Rate of fatal accidents in Mooneys over time

[DXB]

I thought this would be an interesting albeit somewhat grim graph for discussion.  I tabulated the number of fatal accidents in Mooneys per year globally, averaging over 5 year intervals to help smooth out the data.  Obviously there's many factors here impacting these stats other than safety of operation, such as fleet size and number of hours flown.  There is still some interpretable information here.  We are certainly doing better than the peak of 30 (!) fatal accidents in '1968, which seems staggering to me given the relatively small size of the fleet at the time.  However improvement over the last two decades is hard to discern despite the vastly improved resources in terms of flight planning, training, maintenance, avionics reliability, and cockpit data that became available during that period.  

Improvements in rescue, trauma/critical care and thus survivability may also contribute to the favorable trend slightly.

 

Edited September 19, 2021 by DXB

[irishpilot]

@dxb , this is an interesting discussion. When you researched this info, did you come across the registered number of Mooney's per year? The fleet is most likely decreasing in size over the last five years.

Thanks for posting!

Sent from my Pixel 4a using Tapatalk

[DXB]

 

29 minutes ago, irishpilot said:

@dxb , this is an interesting discussion. When you researched this info, did you come across the registered number of Mooney's per year? The fleet is most likely decreasing in size over the last five years.

Thanks for posting!

Sent from my Pixel 4a using Tapatalk
 

I did not - I just used this data https://aviation-safety.net/wikibase/type/M20P

I'd be happy to repost with data normalized by registered fleet size if someone has that data handy, though that information may still be confounded by the unknown true number of operational aircraft and the number of hours flown.

*I also just realized the graph I posted misses all the turbo aircraft - which is here https://aviation-safety.net/wikibase/type/M20T/3

Adding those stats increases fatal accident rates from 1980 onward:

Edited September 19, 2021 by DXB

[irishpilot]

I did not - I just used this data https://aviation-safety.net/wikibase/type/M20P
I'd be happy to repost with data normalized by registered fleet size if someone has that data handy, though that information may still be confounded by the unknown true number of operational aircraft and the number of hours flown.

I wasn't able to find a reliable source of registered Mooney's by year.

Sent from my Pixel 4a using Tapatalk

[0TreeLemur]

If you added in M20T's your bar graph will better represent the fleet.

[DXB]

22 minutes ago, 0TreeLemur said:

If you added in M20T's your bar graph will better represent the fleet.

Done!   Edited graph above...

[Ragsf15e]

Very interesting.  I’m guessing there’s a good mix of factors helping us improve… probably decreased total hours flown being a big part.  Improved weather and flight planning -1990s on.  Improved airborne situational awareness ~2014 on (adsb in).  Maybe improved training and recurrent training mixed in there from ~1995 onwards.

Someone should pull the most recent Nall report graph for all of GA and compare.  I think it looks similar. Definitely an interesting graph, thanks!

[carusoam]

Add in the variable of continuous education…. And enhanced knowledge about your flying machine…

Members of local community flying groups, and MSers, must have some level of improvement in safety…

This is a great way to know what you are missing…. Before that missing item shows up in flight…. 

 

Fortunately, there isn’t enough accident data to go on…. Most days…  (all due respect to the latest lost Mooney airmen in the Ontario area… recently reported around here)

Best regards,

-a-

[0TreeLemur]

Nice work Dev.

Numerous confounding effects I suppose.  The effects of several eras appear on this plot:

1961-1975- Growth of the fleet.   As you make more planes, flown by more new pilots inexperienced in the M20 plus the common bravado/machismo back then, makes for a lot of twisted metal.

1976-1985: Survival of the fittest.  the Mooney pilots that didn't get killed between 1961-1976 were actually a pretty good group, but they flew a lot and continued to take too many unnecessary risks..

1986- present: Era of increased costs, improved training, aging fleet, aging owners.  Sum it all up, annual flight hours per airframe are coming down.  Fewer fatal accidents are mostly caused by poor decision making, which can never be eliminated.  That explains why the graph is coming down to a plateau at about five per year.  Here's a link to a similar thread from earlier this year.

14 hours ago, DXB said:

I

 

 

[A64Pilot]

16 hours ago, irishpilot said:

I wasn't able to find a reliable source of registered Mooney's by year.

Sent from my Pixel 4a using Tapatalk
 

You won’t either, possibly now that the FAA is requiring renewal,but before that it was a mess.

