Brokers make it impossible for scalping EAs to profit. - page 2

New comment
 
🤣🤣That should read brokers make it impossible for profit . Drop all your fancy suffix brokers and find a well established market maker , Pay up front ,All these .I .forex blah blah blah , All designed to take your cash , Retail Forex attracts degenerates and degenerates attract Charlatans ready to trick you at every turn .  Scalping done right is trouble for these guys , thus they move the goal posts . 
 
Gerard William G J B M Dinh Sy #:
Try to determine in some way that your robot is working on the correct TF
The Robot does work on the correct Timeframe. As a matter of fact, Timeframe is the very first input variable. 
 
IMRAN SALIFU:

Brokers make it impossible for scalping EAs to profit. My EA does very well on strategy tester ( with downloaded tick data from the respective brokers and the paid versions of quantum data managers ), but the Ea performs very poorly on real market ticks. 


I then decided to compare the bid and ask prices of the real market to the ones record after the day has ended ( in the download) . So I  developed another Ea that will record all the bid and ask prices that went through the live market per each one minute candle , and then I run the same EA on the strategy tester for the same 1 minute candles.


I was shocked by the results: the candles looked the same, but the bid and ask prices that went through the live market were completely different from the prices in the strategy tester for the same candle. Mind you, this happened even with paid version of quantum data magager. 


Why are the candles the same, but the price fluctuations that built the candle are very different from the real ( live market ) and strategy tester? How can I get the same data the brokers provide on the strategy tester( I need it in a live market , not when the day has ended) ? 

I have attached one of the recorded tick data for both the live market and the strategy tester after the day ended. You can see for yourself that the bid and ask prices are different in every possible way. Fortunately my EA work perfectly with the data on the strategy tester on  all the brokers I used( more that 10) but it never works on any of the data ( bid and ask prices ) provided on the live market. So I am thinking if I can tap into what ever source of data the brokers upload for us to download and use in the strategy tester, then I am good to go!! Is there anyway I can get access to that source?
 
IMRAN SALIFU:
Why are the candles the same, but the price fluctuations that built the candle are very different from the real ( live market ) and strategy tester? How can I get the same data the brokers provide on the strategy tester( I need it in a live market , not when the day has ended) ? 

If you backtest your EA with "every tick based on real ticks" mode (MT5) so it should be same (or almost to be same) as you are trading by your EA on this trading account ... other trading account (especially the situation with demo vs real/live trading accounts) may give the different result but you can backtest with other account type as well.

--------------------

There is one article about:

Testing trading strategies on real ticks

If you need the most accurate representation of history data in the strategy tester, use "Every tick based on real ticks" mode. In this mode, the tester downloads real ticks from a broker's trade server and uses them to display the price development. In case real ticks are absent for some time intervals, the tester simulates the price just like in the "Every tick" mode. Thus, if the broker has all history of the required symbols, you can perform testing of real historical data without artificial modeling.
Testing trading strategies on real ticks
Testing trading strategies on real ticks
  • 2016.08.04
  • www.mql5.com
The article provides the results of testing a simple trading strategy in three modes: "1 minute OHLC", "Every tick" and "Every tick based on real ticks" using actual historical data.
 
Sergey Golubev #:

If you backtest your EA with "every tick based on real ticks" mode (MT5) so it should be same (or almost to be same) as you are trading by your EA on this trading account ... other trading account (especially the situation with demo vs real/live trading accounts) may give the different result but you can backtest with other account type as well.

--------------------

There is one article about:

Testing trading strategies on real ticks

If you need the most accurate representation of history data in the strategy tester, use "Every tick based on real ticks" mode. In this mode, the tester downloads real ticks from a broker's trade server and uses them to display the price development. In case real ticks are absent for some time intervals, the tester simulates the price just like in the "Every tick" mode. Thus, if the broker has all history of the required symbols, you can perform testing of real historical data without artificial modeling.
I am aware of the “1 minute OHLC” and the “Every tick” options on the strategy tester. I always use “Every tick” option. 
 
IMRAN SALIFU #:
I always use “Every tick” option
It is necessary to use "Every tick based on real ticks".
Because "Every tick based on real ticks" mode is using the ticks from the broker, "Every tick" - not.

And please note: the backtesing with real/live trading account may be different if we compare with backtesting with demo trading account (because the ticks are different for demo and for real/live trading accounts).
 

Brokers are not sitting around trying to make every trader lose. There are millions of traders and systems.

In most cases, they simply decide whether your trades go to real market liquidity or a simulated (demo-like) environment, based on your strategy and performance.

In rare situations, if an expert advisor makes very high profits, there could be small execution issues or slightly different data, but this is not something that happens often.

Based on my 12 years experience, Scalping strategies are very sensitive. Even a tiny change in tick data can affect results. That’s why backtests often don’t match real trading.

And your trade result could be change even you use 90% metatrader tick data, or 99% of third-party data provider.

Change your strategy because you cannot change the market and brokers behavior :)

 
IMRAN SALIFU #:
I always use “Every tick” option.

🤦‍♂️

You're using simulated ticks in the Tester. After all of this back-and-forth in this thread, that is your problem across all broker-dealers.

Definitely follow Sergey Golubev's advice in Post #16.

 
On M1 to M5 timeframes, it's very tick sensitive, so you can get the drain quotes. The type of scalping matters here. You can't be hoping to make money fast...unless you are on some hedge fund where high frequency trading is legal.
 
Ryan L Johnson #:

🤦‍♂️

You're using simulated ticks in the Tester. After all of this back-and-forth in this thread, that is your problem across all broker-dealers.

Definitely follow Sergey Golubev's advice in Post #16.

Bear in mind, I even had to subscribe to a third party source for accurate tick data just to be sure. After testing my EA with the third party data, it gave very good results , slightly higher that the results I got from the broker’s downloaded data from the strategy tester.
 That means both the data the brokers provide for download and use on the strategy tester ; and the data provided by the third party sources are profitable with my bot.
 But when I place the bot on real market, it makes losses straight away. That is why I had to develop another EA to record real market data and compare it with the downloaded versions to see if they are the same, but unfortunately they are very different. 

So I thought to myself, if there is a source of data from which the brokers upload for us to download and use, then that particular source is very similar to the third party data source, hence it must be the real source of data, and if I get access to it, I will make my EA trade based on that “original source” and then problem solved. 
1234
New comment