but it no longer exists.
I tried to figure out how it worked, it would look at the statistics per type, airline, country and I think also month of the year (not sure about that one). These would all affect the outcome independently.
So I played around and, if my memory serves me right, it said that if I flew from Quito to Lima in a Nigerian Airlines Tu134 in November (or something along those lines) my likelihood of crashing was about 1:32.000, whereas if I'd fly from Reykjavik to Akureyri in a Finnair A330 in June, my chances of dying were 1:250.000.000.
Not sure anymore about the parameters but quite sure these were the most extreme outcomes I managed to get. Generally, and certainly with realistic input data, it would seem very accurate though. If it wasn't sure (due to lack of data) it would hover around 1:1.000.000 in general.
|Quoting BAxMAN (Reply 5):|
It correctly worked out that none of the flights I was scheduled to take were destined to end up in the middle of the Atlantic or strewn in many bits across a field.
:D gotta love those ex-post analyses!!!
I scratch my head, therefore I am.