Listen:
Check out all episodes on the My Favorite Mistake main page.
My guest for this bonus episode of the My Favorite Mistake podcast is Arnold “Arnie” Barnett. He is the George Eastman Professor of Management Science at MIT Sloan School of Management. With a BA in Mathematics from Columbia University and a PhD from MIT, Arnie is a renowned aviation safety expert who has earned prestigious accolades, such as the 2002 President Citation from the Flight Safety Foundation. His exceptional teaching has been recognized repeatedly by MIT Sloan students, and his career is marked by a deep commitment to advancing the understanding of aviation risk and safety.
In this episode, we explore the nuances of aviation safety, examining both statistical trends and public perceptions. Arnie explains how improvements in safety have continued at an impressive rate—comparable to a “factor of two” reduction in risk every decade—even amidst occasional high-profile incidents. We delve into the concept of risk perception, using analogies like coin tosses to illustrate how random events can appear more alarming than they truly are, especially when amplified by media coverage and human cognitive biases such as availability bias.
The discussion also touches on the importance of data-driven analysis and balanced risk assessment. While acknowledging that rare events can occur, Arnie reassures listeners that the overall safety of air travel remains exceptionally high. He offers insights into how seemingly isolated incidents do not necessarily indicate a broader decline in safety standards, emphasizing that the structured protocols and continuous learning in aviation make flying one of the safest modes of transportation available today.
Links:
- His “favorite mistake” story — Episode 169
- A 2024 article about Arnie's latest research on safety data
Questions and Topics:
- Is it a mistake to assume that recent incidents indicate a decline in U.S. aviation safety?
- Can you explain why people struggle to grasp the inevitability of rare events, as illustrated by the coin toss analogy?
- At what threshold—measured in incidents per million passenger miles—might aviation safety improvements stall or reverse?
- Do connecting flights meaningfully increase overall risk, or is the cumulative danger still essentially negligible?
- Could you recap your earlier story from episode 169 to highlight how a perceived mistake led to valuable safety insights?
Scroll down to find:
- Video version of the episode
- How to subscribe
- Quotes
- Full transcript
Video of the Episode:
Quotes:
Click on an image for a larger view
Subscribe, Follow, Support, Rate, and Review!
Please follow, rate, and review via Apple Podcasts, Podchaser, or your favorite app — that helps others find this content, and you'll be sure to get future episodes as they are released weekly. You can also financially support the show through Spotify.
You can now sign up to get new episodes via email, to make sure you don't miss an episode.
This podcast is part of the Lean Communicators network.
Other Ways to Subscribe or Follow — Apps & Email
Automated Transcript (May Contain Mistakes)
Mark Graban:
Hi, everybody. Welcome to a special bonus episode of My Favorite Mistake. I'm your host, Mark Graban. Our guest today is Arnold Barnett. He joined us back in episode 169.
Mark Graban:
So, if you want to hear his favorite mistake story, I'll point you to that in the show notes. But Arnie is the George Eastman Professor of Management Science at the MIT Sloan School of Management. He's been cited as the nation's leading expert on aviation safety, which is why I wanted to talk to him today. He was recognized with the 2002 President Citation from the Flight Safety Foundation for his truly outstanding contributions to safety, and MIT Sloan students have honored him on 14 occasions for his exceptional teaching.
Mark Graban:
Arnie has a BA in mathematics from Columbia and a PhD in mathematics from MIT. So, Arnie, thanks so much for coming back on the podcast.
Arnie Barnett:
How are you? I'm fine and very happy to be here.
Mark Graban:
Yeah. I— I— I hope you don't look back and think it was a mistake to come on the podcast that first time. I really enjoyed having you here.
Arnie Barnett:
No, it was not a mistake.
Mark Graban:
Good. You know, the topic that's been in the news and on a lot of people's minds is the safety of air travel, especially here in the US. I've heard people comment—even as recently as yesterday—that they don't feel safe flying right now. Is it a mistake to think that air travel in the US has suddenly become less safe?
Arnie Barnett:
I think that is a mistake. First of all, it's just common sense. Yes, it's true that in the last month or two, we've had a number of events, some unnerving and a few fatal. One might look at that and say, “My goodness, I always thought flying was safe—I never gave it a second thought, and now I can’t help but think about it.”
