Modern Generation Accident Statistics

[TyroGyro]

[first forum post]

I am almost ready to commence training, after a few trial flights. I have been watching developments closely for over five years, trying to get a handle on (fatal) accident stats for the modern generation gyros. More out of curiosity than anything, as I have a stats background, and to be able to satisfy myself that I have evaluated the risks as accurately as I can. I do acknowledge that the risks have fallen dramatically since the Bensen days. But by how much?

One of the pieces of the jigsaw I am having to guess at is the average annual hours flown per craft. My ball-park guess is 50, but I have no idea if that's realistic.

Another piece of info I'm looking for is approximate start dates of manufacture, and growth rates. [I estimate about 5% annually]

I am focusing on the Big Three manufacturers (AutoGyro-GmbH, Magni and ELA), although I recognize there are others appearing all the time.

I notice a few users are keeping tabs on the number of craft flying worldwide, and I would estimate around 3700 * in total currently for the abovementioned three types. [About 250 in my own country, the UK]

I've come up with a model to estimate fatals per million hours flown [a common metric applied to other forms of GA and transport in general].

Before I publicly churn out some numbers, I would appreciate comments on my general approach, and best estimates for the hours, etc I mentioned above.

Safe Flying!

UPDATE: * maybe my initial estimate of 4000 was a tad high.

 

[Vance]

Welcome to the Rotary Wing Forum Rod!

Welcome to the Rotary Wing Forum Rod!

[first forum post]

I am almost ready to commence training, after a few trial flights. I have been watching developments closely for over five years, trying to get a handle on (fatal) accident stats for the modern generation gyros. More out of curiosity than anything, as I have a stats background, and to be able to satisfy myself that I have evaluated the risks as accurately as I can. I do acknowledge that the risks have fallen dramatically since the Bensen days. But by how much?

One of the pieces of the jigsaw I am having to guess at is the average annual hours flown per craft. My ball-park guess is 50, but I have no idea if that's realistic.

Another piece of info I'm looking for is approximate start dates of manufacture, and growth rates. [I estimate about 5% annually]

I am focusing on the Big Three manufacturers (AutoGyro-Gmbh, Magni and ELA), although I recognize there are others appearing all the time.

I notice a few users are keeping tabs on the number of craft flying worldwide, and I would estimate around 4000 in total currently for the abovementioned three types. [About 250 in my own country, the UK]

I've come up with a model to estimate fatals per million hours flown [a common metric applied to other forms of GA and transport in general].

Before I publicly churn out some numbers, I would appreciate comments on my general approach, and best estimates for the hours, etc I mentioned above.

Safe Flying!

I feel there is value in collecting the statistics although it is hard with such a small sample.

I don’t feel you are too far off with 50 hours per year although there are people who fly more than 200 hours per year.

In the USA the average experimental aircraft flies 50 hours per year according to the EAA.

How many gyroplane hours people involved in accidents have had?
How much training they have had?
How long ago? are of particular interest to me.

I am interested in all gyroplane accidents because in my opinion the difference between a fatal and nonfatal accident is often a matter of luck.

Thank you for taking on this task. I have no doubt it will help you to become a safer pilot.

Your efforts will help to identify weaknesses in training, aircraft and recurrent training.

Steve on the forum would have a good idea of the total numbers for the aircraft you are interested in.
http://www.rotaryforum.com/forum/member.php?u=14733

 

[PW_Plack]

Rod,

I think you're on the right track. The paradox is that as the fleet gets more hours per year, the number of accidents can be expected to rise, but as an individual flies more hours per year, his personal risks per hour probably drop considerably.

I get the feeling after watching the sport for a number of years that many Bensens were brought out only once or twice a year, (or once every few years,) and the rust on both the machines and the skills of the pilots became an increased risk factor. That seems to be less common with the newer two-place machines, which seem to get flown often, and whose pilots are more likely to have had formal training.

 

[TyroGyro]

Thanks for the responses, so far.

It was difficult enough tracking down records of fatals, but I take your point Vance. Version 2 of this analysis might do that, although I already have some ball parks for total notifiable accidents (x4 of the fatals, I'd hazard, from my trawling of the databases) @Steve_UK 's stats were invaluable on the number of machines, and form a core part of the analysis.

It's a high level overview at the moment, and no doubt can be refined in future to almost any level of detail you want [my training schedule permitting ]

But here's a feel....

