by Patrick Appel

A reader writes:

Some people should not be allowed to play with numbers.

First, driving and flying do not remotely have "about the same death rate per mile" - you're 3.6 times as likely to die in a mile on the road as in a mile in the air or, put differently, you have a 72% lower chance of dying per mile if you fly.  That's a huge difference.  Ask the average person if she'd, say, change toothpaste if she could reduce her chances of getting tooth decay by 72 percent and see what reaction you'd get.

Second, the "per trip" measure is useless because it compares apples and oranges.

The average plane flight is somewhere in the 700-800 mile range, that is, 70-80 times the average trip length for auto travel.  Of course you have a lower chance of dying on any individual car trip.  Heck, you have a much smaller chance of dying from a heart attack or stroke during a 10 mile car trip than you do during a 700 mile plane trip, just because the flight is likely to take 6 to 10 times longer than the drive.

If you want a more valid comparison of the risks of driving to the risks of flying using the per trip measure, consider that the average driver probably is on the road for about 10,000 miles a year.  (Statistics on this are hard to come by, but this appears to be in the right range.)   That's about 1,000 trips, but it would be only 12 to 14 trips if you were flying.  So your likelihood of a fatal driving accident would be 0.00009 (0.09/100) and your likelihood of a fatal flying accident would be from 0.000021 (0.18/8,333) to .000025 (0.18/7,142).  Not coincidentally, these numbers look a lot like the per-mile numbers, and flying is considerably safer.

Oh, and 2005 was an above-average year for air fatalities.  The five-year average from 2004 to 2008 was 0.0021 fatalities per million miles and 0.15 fatalities per 100,000 departures, both around 17 percent lower than the 2005 figures.

There are also, of course, far more non-fatal accidents in cars.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.