Hmmm...I'm going to do some digging around and see if I cant find a rough equation for this...I'm going to have to assume that the only way to accurately guage this would be to put a bar and cfm gauge at both the inlet and outlet of the pipe to determine how much the barometric pressure changes as the exhaust gasses pass through the header design along with the total cfm of exhaust gasses the engine will put out if flowing freely versus how much it is putting out after passing through the header...as KJ said though...the less bends the better...but I'll work on finding out how much difference it would make if you can tell me the pipes ID, and the total number and degree of bends you plan to put into the pipe...I would most likely assume that after the first bend the amount of cfm flow reduction would be a diminishing return as the first bend would dampen the exhaust gas flow down to the maximum allowable amount through out the rest of the exhaust manifold...however...that's not entirely ruling out the idea that the flow could pick up velocity again in the straights. But seeing as how there is nothing to provide additional inertia behind the slowed gas flow aside from the same slowed gas flow, I cant imagine this being the case.

Another thing to consider in this equation is that forcing super heated gasses downward is also sure to slow it's flow...knowing the downward slopes of the header design would also be helpful in determining flow capability...along with the approximate EGT throughout your powerband...AND the ambient temperature of the area you are running the car in (to determine total thermal resistance)in other words...there are A LOT of variables.

Ultimately, the best design to evacuate exhaust gasses would probably be to have straight pipes angled squarely away from the engine at a 40 to 45 degrees upward angle...but I doubt seriously you want your car to look like the dragula car from the rob zombie video lol.

Something like this lol.