DislikedI will do my own visual backtest to see what I see.
However, I am curious what system you suggest as an alternative.
I value your input as to why this will NOT work, but I would like to hear your input on what does work.
PGPBIgnored
I don't have any idea why it doesn't work, but unless the other pairs you trade perform much better than E/U does, I don't believe it works.
Thirty trades is statistically meaningful and the results are exactly what could be expected with a random coin toss entry- an average loss of 2 pips(the spread) per trade.
I wouldn't bother trying to tweak it to fix it. The entry doesn't provide a statistical edge. You would do as well trying to tweak a coin toss entry.
I have seen only two mechanical systems that work. One I developed several years ago, that stopped working when the EUR and the CHF lost correlation.
The other one is this one, which I had nothing to do with:
http://www.forexfactory.com/showthread.php?t=296776