Coredump said:
I disagree. They *have* to be mutually exclusive to an extant.
Look at the example of Combat Expertise from above.
Right now it is very flexible. Every round you can choose to use it, or not. And every round, you can change whether to use +/-1,2,3,4, or 5. It can lead to a lot of calculating. But it is very flexible.
We can simplify it. And say it can be used or not, but is always a +/-3. There, we have added simplicity, taken away a lot of addition, and at the same time removed some of the flexibility.
Now, if you can provide an example for me of keeping the flexibility of having/using Combat Expertise, yet simplifying it.... then I am all ears.
The difficulty isn't having flexibility and simplicity, it's having flexibility, simplicity, and crunchiness.
Take this paraphrase of a rule from Over the Edge: if you do something that gives you a tactical advantage, you get one or more bonus dice. There: all the flexibility of D&d3E, and then some--anything you do has a meaningful mechanical result, the results are differentiated in magnitude, and i've just obviated the need for 25pp of rules.
Or, in our own Four Colors al Fresco, benefit and hindrance dice come in any size from d3 to d50, giving you at least 11 steps in each direction, and there's no reason you couldn't assign multiple of each type.
Or, for that matter, read a superhero comic. The characters have not only infinite options, but infinite shades of degree, and with "mechanics" no more complex than "what makes sense, and what are the consequences?" I'm not bringing that up to be flippant--there
are RPGs that basically play like that, relying on the players rather than tons of rules.
Now, trying to go for that sort of simplicity-and-flexibility under a numbers-heavy paradigm like D20 System--that's a real problem. It will require significant rewrites from the ground up to really pull it off, not just some tricks and techniques with the existing system. The suggestion to ditch bonus types and just allow all of them, or a certain number of them is a great example of this: it does simplify things, but requires a significant change in how things work.