there was a thread started in here a few weeks back that linked over to another forum that was talking about header design and building a custom header. It didn't directly answer your question, because there are so many variables, but it did discuss the answer to your question, and even went into details as to why the computer programs aren't 100% right either. dig through the posts, maybe even a page or two back, and find it. I seem to recall it being titled something obvious.
header design really is a lost art... and subtle differences between two otherwise identical headers can make a notable difference when it comes down to the resulting power.
The thing that I believe Chronik Fab (or at least hope) was trying to say is that unless you're trying to break some record or racing competitively with a huge budget or trying at least trying to squeeze every last bit of power out of your race car engine, then it isn't worth trying to design and build the "perfect" header.
While I did say that those subtle differences make a notable difference, we are talking less than 2 hp... or maybe a few hundred rpms difference in the powerband. Something that the average car guy, or hell, even the above average purpose built car guy would notice.
the saying "you spend 10% of your money making 90% of your power and 90% of your money of the remaining 10%" is very much true when it comes to header design.
So, is that extra 1whp worth the trial and error it is going to take to build the header that produces it?