I'm kind of cringing right now because I'm not sure I should give away some of my ideas simply because when people start to jump on the same idea, then the whole supply/demand balance might shift. But, in any case, when I spent hundreds of hours looking at hundreds of charts, I noticed retracements are often smaller than the moves in the direction of the trend. However, when you're talking about the timing of when to determine a trend, I find that's tricky. I know of at least 3 ways of determining a trend. One is moving averages, the ADX (DMI) and then what Martin Pring calls your peak and trough analysis. Looking at the relative lows and highs is your fastest way of determing a trend. But faster trading signals usually give you more false signals. Victor Niederhoffer wrote a book (somewhat difficult to read) called Education of a Speculator where he said one of the ways you can defeat an opponent is to vary your patterns. I've been watching the EUR/JPY for months now and noticed patterns can change from one week to the next. If you can develop a system that can at least survive difficult periods and make money in optimum periods, I think that would be the realistic way to be profitable.