Hacker News Reader

Comment by NikolaeVarius | Hacker News Reader

Nope, risks are insignificant. It's extremely irrational to be scared to fly on Boeing.

LOL what? It's at least 4 orders of magnitude more risky to fly 737 MAX compared to Airbus A320 (counting deaths per 100k takeoffs). 4 orders of magnitude are not irrational.

That's a relative measure, it could still be the case that the absolute risk is still quite low for each individual. It could very well be the case that the risk per individual is low enough to not cause serious rational concern, while the risk for the entire fleet is high enough to justify grounding it. In much the same way that golfing during a rain storm almost certainly will not get you killed, but if it became a trendy thing to do the number of people being struck by lightning would be significantly higher.

To put a finer point on it, I believe a 737-MAX is still safer than general aviation, which I think most rational people aren't scared of.

https://en.wikipedia.org/wiki/General_aviation#Safety

I become increasingly anxious of flying based on the size of the aircraft, not sure that I'm an outlier there. The smaller the aircraft the less stable the flight will be, which is what most people are afraid of while flying.

Terrible comparison. Number of takeoffs of 737 Max is much lower than other planes, so numbers will naturally be skewed

Why would they be skewed? Stats are stats. I'd imagine say we have a sampling bias if there are less than 1000 flights in total, as in the kinks haven't been worked out, but this is a production aircraft, with 1000s of takeoffs each day, maybe more than 1M in total during program lifetime. Sufficient to gather significant stats.

And that is exactly the issue, that within the very short period while the 737 MAX has been active it has had major incidents.

Naturally it'll be skewed but also naturally it tells something about its risk given the safety record of the aviation industry.

Almost all planes have more crashes in their early years, and many had worse or equally bad early records. It's too early to claim the 737-MAX will be inherently less safe than other planes using a multi-decade long measure.

Which modern plane do you want to compare it with?

The 787, 747-8, A380, A350, A320neo and A220 all have zero fatal incidents. That is what I expect flying a modern aircraft.

Stop being ridiculous.

Out 300 Boeing Max delivered since it was release in 2016, there has been 2 full hull losses with full loss of life. I bet if you do the numbers it is the most dangerous way of traveling. Tesla gets shit on for autopilot and it’s killed 4 people but has 100s of thousand of cars on the road with autopilot.

> I bet if you do the numbers it is the most dangerous way of traveling

That's just crazy talk. If it's more dangerous than a car I'll eat my hat.

According to NYT/Flightaware, there were ~8600 737Max flights per week. Let's call it 8000, and limit our analysis to the past year. Let's also make the very low assumption there are only 100 passengers on each.

8000 * 100 * 50 is very conservative 40 million passenger trips per year. And there were 350 deaths.

Cars have a fatality rate of ~1 per 100 million miles. That means that the average distance traveled needs to be ~1000 miles to be better than cars.

I challenge you to find better numbers estimates, but my very conservative numbers says they're on par or better than cars. I must confess I'm a little surprised how close it was.

I thinks it’s more comparison to compare travel time than travel distance when measuring safety. If we invented teleportation device that could whisk humans billions a mile way to the next star but killed 1/2 of the people that used it, it shouldn’t be considered the safest way of traveling.

Air travel should be much safer than driving. That it barely is in the case of the 737 MAX is itself a huge indictment.

Absolutely 100% agreed. I started with that assumption and figured that even with very conservative numbers I'd be well below the number of driving fatalities. I did that calculation as I wrote that comment and didn't retroactively edit my initial reactions or change any numbers to fit my bias.

I'd love to see someone figure out the real numbers here.

Yeah, I was just emphasizing your point that you arrived at organically while doing the calculations. Definitely on the same page as you.

"Most dangerous way of traveling"" This is the kind of alarmist stuff that gets perpetuated on social media and reddit.

https://finance.yahoo.com/news/boeing-737-max-8-compares-142...

"Well-known and popular commercial aircraft have been involved in far more fatal accidents than the Boeing 737 Max 8. The 3,065 fatalities onboard the Tupolev Tu-154 are more than any of the 47 aircraft models Quartz analyzed. But commercial staples from Boeing, the 737-200 and 747-200, rank second and third on the list, with 2,910 and 1,664 fatalities respectively."

Counting absolute numbers makes no sense, in the long run any plane model will kill an infinite number of people; what matters is the rate - that is, the slope in that first graph - and the MAX8 has the steepest. As they write, "Compared to the planes involved in accidents with the most fatalities since 1966, the 737 Max 8 has had more fatalities in its first years in service than any of the other."

Also, conditions outside of the plane should matter. The Tupolev you listed was typically poorly maintained and flown to places like Irkutsk or Arkhangelsk, taking off from and landing onto short or poorly maintained runways, often in winter, with technology available in the sixties to the Soviets. The fact that more deaths happened sooner on a modern aircraft built with state-of-the-art technology, flown primarily by its largest users on routes like Dallas to Miami, should be somewhat alarming.

There is the comment below about the 4 order of magnitude difference between A320 vs. 737MAX. And there is confirmation bias that can save your life.

Magnitudes mean very little when numbers are tiny in the first place

You aren't one of those who say people should “vote with their wallets”, are you?

Reply on news.ycombinator.com