Check out all episodes on the My Favorite Mistake main page.
My guest for Episode #169 of the My Favorite Mistake podcast is Arnold (Arnie) Barnett. He is the George Eastman Professor of Management Science and a Professor of Statistics at the MIT Sloan School of Management.
Barnett holds a BA in mathematics from Columbia College and a PhD in mathematics from MIT. His research specialty is applied mathematical modeling with a focus on problems of health and safety.
Cited as “the nation’s leading expert on aviation safety,” Barnett was recognized with the 2002 President’s Citation from the Flight Safety Foundation for “truly outstanding contributions on behalf of safety.” MIT Sloan students have honored him on 14 occasions for outstanding teaching.
In this episode, Arnie shares his “favorite mistake” story about blurting out something to a New York Times reporter who called to get his comment about a US Airways crash that had occurred earlier in the day. Even though he regrets saying what he did, it gave him a reputation for being “willing to talk straight” which led a torrent of requests to speak and to be interviewed in venues ranging from radio programs to NBC's Today Show.
We also talk about questions and topics including:
- 1994 US Air had a number of crashes – a “temporary spasm of bad luck”?
- The NY Times article he was quoted in — the “mistake”
- The word “amazing”: “causing great surprise or wonder; astonishing”
- Lesson about talking with the press?
- Are we bad at estimating probabilities in general?
- Bad at estimating the risk of driving vs. flying?
- You wrote an opinion piece in late March 2022 titled “Don’t end the mask mandate for US airlines”
- “ending the requirement now would be a serious mistake.” —> why did you say that then and do you still say that now?
- I saw you give a talk about this — is the Electoral College a mistake? Is it a mistake that can be fixed?
- A simple fix for gerrymandering?
- “MIT now has a reputation of being very much woke”
- Tell us about the Leaders for Global Operations program… you are a popular internship and thesis advisor. Why do you like working with LGO students?
Scroll down to find:
- Video of the episode
- How to subscribe
- Full transcript
Watch the Episode:
Subscribe, Follow, Support, Rate, and Review!
Please follow, rate, and review via Apple Podcasts or 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 become a financial supporter of the show through Anchor.fm.
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 (Likely Contains Mistakes):
Mark Graban (0s):
Episode 169 Arnie Barnett, professor at the MIT Sloan School of Management.
Arnie Barnett (6s):
Well, let me describe it. The, it is a little bit obnoxious as you'll hear to describe it this way, but I'm trying to answer your question, honestly. It is my favorite in this day.
Mark Graban (21s):
I'm Mark Graban. This is My Favorite Mistake. In this podcast. You'll hear business leaders and other really interesting people talking about their favorite mistakes, because we all make mistakes, but what matters is learning from our mistakes instead of repeating them over and over again. So this is the place for honest reflection and conversation, personal growth and professional success. Visit our website at MyFavoriteMistakePodcast.com. To learn more about Professor Barnett and to get links to his articles and research and more. Look for links in the show notes, or go to markgraban.com/mistake169.
Mark Graban (58s):
As always. Thanks for listening. Hi everybody. Welcome to My Favorite Mistake. I'm Mark Graban. We're joined today. You might say my favorite professor from my days at MIT is Arnold Barnett. And before I tell you more about him, I imagine Arnold is what you call you get called when you're in trouble, because most everyone calls you Arnie, right?
Arnie Barnett (1m 25s):
Well, either is fine. My mother always called me Arnold,
Mark Graban (1m 29s):
So Arnie thank you for being here. And Arnie is the George Eastman professor of management science. He's a professor of statistics at the MIT Sloan School of Management. He has a BA in mathematics from Columbia and a PhD in mathematics from MIT. His research specialty is applied mathematical modeling with focus on problems of health and safety. He's been cited as the nation's leading expert on aviation safety. I've heard him describe that way. Many times he was recognized with the 2002 president's citation from the flight safety foundation for truly outstanding contributions on behalf of safety and MIT, Sloan students have honored him on 14 occasions for outstanding teaching.
Mark Graban (2m 16s):
So that's not just me earlier. My, my wife, who was also a student we're we're we're in that camp and we know other students have agreed with us. We're not a statistical anomaly, I guess one other memory real quick and Arnie, you, you, you may well remember this too, the summer of 1997, in that, that first summer of the leaders for manufacturing program class, you, you informed us that our next class was canceled because you'd been invited to a meeting with the vice president. So we all thought that was a perfectly fine reason to be away.
Arnie Barnett (2m 50s):
I, I probably was lying, but it must've sounded like a good excuse.
Mark Graban (2m 57s):
Well, I, the lie was very specific, which makes it believable. This is back when they were looking at the question of bag match for passengers and checked baggage.
Arnie Barnett (3m 5s):
Yeah, I think that's true, but I think I'd probably postponed the class. I don't think I would have canceled it because you people pay pretty sizable tuition loans to be at MIT. So I don't think professors have a right to announce some too busy for you.
