fivethirtyeight / data

Data and code behind the articles and graphics at FiveThirtyEight
https://data.fivethirtyeight.com/
Creative Commons Attribution 4.0 International
16.74k stars 10.94k forks source link

Individual match predictions based on SPI #216

Closed albertovth closed 5 years ago

albertovth commented 5 years ago

Check out this way of predicting individual matches: https://github.com/albertovth/postulated-reality/blob/master/StatFoot.py , based on your latest stats. It was just meant as a data science exercise, which started as an Excel-VBA workbook, and that I later moved to Python. I am well aware that I could achieve the same result, while saving several lines of code... I also know that your model takes into consideration several other factors, to make match forecasts in very particular conditions. The way that I have set up things, only renders forecasts for matches in average conditions. But I think it's fun to use this to make up own match-ups, also between national teams and clubs. I appreciate comments and feedback.

jayb commented 5 years ago

Cool!

SoccerCashUK commented 5 years ago

Hi maybe you could do the same for Club Soccer Predictions from FiveThirtyEight

James

shivekbittu commented 5 years ago

Why am I getting all these mails pls I w at to get rid of us.

On Mon, 11 Mar 2019 at 1:11 PM, SoccerCashUK notifications@github.com wrote:

Hi maybe you could do the same for Club Soccer Predictions from FiveThirtyEight

James

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/fivethirtyeight/data/issues/216#issuecomment-471632736, or mute the thread https://github.com/notifications/unsubscribe-auth/AlQufTeVzQrcQ5xpfEa__BYB0xXaa5j4ks5vVo5BgaJpZM4Zzt9n .