DAN:
Your graphs drive me nuts, since You have watts / RPM and no indication of the generated voltage, at all.
Generated voltage / RPM @ no load is a better indication of the mill power if the winding resistance is known and the GAP Voltage stator - rotor ( at peak RPM) is defined ( which is done by lightly loading the generator - around 10 % of peak power -).
Then a power curve with clamped output voltage as You show them now.
Even if battery clamped, the output voltage should be taken, as well, because the voltage is not "fixed", it may vary several %'s
Maybe if you get a MPPT ballast controller for testing the mill, generating all the parameters including the efficiency.
I have a good design but not the man hours to build it, at the present time.
I understand that You want to reduce the electronics since it is an unknown for You, with electronics, the generator can have a very close AIR-GAP and the winding can be made to produce higher voltage for much better power harvesting with MPPT.
A 1 KW wind mill used with a battery bank "clamp" producing 50 % efficiency, with a MPPT charge controller, the same wind mill will be able to produce around 1700 watts and possible more if windings are optimized for MPPT harvesting.
Regards
Nando