Mark Graban (3m 23s):
Well, it was for the greater good and education is a weird product where sometimes people are happier to get less of it for the same money, but,
Arnie Barnett (3m 32s):
Well, I hope my class was not an example of that.
Mark Graban (3m 35s):
No, no. We get a lot of value and a lot of great lessons out of your class in your teaching. So y'all have some questions that I'll want to ask you later, later on about statistics or mistakes people or businesses or governments make with, with mistakes. But we always start off first here with this question. So I'm going to ask you Arnie thinking back to your career, what would you say is your favorite mistake?
Arnie Barnett (4m 3s):
Well, let me describe it. The, it is a little bit obnoxious as you'll hear to describe it this way, but I'm trying to answer your question, honestly. It is My Favorite Mistake. As you've already mentioned, mark, I have an interest in aviation safety and have written papers on the subject using statistical principles to make sense of safety records. And one of the points I endlessly made was the bad crashes of airplanes, particularly in a place like the United States are extremely rare events. And when events are rare, the observed statistics can be highly misleading.
Arnie Barnett (4m 44s):
You know, let me give an example. Suppose I have a coin that's fair. Okay. And I toss it twice. Well, there are four possible outcomes. Heads, heads, heads, tails, tails, heads, tails, tails, all four equally likely. So in other words, there's a 50% chance you're going to get a double head or a double tail. Now, if someone were to say, look, two heads in a row, the coin can't be fair. That of course would be a highly misleading conclusion because even if the coin is fair, that was not in the least a rare outcome. So that was one of the points I kept stressing when people would say, wow, this airline has two crashes. This one has none. And therefore the one that you really want to fly is the one without crashes as well.
Arnie Barnett (5m 27s):
Even if they're equally safe, the same number of flights under the same conditions. It basically one of the airlines having the two crashes, it's like two, two heads for the fair coin, if they're equally safe. So you shouldn't jump to conclusions. Back in 1994, US Air, it eventually became US Airways before it was absorbed into American airlines. You went, Sarah had had a number of crashes in quick succession. And I have been saying, you know, this could be just a temporary of bad luck. I mean, if you toss a fair coin every now and then you get several heads in a row, but then on, in September, 1994, there was a USair crash at Pittsburgh that, that killed, I believe 150 people.
Arnie Barnett (6m 14s):
And I had not known about it until I got a call that day from a reporter for the New York times, with whom I'd spoken a little bit about aviation statistics in the past. And he said, you know, there's just been a bad crash on US Air. So I blurted out USA or again, that's amazing then of course, having made this holy cow comment, I tried a little bit to walk it back and say, of course, with rare events, you know, things can happen, but he had his quote and on page, one of the times he had the quote us Erica and that's amazing. Well, I started getting comments even though I really didn't want to say that I blurted it out.
Arnie Barnett (6m 57s):
I remember someone wrote to me, it's nice that finally someone is willing to talk straight and, you know, remember what Harry Truman used to say. He wished that there will one handed economist cause he was sick and tired of having them say on the one hand, on the other hand, on the same way, I think people are used to academic saying, we must take into account all the contextual factors, including the historical and metaphysical elements. So basically to have someone say, gosh, that's amazing that struck them as a breath of fresh air. I, I got a torrent of requests to speak that day from newspapers from I guess the radio stations and whatnot shortly thereafter, I was invited to the today show and I was interviewed by Katie Couric about aviation safety.
Arnie Barnett (7m 49s):
And in general, actually I got a lot of attention because of that mistake. And then I think it helped to be honest again, I don't mean to be obnoxious that it helps solidify my reputation as one of these go-to guys. When you have issues about aviation safety statistics. So I was invited to be on a part of projects about mid-air collisions, about collisions on airport runways and especially about aviation security and experiments such as the one you'd mentioned earlier today about whether bags should be matched to passengers in the interests of aviation security.
Arnie Barnett (8m 32s):
And I think it's, I can say literally that my career benefited from that mistake, but it was a mistake. And to the extent that it contributed to the sense that US Airways was less safe, you know, it may be that it did have some harm in the statistic USA or that it contributed a bit maybe just a little bit to the aversion to fly west here, which certainly arose after the Pittsburgh crash, which was the fifth crash in five years. So, and of course it is true that even though something could be a fluctuation, that doesn't mean it has to be.
Arnie Barnett (9m 12s):
I mean, sometimes the reason you have so many crashes is that you're not as safe. But anyway, that is My Favorite Mistake. I will tell you though, I'm not saying this was out of guilt, but in the next 20 years, USA is not only a perfect safety record and character billion passengers without a single fatality, but there were three circumstances where heroic actions by us, airways pilots prevented A terrible including of course the miracle on the Hudson was one,
Mark Graban (9m 44s):
Arnie Barnett (9m 44s):
Were two others. There was one case where the controller said take off in the US Air pilot said, I don't think we will because we're not sure what about that plane that just landed. And even though the controller got very kind of abusive, they said, we're waiting until that plane is at the edits gate. And it turns out that the plane was right ahead of the US Airplane on the runway, in the fog. And if they can follow the air traffic controller, they would've smashed into it. So anyway, so I did write that saying when it was absorbed by American airlines, US Air leaves and triumphs, but there was that mistake, as I said, I blurted something out and indeed had I heard about it earlier that day so that when the reporter called I had a considered response, I wouldn't have said that, but I did wind up saying it and it did wind up help you.
