If you are interested in using historical data dating back to 2013 for data analysis or betting prediction algorithms you can email us at firstname.lastname@example.org for more information/questions.
An explanation of the historical data follows:
We are constantly gathering datapoints every 10 minutes.In order to keep the database small, the odds are kept only if
changed from the last parse. This is a compressed time-series schema.
We include about 10-15 sportsbooks at a time depending on availability and we gather statistics for 37 leagues worldwide.
The current database is available in MySQL and CSV format and currently contains 11k matches and more than 1m datapoints (that are actually more since we are using a compressed time-series model).
The cost of the database snapshot is 499 EURdownload demo
Note that 100% match coverage is not guaranteed for historical data and in some cases matches,results or odds from specific books may be missing but this is benign in big data analysis.