Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
I agree with you. All the complexity and then I can't post more, leaves me a bit suspicious. Every successful trader I've known, would say there is less to trading then meets the eye. Find a simple system that will fit your personality, learn it well and trade it.
What has been described here makes me think of a thread on Forex Factory. Merlin's Amazing Strategy. It's here. Warning, the thread is been running for 7 years.
Unfortunately, he does not have the indicator fundamentals correct, so he can't possibly understand how to apply it and I can't provide the details on how to do it.
Regarding "correlated data" - what type of correlation does the trader care about? Not all statistically derived correlations apply. Any basic knowledge of statistics comes with the understanding that correlation does not imply causation. It cannot assume that a causal link between the derived result(s) and variable X (as one example) exists. Also, what net effect correlations would be found in the ultimate "price" of a currency pair on any given day, and an indicator that generates a scale of relevant significance - meaning: Likely to 'Go Long' with a 78% historical significance? (as just one example)
That having been said, restricted-relative correlations (what I like to affectionately call: Covered Correlatives) can absolutely be found when researching OHLC streams and in any time-frame and they can be found in abundance when and if you know what you are looking for. Let us not forget that statistics is just as much about art as it is science. Though one might have an unconventional way of deriving correlations (or, seeing them), that does not reduce the meaning and/or significance of their existence.
In the world of trading, a "fool" is someone who fails to explore the causation behind a correlated or uncorrelated event and who then relies upon such events in the future for the production of bankable revenue. If that's the definition of a fool, then I'm well far from it.
I'm reminded of James Carville's famous words: "It's the economy stupid." Well, in trading, I like to say: It's the causation stupid. I might not be able to nail causation 100% of the time, but if I can find evidence through a particular kind or class of indicator(s) that "causation" seems to cozy-up and rest its weary head on the lap of a particular set of variables, oh say, 93% of the time on average - well, I can find a way to be happy about that as a trader - even though I cannot prove beyond a shadow of a doubt that "cause" was a part of it.
Call it Statistical Faith, if you must give it a name. Or, just call it Statistical Artwork. But, as long as my account balance continues to grow as close to the model as planned, I'll call it Elvis Presley, if you want me to - just as long as the revenue continues to flow on schedule.
Sure, you can apply a good NN to just about any trading idea, but you don't need to. I don't use an NN, but that is certainly one way to attempt to locate points of optimization.
Read what Constance Brown, has to say about Fibonacci confluence zones. More importantly, read what she has to say about where to locate the start of the range to be retraced by the market. She has some very interesting ideas on why most traders fail to establish the correct range and thus fail to derive the correct reversal zones. She outlines several places where the start of the range should be located and where they should not.
Most people that boohoo this stuff, have most likely not spent the time studying it. I'm an admitted late-comer to the Fibonacci Ratios. But, when I started applying them (the correct ratios) to my own work, I began to see the hidden value in using them at a much deeper level within my trading system. Establishing the correct range, is the first most important thing about using the ratios. Making sure that one is using confluence and understanding the difference between an Expanding -vs- Contracting market dynamics relative to Fibonacci Ratios, would probably be the second most important thing about the ratios.
I'm at the point now where I trade the CD and then the DP legs, where P = Projection. In some cases, I can trade the AB, BC, CD, DP legs. Those are like E-ticket rides at the amusement park, when they work out properly. They sort of make you feel like Superman, in a weird kind of way. Conditions typically are not appropriate for doing all four legs, so when I get one, I consider it a grand slam. I like being involved in all four legs because it affords you the opportunity to get stopped-out on more than you win, yet still walk away with a profit on the attempt. Fewer legs, don't offer that feature. It would be impossible to even attempt without the Fibo Ratios. Most of the bread and butter, however, comes from the CD/DP ratios. Anything I get beyond that is classified as icing, cake, ice cream and strawberries. The breakfast of champions.
But, underneath it all, is the appropriate setting of XA. Unfortunately, it is not where most people initially look.
Book ref: Fibonacci Analysis - by Constance Brown, Bloomberg Market Essentials (hard cover)
You can read my post in the Introduce Yourself section of the forum. That should help you to understand.
