Rel's Forge | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Rel's Philosophy of Trading Discussion 24 replies

Rel's Evaluation 13 replies

Rel's Lab 103 replies

Attachments: Rel's Forge
Exit Attachments

Rel's Forge

  • #1
  • First Post: Edited Oct 23, 2014 8:48am Oct 11, 2014 3:01pm | Edited Oct 23, 2014 8:48am
UPDATE 23rd Nov 2014 : Forge is closed, but questions are still welcomed.

The purpose of this thread is to demonstrate the translation of various research into reality, hence the keyword 'Forge'. The 'Lab' is primarily about research and testing ideas, while the 'Forge' is primarily about implementation and creating tools.

Again 3 simple rules / guidelines for participation of this thread remains the same as from the Lab, but with an additional recommendation :
1-No indicators will be released. No amount of asking/begging will get me to release anything. It is really up to me to decide what I would like to release to the public, not anyone else.

2-Be prepared for the unconventional. If you find that these methods are not for you, please do stay out of this thread. Do be warned that often I am forced to write in point form as there are too many things to expound upon.

3-Do consider the rational, context and background of the tools before commenting. If you have your own tool that is backed up by real research, then you deserve a shot at commenting. After all, this is a virtual forge, so do not assume all tools found here works 'as is'. It is highly recommended that you create your own tools as well.

Recommendation-Do try to read up and spend your own time on the following to get sufficient background on what is being attempted. Trust me... it will help as there is no point answering certain questions if they are answered sufficiently elsewhere.
-Rel's Lab and whatever posts I've done on FF so far (there are a few post outside the Lab thread and iDouble's main thread)
-Any iDoubleStoch / iDouble / WaveTop..etc post that is lying around the internet (99.9% of his advice is worth looking at)
-Ray Barro's blog (tradingsuccess.com/blog/), videos (vimeo.com/channels/639659) and the book Nature of Trends
-MasterForex-v's work (masterforex-v.su)
  • #2
  • Edited 4:30pm Oct 11, 2014 3:54pm | Edited 4:30pm
Markets move from state to state. We had already established this in the previous thread plus there is an edge when we follow price in this manner. IMO the best way to capture this is swing indicator. This means one should get a good basic understanding on how (most) swing indicators actually work.

Consider this picture;
http://i60.tinypic.com/2v3pthz.png

By candle count, its obvious that there are only so few price extremes marked out by the swing indicator. Statistically speaking, it falls around 5%-10%. This means 90%-95% of the time, one should be following the line direction set by the swing indicator, hence 'following the trend'.

The 'major' complaint of swing indicators is that they repaint too much and every extreme is a signal to attempt to reverse trade. Thats the point, instead of being wrong and try to reverse trade 90%-95% of time. The current swing will repaint 90%-95% of time, so that one can track the current momentum and direction of the trend correctly.

However, IMO there is three other 'problems' that is much less understood by majority of ZZ traders (I say this because I see no existing study of this anywhere with clear explanation, other than Ray Barros who did indirectly explain this).

1-The gray boxes are areas of price congestion, before the swing turned its line direction and momentum, liken to a small breakout. If the current swing looks to be ending and one intends to reverse trade, one is better off waiting for the basic turn and trading its breakout. This means the 90%-95% edge drops to around 70%, but it is still a substantial edge. Also, consider that this also means the swing indicator itself is a good way to find these mini price congestion areas leading to a small breakout trend later. Congestion = squeeze market state too. This also means that the swing indicator with these gray boxes mark out what we already understood, that price moves from state to state, even on a smaller scale. Ray Barros calls this type of MTF analysis as '1st and 2nd lower timeframes'.
http://i60.tinypic.com/3498e2b.png

2-There are cases where the swing turn breakout doesn't produce sufficient momentum and price fails to move further. Fortunately, these happen 15%-20% of time and basic understanding of market states on a higher scale will tell you when this is more likely to happen.
http://i59.tinypic.com/34e4yfr.png

