Safe martingale. - page 14

[Viktor Yarovoi]

Martyn can only work on a very small timeframe (1M). It is very similar to scalping, with a return of around 10% per month.

[khorosh]

GaryKa:

...

I gave you the test technology above. You need a "profitable" Expert Advisor - please. Ideally, without taking into account the commission, swaps, and spread - set the stop loss/take profit the same way on any pair in any direction (you can change the direction of one transaction for clarity) - for example 100 points - in buy a little bit more transactions will occur, always, everywhere and on any pair (even on the reverse). I can even mathematically justify it. Does that work for you? )))

If you want to show me more clearly what algorithm is used to enter the market, you haven't done that yet, but you want to prove something. Do you want to enter on a coin? What you have written is enough to conclude that the security of increasing the lot will not be observed under these conditions.

[TheXpert]

artemiusgreat:

If you don't mind, can you justify to me why with equal TP/SL and multi-directional trades BUY is triggered more often?

It's simple, just be smart.

[khorosh]

Viktor Yarovoi:
Martyn can only work on a very small timeframe (1M). This is very similar to scalping, with a return of around 10% per month.

This is a safe anti-martingale variant, while you are probably talking about an averaging martin. These are different algorithms.

[GaryKa]

artemiusgreat:

Heroix:

The system I gave to illustrate simple things that are not obvious even to many seasoned people. Therefore often in some new system, we see its pluses, and arising disadvantages - well they seem to us at first as if quite insignificant.

This fact (about buys) has its place, but we cannot build a profitable system on it. I stated it as a "profit system".

The answer is simple: we are comparing aditives on a relative scale. On an interval scale, 100 points one way, 100 points the other way, they are equally likely. But this is not the case on a relative scale.

A thumbnail example below:

What is a quotation - it is a ratio of quantity of one currency to the quantity of another currency and they are valued equally. The measurements are relative, there is even a reference point - absolute zero.
That is, there is currency A and there is currency B. If we can exchange 2 A units for 2 B units, then the exchange rate (A/B) is 2/2 = 1.
a) Now let's imagine that there was a change in the market and we can exchange 2 A units with 3 B units. B, that is the exchange rate (A/B) is 2/3 = 0.666(6)
b) Or, equally likely, we can now exchange but vice versa 3 units of A A by 2 units of B, i.e. the rate (A/B) will become 3/2 = 1.5
------
i.e. the events when the rate becomes 0.66(6) or 1.5 are equal probabilities. But what about the points (let 1 point be 1/100th of a point)? Well, in (a) we got 1 - 0.66 = 0.34 = 34 points of decrease, and in (b) 1.5 - 1 = 0.5 = 50 points of increase. Now I think it is clear, that at the same SL/TP pips distance TP will trigger more often for Bays, and SL for Towers. Why we cannot make profit with it is also clear.

[Sergey Efimenko]

khorosh:
Well, send me a profitable Expert Advisor, and I'll add a safe incremental lot, and it will be a grail. I'll make such a soup from an axe that you'll love it.)))

And why the hell would a profitable Expert Advisor need anything else? ;)))

[khorosh]

Sergey Efimenko:
And tell me why a profitable expert needs to be screwed with something else? ;)))

In order to make it a real grail. ) But seriously, to improve its performance.

[[Deleted]]

GaryKa:

At first I thought the explanation was logical ... But I had the feeling that somewhere someone was trying to trick me, like in the joke about the new russian and the devil :)

I think the explanation is too mathematical and mathematicians have too many assumptions, if they start to convert to points and earnings, it will be calculated by addition and subtraction, not by multiplication and division, so there can not be a situation where an instrument A = 2 / 3 and another = 3 / 2. You will still need to reduce to a common denominator, and here you will already have 3 / 2 versus, for example, 15 / 10, and here their changes in relation to each other will look more logical. In my opinion, to use points in explanation and at the same time refer to relations of instruments is an attempt to connect a man and a horse and to equate them to a mythical centaur.

And we can't make money on it, most likely because we don't know where the reference point is, and therefore we don't know what is buying currency, conventionally, 2 / 3, and what is selling, conventionally, 3 / 2.

P.S. I am not writing for the sake of argument, but to understand ...

[GaryKa]

artemiusgreat:

Ok, also last explanation and sleep - how much is the eur/usd rate in pips now - 1.09280 = 109280 pips right?

Now give me an answer what will happen sooner if I put a buy and a sell at the current price with hepothetical targets of 110000 pips. А? I'll tell you, the sell will never (at all, even at the end of the world) reach its target, and the buy may be. But according to you ... my answer is not logical and we have all chances to see a negative quote for eur/usd someday (even if in a hundred centuries)

Not for the sake of argument ... But the explanation is more than logical. I assure you. Even many of the most serious mathematical models of price behaviour use such an assumption (maybe heard) as lognormal price increment, not normal but lognormal.

[Roman Shiredchenko]

How did the sabbath end? It's not clear to me....