RAH RAH WALL OF TEXT
Concept: Type Representation
I know you said not to look at typing, but I believe this 'experiment', if you will, could possibly be good for an outlook on the metagame. Plus I believe it could be both fun and challenging for a new way to build a Pokemon. The idea is simple: there are types that are underrepresented in OU. We take two of these types (it doesn't have to be a completely new and original type either, because I actually tried those combinations and there's basically nothing good) and figure out a way to increase the representation of these types in the current metagame. Now this may sound easy, but it's more than likely going to be hard as ****. The reason for this is that in some cases, the type in question is currently only known for doing one or two things good (Poisons for Toxic Spikes for instance), whereas it kinda sucks in a lot of other ways. The problem is, you can't go the 'normal' route of making this Pokemon, because if its niche overlaps too much with others of the same type, then it could easily drive them out (or even worse, the new Pokemon could be outclassed and not used at all) and the representation value could become worse. So yeah, I guess that's it for my explanation. Now to explain what the hell this representation value nonsense is.
My calculation for representation is based on the average usage percentage of a specific type in the OU metagame (based on Shoddy Statistics December of '09), multiplied by a "rep. value", which is simply the amount of Pokemon of each type that sits in OU divided by the amount of fully evolved Pokemon in each type that exist overall (with the exception of Ubers). For instance, there are 8 Water-types in OU and there are 47 overall. The representation number then, is .170. This means that 17% of all Water-types sit somewhere in OU. You then take that number and multiply it by the average usage % of a Water-type in OU (9.875) and bam! You get 1.678325. As for what the number means, well, I haven't exactly figured that out yet, because I haven't exactly figured out an equation that tells what a perfect representation value should be. My view is just that if it's above 2 (or 2.5...I haven't decided yet. I like 2 more though because it makes this more challenging), it's represented enough. You'll see why I say that when you look at the values. And yes I know this number is partially flawed (8 Waters in OU is considered underrepresented...kinda weird), but I think that for the most part the values still ring true. So yeah, here's the list.
1. Dragon (10.769024)
2. Steel (7.518585)
3. Ghost (3.65274)
4. Fighting (3.369294)
5. Electric (2.805456)
6. Psychic (2.700405)
7. Ground (2.611386)
8. Fire (2.17125)
9. Flying (2.1464142)
10. Bug (2.0609)
11. Water (1.678325)
12. Dark (1.6325)
13. Rock (1.3135)
14. Poison (1.22766)
15. Grass (.874888)
16. Ice (.80095)
17. Normal (.7197125)
Also, if someone finds a better and more accurate method (or figures out a prefect representation value to base the statistics off of), then please feel completely free to post it and explain how it works. I'd be more than happy to use that if it's more accurate.
...and sorry if this is kind of badly explained or whatever (I honestly don't know if it is or isn't). It's 3:45 in the morning and I'm dead tired. ;-;