I was using the stats from the article for injuries, not deaths. I believe the quotes you provided used deaths intentionally because the probability of dying in a crash is so low. Injuries still matter (in my opinion).