Martingale again - 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
Good thinking comrade!!!
It's either a small profit but with an almost 100% guarantee of not being drained...
Or 50-100% per month, but with the risk of losing 1-2 times a year. By the way, summer is ahead, and summer is most often a flat area, i.e. the best for this owl.
By requirements to the deposit it is quite reasonable for martin.
I'm 100% positive about splitting a lot into smaller ones. I'm using it myself when working with grid drawers. Although now I prefer to work and test on PAMM account from A****ri because min lot there is 0.01 and max 1000 (very convenient for testing), although the leverage is 1:100, but also quite enough.
As for the exits, I now measured on the chart ... well, even in the screenshot with the last eurik sale exits from purchases were not even in B/B but in + (at first glance they seemed unprofitable).
I also wonder what caused the exiting of positions by this line
I.e. it's a real improvement or more to calm the nerves 8-))
I do not know - rather to calm the nerves - as with this line Owls earn less BUT! often happens that (for example) first sold one order - the price went against - bought one order - the price went against again (that is again down) - sold again (sold two already) -
further gep down - owls in the plus on all orders and closes all - i.e. hung buy was covered by villages and closed with all in place - it does not give the owl to merge (a kind of fuse) - although I am not sure of its necessity - but BE IT ....
you're welcome ... i have an idea but i do not know how to do it - to close out of all orders (one direction) only the last and penultimate (of course at profit) - of course it is not difficult to "cut off" the last two orders,
only if the lot is complex and "one" order is made up of three - this is where classical solutions will not help .... I'm still "scratching my head" ....
I'm sorry, but I do not quite understand the meaning of this approach.
1. If we have closed, and the price continues to move a bit further to the required direction, we could have closed completely, but we are still stuck.
2. If we have closed at the peak and the price has gone again in a bad direction for us, after some time we open another position with a larger lot on the martin. And now we have the first position hanging and somewhere far from it a new one and we need a good bounce to close it in profit (let's say we had 3 levels open and closed 2, we still have 1 initial entry + we have another one opened and now we need to close them both in profit).
Although, yes, maybe it will help, because the further the price goes, the higher the probability of a pullback and we expect it with fewer open levels and with less potential drawdown and more room for manoeuvre. But in general it should be tested on 2008 and 2009, when the markets were strongly volatile.
ALEX_SPB_RU:
I apologise, but I don't quite understand the point of this approach.
1. If we have closed and the price continues to move a little more in the direction we want, we could have closed completely, but we are still sitting on the doldrums.
2. If we have closed at the peak and the price has gone again in a bad direction for us, after some time we open another position with a larger lot on the martin. And now we have the first pose and somewhere far from it a new one and to close it we need a good bounce (for example, we had 3 levels open and closed 2, we still have 1 initial entry + we have opened one more and now we need to close them both in +).
Although, yes, maybe it will help, because the farther the price goes, the higher the probability of a pullback and we expect it with fewer open levels and less potential drawdown and more room for manoeuvre. But in general, this should be tested at the time of market turbulence in 2008 and 2009.
I used to trade with my hands like this - the first two last lots close with good profits
secondly, it reduces the probability of drawdown, and drawdowns may be withheld longer
Third, (for example) you already have a grid of 4 orders - close the last two with profit - 2 orders are left - open a third one - and then close the last two again - one order is left - open a second one - and close it with profit ..... orders are over ....
Both profit will be bigger and less chance of losing ....
like on the screenshot - only with hands ... (bai)
Today, at first I thought, "Bitch, I closed early,"
and now I don't know if the owls are right. ....
I traded by hand this way - first, the last two lots close quite a good profit
Secondly, it reduces the probability of losses, and drawdowns can be withheld for a long time.
Thirdly, (example) you already have a grid of 4 orders - close the last two orders in the plus - 2 are left - open a third - and then close the last two again - one is left - open a second - and close in the plus .... orders are over ....
Both profit will be bigger and less chance of losing ....
like on the screenshot - only with hands ... (bai)
I agree, but on the top screenshot we would have closed all positions in about an hour already at the high of the chart.
I do not dispute that there are certain positive aspects of this approach. But in practice, you can check this only in the tester; otherwise it is difficult to calculate.
By the way, if necessary I can send you the indicator that calculates the maximum failsafe trend at a given level of slippage.
I.e., you can find the longest failsafe trends in the file in Excel, filtering them in descending order and see the dates where it happened and immediately find this segment on the chart and run your EA on it...
I agree, but in the top screenshot we would have closed all positions on May 15 in about an hour already at the maximum of the chart.
I do not argue that there are certain positive aspects of this approach. But in practice, you can check this only in the tester; otherwise it is difficult to calculate.
By the way, if you need, I can send you the indicator that calculates the maximum retracement trend at a given level of slippage.
It's interesting - leave it in your personal message - we'll think ...
But I will have to add settings again - but I would not like to .....
We need a signalling part with as few settings as possible!
if you have any ideas, please contact us and we will check .....
As for the problem of how to extract the last 2 positions if they are open with multiple lots, the first thing that came to mind is the following two options.
Suppose we have open 3 positions of the grid level (the new lot increases 4 times for simplicity) and the first level lot will be 3, the second 10+2, and the third 10+10+10+10+8.