Is there less to trading than meets the eye - sure, for those who have found less to trading than meets the eye. For those of us that have not found less to trading than meets the eye, things are a bit more substantive and that causes complexity. The Boeing 747-8, on the surface, looks rather harmless and uncomplicated. The Bird itself, even looks fairly uncomplicated. Yet, both machines owe their simplistic surface features to an array of complex concepts and ideas.
I use a Digital GUI to trade with on a daily basis. It can't get more simple than looking at a Digital screen to make trade decisions. It has big numbers and text messages that come on screen, telling me what to do and when to do it. The WHY behind the simple GUI is the hard part, and that consists of a real-time database, engine and an integration layer. The Digital GUI is the layer that sits on top of the other three layers, which keeps the details out of my eye and off my screen. I don't need to see a thousand calculations being run every second while trading. That would simply inject a level of confusion into an already mission critical task. So, the single screen Digital GUI, filters and formats all that other "stuff" into something with big numbers, easy to read text and color coded graphics for ease of use. I have charts, but I don't need or use them to execute trades.
Trying to explain that kind of trading system in a forum such as this, would be like trying to unpack the details of a 747, or a Bird, in written form. If you have ever seen the EFIS maintenance manual on a 747, then you realize how impossible a task that would be and that's just the avionics. There are entire libraries of such manuals for the rest of the airframe on location at any maintenance facility authorized to conduct 74 maintenance. The concept behind trying to unpack this system is not the same, but it is fairly similar because it is a fully integrated system and not merely a set of loosely strung together indicators.
Take something like ZUP, for example. Nen, built an MT4 indicator and gave it to the world. It is not a fully integrated trading system, yet look at the massive amount of code the was necessary to build it. It tracks harmonic patterns automatically, but it is not an EA. It does not tell you when to enter the trade, nor does it provide a synthesized Trade Profile with specific entry, dynamic stop and dynamic limit. Yet, if asked, I'm sure that Nen, would not want to come online to hash-out the details of ZUP, as it would most likely consume the vast majority of his time. And, that's for a guy who actually wanted to make his work fully public. I have no intention of making my work fully public, nor was that the purpose of the posts I've made. Unpacking ZUP, would take eons for Nen, and its not even a fully integrated trading system. Case in point.
This thread is about why indicators don't work. My posts in this thread are about why they do. Since I do not use 40 year old indicators, I made mention of what I do use as a matter of record. I was then asked to explain terminology and did so, up to the level that I can. Beyond that, the task of explain a fully integrated system with posts on a trading forum, get evermore impossible with each keystroke. So, I can only talk about the system in conceptual form. Which is what I've done thus far.
The issue was not meant to be my system. The issue in this thread was indicators, the belief by some that they do not work and my belief to the contrary.
Hi... I wish you were inclined to teach your method, as I have found some of your posts intriguing. And FWIW I disagree with an earlier comment, paraphrased... that your posts are a waste of time since it's not fully divulged. Even a little information can lead one in a new and interesting direction.
Are you looking to fundamentals for causation, or something on the charts (maybe volume)?
I never said he had the indicator fundamentals right, but even if wrong it's still an indicator IMHO (and it doesn't even have any influence on usability of the derived data).
You have a very valid point about correlation of data! I think you hit the nail there. What I meant was kind of the opposite: if data has no correlation (ie coin toss) any application of an indicator is useless.
A nice example of correlation not implying causation: ToS
At the end it all comes back to: Never trust statistics you didn't fake yourself!
I have read a number of books on Fibonacci Analysis, including the book by Constance Brown, but also including
- Robert C. Miner: High Probability Trading Strategies
- Larry Pesavento + Leslie Jouflas: Trade What You See
- Carolyn Boroden: Fibonacci Trading
I prefer the book by Robert C.Miner, it is the most complete of the four. Carolyn Boroden's book is similar to Robert Miner's, but less comprehensive and not structured as well. Larry Pesavento's book is a fun read with nice charts, but only treats some of the harmonic patterns.