I tried as a manufacturer to contact owners ref an upcoming AD, and well less than maybe half of the registered owners still owned the airplane. that was Ag and GA is surely better, but still it wasn’t even close.

I also gathered accident data and in truth there was no way to determine actual fleet size either.

‘Total accident rate isn’t really very informative, you need accident rate per 100,000 hours etc, but how do you determine that?

I don’t think we are much safer, just less of us flying fewer hours.

 

I believe and may be wrong but it’s my opinion that the wing leveler on all Mooney’s probably did more to save more people than any other safety enhancement. At least I think the wing leveler was standard for several years?

Edited September 19, 2021 by A64Pilot

[steingar]

I think that graph represents decreased fleet size more than anything else.  The penetrance of in cockpit weather probably helps as well.  

[dominikos]

Also, reading Kathryn’s report you cannot help but reflect on how much longer the current GA fleet will last. Not many can afford the new planes and you can see 10-20 a week getting damaged and destroyed.

[steingar]

Yup, I fully expect to be among the last middle class hobbyist pilots.

[Alan Fox]

The biggest increase in survivability , is the shoulder belt.. 

[Alan Fox]

On 9/20/2021 at 1:29 PM, dominikos said:

Also, reading Kathryn’s report you cannot help but reflect on how much longer the current GA fleet will last. Not many can afford the new planes and you can see 10-20 a week getting damaged and destroyed.

They couldn't afford them 50 years ago , why do you think its any different now ??  Any Idea how many Cirrus , Cessna single , Piper single , Diamond singles , Robinson helicopters , Etc. etc . etc ???  

 

[aviatoreb]

Interesting post revival.

One thing that strikes me of the fact that the accident rate is not decreasing to zero.  The safety of the plane, available data, maintenance, etc etc you might thing the rate should be decreasing to zero.  But there is a relative constant also built in to the rates, the weak link, which is the constant of human stupidity.  There is a certain (large?) fraction of people that just do dumb stuff.  Said plainly, even if we drive to almost zero all the other sources of problems, there will always be a few folks that thinks its fun to buzz their girlfriends house and crash in their front lawn.

[A64Pilot]

16 hours ago, Alan Fox said:

They couldn't afford them 50 years ago , why do you think its any different now ??  Any Idea how many Cirrus , Cessna single , Piper single , Diamond singles , Robinson helicopters , Etc. etc . etc ???  

 

They actually could afford them 50 years ago, as a kid going to the local airport on a weekend you couldn’t wash your airplane unless you got there early. Too many in line. There would be 10 pilots in one of the two FBO’s discussing the exact same thing we argue about now, like reduced power climbs and leaning it out etc. 

So much traffic there were two follow me trucks that would race to the taxiway vying you to follow them to their FBO, of course two full maintenance shops on the field, and one full blown Avionics shop. Full blown pilot school with a fleet of newish Beechcraft’s to train in and rent and several CFI’s on staff, They sold the aircraft and bought new about every two years, who would rent an old airplane?

Same town today there is one FBO locked behind what looks like a prison fence that services the occasional Biz Jet, I think there may be one aging A&P/IA associated with it. Avionics shop is gone, no traffic to speak of on Weekends and you occasionally see some old Dr dragging his Baron out of a T hangar, no rental aircraft, maybe one or two beat to death OLD 172’s but I don’t think so. Incidentally the towns population is more than double what it was in 1970. Back then before you left to go to the airport you called the FBO and they would pull your aircraft out of the hangar and put it on line, it was part of hangar rent, when you got back they would top it off and put it away. No T hangars back then, but then you didn’t pump your own car gas either, or put air in your tires or check oil.

Different, more affluent world for Joe Average

There were more aircraft manufacturers doing a good business than there were Auto manufacturers.

Average airplane was five or less years old, but most pilots were plumbers, electricians, house builders, Pharmacist's and of course Doctors. 

Price of a 172 in 1970 was $12,500, about half the price of the average house.

Median price of house now is supposed to be $380,000, can you get a new 172 for half that now? If you could I bet airplanes would be selling.

https://www.statista.com/statistics/272776/median-price-of-existing-homes-in-the-united-states-from-2011/

Oh, and average age of GA aircraft isn’t 5 years old anymore, it’s more like 50 years old according to the FAA, and there are only a couple of GA manufacturers, most that are still building GA aircraft make their money from building Biz Jets but still build the rare single engine piston out of tradition I guess.

https://www.faasafety.gov/files/events/EA/EA68/2023/EA68121005/EA68121005F.pdf

 

[GeeBee]

The only trouble with this whole thread is it is "accident rate". Which is the number of accidents per flying hour, or the number per capita of the fleet, or the number per Mooney pilot but we have here is only accidents per year without a proper denominator. Bias the number of accidents vs fleet size or even better per flying hour then you would have something to discuss.

