As an experiment this afternoon and evening I am using the (short baseline) earth-electrode antenna. I am being copied by G0LRD (25km) and G3ZJO (79km) so far. It is now 1620z. Initial results suggest G0LRD is getting me very slightly better on the earth electrodes, whereas G3ZJO is getting me slightly weaker. This is comparing the earth-electrode "antenna" with the HF /VHF antennas with strapped feeders tuned against mains earth. In the latter case I resonate the antenna with a 110mm diameter coil with many taps.
One end of the earth-electrode is tied to mains earth. The "far" earth is a 1m long earth rod driven into the soil. The connection to the far electrode is 32 x 0.16mm PVC covered wire running along the fence at a height of about 2m. This wire dog-legs and is about 15m long. The spacing between the "far" electrode and mains ground in the shack is about 12m max. At the old QTH the baseline was more like 20m.
I think the earth-electrode antenna acts a bit like a loop so best results tend to be in the line of the loop. It also means there is not a huge difference between the two systems. You could say they are both equally bad, but the earth-electrode system needs no matching coil. I have optimised the resistive match using a 3C90 toroid (step up) between the transverter and earth-electrode antenna. At the old QTH it looked close to 50 ohms so the toroid was not needed.
When fitter, I'd like to try an earth-electrode system with a much bigger baseline.
There is no doubt that my 472kHz antennas can be much improved. The question is, "how seriously do I want to try?"