This is an early stage of the experiment, but
I believe that there is a lot to say with the lower noise on the
horizontal antenna station

Probably has a lot to do with the particular sites
though. It's quite possible to be near a noise source
that is mainly vertical polarized. In a case like that, it's
possible it could be a problem. But I never saw the difference
in noise levels you are seeing. At the worst, I might see
appx 2 S units, but sometimes it might only be one, or
even other times , nearly no difference at all. Most of
the noise I would see at this location is power line noise.
It seems to effect both horizontal and vertical nearly equally.
Probably cuz much is radiated by horizontal power lines.
I've never tested it, but I think if you are in a noiseless location,
the difference would be fairly small as far as meter readings
just measuring the average atmospheric noise. The reason I
say this is because sometimes I would see little difference in
noise between the two. But other times I might see more.
But you could see small differences just from the increase in
strength of dx signals. IE: if you had T-storms 1500 miles away,
it's quite likely the vertical will receive them stronger than the
horizontal due to the normal operation of the antennas.
Anyway, I don't totally consider what you see as the norm. "4 s units"
You probably have a local vertical noise source nearby. If it's power
line, etc, you might be able to track it down and get it fixed.
I'd be curious to see if you see the same 4 S unit noise difference
over a period of time. Like I say, mine would vary. But noise
never was much of a concern on mine. Never gave it much thought
at all. Kinda weird too being I'm in a big city, in a residential area.

Being mine was elevated at 36 ft at the base, I also had a pretty
good line of sight to any potential noise sources.
The tip of the radiator was at about 68 ft. As far as the VE being
better on the wire, that's probably fairly normal, being he wasn't dx.

Also, as a final note, while your butternut with 20 radials is ok, it
still
isn't quite up to the performance I saw with mine at 36 ft, using a
full size antenna. So I saw a larger signal increase on the dx than you

I bet. Mine was appx equal to a full length monopole with 60 radials,
if ground mounted. I'd have to look, but my ground may be a bit better
too. I'm right on the edge of being in a "30" zone. Of course, raising
efficiency raises s/n equally, but I noticed that I never saw the same
performance I had with the ground plane, when I ran the same full size
vertical on the ground with 32 radials. That antenna was about equal
to my dipoles at 1500 miles. Maybe a small bit better, but not any
2 S units worth like the GP was. So regardless of some saying the
number of radials is not too important, it must be, if you want the
best
performance. Sure made a difference here...
Either that, or elevating it above the surroundings makes the
difference.
Myself, I think it's about 75% the first, and 25% the second...
Elevating
the antenna for sure increased my local ground wave. I could work 50
miles away ground wave easy. I'd have cases in the daytime where I'd
lose locals due to the band stretching out. But I could still nail them
at
S 9 using the GP, where the dipole would be hard to read backscatter.
Of course, if the band was open short, I'd be 10-20-30 over 9 on the
dipole
to the same location.
Anyway, I guess you gotta use what works, but I don't think it's
totally
normal to see a huge difference in noise between vertical and
horizontal
unless something local is the culprit.
MK