[Schllc]

On 1/1/2024 at 10:55 AM, GeeBee said:

The only trouble with this whole thread is it is "accident rate". Which is the number of accidents per flying hour, or the number per capita of the fleet, or the number per Mooney pilot but we have here is only accidents per year without a proper denominator. Bias the number of accidents vs fleet size or even better per flying hour then you would have something to discuss.

“There are lies, dam lies, and statistics.”

Mark Twain

[aviatoreb]

11 hours ago, Schllc said:

“There are lies, dam lies, and statistics.”

Mark Twain

It's a fun quote.  But Mark Twain was a writer and didn't know a damn thing about statistics.  In the hands a of a good data analyst, mathematician, statistician, statistics are an excellent and necessary part of empirically analyzing large complex issues.  But since there are complicated issues, mediocre practitioners can really muck it up and yes charlatans can lie with statistics which means deliberately obfuscate truth in a shrowd of what seems like knowledge.

[Schllc]

1 hour ago, aviatoreb said:

yes charlatans can lie with statistics which means deliberately obfuscate truth in a shroud of what seems like knowledge.

I think this was the genesis of the quote.   I absolutely believe statistical analysis is both valid and helpful, but crap in, crap out, and if you are going to predicate a decision on the statistics, it would be helpful to understand the sample set.

[A64Pilot]

13 hours ago, Schllc said:

“There are lies, dam lies, and statistics.”

Mark Twain

When I lived in Germany my German neighbor told me Germans have a saying

”Statics are like a lady of the evening, if your paying you get what you want”

Sometimes when you see a study it’s often illuminating to see who funded it.

[Pinecone]

Old joke.

Company is hiring a statistician.  Three make the cut to be interviewed.

First one comes in and the manager asks, "What does 2 plus 2 equal?" The candidate says "4 of course."  

Second one comes in and is asked the same questions and also answers 4.

The third one comes in and is asked the question.  He gets up and closes and locks the door.  He closes the window shades.  He turned out the overhead light. He sits back down and leans across the desk and whispers, "what do you want it to be?"

Quess which one got the job? 

 

 

[Ragsf15e]

4 hours ago, aviatoreb said:

It's a fun quote.  But Mark Twain was a writer and didn't know a damn thing about statistics.  In the hands a of a good data analyst, mathematician, statistician, statistics are an excellent and necessary part of empirically analyzing large complex issues.  But since there are complicated issues, mediocre practitioners can really muck it up and yes charlatans can lie with statistics which means deliberately obfuscate truth in a shrowd of what seems like knowledge.

I agree statistics and testing are a powerful tool, but I’m just finishing my masters in economics, and the last class I had on econometrics was disheartening.  There are so many pitfalls, methods, tests, choices, and grey areas that it seems like two things happen… either the researcher can make the end result say what they like (to a certain extent) or someone with less experience will just muff a very complicated situation and draw the wrong conclusions.  Omitted variable bias and selection bias are easy ones to understand and see how they screw up everything.  In the end, I feel like it’s the experience and impartiality of the researchers as much as the data they have.

[aviatoreb]

46 minutes ago, Ragsf15e said:

I agree statistics and testing are a powerful tool, but I’m just finishing my masters in economics, and the last class I had on econometrics was disheartening.  There are so many pitfalls, methods, tests, choices, and grey areas that it seems like two things happen… either the researcher can make the end result say what they like (to a certain extent) or someone with less experience will just muff a very complicated situation and draw the wrong conclusions.  Omitted variable bias and selection bias are easy ones to understand and see how they screw up everything.  In the end, I feel like it’s the experience and impartiality of the researchers as much as the data they have.

I am a math professor - not a statistician - but I am quite experienced in statistics and probability.  Any statistics has some kind of assumption underlying what was the distributions of the random variables from which the data was sampled.  Rarely are those idealized assumptions perfectly true for real world data - but maybe mostly true?  And there are sometimes statistical tests for those assumptions.  Often all that is glossed over and full speed ahead damned the torpedos and use the stats anyway.  With lots of charts and graphs showing outcomes of plugging numbers, the raw data, into what is often inappropriately chosen statistics.  And when shown to an unsuspecting consumer, whether company executives, or public, this is where Mr Twain does have a point.