Friday, June 4, 2010

A magnitude of improvement

I am very excited about the direction my trading has taken in this first week. I've made some simple optimizations to my strategy that help take each currency pair's personality into account. In the month of May, I was putting everything into one box which seemed logical to me at the time. But I soon realized that the currency pairs behave very differently, and they tend to each have their own attitude. It would be better if I could treat the currencies in a more dynamic way, much in the same way one might parent children ;)

For example, I am currently re-testing each pair and incorporating an additional metric that I had not tracked before: the reward ratio of the setup. Reward ratio (aka risk/reward ratio) is expressed as a comparison of a single trade's probability of risk to reward. So if your maximum loss potential on a trade is 50 pips, and your expected target is 150 pips, that would be a 1:3 risk:reward ratio. Risk is always expressed as a 1, so I just call it the "reward ratio" and say that it is 3. This means that for every $1 I commit to risk on this trade, there is a probability I will reap $3 in return (net 200% return). This method of trading allows me to constrain risk to a specific threshold of pain (such as a percentage of account equity) and also target specific profit levels.

If you take the average reward ratio for an entire collection of trades (such as the entire lookback testing period), you get a different figure called expectancy. This figure represents the average expected return on your system for the given pair. In my weekend analysis, I realized that it doesn't make sense to trade pairs that do not have exceptional expectancy. I would rather trade less frequently but with more accuracy. That would be working smarter, not harder.

I'm still performing lookback tests and adding quality pairs to my toolbox. I've decided to set a minimum standard of 3.0 expectancy. Anything less does not make the grade. Here is my list so far:
  • Great British Pound / US Dollar: expectancy 4.5 trading with 6:1 profit targets -- WOW!
  • Euro / US Dollar: expectancy 4.5 trading with 4:1 profit targets -- awesome!
  • New Zealand Dollar / US Dollar: expectancy 3.0 trading with 2:1 profit targets
  • US Dollar / Japanese Yen: expectancy 3.0 trading with 2:1 profit targets
  • Australian Dollar / US Dollar: expectancy 5.0 trading with 2:1 profit targets -- woohoo!
So in laymen's terms, I would describe it like this: The higher the expectancy, the more likely the pattern is to work and the trade to become a winner. The higher the profit target, the more moolah generated with each trade taken. So in my opinion, the best performer in the group so far would be GBPUSD, but truly all of these are superstars!! There were plenty of currency pairs I tested that failed to make the 2.0 expectancy grade.

One of the biggest frustrations for a trader
One of the biggest frustrations for a calm, collected trader who is waiting patiently on the sidelines for the absolute juiciest and most succulent fruit to come along... that flyball that he hopes to hit right out of the ballpark -- is to miss the trade setup. YIKES!

This exact scenario happened to me this week and I'm bummed. I did not do anything wrong and all of my system procedures worked correctly. The computer informed me at 4am that a prime trade was developing, and that I should get involved to take a look. ARG! For whatever reason ... I did not hear the multiple alerts that were reaching my cellphone. I must have been in REM sleep mode or something :( By the time I awoke at 7am, it was too late. I entered my trade order but the pair never retraced to reach my optimal entry.

Typically on the 4hour timeframe when a pattern forms, it seems that there is only a small window of opportunity to execute. The computer does all the heavy-lifting and math calculations so it is not a big deal to wake up and key the trade entry (I don't have to do any thinking except to evaluate how sexy the pattern looks).

To give some weight to the frustration... this pattern formed on GBPUSD which, as discussed above, is one of the absolute best pairs in my toolbox right now. This single trade would have eclipsed the entire month of May due to optimizations made this past weekend. Trades taken on the Great British Pound are expected to catapult my account into the stratosphere!

Alas, I am not too disappointed. Part of being a good trader, I believe, is rolling with the punches. The lesson learned here for me is that I need to prioritize automating the trade-entry process. I think I will make that my major focus and hope to have a prototype ready next week. I would like to give the computer the freedom to identify and also execute on my trade orders, because I do not want to miss a single RIPE trade.