Teach me how to make money. - page 5
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
Got it: the fish is where the dog is buried.
This is a perfect example of how analysis of outcome statistics literally "pulls by the ears" in the direction where "there are fish" ... and "the dog is buried".
There is very little data on distances to stops, but there is the MOST IMPORTANT CHARACTERISTICITY- the persistence of at least one characteristic
Well, I expected that this example would be a good illustration of your ideas.
But! Although I support the described principles, this TS (according to the charts) does not analyze or use outcomes at all.
There is just one indicator (I developed it myself). It "feels" where the fish and the dog are "buried". :)
Stops (TP and SL) are sliding, they change on every bar. That's why they are on the chart at the moment of position closing.
But everything is not perfect. I have profit in some sectors, and not in others. :(
Constancy of one characteristic in this case does not give constancy of profit.
Below are two charts. The same TS, but it is performed slightly earlier (like a backtest).
Balance (in points):
Outcomes:
As we can see, not always the indicator configured in a certain way "catches fish".
I assume that your TCs also show "consistency" only in certain areas.
If it is not so, specify, preferably with examples.
And achieved "consistency" is usually by optimizing for certain areas.
Outside the plot boundaries are usually uncertainty, a lottery.
Even if you keep TP / SL constant, the graph of outcomes will change as the market changes.
That's what I wrote at the beginning: https://www.mql5.com/ru/forum/158349/page2#1031247
Often the situation - "for today" TC statistics is good (option 1), we open by it.
For some time it gives a profit on average. But (almost certainly) the moment comes,
when it stops working. But while we will understand it, we will lose the profit or at best, break-even.
In other words, a profitable TS can "turn" into a losing one, on which one must trade "backwards".
Same with 2nd and 3rd choices.
How to determine the moment of inadequacy, "conversion"?
How do you ensure consistency of characterisation?
Well, I expected that this example would be a good illustration of your ideas.
But! Although I support the described principles, this TS (according to the charts)does not analyze or use outcomes at all.
There is just one indicator (I developed it myself). It "feels" where the fish and the dog are "buried". :)
You were in a hurry to pin your "Balance" illustration to "my ideas"...
In order to just START MOVING towards "my ideas", we will need to add the information about the distances to the stops to the existing LOSS graph. To do this, we must vertically plot NOT ONE unit of "success/failure" data, but a NUMBER of POINTS (tens of points) of success/failure. (got a 40 p. profit - moved the chart line up 40 units,... 10 points of loss - 10 units down).
And now, if you demonstrate such a graph, you can start reflections in the style of "my ideas".
For now, my cheerful exclamation about "fine example" concerns only the pleasant FACT of MAN'S HAVING such data. For some reason, people, for the most part, prefer to have a conversation "on their fingers", ... reaching, at the end of such a conversation, the appropriate logical result of "zero".
mt4trade:
How to ensure constancy of characteristic?
I suggest to look for the answer to this question, having a FULL graph of statistics of outcomes at any signal source.
In the meantime, I want to tell you that it is much easier to find a signal source, which gives a graph with a sufficient constancy of one of the parameters, than a source with a graph with a stable positive or negative slope.
You were in a hurry to pin your illustration of "Balance" to "my ideas"...
Why not!? :) The charts at least illustrate that by "lowering" stops and "increasing" takes we can bring up even a losing curve in terms of balance.
I think this is close to the idea that if the curve is horizontal with equal TP/SL, it may be raised by increasing TP/SL.
So far, without any statistics on outcomes, it's true. Of course, we will continue analyzing it.
The balance charts are exactly the same as the ones we have used in our previous analysis. That is why it exactly meets your requirements:
"we got 40 points of profit - we moved up the chart line by 40 units... 10 points of loss - 10 units down".
So, I was wrong when I said there are no stops. I was just thinking of something else - the TS is sliding, you need constant.
But the real SL / TP on the chart(at the time of closing the position). Would that help?
Why not!? :) The charts at least illustrate that by "lowering" stops and "increasing" takes we can bring up even a losing curve in terms of balance.
I think it's close to the idea that if the curve is horizontal with equal TP/SL, we can raise it by playing with TP/SL.
No, no, dear. Changing the distances to the stops CHANGES the STATISTICS of the outcomes... And is a matter for the TC.
And the basis of my whole conversation here on this thread is to USE the statistics that the TS generates.
mt4trade:
And the balance charts are precisely made IN PUNCH. Therefore exactly meet your requirements:
Well, if it's a purely TP/SL curve, then - well... we can speculate about this graph.
Do you think there are enough statistics on the chart to decide for yourself what it is: upward; horizontal... Or downtrending...?
If the TS changes the stops on the outcome statistics, the statistics themselves will also change somehow.
Accordingly, there should be an analyzer and criteria for how to change stops (in this case; and "usually" - the parameters of indicators).
And apparently we will again measure the results and change them if necessary. But it is unlikely in the live market.
Self-optimisation?Do you think there are enough statistics on the chart to decide for yourself what it is: upward; horizontal... or downward...?
If we are talking about the graph of outcomes - it is obviously (almost straight line) descending. Usually there are all sorts of curves. :)
About the "balance in pips" chart (second version in #) , then generally upward, but no guarantee that it will continue to be so.
So my TS (what's on the charts) is exactly changing stops. But based on market analysis.
If the TS changes stops based on outcome statistics, the statistics themselves will also change somehow.
Accordingly, there should be an analyzer and criteria for how to change stops (in this case; and "usually" - indicator parameters).
And apparently we will again measure the results and change them if necessary. But it is unlikely in a real market.
But my gut feeling is that, in your mind, the WORK of the trading system is the end of the whole chain of events: the TS evaluated the market situation, indicator readings, and maybe something else - and it produced an outcome (buy or sell).
You made a deal, got the outcome, put it in the table (on the chart) ... And ALL (!!!) - as I understand it, you DO NOT NEED ANY OF THIS STATISTICS FOR THE next move.
Am I right...?
And me:
- let's say the CU calculates the current situation... and decides whether to buy... or sell; (without regard to statistics)
- now, instead of immediately opening a position, I decide whether to open in the direction of the signal, or against it. (And guess what I make that decision based on)
- then, when the position is closed, in the statistics I enter NOT the result of a real trade, but the result of a VARIABLE CONNECTION between the signal given by the TS and the real price movement that has occurred. (feel the difference?)
(I was able to explain the situation clearly...?)
:) In which of my systems? :) Some are not involved in any way, because the indicators are constructed so as to give a "weighted" picture immediately. In some systems, the statistics are calculated and taken into account over the years. Some of them are recalculated on the fly 'somehow'. And some are self-optimizing.
I also tried to use an analysis similar to yours. I may have misunderstood something. That is why I am trying to understand it.
- now, instead of immediately opening a position, i decide whether to open in the direction of the signal, or against it. (And guess what is the basis for my decision).
Let's take something on the average - outcomes were positive N times, it means that "statistically" they will be negative M times. Then - "flip".?
If the signal was buy and the price went up - the result is +1? If the signal was buy, and the price went down - the result is what: zero? Or -1? Or the result is written in pips? Then how much? The delta of TP/SL and the real price movement?
Of course, there is something in it. But I do not understand how it works.
Fuck, you guys either make money or don't mess around. The grail is made in silence. It doesn't make noise.