Hacker News Reader

Comment by sriram_malhar | Hacker News Reader

I heard this anecdote in India, and I'm not sure if it is apocryphal. There was a small airline flying turboprops to service a Himalayan region. Someone asked whether it was safe to fly these planes and the response was as hilarious as it was scary: "Yes, sir. It is very safe. They used to have 11 planes, 10 of them crashed and now they are really really careful with the one left" :)

View on HN

Reminds me of a story I was told while learning how to design experiments and a/b tests.

In World War 2, some company was experimenting with the kind of material to use for combat aircraft wings. When they did their initial trials, they saw that the planes coated with the new material came back with more bullet holes. Off hand, that would seem like a reason to dismiss the new material. But, digging into the data, they realized that planes with the old material were actually crashing when they got their first bullet hole and the planes that were making it back, even with bullet holes, all were coated with the new material.

I think that account gets it a bit wrong. Here's what Wikipedia has:

https://en.wikipedia.org/wiki/Survivorship_bias#In_the_milit...

> During World War II, the statistician Abraham Wald took survivorship bias into his calculations when considering how to minimize bomber losses to enemy fire.[10] The Statistical Research Group (SRG) at Columbia University, which Wald was a part of, examined the damage done to aircraft that had returned from missions and recommended adding armor to the areas that showed the least damage, based on his reasoning. This contradicted the US military’s conclusions that the most-hit areas of the plane needed additional armor.[11][12][13] Wald noted that the military only considered the aircraft that had survived their missions; any bombers which had been shot down or otherwise lost had logically also been rendered unavailable for assessment. The holes in the returning aircraft, then, represented areas where a bomber could take damage and still return home safely. Thus, Wald proposed that the Navy reinforce areas where the returning aircraft were unscathed[10]:88, since those were the areas that, if hit, would cause the plane to be lost. His work is considered seminal in the then-nascent discipline of operational research.[14]

That's coincidental; according to Wikipedia, Wald and his wife perished in a plane crash in India.

That seems to just be about picking the correct success criteria. It seems here the key success criteria should be which version of the plane makes it back most often. Which plane has more bullet holes is clearly secondary to that.

Heh, this reminds me of a flight my friend took from a small island country. The side of the plane proudly displayed in big letters "Fly is Safe". I don't think he was reassured...

edit: whoops, can't spell "plane" this morning

If it was the plain it was ok

The plane in Spain falls mainly in the rain.

Reminds me of this anecdote from Guatemala. They are boarding the plane and the women notices that the pilot is very young. She asks, "Are you sure you're old enough to be flying this plane?" He replies: "Don't worry ma'am, I come from a long line of pilots. My grandfather was a pilot and my father was a pilot too."

Took a while :)

I don’t get it! Can someone explain?

The key word is "was"!

To me it simply sounds like all of the unsafe planes crashed and now the safest one is the one left ;)

Damn right... survival of the fittest.

That actually happened in SE Asia.

One airliner had a couple of incidents, including a tail strike which damaged the pressure bulkhead, and then crashed.

That boosted the reliability average of all the other remaining airliners in the region.

Sounds like Nepal airlines

Fake planes beats fake licenses

Way to take a joke and extrapolate it to the truth. Many friends worked hard for their licenses.

> I'm not sure if it is apocryphal

Seems likes they took it with a reasonable grain of salt?

Reply on news.ycombinator.com