Predictive value of divergences?

[Olle]

Hi!

When looking at a chart with an oscillator attached to it, one doesn't have to wait a long time before a divergence between the price and the indicator reaveals itself. Also when looking at history data one can see that after a divergence, there is often a change in the trend, which makes many of us (i guess) believe that divergences is a powerful tool.

Is this just a random consequence of how the indicators are made? Or are there really any predictive values in divergences?

I was hoping for someone to shed some light on this topic.

Thanks!

[hughesfleming]

Good question. Many people will disagree with me but my opinion is that divergences have no predictive value. Someone might see a divergence on a chart that reinforces an opinion about the price action but it is not enough on its own to be a reliable signal. Generally divergences are an artifact of the normalization process and nothing more.

I did some tests a long time ago and the results were worse than a coin toss. You may notice a divergence and a change in trend but by then it is too late to trade on that pattern. Separating non random patterns from noise is very challenging.

[krelian99]

Divergences and trendlines in oscis always look nice, but in live trading you wanna play more safe. They are potential reversals, hit rate differs from osci to osci and markets maybe from 50-70%. There are to many false signals. WCCI-patterns are better, but not still good enough. It seems that forex beginner like divergences, but no old stager use it.

For trend friends divergences are useless, not to say cumbersome. Swing fans have more powerfull signs as harmonic and candlestick patterns. Scalper use break through or out impulses of trendlines and other lines as pivot or fibo levels or whatever they do.