We want to close 2 positions.
Option 1 We are looking for all lots with the needed for us Medjic and look which one has the minimum opening price (now we want to close the buy). We find it and close it, remembering its open price. Again go through lots and look for the price that is no more than the specified delta (well, let's say somewhere around 10-20 points by the old ones). If we have found it, we close it and keep looking until all orders that fit in the delta and with the given slower price have been closed (in our example, there will be 5 orders, 10+10+10+10+8. Ok, we have already closed one position. Now we can again look for the order with the lowest open price and close it and look for more orders within the delta. When everything has been processed, we have closed two positions. The disadvantage is the necessity to correctly estimate what delta should be set because suddenly, when a fragmented opening is triggered by news and the orders are too scattered.
Variant 2. If we take into consideration that the orders are going in ascending order, we should first close the oldest one among them + all nearest ones that are equal to the maximum possible order to be opened, i.e. in our instance, we have closed 8 and then 10+10+10+10. All one position is closed... Now, we again close the oldest one from the series of orders with our Medjik and all its others preceding it with the maximal lot. i.e. 2 and then 10. Everything is closed. The disadvantage is that if the previous order is a multiple of 10, i.e. we do not have another slice (for example, we had 10+10 and now we opened another 10+10+10+10....8), but it is unlikely.
As for the problem of how to extract the last 2 positions if they are open with multiple lots, the first thing that came to mind is the following two options.
Suppose we have open 3 positions of the grid level (the new lot increases 4 times for simplicity) and the first level lot will be 3, the second 10+2, and the third 10+10+10+10+8.
We want to close 2 positions.
Option 1 We are looking for all lots with the needed for us Medjic and look which one has the minimum opening price (now we want to close the buy). We find it and close it, remembering its open price. Again go through the lots and look for the price that is no more than the specified delta (well, let's say somewhere around 10-20 points by the old ones). If we have found it, we close it and keep looking until all orders that fit in the delta and have a given slower price have been closed (in our example, there will be 5 orders, 10+10+10+10+8. Ok, we have already closed one position. Now we can again look for the order with the lowest open price and close it and look for more orders within the delta. When everything has been processed, we have closed two positions. The disadvantage is the necessity to correctly estimate what delta should be set because suddenly, when a fragmented opening is triggered by news and the orders are too scattered.
2 Variant. If we take into consideration that the orders are going in ascending order, we should first close the oldest one among them + all nearest ones that are equal to the maximum possible order to be opened, i.e. in our instance, we have closed 8 and then 10+10+10+10. All one position is closed... Now, we again close the oldest one from the series of orders with our Medjik and all its others preceding it with the maximal lot. i.e. 2 and then 10. Everything is closed. The disadvantage - if the previous order is a multiple of 10, ie, it does not have another piece (for example, was 10 + 10 and now open another 10 + 10 + 10 + 10 ....8) But it is unlikely.
the first option is quite acceptable but there is a certain value "delta" ... we need to look for it ... and you can't do it in the tester because it will practically be equal to zero ... (the tester will place 200 orders in a fraction of a second)
the second option was not clear (I intuitively did not like it) ....
i have the following option - buffer the order tickers and the opening price during the order placing (to be picked up during restart of the terminal, write them into a file) ...
two two-dimensional arrays will be needed for one direction (the last and the last but one trade operation)
follow orders written to arrays ....
I have an idea, but I don't know how to do it, since I am self-taught in programming and learn everything by "scientific method" ...
By the way - you can close the first and the last one - it's also cool ...
the first option is quite acceptable but there is a certain value "delta" ... you have to look for it... and you can't do it in the tester because it will be almost zero ... (the tester will place 200 orders in a fraction of a second)
the second variant was not understood (I intuitively did not like it) ....
i have the following option - buffer the order tickers and the opening price during the order placing (to be picked up during restart of the terminal, write them into a file) ...
two two-dimensional arrays will be needed for one direction (the last and the last but one trade operation)
follow orders written to arrays ....
I have an idea, but I don't know how to do it, since I am self-taught in programming, and learn everything by "scientific method" ...
As for the array, it is the first thing that came to my mind.
But on the other hand, the list of open orders in the terminal is an ordered array with a given mode, traded symbol and order type, so we can select everything we need... Of course, during the first selection they can be gathered in an array to speed up data processing without requesting them from the terminal.
As for the delta - you won't find it in the tester, even if testing at all ticks. Therefore, it has to be not too small and not more than maximal distance between opening levels.
To be honest... I don't quite understand your implementation of arrays.
I'll help you with what I can... let's have a concrete discussion in private.
It was the first thing that came to mind about the array.
But on the other hand, the list of open orders in the terminal is already an ordered array from which we can select everything we need by a given mode, traded symbol and order type... Of course, during the first selection they can be gathered in an array to speed up data processing without requesting them from the terminal.
As for the delta - you won't find it in the tester, even if testing at all ticks. Therefore, it has to be not too small and not more than maximal distance between opening levels.
To be honest... I don't quite understand your implementation of arrays.
I'll help you with what I can... let's have a concrete discussion in private.
ok