3-Understanding of market states, applied for both higher and lower scales, will also tell you where is the safer side of bias (in this case, down) to trade this mini breakout, because the squeeze state (yellow) of price is temporary and can mutate into trend state (green). The key breaks to follow are marked by orange and the silver boxes.
http://i57.tinypic.com/1pjpyu.png
  • #3
  • Edited 6:14pm Oct 12, 2014 1:53pm | Edited 6:14pm
Quoting Relativity
2-Conventional time based charts are just compressed price movements.
What if you wanted to trade the movements within say, a D1 chart? Then you would have to use a timeframe lower than D1 to trade the movements e.g. H1 or M15. From a pip/time efficiency perspective, meaning getting the most pips within the shortest amount of time; isn't that what we all seem to want? I've done countless of studies in the past to determine which is the most efficicent time frame to watch and the answers vary on how volalilte the instrument is. e.g. EURJPY is really...
I left out one more step regarding PTe; one has to compare the results to how well it does against 'Time Progression'.
e.g.
M1 -> M5.
If we are using M1 as the base, the results for M5 are expected to be 5 times larger, in order to see how fast/slow price decays over time.
http://www.traderbrains.com/average-...of-timeframes/

There are 3 types of PTe studies one can do;
PriceTimeSpeedVSTimeProgression
VolumeDensityVSTimeProgression
HighLowRangeVSTimeProgression

... in order to aid identification of 'best' highest timeframe and 'best' lowest timeframe, as so to objectively achieve sufficent overall context for MTF analysis.
Recent results found are interesting. 2 to 3 years back I found W1 to be the most important highest timeframe for context. Things have changed the last 12 months.

http://i59.tinypic.com/2iiya82.png
http://i59.tinypic.com/1il5bc.png
http://i60.tinypic.com/2aexie.png
http://i57.tinypic.com/2pseov4.png
http://i57.tinypic.com/dm2pe1.png
http://i61.tinypic.com/29x830y.png

... that the market is becoming much more saturated and the MN1 timeframe is trying to take over W1 as the best context. I do think W1 still holds pretty well thou, as I still find MN1 pretty erratic. Plus despite of increased volume density, MN1 still could not take over W1 when it comes to price range. But it is at least trying... maybe W2 is a good compromise?

Quoting iDoubleStoch
{quote}{quote} Good work! Just be sure to involve all time frames relevant to your expected profit event horizon for each trade. That is, make sure your analysis covers time frames that are small enough and large enough to be relevant to the average duration of each trade you make. If you are a Swing or a Long Term Position Trader then using Weekly, Monthly, Annual or even Multiple Annual bars of data such as 2 Yr or 3 Yrs bars (you may have to construct them on your own or use a custom indicator that accurately produces custom time frames). As an Intra-Day Trader, be sure to include all lower time frames and higher time frames up to and including the Monthly time frame.
This is IMO how to identify 'market changes' statistically. Therefore I do not believe anything like 'best timeframe to trade'; its all relative, but theres a certain 'range' that one needs to take note.

Also, the 'best' lowest timeframe tends to cycle between M1 (the base), M5, M15, M30 (see 1st 2 pictures). Therefore I think my its best that settings for context remains the same, but I would take note to run this test again around next year Feb ;
Lowest Timeframe : M1
Highest Timeframe : W1
  • #4
  • Edited 6:18pm Oct 12, 2014 6:05pm | Edited 6:18pm
http://forums.babypips.com/free-fore...html#post81015

