top of page
Below is a sample chapter from my book, Mullane's Guide to Learning to Fly: Deep Dives into Select Topics for Today's Student Pilot
Click here to see a full list of chapters and topics covered

Click here to purchase the book at Amazon.

Click here to purchase a signed copy.

Copyright 2023 Patrick Mullane, All Rights Reserved

Chapter 3
How Safe Is This?


Shortly after beginning my PPL training, the worries started to accumulate in family and friends. 

“Aren’t you scared?” was a question I got at a social gathering one night shortly after I’d let people know that I’d done my discovery flight and was now committed to the hours of training it would take for me to become a private pilot.


“It’s safer than driving a car!” somebody else in the circle of conversation responded before I had a chance to. 


But is that true? 

It is true that commercial air travel is immensely safer than driving a car. How much safer depends on your data source. The numbers with respect to airline safety are surprisingly all over the map. This is partly because there’s more than one way to approximate relative safety. In the interest of picking a single, reliable source on airline safety, though, I refer to data provided in a paper written by MIT professor Arnold Barnett.

Barnett published research in Transportation Science in 2010 that looked at the likelihood of death from commercial air travel, but with a bit of a twist. He broke the world of commercial aviation into three groups: first-world countries (e.g., U.S., Canada, Japan), developing world nations (e.g., Singapore, South Korea), and newly industrialized nations (e.g., China, Brazil). He found that the risk of dying in a commercial airline crash overall was one in 3,000,000 per flight. However, when the data was looked at based on the country groupings mentioned earlier, the numbers are very different. The first-world countries had a death risk per flight of one in 14,000,000. In the developing world the number was one in 2,000,000. And in the newly industrialized nations the mortality risk was one in 800,000. 

But even if we take those last numbers and compare them to your chances of being killed in a car accident, you’d still rather be flying from Cameroon to Liberia than driving from Peoria to Chicago. That’s because your lifetime risk of dying in a car accident (in the United States) is one in 107 according to an analysis by the National Safety Council. You’ll note that this number is a lifetime risk, while the aviation numbers I quoted in the previous paragraph are your chances of dying on any given flight. So, we are doing a bit of apples to oranges comparison here. But suffice it to say that no matter how you slice it, driving is considerably more dangerous than flying on a commercial airliner. 

But you’re not flying commercial

But what about general aviation? Many don’t realize that it’s distinctly unlike commercial aviation. First, there’s the level of training. When going through your pilot training it may seem like you’re being put through the ringer, yet it’s incredible to think how little is required of you to get a PPL in the United States.  You need to log just 40 hours in the cockpit of an airplane before getting permission to operate an aircraft with passengers (this is the minimum time; most will take longer). Think about that. Spend the equivalent of one standard work week in the left seat of an airplane, and the FAA might sign you off to act as the pilot in command of an aircraft. Airline pilots, on the other hand, go through extensive training and—perhaps more importantly—extensive recurrent training. And between episodes of that training, they are flying a lot, making it easier to stay current. Second, there’s the fact that most commercial aircraft have two pilots in the cockpit. That’s a lot of redundancy. They check each other’s work and help lighten the workload on any one individual. You own everything when you fly as a GA pilot; in most cases there’s nobody there to tell you that you are making a dumb decision or missed a checklist item. Third, the technology in many commercial airliners is beyond that you’d find in most general aviation aircraft. Onboard radar to see weather in real time, sophisticated anti-collision systems to keep aircraft from running into each other, and two (or more) very powerful engines to mitigate the loss of power in one … all these things and many, many more increase safety substantially. Finally, there is the ever-present understanding among the flight crew that doing something stupid—even if it might not kill you—could cost you your job. If you do something stupid in the cockpit of your private plane, nobody may ever know. Get away with that stupid thing and you may do it again until … one day …

No, general aviation is not like commercial aviation. And, not surprisingly, the statistics bear this out. I found the raw data from the FAA on general aviation fatalities between 2001 and 2020 and used that to do some analysis. I focused on the last eight years of data (2012 to 2019). I did this for three reasons. First, data is inexplicably missing for 2011. Second, safety increased during the full 20-year period, so I wanted to account for the better numbers over the later years’ data set. Third, I wanted to compare the aviation numbers to driving numbers and, as of the writing of this paragraph, the latest reported year on driving fatalities from the National Highway Traffic Safety Administration (NHTSA) is 2019. 

During this time frame from 2012 to 2019, there was an average of 18.4 general aviation fatalities per million flight hours.  What was it for commercial air travel inside the United States? Very small: 0.095 fatalities per million flight hours. General aviation flying is about 194 times more dangerous than commercial flying by this metric. It’s not surprising when you consider how many flight hours commercial operators rack up and that during the period we are considering, there were no fatalities at all in five of the eight years.

That differential between GA risk and commercial risk is shocking. But commercial air travel is so safe, comparing any other cause of death to it looks shocking. After all, that 0.095 number is due to a total of 13 deaths in the eight-year period. In the United States in 2019 alone there were 19,000 homicides, 33,000 motor vehicle deaths, 39,000 deaths from falls, and 20 fatalities from lightning strikes. 

