Electric motors are not rated by the amount of input power applied, nor does the actual power produced in the installation directly equate to the horsepower rating of the motor. Should Tesla redefine the methodology of rating electric motors to fit your criterion (which appears to be to use input power applied in the application), the new numbers would seem to me to less reflective of the actual performance of the cars. Furthermore, it would require them to change the actual rating of the motor anytime they changed the software to apply more input power.
Tesla has changed their advertised power ratings many times thus far.
Do you have an explanation as to why other models are outputting more than advertised amounts of power, based on Inverter kW input ratings vs the P85D which puts out much less? It seems other models almost directly correlate their advertised power vs. actual logged inverter kW, with the only exception being the P85D. I don't see why the P85D is an exception to that rule. Their other models seem to reflect advertised power based on independent logging and it makes sense- but the P85D does not fall into the same logic and one must wonder how they came up with 691HP, plain and simple.