"Owen Duffy"
No one has demonstrated that using equivalent impedances etc
is not a valid analysis of the steady state behaviour.
_________

A reflection is a reflection. The reflection of a ~steady-state r-f source
may produce a different perceived/effective result than if that source
includes transients (modulation), but such does not negate the existence of
reverse/reflected power in the steady-state case.

Decades of experience with analog broadcast TV transmission systems
demonstrate that the reflected power from a mismatch at the transmit antenna
produces an amplitude variation (ripple) and other effects across the r-f
and demodulated video channel bandwidths that are directly related to the
magnitude of the antenna mismatch and the round-trip propagation time of the
transmission line between the tx and the antenna (period = 1 cycle per ~491
feet of air-dielectric line).

This is evident not only from accurate measurements made via a highly
directional r-f coupler sampling forward power at the tx end of the
transmission line, but also from results seen on the screen of TV sets
viewing those transmissions. I suspect, Owen, that you would agree that
this example originates from a "practical" system.

The r-f power supplied even by a CW source is subject to the same amount of
reflected power for a given antenna mismatch, which will have an appropriate
effect on system performance. Whether or not that reflected
power/performance effect is important (or even recognized as existent) is
the issue at hand.

RF