I did an interview with Dave Mabe of Stocktickr a couple of weeks ago. Dave was kind enough to provide me with a tour of StockTickr. He’s done a great job and I think his trading journal service could be of use to a lot of traders.
The functionality that caught my eye was the ability to examine the overall stop placement strategy: are they too loose or too tight? And then I saw that they were in “R”.
R and “expectancy” go hand in hand. It is the brain child of Van Tharp. If you google this word, you see right away that there isn’t much written about it, mainly because in the world of probability and statistics, it is simply known as the “expected value”. [DOWNLOAD PDF from the Chance Team]
[kml_flashembed publishmethod=”static” fversion=”8.0.0″ movie=”https://www.impossibleportfolio.com/wp-content/downloads/Chapter6.swf” width=”600″ height=”777″ targetclass=”flashmovie”]
In the book Chance, Aczel devotes 4 out of 160 pages to the expected value:
Let’s look at an example to show you what this means. Suppose someone offers you an investment that has a thirty percent chance of earning you one thousand dollars, a twenty percent chance of earning you two thousand dolloars, and a fifty percent chance of losing you four hundred dollars. How much is this investment worth? Asked in a different way, what do you expect to make on this investment? The answer is: 0.3 x 1,000 + 0.2 x 2,000 + 0.5 x (-400) = $500. This is not to mean that, on a one-time basis, you should do it — there is a fifty percent chance of losing four hundred dollars here. But long term, if you had the opportunity of making such an investment everyday, it would be worthwhile, since, on average, you make five hundred dollars every time.
Since I was already digging up information on the Kase Dev-stop and Bollinger Bands, I decided to do the same for R. [DOWNLOAD PDF].
I completely agree that the trader ought to know the expected value of his trading strategy. Once again, the problem is in the execution. And it all goes back to where the stop ought to be; that is, “R” must account for the natural range for whatever market and time-frame the trader operates in. “R” should never be calculated by dividing how much the trader can afford to lose by the number of contracts or shares traded.