Arnie Barnett (10m 38s):
You quoted certain things about how I was taken seriously for aviation safety. Ironically, it was what I blurted out, which was not terribly scientific, which may have contributed to that. So strange world.
Mark Graban (10m 50s):
Well, it's a very interesting story, Arnie. I mean, I think it falls into that category. You know, there, there are some favorite mistakes stories that people tell where it's a mistake that ended up leading to some unexpected doors opening or, you know, a career taking a path that might not have otherwise been exactly the same. I mean, I don't, I don't find it that I, I, you know, in a noxious story, I think it's, it's an interesting reflection of, of how a moment like that can lead to things. I mean, I would argue that there's societal benefit to people asking for your statistical analysis, that it wasn't all just about publicity or attention or what have you.
Mark Graban (11m 37s):
I don't think there's too much to apologize for. I, I understand the dynamic dynamic there. Like you said, you blurted it out, but
Arnie Barnett (11m 43s):
Yeah, well, you know, they say, I think Michael Kinsley one said in Washington of GAF is when a politician accidentally tells the truth. So I think there may have been a sense that I was telling the truth. Well, it was true in the sense that look, if you had five crashes in five years, the natural reaction, what is going on here and to say, well, maybe it's just bad luck. Yes. Maybe it's just bad luck, but maybe there's a real problem. So in that sense, I've probably had the same reaction as an awful lot of other people. And so maybe airing it out by making it legitimate, at least to raise the question you could say, maybe that was a good thing because of the experts are always telling us things that we have a sense aren't true.
Arnie Barnett (12m 28s):
Then we pay less attention to the so-called experts.
Mark Graban (12m 31s):
Yeah. Now, did it ever get back to you directly or indirectly with anybody at us, airways mad at you?
Arnie Barnett (12m 38s):
Yes. In particular, there was a pilot from a Southern state who wrote to me saying among other things, he said, you have dishonored me in front of my family. And I was thinking, gee, I'm glad I'm not there. And I'm glad that Massachusetts has gun control and et cetera. And his wife who ironically I believe was an alcoholic started calling me and then actually called the Dean of the Sloan School and said, I thought, you'd like to know that your faculty members are making idiots of themselves.
Arnie Barnett (13m 17s):
You know, actually to be honest, not to complicate the story, that was part based on something I said with Katie Couric, but that would not have, that would not have happened. That, that interviewer not for the, the holy cow comment, because that that's amazing. But there was, I was concerned about that because when you said you have dishonored me in front of my family, you know, you think the next thing you're going to be invited to a dual
Mark Graban (13m 48s):
Or at least slapped across the face with a glove.
Arnie Barnett (13m 52s):
Yeah. But I did, I did write to them actually to the, to the family. And I said, I know you're unhappy. And I, and I, I pointed out actually, when I did talk, I think later that day to Newsweek, I had said that I suspected it was only a spasm of bad luck. So I pointed out to them that I was not, you know, going out of my way to blame less air. And I said, but no, believe me, no one is re rooting for USM now more than I am, because that's sort of a stupid thing to say. I'm sure the people, US Airways were infinitely rooting for it. And rightfully so the ironic thing is it, the crash wasn't US Air fault.
Arnie Barnett (14m 37s):
It was some problem with the rudder of the Boeing 7 37, and any airline that fluid could have been the first to lose a plane because of that problem. And Southwest had more seven, three sevens than US Air. So statistically, you might've thought when the problem manifested itself, it would have been on Southwest or it could have been on United actually United had had a previous 7 37 crash because of the rudder, but it hadn't been recognized at that point. So there really was, it really was an element of bad luck and US Airways employees, some of its experts were instrumental in finding the cause of that crash. So it really did turn out to be an airline that had suffered bad luck, had an, a perfect record from that point on.
Arnie Barnett (15m 24s):
So it's a bet that anyway, as you say, a favorite mistake again, I do, I wish I had been more in control. Yes. But on the other hand, I noticed that, well, sometimes, sometimes you get lucky not to do you want to get lucky in connection with tragedies.
Mark Graban (15m 40s):
Right. Right. And I wonder if there's a mistake in either the use of, or the interpretation of the word. Amazing. I just went and did a quick dictionary search. I think the word amazing is often used in a very positive context. Like, you know, somebody makes 10 out of 12, three pointers in a game and they might say, oh, an amazing performance by Steph Curry. But the dictionary definition, it sounds like this is how you are using the word. At least one definition says causing great surprise or wonder somebody calling astonishing. Which, I mean, that sounds like that really was your reaction, but this is a great surprise.
Mark Graban (16m 23s):
Not oh, great.
Arnie Barnett (16m 24s):
Yes, that's right. I certainly did not mean it to squeal with joy that USR had another crash.
Mark Graban (16m 32s):
Of course, you know? Yeah,
Arnie Barnett (16m 34s):
Yeah. No, no amazing being amazed, being stunned.
Mark Graban (16m 37s):
Sure. Yeah. But saying that's stunning that, that looks differently in print. I'm not trying to wordsmith you here or, you know, but
Arnie Barnett (16m 46s):
Yeah, no, he got the quote, he had a right to it. I didn't maybe if I had said, please don't quote me, he might've decided to oblige, but frankly I forgot to say it anyway. So, and then, so when we talk about how I'm described as a safety expert, it's in part because of that deviation from scientific expertise.
Mark Graban (17m 9s):
Wow. I mean, it sounds like there's also a lesson there about talking with the press you've been interviewed countless times here. And you know, there's this question of asking to go off the record, asking to go on background. I think there's one lesson more broadly here of you. You, you, you can't ask for that permission after the fact, you can't say something and then say, oh, by the way, this is off the record. It seems like it doesn't work that way.
Arnie Barnett (17m 35s):
You know, I have had positive interactions with the press. I very rarely if ever have been misquoted. In fact, sometimes I'm impressed how carefully they quote things. And there was a circumstance mark, where I remember I was doing something that involved aviation safety and I was working on us, an experiment, a security experiment involving I think it was involving, let's say American airlines. I don't think this is a secret now. And I got a call from someone from the Washington post and I said something, you know, and then I realized after I had said it, and after we had hung up that if I were identified as the person who said that it would make it harder for me to continue in the security experiment.
Arnie Barnett (18m 28s):
So I did call the reporter back and I said, I'm incredibly sorry, but could you perhaps basically treated as off the record, even though I hadn't said it and he did. Okay. So I haven't had bad experiences with, with people, you know? I mean, to be honest, when you get people from organization like the New York times or the Washington post or the wall street journal or the, what used to be the major news networks, they're pretty, they're pretty serious. Yeah. So I don't think that I have any, any complaints. And I think sometimes, you know, the media gets a bad rap for the obvious reason that sometimes they say things that people would rather not be sad and there will be obviously tendency, people will say, nevermind what they said, look at, look at how they should never have said it.
Mark Graban (19m 19s):
Yeah. Yeah. One other question, Arnie for you about aviation safety and probabilities. I remember you would talk about how bad we are. We are sometimes at estimating the risk of we're recording this right before Memorial day weekend. And a lot of people might decide, you know, if I, if I was going from Detroit to Chicago, is it safer to drive or to fly? And I remember you, you had some math around that. I'm curious how the math has changed of w w where is the breakeven point between driving safety and flying risk or safety?
Arnie Barnett (19m 56s):
Well, that's, that's an interesting question. Particularly Detroit Chicago is about 200 miles and you know, one of the sources of risk. Now, if you fly is the danger of coming down with COVID, you know, the, the planes are full and now masking is optional. And I know the last time I've flew, maybe 15% of the passengers were wearing masks. And none of the crew, I was wearing a mask. And I thought based on what the CDC was saying, people should have been wearing masks, but they weren't. So in other words, you have to consider it. And when you talk about the risk, part of the risk of flying now is coming down with COVID either in the airport or on the airplane.
Arnie Barnett (20m 40s):
And of course, if it's your own car with your own family, the chances are, there's no incremental risks. And given the 200 miles, it's only love with three hours. It's not that you have to make lots of stops on the way where you would be subject to risks going into stores and whatnot. Anyway, having said that, I would say, I don't think there's any safety advantage to flying rather than driving for that distance. Assuming you were doing it on the interstate, maybe the interstate 90, it might be. And because interstate driving is extremely safe, you know, we have in Massachusetts, the Massachusetts turnpike, which goes from east to west, we're kind of small state.
Arnie Barnett (21m 25s):
It's only about 130 miles from east west, but there are millions upon millions of cars that use it over the course of the year. And in a typical year there isn't a single fatality on the Massachusetts turnpike. So what I'm saying is, if you look at driving like that, you realize that it's really quite safe flying except for COVID is extraordinarily safe. Of course, in the United States, you know, accidents involve crashes have been pushed almost to the break of extinction almost. Yeah. So the, the verge of extinction, I guess I should say, so I wouldn't worry. I, I would think either way is remarkably safe, but if you're flying where an N 95 mask, and don't talk to the person next to you, whatever, don't take your mask off.
Arnie Barnett (22m 13s):
If they give you a snack for that short flight, put it in your bag and eat it, eat the peanuts or the biscuit after you get off the plane. If you do that, I think you're pretty safe.
Mark Graban (22m 25s):
Well, I want to, we'll come back to the mask topic in a second, but thinking about risk, does the risk of an accident when you're driving kind of increase linearly with the amount of time you're on the road, maybe if you've driven a really long day, it might increase or than linearly. Is that fair to assume? Or there it may, it probably depends on all sorts of things, the congestion of the roads and what road it is.
Arnie Barnett (22m 52s):
Well, I think first of all, what you're saying could, well, make sense that you are more fatigued. I don't, I haven't seen data that actually break down accidents by how long the driver had been behind the wheel before the accident occurred. I wouldn't be surprised though, if we could get such data, if we saw something of an effect of what you mentioned, namely, that fatigue comes in. Certainly I think fatigue can be one of the reasons that one has that way. If you look at the causes of accidents, I don't think there are that many that are attributed to fatigue, protect per se. But I certainly think that if you're fatigued, your reaction time is lower, is higher, I should say.
Arnie Barnett (23m 35s):
And so I think there would be greater risks, but I haven't seen statistics that document that,
Mark Graban (23m 40s):
Yeah, I mean, I have, and that's truly like, you know, an MIT statistics professor to point to, you know, seeing the data and not seeing the data that's different than whether the data exists out there in the wild. But there, there is data that I've seen. If you look at fatigue and health care, working the 11th and 12th hours of a nursing nursing shift, that, that errors do a spike in what seems like a statistically meaningful way, which could probably be correlated with fatigue there there's also the risk of correlation versus causation mistakes.
Arnie Barnett (24m 18s):
No, I think you're absolutely right when I say we don't have the data on the other hand, as you also say, things are complicated because let's say you're driving from, let's say Detroit to Chicago and after a hundred miles, you go in and get a coffee. Well, then you might be more alert in the third and fourth hour than you are on the first and second. So in other words, if you really wanted it to consider all the things that come up, you know, it gets very complicated, but I haven't seen the data, but I agree with you. Fatigue itself certainly is not good. Well, I know that when, even when I play these video games, I played text, twist the word game.
Arnie Barnett (24m 60s):
And I know when I play it, when I'm tired, I really don't get as many words per minute as when I'm alert.
Mark Graban (25m 9s):
Arnie Barnett (25m 10s):
And of course, if you're playing a video game while you're driving, that itself is a source of risk.
Mark Graban (25m 16s):
Yes. There, there are laws and regular regulations may be around. Let's say, you know, semi-truck drivers where I believe they are limited. I, I wonder this, this could be just w w we can wonder and let it go at that. If, if there was data or if it just seemed to some lawmakers or regulators reasonable that a certain number of hours per day was the limit on safe driving.
Arnie Barnett (25m 40s):
Well, I don't know the specifics. I think you might say intuitively, it makes sense. And in the absence of data, maybe you should go with what makes sense. You have to remember the trucking industry was direct collated, and that meant that some of these truckers were under an awful lot of pressure to do an awful lot of driving. And the government may have felt it had to step in less, you know, the free market create a situation that was unsafe for the drivers who are being maybe even exploited in a sense of being told you got to get it from Detroit to New York and one day.
Arnie Barnett (26m 20s):
And I don't want to hear any excuses. And if you don't do it, you're not going to get the next time that an opportunity to drive my stuff. So, so I don't know the specifics of that. And I don't know the statistics.
Mark Graban (26m 33s):
Sure. One other question about, you know, flying safety with, or without the COVID dynamic is I could be misremembering. It was a long time ago when I was in your class, but thinking of the risk of, of going up and coming down.
Arnie Barnett (26m 49s):
Yeah. Well, coming down
Mark Graban (26m 50s):
Thinking about a flight of 200 miles versus 2000 miles, so there's still one up one down. Does that have an impact on the relative risks on shorter flights or longer flights?
Arnie Barnett (27m 1s):
Well, it certainly seems you're right. That statistically, most pure accidents take place on landing or takeoff, or I should say that the takeoff climb and the, or the descent landing phase is that in the cruise phase, accidents are very rare. They're not, it's not at all the case they're non-existent, but so in that sense, a 2000 mile flight is not 10 times as risky as a 200 mile flight. However, you see, we have to talk about all the sources of risks. For example, if we talk about nine 11, I don't think it's an accident that they chose transcontinental flights, you know, that we'll be flying at higher altitudes, et cetera.
Arnie Barnett (27m 45s):
So, and they may want to have a little bit more leeway so that if something takes longer or that the plane is not, you know, on the ground at the time something went wrong. For example, a bombing of pan am 1 0 3, this was a flight from London to New York. And they arranged that the bomb would go off about an hour into the flight when it was at, at high altitude. And when you're at a high altitude, the explosive decompression is not survivable. If you're a lower altitude, it is. In fact, you may remember the underwear bomber trying to bomb. What was then a Northwest airlines flight going into Detroit?
Arnie Barnett (28m 27s):
Well, he attempted to ignite the bomb when the plane was at something like 12,000 feet, and it is said, and I believe it to be true that even if he had succeeded that he didn't succeed, of course he would not have destroyed the plane because the force of decompression at 12,000 feet is not enough to destroy the plane. It's just, it creates a hole, but it doesn't cause the plane to plunge, et cetera. So for that reason, you could say it was more important to be, it might be more important to terrorist to go to longer flights than shorter ones. And in the if, to the extent that that's a source of risk, you know, there, you might say the longer flights are a little bit more dangerous than shorter ones taking everything into account.
Mark Graban (29m 13s):
I wanted to ask you about a piece and I'll link to this in the show notes, you wrote a piece in March late, March 31st, 2022, that headline was don't end. The mass mandate for US Airlines, who said ending the requirement now would be a serious mistake. And I asked this to somebody who still does wear a kn 95 mask on board. And you're 15, you're 15% estimates seems pretty accurate. What, what was the statistical basis for saying that then? And it sounds like you still agree with that statement or, or when, when, when would the statistics show, what would you look for in the data to say, maybe that's not necessary at some point?
Arnie Barnett (29m 57s):
Well, part of what I was doing was reacting to what USairlines were saying, and US Airlines do a fabulous job at preventing crashes, but to other sources of risk, I don't think they especially distinguished themselves and United airlines and said that the risk of getting COVID-19 or United flight was, was nearly non-existent quote unquote in Southwest airlines. Not to be out done said that it was virtually non-existent well, it's not non-existent. And what I pointed to in terms of data in that article was the many flights overseas in which we have evidence in the peer reviewed medical literature that an infected person caused COVID to many others on the plane to several others.
Arnie Barnett (30m 48s):
Now, not everyone you'll have to be seated pretty close to the infected passenger. It is true that the planes have good air air, air purification systems. So if I'm in seats eight, a and someone is infected in 20 B, it's extremely unlikely. This will affect me. But if the infected person is in 20 B, that's not great news for the person in 28 or 20 C or a 21 B. So I have done calculation, but I cited several instances there. You know, one of the earliest examples was a passenger on a flight from London to Vietnam here on air, via Vietnam era airlines was infected with COVID-19 and of the 12 business class passengers.
Arnie Barnett (31m 38s):
Within two rows of her, 11 of them came down with COVID on the flight. Okay. And whereas other business class passengers who are further away from her, the risk was I think one and nine or something. And I remember the airlines blessed them or saying, oh, they got infected in the lounge before they got on the plane. It had nothing to do with the plane they had to, but, you know, just haven't really makes sense. I mean, unless you believe that people arrange themselves in the lounge, in the same configuration, as on the plane, why would you assume that only the ones who were close by on the plane are close to enough in the lounge?
Arnie Barnett (32m 20s):
So I think so there was several examples like that and the USairlines, I think, frankly, they pretend that such examples don't exist or they say, well, these are outliers. Of course we don't have actual data from the United States because we have such a poor system of finding out about COVID. I mean, how many people came down with COVID in the United States, in bowling alleys or on local buses or on plans? We have no idea how many got on the vehicles carrying COVID and certainly no idea which people caught COVID on the plane. In fact, even the people who did, they can't be sure.
Arnie Barnett (33m 0s):
I have a neighbor who believes she came down with COVID on a flight, however it's possible. She was infected before the flight or after the flight. You can't be sure. So I, so I was trying to use actual data. They are not in complicated calculations, but just to say, let's not have people telling us the risk is non-existent
Mark Graban (33m 21s):
Sure. Even if you can't calculate it, precisely recognizing it's non-zero is worth acknowledging.
Arnie Barnett (33m 25s):
Yes. Particularly when you have people telling you it is essentially zero, you know, Delta airlines, which early in the pandemic was doing a wonderful job suddenly started saying, now that COVID is like a seasonal flu. We might as well stop treating it as something unusual. And they got a lot of pushback for that and appropriately. So it's not down to the level of a seasonal flu at, at this point.
Mark Graban (33m 54s):
Yeah. So Arnie, one other topic I wanted to ask you about here even real, real briefly, you know, you do a lot of what work in aviation safety, and there's a lot of interesting articles. I'll, I'll link to, for people who want to read more in the realm of elections and politics. I, I heard a recording of a talk that you gave about the electoral college. Is, is that a miss, is the electoral college in your view, a mistake? Is it something that can be fixed in a way that might lead to more just outcomes that may be more accurately reflect the view of a voting population?
Arnie Barnett (34m 35s):
Well, first of all, I have done some work on the electoral college, but I have not taken any view on whether or not it's a mistake. I mean, there, you can have an argument, a political argument about how it overrepresented small states, et cetera, and others might say, no, it prevents a few states from dominating the country. So I haven't taken part of that argument. What I have done is to say, maybe there's a way of coming up with a compromise that all sides can live with. And in particular, what I said was let each state's influence be the number of electoral college votes it has now.