Arnie Barnett:
But there's no reason to believe, for example, that our ability to fly safely now is lower than it was two, three, or even four months ago, when we went month after month without a single incident causing injury, let alone a death. I mean, I wouldn’t expect these events to be a sign of something systematic. When you're dealing with rare events, the laws of probability suggest that long periods of nothing happening will be interspersed with bursts of events—a temporary spasm of bad luck, if you will. I think that's the most reasonable explanation for what's happening now. So, I don't take it seriously or think I'm at greater risk now than I was, say, two months ago.
Arnie Barnett:
Actually, it’s the opposite. For example, I sometimes fly into Washington’s Reagan Airport. We later learned that the paths of helicopters and airplanes were shockingly close together. In fact, The New York Times reported at one point that the prescribed altitudes were only 15 feet apart on final approach. The helicopters had an upper limit of 200 feet, and the planes had a lower limit of 215 feet over the Potomac. Anyway, we're not going to see those helicopters now.
Arnie Barnett:
They’ve been drastically reduced over there. In that sense, we never knew we were at such potential risk. Tragically, now we know there was such risk, and it had terrible consequences. But I think the risk has since drastically diminished—and that’s happened repeatedly. In aviation, we’ve learned hard lessons that have prevented future tragedies, though sometimes we learn in ways that don't involve deaths.
Arnie Barnett:
They used to say that the FAA—the Federal Aviation Administration—had a “tombstone mentality,” meaning they only started making reforms and proposals after a crash. I don’t think that's as true today. There are numerous studies, and people like Tversky and Kahneman have discussed the fallacies we suffer when assessing which risks are large and which are small. It’s natural; we’re human. When you hear about events like these, psychologists refer to it as availability bias—if you see how something could happen, you start worrying that it will happen, even if you know the risk is infinitesimal.
Arnie Barnett:
Okay, so yes, a plane landing at Washington’s Reagan Airport could hit a helicopter, and you might think, “Gee, that could have happened to me.” I’ve flown into Reagan on planes of the same type. So, people think, “That could have happened to me.” And while that’s true in theory, the chance of it happening to you is extraordinarily low. I do think that, at the moment, people are overly upset, especially since we’ve gone so long without any of these events that many didn’t even think about the risk of flying. After all, there are so many risks in everything we do.
Arnie Barnett:
When you think about it, everything we do entails some risk. You go to the grocery store, and the ceiling could collapse and injure or even kill you. But I don’t worry about that when I go into the supermarket. Similarly, when we drive or eat—if you read about listeria in oats, for example—you might think you could be the unlucky one to suffer a consequence, but we can’t stop eating because of that. We have to assume that when risks are close enough to zero, we must proceed as if they were actually zero. That's how we live our lives. And I believe that aviation, which is nothing short of a miracle, is so safe that it should be considered effectively risk-free.
Mark Graban:
So, is there a level of incidents—measured in fatalities or incidents per million passenger miles—where you’d say that safety improvements have stalled or reversed? I’ll include a link in the show notes about how aviation safety has progressively improved over time.
Arnie Barnett:
Well, one can certainly keep looking at the data. I’ve said that accidental plane crashes have been brought to the brink of extinction—though, like many near-extinct species, there can be a comeback. For example, in US automobile safety, the number of deaths per 100 million miles driven was steadily decreasing until it recently took an upturn. So, it’s not inevitable that improvements will continue forever. However, one thing I’ve noticed—similar to Moore’s Law for computer chips—is that aviation safety has improved by a factor of two every decade.
Arnie Barnett:
Looking at worldwide risk, if I pick a boarding pass at random from any passenger in the world, the chance that a person will be killed has been decreasing by a factor of two every decade since commercial aviation began after World War II. That means it’s a factor of two lower than a decade ago and a factor of four lower than two decades ago. Even as it got safer and one might think further improvement would be harder, it has continued at the same rate. I intuitively suspect there must be some lower bound greater than zero that we can’t cross because we’re human, but regardless, safety has improved consistently by a factor of two.
