Sad but quite interesting (French accident statistics)

Sure, there is the old saying of ‘never trust statistics you haven’t faked youself’. Anyway, the point rather was not to use the number of hours flown as the primary parameter of considering someone as ‘at risk’.

The “Killing Zone” book contains dodgy statistics.

A lot of stats do get deliberately faked (drug trials come to mind, with so many drugs barely better than placebo) but just because something can be faked it doesn’t mean that the subject of statistics has no value. One just has to understand the process used, and understand what (if anything) one can “take home” from the numbers.

When it comes to accidents, statistics are difficult to apply to oneself because they come from a population, and in GA it is fairly easy (by self improvement, risk compensation, or just not doing certain stupid things) to slant the odds in your favour.

Peter wrote:

… just because something can be faked it doesn’t mean that the subject of statistics has no value. One just has to understand the process used, and understand what (if anything) one can “take home” from the numbers.

Amen!

Well said Peter.

While the overall risk of flying GA is regularly compared to motorcycling, the major difference is that your risk in GA is almost wholly dependent on your own behaviour (with technical defects that the pilot has no influence on being a lesser reason for accidents, and direct third party influence such as an airborne collision being even less likely) whereas in motorcycling the behaviour of other drivers is a major risk factor (yet anecdotally, most motorcyclists who reach our hospital after an accident, often heavily injured, were not totally blameless for what happened either).

So one’s behaviour is essential to managing the risk. I for one know that I’m not the most skilled or current pilot, so I make careful risk assessments before flying and try to mitigate the risk by flying only in decent weather and setting simple tasks/goals that any PPL should be able to perform. As a renter, the school is also an additional layer of protection because sometimes they simply phone me up on a day I had booked a plane and tell me things like “It makes no sense to fly today, the wind is already at 35 kts and getting stronger”.

My home region also has the major advantage of offering very forgiving terrain in case of engine failure, and a simple airspace structure that does not create much of a distraction. Overall it is an environment where even a low-hours, low-currency pilot can enjoy flying for a few hours a year with very manageable risks. The more current I become, the more willing I would be to leave this sheltered environment. Of course, to stay safe, that always requires a recurrent re-evaluation of the risks at hand and the own skills and resources to mitigate those risks.

MedEwok wrote:

Overall it is an environment where even a low-hours, low-currency pilot can enjoy flying for a few hours a year with very manageable risks.

And this is where I totally disagree with you.

This thread has been about the fact that statistically it makes no odds whether you have 10000 hours, or 1. If the plane blows a cylinder, does it know how many hours the PIC has? No. If the plane has a snapped trim cable does it know if the pilot has enough bandwidth to deal with it? No.

I have witnessed many near misses with low houred pilots in the local zone who bimble about not understanding how dangerous the LFA can be. My personal view is that the circuit, and the LFA are potentially one of the most dangerous areas for flying. The overhead join, the crosswind, the long down wind where a lot of pilots end up in truly dire circumstances and often to the endangerment of others.
MedEwok wrote:

My home region also has the major advantage of offering very forgiving terrain in case of engine failure,

Are you aware of how many pilots kill themselves, or botch the power off landing, in this particular phase of flight?

Currency, training and understanding of SA are your friends. Stats will not assist you here

The “Killing Zone” book contains dodgy statistics.
A lot of stats do get deliberately faked

Aha. Ok, then skip the recommendation to read that book. Like @Peter indicated above, everyone has to decide what, if anything, to take home from statistics. Looking at accidents, the bottom line I see is that it is not the low time pilot who is the main contributor – and this is my point to @MedEwok ’s question / remark

do low-hour pilots like myself really make up a disproportional part of the accident statistics?

But @Malibuflyer clearly has a point as the question of during which time span the total hours have been accumulated got ignored.

So one’s behaviour is essential to managing the risk. I for one know that I’m not the most skilled or current pilot, so I make careful risk assessments before flying and try to mitigate the risk by flying only in decent weather and setting simple tasks/goals that any PPL should be able to perform.

Agreed…. Human behavior & decisions are the largest factor of accidents, hence showing some humility about the own skills has a healthy influence….

Exactly 31.2% of all statistics have just been made up on the spot

Seriously, I hope you’re ok gallois, and that it doesn’t stop you flying too long

This is an interesting, not polemic review of the PA-28 type.

http://www.wanttaja.com/pa28.pdf local copy

Basically stick to simple, relatively modest cruise speed SEP aircraft with modern stall/spin resistant design, maintain currency per club rules and stick to benign VFR or very light IFR if aircraft and you are qualified.

Within this envelope the accident record is probably statistically better than motorcycles. A statistician would need to prove this for robustness.

Apologies for the mother and apple pie input.

I think the best thing a pilot can do is read the accident reports. And if you think the BEA ones aren’t much good then read the AAIB (or any other country’s) ones

Then, when you’ve read say 100 of them and can say to yourself “I would not have made any of those cock-ups”, you will probably live another day.

Population stats aren’t of much use for learning how to not screw up. They can be distorted by all kinds of things.

For example, hypothetically but IMHO not entirely hypothetically, let’s say that every school is teaching people how to fly circuits which are 20nm in length, and with a 5nm final. Those pilots, when having to land somewhere “real”, are likely to kill themselves. So you will have stats showing that landing is really dangerous, but actually the training they are receiving is crap for the real world. You could paint a similar picture for navigation, with crap techniques being taught, but the stats showing lots of people getting lost.

And that’s before you get onto stats involving human behaviour i.e. “social sciences”. This is incredibly hard to do well, not least because one can’t control for various common factors.

A low hour pilot can fly safely and enjoy it. Whether he/she will feel they are getting value for the money/hassle is a different matter – because they will be flying only on sunny Sundays and only down the road. However I have known many people who enjoy “just flying”. They take the plane out of the hangar, fly for 20 mins in circles, and put it back in the hangar. And they are quite happy with that. It is better than playing golf. However I am told there are more single+looking women in golf than in GA and that must be a self evident fact, along with the mass of the electron, etc