Not sure about the higher frequencies but in the HF band we absolutely need
an antenna coupler that matches the impedance to the selected frequency. If
a coupler fails we can barely throw a signal a few miles whereas when the
coupler does it's job we can bounce a signal off of the ionosphere at night
for a few thousand miles.

That's certainly true at the transmitting end. A good impedance match
is needed in order to enable the transmitter to deliver power
effectively into the antenna, from whence it can be radiated.

It's rather less true at the receiving end, at least in the lower-
frequency HF bands. In these bands, the ability to receive a usable
signal is often dominated by the amount of natural and man-made noise
in the band, and not by the receiver's own self-generated noise. Even
with a serious impedance mismatch between the antenna and the
receiver, enough signal reaches the receiver front-end to overcome the
receiver's own internal noise.

If you happen to live in an area which is blessed by a very low
background-noise level (e.g. out in the country, away from power
lines) and you're DXing in the HF bands, then a good impedance-matched
antenna will let your receiver take best advantage of the low noise
level.

If you're SWLing in a city, surrounded by power lines and electric
motors and neon lights and computers, the background noise level is
going to be much higher, and the weak distant stations will be drowned
out by the noise anyhow... and an inefficiently-matched antenna such
as a whip or longwire will give you enough signal to hear the stations
which are _not_ drowned out by the noise.

As an example - if your receiver has decent sensitivity, and a low
internal noise level, you may find that you can hear a signal with
decent audio quality all the way down to S0 or below (if there's no
noise obscuring it). You then find that with an impedance-matched
antenna the band's background noise is S6 or so. Assuming that you
can make use of a signal which is somewhat below the broadband noise
level, let's say that you decide you can copy stations whose own
individual signals are S5 or better, and that lower-level signals are
blanketed by the noise.

At this point, you realize that you can use an antenna which is 5
S-points (nominally, 30 dB) less efficient, and still receive the same
set of stations. With a less efficient antenna, the stations' signals
will be weaker... but so will the external band noise, by the same
ratio, and thus the signal-to-noise ratio of each station will remain
unchanged.

As an example, my ARES/RACES group has a multiband HF setup in our
city's police and fire admin building, which is downtown near the main
commercial-and-restaurant street and the light-rail system. We have a
trap-dipole antenna strung up above the building's roof. On the
80-meter band, the broadband noise level across the whole band is
rarely less than S9! We could probably receive the same set of
stations using an antenna consisting of two coat-hangers and some damp
string!

--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!