[Henry17]

The problem with creating a system starts with the fact you simply don't even understand how to test a system.

By picking every game and not having any weighting system you are creating too much noise. A system would have to be exceptionally good to be detectable given your testing methods. With the amount of noise you are creating it is a real possibility that even if you had a working system you'd reject it.

If you look at my method I pick every game then reject the weaker selections. Picking every game my record over the last 4 days is 13-10 which is awful in hockey. Based on that data I would reject my method.

My games get graded from 1-6 with 1 being the weakest. If we reject just all the 1s now my method is at 11-5. If we look only at the 4 or higher my method is at 4-1.

So had I tested my method using your testing system I would have rejected it. By removing the noise created by the weaker selections I realized that I should only bet 4s or higher and have been making money doing it for years.