Written from the data
Guides
Twelve pieces on what the DOT's on-time database actually supports: when to fly, what causes delays, how much buffer a connection needs, and where the published statistics mislead. Every figure quoted is pulled from the current data period, July 2025 to June 2026, at build time rather than typed in once and left to rot.
Published
Guides
Newark, LaGuardia or JFK: which New York airport actually works
The three New York airports are not interchangeable. Newark Liberty International runs 77.7% on time, LaGuardia 73.4%, John F. Kennedy International 75.7%. Here is what the gap costs.
5 min read · 1,112 words
Chicago: O'Hare against Midway, by the numbers
Chicago O'Hare International runs 71.0% on time and Chicago Midway International 75.1%. The smaller airport wins on paper. Whether it wins for your trip is a different question.
4 min read · 940 words
Washington DC has three airports. The data says they are not equivalent.
Ronald Reagan Washington National runs 73.3% on time, Washington Dulles International 81.1%, Baltimore/Washington International Thurgood Marshall 77.6%. The one closest to the city is not automatically the right answer.
4 min read · 929 words
San Francisco, Oakland or San Jose: the Bay Area airport question
San Francisco International runs 77.3% on time. Oakland International manages 79.5% and Norman Y. Mineta San Jose International 82.9%. The reason is a specific and unusual piece of geography.
4 min read · 919 words
Los Angeles has more than one airport, and LAX is rarely the best of them
Los Angeles International runs 80.4% on time. John Wayne Airport manages 81.5%. In a region defined by driving, the second airport is often the better trade.
4 min read · 922 words
Miami or Fort Lauderdale: the South Florida airport trade
Miami International runs 73.8% on time and Fort Lauderdale-Hollywood International 71.5%. Twenty-five miles apart, one weather system, two different operations.
4 min read · 932 words
Thanksgiving flying: what the data says about the busiest week in US aviation
Thanksgiving delays are not unusually bad. What is unusual is the cost of landing in the wrong tail of them, because there is no spare seat anywhere in the system.
4 min read · 931 words
Christmas and New Year: the two weeks where a delay costs the most
December is the weakest month in much of the US calendar and the last two weeks are the ones with no room to recover. The risk is not the delay, it is the queue behind it.
5 min read · 1,039 words
Summer thunderstorms: why afternoon flying breaks down from June to August
Convective weather is the one delay cause that runs on a daily clock. It builds through the afternoon, peaks between 3pm and 8pm, and does it again tomorrow.
4 min read · 909 words
Winter storms: how US flying actually fails between December and February
Winter delay is not a steady drag. It is a small number of days on which whole regions stop, and an average blends those with three hundred ordinary ones.
4 min read · 938 words
Coast to coast: how the transcontinental routes actually perform
Long flights carry more scheduled padding, which quietly flatters their on-time figures. Here is what the transcontinental routes really deliver, and what the number hides.
4 min read · 949 words
Red-eye flights: the overnight numbers nobody looks at
The overnight departure window performs far better than its reputation. It flies into empty airspace on an aircraft that has nothing to inherit.
5 min read · 1,034 words
The airlines you fly without knowing it
A flight sold as American, Delta or United is often operated by a company whose name you have never seen. The DOT reports it under the operator, and their records differ.
5 min read · 1,022 words
Delta, United, American: the three big networks compared
Delta Air Lines runs 81.6% on time, United Airlines 79.0%, American Airlines 74.4%. The gap is real and smaller than it looks.
5 min read · 1,033 words
Spirit, Frontier and Allegiant: what the ultra-low-cost record actually shows
Spirit Airlines runs 73.1% on time, Frontier Airlines 73.2%, Allegiant Air 74.1%. The interesting number is not the average. It is the variance.
4 min read · 929 words
Alaska, JetBlue and Southwest: the carriers between the extremes
Alaska Airlines runs 79.1% on time, Southwest Airlines 76.6%, JetBlue Airways 72.7%. Three very different operations that get grouped together for no good reason.
4 min read · 929 words
Short-haul shuttles: why the busiest routes in America are the hardest to keep on time
A 55-minute flight has no cruise in which to recover a late pushback. The shortest routes carry the least margin and the tightest turns in US aviation.
4 min read · 959 words
How to book a trip that survives one bad flight
Every piece of advice on this site moves the odds a few points. This is how to arrange a trip so that the flight which goes wrong anyway does not cost you the trip.
5 min read · 1,032 words
The best time of day to fly, according to 7.04 million flight records
Departures between 5am–8am arrive on time 89.3% of the time. The 5pm–8pm window manages 68.2%. Here is why the gap exists and how to use it.
4 min read · 945 words
Late aircraft, weather, or the airline: what actually causes US flight delays
Carriers charge 40% of delay minutes to a late-arriving aircraft and only 6% to extreme weather. The categories are not what most travellers assume.
5 min read · 1,025 words
The 15-minute rule: what "on time" officially means
A flight that arrives 14 minutes late is on time. One that arrives 16 minutes late is not. The threshold shapes every on-time statistic published in the US, including the ones on this site.
4 min read · 963 words
How airline on-time statistics work (and how airlines game them)
Every US carrier above a revenue threshold files flight-level data with the DOT each month. The filing is honest. The incentives around it are not always.
4 min read · 974 words
Worst months to fly in the US, ranked by the data
July is the weakest month in the current data at 71.1% on time. September is the strongest at 83.4%. The 12.3-point gap is the largest calendar effect in US flying.
4 min read · 975 words
Do budget airlines really get delayed more? The numbers
The gap between the best and worst major US carrier in the current data is real, but it is smaller than the reputation gap, and most of it comes from one structural choice rather than from carelessness.
4 min read · 947 words
Connecting flights: how much buffer the data says you need
Nationally, 8.1% of arrivals land an hour or more late. That single figure, not the airline's minimum connection time, is what should set your layover.
4 min read · 908 words
Why the first flight of the morning is statistically safest
40% of US delay minutes are charged to an aircraft that was already late when it arrived. The first departure of the day is the only one with no inbound flight to inherit from.
4 min read · 922 words
Hub airports vs. secondary airports: delay trade-offs
Salt Lake City (SLC) runs 84.8% on time. Chicago (ORD) runs 71.0%. Choosing the second airport in a city is often the largest single reliability decision available.
4 min read · 954 words
What a cancellation rate actually tells you (and what it hides)
A cancelled flight appears in no on-time statistic anywhere. Nationally 1.81% of scheduled US flights never operate, and reading that number alongside punctuality changes several conclusions.
4 min read · 976 words
How to read this site: grades, sample sizes, and honest caveats
Every page carries a letter grade, a flight count and a data period. Here is what each one means, where the numbers stop being trustworthy, and what this site deliberately does not do.
5 min read · 996 words
Your rights on a delayed US flight, and what DOT rules do not require
The United States has no equivalent of the EU's cash compensation scheme for delays. Knowing precisely what airlines must do, and what they merely may do, is worth more than any claim service.
5 min read · 1,065 words