Avalanche - page 254

 
JonKatana >>:

Итог - вы лжете. Это ставит под сомнение все остальные ваши сообщения.


And you refute my posts. Proofs in the studio otherwise you are lying. And prove that with the same price of a pip on MT4 and MT5, you will only close losses on MT4 on one side, i.e. the losses will be less than on MT5. These are your unsubstantiated statements . Bullshit no more . You can't prove it. Hence you are lying.
 
E_mc2 && Mathemat - guys. take my word for it... I've been seriously and long on the subject of louboucher. i may even have some of my EAs saved somewhere.
louboucher is much less resistant to the market than a classic martin. Distribution in series is critical for a lubucher.
Series of several looses, then one profit, another moose, another profit, and several looses again (standard market behaviour, nothing contrived).
Martin may stand, but if you use another profit and one more moose, coefficient will increase exponentially... I am not a mathematician. Maybe Mathemat is able to formalize these facts. But it is really so. Empirically deduced.
So the theme of the laboucher (at least in relation to forex) is even less promising than the standard martin.
 
And the author is on fire as always:) I'm reading this thread at bedtime... I get a mountain of positive emotions:)))
Prove otherwise it's a lie - it's already a proverbial one:))) And he is shown his own posts - he does not notice them... As if they never existed.
The standard answer - prove otherwise it is a lie, but he is not going to prove anything. His assertions must be accepted as axiomatic and beyond question :)

It was especially funny when he wrote about big banks and corporations... I laughed at those posts... And now you're doing the standard: I didn't say that, prove it, otherwise it's a lie...

O-lo-lo


I wonder if the author's last name is Gil by any chance...
 
The purpose of this study is firstly to see what the lots might be like under near-fighting conditions.
Secondly, to see what time it takes to get out of a drawdown. Here, by the way, is an interesting example (base probability of profit is 0.5 - yes, 0.5, no less; first column is volume, second column is trade result):

1 -1
2 -1
3 -1
4 -1
5 1
5 -1
7 -1
9 -1
11 1
10 -1
13 -1
16 -1
19 -1
22 -1
25 -1
28 -1
31 1
30 -1
35 -1
40 1
37 -1
44 -1
51 1
47 -1
57 -1
67 -1
77 -1
87 1
80 -1
93 -1
106 1
96 -1
112 -1
128 -1
144 1
131 -1
150 1
134 -1
156 -1
178 -1
200 -1
222 -1
244 -1
266 1
247 1
230 1
215 -1
252 1
225 1
213 1
201 -1
268 -1
335 -1
402 -1
469 -1
536 -1
603 1
549 1
498 -1
594 -1
690 1
610 -1
722 1
632 1
603 -1
804 1
670 1
671 -1
1006 1

There are 69 numbers in this "completed" series. Ratio of profitable to unprofitable - just 1 to 2, i.e. 33.3% profitable. The series is successfully completed - but what a lot! The reason is that there was a local rather strong decrease of profitable numbers, and it was not so exceptional. But throughout the series, we didn't get a single chance to "complete" the series. Another example from the same sequence, even worse (in these 10,000 trades it was a record-breaker ):

1 -1
2 -1
3 -1
4 -1
5 1
5 -1
7 1
8 -1
11 -1
14 -1
17 -1
20 -1
23 1
25 -1
33 1
28 -1
39 1
31 -1
45 1
46 -1
63 -1
80 -1
97 -1
114 -1
131 -1
148 1
160 -1
206 -1
252 -1
298 -1
344 1
315 -1
378 -1
441 1
395 -1
475 -1
555 -1
635 -1
715 -1
795 -1
875 -1
955 -1
1035 -1
1115 1
1052 1
989 1
955 -1
1115 -1
1275 1
1161 1
1047 1
1030 -1
1345 1
1110 -1
1505 1
1190 -1
1665 1
1270 -1
1825 1
1350 1

The maximum lot is even bigger, although the series is slightly shorter. There are 20 profitable, 40 losing trades in this series.
The most remarkable thing about these series is that all the time this series we are in a loss - and the closer to the end of the series, the greater the losses and balance fluctuations.

Oleg, these are not rare exceptions. If we take into account that in systems with martin it is necessary to make a lot of deals - because the deposit growth is small per deal (small initial lot!), then it turns out that the statistics of thousands of deals is easily accumulated over a year or two. And in this statistics such a crushing series is inevitable.

