The returns for the last nine months! :) - page 6
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Lot or what timeframe does not matter
The point is that the young topicstarter does not understand that he is fitting his results to the GENERATED quotes of the tester.
Counting ticks and working with tester ticks is the height of stupidity. So the topic in this form doesn't deserve a bit of attention.
Max47, your tests in their current form are om p o m a n.
You need to look for tick data of real quotes and conduct experiments on them.
Absolutely! :) But only after testing on the demo!
If we compare the number of trades with the number of trading days and take into account the IRR = 2.2 pips, then the conclusion about pipsing is clear. Usually they are pipsing on lower TFs, but not on dailies. The modeling inside the daily bar may be the reason of euphoria.
... But there is no problem with periods, you can choose any... It doesn't care, because it counts ticks! :)
Ticks are modeled in the tester.
But even if we assume the absence of technical tester, the MOJ in 2p reduces the expected profit to the loss (slippage, requotes...).
And I was passing through, using Lucky
In the tester I was gambling, on the demo account I was making 100% of deposit per session, but alas, now I have to adjust again to the ticks density and the price speed - it's not interesting anymore
Yes, very good, Maxim. Can you reduce the testing period (not days, but, say, hours)?
It's just not really clear whether the tester can be trusted in this case - with the number of trades exceeding the number of bars by a factor of about 3. The tester uses bar modelling. Do you understand what this is?
When tested on any period, the same result is the same. Because the Expert Advisor uses M1 anyway.
What is bar modelling?
I understood that it takes the opening and closing prices for a specified period and forms a bar by them... and also during testing of the formed bars it uses only the opening and closing prices and does not use the prices inside the bar! Right? )
Max747:
What is bar modelling?
Ticks not bars.
We take the youngest available (downloaded) TF, M1 if there is one, and inside it the ticks are modelled (generated) by the tester according to the algorithm conceived and known only to MT4 developers. For scalpers the difference between simulated and real ticks can be considerable.
... uses only open and close prices and does not use prices within a bar! Right? )
Wrong. Two more prices are used: high and low, which in most cases are just inside the bar.
Ticks, not bars.
The youngest available (downloaded) TF is taken, M1 if there is one, and inside it the ticks are simulated (generated) by the tester according to the algorithm conceived and known only to MT4 developers. For scalpers the difference between simulated and real ticks may be significant.
Wrong. Two more prices are used: high and low, which in most cases are just inside a bar.
So I'm waiting for the next working week to be convinced... I managed to finish the Expert Advisor only on Saturday at 3 am)! So I have not tried it on demo yet!
Let's see what will happen on real ticks ...
Here's another visit to the branch...(https://forum.mql4.com/ru/4773) I don't think it's that bad!
Max, the most important thing is a low trade expectation, of the order of the spread. A system is said to have a chance of being stable if the m.o. is of the order of a few spreads.
About the quality of modelling... no, that doesn't seem to be the problem. You have to look at how your trades are done. If they are too "fast", then it makes sense to think twice.
Max, the most important thing is a low trade expectation, of the order of the spread. The system is considered to have chances for stability if m.o.s. is of the order of several spreads.
About the quality of modelling... No, that doesn't seem to be the problem. You have to look at how your trades are done. If they're too "fast", it's worth thinking about.
May I ask more details about mathematical expectation with spreads, how to calculate it? )
Deals are made with a period of at least 2 minutes, but it can be much longer days, weeks (it varies)!