An Automated Trading System (ATS), it is a software that can create orders for you and automatically submit them to the market based on your pre-defined set of rules and trading strategy.
It may also be known as algorithmic trading. Automated trading systems can help you execute repetitive orders with greater speed and magnitude than you could do on your own. Automated Trading System (ATS) is often referred as an Expert Advisor – EA).
EAs operates on a set of rules without greed, fear, ego or bias.
You can make your strategy far more sophisticated than trading manually.
EAs can monitor many currency pairs at the same time, entering more trading opportunities.
In this article, we will discuss the pros and cons of the Automated Trading System (EA). If you want to learn more about the Automated Trading System and coding in MQL4, read our dedicated section MQL4 Beginner's Tutorial | Syntax, Variables, Comments, Functions.
Human traders have severe restraints on their time — they have to eat and sleep and go to the restroom, not to mention other responsibilities such as work, family, and friends. It is next to impossible for a human to trade the entire 24-hour currency market.
He is often forced to trade during certain times of his waking day, but in doing so he is missing out on both entry and exit opportunities offered by other time sessions. The EA does not sleep and rest and can work the entire 24-hour currency shift searching for trading opportunities.
When the EA trades, you can go have fun, work on other things, spend time with friends and family. You are no longer a prisoner to the computer screen and charts. If you want to be more involved, you can spend your time researching for (and coding) new and improved ideas for your EA, and then back-testing and optimizing these ideas. This would be using your time smartly in R&D (Research and Development), instead of wasting it on planning, executing and monitoring of singular trades.
Fear and greed are the biggest obstacles to successful trading. Fear becomes an irrational force when it prevents the trader from taking necessary trades or hesitating too long, particularly after having suffered a losing trade, or it prevents a trader from closing out a bad trade with a loss. Greed, it’s the emotional opposite, can cause traders to make random trades, or hold on to positions longer than their trading system dictates.
Humans are also plagued with ego and bias, seeing what they want to see from the chart and indicators, looking for confirmation for their hunches, and unable to see things critically and objectively. So long as there is real money at stake it is very difficult for a human to overcome these negative emotions and biases. EAs help combat emotions by getting you in and out of the market based on historically tested strategies.
The ability to back-test on quantifiable data is an epistemological paradigm shift in favor of the EA. Prior to that human traders either did not back-test their trading ideas or they attempted to do so in a very awkward fashion. Often they had to resort to the very crude and time-consuming task of printing out reams of historical charts and “eyeballing” the behavior of the rules related to their indicator(s), making notes of the gains and losses in a ledger.
Due to the time-consuming nature of such an activity, they often backtested only on higher time frame charts and preferred not to go back too far in time. Each change in the rule or parameter of the indicator would force a repeat process, making it exceedingly difficult to optimize. With the built-in backtesting and optimization engines, EAs conduct these necessary and formerly incredible tasks in minutes. To assist with back-testing, MetaTrader servers store a huge quantity of downloadable AND free historical data, in intraday format from 1999 till present, and daily or greater from 1971 till present.
You do not need to worry about missing a trading opportunity because you are not staring at the right chart closely enough, or wasting precious seconds entering an order manually while the market moves away from you. The EA uses the speed of the computer to monitor the markets, seek and identify trading opportunities based on coded rules, and execute based on these rules in fractions of seconds.
With EAs, you do not need to start out being an expert trader and programmer. There are plenty of free and commercial EAs that you can run on your own computers. You have to properly back-test each one to see it matches your performance and risk criteria, and you should forward test each one on virtual money prior to trading live. You can then use the free time you now have because the EA is trading for you to learn more about forex, trading and programming so that you can develop your own EAs.
EAs gives you the potential to trade with complex systems that can include multiple conditions for their entries and exits, profit targets, protective stops, trailing stops, and filters, and have them all automated simultaneously. The human brain is very powerful but it cannot consciously multitask dozens of simultaneous operations like a robot can.
Plenty of wise traders council on the importance of having a trading plan and sticking to it with utmost discipline. Most human traders have no trading plan or fail to stick to them. The EA is designed to stick to the plan without exceptions. It is the way to become a 100% disciplined trader overnight even if you are the most undisciplined person in your private or professional life.
You no longer have to be glued to the screen, trying to keep track of each of your favorite currencies. An EA gives you the power to monitor dozens of currencies at once and do it more efficiently and effortlessly than before.
A manual trader really has to work at being a good trader, doing his homework on the markets every day, watching the markets for an opportunity, carefully entering into a trade and then spending a good deal of time babysitting that trade.
