That really doesn’t look like a very good indicator. By my count there are 7 probable losers, 5 probable winners and 2 break evens. The winners do look to be bigger than the losers so it would be profitable overall. But the trend on the chart was strong downward and most of the winners were shorts. My guess is that in a ranging situation, this indicator would fail badly like most trend indicators.
I agree with the above statement. I actually coded an EA with this system and it failed in backtests although it is a good indicator for finding entries in trending markets.