Hello Beno,
You may want to look over FX Sniper's work some more, as his posts have some of the most solid statistic-based data with explaination.
One of the points he did note was a specific stoploss, which I think would be consistent with this style. If not A, then B. The other factors mentioned in this thread would be taken into account.
Edit: This would keep your MM ratio's at more specific numbers and expectant targets.
Also your sample of 4 will have different results compared to a sample of 5, of 6, and so on. Data that doesn't seem to work for a set of 4 may work well on a set of 6, with same calculations.
I too would like to see your spreadsheet, dejnik, if you wouldn't mind sharing
