For those who are not programmers but have ideas for creating EAs, "dedicated to idea traders and programmers" - page 4

Here's another stochastic - Turtles trading system

So, by the second signal we sell at 1.2695 at best, but will have to close at the next stochastic signal, if you follow your strategy, - but not the fact that the price will go down. Stochastic is a peculiar indicator. In my opinion, it cannot tell anything by itself. It is enough to look at the history. Take a walk along the chart. Everything is clear enough. It is especially problematic to use it on significant trends. For example, it can give a good sell signal and then the trend will go up so that if you don't place a stop, you will lose your deposit. But there is no way stochastic will go below 20 for another signal to trigger. It is very good to combine it with parallel channels, such as Doncian channels. It is very good at filtering out fakes.

We are talking about the same thing.
I'm not trying to pull noodles from my ears to others, I just wanted to share my experience about "guessing" trading direction, by correlation labor intensity and reliability it is the most thankless job unlike the mathematics of money management, which is always under our and only our control, it is different whether it fits our depo, trading system, entries and exits, everyone has his own MM.
I quoted the articles as I'm too lazy to write them myself. If they are interested let them look through the Turtles, they are famous people, they made money without a question, but they entered the market on a breakthrough, and the entrance to the breakthrough very, very often does not grow in the trend, as they were able to pump out the cash from most losing positions, those who except for the search direction could not think of anything else, look for an answer to this question, maybe save money and months or years of your life.

PS: And the questions HOW and WHY is desirable to ask yourself.

О5 25.

Well, OK. Let's assume that there really is " money management maths that is always under our and only our control". Could you state its essence in two words? What is the main criterion of money management?

Suppose a trader wants to make money. He sees a story on the screen. What is next? What should he do?

It's hard to talk when the subject is not accepted, let me give you that answer too?
"Suppose a trader wants to make money..."
If a trader wants to earn, he earns, and if the trader does not know how to make money, he is not a trader, at least he is not a trader yet.
Can you state in two words the essence of a trader?

About the tone, I agree. Let's calm down and try to understand.

A trader is a person who trades.
According to my understanding, if a trader wants to make money he must have a certain trading technology.
By technology of trading I understand a set of characteristics that show trader the necessity of entering or leaving the market.
Having the technology, a trader (who actually trades) is busy trying to identify these features in a certain way (with eyes or using indicator tools and Expert Advisors). I believe that this is a defining moment of trading - a trader decides about the fact of entering the market by his/her own criteria.

If his technology is such that it gives more profitable operations than unprofitable ones, then he can make money on this technology.
If the technology is bad and gives many losing trades, the trader loses.

When speaking of the probability of profit, everyone means something different. Because of this there is often a terminological confusion.
By this term, I mean the percentage value of profitable transactions in their total number (determined by results of actual trading in the account or by testing).

It is clear to me that this figure, whatever it is, cannot influence the bet amount because we cannot know beforehand which operations will prove profitable and which will not. This is why transactions cost should be determined in accordance with the maximum series of losses obtained during testing or by trading results. The bet size should be such to withstand (with some reserve, for example, without decreasing below 30% of the deposit balance) the maximum series of continuous losses.

In doing so the stake should be of a constant size (for example, 10-25% of the deposit).
If a trader allows making an aggressive investment (allows himself a big stake, for example, 70-90%) and gets into a period of a large series of losses, the deposit may not survive.
(You can read about aggressive reinvestment in more detail, with clear charts/testing results
in section 1.2 of My first "grail" article )

I cannot agree that there should be more profitable operations than profitable ones, not because of stubbornness, but from the experience of knowing the system which gave losses on 7-8 out of 10 positions and was profitable at the same time.

Another question is how to determine if the system is profitable or not. Profitability is determined not in percentage value, or in any other terms, but in absolute money terms:
(Profit for the period) - (loss for the period + fixed and variable trading costs) = +/- ?
If the resulting value is positive then the system is profitable, otherwise it is loss-making.
For example, trade gave profit 1000, losses 500, but the cost of maintenance was (terminal - 100, news - 200, satellite traffic - 400) 700 and as a result gave a loss of $ 200, it means that this system is not suitable for me at this time.

As for the percentage ratio always equal to a share of the capital it is of course difficult to argue, this is the most popular type of MM among financial managers. This approach allows to significantly improve results by taking profits into play. Other methods often require changes as the account grows, here recalculation is done automatically. This approach allows you to significantly improve your results by incorporating the profits gained into the game. On the other hand, for 7 consecutive losing trades 9 consecutive profitable trades are needed, it is well known, and therefore this method should preferably be used only on systems with clear statistical advantage of profitable trades over losing ones.
Yes, we can not know which trade will be unprofitable and which is not, but we can control the limit of losses at the current rate, for example, trade 25% of the account from 10000 account was bent to 17000, our limit, say, is already at 15000 and then we had a loss, the account went down to 15000, we reduce the trading rate from 25% to 12.5% of the account and trade on, now we have a limit loss of 13000 for example, but the limit for increasing the percentage traded fell to 17000, ie until we return to the limit of loss of the account.If this losing streak continues, we will trade with a smaller rate. Here is a very good example of how you can control the risks, ie reduce losses without any idea of the profitability of the next order and eventually lose money slower than they were earned.

As for the link to the Grail, thanks, I have enough grails written by myself. As a result I have tested the available demos online or micro lots also online as experience tells me that testers will not fulfill their duties soon enough, I decided to check it so he managed to open a sell position, put a stop loss, and use it to work and show profit (stop loss was not modified, stop profit was not) I wanted to post a screenshot and description in a neighboring thread, but I gave up that fatal business.

