But note carefully the (quite correct) caveat in 'OD's calculator which is designed for the case in which;
"... a transmitter which is designed for a particular load impedance, is connected to a an antenna of that impedance, over transmission line of the same characteristic impedance".
So, if the transmitter is 50-ohms, the antenna is 50-ohms and the intervening line is 50-ohms, all will be well but, if one or more of those components does not meet this specification, a more-complete analysis will be required.
That scenario is the basis for some of the calcs.
The 'estimate' of VSWR assumes that line loss per unit length under mismatch is constant. That is not strictly true, but if you want to calculate more accurate results, you need more than VSWR, you need data that allows calculation of Gamma, the complex reflection coefficient (rather than just the magnitude of it which is the basis of VSWR). Nevertheless, the calculation of VSWR at another point is accurate enough for the OM's purposes.