Bends are one factor among many. Gas flow is a complicated science. A well R&D'd system with a couple more bends can easily produce more peak HP than a straighter system that is poorly designed in other respects. I won't even go so far as to say that similar systems for the same bike from the same manufacturer will have an obvious correlation between "total bendage" and horsepower. It depends on what you are designing the system for; top end above all else, or smooth, broad power, or low-end grunt.
The fact that the potential adverse affects of bends are easy to understand does not mean that the number of bends is the central characteristic by which all exhaust systems should be judged. There's a lot more to it than that.
The velocity of gasses in an exhaust system would be governed by the Euler equations, shown at
http://en.wikipedia.org/wiki/Euler_equations. But outside the combustion chamber, the pressures are probably constant enough that we can treat the gasses as incompressible. That means the Navier-Stokes equations are the best for predicting the motion of the gas. You can take a look at them at
http://en.wikipedia.org/wiki/Navier-Stokes_equations. Of course, before you calculate the velocity of the gas in each segment of the exhaust, you need to decide what velocity will give the best performance across the rev range of the engine. Oh, and it's not steady flow in the header; it's pulsed flow.
Now, if you can do all this in your head and decide at a glance which exhaust will give better HP, I'm sure there are a lot of companies that would be happy to save on R&D expenses by simplying hiring you to eyeball their developmental models! :-D
kit:
Linear Mode
