Hacker News Reader

Comment by TillE | Hacker News Reader

I think there needs to be a more selective filter if this data is supposed to mean anything. The evaluated incidents include terrorist bombings, the deliberate Germanwings crash, etc.

View on HN

Think it's difficult to make much of the data even with a filter. A lot of other incidents are due mainly to pilot error, bad weather conditions or inadequate maintenance, but the aircraft don't altogether escape blame for those types of incident. Actual incidents known to be down entirely to bad aircraft design are exceptionally rare amongst late generation aircraft (but some of the more mysterious crashes might be).

The ATR figures are inflated by their flight cycles info being years out of date (and based on knowledge of what ATR supplied to commercial databases, likely incomplete), but you'd also expect them to be higher than some of the other aircraft types purely because of how they're used (flown into smaller airstrips in hilly regions). At the other end of the scale, the 747-400 performs excellently with only two passenger crashes, but it's also had four fatal freighter crashes not counted. I bet the calculated "rate" includes freighter flight cycles though

But above all, airliner fatal crashes are exceptionally rare, and some of the near misses and non-fatal incidents are more indicative of design flaws

My first reaction on seeing the fokker stats was 'well, yeah, it's heavily used in the Alps and Himalayas, of course it's had a lot of incidents'

Ironically, the 727s numbers are probably skewed in the opposite way to the 747 - there were loads of passenger incidents with the 727, but it's been somewhat of a workhorse for freight in the developing world over the last 40 years.

Would those incidents be frequent enough to make anything but a de minimus impact?

If I'm doing the math right, there were only 3 fatalities on a 777 if you don't count MH370 or the Ukraine/Russian shoot-down. That would change the FLE from 2.01 to 0.01.

I understand not wanting to count a shoot-down but why not count MH370? For all we know there was a safety issue with that aircraft that caused the crash.

After almost a decade of investigation the preponderance of evidence points to murder/suicide by the pilot. A safety issue is highly unlikely.

This is a very USA/western-centric perspective. No SE Asia reporting is stating this.

Isn't that because suicide is a taboo subject in many countries? This was at least the case for EgyptAir Flight 990 and SilkAir Flight 185, and IIRC people expressed the same thing about Malaysian culture.

What do they say?

Why? What are the alternatives?

As I recall, there's a fair amount of circumstantial evidence at this point to suggest pilot suicide as the most likely cause.

We basically know that it was suicide at this point.

One of the three 777 accidents is being shot down by a Russian SAM.

Another is the infamous MH370.

Another is the pilot error on Asiana flight in San Francisco.

Another one is the crash at Heathrow, but that was Rolls-Royce’s fault

Off-by-one error detected

The entire dataset is microscopic. I'd say it could make a substantial impact.

A really small datasets shouldn’t be used for statistics. But is it small, considering there are millions of flights registered?

You need a larger dataset if you're dealing with very small probabilities. If you measured millions of events and only saw 1 anomalous event, you have enough data to say that the probability of such an anomaly is extremely rare, but you don't have enough data to compare it to other very rare events.

More concretely: if you flip an unfair coin 1 million times and get a single heads, you know that the odds of getting that heads are extremely low, but you can't yet say what the odds are. It's possible that the odds are one in a million, but it's also possible that you got very lucky or very unlucky relative to the actual odds. You have to have a lot of data in both buckets before you can distinguish.

Millions of flights, but less than ten crashes for most, and it was pointing out that even some of those are incorrectly categorized. When you're dividing A by B to get ratio C, you cannot ignore that A is bad data because B is good, it still makes for a C that is useless.

All this dataset provides is that with small enough samples it is too noisy to draw reliable conclusions.

Depending on plane, it could.

For example, 777 has had very few incidents, but its figure is impacted by MH370 and MH17.

747's number is also skewed (by about 10%) due to 3 incidents involving bombing and missile shootdown.

By and large though, most plane crashes are due to human error, mostly by pilot. Terrorism, as well as maintenance or engineering defects are pretty rare.

Probably, given how rare any fatality in commercial aviation is.

You said:

> The evaluated incidents include terrorist bombings

From the site:

> Excluded would be events where the only fatalities were to crew members, hijackers, saboteurs

The numbers in the "Events" column are not the same numbers as are on the linked Events page, if that makes sense.

Those statements aren't at odds are they? Terrorist events that results in death of at least one passenger are included, despite terrorist activity (presumably) having nothing to do with specific model of airplane.

It includes fatal accidents caused by hijackers or saboteurs, provided they manage to kill passengers and not just themselves

I wondered about that, because isn't the hijacker technically a passenger too? I wonder if their deaths are excluded from the data.

Reply on news.ycombinator.com