To be completely fair, each car requires like 10,000 parts, each with its own supply chain, from all over the world. You only get the car built when you have all 10,000 parts in one place (fremont). Any issues with any vendors, for any reason, will affect the timing of the final product - you are only as fast as the slowest supplier. And that's to say nothing about quality control, unexpected disruptions, the scale which Tesla operates, immaturity of supply chain overall given the infancy of EV, current global supply chain issues, etc.
In the past I have worked at Apple supply chain, which has a world class reputation under Tim Cook, and I can tell you it's got nothing on Tesla, in terms of shear complexity.
So, if you are Tesla, what options do you have regarding EDD? 1) stick with the original parts and wait, even if they are in shortage, as new qualification and internal adjustment processes add to the complexity of the supply chain further (basically the GM/F route), 2) provide the most pessimistic EDD and have customers pleasantly surprised, 3) provide the most update to date EDD based on what Tesla knows about the supply chain real time. Note 2) would still create disappointments at times, as some supply chain disruptions will still happen, furthermore, not sure getting an Oct EDD and all of the sudden being told to pick up the car in March is good for everyone either.
I am not trying to defend everything Tesla does - their customer service could definitely be better - but I do want to point out when it comes to EDD it's not all up to Tesla and their hands are tied somewhat due its huge dependency to its suppliers - and to a certain extent, their own agility, as qualifying new suppliers to replace unavailable ones adds complexity, uncertainty, and execution risk.