Soooo...if the modulator is transformer-coupled, how does the DC input
to the RF amplifier increase when modulation is applied? A transformer
can't couple DC...

Not saying the fellow can't do what he wants, but we need to know a LOT
more about the circuit.

Remember, too, that the half-power rule is for sinewave modulation. If
you use square-wave modulation at 100% depth, you need as much power
from the modulator as the DC input to the RF amplifier. The modulated
output is four times the unmodulated power, for half the time, and zero
for the other half, assuming an RF stage that responds linearly to the
modulation voltage.

For a given modulated waveform, it's easy enough to figure out what the
actual power is, and it all has to come from somewhere. If the plate
modulation is linear (current is linear with modulated voltage) and the
modulation is coming through a transformer, then the average current
from the RF deck plate supply must be constant, and the DC power input
is therefore constant. Whatever power you put in the sidebands must
come through the transformer. But maybe the circuit he's proposing DC
couples the modulation voltage, and thus allows an increase in the DC
current going to the RF deck.

Cheers,
Tom