Topband: S/N versus Attenuation

[Ford Peterson]

I've been pondering Rx antennas and am having difficulty working through the
logic. 

In my station setup, I have one bi-directional beverage (E-W switchable) and
a second Rx antenna making good use of a full-sized 80M elevated vertical.
The absolute output from both antennas on 160M is virtually identical noise
output and provides me with S0-S1 noise on the MKV with no attenuation.  So
switching between the east beverage and the omni makes little difference in
the signal strength of the desired signal arriving from the east.  At this
Minnesota location, during a contest weekend, I will "run" and listen on the
omni to take advantage of all directions.  The majority of stations can be
worked directly without any fiddling.  When working the guys using "bed
springs powered by the Energizer Bunny" I'm in ESP mode. I will fiddle with
the Rx antenna choices and find that the beverages will outperform this
'detuned' vertical on S/N only about 25%-30% of the time.  The rest of the
time, the omni vertical will actually provide better S/N and seems to have
higher absolute signal output.

Although it would be virtually impossible to prove, I do believe the
elevated vertical provides a nice, clean, classic vertical omni pattern but
is simply too short for 160M resonance.  It's about 1/8wL actually.  The
noise output of this antenna is about S0 on Topband and often provides much
better S/N than the beverages.  Of course, I can accomplish the same goal
(noise=S0) on the self-resonant transmit antenna by inserting about -18dB of
attenuation and gain some benefit of S/N but not nearly as good as doing it
at the antenna.

What am I missing here?  The fact that I'm simply using a detuned
(non-resonant) antenna introduces matching losses that simulates receiver
attenuation.  Yet it provides much better S/N.  Would I be even better off
still by chosing to detune the otherwise self-resonant (1/4wL) transmit
antenna during receive?

Thanks for the bandwidth...

Ford-N0FP