Quoting Relativity
Update : The thread's header (http://www.forexfactory.com/showthread.php?t=499243) now has a poll. Just a simple survey to test if there's sufficent interest before I seriously consider starting a new thread for... something special of course. Teaser : {image}
Quoting iDoubleStoch
{quote} No. Since I was never a heads down coder, when I would switch back and forth between excel and MT4 charts to do research, I needed something with a long and flowing visual reference on deep price history when I was doing research using MT4 charts, which is where OS1 came from. Eventually, I found products like iExpertAdvisor and Molanis, both visual IDEs for creating EA. OS1 then became a fixed part of the system after a while {quote} OS1 is not mathematically connected to the TAC-TCD chart. They are two distinctly separate concepts. One is a moving average that includes some mathematical tweaking on my part to produce a very long range view of market history. The other is based on a different kind of technical analysis foundation that I reference as Trajectories and Vectors, but that are not classical (physics and mathematics) in definition. Trajectories are what make up the bulk of all my indicator designs and those are called Delta Class Indicators. The TCD chart that I posted was only shown because a member here indicated that he had been following my work for a few years and he posted an excel spreadsheet with names of much earlier indicators that I created a long time ago. That's how I knew that I could post a TCD chart, and it would not be too foreign to him - though it might be something that he had not created yet for himself. The data you see on that chart, had not be discussed in this thread and is not taken from the B2i or OS1 indicator inputs. Those are Delta Class Indicators and they are something completely different. I have mentioned and even given the logical structure for creating the prototype version of my Distinct Vega Indicator, but I'm not sure whether or not people followed that post. I've also said that I would work with an MQL developer to include Distinct Vega into a B2i indicator, if they could get the B2i indicator to work in all time frames with the appropriate variable inputs (parameters), so the indicator was flexible enough to actually be useful to people. {quote} The nature of the universe is chaos. What gave the universe physical stability were the strong force, weak force, electromagnetism and gravity. The nature of financial markets is chaos. What gave financial markets stability (believe it or not) is fear, greed and sufficient market participation, aka volume. Those three things stabilize price to the extent that market volatility becomes an acceptable risk, as both fear and greed work together in the minds of market participants, holding price in check on a routine basis and proportional to a given amount of volume. When just one of those three stabilizing forces gets out of control the market returns back to its natural condition of chaos. So, real equilibrium would be antithetical to financial markets and could only come about in two conditions: Zero Volume, or Equality of both Fear and Greed. In both cases, no trading takes place, price find real equilibrium and the market dies a painful death. Price simply stops moving. This is why I shifted my view on the Moving Average a long tie ago, and began approaching it as a vector quantity with both speed and magnitude, where the speed component is measured in the angle of descent or ascent, up or down the price scale. This eventually caused me to use only a Double MA, where both of its calculations are considered a single average price dimension. This allows for the magnitude of the MA to be expressed through the distance between both lines. With that arrangement, I can then calculate a form of "gravity" between both lines where the gravitational force is directional proportional to the product of both MA values and inversely proportional to the distance of the square between them. Thus, in my eyes, price becomes tantamount to a body in orbit about its star, the MA. Then we get into what I call the Orbital Mechanics of Price, which is a bit beyond the scope of this thread, but it obviously involves OS1 and explains my tweaking of the routine MA calculation. Essentially, I see the OS1 Double MA as a kind of Binary Star System where price orbits that system. But, again, we are getting well beyond the intent of this particular thread. So, while I do understand what you might mean by equilibrium, for reasons that go well beyond the scope of this thread, I have a slightly different take on what the MA actually means relative to price. But, I do believe that I understand the idea of Buyers and Sellers agreeing on a certain price level and how a single MA might be seen by some as that level. However, you can quickly check that by simply grabbing the Guppy GMMA MT4 indicator (google it) and loading that into any chart. Doing that will reveal that what is considered "agreed upon price" at one moment, is quickly changed the next moment. Thus, the concept of equilibrium as being that "place" or "level" in the market where most participants "agree" on "price" rapidly changes definition and price plows through Guppy levels with nearly each tick as the following chart shows: Guppy GMMA {image} {quote} Yes, OS1 and B2i are just as different from each other as vanilla ice cream and chocolate cake. They go great together but are completely different creations.
http://i59.tinypic.com/2cdbj3s.jpg

Quoting iDoubleStoch
{quote} "come up with a aggregate and optimized cycle of 'internal pressure' VS 'external pressure' behavior." = Density Probability. When you locate those aggregations repeating themselves in history and resulting in movement in price in a particular direction, then you have discovered a Density Probability Pattern. That is called Delta StealthTrading and few can do it correctly. It is really good to see you on the right track. It is so stealthy that it should be illegal. The results are often times just like having insider information before the move happens. Very stealthy. Well, that was probably one of the best summations of how to use Deltas that I've seen anyone write so far. Apparently, you get the concept fairly well and are looking in the right directions while asking the right questions in your work. That pretty much sums up the Delta game. It is all about finding the locations that contain the highest density of probability across all relevant time frames using Delta based indicators. That is the entirety of the game in a nutshell. Well summed. Yes, that is the limit of my public demonstration of the matter as well. Going any further than your summation enters into proprietary territory. The goal was to get people to where you are in your understanding of the right questions to ask themselves. The rest is up for them to discover their own path to using the Delta concept to their advantage. Glad to see someone actually getting the concept and asking the right questions. Well done! If I were still doing protege training, based on that summation, that would have probably been enough to get you picked. Back then, I only selected people who could demonstrate that they understood the concept and the problems that had to be solved. However, I do not do anymore personal training. But, what's really important is that you now know the problem that needs to be solved and that is more than half the battle. Congratulations! It is good to see someone actually understand a new trading philosophy that you have been trying to share with people for years without giving the entire store away. Now, you understand why this is not standard technical analysis.
http://i58.tinypic.com/2lmqng2.png
  • #5
  • Oct 12, 2014 6:46pm Oct 12, 2014 6:46pm
OK. Floor is now opened for questions before I proceed further.
  • #6
  • Oct 12, 2014 7:14pm Oct 12, 2014 7:14pm
Nothing to add just yet sorry Rel but I am very interested in what you are doing. Let me just sit on the sideline for a bit longer...
What's it all about? It's all about money.
  • #7
  • Oct 13, 2014 6:02pm Oct 13, 2014 6:02pm
Updated the indicator to automatically map the most important price levels progressively.
http://i58.tinypic.com/2r286pw.png


