Martingale: the maximum possible chain of continuous losses/profits - page 11

New comment
 
gip:


I was trying to explain to you the mechanism why in forex it's not 1000 years according to the maths, but much earlier.

Roughly speaking, if bears fed on humans, they would be seen more often.

--

Moose, they're carnivores.


Have you read the question, the Odessa answer, to which you quoted?

Also... I was talking about a random number generator, not forex, to make it clearer, I gave you the example of a "clean" roulette with no zeros.

 
sever30:

Phew... The topic was killed, not a single post on the substance of the question... I understand your skepticism and pessimism about martin, the topic is not like: i want to use martin, what do you advise.

i can't help you, i just want to be sure i will get it right.

i messaged him in his personal message no answer yet, maybe he won't show up at all?


the 2003 office limits the number of lines to 65536, therefore the length of the series

 

sever30:

Read. Tried to say it's even worse on forex than the "clean roulette" theory and stuff like that.

 
wmlab:

You can really make money on a martin. ... Unless you make a little money by risking the whole depot and stopping in time.

Yes, it is possible to make money. Even if you reduce risks to zero. If only one condition is that the deposit is an investor's one and that you stop in time. But where is this "in time" and where is the guarantee that the drain will not start from the first deal?
 
maxfade:


office 2003 limits the number of lines to 65536, therefore the series length



well, it doesn't even count chains on a run with 30,000 series
 
gip:

Read. Tried to say it's even worse on forex than the "clean roulette" theory and stuff.


What makes you say that? Please don't take into account the technicalities(requotes, glitches, connection failures, etc.)
 
From the fact that in forex at these levels the statistics don't work, the fit doesn't work.
 

It seems to me that the question of the topic has not been formulated correctly.

This is why the topic is not attended by acknowledged experts who know the math.

On the one hand Max gave an example from his personal observations. how many times a single result fell out (I call it -27). the length of the series. i.e. practice.

On the other hand theoretically there is a probability that all deals in a series (1 000 000) will have the same values .

As for the "calculators", the machine will calculate what the user put into it (initial data) and according to the algorithm, which the user also put into it. Therefore all the results of calculations are very subjective.

Vladimir, for the sake of interest, read the article about the operation and methods of AC. Just refresh your knowledge on this subject. In the light of this article "Bare mathematics" and the statistics calculated on its basis become less relevant.

https://www.mql5.com/ru/forum/129196

 
sever30:

well it doesn't even count chains on a run with 30,000 series

it seems to work for me, and it works with 65,000, I converted it to 2007 office, you can do up to a million and something there

maybe i don't fully understand the problem, but i do NOT get an error, and the calculation goes all the way to the end

 
maxfade:

it seems to work for me, and with 65000 works, converted to 2007 office, you can do up to a million and something there

Maybe I don't fully understand the problem, but I do NOT get the error, and the calculation goes all the way to the end


share please.
1...45678910111213141516
New comment
Reason: