The right to publicly demonstrate the conclusions of a theoretical forum thread on a live account - 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
I take my bot and test it in "checkpoints" mode. I get16 million quid out of10 quid on the input and five years later I get 16 million quid on the output. https://www.mql5.com/ru/forum/261503/page28#comment_8015775 . And it works for all pairs and for all timeframes without any overoptimization. Then I test it in "all ticks mode" and get a plum.
In this case my tests shown above show results in"at opening prices" modefor bars on TF H4, while in"all ticks"mode the result is the same or does not have significant difference. A very large profit on tests should be alarming - this is the reason to look for gross errors in the logic of the Expert Advisor. The maximum profit should be commensurate with the maximum drawdown. There is no free money in the market. Without drawdown, there is no profit. The values of the maximum profit or maximum drawdown do not mean anything. It is important to maximize their ratio - the recovery factor - which is the most important indicator of trading process stability. Have you ever asked yourself why such fabulous, out-of-the-market results are obtained? The biggest mistake traders make is expecting super profits from the market - that's what almost 99% of traders suffer from. I do not know who or what has made them believe in the possibility of such a miracle, and this expectation is rooted not only in their minds, but also in the subconsciousness of traders and investors, which is impossible to wash from their brains. Gentlemen traders, wake up, there is and can be no super profits in an exceptionally perfect market - it is the source of all woes.
.
Theory + test report for last 5 years ( minimum ) with test proof of non-clearance from 2000 kind to present time + proof in real world for 2-3 years ( minimum ).
2 years of real is still a bit long... In 2 years the THEORY may change somewhat, and then this time will simply be lost due to incomplete confirmation...
Monthly and/or quarterly and annual trends can also be traced over this time period. The theory is based on fundamental assumptions about the very mechanism of a competitive market that no one can change. Of course, there will be changes in the outcome of real trading, but they do not have to be critical or catastrophic. ATS following a failure will more than repay the loss. Temporary changes are a wound on the body of the market. The circumstances that led to these changes should voluntarily return the market to normal with the return of the temporary loss to ATS. For example, the ATS is now suffering temporary losses due to Trump's statement about the undesirability of an interest rate hike, but, for now, the adviser is confident of continuing the former, established, mode of market functioning and continues to bend his point of view on the market. 2 years of real testing and maybe 5-10 years is necessary to prove the stability of the ATS, otherwise, the opponents will reduce everything to a fluke.
5 - 10 years is necessary to prove the stability of ATS, otherwise, the opponents will reduce everything to a fluke.
5 years of test is ok, but demo or real confirmation should coincide with the test... If the theory changes, the confirmation of the real will not coincide with the test (because of the long period of real time), and this is not a coincidence ...
And confirmation of the theory is not an indicator for some "opponents", and those who need it will believe your theory ...
5 years of test is ok, but demo or real confirmation should coincide with the test... If there are changes in theory, then there will be no coincidence between the confirmation of the real and the test (because of the long period of real), and this is not a coincidence...
And the confirmation of the THEORY is not an indicator for some "opponents", and who needs it - will believe your THEORY...
Sergey, I'm not going to make changes to the theory, let it be what it is. FV >8 - E, what better can we expect from the theory? As the proverb says, one doesn't look for the good from the bad. The main thing in the test is the daily discussion with forum members about the market situation and how an EA reacts to this situation. We will daily display the results of the test since the start of the Expert Advisor and compare them with the real trading conditions:
Tester:
Real:
In both cases the profit is around 11%. Both tester and real show that the ATS shows temporary losses. The only difference: Tester keeps saying that the absolute, it's the same as the maximum drawdown was 8.9 per cent, but actually turned out 5.6 per cent.
Monthly and/or quarterly and annual trends can also be traced over this time period. The theory is based on fundamental assumptions about the very mechanism of a competitive market that no one can change. Of course, there will be changes in the outcome of real trading, but they do not have to be critical or catastrophic. ATS following a failure will more than repay the loss. Temporary changes are a wound on the body of the market. The circumstances that led to these changes should voluntarily return the market to normal with the return of the temporary loss to ATS. For example, the ATS is now suffering temporary losses due to Trump's statement about the undesirability of an interest rate hike, but, for now, the adviser is confident of continuing the former, established, mode of market functioning and continues to bend his point of view on the market. 2 years of real testing, maybe 5-10 years is necessary to prove the stability of the ATC, otherwise, the opponents will reduce everything to chance.