The same levels but using an older version of my swings. Lets see how this turns out.
http://i58.tinypic.com/f1ibsp.png


Will do another post or two soon.
  • #8
  • Edited 4:14am Oct 14, 2014 3:52am | Edited 4:14am
Quoting TheRealThing
....
The "1 pip" experiment is extremely useful to almost everyone who has tried it. And from completely different perspectives. It is very inexpensive to find out so much, and potentially profitable.
Trading for Just 1 pip is Difficult
I've received dozens of emails from traders who have tried the experiment and have told me their results and their reflections. Almost every one of them related how much harder it was to get one pip than they had previously thought. (I'm remembering the recent post of a questioner about how difficult could getting...
I had done this experiment before myself and I can vouch everything TheRealThing said here. Especially this part;
Quoting TheRealThing
You will have beaten the trading firm spreads which average about a pip or more and will have learned how important identification of the prevailing trend is. You will find that even searching out congested price action for obtaining quick 1 pip trades is a timebomb waiting to go off.
1 pip breakout is a very simple but important concept when it comes to using swing indicator. Knowing that a swing indicator tracks momentum and direction, having correct identification of prevailing trend at least 70% to 90% on your side is helpful.
Quoting Relativity
{quote} Pardon me but I do know exactly what I am talking about. There's nothing fishy about the claim. That's also why I said 'The only question left from here is how much off the 95% it will continue length wise?' You can be surprised that '1 pipper systems' can actually act as proper triggers as a good start, with a basic edge of 80%...
Remember this?
http://www.forexfactory.com/showthre...79#post7722179
http://i62.tinypic.com/taq6mg.jpg

...and this?
http://www.forexfactory.com/showthre...31#post7724431
http://i58.tinypic.com/2ewjg4y.jpg

The aqua lines are the key lines that swings have to break before they commit an opposite turn of line direction. The aqua lines are in effect, the gray boxes here, except that the aqua lines show 2 to 3 layers worth of key lines, while gray boxes only show 1 layer. Layers implying MTF/MPF periods used e.g. 5 period swing + 25 period swing + 125 period swing.
http://i60.tinypic.com/3498e2b.jpg

Therefore :
http://i58.tinypic.com/11tpd8o.png
1. For a 3 layer MTF/MPF, there are 3 key lines that have to be broken before all 3 commit to an opposite turn of line direction.
2. And all 3 key lines fall approximately around the same level.
3. And we had already established that at least 70% of time, price will continue to move forward after the line break. (although the hard statistics say 89%, I chose to be conservative)
4. We also define a line break as 1 pip breakout.
1 + 2 + 3 + 4 = ?


