4

I was recently flying with an ex commercial pilot. He told me that the only person he'd flown with that was worse at maintaining a heading was another ex A320 pilot with 20k+ hours. His reasoning was that both him and I watch SkyDemon instead of picking a point on the horizon.

It got me thinking, for commercial pilots, do total hours really reflect the pilots skill level?

Take Bob and Barry:

  • First Officer Bob flies an A320 and does the same 30 minute domestic route every day. He averages 8 take-offs and landings a day and has 2000 hours logged.
  • Barry is a captain with 20,000 hours logged, also flies an A320 and only does long haul flights averaging 13 hours each way. He performs an average of 1 take-off and landing per day.

If total hours aren't the best measure of skill, what other metrics might best reflect a pilot's skill level?

fooot
  • 72,860
  • 23
  • 237
  • 426
Cloud
  • 13,280
  • 8
  • 81
  • 150
  • 4
    I seem to remember that skill fade of long haul pilot is an often studied area of flight safety. I'll find some links. – Jamiec Jan 29 '24 at 14:26
  • @Bianfable NO that does not - I'm not asking what hours can be logged, I'm aware of that. I'm specifically asking about how to measure pilot skill. I have edited the question – Cloud Jan 29 '24 at 14:44
  • Thanks @Jamiec - as I have added to the question, it would be useful for others I think to state what metrics ARE a good determinant of skill level. – Cloud Jan 29 '24 at 14:45
  • 4
    what do flight hours matter if one is a child of the magenta line? – jao Jan 29 '24 at 14:55
  • 1
    Probably there is no metric (or an opaque one). Hours is one, but landing are also a different metric (short haul pilots may get more experience on the most difficult part of flight (under 10 000 feet). Night, instrumental, busy spaces, etc. And most important ability is about dealing with emergencies (so commercial pilots with flight simulators and training are best suited with emergencies) – Giacomo Catenazzi Jan 29 '24 at 15:38
  • Competence is directly related to the size of a certain organ which is directly related to hours logged. How do you not know this? – quiet flyer Jan 29 '24 at 18:01
  • 1
    Are you talking about basic flying skill or command skill, which encompasses a zillion variables beyond stick and rudder? – John K Jan 30 '24 at 02:36

2 Answers2

8

Total hours is a good measure of skill, but perhaps not the best measure of skill. That is usually attributed to hours on type. Also important to note that, at least for this answer, I'm conflating "skill" with "lack of accidents". That's not necessarily a good measure of skill.

Back in 2013, Aviation Safety magazine posted a good piece on exactly this subject. Some relevant excerpts:

In 2010, 96 percent of the 1500 civilian aircraft accidents in the U.S. involved general aviation. According to the NTSB’s 2010 Annual Review of Civilian Aircraft Accident Data, the average pilot pursuing personal flying and who was involved in an accident had a “total flight time of 2863 total hours, with a range of 20 to 31,270 hours. The average time in the type of accident aircraft was 460 hours, with a range of 1 to 10,000 hours.”

Anybody with even a modest grounding in statistics can see a standard deviation and data variance big enough to fly a Twin Beech through.

So clearly, pilots of all skill levels measured by "total hours" have accidents.

A 2002 epidemiological study of airline pilots by researchers at Johns Hopkins University found experience, measured by total flight time, offered a significant protective effect. With adjustment for age, Part 121 airline pilots with 5000 to 10,000 hours had a 57-percent lower risk of a crash than pilots with less than 5000 hours

Total flight time shows significant reduction in accident rates, however insurance rates tell the real story, as these are the people who have actuarial reasons to calculate risk factors.

Based on writings and policies of the aviation insurance industry, pilot experience in a specific aircraft make and model is considered a better measure of competency than total flight time. Insurance underwriters can favorably rank a 1000-hour pilot with 500 hours in the make and model with a 10,000-hour pilot who has no make and model experience.

Jamiec
  • 32,939
  • 8
  • 125
  • 156
  • 2
    I would add to this that these metrics are used partly because there really isn’t a great way to measure skill directly in an objective manner. It’s the same general issue that you run into with almost any large complicated skill-based task that can’t be trivially automated (because if it can be trivially automated, you know exactly what skill is needed). – Austin Hemmelgarn Jan 30 '24 at 00:05
3

Hours equate to experiencing a range of different scenarios and possible outcomes. Undoubtedly, there is learning that has taken place to mold the airman into the pilot they are today. Some of the only correct thought processes can be achieved through experience.

You might remember the saying.
There are old pilots and they are bold pilots but there are no old, bold pilots.

Older pilots, through experience, have learned more thoroughly what it means to stay safe in many different situations.

With that said, I don't necessarily think flight hours are a good barometer for pilot skill. I have seen and trained pilots who were phenomenal at 500 hrs total time and I have seen pilots who could barely pass a checkride with 10,000 hours.

Checkride pass/fail rates are better determining factor on pilot skill. It is only by observation that one can assess pilot skill. There are numerous industry evaluators trained to assess such things.

I would look at checkride pass/fail rates for each pilot. I would also drill down to the specific make and model checkrides to determine competence in a specific airplane.

Small airplane pilots are required to have a flight review every two years. How long did the ground session take and how long and how many flights did it take to receive a sign-off?

Part 121 and 135 as well as multi-crew airplanes (61.58) require a checkride every year. What were the results of those checking events?

wbeard52
  • 12,679
  • 7
  • 38
  • 78