To conduct a test on MT4 is meaningless. We should conduct a test on real ticks. And this can be done only on MT5.
Even if you find real ticks from somewhere, you must find out where they come from. Because every broker knows that there are different types of accounts. Their tick values are also different.
Therefore, the testing should be conducted on the account, where you are going to test your bot. Because, in the mode of real ticks MT5 downloads exactly those ticks on which MT5 is opened.
Even if the broker is the same, the same robot on different types of accounts shows different results, because different types have different trading conditions.
Is it so hard to understand ??? It's"Elementary,Watson!":)
Sergei, I'm not going to change the theory, let it be what it is. FV >8 - E, what better theory would you expect? As the proverb says, one doesn't look for the good from the bad. The main thing in the test is the daily discussion with forum members about the market situation and how an EA reacts to this situation. We will daily display the results of the test since the start of the EA and compare them with real trading:
I have an even better one without your theory, compare it on real ticks.
Inhonor of my 70th birthday, I will deliberately give up commercial secrets and make them available to all comers, in case I discover a persistent pattern that does not harm traders, while respecting the obvious ATC requirements, most important of which is not to interfere in any way with the manual operation of the ATC, no matter how illogical the EA's actions may seem to a trader.
Thank you very much for that).
By no means do I want to troll you Yusuf, but it seems that belief in fairy tales can be at any age. The words "...in the event of finding an enduring pattern..." sound like the words "...in the case of finding the grail... ".
Many traders do not want to accept the fact that trading, is a daily and hard work, which only pays off, with the right level of professionalism and realism.
The search for a magic wand (formula, pattern) leads to a world of illusions. A trader throws all his efforts into searching for the sacred and secret formula and does not take advantage of what can really be effective in trading. It does not promise rabid rewards.
If you look at modern algotrading, you can easily see and understand its condition. Modern algotrading is in the throes of a grail search. It is because of this, programs do not develop normally. They get stuck on the initial level and do not go further.
Why? - Why create effective and useful tools in a program, promising a small, but sure growth (ONLY in case of constant trade control and analysis), if there is a grail?
Especially since this toolkit asserts a rigid and realistic concept of trading. It contrasts it with a magic wand with which nothing is needed. Therefore, No - the magic wand is better!
I believe, that with the colossal possibilities to automate human routines and the huge potential, algotrading has been stagnating for years. Because most people don't want to accept reality. They don't want to work and interact with a robot.
They are looking for a magic formula, looking at trading through rose-coloured glasses.
In this case, in my tests shown above, the results in the mode"by opening prices" of bars on TF H4, while in the mode"all ticks" you get the same result or no significant difference. A very large profit on tests should be alarming - this is the reason to look for gross errors in the logic of the Expert Advisor. The maximum profit should be commensurate with the maximum drawdown. There is no free money in the market. Without drawdown, there is no profit. The values of the maximum profit or maximum drawdown do not mean anything. It is important to maximize their ratio - the recovery factor - which is the most important indicator of trading process stability. Have you ever asked yourself why such fabulous, out-of-the-market results are obtained? The biggest mistake traders make is expecting super profits from the market - that's what almost 99% of traders suffer from. I do not know who or what has made them believe in the possibility of such a miracle, and this expectation is rooted not only in their minds, but also in the subconsciousness of traders and investors, which is impossible to wash from their brains. Gentlemen traders, wake up, there is and can be no super profits in an exceptionally perfect market - it is the source of all woes.
.
There are no mistakes there (in big profits). Everything is elementary, "Watson". The lot is simply proportional to the deposit, and that gives exponential growth of the latter (in theory, of course, or in the tester, where there are no DC counteractions).
There are no mistakes there (in big profits). Everything is elementary "Watson". Just the lot is proportional to the deposit, which gives exponential growth of the latter (in theory, of course, or in the tester, where there is no DC counteraction).
I dare say you are grossly mistaken. The very fact of large profits in testing indicates the presence of gross errors in the logic of the EA, which in practice is not observed and in principle cannot be achieved in a perfect market. Try to test the EA's abilities with a constant lot.It's not that elementary, "dear Sharlock".