People that kill people behind the wheel are often sober.
41,000+ die every year in the US from fatal car crashes
13,470 of those were alcohol related
So, statistically speaking 33% of people involved in traffic related fatalities, alcohol played a role. That leaves 66% of all traffic fatalities in the year of 2006 not alcohol related. So, to try and use fear mongering or blanket statements about me potentially killing someone. I am twice as likely to do that when I am sober. It has a lot more to do with everyother factor involved with the one person then you care to admit. Such as the quality of vehicle, driving conditions, driving experience, driving habits, knowledge of areas, traffic volume, vehicle design, etc. I admit that alcohol will affect a person's ability to drive more then any one of these reasons, but to not consider them is just ignoring facts. That said, all these other conditions were ideal.
The fact that you assume a person who has been drinking is automatically to blame also skews your opinion of the facts. But I imagine your response would be, "they shouldn't be on the road to get in the accident anyway". Maybe they aren't and the three-year-old on the sidewalk that they stopped the other driver from hitting gets killed instead. See, fear-mongering goes both ways.