I estimate the global risk on these three machines collectively to be about 38 fatals per million hours flown, perhaps 3x the estimate for GA generally (in the US, at any rate). However, that is something like a 10-15 fold improvement over the "bad old days" of homebuilding/self-training. UK stats seem to be somewhat better than average, fwiw.

Here is the spreadsheet, where you can alter a couple of input parameters.
https://docs.google.com/spreadsheets/d/1o8Z28sy8UJYsyBMsZpmKiJbcsv8-6jLJPrtZYw8H6tc/edit?usp=sharing

So far, so good. However, I would also think about the "Alternative Analysis" at the bottom of the Methodology sheet... VERY sobering...

It's not perfect, maybe a bit like Fermi's "How many piano tuners are there in Chicago?" question for his students. But it's a start...

Feel free to shoot me down guys.

UPDATE: I've added a list of accidents upon which the study is based, to the sheet. It should be editable by those with the link [above]. Those with more understanding than me might like to update the info. I've added a new column for "Classification", which might be helpful, if anyone knows a standard nomenclature. Btw, looks like we're a lot better off not flying in either France or South Africa...

 

[HighAltitude]

Another important stat that is very difficult to calculate is what percentage of all accidents are fatal. Another trend seems to be that a pilot error that is recoverable in a fw, is a fatal error in a gyro most of the time.

My gut feel tells me that the percentage is very high, maybe 80% for both scenarios above.

I won't jump in for two reasons, my home altitude and the fact that pilot errors seem to result in death (very unforgiving).

 

[Smack]

So TyroG,
I don't quite understand your goal here.
You say you are about ready to commence training, but you want to run some statistical analysis first?
So, if you reach some critical number, are you NOT going to commence training?
If you happen to drive an automobile, did you also run the analysis for how many people die in car accidents BEFORE you started driving?
Really, yes, there is some risk as there is in all of the really fun stuff one can do, but don't let that hinder you. Get a good Instructor and wade in.
Don't focus on general stats that don't directly affect YOU.
Welcome.
Brian

 

[TyroGyro]

Another important stat that is very difficult to calculate is what percentage of all accidents are fatal. Another trend seems to be that a pilot error that is recoverable in a fw, is a fatal error in a gyro most of the time.

My gut feel tells me that the percentage is very high, maybe 80% for both scenarios above.

I won't jump in for two reasons, my home altitude and the fact that pilot errors seem to result in death (very unforgiving).

I don't think the stats bear that out Tim, at least for the three modern machines included in this analysis. During collection of the fatals data, I did make (cursory) note of the total number of accidents recorded on
https://aviation-safety.net/wikibase/wikisearch.php

I think the figure was 163, including the 47 fatals, a ratio of 3.5.

That does not seem drastically different from the ratio of about 5 recorded here for (US) GA.
https://www.aopa.org/about/general-...l-aviation-safety-record-current-and-historic

Such difference that there is may be down to recording methodology differences, the completeness of the wikibase, or chance.

The ratios look broadly similar to me.

Also, a notable characteristic of gyro accidents is that many occur on the ground! [and are non fatal.] As for the fatals, it is very rare to find one that involves an "unrecoverable manoeuvre" in the air... [more analysis required on this subject, however]. Gyros... stable in the air, UNstable on the ground?

 

[TyroGyro]

So TyroG,
I don't quite understand your goal here.
You say you are about ready to commence training, but you want to run some statistical analysis first?
So, if you reach some critical number, are you NOT going to commence training?
If you happen to drive an automobile, did you also run the analysis for how many people die in car accidents BEFORE you started driving?
Really, yes, there is some risk as there is in all of the really fun stuff one can do, but don't let that hinder you. Get a good Instructor and wade in.
Don't focus on general stats that don't directly affect YOU.
Welcome.
Brian

Thanks Brian, I think you may be over-analysing my comments a bit!

As I've said, I've been quietly watching developments for quite a number of years, wondering if the gyro environment was for me, not only in terms of time, money and commitment, but also in terms of my personal "comfort level" and understanding of risk, given the woeful historical accident record of the previous generation of machines.

The anecdotal and non-rigorous impression gleaned from the performance of the New Generation machines over the past decade is rather encouraging, sufficiently bright for me to decide to finally "go for it."

