Saturday, February 15, 2014

Don't bring the risk of ruin into play

Based on the current performance statistics here, the average win is +1.99R on each position, and the average loss equates to -0.39R per trade. This therefore gives a profit factor of 5.1 (1.99/0.39), so therefore the size of my average win is 5.1 times bigger than the average loss (all expressed in R).



Due to the particular stop methodology used, it is rare that losses occur of 1R or more. When they do happen, they are a result of a gap through a stop level, normally after a trading update or earnings release.

The initial stop distance and consequently the position size is based on a volatility measurement. What would happen if this metric was reduced and a tighter initial stop used?

In theory, if the volatility measurement was cut in half, while risking the same percentage of equity per trade, the the position size would be doubled. This would result in the size of the average loss increasing to -0.78R, and the size of the average win would also double to +3.98R.

Now, based on the current stats lets see what happens as a result to the system expectancy:

(37.13% x 3.99R) - (62.87% x 0.78R) = 1.48 - .49 = +0.99R per trade - i.e. double what it is now.

This would indicate that, should the tighter initial stops be used, the the returns could be doubled! Sounds great, but there are some major pitfalls here.

In the sample of trades taken (just over 270), the biggest loss has been -1.98R. Had the tighter initial stop distance been used, with the position size adjusted accordingly, then the biggest loss would have correspondingly doubled to almost 4R. The question is, would you feel comfortable with that?

Say you had suffered a 5R loss on a trade, when a particularly poor earnings report came out. Had THAT loss been doubled, then you've suffered a 10R hit on your account overnight. If you are risking say 2% of equity per trade, then that's a fifth of your equity gone in one go.

Big losses can also happen if you adopt an aggressive pyramiding strategy on a position going in your favour. You should ensure that you have an overall portfolio risk element incorporated into your strategy.

Typically when using a trend following strategy, the overall win rate can be in the range of 30% - 40%. So you can expect a run of losing trades from time to time. As a result, the resulting drawdown will be faster and deeper by using a tighter stop.

In The Way of The Turtle, Curtis Faith talked about the Whipsaw strategy that was used by some of the Turtle traders - in essence, they used much tighter initial stops than their standard systems, but they also reduced the percentage risk per trade by a corresponding amount, which equalized the overall risk out.

In our example here, if the tighter stops were going to be used, then the risk per trade should also be reduced by half. In monetary terms then the number of shares traded, or pounds per point on a position if using spreadbets or CFD's, would be the same as before.

Trading is a compromise between seeking to maximise your profits, while keeping losses or drawdowns to a minimum. If you want to attain the rewards, you have to be prepared to accept the potential losses that go with it. But what you don't want to do is to push the limits so far so that while yes, you can potentially earn bigger returns, you don't want to bring a major drawdown or even the risk of ruin into play. Because of that, I will be sticking to my existing volatility measurements, position sizing, and overall position limits.

No comments:

Post a comment