Wednesday, August 28, 2013

Trading, expectancy and R

Yesterday I had the pleasure of meeting a trader new to trend following for a 1-2-1 training session. He has traded for a while in the forex arena doing some very short term trading. To make money, the method he was using was reliant on a high win ratio - the stops were placed five terms further away than the profit target on each trade. 

When using such a system, to simply achieve a break even performance (and before any costs are factored in), your win rate has to be in the region of 85%. If you use such a method, it is quite possible you can have a run of small winning trades for most of the week, only to suffer one or two bigger losses at the end, which will eliminate all those previous profits and all your hard work over the preceding days.

Trend followers approach the question of expectancy and achieving an overall profit from the opposite direction.

Typically, a trend following system achieves a win percentage around 40%. Anything over 50% is unusual. But, the size of the average win is significantly larger than the size of the average loss. As a result, even if you lose on more than half your trades, you can still make money overall.

Expectancy is calculated by multiplying the win percentage by the size of the average win, and then deducting from that the sum of the loss percentage multiplied by the average loss.

As an example, my performance statistics tonight show that my win ratio stands at just over 42%, yet my profitable trades are 3.76 times larger than my average loss. When calculated this shows the positive expectancy of the system, which has been achieved in periods of high and low volatility, uptrends, downtrends and periods where no trends have been apparent. In short, it is a robust system.

The basic expectancy calculation is normally done using a fixed monetary amount of risk per trade, however I change the calculation so that it is based upon R.

R is the amount of risk I am prepared to put up on every trade, and is calculated as a percentage of my overall cash equity.

So, if I happened to risk £100 on a trade, and I made £300, then the profit equates to 3R. If I had lost £50 on the same trade, the this equates to a loss of 0.5R.

By using what is called the fixed fractional method of position sizing, then the monetary value of the risk on each trade will increase or decrease depending on my trading results. In my own case, I risk 2% of my equity on each trade, and this is recalculated after every trade I close. If I am making money, then the monetary value of the risk per trade will increase - if I am losing, then it will reduce. But, it will remain fixed when expressed as a percentage of my equity.

By using this method of position sizing, combined with the positive expectancy generated by the system, then I can use the power of compounding to increase my equity over time.

One thing which has helped some traders I work with to improve, is to ignore the monetary aspect of their trades. They now think in terms of R. Some have even removed the profit/loss column from view on their trading platform! 

While some may see that as extreme, they are working on the theory that, providing they follow the system rules, which have a proven positive expectancy, the profits will take care of themselves. But, to help them achieve this, they are focusing on the process (the system rules), not the outcome (profit). Those who have read the story of the Turtle traders will know that they went through the same process as part of their training from Richard Dennis and William Eckhardt.

No comments:

Post a Comment