There's a difference between backpressure and velocity.
High velocity gas flow is what you want; this creates scavenging effect. Many people confuse this with back pressure. It's not the same. Back pressure means work is required to push out those gasses, while scavenging means the gas is moving fast enough that it's pulling itself out(means less work needed to pull out the exhaust gasses, thus more hp getting to the wheels).
I'm sorry, pulling itself out ? Free energy ? Think about this statement. You are correct: backpressure and velocity are different, the two aren't mutually exclusive . With too much backpressure, scavenging exhaust isn't a matter of "more work", fluid dynamics applies: It just doesn't move. More work or not, it doesn't move. And what is proving this "work" anyway ? It's just a higher area of pressure than surrounding areas, which would tell us it would RUSH to the exit of the open system - the exhaust tip. So by this argument, the backpressure argument wins.
And it won for like 60 years.
What keeps the exhaust pressures higher in a pipe isn't volume, it's wavelength. The compression of the sound (pressure level sound, not audible sound) creates troughs and compressions of medium (Exhaust gases). When the valve opens and closes, it releases a burst of material and thus a pressure wave that moves through the pipe. With that burst comes air pressure, fine particulate and sound - or "pulse". By measuring this pulse you can get a waveform, complete with frequency and amplitude measurements. That frequency can be measured in inches. If the amplitude is below one inch, then a pipe over 1 inch will change it (As it "bounces" off the side of the pipe... horrible explanation, but it's early in the morning)
The plates in the Supertrapp are spaced very closely not to "create backpressure" but to bounce the wavelength back down the pipe. When you bring the pulse fully out of phase, it will flow wonderfully, but the valve will see a out-of-sync pulse when it produces the originating pulse. If the reflected pulse (reflected by the end cap on a Trapp) is brought back to the valve in sync, the valve will produce a pulse that is more scavenging. So it's a matter of velocity (Out of phase) or scavenging (in phase at the valve)
The Supertrapp was invented by Paul Moller, inventor of the Moller Skycar, or his Freedom wankel engines. The supertrapp was meant to mate to the wankel, where instead of Valves producing the pulse, the passing apex seal does. By keeping the exhaust pulse "in pulse" the exhaust doesn't get pushed back into the passing exhaust stroke chamber.
And yes, we have ran an open header. Worked very well actually; RMF "big tube" header with megaphone.
Too loud though.
You can do lots of things with an open header. The hotrod guys have known this, but we are severely hampered by under-hood space. If we could bring a pipe out of our heads and extend it 12 inches forward, before any bend - We too would see amazing differences in torque. Lots of Neon guys take advantage of their clearance at the firewall with an almost STRAIGHT header pipe. It's a thing of beauty.
Once you add bends to a pipe, your valve needs to work against the pulses. If the pipe is straight, then the pulses will leave the pipe regardless of any muffler.