How is the viability of a trade idea determined?
There has been a gradation of ideas in the English part of the forum (and in English-speaking forex in general) since the turn of the century:
- raw ideas, and
- and developed ideas.
There is a post in the thread with a brief explanation - this #30
- 2019.04.09
- www.mql5.com
There has been a gradation of ideas in the English part of the forum (and in English-speaking forex in general) since the turn of the century:
- raw ideas, and
- and developed ideas.
There is a post in a branch with a brief explanation - this #30
Here is another short branch where the topic starter worries about the fact that the coder in Freelance can use his idea (and sell it) -
https://www.mql5.com/en/forum/281029 (there is also a brief explanation between raw ideas and worked out ideas).
Alternatively, when testing a promising idea there should be trades when, after the opening, the price quickly goes towards the opening and the profit rushes into the sky. Immediately, without entering the "stop" zone. And these deals should be numerous, not just one in five years. In this case, we may assume that something worthwhile has been found. Then, one may try to separate good deals from bad ones, to filter them.
The test does not necessarily have to be positive. That is the formally good system can be even lossmaking in the test because of the "preponderance" of bad trades. The main thing is that there must be some promising deals "inside" the system that can be filtered later. And it would be better if all this would happen on all instruments available for testing.
Here are the two main points - there are good deals and they are more or less evenly within the whole test period on all the instruments/periods. Even if almost all the instruments show losses or small profits and even if equity charts are ugly, it does not matter at the initial stage. There are just a lot of noisy trades, the chart chattering, sometimes accidentally taking profits and sometimes taking losses.
(During tests, on some timeframes/instruments, the equity chart may even turn out beautiful by accident. But it has happened this way, it does not mean that the system has been designed for this symbol and timeframe. It just turned out to be temporary and accidental. How many people have been misled by it, because it is in fact a "random coincidence" and they are trying to find it by combining, testing and optimizing).
There is no sense at all in measuring sharps, profit factors and being horrified or enjoying equity. It's probably impossible to come up with such "parameters" at all that allow one to understand from a test whether one has "found" or not. You cannot see a "diamond" in such a way but it is easy to trash it. Trading is for the assiduous :))) Although, I certainly understand and heartily approve the desire to outsource the routine work to the "machine".
Let's say some market pattern has been discovered, or it just seems like a pattern. The main thing is that there is a certain trading idea that you want to test.
The idea should be tested in practice, exclusively. With real money.
If it is profitable, there is a high probability that you will be pushed out of this niche.
How do you know which idea is worth developing and which can be discarded straight away?
The idea of developing a long-term profitable system can be discarded straight away. It is worth to think about becoming a professional participant, and then long-term promising ideas may open up, quite unexpected.
Nobody will tell you about them on the forum, ever. So, dive into reality and listen to those who have proven in practice their skills and ability to earn money, rather than every cross, including me ;)
P.S. This is not a topical killer, this is my opinion, which has a place as any other. You can throw around tester ideas, but they will not bring you practical benefit. And the level of damage from such practices will increase the faster the less you have real knowledge about real markets and their possibilities, especially, I stress, as a professional participant.
I'll tell you about myself. I suspect there may be other options.
Like many people here, I have tried a lot of options. And decided to focus on finding signals rather than building grids, martin and pyramids.
Further studying the indicators I came to the conclusion that there are no fish in many places. The vast majority of indicators in the terminal, kodobase and market are based on average lines. Using these indicators, one can only try to catch the period of price fluctuations in order to enter in time. But, as the period is constantly changing, the strategy working on these indicators works in a small time frame.
What I selected for myself: there are trends that last more than three to five days. This is a global trend, the price will move in their direction one way or another. There are trends from 1 to 3 days, these are pullbacks of the global trends and return to the continuation of the global trends. There are levels, these are usually round prices. There are extrema formed earlier, to which the current price approached, and which this price will try to break. Well, that's all there is to it.
I see the prospects in it and I move in this direction, but the final result is not there yet.
The prerequisites for a correct IDEA:
1. A test over a period of at least 2-3 x years yields positive final results.
2. Number of profitable transactions - not less than 75%.
After that, the strategy can be polished to improve performance...
The prerequisites for the right IDEA
Sergei, do you already have a result on your conditions? Interesting.
Sergei, do you already have a result on your conditions? I wonder.
You think it's such scary conditions for nothing...
https://www.mql5.com/ru/blogs/post/721548
- 2018.12.09
- www.mql5.com
The prerequisites for a correct IDEA:
1. A test over a period of at least 2-3 x years yields positive final results.
2. Number of profitable transactions - not less than 75%.
After that you can polish the strategy towards better performance ...
Right... 3 years on MN1
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Let's say a market pattern has been discovered, or it just seems like a pattern. The main thing is that there is a certain trading idea that you want to check. Since this forum is mainly a programmer forum, most likely, in order to check it, you will write an Expert Advisor and run it on the history. (If not, please write how you check it, I'm very interested).
So. What do you need to get after the test to get the conclusion that the strategy has a good future and that it is worth bothering with it and trying to develop it further? By what signs do you understand that it makes sense to try to improve entry points, add some filters and so on, without changing the overall concept of the trading system.
After all, surely even a profitable trading idea in the first iteration does not immediately look like a diamond. Probably at the first check, without any optimizations and adjustments of parameters, you will not see how the graph of balance and funds will go through the sky. So what do you think a test chart of a promising trading strategy should look like? What indicators do you pay attention to when assessing the quality of the system? How do you know which idea is worth developing and which can be discarded immediately?