As a software developer and game programmer for many years, it is my opinion that the best tools for designing a new game are felt tip pens, some cardboard, and a pair of scissors.
If your game is designed and play tested to the point where only some fine-tuning is being considered, then start thinking about software.
If your game fits one of a few very specific moulds, there are tools that can get a prototype flying with relatively small effort. Perhaps the best example is ZOG for standard board games. It's really pretty amazing how many games have been implemented using ZOG. Even there you're talking about a pretty serious effort by a programmer familiar with the tool. I'm sure there are some other examples of highly evolved tools for specific types of games, but most of them will be in-house tools not publicly available products. (Ie; The people who do Sim-xx have great tools for whipping out another sim game.)
However, no mater how you slice it, thse aren't tools that are appropriate for much experimentation. To use them, you have to incur costs in time and either money or your own effort that are a substantal part of the effort to produce a finished product.
If your game doesn't fit into one of these ZOG-like moulds, you're even further back. I can guarantee that anyone who has ever programmed any game immdiately starts thinking about creating generic tools to do more of the same kind of games, but it's a really, really hard thing to actually do. My best advice would be to lay hands on the sources for any game with anything like a similar structure, and start there.