This reasoning is wrong.
This seemingly understandable scheme is actually a mistake.

There is no point in lowering the bet on the basis of previous losses. You can start to work, and the bet is low.
It is the other way round. If there is a winning streak, you should not raise the bet, because a series of losses can be incurred.

Here is an example of faulty reasoning:
When you flip a coin, there is a 50% chance that you will get an eagle.
If we have a series of 37 heads in a row, we can justifiably assume that (to keep the total number of heads and tails at 50%) there must be a series of 37 heads or a number of smaller series that add up to 37.

However, it does not follow from this reasoning that the probability of getting an eagle on the closest tosses increases (or changes in any way)! It always remains 50%.

Suppose we are going to flip 100 times.
We have flipped once. It is heads. What is the probability of tails on the next flip? 50%.
We have tossed 10 times. Once again all the heads have fallen out. What is the probability of tails? Same 50%.
We rolled 99 times. Again (although rare, it can happen) all heads fell out. What is the probability of tails? The same 50%.

So what's the point of limiting the evaluation of these events (assuming the dealer gives-returns $1 for a guess/error)?
The probability of the next event remains 50%.

Changing the value of the event estimate only makes sense if we have a criterion by which we can judge the change in probability of future events. For example, if a bimetallic coin is tossed, one side of which is copper (heads, for example) and the other side (tails) is bronze.
If such a coin is tossed in the air, then due to the different density of the metal, the probability of the heavier side falling down is greater.
If the same coin is tossed in a vacuum, the probability is again 50%.

The result of tossing a bimetallic coin will depend on the environment (water or vacuum). This is the factor to take into account when betting on heads or tails.

If the probability process (for a bimetal coin - approx. 50.1% vs. 49.9%) is estimated by raising or lowering the contract rate with the dealer, it will not result in anything, because the probability of flipping the coin is not dependent on the event history, but on the environment.
Again, if the stakes are too high, a series of tails can empty the wallet.

The trader's job comes down to correctly identifying the "habitat" of the currency instrument, and in no case is it to be guided by the number of recent losing trades.

This statement is not always obvious and not immediately obvious. But it is true.

----------------
To make it complete, you can add that it is possible to create an Expert Advisor, which, for example, will frequently open and close positions, and at the same time, for example, only buy. Based on the results of trading, we can easily understand that the wave is going down or up, and react to such trading by decrease or increase of rates.

However, this example cannot be considered as a case confirming the right to life of the initial statement either.
In this example, the key concept is the length of a wave that the trader assumes will last by inertia for a certain period of time.
In this case it will be a trading criterion.

If the trader does not make a mistake in estimation of the wave length, he or she can earn.

The trader's main job comes down to choosing the right criteria to make a forecast.

If every single flip of a coin is based on a 50/50 chance, what criteria are we talking about?
What data do you think you need to have in order to correctly determine the "habitat" of an instrument?
How can you estimate the "length of a wave", when no one can say exactly where the market is at any given moment of time?
If a trader makes no mistakes and gains profit, where does the profit come from? From another trader who is mistaken this time? 50/50?

PS: I am FOR finding this very "criterion", but my life experience says only one thing - the good criterion is always in the past :)
And my reasoning is correct. If I have incurred a loss I would rather decrease my profit than lose more, because the trader's code says "It is not important what your profit is, it is important what your loss is", and there is also the golden saying "Trends tend to continue more often than change direction", but trends are the same, but deposits are different, so ignoring this aspect of trading and moreover insisting on your being right is very suspicious.
Imho :)

If every single flip of a coin is based on a 50/50 chance, what criteria are we talking about?

This process is studied mathematically to a large extent. There is no criteria apart from the probability itself, so you should not bet on such a process, because the overall result will be minus the spread.

What data do you think you need to have in order to correctly determine the "habitat" of the instrument?

I believe it is necessary to have historical data and to assume some model (or several models) of market behaviour. By analysing the models, one or more defining criteria can be identified, indicating the future development of the market.

How can you estimate the "wavelength" when no one can say exactly where the market is at any given moment in time?

Saying no one is probably not fair. I think the figure is 95-99% of traders. These are traders whose educational level does not allow them to solve problems of this degree of complexity. In mathematics, however, a number of constituent issues have been solved. There is, for example, correlation analysis and pattern recognition algorithms. If we do not know all this, it does not mean that no one else does.

If a trader is not mistaken and earned, where does the profit come from? From another trader who was mistaken this time? 50/50?

The market is a closed system in the sense that money flows in it like water from one vessel into another. How much someone has won, so much others have lost. All participants in the market trade with each other. It is a market.

And my reasoning is correct. If I started losing, I would rather decrease my profit than lose more, because according to the trader's code "It is not important what your profit is, it is important what your loss is", and there is also the golden rule "Trends are more likely to continue than to change direction", but trends are the same and deposits are different, so ignoring this aspect of trading and moreover insisting on your own correctness is very suspicious. Imho :)

Wrong.
The main mistake is that you cannot tell what "losses have started" means. It is not a ball that, if it has already begun to roll down a hill, will not stop until it reaches the bottom. I can understand that we would all like to know when losses have "started". But the trouble is that when we (suddenly for some reason) decide that "losses have started" or "profits have started" the situation can be reversed.
Of course, losses must be limited. But not by reducing the value of subsequent knowingly losing orders:)