Airline Overbooking: The Math Behind Bumped Passengers

Why Your Boarding Pass Isn't a Guarantee

You've paid for your ticket, checked your bags, and made it through security. Then comes the announcement that changes everything: the flight is full, and you've been bumped. It feels personal, almost like a mistake. But here's the uncomfortable truth: airline overbooking is not an error. It's a calculated, deliberate strategy built on probability and profit margins, and it happens far more often than most travelers realize.

Overbooking means selling more tickets than there are seats on the plane. Airlines do this because they know from years of historical data that not every single person who buys a ticket will actually show up. People miss connections, get sick, change plans, or simply oversleep. If the airline sold exactly 200 tickets for a 200-seat aircraft, they'd risk taking off with empty seats, which means flying at a loss. So they sell 220 tickets for those 200 seats, betting that a predictable number of passengers won't make it.

The same logic applies beyond airplanes. Hotels routinely overbook their rooms for the same reason, hoping that a few no-shows will keep them from losing revenue on empty beds. It's a numbers game played across the travel industry, and the stakes are real. Estimates vary from year to year and depend on how the count is done, but air travel consumer reports suggest that tens of thousands of people are denied boarding annually despite holding valid tickets.

Here's how the math works out in practice. Imagine a flight with 200 seats and a ticket price of $200. If the airline sells exactly 200 tickets, they earn $40,000. But if they sell 220 tickets and 15 passengers don't show up, that's 205 people arriving for 200 seats. Five passengers get bumped, and the airline compensates them, say $500 each, which costs $2,500. But they collected $44,000 in ticket sales. Take away the compensation, and they're still ahead by $1,500 compared to selling just 200 tickets. That's the core of airline overbooking: a small number of unhappy passengers is cheaper than a plane full of empty seats.

To make this work, airlines lean heavily on statistics. They track show-up rates meticulously, and a commonly cited average is that each ticketed passenger has about a 90% chance of arriving for their flight. That figure shifts with route, time of day, and season, but it gives the airlines a solid starting point for their calculations. They're not guessing. They're running probabilities on every single departure.

The Simple Math Behind Selling 220 Seats on a 200-Seat Plane

Picture a 200-seat airplane and an airline that sells 220 tickets at $200 each. If 185 passengers actually show up, the airline collects 220 times $200, which comes to $44,000 in revenue. That is the core arithmetic that makes airline overbooking so attractive.

Now compare that with the conservative approach. If the airline had sold exactly 200 tickets, the math would give 200 times $200, or $40,000. The difference between those two numbers is $4,000. That is the money the airline would leave on the table by playing it safe, assuming some passengers never board.

But what happens when too many people do show up? Imagine 205 passengers arrive for those 220 booked seats. Five of them get bumped. If compensation for each bumped passenger runs $500, the airline spends 5 times $500, or $2,500, to make things right. Subtract that from the $4,000 gain and the airline still walks away with $1,500 in extra profit.

The logic holds together because no-shows are not rare. Travelers cancel, miss connections, or simply change their minds. An airline that refuses to account for that reality flies with empty seats, and an empty seat generates zero revenue. Paying a few passengers to take a later flight often costs far less than the money lost on vacant rows.

This is why the practice exists in the first place. Airlines are not gambling blindly; they are running the numbers. The $1,500 net gain in this example comes straight from the gap between what overbooking captures and what compensation eats away. It is a calculated trade, and the arithmetic checks out under these assumptions.

How Airlines Decide How Many Extra Tickets to Sell

Airlines lean on a branch of mathematics called the binomial distribution to figure out exactly how many extra seats they can safely sell. The core idea is simple: each ticketed passenger is a trial, and showing up for the flight counts as a success. By modeling thousands of passengers this way, carriers can predict with surprising precision how many people will actually walk through the boarding gate.

The whole system rests on historical data. A commonly cited average is that each ticketed passenger has roughly a 90% chance of showing up for their flight, though that number shifts with routes, seasons, and fare types. An airline selling 220 tickets on a 200-seat plane knows that, on average, about 15 to 20 people won't make it. The trick is figuring out the odds that more than 200 will show up, because that's when things get expensive.

This is where the binomial distribution earns its keep. It calculates the probability of getting a specific number of successes across a fixed number of trials. For a 100-seat aircraft with a 5% cancellation rate, the math gets very concrete. Selling 101 tickets gives only a 0.56% chance of overcapacity. Bump that to 102 tickets and the probability rises to 3.40%, still comfortably below the 5% threshold most airlines use as their comfort zone. Go one step further to 103 tickets, though, and the risk jumps to 10.65%. That single extra ticket nearly triples the chance of having to bump someone.

The foundation for all this was laid in the 18th century by a statistician named Abraham de Moivre. He noticed something remarkable while working on gambling problems: as the number of trials grows, the binomial distribution starts to take the shape of a smooth, symmetrical curve. That curve is what we now call the normal curve, or the bell curve, and it became the backbone of modern probability theory.

A coin flip makes the concept intuitive. Flip a coin ten times and the most likely outcome is five heads, represented by the tallest bar on the distribution chart. Four or six heads are nearly as likely, while zero or ten heads are extremely rare. The same logic applies to a flight full of passengers. Most will show up, a few won't, and the airline just needs to know where the risk of overcapacity starts to outweigh the profit from selling one more ticket.

The 102-Ticket Sweet Spot: A Step-by-Step Calculation

Let's make this concrete. Imagine a 100-seat aircraft and a 5% cancellation rate, meaning p = 0.05. The airline's goal is to keep the probability of overcapacity below 5%. Selling 102 tickets does exactly that, with a 3.40% chance of having more passengers than seats.

Here's how the math actually plays out, ticket by ticket. If the airline sells 101 tickets, the probability that more than 100 people show up is just 0.56%. That's remarkably safe, but it also leaves money on the table. Bump it to 102 tickets and the risk rises to 3.40%, still comfortably under the 5% threshold. Push to 103 tickets and the probability jumps to 10.65%, blowing past the limit. The optimal number for this flight is 102 tickets.

The pattern becomes clearer when you see the full range. Selling 104 tickets brings a 23.08% chance of overcapacity. At 105 tickets, it's 39.24%. The probabilities climb steeply from there: 106 tickets at 56.22%, 107 at 71.21%, 108 at 82.67%, 109 at 90.40%, and 110 tickets at a staggering 95.08%. Each additional ticket adds significant risk once you cross that sweet spot.

This simple model gives you the core idea behind airline overbooking, but real-world decisions get more complicated. Advanced approaches factor in the cost of bumping a passenger. If compensating denied boarding passengers becomes too expensive, the airline overbooking strategy stops being worth it. The financial risk simply outweighs the benefit of filling more seats, and the airline's reputation takes a hit on top of the monetary loss.