Wow - That brings back some memories
...
There is a flow bench at work, but would have to see if they'll let me use it. Probably flow in from each side and measure the flow of each individual runner to prove/disprove this theory. I think RyanP mentioned this idea at some point.
My opinion (still) is that the two dyno runs were scaled differently somehow - maybe an RPM difference, one was strapped down tighter than the other so more tire rolling resistance, etc. The more/less vertical offset of the two curves seems to support this theory (but I may be just drawing circles around the dart on the wall too
). General theory says that a component like this won't have an overall effect across such a wide powerband. It might help the bottom end, but hurt top end, etc.
In the end it's a singular pair of data points with insufficient data and uncontrolled variables to draw any meaningful conclusions.