Long ago when I had a real job I was "safety director" of a trucking company. I went to a couple of accident investigation courses at Penn State and Michigan State. It is actually pretty simple and common to determine whether or not a light was on when impact occurred. Other reconstruction can be somewhat speculative, but there are methods of reconstruction there, too. Typcially insurance companies do this to some degree to limit their exposure. I agree that the "facts" of this case are improbable, at best. With a fatality like this caused by a rear-end collision, the insurance company would probably have information they gathered from police or on their own to limit damage to themselves. I'm not buying this story at all as written here.