We're not done analyzing this data; not even close. But, in doing our analyses we've done some really useful things:
groupby()function were powerful allies in understanding our data.
ggplotpackage to help us visualize the cumulative distribution of connection events.
time deltas: a way of seeing how much time elapsed between successive events. With their help we got a better sense of where the most active parts of our dataset were.
What remains, for starters, is to figure out exactly what's wrong. But look at how far we've come! Using open source tools, expressive scripting, and powerful graphics we've unearthed some really important things about our data. Ultimately, I hope this book has given you a taste of what's possible with a game data collection framework and scripted, reproducible analyses.
Python on, my friends.