The book by Constance Brown is peculiar, because she uses strange points of reference. I have partially understood what she wanted to communicate, but I think that the ideas are too complex to be applied. Fibonacci analysis basically works as a self-fulfilling prophecy. Drawing lines that nobody else will detect is not the way to go.
I have observed fibonacci lines a lot. The self-fulfilling prophecy can be observed in real-time. When a major line is approached price will snap back immediately after having it touched. Now there are three options
- the snap back will be permanent, that is the retracement or expansion observed will show the exact Fibonacci ratio
- the snap back is followed by a testing of the value zone established by the last expansion bar that formed prior to hittiing the Fibonacci line
- the Fibonacci support or resistance will be broken
But this is more like a game than science. I first thought that Fibonacci ratios, which underly a number of natural phenomenons are also an organizing force of the markets, but this is not the case. Fractal dimension can take any ratios, and non-Fibonacci ratios are as likely as Fibonacci ratios.
I missed those three as I got a lot from Constance Brown.
She is verbose, like me and that's one of the reasons why I can relate to her writing style so well. She'll be discussing a particular ratio concept and then take you to ancient Mesopotamia, for a short lecture on the origins of the numbers themselves, before bringing you back to the charts. A well-rounded author and I appreciate writing styles like that. However, I do understand how that can also cause some confusion when it comes to remaining on track with what the author is attempting to convey. Constance, is a very sharp Woman - I can tell that by her depth of knowledge which goes well beyond mere trading. But, she does make it clear out the outset that you have to basically (essentially) re-learn how to plot your Fibonacci range starting price level, as that (she claims) is the first and biggest mistake that most traders make. Most traders take the absolute Swing Low/High as the starting price level. She says the vast majority of the time, that is not the Fibonacci range initialization level.
Now, I add other things to what she shows, but when I began using some of her initialization levels, like the second re-test of the absolute Swing Low/High, I began finding better confluence zones at the other end of the spectrum. Her writing style is a bit awkward, mostly because of the way she jumps around. But, she does seem to have a solid understanding of where the Expansion and Contraction price action initiates and how to capitalize on those transitions when they do occur by establishing the correct starting level for the range.
I'm personally seeing far too much harmony in other areas of my system using Fibonacci ratios, to call it a fluke. If I remove the ratios from my current body of work, I don't get the same level of precision that I enjoy today. There is a lot more to "price" than meets the eye. You can't just go take a Fib Tool, draw a line, plot some retracement ratios and then expect dollars to fall from the sky like manna from heaven. The question is: Retracement of What Exactly? At least, that should be the question that traders ask themselves before using the Fib Tool. The second question should be: Retracement from Where Exactly? Constance, covers these questions in pretty good detail.
Simple Swing Highs/Lows don't tell the trader enough about whether the market is Expanding or Contracting. There can be Swing Highs/Lows inside of [i]contracting markets[/b]. The inverse is also true. If the market is Horizontal at the time the Fibo starting level is plotted, then the trader will typically say: 'Gee! This Fibo stuff really is amazing!' But, next week, when the market begins to Expand or Contract, and the trader using the same Swing High/Low to assert the starting pint for the Fibo range, the trader will find a [b]lack of confluence and thus a lack of responsiveness to retracement levels.[b] This is extremely important and this is what Constance tries to explain. Where the trader initializes the Fibo range, depends on what type of market the trader is observing. And, there are only three (3) market types with two (2) behavioral patterns. (Bullish, Bearish or Flat, and Expanding or Contracting).
She's difficult to follow sometimes (I agree), but worth trying (reading) again.
If the range is established correctly.
Markets often Double or Triple Top/Bottom before reversing. These are zones where Fibonacci levels can be seen undergoing contraction before the next expansion.
Market has just entered a new expansion cycle.
The three (3) answers I just gave lend themselves to scientific exploration. Remember, there can only be three (3) market types and two (2) market phases (named above). That means that there can only six (6) different forms of Market Behavior. This helps you establish the technical matrix in which you can conduct your research/testing/analysis/etc. Price is perpetually moving between these six (6) different behavioral regimes and in doing, price exhibits normative patterns that can be detected and even predicted with a fairly high degree of probability, but only if these behavioral regimes are understood to exist. If one does not see them, then one cannot possibly set-up tools to measure their dimensions. Price cannot do anything else. If it does, then the market ceases to exist. Remember, there can be no trans-lateral movement in price, only Up, Down and Sideways, while Contracting or Expanding. That a six dimensional structure to measure and Fibonacci ratios, when applied correctly, can really help to identify which dimension is active.
