RE: Real Account Doesnt work

[Michael R. Bryant]

Can you elaborate on what happened in real trading and how it didn’t match the MT4 back-test? Were there execution problems, order generation problems, or something else?

 

Mike Bryant

Adaptrade Software

 

Subject: Real Account Doesnt work

 

Hi,

 

I am trying several strategies in forex market generated by builder. The backtest is very great in builder and in MT4, but in real account doesnt work.

Simply blow up my account in 20 days.

 

How can I trust in this software?

 

Anyone has a real forex account with profit?

--

[mandelmus]

Ya, there are so many challenges to getting builder strategies to work real-time with metatrader at your broker.  If you are new to Metatrader and/or backtesting, etc, you have some work ahead of you.

Have you read my posts?
https://groups.google.com/forum/?fromgroups=#!topic/adaptrade-builder/4ck3-PtiGmg
https://groups.google.com/forum/?fromgroups=#!topic/adaptrade-builder/10mErvTMQpc

To begin with, you must develop all your strategies, regardless of time frame, using all available tick data.  If you don't know why, read Birt's blog here ... http://eareview.net/tick-data ... I use his recommendations with the software Tickstory Lite and it worked great.  If you want some additional step-by-step instructions and tips, let me know and I'll post them. 

Unless your broker is DukasCopy, since Tickstory Lite uses DukasCopy data, your strategies built on DukasCopy data won't be an exact match with your current broker's data, but should be close enough in most cases. My Forex broker is FXCM and their historical tick data is useless and/or non-existent. 

More importantly, I hope you realize that many Forex brokers (not all) hedge against their clients either at the server level or at the client level.  In other words, your own broker is NOT on your side.  Watch this video on the MT4 Virtual Dealer Plugin that many brokers use ... this is old news.  I know there is a lot of debate about this issues, but it is likely that if your builder strategy is successful for several few weeks, your broker may eventually learn your strategy and find ways to trade even more against you.  There has been a lot of talk over the years about brokers purchasing FAP Turbo themselves and using those signals to trade against their own clients who also trade FAP Turbo.

If your Builder strategy places protective stops via MT4 code, your broker will hunt for and see those stops and hit them.  That's why I asked Mike to hide our stops ... in other words, have Builder either not place the stops via MT4 code or place fake stops way outside of our desired stops and then auto-exit when the price meets the "real" stop.  Several EAs I have used had to have that feature to trick broker computers or they would never have made money.

This is why it is so frustrating trading Forex with any broker -- I don't trust any of them.  I have been trading Forex manually and with EAs for over 4 years with a variety of brokers.

Out of the dozen or so strategies I created with Builder a few months ago, only a few worked as expected going forward in real-time.

[enlightenor]

I appreciate your methods to prevent brokers from hunting for stops to hit them and trade against you. When you say "and then auto-exit when the price meets the "real" stop", do you mean trigger an exit at market order, or what?

Thanks,
Steven

[Ell]

John, what kind of tests did you make on the strategies before using them. Did you forward test them? All strategies developed using historical data er fitted and only a tiny percentage of them works in actual trading. If you are on top of this trading forex intraday you should not expect to make money and the chances of losing it all are high. Just ask yourself before trading: "Who is going to lose money in forex so that I can win?". Nobody is willing to lose money, especially your broker who is your counter-party in most cases. Try to use AB in a more effective way. AB cannot defy the markets and their rules. This is my opinion of course based on my professional experience. 

[mandelmus]

Basically, we should have the strategy place a "fake" stop way outside of the strategy's "real" stop.  That should give our strategy enough room and time for a surprise exit before the broker takes us out.

[NewJ]

Using the forex example strategies found in the trial AB, the 99% quality back test results of MT4 was not matching with the AB. Would really appreciate if you can post the step-by-step instructions and tips. Thanks.

NewJ

[YUKUN ZHAO]

SPREAD affect a lot....

在 2013年5月14日星期二UTC+8上午3时39分27秒，John写道：

[Fadi Al Khoury]

I have the same problem as NewJ. The spread and slippage shouldn't effect much in my case since my strategy works on the daily chart. What do you mean? Did you get metatrader to match? If so, please, could you post your exact procedure for getting them to match.