Game design theory gets a terrible reputation. Between the fights between rules light and rules heavy, lethal and forgiving, and a number of binary debates. Some people don't know enough to care, some care too much. In the end, it leaves the number of solid discussions on the topic I've had to a managable dozen or two.
Those who dislike theory seem to like chanting 'fun is more important than anything' missing that bad design saps fun from a system easier than anything else. In a discussion on this very topic, discussing the balance of Fun, Simplicity, Balance, and Depth, in which one person presented Fun > Balance > Depth > Simplicity, Something Awful user Angry Diplomat said:
"A more sensible way to present it would be a Venn diagram with three circles marked 'Depth', 'Simplicity' and 'Balance' and the bit where they all overlap is marked 'Fun.'"
This sums up my theory on game design perfectly.
Complexity, in general, is the enemy of fun. The more complex it is to get into the less people play, the less fun it is over all. But without depth, simple fun fades quickly. With depth, there's always something new to explore. And balance is its own issue. An unbalanced game is bad since the people who pick the 'better' choice is going to be automatically better and have more fun than the others.
So my goal is simple: Simplicity + Depth + Balance. The game I intend to make should be all of these things.