Saturday, April 27, 2013

Performance metrics

One way of comparing trading systems is to look at various performance metrics. I keep a track of a number of these, to ensure that nothing untoward crops up in my trading performance, which would warrant further investigation. Some, such as overall expectancy, profit factor, win ratios etc., have been discussed on here before (refer to this post).

Another such metric is the Calmar ratio. Although there are some variations to it, in its simplest form this ratio compares the relationship of annual returns to the maximum drawdown, over the period of time (normally 36 months). For example, if you had achieved a 50% annual return, while suffering a 25% maximum drawdown, the Calmar ratio will be 2.0. It is generally accepted that a ratio of 2-3 is very good. So, this single number tells you a lot, and is useful for comparing different trading techniques.

The Calmar ratio is very similar to the more common MAR ratio - the difference is that the Calmar ratio looks back at the performance achieved solely over the last 36 months, whereas the MAR ratio looks at the performance from inception.

One way to have a high Calmar ratio is to cut your losses short (minimising your drawdowns) and to let your profits run (so that in terms of multiples of what you risk per trade, they potentially can be many times bigger than the losses incurred).

Any losing positions I generally cut within 24 hours of being initiated. I want my positions to be in profit from day one, otherwise they are not performing as anticipated. This also minimises the risk to my capital. After day one, my stops will take me out of a position - the sole major risk to my capital from that point is from an overnight 'gap' through a pre-set stop level. Due to my risk parameters and position sizing, the majority of my losses now are in the region of 0.2R to 0.5R. I don't suffer a 'slow death' trade where price drifts against me, ultimately triggering a 1R loss.

It is also important to note that it is a lot easier for an individual trader to enter and exit relatively small positions in a stock or instrument, compared to a large hedge fund who would be trading substantially bigger positions, which brings its own problems and may skew the results to a degree. But the principle of achieving the maximum your returns, while keeping your drawdowns as small as possible (and keeping a track of it) is applicable to all market participants.

I have calculated the Calmar ratio on the trades I have called via Twitter since July 2012. Granted, this is for a period of less than one year, let alone three, but it is a starting point, and I will be tracking this myself going forward. I have calculated the ratio based on my actual performance achieved in this period (which is shown on the blog sidebar) and have not annualised it.

Given that in the same period the biggest drawdown suffered has been less than 5% (and which can be seen in the equity curve chart in the blog post linked to above), you can work out what my Calmar ratio is. Bearing in mind what is perceived to be a good ratio, I'll take it... :)

No comments:

Post a comment