Machine learning is a branch of computer science in which we try to make computers learn from data. There are plenty of algorithms and parameters to choose from. For this blog I use a combination of 2 such algorithms, but I keep experimenting with more and also with better parameters. So that's why on the blog's description it reads that these suggestions come automatically.
As far as the stake is concerned, they are based on probabilities that come with the algorithms. However they are not automatic, like the suggestions. I put greater stakes to greater probabilites and vice versa.
There is also a threshold regarding odds, so you will not find any odds lower than 1.70 on this blog. Usually (as you may have noticed) I try to keep suggestions with odds greater than or around 2 (the average odd of this blog is around 2.25). That means that there is a greater margin for error, since a winning percent of around 50% is enough to make profit (if you put the same stake at each match of course). That's beacause the losses of one match can be overcome by the profits of another. But by having a good staking strategy, profits can be significantly greater.
In order to keep these suggestions free and still have an income, I am only counting on your signups or donations. So if you feel like bying a book, or betting online, please do it from my referrals.
If you have any further questions or suggestions (especially for the staking part) feel free to contact me by email or by posting a comment.