The Edge In Chart Patterns, Part 2 by Giorgos E. Siligardos, PhD
Part 1 of this article showed how you can use a trading system to define the edge that a chart pattern gives you. This concluding part will elaborate on three practical issues relating to this subject: market frictions, sample populations, and capital restrictions.
Last month in part 1, I used the term profit factor (PF) to define the reward/risk part of a trade based on the pattern and then showed that the arithmetic average (mean) of many historical PFs under various market conditions provides what I called the historical edge of the pattern. Inferential statistics can then be used to provide a confidence range for the edge. I also provided an example of how you can determine an edge in the cup pattern. In that example, I performed an identification algorithm on daily charts of S&P 500 stocks from 1982 to 2014 to show a total of 3,991 cup formations. I backtested a specific trading system to show if “betting” on the cup pattern’s ability to produce uptrends gives the trader an edge. The results showed that betting on that pattern’s ability gave a historical 16.6% edge with an implied 99% confidence range of 9.1% to 24% as an estimate for the overall edge of the pattern. In this second part of the article, I will elaborate on three practical issues of the subject ...