Not the Grail, just a regular one - Bablokos!!! - page 223

 
Talex:

Joker found his system, after making a filter based on this thread (incremental modulus difference). The increment modulus difference, as far as I understand it, shows volatility narrowing and the last trades show a channel, which is narrowing (my channel is built on 2 sigmas)

Here are all three trades



Let me correct right away - difference in increment moduli has nothing to do with the system but in practice it is also a working tool.

The incremental modulus difference is just a special case of the Penny paradox, where you always have a 75% and 25% chance of winning and 25% chance of losing respectively in two combinations of spins. ( I gave a link to the penny simulator in this thread. Here it is: https://www.haverford.edu/math/cgreene/390b-00/software/CoinFlip.html ). A 75% probability of winning on SB is just a huge probability advantage.

Why can't a penny be applied in this case, even though you really want to?! Even synthetic trading with seemingly known resistance and support spread targets does not guarantee profits and limits losses within given limits.

For those who are into combinatorics, go straight to binary options, it will work there.

 
Joker:


Let me correct right away - incremental modulus difference has nothing to do with the system, but in practice it is also a working tool.

The incremental modulus difference is just a special case of the Penny paradox, where in double combinations of spins there are combinations in which you always have a 75% and 25% probability of winning and losing, respectively. ( I gave a link to the penny simulator in this thread. Here it is: https://www.haverford.edu/math/cgreene/390b-00/software/CoinFlip.html ). A 75% probability of winning on SB is just a huge probability advantage.

Why can't a penny be applied in this case, even though you really want to?! Even synthetic trading with seemingly known resistance and support spread targets does not guarantee profits and limits losses within given limits.

For those who are into combinatorics, go straight to binary options, it will work there.


About the Peny paradox, as I understand it. 0-eagle, 1-crack. There is a combination, for example, 000, if I bet on 100, with a high probability (87.5%), my combination will fall out FIRST. And that makes sense, because if I don't get 000 the first three times, and there's only a 12.5% probability, then my combination wins. But we have to guess the LAST combination. Peni's paradox makes it clear that some combinations have a "fall first" advantage over others, maybe the paradox works the other way too? I don't know, but I think it's unlikely.

Although, yesterday I made a filter according to Used's description in this thread and debugged it, haven't tried it yet, not enough time, but during debugging I looked at its work a bit. Most of the time nothing interesting, but in some moments it has shown the probability of movement in one direction at 65-80% levels, though I haven't watched the realization of benefits, I had no time for that, but it's a matter of time. ))

As for trading in the channel the question remained for me "Co-integration can be performed visually, using the rollover of currencies on the relative motion chart, by throwing out all those that do not fit into the channel we are aiming for". (Copyrigth Joker), i.e. in selecting the "right" instruments according to Joker.

 
Talex:

About Peny's paradox, as he understands it. 0-eagle, 1-crack. There is a combination, for example, 000, if I bet on 100, then with a high probability (87.5%), my combination will fall FIRST. And that makes sense, because if I don't get 000 the first three times, and there's only a 12.5% probability, then my combination wins. But we have to guess the LAST combination. Peni's paradox makes it clear that some combinations have a "fall first" advantage over others, maybe the paradox works the other way too? I don't know, but I think it's unlikely.

Although, yesterday I made a filter according to Used's description in this thread and debugged it, haven't tried it yet, not enough time, but during debugging I looked at its work a bit. Most of the time nothing interesting, but at some moments it has shown the probability of movement in one direction at 65-80% levels, though I haven't watched the realization of advantages, I had no time for it, but it's just a question of time. ))

As for trading in the channel the question remained for me "Co-integration can be performed visually, using the rollover of currencies on the relative motion chart, by throwing out all the ones that do not fit into the channel we are aiming for". (Copyrigth Joker), i.e. in selecting the "right" instruments according to Joker.


keep it simple:

For HH combinations the winning combination will be TH ( with 75% probability )

For TT combinations the winning combination will be HT ( with the same 75% probability )

How is this used in practice? - Very simple: on SB if we meet a combination of HH, the next trade is T... . And vice versa: If we get TT on the SB, our next trade will be H... With the number of spins tending to infinity, this theory one hundred percent works. From a practical point of view I will only say that refinancing is extremely undesirable at this way of trading even with binary. But the constant movement of the winning curve in the required direction is guaranteed in this case.


