MIT Professor: Flights Booked to Capacity May Present 1-in-4,300 Chance of Contracting COVID-19

MIT Professor: Flights Booked to Capacity May Present 1-in-4,300 Chance of Contracting COVID-19
Joe Cortez

A new pre-print thesis from an MIT professor suggests a flyers could face a reduced risk of getting COVID-19, even of the aircraft is booked to capacity. Using an equation model, the odds of catching COVID-19 from a flight booked to capacity may be 1-in-4,300, while flying with an empty middle seat could reduce those odds to 1-in-7,700.

For those concerned about airlines booking to capacity, a new model suggests chances of contracting the novel Coronavirus may be low when aboard an aircraft. A pre-print study submitted by a Massachusetts Institute of Technology (MIT) professor suggests the chances of catching the viral infection from middle-seat flyers could be low — but flying with an empty seat between flyers may reduce it even further.

An Emphasis On “Rough” Numbers for Full Flights

Since the beginning of the COVID-19 pandemic, health experts continually asked the public to wash hands, wear facemasks and social distance by at least six feet wherever possible. To those ends, airlines have instituted mandatory face mask policies on all flights, but are split on appropriate social distancing measures.

A 2020 CarTrawler/Ideaworks study found that while Delta Airlines, Southwest Airlines and JetBlue were committed to blocking middle seats to promote health measures, American Airlines, Spirit Airlines and United Airlines were booking flights to capacity. Does blocking middle seats actually promote social distancing on aircraft?

According to Arnold Barnett, the George Eastman professor of management science and statistics at the MIT Sloan School of Management, the odds may not change significantly in either case. In a pre-print thesis, his math suggests that flyers may not face an overall severe risk of getting Coronavirus from sitting on a commercial flight.

“Answering that question [the effectiveness of social distancing on aircraft] entails major complications and uncertainties, which can easily lead one to throw up one’s hands,” Barnett writes. “But even a rough approximation of the risks at issue seems preferable to clashes of unsubstantiated conjectures. This paper strives for such an approximation, with an emphasis on the word ‘rough.’

Using an equation that take into consideration new Coronavirus infections by state, face mask usage aboard aircraft and passenger load, Barnett suggests that the risk of contracting COVID-19 from full flights could be as high as 1-in-4,300. When the middle seat isn’t booked, that risk could drop to 1-in-7,700.

“These estimates imply COVID-19 mortality risks to uninfected air travelers are considerably higher than those associated with plane crashes,” Barnett writes in the study abstract. “But probably less than one in 500,000.”

Early Report Still Requires Review

While Barnett’s report presents a “rough” view of risk, the report is incomplete. First, the professor acknowledges that the equation model only considers contact between those on airplanes. The report does not consider 2019’s average aircraft load factor of 85.1 percent, or the odds of coming in contact with someone who may have COVID-19 during boarding, deplaning, or in a lavatory. The equation model also assumes everyone on a flight will wear a mask, and does not take into account talking or other social activities while flying.

Moreover, the paper is still in pre-print phase, and has not been certified through peer review. The Cold Spring Harbor Laboratory, which hosts the pre-print server where the study was published, warns studies “…should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.”

FlyerTalk does not consider this pre-print study as “established information,” and may revise this story after the paper goes through peer review.

But if Barnett’s math is upheld, the professor suggests that there could be a lower risk of contracting COVID-19 if middle seats remain open through the pandemic. It remains unclear how much of that risk could be reduced, and if the chances of “…getting infected during a flight is any higher than the risk associated with everyday activities during the pandemic.”

“The calculations here, however rudimentary, do suggest a measurable reduction in COVID-19 risk when middle seats on aircraft are deliberately kept open,” Barnett writes. “The question is whether relinquishing 1/3 of seating capacity is too high a price to pay for the added precaution.”

View Comments (27)


  1. SixMileMtn

    July 14, 2020 at 11:45 am

    Does the research take into account the covid positivity rates of the origin points of the flights? For instance NY’s positive test rate is presently about 1% and Florida’s is about 18%. And Miami-Dade’s is about 26%. Surely some factor for rate of infection based on geography should be considered.

  2. Dave510

    July 14, 2020 at 2:58 pm

    He should also calculate what the additional price pressure coming from an empty middle seat would reduce load factor by, and how the decreased load factor will reduce spread further.

  3. KenTarmac

    July 14, 2020 at 6:23 pm


  4. sekhmet101

    July 15, 2020 at 4:09 am

    He should stay in his lane and let the infectious disease doctors and virologists deem what the odds are.

  5. riku2

    July 15, 2020 at 5:04 am

    >>chances of contracting the novel Coronavirus may be low when aboard an aircraft<<

    Surely the high/low depends on how prevalent the virus is in your normal environment. I live in Finland with very low number of virus cases so even 1 in 4300 is probably much higher than I face in day to day life on the ground. For residents of Texas and Florida though the chance in the plane might be lower than normal life.

  6. HomerJay

    July 15, 2020 at 5:40 am

    This is nonsense. If the guy in the middle seat is positive and coughs on a five hour flight….

    I would NEVER take this prof’s advice at a sports book in Vegas, and that’s for sure.

  7. willieron

    July 15, 2020 at 6:05 am

    The risk of contracting it when no one on the plane is infected is 0%. What is the risk if 1 person is infected? 2? What if the person sitting next to me is infected? I get that he’s calculating an overall number that incorporates the odds of an infected person being on the flight, but what am I supposed to do with that information?
    The odds of someone flying who has been exposed and may be infected would seem to be higher than the overall population, seeing as people who are basically locked down or are limiting contact and travel won’t be flying.

  8. Xandrios

    July 15, 2020 at 6:05 am

    Keep in mind that having the middle seat empty will also simply reduce the number of people around you. Leaving the window or isle seat empty would result in the same. Fewer people around means lower chances.

  9. overdahill

    July 15, 2020 at 6:51 am

    I have stock in the Brooklyn Bridge and would like to sell it. It is a good value and likely to appreciate in value at a rate analysis indicates of about 8% per year.

    Any one interested please contact soon as the number of shares are limited and there is likely to be very
    high demand.

    I am sure that anyone here that finds the article to have merit would have an investment interest.

  10. enggeol

    July 15, 2020 at 6:53 am

    The risk of catching the virus is but part of risk management – you have to factor in the impact as well. If impact is low then a 1 in 4300 is fine and yes death is low probability for the younger members of society. Mortality however is NOT the only impact. It is becoming clearer here in the UK that quite a lot of those recovering are getting long term life changing impact on their bodies the most obvious being scarred lungs. Factor that in the equation and it would be interesting to see the risk of severe impact including death and then one can decide whether that is a risk one wishes to really take.

  11. KRSW

    July 15, 2020 at 6:57 am

    Haven’t we had enough bad information with mathematical models and COVID-19 to realize that having baboons throwing poo at a dart board is probably a more accurate estimate? Imperial College of London used this method, but with blindfolded baboons.

  12. Lackey99

    July 15, 2020 at 7:45 am

    I read the study yesterday. It’s WORTHLESS. It does not take into account the type of mask people are wearing and the amount of time they take it off during the flight. It also does not take into account how much outside air is being introduced into the airplane or if the airplane has a purification device (not a filter, but a purification device). UTTERLY WORTHLESS.

  13. jimbous

    July 15, 2020 at 7:59 am

    How does the old saying goes:
    LIes, Damn Lies, and Statistics?

    Now, let’s have some fun and statistically figure the chances of getting infected outdoors with a greater than the 2 meter personal distance but without a face mask.
    Surely, the chances will be far lower than the above airplane numbers.
    Maybe the chances will be about the same as winning the Powerball lottery jackpot, about 1 in 292,201,338, or the Mega Millions lottery jackpot, about 1 in 302,575,350?

    Surely, the MIT brains can figure that out!

  14. jammex88

    July 15, 2020 at 10:00 am

    Although ppl seem to not want to believe it, the risk of getting infected by a virus on a plane is low due to the constant filtration of the air. Is it possible? Of course, anything is possible. It’s not likely.

  15. xlnja

    July 15, 2020 at 10:36 am

    I can’t believe how many people are freaked out by this virus…totally irrational. The numbers do not justify it. Our society has become so weak and easily manipulated. We’ve gone from the greatest generation where young men voluntarily threw themselves to near certain death on D-Day to worrying about a virus that has a death rate of 0.2%. SMH

  16. android

    July 15, 2020 at 11:46 am

    This is a classic example of statistics posing as facts when actually it’s just opinion!

    There are so many assumptions built into any statistical model that at the end of the day most models are just the author’s opinion. I would be very suspicious that this author has a vested interest in encouraging air travel.

    Let’s look at the assumptions. The most important assumption is the viral status of the people sitting closest to the passenger. Unfortunately, this is highly variable depending on where that person is living. Some states have a very low percentage of infected people, and some states, like Florida have a very high percentage. If I were flying from Miami, for instance it might be reasonable to assume that one in five people could potentially be positive.

    But ultimately, it’s a binary issue. The person sitting next to you, or if you’re in a middle seat the person sitting on either side of you, is either infected or not infected. If they are infected, then the problem becomes how likely is it that you will become infected. It is clearly higher than these odds. People do not wear masks the entire flight. It is impossible to drink fluids or eat while wearing a mask, and if you and your seat neighbor happened to be eating at the same time, then you are both completely vulnerable to each other.

    He also made an assumption that people traveling are less likely to be symptomatic, but I think this is a mistake. The other assumption he failed to make which he should have made is that there are different categories of people out in the world right now in terms of risk-taking. Most people who are trying to avoid risk are also avoiding flying. So the population of people on an airplane is preselected for people who are taking more risks in general. That makes it even more likely that they are positive for this coronavirus.

    Anyway, if his model is off by a factor of 100, then the risk would be one in 43 for a full flight, which means that there will be several infections passed on on every single flight. That’s simply unacceptable risk.

    The bottom line is simple. The government should subsidize airlines by buying all the middle seats on any full flight. Then leave them empty. Meanwhile, if you don’t need to fly that don’t fly. And if you do fly, make the simple assumption that the people sitting near you are all positive for this virus. That means leaving your mask on the entire flight, bringing straws so that you can drink fluids without taking your mask off, and skipping food.

    Good luck. Stay safe!

    PS: The virus doesn’t have a death rate of 0.2%. In fact, in many vulnerable groups, (the aged, people with diabetes, heart disease, etc.) the death rate is much higher, as high as 2-5%. Please don’t pass on false facts. Yes, if you are 20 years old, and perfectly healthy, you will survive, but even these patients are showing longer term harm from the illness.

  17. RalphW

    July 15, 2020 at 12:21 pm

    The paper is a rough estimate; an approach;. nothing more.

    From the paper itself:

    “What does the evidence suggest about the wisdom of a “middle seat” policy as a safety measure?

    Answering that question entails major complications and uncertainties, which can easily
    lead one to throw up one’s hands. But even a rough approximation of the risks at issue seems
    preferable to clashes of unsubstantiated conjectures. This paper strives for such an
    approximation, with an emphasis on the word “rough.”

    It is an easy read. Anyone who got a ‘C’ or better in basic statistics should be able to digest the equations and the rationale.
    It really would be easy to modify and/or provide additional Q terms for their own “what-if” scenarios. .I may do that for fun later today.

  18. rjpjr

    July 15, 2020 at 12:47 pm

    I think this prof is out of his expertise. At the same time, I’d be grateful for a link to an appropriate study with actual infection numbers. With all the screaming about commercial air travel being the guaranteed way to catch coronavirus, does anyone else find it odd that we are not hearing about this in anything other than a speculative manner? For example, where are the articles incorporating statements such as: CDC data have revealed that (insert percentage here) of coronavirus-positive individuals were on an aircraft during the week prior to their test. Furthermore, contact-tracing data show that (insert percentage here) of coronavirus-positive individuals were in contact with an individual who had recently traveled by air.

    Please don’t tell me we don’t have enough data – the only time we will have enough data is when the virus is no longer a threat.

