Exactly the point i have been trying to make. getting a lot of push back on it though in this thread, not sure why!
Of course, the available rated miles are an estimate generated by the BMS of how much energy is contained in the batteries, and Tesla tries to make sure it is as accurate as possible (so it doesn't drop from 30 rated miles to zero instantly, for example).
However, the fact that the rated miles available is an estimate does NOT mean that the number of Wh per rated mile differs on a per-user basis. There is no evidence that it does, for those vehicles with a properly functioning BMS. People can measure it.
How to square this? (This is speculation on my part.) I think basically, there is a bit of "slop" at the bottom of the battery (below zero rated miles), which Tesla hangs on to. When you get close to or below zero, the car will shut down at some point, but there will be a little energy left. If there are errors in the BMS that are during the discharge (would be "discovered" at the very end!), a little bit of this margin might be "dug into" to make sure that you do get the predicted available energy out of the battery that is shown to you (if everything is working properly) and you don't get premature shutdown.
Then, when you recharge, it will "simply" start counting how much energy has been placed into the battery. The number of rated miles displayed is this energy divided by the Wh/rmi constant (on the charging screen 245Wh/rmi is used; not the same as the discharge constant). It's far from simple, of course, since there is rebalancing, etc. that takes place. But that simple formula applies still.
This is presumably why "cycling" of the battery is sometimes recommended - it allows the BMS to start from a low state of charge and really keep track of how much energy it put into the battery - which sometimes results in a better count of the energy available (and a change in your projected 100% rated miles).
But the situation remains that, as far as I know, everyone with an AWD will measure 230Wh/rmi during a discharge (when the test is conducted in a valid manner). There are of course probably a few outliers where the BMS is really confused, where this relationship may not hold for a particular discharge. But I personally have not seen that data from anyone posted here.