Shape is only one factor in determining ballistic coefficient.
The sectional density also matters
Sectional density is the frontal area divided by the mass of the bullet. For a fixed shape
That number increases with the square of the diameter divided by the mass of the bullet which increaases with the cube of the bullet's diameter.
Even for a fixed size and shape the sectional density increases with the density of the material the bullet is made of. An aluminum bullet of a fixed size and shape has a lower sectional density and a lower BC than a bullet made of lead and that's lower than one made of tungsten or DU.
For a mathematical description of the relation of bullet ballistic coefficeents, sectional density, drag functions and effects of bullet shape I reacommend "Modern Exterior Ballistics" by Robert F McCoy of BRL (US Army Ballistics Research Lab, ISBN 0-7643-0720-7. Its very interesting to read even if you're not up to the calculus and trigonometry used thoughout. It has excellent Schlieren photos of the shockwaves which form around various bullet shapes at transonic velocities which gives good insight to why G1 shaped bullet fly differently than the G7s . The book also has the math describing bullet stability and wind deflection.
Robert McCoys computer programs McTraj, McDrag, and McGyro contain the basic math used by just about all popular computer ballistic programs used today. It also describe six degree of freedom bullet models and the effects that it precticts which the point mass models don't prectict aat all. It also shows why the 6DOF modes are not used for practical shooting. The book may cause headaches trying to follow it and if you do it's likely to cause you to loose faith in the ballistics programs you presently use. At least it will give shooters a better understanding why their bullet's path doesn't match their PDA or or cell phone ballistics program exactly.