7 comments

[ 4.9 ms ] story [ 24.9 ms ] thread
If I'm looking at on-time stats to see if they should have a bearing on which airline I fly, I care a whole lot about the definition of "on-time." I don't care about being 15-30 minutes late, but I do care about not being 2 hours late.

Again, this sort of thing should be adjusted for routes and airports, but depending on your purpose in flying a route (going home, where your parents will pick you up? flying to LAX to catch an intercontinental flight with a tight connection?), a more granular breakdown is important.

Yeah, it would be cool if they could calculate the "expected minutes late" for every carrier, broken down by route.
You would really need to see the distribution as you can plan for a consistently average delay, but unpredictable arrival can be a real pest.
I suspect this could be another service that FC offers–choose your desired route before you buy your ticket and find out who has the best on-time %

Partner with Orbitz/Expedia/Whoever so I can see it while I shop ... it would definitely influence me to use that particular travel site.

Kayak has this today, though I'm not sure where the data is sourced from. Flights get a little '%' icon if they have less than 75% on-time.
I wonder how much analyses like these help FlightCaster get good PR and better search rankings. It seems that (1) these analyses would be of interest to publications targeting potential FlightCaster users (travel-related, etc.) and (2) it would be hard for a journalist to reference such an analysis without providing some background on FlightCaster's services.
FlightCaster (and SeatGeek) are a new breed of startup that uses statistical modeling to empower consumers in an industry that they usually have little to no say in.