Backtesting of forex trading strategies is popular among traders, but popularity does not equate with efficiency. Due to the problems inherent in using historical data to test a strategy that will be employed in the future, many times the results obtained from backtesting are misleading at best, and nonsensical at worst. Let’s take a look at the disadvantages of using backtesting in the evaluation technical strategies.
Backtesting ignores broker-specific issues
Backtesting assumes that our broker is perfect. His software is stable always, his quotes are correct, and there are no connection issues as we enter our trading orders. Although this may be the case in short term trading, in the long-term, when sometimes hundreds of trades are involved, this issue can be an important variable determining the profitability of a trading strategy. Backtesting takes the market data in a sterilized form, and does not consider the impact of such broker-specific issues.
Backtesting ignores the impact of emotions, assumes a robotic trader
If you’re backtesting an automated strategy, you can still manually intervene at any point if you’re not emotionally at ease enough to let the automated system do its work. If you’re backtesting a strategy that will be manually executed, the problem is even more severe, because backtesting obviously ignores all problems that can be caused by the emotional status of the trader. As such, the results of such backtesting have very little value in actual market conditions where mental pressures can be difficult to control.
The market is chaotic, rules change all the time, nothing works on a consistent basis
The main problem of using historic data for testing a strategy is caused by the constantly changing nature of the markets. The market is chaotic; not only because prices change all the time, but also because the mathematical rules that create them change as well, and are for the most part unpredictable. It is also thought that the price action has the Markov Property, where future values are dependent on only the present value of the price quote. If this is the case, the natural conclusion should be that the historical progression of price trends and the market action has very little significance for future prices. Indeed, it is highly likely that the patterns of the past will never be repeated again in the future, due to the chaotic nature of the price action that we just discussed . By using backtesting we establish that a strategy could have been profitable in the past, but say nothing about its future performance.
Backtesting is good for education, but not for validating strategies
Backtesting is most useful for getting used to the basics of the market at no risk. You can try all kinds of configurations in this way in order to see how you react as they progress, without making faulty assumptions about their profitability. In other words, backtesting can be good for improving our own attitudes, and improving our understanding of indicators and technical trading, since there is no risk of loss in using it.
Backtesting-optimized strategies are almost certain to fail
We just noted that the price action is chaotic. If the price action is chaotic, any tiny change in the starting conditions can result in greatly differing future scenarios, and any past scenario is by definition highly unlikely, if not impossible, to be repeated in the future, in the exact form. And yet, some traders, and developers try to optimize their strategies to generate the best conditions in the markets of the past. Other than losses, what can be the result of optimizing your strategy for a scenario that will not be repeated in the future?
Backtesting creates a faulty mentality
In most cases, the key to a successful forex trading career is a good money management strategy. There are countless technical configurations that can generate good results in trading, but the rules of money management are strict and crucial. Backtesting creates and sustains the false perception that there’s a Holy Grail to trading. That you can discover a technical configuration that will generate good profits at all times. This belief, however, is wrong. Success in forex is always about managing losses efficiently, and backtesting in search of perfect strategies stands in the way of realizing this fact.
Backtesting is a marketing tactic
Let’s also remember that backtesting is not a scientific method. It is for the most part a marketing tactic employed by many different kinds of traders in order to increase the appeal of their products. This is not wrong by itself, but attaching too much credibility to a marketing method, is probably not a very great idea.
Money management is more important than technical strategies
Backtesting is all about technical strategies. But experience shows that it is possible to create an enormous number of technical strategies and use them successfully for profiting in the market. There’s no justification to the belief that one technical strategy is better than other, and it is often the case that a strategy that functions well one day fails on the other. The key, therefore, is not in trying to optimize the strategy, but in ensuring that you can manage the losses registered during trading. And backtesting teaches nothing about this aspect of trading.
Actual trading under real market conditions, and not hypothetical results, are the basis of a meaningful evaluation of results.
Let us conclude this article by noting that there’s no lesson, demo trading, or paper account that will substitute for the uncertain and sometimes tense atmosphere of a real trading experience. Backtesting will generate neat and tidy results helping you assume false confidence about whatever scenario you have devised mentally, but it is extremely unlikely that this confidence will survive for long in the markets. If you seek to make use of backtesting, of course you’re free to do so. But if you want to improve your results, make sure that you maintain a skeptical attitude to the results generated by it, and always confirm your thoughts with real market action, and with a conservative attitude. And make sure that you ignore the claims about back tested magical systems that never fail found everywhere in the wild world of the worldwide web.
