In the last few years, I was taking trains pretty regularly. I’m talking about a-few-times-a-week kind of regularly. I started hearing myself complain about delays all the time, maybe as often as I complained about my PhD. Okay.. maybe not quite that much.

Was my dissatisfaction justified, or was my perception just distorted by a few incidents?

To find out, for 9 months I recoreded some data everytime I took the train. I noted the scheduled departure time, the actual departure time, the scheduled arrival and the actual arrival time. I gathered about 120 data points. Here is what I learnt from this little experiment.

The plot above shows the difference between when the train was meant to depart/arrive and when it actually departed/arrived. It seems that most trains departed on time, and rarely with more than 10 minutes delay. Not too bad, actually. With arrival times there is a wider distribution, as you would expect. Often trains arrive early and only about 10% of my journeys were delayed by more than 10 min. I arrived to my destination really late only a couple of times.

How do morning journeys compare to evening journeys?

From this data, it doesn’t look like there is a considerable difference between the two. There are a couple of late outliers, but since there are only 4 journeys that arrived more than 20 min late, I can’t really comment on the morning vs evening difference.

All in all, one conclusion is expected: I need more data. The other conclusion is, more surprisingly, that I probably shouldn’t have complained that much!