This arduous process is repeated for every trade, day after day. It is a lot of work and is very stressful. The EA takes over these tasks effortless and reliability, and you feel a great weight lifted off your shoulders in terms of work and stress.
An EA needs to run on a decent computer at home with reliable internet, or on an affordable Virtual Private Server (VPS), 24 hrs a day, in order for it to behave properly and take all the trades it was designed to take. The good news is that most internet connections are up 99% of the time, and there are plenty of low-priced and reliable VPS providers out there.
Many commercial EAs are designed to work best on brokers with a tight spreads, particularly so if it is a scalping EA. If it is a scalping EA, you will have to investigate the spreads of your broker to see if they are low enough for the pairs your EA trades, in addition to asking your broker if it allows scalping.
Most EAs are coded for one pair only, although some work on multi-pairs. The reason is, different formulas and conditions work differently across different pairs, despite the >80% daily correlation between many of them.
Most EAs are mathematically and technically based, and work best on trending markets, so they are vulnerable when the markets become sideways. When the markets enter into low pip range, sideways activity, EAs tends to get chopped up.
Manual intervention of the EA is not advisable: usually, the creator of the EA has a better understanding of its inner workings and has hopefully done all the appropriate back-testing and optimizations. Traders who want to fiddle with exiting trades according to their own whims or rules invariably end up degrading the EA’s full potential.
It is very difficult to get an EA to read and decipher fundamental data, such as economic and political events. Hopefully, the result of the event is impeded in the price action, and then the EA can read it. But a sudden extreme event that quickly reverses the market direction can cause an EA to get stopped out, and/or enter too late in the new direction.
It is difficult for any EA strategy developer to plan and design a robust EA as there are few very good examples (or design methodologies) of such. Most do not have a proper method and criteria, and there are numerous coding and optimization pitfalls, such as over-optimization and curve-fitting.
It is a sad fact that 95% of EAs end up failing, either losing a bunch or just barely staying alive. Interestingly enough, about 95% of manual traders fail as well. The EA has a number of potential edge factors over a manual trader, without all the emotional and physical limitations of a human, and yet in the end, the markets are so exceedingly difficult most EAs cannot win against it.
It helps if the EA developer is both an excellent coder and an experienced trader with the humility to feel one knows hardly anything at all about both. This is a hard juggle. Too often one finds excellent coders who have little experience in trading, or experienced traders who cannot code. Even when finds a combination of both, the experienced trader who can code well, there is still the ego factor to deal with.
Pride and ego, while good motivators for craftsmanship, are dead-end allies to learning and growth. It helps to be humble coder and trader with a passion to learn and renew and grow one’s ideas about the markets and how to code for them.
In summary, EAs can trade 24 hrs per day with greater consistency of one’s trading plan, acting on predefined entry/exit conditions. Metatrader 4 (MT4) is such a popular platform that you spend a lifetime investigating and researching (sharing and purchasing) thousands of EAs in numerous online forums. And there is an endless supply of education out there to code your own EA.
Whether you borrow or create your own EA, it is imperative that you test it on plenty of historical and forward data. MT4 allows you to perform EA back-tests to see how they work in the past, and if the back-tests prove promising, one can test the EA in forward market conditions, preferably in non-expiring demo accounts until enough confidence and understanding of the EA is acquired to test it on real accounts. Contrast this approach with borrowing or creating a manual system.
The manual system would be very difficult to back-test properly other than the eyeball method that is itself besot with bias. Furthermore, the manual system would require the human trader to sit in front of the computer for countless hours looking for the trade setups and exit opportunities.
The EA, in contrast, frees up the trader to do other things. The trader does not need to be physically around, and so can be free to focus on his day job (most day jobs are far more reliable methods of earning money than trading), hobbies and relationships. Also, that free time can be spent improving trading and coding skills.
Trading with EAs is not in itself going to make you rich and more than you likely they will make you poor if you trade them in real accounts. I would estimate that 95% or more of the EAs one finds or purchases or creates will turn out to be unprofitable over time.
That is an unfortunate fact. An EA can reduce a large number of problems that beset human traders (such as limitations of time and speed, greed, fear, ego, diligence, consistency to trading plan, etc.), but it is very hard to construct a robust EA that has not been curve-fitted, and the markets are brutally tough, with a high degree of unpredictable volatility and randomness that degrade the performance of the best conceived robots.
But despite this horrible reality, I still like the quest of finding and/or creating promising EAs to test in unexpiring demo accounts until I find or create that pearl in the pile. If the markets were not so terribly tough, we might all be able to find or develop a bunch of EAs to make us richer, but at the same time the trading world would be perhaps more dull, as there would be less challenge, less propensity to dream the impossible dream and fight the unbeatable foe.
