I'm puzzled... EAs don't work on real accounts... maybe we should change broker? - page 5

 

Yes. There are bids and aska's there. and even my DC's. And I can now see what really happened and what quotes came in. By the way, I have noticed on ducas that there are very rarely ticks more than one tick per 0.1 second.

 
Reactor555:

I don't really calculate anything in the owl. I only modify orders and place new ones. Sometimes there are situations when several orders need to be modified at once. In this case there are some time delays, because it takes time to modify an order.

I set a simaphore myself and new ticks are not processed if the old ones have not been processed yet.

Admittedly, I still don't fully understand what happened that day on the second situation. Having searched more thoroughly in the logs I found out that EA has sent commands to modify ticks, but all of them were rejected. Taking into account the fact that the distance between all of them was significant, I do not understand why they all were rejected. The only explanation can only be an increase in spread at the moment of news movement. So for the future I decided to just switch off trading at the time of news. And the only thing left to do is to pray that the spread will not just increase for no reason at all.

Probably, it was not spreads that increased, but distance to stops ("stop level"), and the updated take-profits did not fall within the required range.

 

All this suggests that on strong news it is better to shut down trading. Yesterday's canadian was once again an eloquent demonstration of this.

 
Reactor555:

Played around with TDS for a while. Unfortunately, it does not give bid-ask data. It simulates situations with floating spread, and allows to set an arbitrary spread, which by the way is also very useful.

I can't understand what is Spread multiplier and Spread addition in Spread settings.

If I understood correctly, then Minimum and Maximum Spread should be set not to 1 and 2, but to 10 and 20... right?

Forget that nonsense, it won't give exact tick data anyway, because you download it from one provider and trade on another. Use MT5, everything is already there, time to forget about mt4
 
Maxim Dmitrievsky:
I don't have any idea what to do with this nonsense, it's not going to give exact tick data because you download it from one provider and trade on the other. Use MT5, you have everything there already, it's time to forget about mt4

Already started learning MT5.

But really ...

All our history tests are nothing more than complacency. Writing an owl and fitting it to history is not that difficult. But the truth is that income in the past does not guarantee income in the future. It's impossible to foresee everything. My goal is for the owl to live for 2.5-3 months. During that time she should do 100%. On tests over a year and a half it shows a minimum of 3200% on one pair with a maximum drawdown of 25%. And I have 10 of these pairs in the works and I plan to add more. But in fact the first experience ended on the 20th day with the loss. The idea that I earned 40% in less than 20 days almost does not warm me up.

 
Reactor555:

Already started learning MT5.

But really ...

All our history tests are nothing more than complacency. Writing an owl and fitting it to history is not that difficult. But the truth is that income in the past does not guarantee income in the future. It's impossible to foresee everything. My goal is for the owl to live for 2.5-3 months. During that time she should do 100%. On tests over a year and a half it shows a minimum of 3200% on one pair with a maximum drawdown of 25%. And I have 10 of these pairs in the works and I plan to add more. But in fact the first experience ended on the 20th day with the loss. The idea that I have earned 40% in less than 20 days is hardly warming.


If I understand the strategy is based on tick data, it cannot be tested in MT4, it's a tester's grail. If you test it with real ticks in mt5, then we'll be able to talk about something. Add slippages that are always present in real trading
 
Maxim Dmitrievsky:

In mt4, these strategies cannot be tested at all, they are testering graals) Load them with ticks or not. If you test it with real ticks in mt5, then we'll be able to talk about something. Add slippages that are always present in real trading

I'll be sure to check it out. Turns out that mt4 can't just be recompiled into mt5. But I will check it for sure. I will only finish everything in mt4, so as not to get caught in spread widening, I will try it once again on real account. I will try again on real account and at the same time I will start transferring everything to MT5.

By the way, no, not on tick data. This is not scalping. Plus or minus a few pips does not threaten me. It's the widening of spreads by a factor of 3 or more that threatens me.

Slippage is not for me. My trades are all against the market.

 

A converted Briton from 27 March to date. On tests it's just a fairy tale. Maximum drawdown on 0.01 lots is only $60 at 1:500.

Files:
GBPUSD.jpg  50 kb
 
Reactor555:

A converted Briton from 27 March to date. On tests it's just a fairy tale. The maximum drawdown on 0.01 lots is only $60 at 1:500.


This will not work on the real, I tell you right away, do not waste your time )

 
Maxim Dmitrievsky:

It won't work in real life, I'm telling you right now, don't waste your time.)


Don't get your hopes up. Tell me why)))