The idea for this site was born when I was learning basics of Data Science, with programming languages which were then new to me. I needed a set of data to explore and practice the skills. For a normal football loving guy who had been enjoying fantasy football for many years (in a casual way) - the data-set of fantasy football made perfect sense. The tedious learning process was suddenly a lot more fun when I was exploring data from English Premier league, and I could do it for long hours without getting bored or tired.
From a programmer's point of view, this project presented all sorts of challenges which could be identified with the challenges of Data Science, like:
- The data set was dynamic - it would change every game week
- The job required:
- gathering open data from directly from Web,
- cleaning and formatting,
- visualization and publishing, Which along with some statistical modeling techniques covered most of the things I was trying to learn in the process
- The data set was small enough to fit in my regular laptop (700 to 800 registered players, 20 teams, 38 game weeks in total) while it could be made really big to learn/practice some aspects of DS which required larger data sets (try Combination formula C(n,r) = n! / ( r! (n - r)! ) to find out all possible XIs of about 700 players )
If you have any comments, suggestions or helpful ideas then please post them in the comments sections of relevant posts. Alternatively, you can also contact me at email@example.com or via twitter at @fplAnalytics for general queries or issues.
Good luck for the season 2016/17