The main question is: what to do?

P.S. I will not post the code yet because I still have errors. But in this series, they do not seem to be visible.
 
JonKatana >>:

Об этом я писал - читайте внимательнее.


A link to what you wrote. Otherwise you're lying, and you didn't write anything.

Two cheeks in MT5 is cool. That's even more margin than with partial compensation))) Two different accounts will take margin for each open lot at full)))) I can imagine when they run out of free funds)))

That is, you open a trade at 0.1 lot in one account, and place a 0.2 lot order on the opposite side of the channel at 40p in the other account )) If this works, on one account there will be 0.1 lot, and on the other one 0.2 lot. Of course they took a margin in both places. And the spread, too. That is, as for 0.3 lots. You look and see. In one account the 0.1 lot is a loss. And 0.2 lot is in profit. The price passed 40 pips from the border of the channel... The first lot will be -80 pips. And the second lot from the border of the channel will be +40 pips.
Let's calculate. Lot 0.1 on the first account at a loss of 80 pips = 80 quid.
Lot 0.2 from the channel boundary will be +40 pips. That's 80 quid. We have a b\u.


The same on one account.
Open 0.1. We got a 0.2 pending on the border. The pause is triggered. The first lot takes out a stop loss of 40 pips. The lot 0.1 is opened, the price passes 40 pips. We have on the first one - 40, on the second one + 40 pips. Exactly the same way we have boo.

Difference. In the first case, the idiot Catala paid the margin for 0.3 lots. And the spread is also 0.3 lots. In the second case, you paid margin for only 0.1 lot. The spread is also for 0.1 lot. The difference even in the first step is three times. The result for the deposit is the same b\u)))) But not exactly boo. You paid the spread for two accounts as for 0.3 lot. That is 0.2 lots more. Therefore, the amount of this spread in two accounts will be less. Even if orders are closed in B/B))). Only at the expense of the spread will be lost. Voprosy...when Opolzny on two accounts in MT5 not enough deposit and will be plum)
What an idiot.
 
E_mc2 - calm down:) This creature is from Sirius. It's already been understood by everyone and no one takes his remarks seriously any more... The creature has learned a few phrases... "prove otherwise it's a lie", "read carefully" and a few others... ...and pokes them in and out... I'm reading this thread purely as a joke. It's been a long time since other threads have been threaded here. Just topics that aren't worthy of a separate thread... His eyes are triangular, he has three ears, eight toes on false legs, and his skin is mottled violet... typical Siriusian.
 
Mathemat >>:
Цель этого исследования - во-первых, увидеть, какими могли бы быть лоты в условиях, близких к боевым.
Во-вторых, посмотреть, какое время требуется для выхода из просадки. Вот, кстати, интересный пример (базовая вероятность прибыльной равна 0.5 - да, 0.5, не меньше; первый столбец - объем, второй столбец - результат сделки):

1 -1
2 -1
3 -1
4 -1
5 1
5 -1
7 -1
9 -1
11 1
10 -1
13 -1
16 -1
19 -1
22 -1
25 -1
28 -1
31 1
30 -1
35 -1
40 1
37 -1
44 -1
51 1
47 -1
57 -1
67 -1
77 -1
87 1
80 -1
93 -1
106 1
96 -1
112 -1
128 -1
144 1
131 -1
150 1
134 -1
156 -1
178 -1
200 -1
222 -1
244 -1
266 1
247 1
230 1
215 -1
252 1
225 1
213 1
201 -1
268 -1
335 -1
402 -1
469 -1
536 -1
603 1
549 1
498 -1
594 -1
690 1
610 -1
722 1
632 1
603 -1
804 1
670 1
671 -1
1006 1

В этой "завершенной" серии 69 чисел. Соотношение прибыльных к убыточным - как раз 1 к 2, т.е. 33.3% прибыльных. Серия успешно завершена - но каков лот! Причина - в том, что произошло локальное довольно сильное уменьшение частоты прибыльных, и оно не было таким уж исключительным. Но на протяжении всей серии у нас не было ни одного шанса "завершить" серию. Еще пример из той же последовательности, еще хуже (в этих 10000 сделок это был рекордсмен ):

