This is a great article, with hard-to-find details and excellent analysis of how different factors effect both calculated/predicted and real-world results. I disagree that Tesla should show only percent of charge, and drop any mileage prediction. Drivers care about how far they can go. Showing only the charge percentage just means that drivers will try to make rough mental calculations of range, that will be less accurate than Tesla's current distance predictions. A better solution, in my opinion, would be to display a range estimate from "Tesla’s actual prediction algorithm", which as Brandon says, is excellent. Finally, my experience with using the Tesla mobile app for service appointments has been poor more often than it has been good. The bad experiences include multiple app-scheduled appointments requiring driving to a Tesla Service Center, to be told on arrival that the parts were not in, and sometimes, not even ordered yet. The worst example of relying on automation instead of human understanding, was after a tire failure, when the app arranged for us to be towed to a Service Center 70 miles away, with our arrival time falling well after they would be closed, on a Friday evening. I suppose the app wanted us to sleep in the car for three nights, until Monday morning. At which point, I doubt that the Service Center would have had a tire to match the other three.
Freaking phenomenal article. I know all this stuff but you put it into words. Bravo.
I don’t know all this stuff but read this because of a semi-snarky comment I made on threads. Thank you. This makes real sense to me now.
This is a great article, with hard-to-find details and excellent analysis of how different factors effect both calculated/predicted and real-world results. I disagree that Tesla should show only percent of charge, and drop any mileage prediction. Drivers care about how far they can go. Showing only the charge percentage just means that drivers will try to make rough mental calculations of range, that will be less accurate than Tesla's current distance predictions. A better solution, in my opinion, would be to display a range estimate from "Tesla’s actual prediction algorithm", which as Brandon says, is excellent. Finally, my experience with using the Tesla mobile app for service appointments has been poor more often than it has been good. The bad experiences include multiple app-scheduled appointments requiring driving to a Tesla Service Center, to be told on arrival that the parts were not in, and sometimes, not even ordered yet. The worst example of relying on automation instead of human understanding, was after a tire failure, when the app arranged for us to be towed to a Service Center 70 miles away, with our arrival time falling well after they would be closed, on a Friday evening. I suppose the app wanted us to sleep in the car for three nights, until Monday morning. At which point, I doubt that the Service Center would have had a tire to match the other three.