Safety at large flight schools vs. small schools / clubs

[Beaker]

Per a report from AOPA, the overall flight instruction fatal accident rate in light fixed-wing piston airplanes is 0.26 per 100,000 hours. This is based on data from 2015-2019, and is a significant improvement from the early 2000's, most likely due to improved avionics and the adoption of ADS-B. I poked around in the most recent data from 2020 to now and that rates still appears to be accurate.

Now riddle me this. The largest flight school, ATP, reports that their fleet operates 581,000 flight hours per year, but in the past five years they have only had 2 fatal accidents. One of those is an apparent suicide which I think should be removed from the discussion. So that places them somewhere around 0.03 fatal accidents per 100,000 hours. Large uncertainty bars there since I do not know how much their fleet flew in previous years, or if they are doing some funny accounting by what they mean regarding "flight hours" on their website.

Nonetheless, it seems consistent with another large school. Embry-Riddle has only had one fatal accident in the past 20 or so years, and that was solely due to the Piper McSnappy wing design defect. Their website indicates around 150,000 flight hours per year, which gets us to a similar figure of 0.03 fatal accidents per 100,000 hours.

We are flying from the same airports doing the same things, yet the safety disparity appears real. I do not have experience with these programs as I have always been at small schools or flying clubs. What is the secret sauce? How can those of us over at the smaller hangar achieve the same results?

Sources:
EXECUTIVE SUMMARY
Flight Training Facts / ATP Flight School
Aircraft and Simulators

 

[ahw01]

But GA isn’t just training?

 

[Beaker]

But GA isn’t just training?

Sorry, not following. All the numbers I mentioned are specifically regarding flight training activities.

 

[ppragman]

We are flying from the same airports doing the same things, yet the safety disparity appears real. I do not have experience with these programs as I have always been at small schools or flying clubs. What is the secret sauce? How can those of us over at the smaller hangar achieve the same results?

lot's of rules at the bigger schools

 

[killbilly]

One major contributor to a good safety record is standardization. Larger schools and programs like those place a very strong emphasis on doing things a very particular way across the board. Flying profiles, sticking to specific checklist procedures and risk management actions. They generally try to emulate the kind of standard procedures we follow in a 121 airline.

Clubs are a mixed bag. Some are more standardized than others.

 

[bp]

are these pilot error accident rates? not sure how ATP and Riddle do maintenance, but 61 GA fleets I am sure would be a mixed bag of aircraft age and general upkeep

 

[Beaker]

are these pilot error accident rates? not sure how ATP and Riddle do maintenance, but 61 GA fleets I am sure would be a mixed bag of aircraft age and general upkeep

It is a total accident rate including all causes. That being said, the AOPA report indicates system failures are only a minor contributor. The majority of the accidents are encompassed by loss of control, midair collision, and CFIT. I could see fleet age playing a factor in reducing such accidents to the extent the more integrated glass displays promote better situational awareness.

 

[drunkenbeagle]

We are flying from the same airports doing the same things, yet the safety disparity appears real. I do not have experience with these programs as I have always been at small schools or flying clubs. What is the secret sauce? How can those of us over at the smaller hangar achieve the same results?

Not flying into IMC, going to a very limited number of airports, and having students that are mostly full time removes a lot of risk I'm sure. Aircraft types almost certainly must play a role. The big box schools don't do a lot of tailwheel, aerobatic, floatplane, or other instruction in more exotic types. There's also more risk there.

Instructor training might have something to do with it as well. There isn't much standardization between instructors part 61, and I've never seen much training for them anyplace I've been.

 

[Beaker]

I read through the NTSB reports for fatal instructional accidents in non-amateur-built single reciprocating engine airplanes from 2019 onward. This covered through August of 2023, as I believe reports for accidents after that date are still being finalized. There were 51 such airplanes involved in fatal accidents during this time period. Some observations:

• 25%, were operated by a part 141 flight school. This appears to just be representative of the number of part 141 vs. part 61 operations. Per the AOPA flight school finder, only 19% of flight schools are part 141.
• Carburetor ice was implicated in 10% of the accidents. Any flight school that operates a newer, fuel-injected fleet eliminates this risk.
• 14% of the accidents were a discovery flight. I do not know how many discovery flights there are vs. normal training operations, but this strikes me as overrepresented. About half of these accidents can be attributed to non-standard instructor pilot actions, such as ignoring takeoff performance limitations or engaging in mountain flying without training or the school's approval. In at least two of the cases, the student locked up on the controls during a takeoff or go-around and stalled the airplane into the ground. Through a surviving witness or unintentional radio broadcast it is clear that the instructor was fighting for control of the airplane, but lacked sufficient time to recover. Perhaps students should not be allowed on the controls whatsoever during critical phases of discovery flights.
• 98% of the accidents occurred in visual meteorological conditions. The only IMC accident was an unauthorized student solo cross-country. While schools may have different policies regarding flight into IMC, those policy differences do not appear to affect this accident dataset.
• Similarly, 94% of accidents occurred during the day.
• 92% of the airplanes were tricycle-geared. Of the four that were not (three tailwheel and one floatplane), the type of gear was a causal factor in only one accident, where a tailwheel airplane had a failed go-around after departing the runway during landing.
• None of the accidents involved purposeful aerobatic training in an aerobatic airplane (such as a Decathlon, Pitts, etc.). One accident involved CFI spin training in a Cessna 152 where one gets the impression of the blind leading the blind, as the instructor pilot only had a couple of hours experience in spin training. This was a university program, to boot. My takeaway is to only do spin and upset recovery training at a place specializing in aerobatics.
• 10% of the accidents were collisions with other airplanes. In about half of those, at least one airplane was not talking on the CTAF and/or did not have ADS-B. About half of these also involved operations to parallel runways.
• Failure to wear a restraint, or the lack of a shoulder harness, was a factor in 4% of the accidents.
• 8% of the accidents are due to lazy maintenance. Some issues that led to the accidents, such as failed control cables or a cracked muffler, would have developed over time and been visible during required inspections. However, inspections were completed without any corrective actions.

