Well I finally got out of bed, thanks for the wake up call Dr. G, and can only comment on this subject based upon my old friend, Daniel Bernoulli, who's law , when interpreted for this situation, says that air that is at a higher velocity is at lower pressure than air that is at a lower velocity when flowing around a moving body. That being said we need to consider the air flow around a typical wing tank when in close proximity to the ground. The air stream hits the nose of the tank and splits, the air that goes over the top does accelerate in velocity because it is required to move upward to go around the tank, that air is not restricted from moving by anything except the atmosphere above it. The air that is going below the tank is restricted by the ground plane which means that it must accelerate to a higher velocity than the air going over the top therefore it is at a higher velocity and therefore a LOWER pressure i.e. you are generating down force.
Several things can happen to a tank related to ground clearance. If you have the tank very low this can cause the air flow along the centerline to begin to interfere with the boundary layer of air at the surface which can completely restrict air flow beneath the tank. This will typically cause the generation of large vorticees that can interrupt the hopefully attached air flow on the rear sections of the tanks lower half and cause separation of the air flow in this area which will increase greatly drag. As Dr Goggles says I do not like flat bottoms as they are great generators of vorticees all of which generally increase drag. The minimum ground clearance for various shapes as published by Katz, Tamia, McBeath et all are not something that they just pulled out of their collective a$$es, they are based upon both mathematical and experimental data that is proven and reliable. For a "torpedo" shape (typical tank) the minimum is 3 to 5% of the total length of the shape. i.e. if your tank is 100 inches long then the minimum ground clearance should be between 3 to 5 inches.
There certainly are flat bottom cars that are fast, witness Poteet and Main, Bobby Moore, Jack Costello's cars but there are also some round bottom cars that are pretty fast, Seth Hammond, Al Teague, Charles Nearberg, Dennis Varni, the Contrivance Engineering modified roadster. As far as I am concerned the big advantage of a flat bottom car is that they are alot easier to build.
Rex