Arnie Barnett (35m 15s):
So to the extent that that benefits a state like Idaho, fine, let's keep that in place. However, in a given state, instead of giving all the electoral votes to the winner there and all or nothing, divide the electoral college votes based upon the split of the popular votes. So if you have a state, for example, with four electoral votes and the two candidates essentially get very close to 50% each, then maybe a two to two split, better reflect what happened in the state. Then for zero, for the one that's slightly better than the other, where one half of the population is exalted over the other half.
Arnie Barnett (35m 55s):
And I worked out what would have happened if we had this system in place, you know, based upon the patterns we saw and argued that it really had some of the advantages that people that people want to popular vote. I said, you know, look in practical terms, you essentially have a popular vote. If you actually look at the outcomes under this system, they very much resembled the overall popular vote. In fact, I think it was a factor of 50 closer to the popular vote than the actual electoral college vote. At the same time, the small states get to keep the advantage they have. Now, in fact, ironically, it actually increases their advantage, but that's a technical issue.
Arnie Barnett (36m 38s):
But anyway, so their era was just saying, here is a possible revision, which might actually make a lot of people happier. You'd be closer to a popular vote. You would pre protect those aspects of the electoral college that people don't want to see go away. And I've just, you know, I had the, you know, threw the idea out there and, you know, I don't think, I don't think we can assume it's going to happen very soon.
Mark Graban (37m 7s):
Sure. And is that small state advantage? For example, Wyoming, with the very, very small population gets two electoral votes, correct?
Arnie Barnett (37m 15s):
It gets three, I get three gifts. It has two senators and representatives.
Mark Graban (37m 20s):
Yeah. That's part of the advantage is having two senators, just the same as
Arnie Barnett (37m 25s):
That's right for us. That's, that's really why it's, non-linear the number of members of the house is roughly proportional to the state's population. But the two senators for a small state mean a state that really would only justify one member of Congress in the house gets three. But again, that really in the scheme of things, if you actually look at what happens, you really shouldn't go to great extremes to try to take that away. And also the greater way to the smaller states, doesn't really redound to one party's favor because the smaller states include Wyoming. You mentioned that North Dakota have states that go very heavily Republican, but it also includes states like Delaware.
Arnie Barnett (38m 10s):
You know, Washington DC has three electoral votes. It's the most democratic enclave in the solar system. So in other words, the small states basically are pretty much matched Republican Democrat. So, so doing something that would involve taking their slide influence, extra influence away, you know, it, it just isn't worth it. You don't need it to get close to your objective. And I think that's what I tried to do basically to say, well, mathematically, can we come up with a solution that for non-mathematical reasons is attractive
Mark Graban (38m 43s):
And there's a difference between the math being correct. And the argument of an opinion of which is better red state versus blue state, large state versus small state, that's different than the math that might support one choice or another.
Arnie Barnett (38m 60s):
No, that's true. And not only is it true, but I think if you do offer sort of a technical solution like this, I think it's important that you see neutral. If it sounds as if this technical solution, which sounds very abstract is really an excuse to benefit. Let's say the Democrats, then people will call you out for that and say, you know, you pretend to be neutral, but in reality, it's a scheme. It's a scheme that benefits you. But I think what I try to argue is actually, that's not true. It's not true that in the systematically benefits one party,
Mark Graban (39m 34s):
Right. And I'm not really asking this or, you know, but, but people might assume, oh, well, you know, professor Barnett coming to us from the people's Republic of Cambridge must obviously view, you know, so yeah, that probably enters into people's analysis. Even if you're trying to, even if you're being completely neutral and people will make all sorts of bad assumptions.
Arnie Barnett (39m 55s):
Well, I'll tell you, it used to be that MIT was viewed as technical apolitical. Harvard was always viewed as you know, Senator McConnell, judge John McCarthy called it the Kremlin on the Charles, but MIT used to have a reputation as being sort of apolitical and scientific, I think. And I think it's unfortunate. MIT now has a reputation of being very much woke. You might say that it does seem to have a tilt toward the left. So I think that's unfortunate and whether or not people think that when I make these suggestions, that outlook, I can't do anything about that, but I can say, why don't you look at what was, what I'm saying and see whether or not there was you sense bias.
Mark Graban (40m 44s):
Yeah. It's also a university as they, I think I've heard at least once the university that World War II built, and there's a long history and association with no defense systems and so different people could have their own view based on some slice of data or perception.
Arnie Barnett (41m 2s):
I think that's true. And in fact, MIT recently restored the requirement for the just submit AT scores. And I think it got a lot of praise from people like, like the Wall Street Journal for at least if not, for acknowledging that sometimes objective measures of talent may have some value. So I think MIT is not viewed the same way as Harvard, but I also think MIT is not viewed as a political in the way that it used to be
Mark Graban (41m 34s):
One quick story that that comes back to mind when I was there on campus in 1997, there I was in one of the engineering building coffee shops. And there were what seemed to be kind of a gaggle of professors emeritae who would get together and have coffee and shoot the breeze. And I remember distinctly overhearing a couple, these, these seasoned professors complaining about the quote unquote well-rounded students and that they had no interest in how many clubs or activities somebody was in. And, and one of them, I think this is a direct quote, cause this got burned in the memory. He said, nerds make for good students. It was I think, coming out pro nerd, but
Arnie Barnett (42m 17s):
Yes, yes. I think that's true. I think we'll get into college now. You know, they do ask for all kinds of things about extracurricular activities. And that may be unfortunate if the aim is to take people let's say at MIT were most likely to benefit from an MIT education. I think the fact that you played the trumpet may not be as important as whether or not you're really good at math.
Mark Graban (42m 49s):
Sure, sure. Well, one other final question for you, Arnie, you know, I want to give a quick plug to what's now called the Leaders for Global Operations program that I was fortunate to attend. There was no way I didn't have the test scores. I wouldn't have been admitted as an undergraduate, but I think I've benefited greatly from the LGO program. And you know, you, you advise a lot of students who are doing their internships in their, in their thesis. So it kind of open-ended question here, you know, tell us what, what you enjoy. This was kind of a leading question. Forgive me, but what you enjoy about working with the LGO students?
Arnie Barnett (43m 25s):
Oh, I do enjoy working with the LGO students. Well, my stood probability course is shorter now because it's believed that many of them have already had some probability and they need to learn machine learning now. So in fact probability was, was truncated. So in addition to statistics, they can get machine learning. And I think on balance, that's the right decision, but it would reduce the teaching. I do, but I look I've had wonderful experiences and indeed being an internship advisor, you know, has not only been interesting in its own, right, but it's actually made me a better teacher.
Arnie Barnett (44m 9s):
And in fact, I remember one particular example. I had an intern who was at some paper mill in Northern Wisconsin. And you know, in, when you teach probability, there's the classic birthday problem. You know, if you have 50 people in a room, what's the chance. I know two of them share a birthday and birthday. Sharon turns out to be more probable than you would first have thought, but it sounds a little bit like a party game. You know, what's your birthday? Well at this paper mill, they actually had a problem that when they produced the paper, there was certain random tears that would arise in the paper in the page. And if two random ties, tears were close together, the paper was weakened in such a way that you couldn't really sell it because when people try to write on that page would sort of rip apart and they said, this is just unfair.
Arnie Barnett (45m 1s):
Why does it happen so often when there are just a few tears randomly, randomly in the paper and the response to them actually was related to the solution to the birthday problem. Namely, even though there's no need for overlap, the laws of probability say there's a very good chance. You'll have some. So I enjoyed enormously working with students on projects like this because I learned things. It gave me a certain amount of street cred for these things. And the students themselves are just a delight to teach really nice people. The summer is a wonderful time for, for that. So I have been very happy with the LGO program and indeed on 9/11, I was in Seattle at an LGO conference,
Mark Graban (45m 47s):
Alumni conference, the
Arnie Barnett (45m 48s):
Alumni conference, and I might talk was titled “air safety, end of the golden age?” and I had prepared it before the what happened that morning, et cetera. So it almost seemed as if, when I said enter the golden air question mark. I think some people thought there was no need for a question mark, et cetera, but no, I've had very, very good memories of the program. It continues to be a very strong program and continues to have lots of support in industry. And that's because it's a good program. Yeah.
Mark Graban (46m 23s):
Well, and, and it's, it's at the, at the risk of sounding like a circular definition, part of why it's a very good program is professors like you who are actively involved and go above and beyond. And I, you know, I think of, you know, Steve Graves and the late Don Rosenfeld as just a couple examples in, there are many, many more. So the, the, the, the passion for the students and the teaching and the learning, and it's very, very appreciated. So thank you for that Arnie. And thank you so much real pleasure that we could do the podcast here today.
Arnie Barnett (46m 56s):
Well, it was a pleasure being here with the podcast, and I wish all of your listeners and those who watch it to have very safe flights and happy landings.
Mark Graban (47m 6s):
Thanks, Arnie. And so, and again, if you want to find, you know, different articles that Arnie has written either from an academic perspective or times he's been cited in the news media or written op-eds, I'll put links to those in the show notes. So thanks. Thanks again. Well, again, thanks . To, for being a great guest today, to learn more about him, to find links to his work. And there's actually a link to the New York times article in his, my favorite mistake story. You can find all of that in the show notes, or go to markgraban.com/mistake169. As always. I want to thank you for listening. I hope this podcast inspires you to reflect on your own mistakes, how you can learn from them or turn them into a positive I've had listeners tell me they started being more open and honest about mistakes in their work.
Mark Graban (47m 57s):
And they're trying to create a workplace culture where it's safe to speak up about problems because that leads to more improvement and better business results. If you have feedback or a story to share, you can email me MyFavoriteMistakePodcast@gmail.com. And again, our website is MyFavoriteMistakePodcast.com.