The similarity system | Page 602 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Similarity: A Practical Approach 859 replies

Similarity Trading - The ZZBB Method 215 replies

Rainbow similarity 256 replies

Similarity Stuff 17 replies

Directional Similarity 0 replies

Attachments: The similarity system
Exit Attachments

The similarity system

  • #12,021
  • Nov 11, 2014 3:02pm Nov 11, 2014 3:02pm
Can any of our wonderful coders here help me code an indicator that ALERTS and draws a horizontal line to mark whenever a fractal is engulfed by a candle close BEFORE another fractal prints in the same direction?. Hope it's clear. :nerd:
Thanks guys.
Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDM30.png Size: 41 KB
  • #12,022
  • Nov 11, 2014 3:14pm Nov 11, 2014 3:14pm
Quoting Barbaresco
{quote} M1 Update: All of the TZ strategies I have tested seem to have the same problem - W:L is good or great, but R:R is bad or terrible. Increasing R:R kills W:L. So, why not try to use this to our advantage. If small wins can't keep us ahead of the large losses, why not take small losses and get propelled by the large wins. So, with the quoted example from above, instead of running a single trade with the offset of 25 pts for avg +45 pts/wk, I checked running the strategy in reverse with multi-entry up to the 25-pt threshold. With 1 trade every...

Been trying to follow your posts on the strategy. Little difficult without charts as I'm very visual. But definitely looks interesting. Will be watching for further updates.
  • #12,023
  • Nov 11, 2014 3:18pm Nov 11, 2014 3:18pm
Quoting Barbaresco
{quote} M1 Update: All of the TZ strategies I have tested seem to have the same problem - W:L is good or great, but R:R is bad or terrible. Increasing R:R kills W:L. So, why not try to use this to our advantage. If small wins can't keep us ahead of the large losses, why not take small losses and get propelled by the large wins. So, with the quoted example from above, instead of running a single trade with the offset of 25 pts for avg +45 pts/wk, I checked running the strategy in reverse with multi-entry up to the 25-pt threshold. With 1 trade every...
R:R is not bad when you consider TZ as zones with a calculated maximum deviation under "x" circumstances. Each level above or below has a corresponding % probability chance. Knowing these % should be the solution in avoiding "bad" R:R's. It is a simple concept, the higher the % gets(the more price reaches max deviation), the more (small sized) long or short positions will be added into play. Adding smaller chunks, one by one, instead of having the whole thing at one level, is more efficient in this case.
  • #12,024
  • Nov 11, 2014 4:39pm Nov 11, 2014 4:39pm
Quoting jmn5611
{quote} I had my zone cleared today too. There is very little doubt remaining that the principle of similarity and of transient zones is valid.
Nice observation friend Our next focus will be to add the foundation for delay! That will get us into trends, short or long , very early!!!
  • #12,025
  • Nov 11, 2014 4:43pm Nov 11, 2014 4:43pm
Quoting ephisi
{quote} Hey eurusdd, i see you have started dropping new gems. Dont even have time to reply on the thread anymore. I have been preoccupied with TZ studies and testing. Building my system. So far so good as can be seen from my trade explorer. I still have a lot to learn though. God enrich you more and more
Nice! I am not really back. Just dropping a few signals here and there!
  • #12,026
  • Nov 11, 2014 4:54pm Nov 11, 2014 4:54pm
Quoting Robertk
{quote} R:R is not bad when you consider TZ as zones with a calculated maximum deviation under "x" circumstances. Each level above or below has a corresponding % probability chance. Knowing these % should be the solution in avoiding "bad" R:R's. It is a simple concept, the higher the % gets(the more price reaches max deviation), the more (small sized) long or short positions will be added into play. Adding smaller chunks, one by one, instead of having the whole thing at one level, is more efficient in this case.

I haven't found the probabilities to hold when looking at scenarios with tradable ranges. The probabilities look great because the non-tradable bars are treated as wins in the stats. Larger tf's look better because the avg occurrence of a setup offers a tradable range, albeit most of the time with poor R:R. Ex. trading the 2nd occurrence of a price in a PTZ after the first occurrence became a confirmed TZ. This is supposed to yield 0.9% loss, given an h that yields 3%. In my tests, it won only 42% of the time using a minimum TP of 5 in M1, i.e. I only entered the trade if the PTZ was min 5 pts away from the price. That 5 pts brought 99% down to 42%...

I think that the probability for a price to become h-transient or h-recurrent can be thought of as a vector of time and price, and the attractive or repulsive force of the transient or recurrent price is subject to logarithmic decay with both elapsed time and the net change in price. If the vector pierces a particular arc of a circle with center [time(PTZ) ,price(TZ)], the probability for that trade falls below 50%. Perhaps the sum of all of the potential arcs is price (efficient market), and I wish I was still 20 yrs old so I could attempt the calculation lol...