Quoting Relativity
When I decided to tackle time, I was taken aback by the following;
1-There are too many timeframe sets to consider. e.g.
Ray Barros uses the approximate x5 rule to construct his timeframes, since we have 5 days in a week. e.g. H1, M290, D1, W1, MN1
Yet non-standard timeframes does exist and x5 rule doesn't seem to apply for all situations.
Timeframes can also be subconsciously enforced by trading software companies. I've always asked myself why MT4 uses M1, M5, M15, M30, H1...etc, which doesn't obey the x5 rule Barros has.
There are so many timeframe sets to explore that I have no idea where to start. Therefore, what I've tried to do is to reduce the number of vectors to be explored to the key ones. I define the key points as 'swing high and swing low'. Nothing about open and closes. Maybe my issue is that I had not understood the significance of open and closes sufficiently?
Quoting Relativity
-Why are we trying to map and model the market using fractals, which are of geometric progression, but the order of conventional time based charts (M1,M5,M15,M30..) are not in geometric progression or even fibonacci sequenced (M1,M2,M3,M5,M8,M13,M21,M34...etc)?
Imagine this for a N layer MTF/MPF.
Imagine the layers are not 5 period -> 25 period -> 125 period ...etc, but 1,2,3,4,5,6,7,8 and so forth until reaches N
  • #9
  • Edited 8:28am Oct 14, 2014 6:41am | Edited 8:28am
Quoting Relativity
OK. Floor is now opened for questions before I proceed further.
I thought I would ask my question here : )
I am not sure how to read the price levels that your chart is showing :
Quoting Relativity
Updated the indicator to automatically map the most important price levels progressively. {image}
I can see, for example, that the yellow horizontal lines are the levels with most occurences and that the price may be attracted by them more than the lesser occurences, but I am not sure what these levels represent regarding the swings or priceframes.
The possibilities I considered end up being wrong after some observation. So, what do these stats represent ? ^^

EDIT: I watched the screen video, but it did not help me yet to get this. I only noticed the 'up' and 'dn' changing on the upper left corner and the levels updating when the swing repaints or that a new swing is drawn.

Quoting Relativity
Therefore : {image} 1. For a 3 layer MTF/MPF, there are 3 key lines that have to be broken before all 3 commit to an opposite turn of line direction. 2. And all 3 key lines fall approximately around the same level. 3. And we had already established that at least 70% of time, price will continue to move forward after the line break. (although the hard statistics say 89%, I chose to be conservative) 4. We also define a line break as 1 pip breakout. 1 + 2 + 3 + 4 = ? {quote} {quote} Imagine this for a N layer MTF/MPF. Imagine the layers are not 5 period...
1+2+3+4=?
For this one too, I must admit I am confused and that my small brain does not know what to do with it.
Let's say we are at the 'decisive' price level (if the price goes 1 pip down, the 1pip breakout occurs and the 3 swings take the down direction ). I am divided between two things ( probably they are wrong assumptions as the reasonning is not clear at all in my mind, hopefully someone will correct me ) :

- We may play the breakout, and have 70% probabilities to get 1 pip profit if the breakout indeed occurs ? (as 70% of time price will continue to move forward after the line break )
- But also, maybe we can play the reversal, with a very small stop for a nice reward ? As the breakout did not occur yet, the swing is still Up, there are more probabilities (if we take nothing else in consideration) for the swing to continue than to stop, so i can enter a buy order with 1 pip stop with for example the breakout of the current swing as a target ? ( by writing it, it seems a bit strange, but who knows... )
  • #10
  • Oct 14, 2014 6:51am Oct 14, 2014 6:51am
Quoting Relativity
Updated the indicator to automatically map the most important price levels progressively. {image} The same levels but using an older version of my swings. Lets see how this turns out. {image} Will do another post or two soon.
Looks pretty good to me....
(click link to download video file named prog.mp4)
http://www.yourfilelink.com/get.php?fid=981286

I do like my 'older' work... but I still find it pretty single dimensional.
http://i59.tinypic.com/21j7t6d.png
  • #11
  • Oct 14, 2014 8:22am Oct 14, 2014 8:22am
Quoting 030985
I thought I would ask my question here : )
I am not sure how to read the price levels that your chart is showing :

I can see, for example, that the yellow horizontal lines are the levels with most occurences and that the price may be attracted by them more than the lesser occurences, but I am not sure what these levels represent regarding the swings or priceframes.
The possibilities I considered end up being wrong after some observation. So, what do these stats represent ? ^^
You got it right. The percentages/stats are just the against N layers, depending on the type of level. I've already revealed the one (1) type of level in the last few posts, which is the swing turn breakout level. I call it 'CurrSwingPostBreakoutTurnLevels'


Quoting 030985
1+2+3+4=?
For this one too, I must admit I am confused and that my small brain does not know what to do with it.
Let's say we are at the 'decisive' price level (if the price goes 1 pip down, the 1pip breakout occurs and the 3 swings take the down direction ). I am divided between two things ( probably they are wrong assumptions as the reasonning is not clear at all in my mind, hopefully someone will correct me ) :

- We may play the breakout, and have 70% probabilities to get 1 pip profit if the breakout indeed occurs ? (as 70% of time price will...
No.

