Actually, flying is very safe.

[msjh]

An interesting snippet from "The Spectator" magazine.

The loss of a second Malaysian Airways airliner means that the number of worldwide casualties in civilian air disasters — 827 so far this year — has already exceeded that of last year, when 459 died. But it does little to disturb the steady downwards trend. These are the deadliest years for air passengers since 1945:

1972 3,344 deaths
1985 2,962
1973 2,815
1999 2,798
1974 2,621
And these are the least fatal:

2013 459 deaths
2004 767
2012 800
2011 828
2008 952

This, of course, will be of no comfort to those who have lost loved ones but may help us remember flying is still extraordinarily safe.

[cockney steve]

On their own, these statistics are pretty meaningless. Comparing the number of deaths with everyday situations to which Joe Public can relate,is a far more relevant benchmark.

It is self-evident that Flying is safe- were it not so, there would not be so many willing to sit at the pointy-end, guiding the Aircraft to the scene.....
Flying would be like oil-well fire -extinguishing , the province of a very few.
The fact that Commercial Aviation is a mainstream worldwide form of transport, would tend to support this view.

So, what's new?

[susier]

You need a graph. Obvs 2014 is (hopefully) a bit of an anomaly in the general trend, and it isn't shown here, but most people would be able to see the bigger picture from this.


I imagine the topic will be moved though

[acmech1954]

It would also be interesting to see the detailed number of flights/passengers carried proportional to casualties.


As if by magic the graph appears

[barit1]

It's meaningful also to break it down by geographical region.

[Nakata77]

But?

What if we do the calculation based on fatal accidents per unit of aircraft manufactured... Isn't it then alarmingly high?

For example if only 1,000 aircraft (of a given type) are produced, maybe 1% will be involved in fatal accidents over their life time.

Can we do the same calculation for ship, train, car, bus etc? Tens of thousands of cars, busses and trains are produced for example.

[HighSpeedAluminum]

Here is a risk statistic from US DOT.

http://www.phmsa.dot.gov/portal/site...gnextfmt=print

Every day ordinary folks should be able to grasp this statistic when it comes to risk of death in an air carrier accident. You would have to fly every day for 55,000 years (yes that number is correct) before you would be involved in a fatal aircraft accident.

[msjh]

What if we do the calculation based on fatal accidents per unit of aircraft manufactured... Isn't it then alarmingly high?

For example if only 1,000 aircraft (of a given type) are produced, maybe 1% will be involved in fatal accidents over their life time.

I don't have the figures, but I doubt it. The chart (post 3) shows a substantial decline in fatalities in the last four decades while the number of airliners has increased enormously.

[DaveReidUK]

You would have to fly every day for 55,000 years (yes that number is correct) before you would be involved in a fatal aircraft accident.

It's very unlikely that any passenger ever killed in the history of air transport has previously flown 20 million times.

Hint: the word you are looking for is "average"

[Evey_Hammond]

I've seen similar stats before and know that when compared to, say, driving a car, you are more likely to die in a car crash than a plane crash. But whilst the statistics are logical, for many flying (& dying) are more emotional .

I also wonder if the perception that "flying is dangerous" may come about due to the sheer number of people who can die in a single plane crash (e.g. in the hundreds) whereas it's (usually) much less in a car crash.

(unfortunately none of the above apply to me - I'm terrified due to "control issues" )

[cockney steve]

also wonder if the perception that "flying is dangerous" may come about due to the sheer number of people who can die in a single plane crash (e.g. in the hundreds)

ERR....Bus, Coach, Railway, passenger-ship (costaconcordia?)
Ferry (Herald of free enterprise).......just a few examples where one does not have "control"....At least,the operators do their best to make sure their operators (drivers) are reasonably well trained, to beincharge of vehicles/vessels costing hundreds of thousands or even billions and carrying from ~50 to ~500 Pax.

Whereas any semi-literate cretin can buy a car and take his /her chances.
My late partner drove for around 40 years, Provisional license, no "l" plates, never got a "tug" and , AFAIK, never bothered with lessons or a driving test!.....There are plenty similar on UK roads that can't even speak the language!

For goodness sake, get your priorities right and worry about real risks!

[barit1]

Nakata77:

What if we do the calculation based on fatal accidents per unit of aircraft manufactured... Isn't it then alarmingly high?

For example if only 1,000 aircraft (of a given type) are produced, maybe 1% will be involved in fatal accidents over their life time.

"per unit of aircraft manufactured" isn't too meaningful when the utilization of the aircraft is considered. I'm not up to date on annual numbers, but I wouldn't be surprised if a T7 might spend 50% of its time airborne - I.E. 4000 hours per year. How many cars are driven 4000 hrs in a year? Right away an apples-to-apples comparison doesn't work!

[CrashManII]

I also wonder if the perception that "flying is dangerous" may come about due to the sheer number of people who can die in a single plane crash (e.g. in the hundreds) whereas it's (usually) much less in a car crash

When my career started to depend on having to fly frequently, I had to overcome my fear of flying. Read a couple of really good books that were incredibly helpful. One mentioned that you rarely see coverage of a fatal car crash....but if an airplane has any trouble and footage is available they'll play it a million times. The visual imagery "locks" in your brain giving one the impression that planes crash all the time. They don't.

They also suggested not watching repeated showings of any air crashes. Just change the channel.