Monday, October 27, 2008

Automated Trading systems

As I have said before, I have never been a great fan of automated trading systems but the past 6 months or so of testing various Expert Advisers (automated systems) has persuaded me that this is probably going to be the future of trading. These systems have become sought after trading tools for traders at all levels including part time amateurs and professionals alike. That's not to say any of these systems are the be all and end all of trading, it will still require astute professional traders with proven track records and trading systems to design systems that work.

As with all trading systems there are pro's and cons.

Some of the major advantages have been the fact that the EA can find trades I would not ordinarily do as I have to sleep sometime and cannot possibly watch the market 24/7 unless I employ someone to do that for me. A good EA can become a cheap faithful and loyal employee.

Using an EA removes all the emotions from the trading process. As humans we are often subject to different emotions. When things are going our way, the adrenalin is pumping and we cannot wait to execute the next trade. However, have three losses in a row and it becomes difficult to follow your tried and tested system and doubts start to creep in and we start looking for a more reliable method.

The speed at which the EA can execute trades and make decisions is far faster than the human can and they are consistent in executing the trades giving an even spread of trades 5 days a week. This consistency is important because when we have a bad run and stay out of the market suddenly things go our way and we miss the trade.

Obviously the downside is changing market conditions, EA's are programmed to trade specific events, (maybe the cross of moving averages, break of support or resistance or a host of other triggers.) when it suddenly fails we have no Idea how to fix it other than to turn it off until market conditions change.

The EA I was testing in my previous article performed very well for three consecutive months, producing consistent profits daily. When the big move started in August it literally blew the entire account in a week. Looking at equity curves on promo material often doesn't reveal the entire truth, as with the previous test. Though the profits were consistent the actual risk is not visible on the equity curve as only the final result of the closed trades become part of the equity curve.

The reality was that often the account was down sometimes up to 10k before the profits were realised. A 30 to 40 percent draw down is just too large a risk on a 25k account and the warning bells should have sounded that sooner or later it would not recover and the account would be blown.

I have tested a number of systems over the past 6 months and have only found one that has stayed in profit profit regardless of market conditions and that one is Forex Auto Pilot. Having said that I do not believe it can live up to its advertising hype and produce the sort of returns it claims over a 12 month period. The operative word is('it stayed in profit') it did not make buckets full of money.

I have yet to find one that operates on sound fundamental trading principals and a review of these principals mean.

1 ...... Risk for most the risk is an input figure by the user rather than a figure based on an equity calculation that uses the difference between the entry point and the previous high or low to calculate the actual risk and adjust the lot size accordingly.

2 ..... Take profits again are mostly arbitrary input figures by the user rather than a decent risk to reward of 1.5 to 2 times the risk factor.

3.... They all seem to be focused on the entry with scant regard to risk or a decent risk reward ratio.

All these systems can be used on demo and rather than entice people to first lay out good money. A protected time limit demo EA could be offered for traders to first evaluate before making a purchase. If the Demo version lives up to its reputation then I am certain people will rush to buy the system.

I do believe though that the really good ones will stand out head and shoulders above the ordinary ones and will become a benchmark for developers to strive for. I won't go so far as to say that Automated systems will entirely replace humans but I can foresee a time in the future when many institutional traders will be replaced by automated systems.

Commando Ea

I was sent a copy of this ea to test about 3 months ago and tested it on demo for the full 2 months. The system is based on the martingale system, as recommended I traded the four majors simultaneously. on a 10c 5 k account and on a 25k 1$ account.

On the 1 $ account it produced profits of up to 13 k in a week and the 10c account a record week of $1300. Whilst the profits sound impressive they don't show the ever present risk in trading the Martingale system. Like all Martingale systems it piled on the profits during the quiet periods but had big losses when the market trended in any direction. Week after week it managed to recover the losses and then some but would then give it all back the moment there was a major move.

After 2 months the account finally blew itself out. It does have a 200 pip stop loss so the pyramiding effect was not infinite. It also has a timer that can be set to stop trades at particular time which helped to eliminate many of the large runs which might have brought about the accounts demise a lot earlier.

Maybe when the Forex market gets back to normal, this might be a good EA because it certainly could pile on the profits. Then again what is normal in a Forex market that changes daily. Because it is based on the martingale system and doubles lots against the trend the risk reward ratio is always against the trader and sooner or later will probably blow the account.The cost of this Ea has been reduced from $145 to $75

Money Grid
The same developer also sent me this EA about 2 weeks ago, also based on the martingale system but with a few changes. Instead of doubling up a set number of pips from entry it used increasing increments against the trend.

So if the first Grid was 20 pips the next double was at 30 pips then40 and 50 and so on. This definitely had the effect of reducing the number of lots placed in the market but made it equally difficult for the market to retrace enough pips to take profit.

The recommendation was to trade the 4 Majors and the Aud/Usd and Cad/Usd pairs simultaneously on a 1c account with a min of 5 K equity. After 2 days of trading it had piled on about $200 in profits, then took a 600 $ loss on the first big run so I reduced it to the 4 major pairs. I simultaneously ran a 500k demo trading one Mini lot at a a time on the four major currencies. It only took 5 days to wipe out both accounts completely.

Money Grid sells for $165.

The developer of these two systems Pro-FX-Experts is a good programmer and if you have any good Ideas for an expert adviser he can code it for you at a reasonable price.