At first glance automated trading — as its very name implies — supposes absence of any psychological bias. Indeed, if a trading strategy is automated, how can human affect its performance?
However a human is always a human, and it has a number of features which help him in everyday life, and greatly disturb when it comes to unusual activities like trading or investment. One of these features is that human tends to be in doubt about anything he does. You know, it’s like you bought a new Ferrari and then entertain the idea of what if you went to the Caribbean instead. Or you bought 1000 gold contracts in July 2008 and then, in December the same year you ponder upon whether a Madoff investment would be wiser. Human always discusses things like that internally, this discussion lasts forever till death. The result is that sometimes human is prompted to conclude that he definitely did something wrong and, what is worse, the time to fix it is right now.
Trading is that special because it is always obvious when to buy and when to sell, only some time later. Thus the temptation to follow one’s intuition is sometimes so great that it becomes unbearable.
Unfortunately automated trading needs human supervision. We all live in physical world and so far all automatics we create require some degree of control. Automated trading boxes are no exception, as, for example, server failures require human intervention. And as long as there’s the smallest chance to interfere with the system logic, be sure that it will be done.
To come from abstract speculations on psychology to more solid grounds let’s consider a real story which illustrates how severely can psychology affect the performance of automated trading.
To respect all parties I will keep all names confident. Moreover, I would even prefer that you thought these are just imaginary characters.
So, the story starts in December 2010, when two investors were offered to allocate some money to the spot forex programme. I presented them with exactly the same information and was open to any discussion. In a short while the first investor — let’s call him Mr. Nonchalant, as he has always been more or less prepared to any unpleasant moments which inevitably appear in trading — allocates some amount and the work starts at the very beginning of 2011. The other, Mr. Smart, as he is always thinking that he can outsmart a trading strategy in particular and the market itself in general, is still considering whether he likes the performance data provided and looks for any additional proof of the future performance.
By end of January the live equity dips into a drawdown. I am having an unpleasant discussion with Mr. Nonchalant, but the decision is to continue as in fact neither of the key metrics were outside of the predefined tolerance corridor, and risk limits were not exceeded. Mr. Smart is proud of himself as he provisionally avoided unnecessary risk.
About early February the equity starts to grow, and by end of February it looks like this growth is consistent. Mr. Nonchalant is very glad, anticipating full recovery in the nearest future, and Mr. Smart likes to think that this growth is just temporary, and soon the equity will turn south again.
Around mid-March it’s obvious that the drawdown is fully recovered, and the strategy starts to bring gains. Mr. Nonchalant completely forgets about the account, and Mr. Smart decides to give the strategy a try, with a small amount.
Some time passes on with a nice and steady growth, and around late April Mr. Smart decides to increase the trading size as he is now confident in the strategy and wants to earn more. Almost immediately after that the equity curve first stops rising and then turns into a local drawdown. Mr. Smart waits for some time watching his profits first diminishing, then evaporating and turning into losses, and then decides that the market has finally changed — that’s what all analysts or other professional market oracles have always warned him about! — and orders to stop trading.
This time the drawdown is not that prolonged and not that deep at all, and just a month later the equity resumes its growth. Mr. Smart says that in summer all markets are slow, so he will wait for some better time.
In early August, when the live equity on Mr. Nonchalant’s account exceeded 50% annual growth, Mr. Smart resumes trading, again with small amount. By the beginning of September there happens a very sharp rise of equity, and Mr. Smart immediately increases the size. A couple of days later follows the drawdown, as sharp as the preceding growth, but as now the trading size is twice as large as it was, Mr. Smart experiences a deep drawdown instead of being just around 0 (if counting from September).
The story goes on and on, and only in Novermber Mr. Smart understands the importance of following the single set of rules, the importance of being consistent not only in trading, but also in money management or investment. But — alas! — it was a bit too late, and the strategy haven’t brought any substantial income by end of last year any more.
So, let’s summarize the smart actions of Mr. Smart. He increased the size when he started to feel that he’s about to miss the train — and always missed it. He suffered from greater drawdowns while having modest returns. In other words, he cut profits and let losses grow.
Then bottom line: the strategy performance on the Mr. Nonchalant’s account is about 120%, and Mr. Smart proudly won about 20% for about the same period of time.