For closure, my solution to this was to use a very long wire and make my voltage measurement connection as tight as possible in the highest current carrying area (the underside of a wire twist). I also used a nice fluke multimeter to measure the current but it didnt seem to be varying much (with my long wire the voltage drop was much higher so maybe having a larger load helped the PSU).
The calculated resistance seemed quite reliable to determine temperature rise.
The biggest issue I had though was that the temperature coefficient of resistance in copper its fairly different from different sources, ranging from 0.38% to 0.42%