I hadnt considered the venturi effect caused from the gas pulses of each pipe passing into the collector...yet one more thing to toss into the variables mix lol >.<

Needless to say, you are correct in assuming the only way to get a flawless header design is by trial and error. There are entirely too many things to consider for it to be cost effective.

The purpose behind my illustration was to show the intricacy and complexity of determining exactly and precisely the differences made by the bend itself. Ambient temperatures may not make a noticeable difference in power increase on the engine, however every 0.0001 bar or cfm of decrease should be accurately accounted for in determining exact differences if you're going to go so far as to concern yourself with the bend ID's impact on overall flow in a mandrel bent header. To be completely accurate it would need to be checked repeatedly in a controlled environment with precision equipment. Again, all in all...it's really just not worth it...this is the sorta stuff those guys who are busying themselves creating extreme output ceramic 4 cylinder engines worry themselves with.

And the rearward swept angle of the pipes on drag engines remain exposed to passing air as the car moves forward...creating the same venturi effect as the collector pipe...only much more dramatic. Anything that spoils airflow creates a small vacuum behind it...kinda like when you draft a semi on interstate to conserve on fuel.