Comparing GA flying to commercial flying, then, seems unfair. How then does GA flying look compared to an activity virtually all of us partake in: driving? It’s very tricky to make a one-to-one comparison. As you’ve noticed, fatalities in aviation are based on flight hours. And even that is an educated guess when it comes to the GA world since there is no formal tracking mechanism of hours flown. Driving fatalities, by contrast, are measured in hundreds of millions of miles driven. During our same reference period of 2012 to 2019, there were 1.1 fatalities per 100 million vehicle miles driven in the United States. That may seem like a low number, but keep in mind that the total miles driven in the U.S. every year is about 3.2 trillion, so the total fatalities figure is quite large: around 32,000 deaths a year during the period. 

One method some have used to try and make a comparison between driving and GA flying is to convert hours of flying into miles flown. If we assume an average ground speed of 140 mph (around 120 knots) for general aviation aircraft, that means that 1,000,000 flight hours equates to 140,000,000 statute miles traveled. Our 18.4 fatalities per million flight hours would therefore be the same as 18.4 fatalities per 140 million statute miles traveled, or 13.1 fatalities per 100 million statute miles traveled. This is obviously not a heartening number; it’s about 13 times higher than the car crash numbers which, as a reminder, were 1.1 fatalities per 100 million statute miles traveled. But again, we are assuming that the “deaths per 100 million miles flown” number we backed into for aviation is a good way to do a comparison. It may not be … more on this in a minute. 

What about motorcycle riding? According to the NHTSA, all you easy riders out there face a fatality rate of about 25 deaths per 100 million miles traveled. This is nearly two times greater than the number we just calculated for GA flying and nearly 25 times greater than driving a car.

But some good news …

We are therefore left with numbers that tell us that the risk of GA flying is about halfway between the risk of driving a car and the risk of driving a motorcycle. 

However, there’s a really important point to consider when thinking about these comparisons. And it has to do with this: you could die in a car or on a motorcycle due to no fault of your own. You could be happily driving through a green light with your dog in the passenger seat when a driver who has the red light is distracted while texting his girlfriend and plows through the intersection, sending you across the rainbow bridge with your dog. You’re not in complete control of whether you have a safe drive or motorcycle ride.

But I’d argue—and I think the data supports—that you are almost always in control of whether you have a safe flight. 

How can I say this with such confidence? You probably already know: most aviation fatalities are the fault of the pilot. I looked through data that the Aircraft Owners and Pilots Association (AOPA) puts together in something called the Joseph T. Nall Report. Joseph T. Nall was a member of the National Transportation Safety Board (NTSB), the organization responsible for investigating aircraft crashes. In a stroke of sad irony, Mr. Nall was killed in 1989 while flying as a passenger in a small plane in Venezuela believed to have been chartered for a sightseeing flight. The report now named after him includes detailed data on the number of accidents each year, how many were fatal, and the cause of those accidents. While the numbers bounce around a bit year to year, somewhere around 75% of fatal crashes are classified as “pilot-related.” The report defines these crashes as “accidents arising from the improper actions or inactions of the pilot.” The “mechanical/maintenance” category is often below 10% of fatal cases in a given year, and the balance of cases are classified as “other/unknown.” 

Let’s assume, conservatively, that 70% of all fatal crashes over the eight years we discussed earlier were due to pilot behavior. That would mean that of our 13.1 fatalities per 100 million miles flown, only about four of them would be due to an act of God—an engine coming apart at a point where there was no chance of recovery, a wing falling off due to a manufacturing defect, a bird strike ripping off the tail. While this number is still about three to four times greater than the fatality rate in automobiles, it gets close enough to be comparable. And it’s well below the risk of riding a motorcycle. 

I want to emphasize one last time that these numbers only point us in the direction of understanding risk; they are not precise in any way and, in fact, have some big flaws. For example, consider that most fatal GA accidents occur during takeoff and landing. An NTSB study found that between 2013 and 2018, 23.6% of GA accidents occurred during takeoff and 40.2% during landing. With that information in mind, imagine two pilots flying from the same airport with the same skills, the same decision-making ability, and the same appetite for risk. They are essentially identical twins flying Cessna 172s in exactly the same way. Their airplanes will not fail mechanically; they are 100% reliable. Now imagine that one of our pilots takes off, does just one lap around the pattern, and then lands, flying a total of five miles. The other takes off, flies 100 miles away from the airport, turns around and comes back, and lands in the exact same conditions (same winds, lighting, cloud cover, runway condition, etc.) as the first pilot. Each of our imaginary twins had one takeoff and one landing under the same conditions. Therefore, they faced substantially the same risks. Yet one flew 40 times more miles and about two hours longer than the other. Thus, measuring accident rates or fatalities by miles flown—or even hours flown—is convenient, but not precise in any real way. 

But it’s all we’ve got. And it does allow us to understand relative risk in a directional nature. Remember that in the eight-year period we discussed earlier, there were a total of 13 deaths due to commercial airliner crashes. In 2020 alone, general aviation registered 332 fatalities. And GA pilots flew far fewer hours and miles than their airline counterparts.  

No matter how you look at the data, the key takeaway must be this: GA flying has risks, but they are manageable. And managing that risk depends almost entirely on you and your decision-making. 


Click here to purchase the book at Amazon.
Click here to purchase a signed copy.

bottom of page