As was said, soaking the soil around the rods doesn't help much in drought times. The soil around the rod becomes conductive, but 20' away from the wet area into dryness and the resistance to the pulse's current flow skyrockets up near infinity.
I'm sure you know this, but the shock doesn't take place until that pulse makes it back (at near light speed) to the negative terminal on the circuit board of the charger.
Lots of people think all that has to happen is the pulse has to go thru the animal and into ground, but not true if moisture isn't present in the soil.
The soil, from the point where the animal touches the hot wire, all the way back to the rod itself, and up the charger's clip-on wire to it's negative terminal on the circuit board all make up 1/2 the circuit--the return pulse circuit..the animal is also part of the return circuit. (Actually, it's the moisture in the soil that makes up most of the return circuit)
If you check it in the early morning when a heavy dew is on the ground, don't be surprised to find the system is working right, but by mid day, when the surface moisture is gone from the grass, there is nothing to conduct the return pulse.
Adding a 2nd or 3rd wire all the way along the fence tied only to the charger's negative terminal takes the dry soil out of the circuit, and takes the place of the soil as a return path to the charger's neg side. The animal does have to touch both the hot wire and the ground wire tho.
If you do it right, when rain returns, the system will work as a conventional system, with the wet soil again becoming part of the return circuit and the animal only has to touch the hot wire. At the bottom of this post is but one of several ways to do it.
I have one area, that no matter how dry it gets, always works on a conventional set up. That's because it's red clay--remnants of iron ore all the way to surface. It conducts that return pulse wet or dry..the iron particles are the conductor..
