Marcus,
I think the 'right' answer would be "both", sort of, since what you
are doing is looking for the maximum signal strength. Since the input
impedance of a 1/4 wave radiator isn't 50 ohms to start with, you are
also doing some impedance matching by length of radiator adjustment.
You seem to be getting the most benefit (signal strength) from using
'different' measuring devices than 'normal', but it's working, right?
And when you think about it, that's what most people do when they
'tune' an antenna, just different measuring devices. 'Resonance' and
impedance matching trade off, sort of. Is that the 'best' possible
way to do it? Mmmm, no. Is it a practical way to do it? Mmmm, yes,
sort of, if you've got the required devices.
- 'Doc

(Lots of 'wiggle-room' in that answer, not very satisfying huh? Oh
well...)