Owen Duffy wrote in
:

....
So, if the feed point impedance was 18-j142, and you used 38' of RG58
(for example, I could not find that info), then you would expect to
lose 70% of the power out of the ATU as heat. Of course, it is likely
that the power out of the ATU is somewhat less than the transmitter
output as the ATU load is some 7+j10, relatively low R (but typical
ATUs tend to be more efficient on the higher bands).

Did the Smith chart reveal any of that?

Well, subject to how carefully you plotted the data, you might get a
reasonable answer for the line input impedance (as seen by the ATU), but
the Smith chart will not directly read the line loss with a mismatched
losd.

With a lot of care in plotting, you might scale off the values of G at
line input and output, and find the ratio Vload/Vin from the chart to
calculate the ratio of Pload/Pin and so find the loss under mismatched
load.

The line loss in this case is just over 5dB, more than 10 times the
0.5dB loss you would observe with a matched load.

An exercise for the reader is to use a Smith chart to find the loss
under the mismatched load with RG213 where the matched line loss is
about 0.31dB.

Owen

PS: if you had in mind using the handy dandy "Additional loss due to
VSWR" graph in the ARRL, the line input VSWR is 7.7, load VSWR is 23.7.
If you get the same loss figure are mentioned above, it is by accident
of the scenario rather than soundness of the method.