Arnie Barnett:
In fact, even if you look at the last few months—including that terrible crash in South Korea where a plane hit a concrete barrier, burst into flames, and resulted in 179 deaths—even including that data, the overall risk is about a factor of two lower than it was for the previous decade. So, while it is theoretically possible for the trend to stall or even reverse, I don’t anticipate that happening. The one thing I do worry about is deliberate attacks on aviation.
Arnie Barnett:
Such attacks can occur in several ways: through old-fashioned terrorism, or even if a plane is shot out of the sky—perhaps accidentally, when people mistake it for a military plane in a war zone. It’s a strange concept: a deliberate act executed in an accidental manner, especially if a pilot makes a grave error.
Arnie Barnett:
I think the current consensus is that Malaysia 370—the plane that disappeared—was destroyed by its captain. It’s not definitively known, but many reasons point toward that conclusion. We also had a plane shot down late last year over Russia—related, albeit accidentally, to the ongoing conflicts there. These are the kinds of events that could conceivably occur again in the US, although we haven’t seen any aviation terrorism since 9/11.
Arnie Barnett:
And 9/11, after all, was roughly a quarter-century ago. Before 9/11, if you looked at the data as I have, there were several events in quick succession—like Pan Am 103, which was destroyed by a bomb, along with several other bomb-related incidents in that period. Then it stopped after 9/11. Could such events resume? I hope not.
Arnie Barnett:
Certainly, security is better now than it used to be, but you never know. Worrying about potential risks—even those that haven’t yet materialized—introduces a new level of anxiety. Instead of being anxious about actual events, we might become anxious about hypothetical ones. While it’s important to think about potential risks, I believe the fact remains that aviation is so safe that we can count on it.
Mark Graban:
Yeah. So, what I hear you saying as a final point is that the risk of something bad happening on a particular flight—or even over a lifetime of flying—is so close to zero that it’s essentially zero. I’m trying to remember if this is something you discussed in class or just the general aviation safety notion that the most dangerous moments are takeoff and landing. You might think that taking a connecting flight, with two takeoffs and two landings, would be twice as dangerous as a direct flight. But you’re saying that two times zero is still zero.
Arnie Barnett:
Well, I think you’re right. If you look at when these events occur—say, the crash at Reagan happened on final approach, and the previous crash with double-digit fatalities in Buffalo occurred during landing amid a storm when icing conditions caused instability—then you’re right. The risk, particularly the accidental risk, is mostly on takeoff or landing.
Arnie Barnett:
But as you say, the overall risk is extremely close to zero. One thing to remember is that every day around the world, 12 million people board airplanes—about 5 billion per year. If you work it out, that’s 12 million a day, and on the overwhelming majority of days, not a single person is injured or killed.
Arnie Barnett:
So, even yesterday—for example, 12 million people flew with zero incidents, as far as I know; the day before, 12 million, zero, etc. The numbers are extraordinary. Another way of putting it is that, at the current level of mortality risk in the United States—or in comparable countries like the UK, those in the European Union, Mainland China (which has an excellent safety record), Australia, Canada, New Zealand, Israel, and Japan—if you were to take one flight a day, on average you could fly for 300,000 years before succumbing to a fatal crash. And I don't mean to be pessimistic, but I don't think we're going to live 300,000 years.
Mark Graban:
Yes, I think that's fair to say. Final thought: I was thinking about you and the story you told back in episode 169 when you mentioned that, back when US Airways experienced a spasm of bad luck with several crashes, a reporter called you and you said, “That's amazing.” You didn't mean it in a positive way.
Mark Graban:
Maybe you could recap that, and I invite people to listen to the whole episode. You used the word “amazing” in its true sense—meaning shocking.
Arnie Barnett:
Yes. Amazing, meaning shocking.
Mark Graban:
Yes.
Arnie Barnett:
Certainly, I was not expressing joy upon hearing about the Pittsburgh crash.
Mark Graban:
No. I recalled that, if I remember correctly, it was Mary Schiavo on CNN—who I believe was the former head of the National Transportation Safety Board—using that same word, in the same tone that you meant it: that it was shocking, not great.
