Belarus Is Bust
Belarus Is Bust
If you read my previous article, you will be fully aware that my latest focus has been attached to events in Germany and the restart of the Bundesliga, a restart that has fuelled by bid for data driven betting excellence.
While at the same time, I’ve also been trying my luck in Belarus and attempting some data driven football predictions, something that admittedly has not gone well – if we’re honest it has not gone well at all.
NO PRIOR KNOWLEDGE
The basis of this project was crafted from two main standpoints:
1) There was no other league to work with
2) I wanted to purely test the models that I am using
With this mind, it really was a case of working in the dark and seeing whether or not ‘BelaBot’ could shine some light and in turn provide me with some much needed financial relief. A plan that has been rather fruitless.
Last week, I once again served up another raft of predictions across two model variants and it is time to see how they fared. First up, lets go with the standard six match form guide:
Now with their being seven matches last weekend, there was less overall to win in terms of a big all in bet. However, there was also less matches to get wrong but even with that being the case, it was still an absolute disaster.
‘BelaBot’ reutrned just one correct data driven football prediction, nothing short of useless and even with the innovation of home weighting, it was only BATE Borisov (again) who managed to get me over some form of line.
Which means now we’ve reviewed that one, lets see how its younger, slimmer brother fared:
Better, but still not good enough. Admittedly there was the perfect start as Torpedo Zhodino got off the mark with a home win but from there on in, it was once again BATE Borisov who eased the strain and this meant just two out of seven were correct.
Which on this evidence suggests that Belarus is bust and therefore I will focus my full attention (for the forseeable) on the Bundesliga, because it seems as if having no insight in a league whatsoever, is actually a massive hindrance.
The most galling thing about this, is that in the four weeks of operation, the main model in play was worse than gut-instinct. Even with just bliind picks in the very first week, I did better than any data driven football predictions.
WHAT WAS THE POINT?
That is something that I’m now asking myself, admittedly it was not a great amount of time lost or money for that matter, but it does show that this particular project was incredibly wide of the mark and lessons will need to be learned.
With that said, there are at least the frameworks for some sub-models going forward and the introduction of additional home weight and less form, could come in useful for some of Europe’s major leagues, therefore all is not necessarily not.
But as they say, sometimes you have to know when to hold them and also when to fold them and with that in mind, I never want to watch another second of Belarusian Premier League action – that is me absolutely done. Project stopped.
Happy punting and thanks for reading. Dan
(THESE ARE NOT TIPS PER SE, THESE ARE JUST DATA DRIVEN FOOTBALL PREDICTIONS IN A PURE TEST AND LEARN MODELLING CAPACITY)
If this has grabbed your interest and you would like to discuss/feedback then please feel free to drop me a message at firstname.lastname@example.org. While I am always looking for new football/data projects to work on and if you feel that my skills would be of use, I can be contacted at the same address.