I noticed there is very little on Babypips relating to automated trading so I thought I would start a thread explaining how it all works, and how to get set up.
Before I start I need to disclose that I run an online trading room where we offer automated systems, however this thread is not an advertisement. It is purely educational, and I welcome any questions any of you may have relating to how to trade Expert Advisors and signal services.
When I started out Forex trading in 2010 I only learned and practiced manual trading. I even used the school of pipsycology for much of my education. After a period of about 12 months (of losses) a friend of mine showed me how he was trading forex through automated software. I found this incredibly fascinating! The concept of a computer system trading the market for me, or that I could design a system and completely automate my strategy caught my imagination.
Since then I have researched, practiced and traded with dozens of Expert Advisors (software used to automate a strategy, aka a robot or EA). Most of them resulted in losses, but through these experiences I have learned to identify quality systems. At the moment I’m trading around 10 different automated systems across accounts totally close to $40,000. Some of the systems are from commercial sites and others are my own creation.
If anyone has any questions on how to start out in automated forex trading or designing their own system (don’t ask me programming questions, I outsource the coding) please let me know. This is something I’m quite passionate about.
Right now there are thousands of poor products across internet giving misinformation, and nowhere for traders to get all the data they need so they can learn how to start automated forex trading. Hopefully this thread can go part of the way to offering advice on the subject.
It’s actually easier and far less stressful than manual trading, but it takes a while to work out how it’s done.
Haley, I suggest looking for some independent and verified forward test results. Backtesting is all well and good, but nothing provides better proof that Forward tests. If an EA has a proven track record with forward testing, and the testin site can tell you their setup (Broker/account ETC) then you can match their setup and ensure you get very close, if not identical results.
I have been interested in created my own automated trading systems. I will be looking forward to your posts. I would love to see how you balance optimization and over curve fitting, as well as what fitness functions you use and evaluating overall robustness of a system.
Uhmmm… you can’t find an EA with “proven record with forward testing”, unless you travel in the future. Even if the results are of the very recent past, it is still “backtesting”. By “forward testing” we usually mean demo trading for some time so that you can make sure an EA works under the current conditions, but by the time it is “proven” it has already become “backtesting”.
The expression “proven forward testing” is an oxymoron.
It has backtest and currently demo-ing/forward test results. If you interested, the site will also help you set up a test system so you perform backtests yourself.
Hmm… Bots or EA’s. If you have several years in and your strat is long term positive then its your money right? Code it and take a flier. But consider this, an EA will work perhaps pretty good in the short term. But the market is never one way. Trending and rangebound and over what tf? The goal posts constantly shift. Does your EA work on only one or both? Does it factor in multiple tf’s and their trend and range?
I’m old school and figure there’s no better EA than whats between your ears. Trade blind faith in an EA or trust your long term experience? :33:
EDIT: By the way. There’s one doing the rounds that does look good! For how long remains to be seen?
That series of lessons, although good, frustrated the living hell out of me while trying to learn MQL4. If you have no programming experience, it’s hard to learn. I actually got more out of C++ For Dummies to begin with. There are so many important pieces to the puzzle that are just plain left out. Mostly because they are taken for granted that they are understood. And the MQL4 book? Fuggetaboutit. I know now, but there’s a MILLION easier ways to explain things than the way they are presented in the book. Good coders aren’t necessarily good communicators.
Once I understood the basics, then the Coder’s Guru stuff started making more sense. But until then, it might as well have been Swahili:D
Carter, your argument is definitely valid. But there are advantages to using algos. For example, as you said the market is dynamic, and is always changing. But that;s just the thing, even as a manual trader, your trading system may face a certain degree of drawdown. And as such an algo is much the same. But in my eyes, algo trading has the edge here; it’s very easy to adapt. For instance, how hard would it be for an individual to learn and practise a new trading system compared to running a different algo better suited for the current market.
Without giving away too much, I see my ATS as a Swiss Army knife. It has many tools AKA algos that I can choose from to combat any situation and I am constantly developing new algos and updating/monitoring my current ones. Just because I have a knife, screwdriver, chisel (I dunno what’s in a Swiss Army knife to be honest), it doesn’t necessarily mean I’m going to use all my tools at once. Rather, it is the idea that I have a wide array of tools to select from, and hopefully there will be a tool for each challenge I am faced with. So if you look at algo trading like that, we can see that algo trading does have a distinctive edge.
I’ve been playing around with programming languages since I was 12, first with creating my own RPG games (Playerworlds Client), then webdesign and now for trading. So I have a VERY BASIC understanding of VBScript, Java, C, C++, Python, PHP, HTML, CSS, and SQL. MQL4 is pretty much exactly like C, so I can pick out the bits and pieces and that’s how I’ve been learning.
Lol! Honestly, the easiest way to learn MQL4 is just to learn C first, then look at some sample code from indicators, scripts, then EAs. I got a big VISUAL (;)) EA project in progress right now, I’ll make sure to show you.
