17 fatalities, 736 crashes: The shocking toll of Tesla’s Autopilot, the Washington Post announces, in a long piece by reporters Faiz Siddiqui and Jeremy B. Merrill. What does Tesla dude Elon Musk have to say?
Tesla CEO Elon Musk has said that cars operating in Tesla’s Autopilot mode are safer than those piloted solely by human drivers, citing crash rates when the modes of driving are compared.
Well, is that true? What are the data? Siddiqui and Merrill don’t bother to tell us. But they are on the search for “patterns”:
Tesla’s 17 fatal crashes reveal distinct patterns, The Post found: Four involved a motorcycle. Another involved an emergency vehicle.
Uh, really? You can obtain statistically valid results out of a population of 17? I guess the data “suggest” that Teslas have a harder time “seeing” motorcycles than emergency vehicles, or maybe that riding a motorcycle is a lot more dangerous than driving an emergency vehicle, or maybe that cyclists are hot dogs who get what’s coming to them. Who knows?
There’s nothing in Siddiqui and Merrill’s article to “prove” that Teslas operating on Autopilot are more likely to be involved in an accident than driver-operated cars. The implicit assumption, of course, is that since self-driving cars are “new”, the accident rate, and even more the fatality rate, should be zero, which is ridiculous. For the record, 42, 915 Americans died in traffic accidents last year, and damn few of them died thanks to “Autopilot”.
WashPost folks are not the only reporters who have trouble with “numbers”.
