Automatic trading

I have heard that some traders after developing their own working and successful trading systems they programmed it for the automatic trading.And the trading robot or software will automatically execute trades once certain requirements have been met like price levels ,indicators etc. But still it can,t replace human intelligence and the ability to decide according to the unforseen market situations etc which are part of our everyday trading life . So how reliable they are if anybody is using such a system please share.And what about those thousands of softwares available online are they all scams?

Learn to manual trade. For example http://www.forexfactory.com/showthread.php?t=236210

Most automatic trading is crap. In fact i haven't found one that works, except for one that gives 86% drawdown , but works for 10 years , placing one trade per month on eurusd! Im sure its possible, i just haven't found one that works. And ive looked

My goal is also to automate my trading. But im crappy at manual trading and ccant make a consistent profit. I know its possible, cause other people have done it. So in order to automate this i need to become consistent. I plan to record all myt rades once i become profitable and then try to get an ea to take the same trades based on whatever method works for me. That way, if i m consistent for one year, ill know it might be possible to program an ea to be the same during that year.

Ill figure it out one day! Itll be fun

Given most EAs are developed by people who don't trade for a living to begin with (evidenced by the fact that they're selling you software), it's hard to believe that that very software is going to make you any money. The vast majority of EAs use very simple logic; for example, moving average cross software that comes out null in the long term, negative if you factor in fees. I'm not saying that EAs can't make you money - but I think developing an automated system would be much harder than learning to trade manually in the first place. The one benefit of an EA is to show you that very simplistic indicator-based strategies won't work long term, though. I believe to extract profit from the market, you have to observe some of the more subjective features such as supply and demand and price action at these levels. Given how subjective these things are and how conditions vary across markets, it's very difficult to program something to identify what you readily identify with your eyes.

TL;DR: Forget automated trading until you know what you're doing manually. By that point you'll have realized what it takes.

When you refer to automatic trading , you are refering to expert advisor or EA. Remember EA are programmed based on technical indicators, and all technical indicators are made of mathematical algorithms of what already happen. This means all technical indicators lag. EAs may seem to work in some cases especially on trending markets, but the problem is that markets only trend about 5 to 10 percent of the time. So far in my trading time I did not found a EA that make money in the long term and didn't know any trader making money with an EA either. The best way to make it in this bussines is to work hard, develop a plan, and paper trade it until gain confidence. This will take long time, there is no easy path to riches here, but at the end if you stick with it the rewards will be limitless.

of course it's possible and happening everyday. current estimates suggest high-frequency trading, a specific type of algo trading comprises 50% of the volume on the NYSE. But if a black box truly works well it holds far greater value in it's ability to trade than it's ability be sold for $97--so the good ones usually aren't floating around on the internet.

Hi,

The best way to make profits in Fx is to learn to trade manually. Learn to trade by using price action.

From my experience, automatic trading simply does not work, at least it didn't work for me. Abt 1 yr ago, when i newly joined Forex i wanted to make some quick cash so i invested on an automatic robot to do trading for me, but in the long run i started losing money and eventually i stopped using it.

Since then i started to learn FX by myself, there are many free courses that teaches us how to use price action to trade, in my opinion very useful.

I'm still relatively new to Forex, but i have learnt a couple of methods/strategies that works for me and i only use these methods to trade.

Remember all you need is only one successful trading method to make consistent profits. There are many strategies shared in this forum by fellow traders, you can try them out in demo acct, and hopefully u will find one strategy that works for you

But if you still intend to invest on any automatic trading stuffs, at least learn the basics of FX first so that at least you can roughly see whether your robot is doing profitable trades for you.

Good luck,

I can agree )

share it just like that? you must be joking.

Yes 98% of the ones flying around online are not good

kos

yes

I'll be honest with you. I have used the following and been successful: http://forexsb.com/wiki/start

I think the reason most automated systems fail is perhaps two reasons, firstly most rely on lagging indicators that are rooted in the past not the here and now. You don't need to be a genius to realise that trading on past price movment is like trading on a chart with a 15 minute delay. Most programmers have no idea of trading and cannot be bothered to work at it. Programmers want a quick hit, find a method then let the computer try all different combinations to find the answer, sadly markets are like women the constantly change their minds and although they seem unpredictible, they are in fact quasi predictible.

What you need is a good programmer, a trader and new tools that do not use lagging data but trade in the moment. As humans we can see and decide given vast amounts of information, machines need to be programmed to make the same judgments we humans make and that requires understanding. Which to be honest a moving average cross will never have. I also think that programmers tend to look for scalping type systems, most of the programmers I met wanted to trade 5 minute or 1 minute charts, but never looked at the real picture which is a must for successful trading.

As retail traders we cannot scalp .2 of a pip because of spread, we cannot front run orders as banks do, so in order for us to make money we have to be even smarter then the guys that work for the banks, because they have access to lots more information and execution advantages we will never have.

I once designed a scalper for the GBPJPY, it used a rolling percentage of its capital and just took three pips, when I back tested it it made hundreds of thousands of dollars, but you needed to get in for virtually no spread and that is impossible in the real world.

All the best


Mike Haran

Regarding automated trading...yes, there are traders out there who are successful using automated strategies. Yes, it is possible. Some of them are very simple, some are very complex. How do I know this? Because I've been doing it for almost 3 years now. No, I will not share the system on a public forum. No, I will not share my account statements on a public forum. No, I would never sell it on Ebay.