Having a stats background, I thought I may as well do a formal analysis with some hard numbers, and this forum seemed the best place to post my research. As I've intimated, this was more or less an afterthought, just out of curiosity and my general love of numbers, undertaken for the sake of completeness.

Broadly, the numbers are about where I would expect them to be, and certainly don't discourage me from beginning training, with one or two caveats. They certainly encourage me to engage in further analysis which, as Vance rightly says, can only be for the good in my future flying...

 

[TyroGyro]

Get a good Instructor and wade in.

An interesting question that got me thinking.

All things being equal (and we know they never are!) would the wise owls here, knowing what they know now, prefer to be trained by an instructor whose pre-gyro aviation experience was primarily:-

a) fixed-wing
b) helicopters
c) just gyros

Or is the question of no relevance?

 

[Abid]

Hi
I did something similar in 2012 and at that time went to Europe to collect data and stats from the clubs that govern Gyroplane flying there. It was difficult to get proper stats from France and Italy but I got something. Your percentages look similar to what I had come up with. I don't remember the exact numbers now but it was clear via stats that
1) The gyroplanes with horizontal stabs and decent designs were safer
2) pilot training and pilot attitude and decision making was the determining factor in fatal accidents than anything else. Hence pilot training and judgment were the weakest link in the chain

To be honest I was literally afraid of gyroplanes till I did that and only after that I decided it really was not the machines of today.
Welcome and best of luck. Stay safe and just fly conservatively and safely and maintain your equipment.

 

[WaspAir]

An interesting question that got me thinking.

All things being equal (and we know they never are!) would the wise owls here, knowing what they know now, prefer to be trained by an instructor whose pre-gyro aviation experience was primarily:-

a) fixed-wing
b) helicopters
c) just gyros

Or is the question of no relevance?

It shouldn't matter, if the instructor knows his/her stuff.

I started in gliders, and I think it gives excellent insight into the behavior of the atmosphere and of airfoils, but that by itself is probably a smaller advantage than a good pairing of the right instructor with student, who communicate well with each other.

 

[Vance]

A long non answer to a short question

A long non answer to a short question

An interesting question that got me thinking.

All things being equal (and we know they never are!) would the wise owls here, knowing what they know now, prefer to be trained by an instructor whose pre-gyro aviation experience was primarily:-

a) fixed-wing
b) helicopters
c) just gyros

Or is the question of no relevance?

I find my limited experience in helicopters and fixed wing aircraft helps me to recognize where unreasonable fantasies and or bad habits come from in my clients.

Even with no aviation experience most clients have formed ideas about how a gyroplane flies based on what their friends say or what they have read.

Clients with fixed wing or helicopter experience have specific things that they need to unlearn so my limited experience is helpful to identify what they are specifically doing in the gyroplane.

I don’t know if more other than gyroplane flying experience would be helpful to me or my clients.

I find great value in books like Stick and Rudder and would recommend it to anyone learning to fly.

I find flying a variety of gyroplanes to be helpful in separating aircraft caused challenges from pilot challenges.

I have flown with 26 CFIs and find the quality of instruction varies wildly. I have learned something from each of them.

My favorite instructor was helicopters only.

My primary gyroplane instructor taught in everything that flies and I feel he did a good job instilling aviation culture. He is still one of my flight instruction mentors.

My worst two flight instructors were gyroplane only and fairly low time gyroplane pilots with less than 200 hours each.

I have had gyroplane instructors teach me to fly a gyroplane like a fixed wing and I feel this is counter productive. Faster than ten knot landings and rotating on take off at some specific air speed are two things I feel are bad form for a gyroplane flight instructor.

 

[birdy]

WTF!!!
Instructing in gyros, with less than 200 hours?
Little supprise so many gyros get bent.

 

[Richard Anderson]

I got my Private Pilot fixed wing in 1977 and my Gyroplane Sport Pilot endorsement 4 years ago. Have had many instructors, most good. I've flown Cessnas, Pipers, Luscombes, Quicksilver, Mitchell Wing, Tierra, Minimax, Challenger, Magni, Calidus, MTO, Sparrowhawk aircraft. Learned something about flying from each one of them. Have had 6 unscheduled landings. Have made hundreds of friends through flying. Have had 16 friends die in aircraft. I have life insurance policies that will pay my bills even if I die in an aircraft. I try to be as safe as I can. I fly because it is something I want to do. My advice is learn as much as you can, fly as much as you want.

