Thursday, May 27, 2010

Consistency is key

As with most endeavors, it makes no sense to keep going around in circles if it isn't working (ie, the definition of insanity). But for a trader, consistency is one of the most important accomplishments one can aspire to. We want to realize our "edge," and be able to repeat it over and over in the way that only a dynamic and genius human brain could. We are charged with the task of check-and-balance, of constantly reviewing our execution to see if it can be tweaked or improved. The ultimate goal is consistency. We need a performance metric that can be measured statistically over time, and one that we can use to gauge our market prowess. While discretion is important to a trader, I believe that for review and measurement purposes, we need to extrapolate our execution scorecard in a way that is as objective as possible. In fact, in cases where subjectivity cannot be avoided, I always err on the side of failure, and would rather be pleasantly surprised by reality.

This week marks the end of May, and the first month of a significant new trading strategy for me. I have been very focused on closing efficiency gaps and trying to listen to the tick of the markets. I could argue that this month is not "typical," but then what is typical?? I don't think it is wise to try to define typical or normal -- to put the markets in a one-size-fits-all box. I would rather hone my skills with a strategy that is dynamic and intuitive enough to give me an edge under changing and dramatic conditions. After all, we have not lived through an economy like this yet in our lifetimes, and we certainly cannot assume anything at this point.

A lot of people made bank in the run-up to the Internet bubble bursting circa 2000, but only the smart money managed to live through it (because the rest went bust). I am in this for the long haul so consistency is more important to me than even pips or dollars; I'm not saying they aren't important, just not giving them any focus. I believe that once a trader's strategy becomes second nature (ie, he has mastered his game), then perfect practice will result in a perfect execution. Speculation is an art form, and should be held with such regard. Of course mastery of this art leads to growth, in more ways than one.

So the metric I've chosen to follow, the benchmark by which I will rate my system's success, is called expectancy. In a nutshell, this metric simply gives you a reading of how effective and profitable a strategy is over time. Van Tharp talks extensively about expectancy in his books. This single number gives you a birdseye view of your entire collective strategy. It can help you determine which instruments are worthy to trade your rules with. In my case, I will use expectancy to do these things:
  • Identify which currency pairs I can trade with highest success probability before taking any trades;
  • Track ongoing system performance and make my own "circuit breakers" to give me advance warning of when a pair goes "out-of-spec" -- because I believe that most strategies do not repeat forever, at least without some degree of morphing. My goal is to be nimble enough stop trading before I can see it visually and obviously in my account equity.
In measuring expectancy from the lookback tests, I can see that some pairs perform exceptionally better than others. For example, I'm finding that the Pound/Dollar and Euro/Dollar have a far better expectancy than Dollar/Swissy. I don't exactly know why, but I'm not sure that I really care either. What matters most is that I have a benchmark to look at, and I know that I need to focus only on pairs that have the highest-probability success patterns.

I'm also currently debating whether it is a good idea to add a leverage weighting to the higher expectancy trades. For example, if I know that Pound/Dollar has twice the expectancy of say NewZealand/Dollar, perhaps I should double the gearing when I place the trade? I think for the upcoming 30 days I will not make this change, but instead will observe what kind of impact that could have. In the back of my mind, I'm trying to justify why NOT to add expectancy weights. In laymens terms, I'm basically arguing: if I know a trade is twice as likely to be successful, then why shouldn't I take twice as much risk, for the opportunity to make twice the reward? It would be an objective way to exponentially grow the account based on a consistent strategy.

Speaking of gearing and leverage, I've started using this different calculator due to some weirdness in lots calculation I noticed for cross currencies like Euro/Swissy. I am now using this calculator which gives me the exact micro lot figure I need to enter into my broker platform. I am still planning to automate this, but the math has me befuddled and as such I have not prioritized it. Once I'm able to consistently work these formulas on paper, I can pretty easily program the computer to calculate them in realtime. There are 3 circumstances requiring 3 different formulas, as I understand:
  • direct rate
  • indirect rate
  • cross rate
Below is an example of how I use this new online calculator to get an exact micro lot calculation prior to entering new trades: