To believe otherwise is either:
    a) Informed nitpicking about wording or details, or
    b) Underinformed innumeracy.

Flying in the United States is currently well over 20 times safer than driving, on a per-mile basis. This is for the average person using the average road and the average commercial airplane.

If you boarded a commercial airliner in the U.S. every day for the rest of your natural life, on average you would have to live 29,000 years before experiencing an accident.

Even then, the odds are good you'd survive.

An overwhelming majority of airplane accidents (of the fatal variety) are caused by pilot error, usually controlled flight into terrain. In general, it's not the mechanics or the bombs or the weather that kill you. Also, a disproportionate number of accidents occur at takeoff and landing, and not at 30,000 feet as accipiter would have you believe.

An accident occurs on a commercial airliner approximately once every 2 million flights. According to the New York Times, the transportation-related fatalities expressed in deaths per 100,000,000 miles traveled (in 1992) were:


    Mode of transport   Deaths per 100M miles
+------------------------+------------------+
| Commercial airliners   |        0.006     |
| Bus                    |        0.022     |
| Rapid transit - rail   |        0.029     |
| Car                    |        0.849     |
| Truck                  |        0.898     |
| Bicycle                |       21.235     |
| Motorcycle             |       22.909     |
| Pedestrian             |       84.030     |
+------------------------+------------------+

Air travel is getting safer every year, and has been for quite some time. That said, risk assessment is not simple. If you're a teenage male, or you like to mix the liquor with the interstates, you're certainly much more likely to die in a car than on a plane. If you primarily use rural roads, are in your thirties, are female, and mostly take very short flights (which spend a disproportionate amount of time taking off and landing, compared to cruising on long flights), the numbers come out about equal, slightly in favor of the car journey.

If you compute deaths worldwide on a per-trip basis, you're slightly better off taking a car, partially since so many short car trips are taken on low-fatality-rate roads. Also, the fatality risk for 'high-risk' drivers is 1,000 times higher than the fatality risk for 'low-risk' drivers, according to Risk Analysis.

Keep in mind that well over half of all drivers in the U.S. rank themselves above the 50th percentile in driving skill and safety. Believe it or not, the odds are tautologically 50-50 that you're a below-average (with respect to safety, or, I suppose, anything else) driver (although, reading this, online, on E2, changes that substantially I bet, though in which way I have no idea).

It doesn't take too much to get yourself killed in a car, by someone else, with little or no chance at avoidance. It happens all the time to reasonably adequate drivers. If you think your mad skillz are going to keep you alive in the face of Johnny Sixpack, you're wrong.

As long as your statistics aren't coming from the airline industry, and you maintain an open mind with respect to risk analysis, you can still safely conclude that:

all other things being equal, traveling a mile by air is much, much safer, today, than traveling that same mile by car in the U.S.

2/19/01: accipiter's writeup, which inspired this rant by loudly declaring the title of this node to be a myth, was nuked.