It has been a couple of decades since linear algebra for me, and I may also focus too hard on a tree and never see the forest, so perhaps I'm still missing something. If that is the case, please let me know.
  • #12,027
  • Nov 11, 2014 5:11pm Nov 11, 2014 5:11pm
Quoting Eurusdd
{quote} Nice! I am not really back. Just dropping a few signals here and there!
Bless you all the same sir. This really needs to be on wikipedia. youre a legend in the making. i have not started applying probabilities and subsequence theorem. i havnt had the time to learn them. My TE results is strictly based on TZ without probability and subsequence theorem. Imagine what i can do when i fanally learn probabilities and subsequence theored. This is too good to be true. I never traded any pair other than eurusd. but at first attempt at trading other pairs i cant believe my eyes. awesome
  • #12,028
  • Nov 11, 2014 5:13pm Nov 11, 2014 5:13pm
Quoting Eurusdd
{quote} Nice observation friend Our next focus will be to add the foundation for delay! That will get us into trends, short or long , very early!!!
waiting patiently. my system is based on short term and long term swings as can see om my explorer. cant wait to see what delay is all about
  • #12,029
  • Nov 11, 2014 5:36pm Nov 11, 2014 5:36pm
Alas! We have made it to Rome..... 99.33% of levels drawn are cleared...
Attached Image (click to enlarge)
Click to Enlarge Name: Screen Shot 2014-11-11 at 4.34.34 PM.png Size: 35 KB
Life's Good & Why Wouldn't It Be? Im a Pirate After All!
  • #12,030
  • Edited Nov 12, 2014 10:29am Nov 11, 2014 5:43pm | Edited Nov 12, 2014 10:29am
Quoting LITEchild
Can any of our wonderful coders here help me code an indicator that ALERTS and draws a horizontal line to mark whenever a fractal is engulfed by a candle close BEFORE another fractal prints in the same direction?. Hope it's clear. :nerd: Thanks guys. {image}
I have attached the indicator.

Attached Image (click to enlarge)
Click to Enlarge Name: USDCADH4.png Size: 73 KB

Attached File(s)
File Type: mq4 LineFractal.mq4   17 KB | 841 downloads
  • #12,031
  • Nov 11, 2014 5:48pm Nov 11, 2014 5:48pm
Quoting PiratePip
Alas! We have made it to Rome..... 99.33% of levels drawn are cleared... {image}
how do you get those stats in the upper left corner?
Humble & Kalcker CLO2 = Covid killer
  • #12,032
  • Nov 11, 2014 5:59pm Nov 11, 2014 5:59pm
Quoting PiratePip
Alas! We have made it to Rome..... 99.33% of levels drawn are cleared... {image}
Awesome job!

How are the stats when you increase the number of bars to lets say around 100,000 in M5? Could you post some stats for large number of bars in different timeframes and currencies. I'm really curious how good the subsequences are. Maybe it might be useful to also see the drawdown. What I would do is save all the trades to a CSV file and then check the distribution in excel so I can get an understanding of the drawdown.

Did you do it similar to the way Kiads did it?

Thank you!
Focus, Patience, Determination & Order in chaos
  • #12,033
  • Nov 11, 2014 6:01pm Nov 11, 2014 6:01pm
Quoting saver0
{quote} Awesome job! How are the stats when you increase the number of bars to lets say around 100,000 in M5? Could you post some stats for large number of bars in different timeframes and currencies. I'm really curious how good the subsequences are. Maybe it might be useful to also see the drawdown. What I would do is save all the trades to a CSV file and then check the distribution in excel so I can get an understanding of the drawdown. Did you do it similar to the way Kiads did it? Thank you!

I will try and see what I can do to upload some more data.

According to eurusdd the as the sample pool goes onto infinity the number will surely just get closer to 1(100%).

Time to figure out how we can prove this now.

Pip
Life's Good & Why Wouldn't It Be? Im a Pirate After All!
  • #12,034
  • Nov 11, 2014 6:08pm Nov 11, 2014 6:08pm
Quoting PiratePip
{quote} I will try and see what I can do to upload some more data. According to eurusdd the as the sample pool goes onto infinity the number will surely just get closer to 1(100%). Time to figure out how we can prove this now. Pip
Thank you! Can't wait to see the results.

Also, this looks very different from SSxux99v2 from your thread. Do you have a modified version or is it completely different?
Focus, Patience, Determination & Order in chaos
  • #12,035
  • Nov 11, 2014 6:11pm Nov 11, 2014 6:11pm
Quoting xux99
{quote} I have attached the indicator. {image} {file}
Ooh lala! You have a kind heart Xu! Really appreciate this. Tnx man!
  • #12,036
  • Nov 11, 2014 7:16pm Nov 11, 2014 7:16pm
Quoting saver0
{quote} Thank you! Can't wait to see the results. Also, this looks very different from SSxux99v2 from your thread. Do you have a modified version or is it completely different?

This is a completely different version(currently private until further notice).

The probability for the ssxux99v2 is about 85-89% on most pairs so it is still extremely powerful.

Pip
Life's Good & Why Wouldn't It Be? Im a Pirate After All!
  • #12,037
  • Nov 11, 2014 7:18pm Nov 11, 2014 7:18pm
Quoting Eurusdd
{quote} Nice! I am not really back. Just dropping a few signals here and there!
Hello EURUSDD....

Great thread...great insights...great teachings...I could finish only 44 pages....of....602,,,in 2 days....

Can you point me to a faster way to go through your great work ...is any summary important posts / pages that i may learn first....

My respects to all contributors here...

regards..

T4xr
  • #12,038
  • Nov 11, 2014 7:25pm Nov 11, 2014 7:25pm
Quoting T4xr
{quote} Hello EURUSDD.... Great thread...great insights...great teachings...I could finish only 44 pages....of....602,,,in 2 days.... Can you point me to a faster way to go through your great work ...is any summary important posts / pages that i may learn first.... My respects to all contributors here... regards.. T4xr
Click on EURUSDD's name and it will take you to his profile page, from there you can click on posts or threads started.
  • #12,039
  • Nov 11, 2014 8:21pm Nov 11, 2014 8:21pm
Quoting PiratePip
Alas! We have made it to Rome..... 99.33% of levels drawn are cleared... {image}
Please give us here your bombs
  • #12,040
  • Nov 11, 2014 8:24pm Nov 11, 2014 8:24pm
Quoting Exciter135
{quote} Please give us here your bombs

Need to clear this up, They are not mine and unfortunately I can not sure them for the time being until we hear from cheif.
Life's Good & Why Wouldn't It Be? Im a Pirate After All!
2 traders viewing now