 

[SGK]

I got my Private Pilot fixed wing in 1977 and my Gyroplane Sport Pilot endorsement 4 years ago. Have had many instructors, most good. I've flown Cessnas, Pipers, Luscombes, Quicksilver, Mitchell Wing, Tierra, Minimax, Challenger, Magni, Calidus, MTO, Sparrowhawk aircraft. Learned something about flying from each one of them. Have had 6 unscheduled landings. Have made hundreds of friends through flying. Have had 16 friends die in aircraft. I have life insurance policies that will pay my bills even if I die in an aircraft. I try to be as safe as I can. I fly because it is something I want to do. My advice is learn as much as you can, fly as much as you want.

16? Wow! Which country? I've been around since 1981 and during this period not one of my friends died in an aircraft (if I don't reckon with two shot by a missile), knock, knock. And I have flown in most EU countries, many classes, PIC on 78 types, 3000+ hours. Does it statistically mean that I have few friends or that your country has a problem?

 

[WaspAir]

Happily, that number is not typical for civil aviation here.
I have lost only one friend to an aircraft accident since 1973.
For contrast, I have lost three in mountaineering since 2010.

 

[Vance]

16? Wow! Which country? I've been around since 1981 and during this period not one of my friends died in an aircraft (if I don't reckon with two shot by a missile), knock, knock. And I have flown in most EU countries, many classes, PIC on 78 types, 3000+ hours. Does it statistically mean that I have few friends or that your country has a problem?

Richard Anderson currently resides Eureka, Illinois, USA according to his profile.

There are lots of fatal aviation accidents in the EU Roman.

Just looking at the last ten years there is more than one fatality a year just in Sweden with a total population of around 10 million and not a lot of GA activity.

You figure it out Roman.

 

[SGK]

You figure it out Roman.

I think that I figured it out. Short Google search gave me following results (I'm sure that you can make different conclusions based on which sources you have access to):
Fatalities in general aviation:
EU, population today 741,2 million: 1036 fatalities during period 2010-2014 = 207 fatalities/year.
USA, population today 324,8 million: 400/year (someone's approximation, is it realistic, which period?)

This gave me approx. four times higher fatality rate per capita in USA then in EU.
Still, 16:0 is much more but I admit that my social life could be better, I'm socializing mostly with gyro pilots nowdays.

 

[Vance]

I think that I figured it out. Short Google search gave me following results (I'm sure that you can make different conclusions based on which sources you have access to):
Fatalities in general aviation:
EU, population today 741,2 million: 1036 fatalities during period 2010-2014 = 207 fatalities/year.
USA, population today 324,8 million: 400/year (someone's approximation, is it realistic, which period?)

This gave me approx. four times higher fatality rate per capita in USA then in EU.
Still, 16:0 is much more but I admit that my social life could be better, I'm socializing mostly with gyro pilots nowdays.

I am not going to take the bait Roman.

I think of all pilots as part of a magical experience with their passion unfettered by borders, language and culture.

There are bound to be exceptions.

 

[TyroGyro]

I think that I figured it out. Short Google search gave me following results (I'm sure that you can make different conclusions based on which sources you have access to):
Fatalities in general aviation:
EU, population today 741,2 million: 1036 fatalities during period 2010-2014 = 207 fatalities/year.
USA, population today 324,8 million: 400/year (someone's approximation, is it realistic, which period?)

This gave me approx. four times higher fatality rate per capita in USA then in EU.
Still, 16:0 is much more but I admit that my social life could be better, I'm socializing mostly with gyro pilots nowdays.

We don't measure accident statistics against the number of bods sitting on the ground. We measure them against the number of people (machines) flying, and GA is far more prevalent per capita in the US than in Europe [although admittedly in long-term decline since about 1979].

The US and Western Europe are about the safest places in the world for GA, with comparable fatal accident rates of around 11 per million hours flown. And falling...

i.e. if such a thing were possible:-

If 1 million pilots all took off for an hour's flight, you would expect 999,989 of them to live to fly another day.

Or, if you flew 50 hours a year, you might reasonably expect to fly for 1,260 years before you "bought the farm", as a 50/50 bet...

Although, as I've posted above, New Generation Gyros still seem to be somewhat less safe than overall GA. [further analysis required, since this study is based on global statistics, including accidents in places like Kazakhstan, Russia and Saudi Arabia, not normally known for their "western" standards of training, safety or statutory supervision]