that start to emerge work.
2) Show more discipline in keeping ideas simple. Easier said than done, but something I'm working on.
3) Generate more simple ideas, creating a large number of them, and not getting too attached to any one.

This last one is where I'm going with all this. It seems a little unintuitive, essentially a strategy of quitting. To explain: I think its far, far too easy in design in general, and game design in particular, to get hung up on a given idea, work out some additional rules/constraints/decisions to make it work, get locked into those ideas, and find yourself in a failing state in the space of possible designs, without the will/wisdom/wherewithall to know how to salvage what was working. In fact, in the midst of working on a given idea, I'd say its nearly impossible to even throw away everything other than the core idea and try a different direction on it, let alone retreat to a more complete, later, partially successful state.

But that idea was good - what's needed is a way to keep it, but to get a clean conceptual start on it. I wonder if rather than hashing out a single idea, it would be wiser to create a large number of ideas, take a stab at working out each, but then log them, move on, and use the time working on newer ideas as a palette cleansing period to take on those that have fallen by the side.

Perhaps I'm not succeeding in explaining the motivations behind this process, let alone the process itself. But I want to find a way to:
1) ride that enthusiasm of an idea, which is fun
2) explore it a bit
3) but not keep pounding on it until its broken
4) find away to return to it later

In short, going a bit more breadth first with my design process for a bit - increasing my chances of keeping the core good ideas straight and sticking to those simple and innovative games without letting the need to make them work distort them. This might involve a more rigorous approach to my sketchbooks and filing, or might emply this blog somehow. We'll see.