[nyc111]

> When waiting for a bus that comes on average every 10 minutes, your average waiting time will be 10 minutes.

This is very ambiguous. Unless he gives a time frame the numbers do not make sense. Average in a week? Average in a year? This is not how it works in real life.

And I cannot accept his premise. My experience tells me that, in New York, when I used to take a bus to work, sometimes the bus was coming as I was walking to the stop; sometimes I would wait a long time. Sometimes not very long. There was no observable bias.

[pieguy]

In statistics, "average" often means "expected value". No time frame is specified (although you could consider it an infinite time frame). With a small sample size your actual average might not be 10 minutes, but as your sample size grows, it will tend toward 10 minutes.

[twtw]

> it will tend toward 10 minutes

If you are talking about spherical-cow style poisson buses, yeah (that's what the author means by "reasonable assumptions). But as the author concludes, bus arrival times are not well modeled by a poisson process.