However, it has almost nothing to do with practical spread-trading and this branch, in most cases it is purely for binary trading. Although double testing the lower or upper spread channel also means a probable reversal (with the same 75% probability). But I don't use this in practice.

 
Joker:


Let me correct right away - the difference in incremental moduli has nothing to do with the system...

What about this

Joker:
...

The patterns here are the state of the market in the past (the previous states of spread instruments relative to each other: a>b or a<b). In the Expert Advisor's code, it is translated into 011110101 ).

...

 

To Joker:

***** if we meet the combination HH, the next deal is T*****
Alas, even the Excel PRNG is not right, otherwise life would be too easy:) It's 50/50 as usual.
You could also bolt on a martini, as the main guru (NeColla) said. But that's another matter.

 

Joker! What's the probability of the synthetic you're trading? Is it 99%? The statistics have already accumulated a lot, apparently.

If you build classic (MNC, multifactor regression), it often does not return to the middle of the channel (10-20%).
Seka's deal was hung for 1.5 months with drawdown of more than 3 profits. In the real world, that's supposed to be a stop loss.

Trading a border breakdown will result in less frequent profits or just the same 10-20%.

Apparently there's magic in building a synthetic.

 
Joker:


Don't complicate it:

For HH combinations the winning combination is TH ( with 75% probability )

For TT combinations the winning combination will be HT ( with the same 75% probability )

How is this used in practice? - Very simple: on SB if we meet a combination of HH, the next trade is T... . And vice versa: If we get TT on the SB the next deal is H... When the number of spins tends to infinity, this theory works hundred percent....


Here is the file, it shows that with HH, the next deal is 50/50%, though. ((( , there even on second sheet made HH and still next 50 on 50%.

How it works for you is beyond me.

P.S. The file is made in OpenOffice, saved in excel, can that in it it should change.

Files:
comb.zip  57 kb
 
Talex:

Here is the file, it shows that with HH, the next deal is 50/50. ((( , there even on the second sheet did HH and still the next 50/50.

How it works for you is beyond me.

P.S. The file is made in OpenOffice, saved in excel, can that in it should change.


Greetings!

I have looked at your file as far as I have had time. In my opinion, you have a mistake, namely, I do not see work with spins.

Please study the theory (there is information in wikipedia, + practical example of application here: https://www.haverford.edu/math/cgreene/390b-00/software/CoinFlip.html)

 
Joker:


Greetings!

I looked through your file as much as I had time. In my opinion, you have a mistake, namely I do not see the work with spins.

Please study the theory (there is information in wikipedia, + practical example of application here: https://www.haverford.edu/math/cgreene/390b-00/software/CoinFlip.html)


Please explain, what do you mean by "work with backs"? And in the link that you cited incorrectly calculated the average length of the series, I counted as such, the first three series. I can't right now, but later, if I don't forget, I will calculate the average length of the series myself.
 
Joker:


Greetings!

I looked through your file as much as I had time. In my opinion, you have a mistake, namely I do not see the work with spins.

Please study the theory (there is information in wikipedia, + practical example of application here: https://www.haverford.edu/math/cgreene/390b-00/software/CoinFlip.html)


Good day!
If you can, a link, please. I couldn't find it, the word "spin" comes up with definitions from quantum physics. In game theory I also could not find any definition. If by "work with spins" you mean what was stated in file before, in which you named columns, then everything is somewhere so, as you write "...For combinations of HH the winning combination will be TH (with 75% probability)...".

But by TH you mean a LOT of combinations, as far as I understand it. Anyway, made a filter, which I think is what you are trying to tell us all about. If you are interested, I can send it to you in my personal message (I can do it at MQL5). I just have to check it, it's a question of time, which is always in short supply.