  19. jimbous

    July 15, 2020 at 12:51 pm

    It was Clint Eastwood who said that America is being taken over by the “pussy generation”.

  20. myisland

    July 15, 2020 at 2:20 pm

    Lots of bad data is used to create bad statistics to be used politically to lock down society.
    For example:

  21. snidely

    July 16, 2020 at 8:48 am

    A more accurate COUNT would be to get a COUNT of the number of on plane employees that have been affected by Covid 19 compared w. general population ages 22 – 65.

  22. c1ue

    July 16, 2020 at 9:29 am

    This is nonsense.
    The professor takes an equation and then adds in WAG estimates on the variables.
    The resulting formula is 100% unvalidated vs. real world data.
    Yet another example of Drake Equation type flim flam.

  23. Gizzabreak

    July 17, 2020 at 1:34 pm

    “… 1:4300 …” to “… 1:7700 …” Sounds better than ‘on the street odds’ … at least in some cities/countries. Maybe the long term solution will be to issue everyone with their own aircraft.

  24. glob99

    July 17, 2020 at 8:33 pm

    Here is a thought experiment on the effectiveness of an aircraft’s HVAC system. If a person smoked a cigarette at his/her seat how many people around the smoker would smell the cigarette smoke? The stronger the smell, the greater the chance of infection.

  25. talkandfly

    July 19, 2020 at 4:05 pm

    For those of you who feel they know all the facts, I am happy you sleep well at night. For those of us dealing with this illness every day, we don’t sleep so well. It is really too bad this has become so politicized. That, fellow FTers, is why we are having so many issues.

    Fact, while most people who get infected, even in the high-risk groups, do get over the acute infection, many are left with long-term consequences including lung, kidney, heart, brain and muscle problems. There are many more of these folks than there are folks who have died. The understanding of the consequences is still evolving.

    Fact, local health care systems are being overwhelmed. In our community, hospitals are full and a significant portion of the beds are occupied by Covid patients.

    Fact, just as happened in NYC a few months ago, in our community, morgues are filling and refrigerated trucks are starting to be recruited for overflow.

    If you need to travel, do so with a great deal of thought. Protect yourself. Protect others around you. Even if you do not believe the severity of the issue, at least practice courtesy by respecting the fear or caution or common sense (you decide) of those around you.

  26. hedoman

    July 24, 2020 at 8:38 am

    And the survey says…..after reading these comments, 1 in 2500 chance of people with a brain buying into anything this paper is selling.

  27. carlosdca


    July 24, 2020 at 9:17 am

    Besides what everybody has commented, to add insult to injury, we are getting the paraphrased version by Joe Cortez (the author of this article), which then turns this into completely fake news, as we don’t know how much of Joe Cortez biases are incorporated in the article.

    Personally, I want to see a different study

    My non-peer reviewed study tells me that:
    – The current air-traveler is:
    1) someone that HAS to fly for work and has to interact with lots of random people. Therefore, their chance of having COVID is 100 times higher than someone that works from home and only goes out to the grocery store.
    2) someone that believes he won’t get it. And just as they are flying planes, they are also participating in multi family back yard parties, religious services, etc. or someone that is going to a county where restaurants and bars are open, and is probably very bad at social distancing and wearing masks. This one has covid.
    3) someone that had a family need (taking care of aging parent in another city, split custody families, etc) and unfortunately has to comingle with lots of strangers. Higher risk of covid.

    So, according to my study, the probability of someone having Covid in a plane is MUCH higher than that of the average population.
    Therefore, the MIT study referenced above is absolute CRAP.

You must be logged in on the FORUM to post a comment Login

Leave a Reply


More in Uncategorized

Cuomo Wants COVID-19 Passenger Testing at New York Airports

Joe CortezAugust 25, 2020

Lufthansa Announces Revived Schedule For June

Jackie ReddyMay 18, 2020

Cathay Pacific to Lay off 300 Cabin Crew

Meg ButlerApril 17, 2020

Copyright © 2014 Top News Theme. Theme by MVP Themes, powered by Wordpress.


I want emails from FlyerTalk with travel information and promotions. I can unsubscribe any time using the unsubscribe link at the end of all emails