An examination of the head and shoulders technical pattern; A support of the technical analysis’s subjective nature

It is being argued that Head and Shoulders technical price pattern although it occurs seldom it is worth to trust. In this Paper we use the Brownian motion to generate time series of different characteristics (drift, volatility) and we apply on them a matlab script for the identification of the head and shoulders pattern. After ensuring that the aforementioned technical pattern is being identified in random generated time series, we look for the profitability of the pattern through three different trading strategies. In a mechanism, like Brownian motion, of producing time series in a “random” way we should not expect to be able to forecast the further evolution of the price. On the beginning our results show that we come up to a fair game, which means that we cannot profit from the strategies used as a whole. Examining each strategy separately, looking at the way the variables, drift and volatility rate, affect the returns, and after taking into account general rules and principles of TA, the question we bring into discussion is whether investors selectively focus on specific profitable cases of the Head and Shoulders pattern. This results in the misleading conclusion that the pattern is seldom observed but profitable. Is it possible that the pattern is being identified more often having no predictive power at all?