Yes, I would guess most publicly available automated systems are rubbish...but I'm sure there are some that are at least short-term profitable and even a few that are long-term profitable.

Most automated trading systems fail because they do not take all market conditions into account. You need to learn how to define all (most) possible market conditions and stress test your strategy against these conditions. How many market conditions are there? Well, that would depend on how you define them and how your strategy trades. Maybe there are only 2 market conditions, maybe there are 100s...it all depends.

For example, you may come up with 5 different market conditions:

Hyper Up Trend
Up Trend
Ranging
Down Trend
Hyper Down Trend

Your goal during a stress test is to determine how your strategy will perform during these different market conditions. Is it profitable under all 5 conditions? If not, how big of a loss is expected? You can then use this information accordingly.

The real challenge is how does your system perform during new market conditions that you did not think of.

No, you do not need discretion/intuition to be successful trading Forex...it is possible to trade a fully automated/mechanical system. You just need to tell the program what to do, how to do it, and when to do it.

All this talk about lagging indicators. There is only one thing on a trading chart that is current and that is the most recent tick (and some may say even that is lagging). Everything else is in the past. And I don't think anyone bases trading decisions solely on the current tick. Everyone (whether they admit it or not) uses past (lagging) data to form trading decisions.

Of course, some indicators lag more than others. There are some indicators that try to predict the future...but you'd have to think they are also based (in some way) on past data.

To the guy who said the "markets only trend about 5 to 10 percent of the time", I'd really love to see some data to back this statement up. Because I read these types of statements all the time but no one ever gives any data to prove these claims...these numbers seem to be pulled right out of thin air. What markets? What timeframe? How do you define a trend? How do you define a non-trend?

One thing i don't like with current forex education is that they stress following rules as necessary and then you have eas which miserably fail when they follow rules strictly Noone ever mentions the key element which is necessary. Maybe rules based on human visual recognition is better or what? One interesting option for exploring creating a profitable ea andthe difference between humans trading and eas is that of the cowabunga system. The cowabunga system has documented all its trades for 2007 and 2008. It is profitable. Yet all the cowabunga eas fail miserably on the same time period. Why? GOOD QUESTION! Comparing the trades made by the ea to the recorded trades would be interesting . But ive decided i need to learn manual trading first, eas later.

I present you with two crappy eas which still manage to eak some money out of the market on a daily timeframe on only the eurusd (dont ask why idk) for 10 years straight. Large drawdown (30% on turtleand 60% on lowtrader). The turtle ea only wins like 30% lol. Take a look at zulutrade and see how much better equity curves can look with less drawdown. If these eas can manage it , anyone can manage it. Now, i still haven't found an intraday ea that even manages to eak out some money over a long time period. Probly cause there are many more market conditions on the smaller timeframe that have to be taken into account. Unless of course you take a look at martingale eas which will eventually blow you up.

I hope this provides some inspiration that its not impossible Far from it! All you really need to look at is other peoples accounts to see how possible it is. have fun!

turtle



lowtrader




If i had a profitable intraday beast ea, i would share it. but i dont. its so sad oh well! this is my new hobby/potential job if i get good. itll be great fun

noone else wanna show an ea that actually gains profit?

A good, solid, profitable EA is worth millions of dollars. You expect someone to share it for free??

haha not really unless some generous soul wants to for a late christmas gift lol i would but im giftless and bereft of good solid eas

I see a lot of incorrect information on this thread. Algorithmic trading works, and just because you don't understand it, or have been burnt by a trading system that you purchased, doesn't mean that it is universal that no one can use it. In fact, nearly 65% of all volume on major exchanges is done through algos. The market has evolved very much in the past 5 years, and in order to survive in this business, you need to adapt, otherwise you won't survive. I see discretionary traders trying to trade on short tfs like 5m and 1m and I cringe. The algos control those tfs, and as a discretionary trader you don't stand a chance in the long term trading against people who have very large capital resources, unlimited talent pools, and data mining and flow much faster than you ever would be able to gain access to. The algos know exactly what you are doing, thinking and feeling in the shorter timeframes. The retail traders who use algos or "EAs" as they like to say, are simply outmatched, as they are competing against far more sophisticated algos that will always come out on top. It is a complete losing battle, stay away.

You need to develop, or use a system that fits your personality and your risk tolerance. It is very naive to think that you can just buy an "EA" and turn it on and leave it on and it will make you money. Those systems are for new suckers just coming into the game that don't have any experience so they will rely on others to make a quick buck. However, those schemes are quite the opposite. The newbie traders pays a bunch of money for a worthless product, which the seller reaps the benefits. Ask yourself this question: If the EA that they are trying to sell me is so profitable like they claim, then why are they selling it to me instead of using it for themselves?

There are ways, however to be successful in algo trading from a retail standpoint. I will personally say from experience that it is possible. You need to base your trading decisions on larger timeframes, to the point where you can identify trending markets or ranges. You can either choose to use algos that operate on higher timeframes, or you can use a systematic hybrid approach where you can base your bias, whether it be long/short/range, on a larger timeframes, then use your according algos which will trade in the direction of the trend, or will trade a range.

My thread is dedicated to my style, which is an algorithmic hybrid strategy. http://www.forexfactory.com/showthread.php?t=400914

You have added the buzz word "Algorithmic" to your system to come up with something sounding fancy. Your system has nothing to do with algorithmic trading. It's just momemtum and moving averages rehashed to make it appear super sophisticated.

I'm not even going to entertain your ignorance.