So both these replies suggest EV makers only have to present one number? And non other details?
All car manufacturers (doesn't matter what propulsion they use) must, by law, quote the official test data, which here is WLTP (in the USA it's EPA). Car manufacturers can choose to augment that with additional performance guide data, provided they word it carefully such that it doesn't fall foul of the various advertising rules.
In practice, few manufacturers bother to quote anything other than the official test data, plus, pretty much every manufacturer (and I very much doubt that Tesla are any different) "game" the test process to some degree, in order to get the most favourable result. "Gaming" the test process is allowed, to some degree, and consists of everything from making sure that test vehicles are as free running as possible (things like adjusting them carefully to reduce brake and bearing drag, making sure tyre pressures are absolutely spot on, wearing in the tyres to reduce rolling resistance, polishing the bodywork and making sure all panel fits are perfect to reduce drag, etc).
The problem is that virtually no one will ever be able to drive a car that is absolutely identical to the test car used, and in the same way as the test is conducted. As already mentioned, air temperature alone will change range/economy for all vehicles by about 9% over a 0°C to 25°C range (dry air, at sea level).
One of the biggest shortcomings in official testing, in my personal view, is that it takes little account of the extremes encountered in normal use. For example, a series of short (under 10 mile) trips made from a cold start each time, in very cold weather, might be a good way to highlight differences caused by cabin and battery heating efficiency (this is a seemingly big factor for EVs in general). I believe there is a good rationale for changing the WLTP test to include a standardised winter test as well as a standardised summer test. The test used at the moment is pretty much a summer test, as, AFAIK, it doesn't include testing after an overnight cold soak, to include the heater initial consumption, in fact the test car is allowed to be heated to a uniform temperature throughout, including the battery pack, before the test, I believe.
With conventional cars, where at least some of the winter range/economy hit is mitigated because heating comes from waste engine heat in part, the seasonal difference may have been seen as less significant. As more and more EVs become available I believe there is a very good case for a more representative test, so that new owners aren't so surprised at the loss of range seen in winter. This might also push manufacturers to do things like providing better insulation to their cars, plus more efficient heat pumps, or even adopt the original Prius system of using a vacuum flask to store coolant/heating fluid heat overnight for use when preheating the next morning.