I think the markets are "organized" by something completely different. However, Fibonacci ratios can act as genetic markers of the six (6) dimensions of behavior outlined above, IMO. One way to prove this to yourself is to take an M1 chart, reduce the window size to just a couple inches across (in width), zoom out as tight as you can get so that the bars appear as vertical sticks and then simply observe price action through this 2-3 inch M1 window.
If you watch long enough (it takes a while) you will begin to see that price repeats some of the same exact patterns, over and over and over and over and over again, without using any indicators or plotting any lines on the chart to observe them. You will start to be able to see double tops and double bottoms form before they break. You can see the Three Drives Pattern forming before it happens. You can see a whole lot of lower time-interval Harmonic patterns taking shape right before your eyes - all with no indicators drawn on the screen.
This exercise clearly demonstrates that patterns in the market do exist. And, where there is pattern, there is structure and where there is structure, there has to be a non-coincidental causality predicated or traceable through some kind of mathematical probe. Statistical Alchemy, Statistical Harmonization? I really don't know the answer to that (yet). But, that these patterns are observable and predictable to a fairly large degree when approached correctly, is unquestionable in my experience.
Anything that moves price is causation. Interbank is in control, not the retail component of FX. BTW - I'm an FX trader. I don't do equity options anymore and have not for years now. Fundamentals are by definition, baked into the Technicals and in FX, volume is impossible to really get a grip on, given the decentralized nature of the counter-party transactions that make it work. There is no central clearing in FX. Most importantly, as far as Volume accuracy is concerned, there are too many disparate sources of liquidity from intermediary to intermediary, so the volume that I see through my bank's platform, will not match the volume seem through another retail platform, or even another bank. FX is the purest form of OTC out there right now and if it did not have the massive daily liquidity (trillions), plus the reliability of strong global banks providing the backbone, I wouldn't want anything to do with the business - because it would then be nothing but a Globalized OTC crap shoot.
No - instead, what I look for are Delta Patterns in the OHLC stream. The differences in price points between various combinations of OHLC streams across multiple time-frames, are what my indicators are predicated on. My system looks for patterns and pattern matches in these Deltas, as well as patterns that have not yet been completed. No different than any other pattern completion approach to trading, except mine are not based on actual price points in the markets, but the delta between price points in the market. That gives me a whole different type of technical indicator to work with, because I can then take that Delta and basically run any kind of algorithm against it, to help me find correlation between Delta and Price (what you guys call "price"). For there, the system further extrapolates a Price Cluster, which essentially take "price" and expresses it as a Wave going through a Cycle.
That's what I look for in the market. Or, that's what my system searches for in the market, more to the point. I don't really care about price at all, quite frankly. Honestly, I hear people on radio talking about the Fed this, the BOJ that, and the ECB the other - I quite frankly really don't care. I just want price to move. I don't care which direction - I just want it to move and I want it to move with regularity and robustness.
It is sort of like that line in the movie Predator: "If it bleeds, we can kill it." Well, the way I see the market is this: If it moves regularly, I can map its behavior. If it stops moving - my system is blinded and handicapped. But, if the world's currency markets stop moving, the entire planet is in very serious trouble. That's one of the other reasons why I favor currencies over stocks/derivatives. Until the world moves to a single global currency (and that will happen one day), then I can always count on some country having the Kingpin Currency, to which all others will be measured and paired. That's a market with the kind of stability a trader needs. Deep liquidity, global reach and truly too big to fail (as long as you stick with the majors).
Thanks for sharing your ideas, but your answer feels more like a "grand theory of everything" than something that can be discussed. Describing markets via 6 different states by using direction and volatility as qualifiers, or using harmonic ratios to analyze price action are just basic concepts.
How do you apply these? Could you show us some examples, charts or trades derived from it?
