Ask Colin

When is a divergence completed?

This is an excellent question. It has not been made clear in my articles. Nor have I been able to find an answer elsewhere after searching some textbooks. Most commentators in textbooks show you examples in the middle of the chart when the divergence has already been clearly completed. However, an analyst or trader is required to act on the right hand side of the chart. As you rightly ask, when is an apparent divergence actually complete?


Since we are dealing with trends, it is tempting to simply fall back on the definition of trends. In an uptrend, a peak can not be definitely called until the price either falls below the previous trough (the end of a trend) or has made a subsequent trough. That trough in turn is only definitely in place once the price has risen above the peak we are concerned with. This can be reversed for a downtrend. The problem with this is that it is all too late to be of practical use except by a pedant teaching students.


The practical answer is that some judgement is needed. If we are looking for a divergence in an uptrend, we may be inclined to act on a potential divergence. So, if we have a series of higher peaks in an uptrend and in an oscillator and then we notice that the oscillator is nowhere near its last peak as the price turns down from a peak, we may conclude that a divergence is likely to be formed. Of course, we may be wrong sometimes in hindsight, but at least we will have protected our capital if we turn out in hindsight to be correct in our assumption.


You asked specifically about Stochastic divergences. In my experience, unless there is a Lane’s Classic Divergence, which involves three peaks or troughs, most divergences could have been correctly diagnosed as soon as the indicator turns at the second peak or trough. If we acted on such a signal and then Lane’s Classic Divergence formed, we would have only been premature on acting when an even more powerful signal unfolded. Some may argue that we had acted too soon, in hindsight. However, this is heard from people who are still trying to pick the exact bottom or top of the market. I would ague that, especially at tops, it is better to be early on the odd Lane’s Classic Divergence, than late on the others, all of which would cost us money. Lane’s Classic Divergence is explained in the Stochastic article in the Technical Analysis Articles section of my members web site as well as in some textbooks.