Looks like they used the same data that they use to provide insurance quotes, and then the article got popular, and they didn’t want to fight pissed off car manufacturers, so they chucked a disclaimer on the article.
Using it to provide insurance quotes would be appropriate. Using it to determine which vehicles crash the most is not.
Using their 'formula,' if I drive a Honda Civic and get rear-ended, insurance totals the car, and I use that money to go buy a Model 3, they will count me as a "Tesla crash statistic" because I'm getting a quote for a Tesla with an accident on my record. This is idiotic because A) I wasn't driving a Tesla during the crash, and B) I wasn't even at fault for the accident, yet they still count it as a Tesla crash.
Not to mention other issues with the article like Pontiac, Saturn, and Oldsmobile being some of the "safest" cars on the road even though those companies built shitty cars and went out of business 10-15 years ago.
The company put the disclaimer on the article because it's junk data and misleading conclusions.
That report listed stuff like "bicyclist collides with stationary parked and shut-off car" because they were super harsh on the assisted driving potential.
That’s just because these drivers actually have insurance report their crashes. Unlike all the Nissan Altimas driving around town with temporary tags and no insurance
A buddy of mine got an EV but had the cash and tested like 30. Kia, Hyundai, ToyotaHonda, BMW, Mercedes, Audi - everything! - and Tesla. Took him 3 months almost.
In his words: there are three main classes of EV: the consumer range from Kia to Honda, the sport range from Audi to BMW, ... And Tesla.
Tesla has all these problems like Musk, and a boring UI and all, but its assistance is still head, shoulders and belt ahead of anything else on the planet. The BeeMercAudis are neat, but, still ... no. Not even close.
And he was a Tesla hater! Hates their boring same-ness and ipaddery shit UI and lack of knobs, etc.