The best and worst airports for flight delays, according to government data

Published on November 21st 2019

Domestic air travelers flying through Newark Liberty International Airport are more likely to face delays than those flying through the nation’s other large airports, based on the most recent year of data from the U.S. Bureau of Transportation Statistics (BTS), part of the U.S. Department of Transportation.

Data on more than seven million commercial domestic flights between Oct. 1, 2018 and Sept. 30, 2019 show that the average flight within the US arrives at the gate 14 minutes after a scheduled arrival (excluding cancelled or diverted flights). About 7% of flights arrived more than an hour late.

Across the nation’s 30 busiest airports—each of which had more than 69,000 arriving domestic flights during this period—Newark (EWR) ranked the worst for the percent of flights arriving late and the average length of delay. These 30 airports account for 65% of the nation’s domestic flights.

Twenty-nine percent of flights departing from Newark arrived at their destination late, with the average departure from Newark arriving 22 minutes late. Of flights coming from somewhere else to Newark, 27% arrived late, with an average delay of 26 minutes. The Department of Transportation defines a late flight as a plane arriving at least 15 minutes after its scheduled arrival.

Other large airports, including Chicago O’Hare International Airport (ORD) and New York’s LaGuardia Airport (LGA), also have a record of late flights. About 25% of flights departing from both airports were deemed late, with the average flight arriving more than 19 minutes late.

Newark, LaGuardia and San Francisco International Airport (SFO), rank the worst in terms of arrivals at the airports. More than a quarter of all flights arriving at the three airports were late.

There are some regional differences between the 351 commercial airline-served airports in the United States. Comparing airports across the Census Bureau’s primary regions — Northeast, Midwest, South and West — showed that airports in the West had both the shortest delays and lowest likelihood of lateness. Northeastern airports are more likely to experience delays, and longer ones, at that.

The scatterplot below shows where each airport stands in terms of percent of flights late and the average delays per flight. The size of the circles represent the number of flights arriving at each airport.

There are more people flying now than ever: more than 87 million flew domestically or internationally in July, the highest air travel total for a month since at least January 2000. There were 8.7 million domestic commercial flights in 2018, the largest annual total since 2008.

Despite this increase in air travel, the on-time arrival rate for commercial domestic air travel has hovered around 80%. Last year, 83% of flights reached their destination on time, down from 84% in 2017.

On-time flights by year

The on-time rate during the first nine months of 2019 was 82.2% for a nearly seven million flights.

Air travel delays are largely seasonal and the BTS data shows how the reasons for delays change throughout the year. Air traffic is busiest during the summer months, and in 2019, that’s when customers were most likely to experience delays of 15 minutes or more. In July, 17% of flights were late by 15 minutes or more. Those flights were delayed on average 76 minutes, about 40% of which could be attributed to late-arriving aircrafts. There were spikes in other reasons for delays, too, including air traffic, security breaches or carrier issues (such as aircraft cleaning, baggage loading, or fueling). Average delays in October 2018 for flights delayed 15 minutes or more were just below 60 minutes.

Travel volume surges around the holiday season, but Thanksgiving week in 2018 was actually a below average week for airport delays. While other weeks had average flight delays of approximately 14 minutes, Thanksgiving week in 2018 had an average delay of 10.5 minutes. This may partly be due to the very low flight volume on Thanksgiving Day, Nov. 22, which had the fewest number of scheduled flights of any single day between during the 12 month period. The Sunday after Thanksgiving, however, had an above average flight delay, potentially due to high passenger volume.

Share Newsletter

Methodology

Data was aggregated from the Bureau of Transportations Reporting Carrier On-Time Performance page, where each month was pulled individually as a file and then merged. For arrival delays, we used the variable ArrDelayMinutes, which represents the difference in minutes between the scheduled and actual arrival time, and sets early arrivals to 0.

Sources

Bureau of Transportation Statistics