1 -1
2 -1
3 -1
4 -1
5 1
5 -1
7 1
8 -1
11 -1
14 -1
17 -1
20 -1
23 1
25 -1
33 1
28 -1
39 1
31 -1
45 1
46 -1
63 -1
80 -1
97 -1
114 -1
131 -1
148 1
160 -1
206 -1
252 -1
298 -1
344 1
315 -1
378 -1
441 1
395 -1
475 -1
555 -1
635 -1
715 -1
795 -1
875 -1
955 -1
1035 -1
1115 1
1052 1
989 1
955 -1
1115 -1
1275 1
1161 1
1047 1
1030 -1
1345 1
1110 -1
1505 1
1190 -1
1665 1
1270 -1
1825 1
1350 1

Максимальный лот еще больше, хотя серия немного короче. Прибыльных в этой серии 20, убыточных 40.
Самое примечательное в этих сериях то, что все время этой серии мы находимся в убытке - и чем ближе к концу серии, тем больше убытки и колебания баланса.

Олег, это не редкие исключения. Если учесть, что в системах с мартином надо делать много сделок - так как рост депозита небольшой за сделку (малый начальный лот!), то получается, что статистика в тысячи сделок легко набирается за год-два. И в такой статистике неизбежны такие сокрушительные серии.

Главный вопрос: а что делать-то?

P.S. Код пока не выкладываю, т.к. еще есть ошибки. Но в этих сериях их, похоже, не видно.


Well, what to do... You have to have a TS with 40% of trades. You're the one who calculated at 33% profit trades. 33% for a Labouchere, that's just going to zero. And if you stuff this series with 40% profit trades. Wouldn't that make a difference? That's one. And two. Profit should be at least a little bit bigger than loss. Nobody makes us to stand till the end of the series if we've gone in the profit. It's quite possible that with 40% of profitable trades and with a profit greater than the loss, we'll be able to leave the series with profit or even in O before the series finishes. In your example it's not possible.
I don't think the drawdown would be big. Well, relatively big... I mean, every couple of trades, we make a profit. And a very decent lot. I mean, it won't be the same as if we were standing on a classic martin. After all, we'll fix profits, and that should reduce the drawdown...

I am interested in the question about the ratio of stop to loss. This is how it will change the layout? If the profit is larger than the loss even by 1 pip. It means that it raises % of profitable deals. That is, this extra pip will be with the number of trades of say 20. At take in 20 pips. It will bring in 20 pips. This is like an extra profitable deal. It will improve profit/loss ratio.
 
Oleg, I wrote in the very beginning of the post: probability of profitable trade is 0.5, i.e. TS with 50% profitable trades. Those were just two pieces of a total series of 10,000 trades. The chunks are quite revealing.
 
Mathemat >>:
... Это были просто два кусочка из общей серии в 10000 сделок. Кусочки вполне показательные.

I'm getting the same thing. I am trying to generate trades in different ways, but I cannot get rid of failed series.

Bernoulli gets everywhere. :)

I'm not generating with mclw random function, but based on real quotes, setting all sorts of opening conditions. Closing by SL/TP (they are equal).

Preliminary conclusion: classic Martin and LaBoucher are different ends of the same cudgel called boomerang.

lexandros is right in his conclusions.

 
lexandros >>:
E_mc2 && Mathemat - парни. поверьте на слово... долго и серьезно мусолил тему лябушера. даже возможно где то сохранились советники.
Лябушер - гораздо менее устойчив к рынку чем классический мартин. Для лябушера критически важно распределение в серии.
Серия из нескольких лосей потом один профит, еще один лось, еще один профит, и опять несколько лосей (стандартное поведение рынка, ничего надуманного). Мартин выдержит, лябушер - стопроцентно приведет к коле при таком раскладе
Ибо по лябушеру, после профита, мы удаляем крайние числа в последовательности, на следуюшей серии лосей - увеличиваемся значительно быстрее, один профит, и опять лось - коеффициент увеличения вырастает по экспоненте... Я не математик. Возможно Mathemat сумеет формализовать эти факты. Но это дейтсвительно так. выведено эмпирическим путем.
Так что тема лябушера (по крайней мере по отношению к форекс) - еще менее перспективна, чем стандартный мартин

lexandros,

I don't think it's correct to say so, without reference to the TC. I think if the TS is more often right than wrong this method is better than classic. How many knees will classic martin with $2000 deposit and 0.01 lot hold? 12-15? This one is supposed to hold longer. Well, I will have to write a full-fledged Expert Advisor. Once again, the TS itself is important in this method (not in classic one).