The performance layer for air travel.
Wingmark Analytics is an aviation data company. We study how flights, connections and airports perform in practice, using the records published after each flight has operated, and we turn that history into insight you can act on.
What we do
A timetable tells you what an airline intends to do. It can’t tell you how often a particular flight leaves late, how often a tight connection is missed, or what happens to a passenger when it is. Those answers are in the operating record, and that is where our work starts.
The data
Scheduled flights in our model
US DOT on-time data, August 2025 to July 2026
Possible connections back-tested against what actually happened
GateCast back-test, checked 3 October 2026
US airports covered
Domestic network
How we work
Every figure we publish comes from a named dataset and time period, and if we can’t source a number, we leave it out. We test our grades against what actually happened on flights the model hadn’t seen, so we know what each grade means in practice.
Our product: GateCast
GateCast grades the flight connections it covers, A to F
Possible connections graded A would have been missed about 7% of the time on our estimates; F, about 28%.
Source: GateCast's check of its grades against past US flights, on weeks the grades were not built from. About 1.25 million possible connections. Checked 3 October 2026. “Missed” means real delays against an assumed transfer time; BTS has no passenger records.