In depth details of the data are available here, but I will also quickly point out a couple of the main limitations. The data is binned by 15 minute intervals, if there are a very small number of taps for an interval (< 18) it will appear as 0 in the data and there were two datasets, each covering 7 consecutive days.
The dataset mapped covered 8 August to 14 August 2016, the week after the retirement of paper tickets.
The map I have created displays the difference between Tap Ons and Tap Offs. This was to convey the sense of where/when people are coming from and going to. The number of passengers displayed is the average for that 15 minute period over the 7 day period.
This displays the morning commutes clearly, lots of people getting on at suburban stations and then disembarking in the city being the most obvious pattern.
It also suggests significant flows of travellers tapping off in places like North Sydney, Macquarie Park/University and Parramatta, at the morning peak.
The commuter pattern seems less clear in the afternoon peak, with significantly more stations flickering between red and blue (ie. having more passengers getting on or off), but you still get the sense of people getting on in the City, North Sydney, Macquarie and Parramatta and off at suburban stations.
I am happy with this visualisation and feel it demonstrates a lot of information quickly and intuitively. I am hoping there will be further releases covering larger time frames, as I suspect this would result in a smoother more pleasing animation, as well as just giving a more robust impression of overall travel patterns.
Data from Transport for NSW used under Creative Commons Attribution 4.0 International (CC BY 4.0) Licence.