I've seen where two carbs that made virtually the same power on the dyno still ran .2 & 2 MPH differently on the car. Whatever looks like the most worth testing based on the dyno results will still get flogged on the car to prove it.
And that is what I was wondering. If the criteria isn't clear going into your test, you may pick the wrong carbs to test. To me, it would seem that the best criteria would be area under the curve, in the RPM range encountered during the run; the right carb for a narrow range is likely different than for a wider range. Not knowing any of those details, my vote would be for 1, 4, and 6B.
