That's where I messed up, if I had known then what I know now I would have tried swapping the pickup leads first. Or... are you saying the rotor-to-cap phasing shouldn't have been off even with the leads reversed?
Are you thinking of trying this combination?
1. original reluctor position
2. original phasing
3. set box to back to falling edge
4. reverse the polarity that is currently at the pickup coil
5. set timing
Another way to check rotor phasing is to sacrifice a distributor cap by cutting, drilling and or die grinding a wndow in the cap. This will tell you if the phasing between the rotor and cap is correct. I prefer to use #1 cylinder (because I won't have to change the timing light pickup to set engine timing) but if that doesn't give you a clear line of site any cylinder will do. Also some times a bright mark placed below the rotor tip will make it easier to see.
I have a few windowed caps on hand at the dyno for when a quick check is needed. Just clip the timing light to the plug wire that is windowed and shoot it at the rotor through the window while the motor is running. You can observe the rotor position at different rpms. With the vac advance unpluged, the rotor should stay steady and lined up with the contact on the cap. It might be slightly to one side but when the advance is plugged in and activated it should pull it closer to the center of the contact and maybe even sweep to the opposite side of the contact when full vac advance is applied. Good luck! I hope you find your problem.
Ps: Pay no attention to the epoxy holding the cap together in the picture. It was new cap damaged in shipping and me being cheap (to the point of impractical) spent enough time and money on epoxy mending it back together that I should have just bought a new cap. Here's what happened. When I told my buddy I was thinking of trying to epoxy this cap back together to make a cap for testing phasing, he said " that will never work" . So....