Avalanche - page 210

 
Svinozavr >>:

Да не. Он даже арифметику не знает. Не бывает таких тупых ботов. Это, все-таки, человек!!!
Куда там до теста теста Тьюринга. Сложение-вычитание бы освоить.


Hehe... no... he knows the maths... but it's selective... When the calculations go in his favour, he does it right... and when it's not in his favor, the maths from Jupiter kicks in...
So if there's a bot here, it's very selective... Bot programmers should use this thread as a teaching aid:)
 
E_mc2 >>:



If, as you say, the flat is followed by an avalanche in a special way, then, thinking like that, you can kill any system, not just an avalanche. About the swaps - I agree.

 
d22 >>:

если, как ты говоришь, флет как спецом будет за лавиной следовать, то, рассуждая так, можно убить любую систему,не только лавину. На счет свопов - согласен.


You're not quite right... You're not quite right... Definitely any TS works better either on a trend or on a flat. Whoever learns how to clearly detect the transition from a flat to a trend - think you've found the grail...
The question is that if for normal TS - a loss on a trend or a flat - only a normal loss (which absolutely any TS provides) and this unsuccessful period is replaced by a successful one, and there is a recovery period and further profit.
This is the fundamental difference between martin-like strategies. For them, one single bad period is tantamount to a complete collapse and total loss of the deposit.
 
As an analogy... Get in a car. accelerate to 5 km/h drive at this constant speed... in 500 metres a concrete wall....... Effect? Broken bumper, crumpled fenders, etc... you go to a repair shop - fix it all and you can repeat it endlessly until you get bored...
Sit in the same car and constantly press the pedal and drive along the same trajectory... the same wall ahead... Effect? The car is unrepairable... and you're probably dead...
Conclusion... a 500m section you will certainly pass many times faster (similar to a profit build-up) but the finale will be sad.
So... in any martingale... including an avalanche - building up lots is like building up the speed of a car... there's a certain limit where you can't stop the car... and stopping it is simply a disaster without the right to survive.
 
d22 >>:

если, как ты говоришь, флет как спецом будет за лавиной следовать, то, рассуждая так, можно убить любую систему,не только лавину. На счет свопов - согласен.


I'm not saying it will follow. The point is, how do you know if it's over? And it's time to place an order? You know, there's such a thing as an expanding triangle. That's the main thing. If you just know when a flat ended and a trend started. Then we don't need Avalanche at all. If you know when the flat ends and the trend starts, you'll never even get into these boats. The author's phrase is simple - when the flat ends and the trend starts, so if I know it, why mess with a landslide? How do you know when the flat ended and it's time to place orders? And if you place an order, but the flat isn't over - why do you go back to the lock? When will it end? Well, if you think about it that way we can say - when the trend is over, let's stop trading in an avalanche! When is it over? - So the trend is over, wait for the flop, the flop is over, start trading... you guys are simple... or not simple...
 
JonKatana >>:

Хоть на десятом. Вы просто после третьего разворота блокируете "Лавину" и пока цена не отойдет от внешней границы коридора хотя бы его ширину или две ширины - новый коридор не создаете.


She stepped back a width (or two) and entered the corridor again, then stepped back a width again and entered again.... and again. What should I do in such a case?
 
Xadviser писал(а) >>

Step back a width (or two) and enter the corridor again, then step back a width again and enter again.... and again. What to do in such a case?

That's how it should be :)
Katana is now preparing two publications... when she went in and didn't go out, you take LinexCutor
and when it's out, the avalanche... it's obvious you didn't read the topic first, otherwise there would have been fewer
serious questions
 
I support the lock option.
Colleagues have posted an EA and it would be a good idea to add a parameter specifying the order (knee) at which we enter the lock.
Once we have entered the lot and the market has moved to some sufficient range, we do not create a new one, we open it by the principle of an avalanche. Maybe it will work.
If after the shift, the price remains in the corridor, then we restore the lock and wait for the next shift.
 
torgash >>:
Поддерживаю вариант с локом.
Коллеги выкладывали советник, к нему неплохо было прикрутить параметр с указанием порядка (колена) на котором мы входим в лок.
После того как зашли в лок и рынок сместился на некий достаточный диапазон - раскурчиваем лок по принципу лавины НЕ СОЗДАЕМ новую, А РАЗЛОЧИВАЕМСЯ по принципу лавины. Возможно будет толк.
Если же после смещения цена остается в коридоре то восстанавливаем лок и ждем следующего смещения.


Only two(!) such transitions are enough to increase the bet by a factor of 1000 according to the specified principle (on the third corridor)
What does it mean to unlock on the principle of an avalanche? Closing positions by doubling them at every step?
 
Xadviser >>:


...
Что значит разлочиваться по принципу лавины? Закрывать позиции их удваивая на каждом шаге?

Detach orders according to the same principle as for entry. On the upper border we cut off one 1 sell, went to the lower one - cut off 2 buy, etc. Generally, the tests show that two adjacent flats do not occur so frequently. It will be enough just to determine the safe order of knees for the deposit and define the yield.

Reason: