The study needs to stratify the results by the typical driver that buys a Tesla and conditions/situations when Teslas are driven. Already, using their numbers,

The study needs to stratify the results by the typical driver that buys a Tesla and conditions/situations when Teslas are driven. Already, using their numbers, just having a Tesla with no safety features enabled drops the accidents rate to 1 in 1.4 million. Is it hard to believe that much fewer Teslas are used to drive home at 2 am, with five distracting passangers, after a night of partying?