, which I'm attaching here. It seems that as mileage increases the standard deviation of battery range lost also increases. What are your thoughts about all this?
I think that's partly expected, but mostly due to the method that Stats uses for gathering this data. If the data were filtered to only show projections of range from SoCs above 85%, and at temperatures above 50 degrees, I think this would look a lot tighter. There are a lot of errors introduced by low SoC extrapolations (that's the source of the initial spread - there is no evidence I've heard that there are ANY vehicles that show over 310 rated miles (under the old system) when new (obviously they do now with 2020 vehicles), and there are no reports of brand new vehicles with 300-305 rated miles that I am aware of), and because at low temperatures, the way Stats gathers data, the 100% projected range tends to go off of a cliff (which is not real). (Related to the two different API SoCs I guess.)
That being said, my current theory is that all vehicles are somewhat pinned (by software) at 310 rated miles when new, and only show degradation when capacity drops below 76kWh (for 2018/2019 AWD vehicles). So that behavior alone would lead to increased spread as time goes by, as different vehicles degrade at different rates (or may even have somewhat different starting points).
The original source of the concern is that we have 2 performance model 3s., 1 with 12000 miles one with 18,000 miles. Car with the Lesser mileage shows no range loss at all or perhaps maybe 1 mile 309 / 310, Well these slightly older car with 6000 more miles is at 290.
Yes. My guess is that your car showing no capacity loss probably just started at a higher max SoC than the other one (even though they both showed 310 rated miles - I've outlined my (unverified) theory on that elsewhere). To some extent it could be related to cycle count, type of use, etc., of course. I would expect that second vehicle to show more capacity loss relatively soon. But it may always be better.
) Tesla's wheel size 'adjustment' of estimated range (obviously a boost for the 18in Aero Wheels and debit for the 20 inch wheels w/PS4S. Etc.)
For the rated range number, this only applies to 2020 vehicles. For the earlier vehicles, it only affects the Trip Planner projections.
But I wanted to get your feedback on whether you think this collection of factors is reasonably comprehensive or whether there's something missing from the list.
Your list seems pretty complete.
So, as covered above, related items I would add:
a) Initial capacity likely differs from vehicle to vehicle, but this is hidden (start point can be estimated from how long the vehicle takes to show capacity loss). So that results in differences from vehicle to vehicle.
b) People erroneously project their projected full SoC numbers from SoCs below 80% (too much error), which results in incorrect reports.
Right now I can't think of anything else to add.