Quoting Relativity
4. We also define a line break as 1 pip breakout.
Logically 1 pip is needed to break any line.
The 70% edge is price to go beyond more than 1 pip at the breakout.
Inversely, you only play the reversal 30% of time.
These percentages are of a single MTF/MPF layer. These percentages will change when we do it on multiple layers. Think of it as confluence of price levels that are logically derived and naturally occurring according to the current state of the market.
  • #12
  • Oct 14, 2014 8:34am Oct 14, 2014 8:34am
Updated the indicator to improve the visuals. Currrently I have 4 types of levels.
http://i60.tinypic.com/25jytxk.png

I think I will add a couple more that is related to the 1-2-3 Trend state and Expand state. These are easier as these levels are breakout based.
Reverse/Resume state and Squeeze state will need different logic from breakout as these are attraction based.
http://i57.tinypic.com/2nitmc6.png

I currently have 3 breakout levels and 1 attraction levels.
Breakout = internal pressure. Attraction = external pressure.
  • #13
  • Oct 14, 2014 9:02am Oct 14, 2014 9:02am
For those who want to investigate the statistics, an excel file is as attached.

If someone can read and understand this correctly, I would actually be delighted =) especially if you can post your thoughts here. In a forge... its all about being busy creating and refining tools, so while I am at it, I suppose I would be concentrating on the task at hand and talk less (yeah right LOL =d)
http://i62.tinypic.com/1qlfmq.png
Attached File(s)
File Type: xls CurrSwingPostBreakoutTurnLevels.xls   363 KB | 694 downloads
  • #14
  • Oct 14, 2014 5:45pm Oct 14, 2014 5:45pm
Levels for today. Nothing is better than observation and experimentation. I think I need to improve on attraction levels, to project price outside current ranges.
http://i62.tinypic.com/24brg9g.png
http://i59.tinypic.com/6y2l9i.png
  • #15
  • Edited 8:40pm Oct 14, 2014 8:12pm | Edited 8:40pm
Post mortem of yesterday's movements and reading of what can happen today.
http://i59.tinypic.com/vzyfdy.png

http://i61.tinypic.com/35j9xkx.png

I say prevailing trend is still down, but its not much and has weakened. Any attempts for long should be scalps or sit out.
Market is now developing into squeeze state, as you can see how the levels position themselves, compared to yesterday's.
  • #16
  • Oct 14, 2014 9:48pm Oct 14, 2014 9:48pm
Rel, as someone who is using zig zag extensively can you tell me your thoughts on this indicator.

http://www.mql5.com/en/code/7050

I have played with it a little bit but not a lot. I have never really used zig zags before. Is it something that would compliment your work or junk?


Also I would love to understand your charts a bit more. I get the gist but can you explain all the percentages/numbers on the right please? Is it the frequency of the prices distributed over N bars?
Thanks
What's it all about? It's all about money.
  • #17
  • Oct 14, 2014 10:16pm Oct 14, 2014 10:16pm
Quoting Innate
Rel, as someone who is using zig zag extensively can you tell me your thoughts on this indicator. http://www.mql5.com/en/code/7050 I have played with it a little bit but not a lot. I have never really used zig zags before. Is it something that would compliment your work or junk? Also I would love to understand your charts a bit more. I get the gist but can you explain all the percentages/numbers on the right please? Is it the frequency of the prices distributed over N bars? Thanks
To answer your zig zag question: yes.
zig zags and waves as I understand them is the core principle behind all this work. A few of the main questions are:
1. How can you know if the zigzags are correctly identifying swing end points?
a. what counts as "correct"?
which of the following 2 pictures "correctly" maps out price?