Arnie Barnett:
Yes, yes. But on the other hand, as I mentioned in podcast 169, I did consider it a mistake. It was, I think, a natural reaction—a fluke, perhaps—but unfortunately, that remark was used as if to imply, “Don't treat this as a fluctuation; this wouldn’t have happened on its own.”
Arnie Barnett:
It was amazing. So, let's not dwell on it. That was a mistake. It turned out that people at USAir—which later became US Airways and is now part of American Airlines—were the ones who helped discover the subtle problem with the rudder that caused that crash, which also caused a previous crash on United. The rudder has been corrected since then.
Arnie Barnett:
So, not only were they victims of bad luck, but the issue could have occurred in any Boeing 737, and yet it didn’t, even though US Airways had one of the largest fleets of 737s. The laws of averages would suggest that a larger fleet would be more likely to exhibit such a fatal flaw, but it didn’t happen that way. So, it truly was a fluctuation. But yes, I think it was a mistake to yell “holy cow.”
Mark Graban:
Well, I appreciate you sharing that story and for coming on to provide reassurance that you wouldn’t be any more worried about flying with me—or as a listener—today compared to a year or even 10 years ago. Thank you for those perspectives. I’m going to put a link in the show notes to an article on the MIT website from late last year with the data you mentioned—this parallel to Moore’s Law in aviation. I encourage people to check it out and use it to reassure anyone with a new fear of flying. So, thank you for sharing those perspectives and for being our guest.
Arnie Barnett:
If I can share one more anecdote: I once got a call from someone—and these days, with the web and all, people know you even if you don’t know them—who was thinking of going to the UK for her grandmother’s funeral. She was very frightened about flying. I said to her—and I was surprised at my own rudeness when talking of mistakes—”May I ask how old you are?”
Arnie Barnett:
She said, “34.” I said, “Do you realize that over your entire lifespan, British Airways has never had a fatal crash?” She replied, “Oh, I didn’t know that.” So I said, “I'm sorry for the circumstances of your trip to the UK, but I don’t think you should cancel it.”
Arnie Barnett:
I later learned that she didn’t go. Fears are deep, and while I respect them—I know many people are afraid to fly and some will take ships or trains instead, even though statistically that puts them at greater risk—I really believe this is something we shouldn't be afraid of.
Mark Graban:
Well, thank you, Arnie. Again, we’ve been joined—on a repeat, return visit—by Professor Arnold Barnett from the MIT Sloan School of Management, and rightfully so, he is a patient aviation safety expert (though I mistakenly said “patient safety expert”). That was my mistake.
Arnie Barnett:
We could certainly use many patient safety experts. That's an area where the risk is still much too high.
Mark Graban:
That is true as well. So, Arnie, thank you again.
Arnie Barnett:
Oh, thank you so much for having me.
Actionable Takeaways from the Podcast Episode
Implementation: When planning air travel, focus on the historical safety record of airlines and recognize that current safety measures are stringent and continuously improving. For example, choosing airlines with exemplary safety records can further mitigate rare and unanticipated risks, providing peace of mind during air travel.
Stay Informed, Not Anxious:
Guidance: Understand that aviation safety is statistically very high, even amidst rare bursts of incidents. Arnie Barnett explained that the risk levels are closer to zero than perceived, as discussed with the “availability bias” and the statistical explanation regarding recent safety events.
Implementation: Arm yourself with reliable safety data and trends when feeling anxious about flying. Remind yourself that the perception of risk is often inflated by isolated incidents, which are outweighed by the overwhelming safety record of aviation.
Leverage Sound Arguments:
Guidance: Use metaphorical insights, such as Barnett's coin analogy, to effectively communicate the nature of rare events. This helps in understanding that a sudden occurrence of incidents does not necessarily indicate a systematic issue.
Implementation: When discussing risks, employ similar analogies to articulate to others that rare sequences of events do happen without indicating a broader change in safety. This approach can not only assure others but also help you internalize and remain calm about flying.
Focus on Broader Safety Trends:
Guidance: Be aware that deliberate attacks are an area of potential concern, but reassurances on trends are clear from Barnett, who discussed continuous improvements and the extraordinary safety levels in countries like the US, UK, and Canada.