The confusion is yours. Radiative forcing is measured in terms of the
change in radiative forcing since 1750 purely because tis is the most
useful measure for understanding the effect of CO2 concentration changed
by human processes related to industrialisation. The model is used to
estimate the degree of radiative forcing in subsequent years relative to
1750. The model simply measures radiative forcing. It measure
something that exists. If it is zero at year one that is the same as
taking a reference year as the base of measurement For example a
temperature of 0 degrees Fahrenheit or Celsius does not indicate that
heat in the atmosphere does not exist. Zero degrees Kelvin represents
the absence of heat energy, but that does not mean that heat energy
does not exist, just that it is not present at that measured
temperature. You are thus mixing up a measurement scale with a physical
state. It is quite possible that if CO2 concentration is reduced below
that of 1750 that the model would produce a negative measurement. Just
as a really cold day can produce a negative temperature compared with
the defined 0 degrees F or C.
Radiative forcing, like many other phenomena, cannot be measured
directly because there is no suitable instrument. It requires
mathematical transformation to filter out other influences to produce
the measurement number. The problem is not that it does not exist.
Your argument that the measurement of radiative forcing does not provide
evidence of a greenhouse effect is what is a real stretch. It has
been shown that higher atmospheric concentrations of CO2 result in
increased heat energy being trapped inside the atmospheric envelope.
Radiative forcing measures merely show how this varies with CO2
concentration. The model that is used for historical measurement can be
used to project future estimates of radiative forcing given differing
concentrations of atmospheric CO2.