In the brief play with numbers before I started writing formulas, I found that caliber alone accounts for about 85% of optimum bullet weight. Mostly because I wanted to keep SD over .24 (English units)

In the calculations I had a threshold value of .23 SD.
Then using case capacity, bore volume (assuming equal length barrels), cross sectional area, and potential energy, I calculated the additional bullet weight to achieve optimum momentum. This tends to select for bullets with lower ballistics coefficients than the max recommended but does allow for better delivery of energy throughout the range of efficacy.

There are some errors in my math up there (typos of actual caliber values.) I have fixed them since.

I don't think there will ever be a truly good index of bullet performance for the very reasons that you list especially given that there is so much variety in bullets and infinite variety in bullet placement.

However, given the same test media and using cast lead bullets, it is an excellent indicator of penetration depth and channel diameter.

Last edited by Aileinduinn; 07/10/09.

...new laws were most numerous when the commonwealth was most corrupt. ~ Publius Cornelius Tacitus