4 ms·

For comparison, about 10,000 unrestrained motorists die in fatal vehicular accidents every year, and millions are injured. https://www.cdc.gov/motorvehiclesafet

For comparison, about 10,000 unrestrained motorists die in fatal vehicular accidents every year, and millions are injured. https://www.cdc.gov/motorvehiclesafety/seatbelts/facts.html https://www.cdc.gov/motorvehiclesafety/seatbelts/facts.html

wallace_f 10y ago
So according to this answer[1], ~1.7MM fly/day. Making a quick guess that from '80-'08, Americans average 1MM flights/day, then odds of serious injury are ~1 in 95,000. Odds of death are 1 in 9,333,333. Assuming the possibility of taking 3 'average' flights/day, then you'd need to board flights non-stop for 8,523 years to die, on average. 88 years to be injured. [1] - https://www.quora.com/How-many-people-fly-domestically-in-the-United-States-each-day https://www.quora.com/How-many-people-fly-domestically-in-th...
- gingerbread-man 10y ago
  Of course, general aviation is much more dangerous. As a general rule, the smaller the plane, the riskier it is, partially due to the corresponding increase in regulatory scrutiny as the size of the aircraft grows. Private pilots need a minimum of ~50 hours for their license. Most newly-hired jet pilots, private and commercial, have at least 1500 hours, and many have more than that.
  - mikeash 10y ago
    Larger airplanes are also much more capable (better instruments, things like anti-icing systems, ability to fly above weather), far better maintained on average, fly into better airports, have better flight planning resources, and more. Note that when looking at airline safety, you'll get misleading numbers if you go back too far. Safety has improved remarkably in the past couple of decades. In the 80s and 90s, a couple hundred fatalities per year was about typical in the US. In the last decade, the typical number is zero. There hasn't been a fatal crash involving an American airliner for over eight years, and you have to go all the way back to 2001 (November, not September) to find one involving a major American airline. Even if you include foreign airlines operating in the US, that only adds one fatal crash and three deaths to the recent total.
    - gingerbread-man 10y ago
      The challenge with measuring the safety of general aviation is that it is much more behavior-dependant. All of the US airlines use the same standards for deciding when to fly/ not fly for weather, have similar maintenance programs, training requirements, etc... When it comes to private air travel, it's all up to the pilot and the insurance company.
      - mikeash 10y ago
        I imagine GA follows something like a Pareto distribution, with 10% of the pilots involved in 90% of the incidents. That 10% would get retrained or weeded out in the airlines, and the rest get much more consistency on all the rest of the stuff you mention. So yeah, there really isn't any one "the safety of general aviation." It's a huge spectrum. Another challenge is reporting. The FAA knows exactly how many flights and how many hours the airlines did last year, but GA activity is much harder to gauge. It's hard to figure out what safety looks like when you don't have a good figure for the denominator.
        gingerbread-man 10y ago
        From my limited experience working with them, the general aviation insurance industry seems ripe for startups. Data science is a huge deal in insurance, but the companies insuring Cessnas and Pipers don't seem to have fully caught on.

wallace_f 10y ago

So according to this answer[1], ~1.7MM fly/day. Making a quick guess that from '80-'08, Americans average 1MM flights/day, then odds of serious injury are ~1 in 95,000. Odds of death are 1 in 9,333,333. Assuming the possibility of taking 3 'average' flights/day, then you'd need to board flights non-stop for 8,523 years to die, on average. 88 years to be injured. [1] - https://www.quora.com/How-many-people-fly-domestically-in-the-United-States-each-day https://www.quora.com/How-many-people-fly-domestically-in-th...

gingerbread-man 10y ago
Of course, general aviation is much more dangerous. As a general rule, the smaller the plane, the riskier it is, partially due to the corresponding increase in regulatory scrutiny as the size of the aircraft grows. Private pilots need a minimum of ~50 hours for their license. Most newly-hired jet pilots, private and commercial, have at least 1500 hours, and many have more than that.
- mikeash 10y ago
  Larger airplanes are also much more capable (better instruments, things like anti-icing systems, ability to fly above weather), far better maintained on average, fly into better airports, have better flight planning resources, and more. Note that when looking at airline safety, you'll get misleading numbers if you go back too far. Safety has improved remarkably in the past couple of decades. In the 80s and 90s, a couple hundred fatalities per year was about typical in the US. In the last decade, the typical number is zero. There hasn't been a fatal crash involving an American airliner for over eight years, and you have to go all the way back to 2001 (November, not September) to find one involving a major American airline. Even if you include foreign airlines operating in the US, that only adds one fatal crash and three deaths to the recent total.
  - gingerbread-man 10y ago
    The challenge with measuring the safety of general aviation is that it is much more behavior-dependant. All of the US airlines use the same standards for deciding when to fly/ not fly for weather, have similar maintenance programs, training requirements, etc... When it comes to private air travel, it's all up to the pilot and the insurance company.
    - mikeash 10y ago
      I imagine GA follows something like a Pareto distribution, with 10% of the pilots involved in 90% of the incidents. That 10% would get retrained or weeded out in the airlines, and the rest get much more consistency on all the rest of the stuff you mention. So yeah, there really isn't any one "the safety of general aviation." It's a huge spectrum. Another challenge is reporting. The FAA knows exactly how many flights and how many hours the airlines did last year, but GA activity is much harder to gauge. It's hard to figure out what safety looks like when you don't have a good figure for the denominator.
      - gingerbread-man 10y ago
        From my limited experience working with them, the general aviation insurance industry seems ripe for startups. Data science is a huge deal in insurance, but the companies insuring Cessnas and Pipers don't seem to have fully caught on.