About DelayGuard
Built by a pilot who's seen every delay
Knowing your delay risk before you book is more valuable than finding out after you've boarded.
DelayGuard (www.delayguard.ai) is a flight delay prediction website for US domestic flights. It is not the DelayGuard add-in for Microsoft Outlook, which is an unrelated product.

Matt van Arnhem
Pilot and founder of DelayGuard
Based in Amsterdam
As seen in: Yahoo Travel (October 2026)
Matt was quoted in Yahoo Travel's guide to flight cancellations on the early warning signs of a cancelled flight.
As seen in: Travel + Leisure (November 2026)
Matt was quoted in Travel + Leisure's piece on why some airport codes end in X.
As seen in: The Boutique Adventurer (November 2026)
The Boutique Adventurer used DelayGuard's analysis of US government on-time data for its Thanksgiving airport delay piece.
I fly commercially. I've sat in the flight deck through ground stops, ATC reroutes, and cascading network delays across hundreds of sectors. I know exactly what causes a late departure — and more importantly, I know when it's avoidable.
DelayGuard started as a frustration I couldn't fix from the cockpit. Passengers would board already stressed about connections, often on flights I could have told them — hours earlier — had a high chance of delay. No consumer tool existed to give them that information before they booked. FlightAware tracks delays after the fact. Flighty works day-of on iOS. Nothing gave a pre-booking probability for any flight, on any date, on the web.
So I built one. I combined real BTS on-time performance data with the operational patterns I understand from flying — how hub congestion compounds, how weather at one airport cascades to another three time zones away, how a 6am departure behaves completely differently to a 6pm one on the same route. That expert layer is baked into the model's pre-departure signals.
The result is an ML model trained on 2023 to mid-2025 BTS data: nearly 4 million real US domestic flights, using only what's known before departure, and tested on later flights it never saw — built by someone who's lived the delays, not just modelled them.
"DelayGuard was built on a simple belief: knowing your delay risk before you book is more valuable than finding out after you've boarded."
1.33M
Unseen test flights
Oct–Dec 2025, never seen in training
3.95M
Training flights
BTS TranStats 2023–mid-2025
149
Input features
Weather, ops, history, live
80
Airports tracked
US domestic hubs
Built on official data
Every prediction draws from verified, authoritative sources — cross-checked against the operational realities I see from the flight deck. No scraped feeds. No guesswork.
US DOT Bureau of Transportation Statistics on-time performance records (2019–2025) — the authoritative source for every delay rate and airline ranking on this site.
Live METARs, TAFs, and SIGMETs for origin and destination airports — the same aviation weather feed used by dispatchers and flight crews.
Real-time ground delay programs, ground stops, and National Airspace System advisories — the earliest leading indicator of systemic airport delays.
Primary live source for inbound aircraft status, departure/arrival delay minutes, and cancelled/diverted flags — used to detect cascade risk and surface ground-truth status changes.
Delay-cause breakdown (carrier, weather, NAS, late-aircraft) and route schedule lookup — used to attribute upstream delay reasons.
Get in touch
I read every message. Whether it's a bug report, a feature idea, or just feedback on a prediction — reach out directly.
Try a prediction
Check your flight's delay risk
How the model works
Full methodology and data pipeline
Go Premium
Unlimited predictions from $3.25/mo. The alternatives list and a trip PDF are included; 1 credit unlocks either one.