2. Is there a way to forecast where the wave will go currently (end point) and where the next wave will go?

simple questions, and looking for simple answer. But the process to get to such a point is often not so simple.. Some of the work being shown here for example is showing possible zones of importance which you can think of in a variety of ways: support/resistance, push/pull gravity, "trap zone" and "launch pad" points, etc.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 1.PNG Size: 80 KB
Click to Enlarge Name: 2.PNG Size: 84 KB
"Mrrraggglhlhghghlgh"
  • #18
  • Oct 14, 2014 10:53pm Oct 14, 2014 10:53pm
Quoting Innate
Rel, as someone who is using zig zag extensively can you tell me your thoughts on this indicator. http://www.mql5.com/en/code/7050 I have played with it a little bit but not a lot. I have never really used zig zags before. Is it something that would compliment your work or junk?
Back then when I was still searching for patterns, maybe coinciding with my first few versions of work, perhaps this would had become useful. Now its just something that I don't think I would look into anymore. I will not call it junk. I merely call it evolution; that my work only gets better each day, improving accuracy while avoiding pitfalls of over-optimization / curve fitting. Its definitely not easy to design something of that level, but I do think I have something now.

But for anyone who wants some basic idea as a start, yes, I suppose it is very helpful for their own research.

The patterns this indicator searches are single dimensional, single layered and mutatable. That's good, until one realizes that price and time are multi dimensional and multi layered. It doesn't mean what the indicator is useless. In order to understand how price moves on a multi dimensional basis, IMO one should at least learn how price moves on a single dimensional 1st.

Here's a picture to help us understand what I had observed with care and detail in the lab thread ;
http://i58.tinypic.com/6ga06o.png
All my old work is single dimension swings and patterns; look at the 1-2-3 trend pattern on the left. Granted, there's a lot of learning and usability value there. But to truly measure these movements, one has to consider every single dimension of force and effect at work, that generates these swings and patterns. The pattern on the right looks exactly the same as the left, but the forces at work that drive the creation of this pattern are different, especially for the 2nd and 3rd swing.

Top that with what advice iDoubleStoch recently gave... I figured that I would need to evolve my work further, hence opening the forge thread

Quoting Innate
Also I would love to understand your charts a bit more. I get the gist but can you explain all the percentages/numbers on the right please? Is it the frequency of the prices distributed over N bars? Thanks
It is the frequency of line breaks that has already happen over N layers of swings, termed 'CurrSwingPostBreakoutTurnLevels'. For a single swing to turn its direction, it must break its own line to commit to the opposite direction (explained in the earlier posts)
  • #19
  • Edited Oct 15, 2014 12:45am Oct 14, 2014 11:25pm | Edited Oct 15, 2014 12:45am
later

Quoting Relativity
{quote} Back then when I was still searching for patterns, maybe coinciding with my first few versions of work, perhaps this would had become useful. Now its just something that I don't think I would look into anymore. I will not call it junk. I merely call it evolution; that my work only gets better each day, improving accuracy while avoiding pitfalls of over-optimization / curve fitting. Its definitely not easy to design something of that level, but I do think I have something now. But for anyone who wants some basic idea as a start, yes, I suppose...
My Best Trade is Next Trade!
  • #20
  • Oct 15, 2014 12:09am Oct 15, 2014 12:09am
Quoting LiquidGenius
{quote} To answer your zig zag question: yes. zig zags and waves as I understand them is the core principle behind all this work.
A few of the main questions are:
1. How can you know if the zigzags are correctly identifying swing end points? a. what counts as "correct"? which of the following 2 pictures "correctly" maps out price?
Yes ZZ and waves are still the core principle. I've considered the degree of correctness and figured a few things out. It wasn't easy to create my own swing formula. Unfortunately, if I am to put the methodology of 'correctness' in public, it means exposing my swing formula, which is something I am not prepared to do. The formula is actually very simple and has its roots from the PH-L / PL-H TCD, combined with a small part from Barros Swing's formula, but that is as far as I can describe.

Quoting LiquidGenius
2. Is there a way to forecast where the wave will go currently (end point) and where the next wave will go?
Yes there is a way. Barros uses Std Devs. I intend to attempt to bring that to another level (soon).

Quoting LiquidGenius
Simple questions, and looking for simple answer. But the process to get to such a point is often not so simple..
Some of the work being shown here for example is showing possible zones of importance which you can think of in a variety of ways: support/resistance, push/pull gravity, "trap zone" and "launch pad" points, etc. {image} {image}
Yes its not easy, even to get to some basic edge and baseline. Good keywords here to describe these levels/zones of importance! =)
0 traders viewing now