Short Term Randmoness and Long Term Statistics : Learning to Evaluate the Live Results of a Trading System

It is not a surprise that so many people lose money with automated forex systems when you realize that most new traders do not actually know how to evaluate the live results of a trading strategy. Evaluating the results of a trading system is not only looking at them and figuring out the average monthly or yearly return, it is about interpreting the actual significance of the results shown  making an accurate and quantitative analysis that weights the profitable aspects of the system against the possible losing scenarios. Learning to evaluate a system’s live trading results is perhaps one of the most important skills a new trader needs to succeed in the long term, especially when he or she wants to succeed through the use of mechanical manual or automated trading strategies.  Within the next few paragraphs I will explain to you the basics of how you can evaluate  a trading system to get a good idea about its potential for long term profitability, its risk aspects and the weight of the results.

The first thing we must understand in order to evaluate trading strategies is how the distribution of returns work on the forex market. The outcome of trades in the short term is determined almost entirely by random chance while in the long term the “edge” of the trading system starts to show as an accumulation of positive outcomes resulting from seemingly random trading occurrences. This is analogous to what happens within a casino. When you start to play the roulette at a casino the outcome of bets in the short term will be random and therefore you can place a bet, win and get out of a casino with a profit. However, in the long term the probability of winning is lower than that of losing and therefore, as a statistically significant amount of results accumulates the house inevitably wins.

When you use a trading system you are playing by these same rules. In the short term the trading of a system can be profitable or unprofitable almost entirely by chance while in the longer term any “edge” that tilts the balance towards the winning side will make the system accumulate profits while the opposite case – or the absence of an edge – will lead to an account wipe out. What we see here is that in the short term almost any outcome is possible while in the long term only a positive outcome is possible if a system has an actual statistical edge on the market. This clearly explains why you can see someone pull amazing returns in the short term – like in EA competitions – since the results in the short term can effectively be anything as they are determined almost entirely by chance. This is why the length of time a system has been trading is a very important characteristic to take into account as the short term results of any strategy can trick us into thinking that a strategy has a positive edge when in fact the profitable results are only a result of random chance.

But how long is long enough ? I have mentioned several times within this blog that I consider long term results only those of a system that has been in the market long enough to have experienced a significant cycling in longer term volatility, meaning that at least 5 years of trading are necessary to hint at the existence of a positive mathematical expectancy in live trading. Since this is a very long period of time the best possible solution is to use reliable simulations and then validate them through 6 months or a year of live trading results compared to simulations of this same period. Of course, the final test will be a period of 5-10 years under live market conditions and a system that survives to such a test is very likely to have a positive statistical edge on the market.

When taking into account the amount of time that is needed for a system to achieve live results that are long enough to point at a significant edge we are obviously left with almost no mechanical systems that achieve real profitability. All systems sold online lack this amount of live testing and none of them show 10 year reliable backtests coupled with a 6-12 month live testing to compare back/live testing consistency (how well the results of a live, real money test match with those of a back-test of that same period). Watukushay FE – a free system I made – is the only free system I am aware of that is able to fulfill this criteria with 10 year reliable simulations, a test of consistency and real money, live investor access verifiable results.

As you can see the whole way of selling systems right now and tricking people into believing something is profitable is through the use of random short term results in order to fake the appearance of a positive statistical edge (something which is also done with unverified simulations from time to time). This works because the return of this random results is effectively almost unlimited since systems that lack any kind of edge usually only lose money in the market as a function of the spread so they can have apparently long periods of profitability before wiping an account. This can be confirmed by a simple experiment you can do. If you run 100 demo accounts with a fixed TP and SL target that enter the market at a random place once every week you will note that a small percentage (perhaps 10%) will show very profitable results after a month just because they were “lucky enough” to accumulate a significant percentage of favorable positions by chance. Of course, in the long term all the systems will lose money as a function of the spread costs but in the short term results that show high profitability are possible.

This becomes even worse when you consider systems that “postpone risk” which simply use very unfavorable risk to reward ratios or progressive money management techniques to increase the amount of profitable trades achievable merely by chance. If you repeat the above experiment of random entries with an SL to TP ratio of 100:1 you will note that after a year a much larger percentage of your accounts will be profitable (perhaps even 30%) because you “postponed risk” by making the reaching of the TP by random chance much more likely than the reaching of the SL. Of course in the longer term – since no true edge is present – some very unfavorable stop loss value (or sequence of them)  will be hit and the account will be wiped out.

Evaluating a trading strategy is therefore much more than simply looking at how much money it has made and how much draw down it has had. Evaluating a trading system is about looking at the meaning of the results, the significance of the profit and draw down levels achieved and how these results point or do not point to the achievement of a true positive edge over the market. Achieving a positive edge is not something terribly hard to do from a technical point of view but it does require a lot of knowledge and preparation and the trading of a strategy with a positive edge is always extremely hard due to the inevitable draw down periods and other psychological hardships a successful forex trader has to endure.

If you would like to know more about how you too can design systems that are likely long term profitable with sound trading techniques and an in-depth evaluation of their trading characteristics please consider joining, a website filled with educational videos, trading systems, development and a sound, honest and transparent approach towards automated trading in general . I hope you enjoyed this article ! :o)

Print Friendly, PDF & Email
You can leave a response, or trackback from your own site.

Leave a Reply

internal_server_error <![CDATA[WordPress &rsaquo; Error]]> 500