The model data is used predictions - IE: the storm will reach this location on this day and at this intensity in the future, but once a storm can be observed directly, in real time, either by aircraft or ground stations you no longer need predictive modeling for wind speed. Same goes for the wind field, they know what the winds are, we literally have hundreds of weather stations on the land and sea these days, so I strongly feel they should adjust the storm graphics to reflect that data. Future intensity forecasting has always been challenging while track forecasting has gotten way more accurate in comparison: https://www.nhc.noaa.gov/verification/verify5.shtml

...and correct on the winds recorded at the Cape.

I actually believe Andrew is what started this betting-the-over trend in wind speed reporting. Better to say its going to be worse then get caught with your pants down using a weaker forecast, basically its a CYA move. Andrew was a 5 but the NHC didn't know enough about such monsters because they were so rare, so call it a 4. Then as Andrew made landfall the instrumentation failed... wind levels were literally off the charts. Years later with better tools and reviewing the data they changed its official status. Part of the driving force behind the revisit was trying to put the damage into historical perspective and get the building codes adjusted to handle possible future storms. Many people cried foul at the Cat 4 rating due the intense damage, they even brought in tornado specialists to explain what they were seeing because the hurricane scientists had no frame of reference for the devastation.

Sources:
https://www.aoml.noaa.gov/hrd/hurdat/andrew.html
https://research.noaa.gov/article/ArtMID...urricane-Andrew