When the Number Isn’t 100%

All my devices—Kindle, phone, laptop, headphones, iPad, and iPad stencil—are at maximum battery capacity. Their chargers are neatly coiled in a billfold-sized makeup purse. Everything is stuffed into my backpack, ready to accompany me on a cross-country trek to my motherland. That’s in the midst of historic flooding, with an emergency declaration coming from the president earlier today.

Where I’ll stay is safe and also interesting. I’ve been receiving pictures from my parents of the landscape in their backyard: a lake. Both of their neighbors’ docks are completely submerged—an odd thing, especially during a recent sunset, where the large wooden real estate no longer blocked the sunset’s orangey-pink, bright and swirling hues.

My sights are set on the 5.5-hour flight and easing my white knuckles back to a more natural color.

Seeing a parade of lights in the sky tonight, all routed to Atlanta’s airport, I thought: See all those planes? All departed and landed safely today.

I next thought to look up statistics of successful flights, but then rethought the search inquiry, modifying it to passenger plane success.

I didn’t look—likely because it wouldn’t ease my fear. Especially if the number wasn’t 100%.

Ok, so I looked. For all you nerdies out there, here’s data from a Reddit user that (sorta) makes me feel better:


“Based on this, you got 5 fatal accidents for 32.9 million flights (5-year average).

Assuming every flight has the same chance to crash (eg. there are no airlines with better and worse maintenance or similar), a fatal crash has the probability of 5/32.9e6 = 0.000000152 = 0.0000152%.

The probability of you surviving a single flight is the inverse of that, so 1-5/32.9e6 = 0.999999848 = 99.9999848%

To survive two flights, you have to survive the first one AND have to survive the second one, so the probability of you surviving two flights is (1-5/32.9e6) * (1-5/32.9e6) = (1-5/32.9e6)2 = 0,999999696 = 99.9999696%

The probability of you surviving three flights is (1-5/32.9e6), and so on.

So the question is how many flights should you take so the probability of surviving all of them is 90% (so you have 10% chance of dying).

We would need to solve (1-5/32.9e6)x = 0.9 you can get the logarithm of both sides, so x = log(1-5/32.9e6, 0.9) = 693272

So you would need to take a flight every day for 1899 years to have 10% chance of dying or for 181 years to have 1% chance of dying.

Known issues:

  • I think the data includes many types of flights, not just passenger jets (which AFAIK the safest)
  • The data may be skewed, as it includes the duration affected by covid lockdowns
  • Probably deaths by hour flown or deaths per miles flew would be a better estimation
  • Assume every flight has the same chance to crash and every flight is independent of the others
  • Assume everybody dies in a fatal accident
  • edit: assuming the safety of flying doesn’t change over time (but it got better even in 10 years time frame)”

See you tomorrow in the PNW! Love, Jaclynn

Leave a comment