In sum, the vast majority of training accidents occur in boring airplanes during the day and in good weather. I feel that I could mitigate 50% of these accidents through commitment to appropriate policies and procedures, such as what is likely duly enforced at the successful large flight schools. That leaves the other 50% as the boogeyman: unexplained loss of control inflight. It happened to solo student pilots, but also flights with instructors on board who had thousands of hours of experience. It is hard for me to picture how to eradicate this, given it is a big focus in training already. Maybe it is this:

having students that are mostly full time removes a lot of risk I'm sure.

Perhaps the immersive, full-time nature works in the background to improve the performance of all involved.

 

[drunkenbeagle]

In sum, the vast majority of training accidents occur in boring airplanes during the day and in good weather.

That's also when the vast majority of training flying happens. It's pretty lonely up there at night in a piston single. Even lonelier when there are clouds.

 

[drunkenbeagle]

. What is the secret sauce? How can those of us over at the smaller hangar achieve the same results?

I treat every training flight like it is the one that is going to kill me. Discipline, procedure, flying the 1000th time like it was the 2nd time doing something. I don't know, the longer I've done this, the more careful I have become. And I was already pretty careful.

 

[///AMG]

Scott Crossfield died in a (edit) Cessna 210, after flying into convective activity. That’s a wild reality check when you think of the things that he did in his military/North American/NACA career. If you can fly the X-15 on a bunch of manufacturer test flights, surviving a crash landing, as well as a a test stand rocket engine explosion of the anhydrous ammonia/LOX tank that sent your cockpit down range a hundred feet, and die decades later in a simple GA plane……..that should give any of us a healthy respect for any flying machine

 

[ppragman]

Scott Crossfield died in a (edit) Cessna 210, after flying into convective activity. That’s a wild reality check when you think of the things that he did in his military/North American/NACA career. If you can fly the X-15 on a bunch of manufacturer test flights, surviving a crash landing, as well as a a test stand rocket engine explosion of the anhydrous ammonia/LOX tank that sent your cockpit down range a hundred feet, and die decades later in a simple GA plane……..that should give any of us a healthy respect for any flying machine

Getting high utility out of airplanes that weren't really designed for the operating environment they were being used in tends to be risky AF... and GA is doing that every day in spam cans designed (and often manufactured) well before most of the pilots were born.

 

[NoStar]

I've seen big schools get better safety simply by sticking to strict checklists and limiting risky flying, which small clubs can copy too.

 

[NoStar]

I've seen big schools get better safety simply by sticking to strict checklists and limiting risky flying, which small clubs can copy too.

and ps...
standardization and a focus on fundamentals are the real keys at big schools. I found that drilling the core knowledge through a solid ppl written test prep helped me build that same procedural discipline.

It forces you to think systematically about weather, performance, and regulations before you ever get in the cockpit, which mirrors that airline-style mindset.

 

[///AMG]

I think a big difference is a focus on basic air work. I was a part 141 guy from the start. Did my PPL that way. And then my flying partner student person that joined me for commercial ticket stuff, had done his PPL at a 61 school. His BAW was beyond atrocious. Couldn't hold an altitude within +/- 1000 at all. I think he just didn't know how to trim an airplane at that point. He was a smart dude though, and figured it out real quick. I'm sure he is a TOS CA at some major at this point, this being 25 years ago. Anyway, point being, I feel like 61 schools tolerate a lot of crap, because they need the revenue because they are small. Ironically, the only two GA accidents I personally knew were 1) a homebuilt who decided to do his test flight over mountainous/forested terrain when the fuel filter/fuel system clogged, and 2) my wife's little sister's friend who decided to fly high AF on pot trying to takeoff well above max GW on a short grass strip into a line of trees. Both, or all I should say, are resting in peace underground now. That's the kind of dumb • that this profession doesn't tolerate.

 

[MidlifeFlyer]

I see two things that have been mentioned. Standardization of instruction and procedures is a big factor. You can see that even outside the big box school environment. Cirrus had a horrible safety record in the early days. Standardized instruction and providing it as part of the purchase package made a big difference there.

The other is a student body that is basically full time with a professional career goal. That dovetails nicely with the first one, since the pro world is, pretty much by necessity, standard-procedures oriented.

I don’t know to what extent it’s universal, but there’s often a “team spirit” factor in a big school. There’s an ATP at my home base, so I see and talk with their people. Seeing 4 sitting at a table during a checkride debriefing picking up tips and discussing best practices is not at all unusual.