If you want to fail as a trader, study TA

[The Expert]

I take it that both of you have not got it yet thencheesy:

:smart:TE:smart:

 

[jon1971]

I take it that both of you have not got it yet thencheesy:

:smart:TE:smart:

Just edited that post coz it was a bit rude...

Got what? All price quotes are an illusion, right? What is this, question your reality?

Well if you want to take it to 'n'th degree then all paper money is an illusion...

...You should read The Creature from Jekyll Island : A Second Look at the Federal Reserve, you'll like it..

http://www.amazon.com/Creature-Jekyll-Island-Federal-Reserve/dp/0912986212

Where does money come from? Where does it go? Who makes it? The money magicians' secrets are unveiled. We get a close look at their mirrors and smoke machines, their pulleys, cogs, and wheels that create the grand illusion called money. A dry and boring subject? Just wait!

 

[The Expert]

Just edited that post coz it was a bit rude...

Got what? All price quotes are an illusion, right? What is this, question your reality?

And still has not got it cheesy:

I can just imagine what type of charts you look at

:smart:TE:smart:

 

[The Expert]

Well until I got to grips with the psychology of trading I wasn't really making any profits!
This book was one of many books and things that helped with that. let's say discipline is much more difficult to master than a trading method.

I mentioned the book here because this is in the psychology section and this book is probably much more useful as a psychology book than a "this is exactly how I do it" book.

One man's GAIN is another man's LOSS

 

[The Expert]

Just edited that post coz it was a bit rude...

Got what? All price quotes are an illusion, right? What is this, question your reality?

Well if you want to take it to 'n'th degree then all paper money is an illusion...

...You should read The Creature from Jekyll Island : A Second Look at the Federal Reserve, you'll like it..

http://www.amazon.com/Creature-Jekyll-Island-Federal-Reserve/dp/0912986212

Where does money come from? Where does it go? Who makes it? The money magicians' secrets are unveiled. We get a close look at their mirrors and smoke machines, their pulleys, cogs, and wheels that create the grand illusion called money. A dry and boring subject? Just wait!

The Expert only reads for tactical purposes and sees no value in such a book.

However, your sharing intention is acknowledged, as some may be interested:?:

Did you get it YET

 

[Traduk]

Totally agree re. part of a larger view, this seems pivotal in defining an effective TA strategy, and it's what the anti TA crew is seeming not to acknowledge.

Sounds fairly similar to my approach, 'cept I use only 2 TFs the set-up and the higher for confirmation, I am looking for %K hooking or a possible mini-divergence pattern, I use a slightly different approach with %D (both on short settings), I would consider stochastic to be only one of five components that I use for decisions and I would score the %K and %D patterns with a max possible score of 20% towards my overall trade probability.

Also, I would wait and possibly disregard the initial pull back as I want to see what I define as a trend being in place, before getting in (which basically involves angle of MA & MA crossovers).

I then only get two bites of the cherry as I would not enter what I define as an over extended trend (which means that I may sometimes miss super trends, but I don't care - like I said I only play the probabilities), I've cannibalised Elliott and use it JUST to gauge average trend structure, I'm not into third wave extensions and all that malarkey, I leave all that to the true Elliott Wave casualties.

I must stress that I use structure as its constructed by market participants in real time. It is the core of everything and over-rides all else. Stops are structural and involve participants to change trend to take me out.

Stochastics have a variety of uses when tied to the fractal structures of interest but their drawbacks (which you appear to know) means they are useful in some situations but not others. They are a useful tool in the arsenal.

Your use of MA's suggest that you primarily want to trade trends and seek quite a lot of confirmation. If what you trade trends a lot and in meaningful runs then fine because chop is where most money is lost.

If your rebuild is via cherry picking trends in markets that have meaningful runs you should be OK providing you are happy with the testing that you have done. Good Luck.

BTW if busting accounts is a badge of honour, I busted three and they were big by today's standards. Resolve, determination and loads of lateral thinking turned that around and rendered those losses as needless tuition expenses. I could have done with the help of an Expert then but none were around. Nothing changes does it

 

[Traduk]

And still has not got it cheesy:

I can just imagine what type of charts you look at

:smart:TE:smart:

I use CQG and as you may be aware it is not exactly cheap. I apply a rule as with everything in trading that as a business my total costs must represent a small proportion of earnings. You can work out the rest as you are purportedly the Expert in deductive thinking.

What does The Expert use?. Promise I will not laugh if its a freebie but maybe I will.

You have been busy tonight. Been celebrating your big win?. The year is young and with luck you might get another in 2010.

 

[The Expert]

I decided to do a little test.

It took me exactly 4 minutes and 38 seconds to find the book, and, the other two have still not GOT IT YET

SUMMARY
The American dollar has no intrinsic value. It is a classic example of fiat money
with no limit to the quantity that can be produced. Its primary value lies in the
willingness of people to accept it and, to that end, legal tender laws require them
to do so.
It is true that our money is created out of nothing, but it is more accurate to say
that it is based upon debt. In one sense, therefore, our money is created out of
less than nothing. The entire money supply would vanish into the bank vaults and
computer chips if all debts were repaid.
Under the present System, therefore, our leaders cannot allow a serious
reduction in either the national or consumer debt. Charging interest on pretended
loans is usury, and that has become institutionalized under the Federal Reserve
System.
The Mandrake Mechanism by which the Fed converts debt into money may
seem complicated at first, but it is simple if one remembers that the process is
not intended to be logical but to confuse and deceive. The end product of the
Mechanism is artificial expansion of the money supply, which is the root cause of
the hidden tax called inflation.
This expansion then leads to contraction and, together, they produce the
destructive boom-bust cycle that has plagued mankind throughout history
wherever fiat money has existed

 

[The Expert]

I use CQG and as you may be aware it is not exactly cheap. I apply a rule as with everything in trading that as a business my total costs must represent a small proportion of earnings. You can work out the rest as you are purportedly the Expert in deductive thinking.

What does The Expert use?. Promise I will not laugh if its a freebie but maybe I will.

You have been busy tonight. Been celebrating your big win?. The year is young and with luck you might get another in 2010.

CQG, and I bet you used to use e-signal as well

You are confirming how stupid you really are:cheesy:

The Expert uses the best package available to man, his BRAIN

I think we will call you Tradmuk from now on

What a bunch of silly little "traders" pop up on these sitescheesy:

:smart:TE:smart:

 

[The Expert]

I must stress that I use structure as its constructed by market participants in real time. It is the core of everything and over-rides all else. Stops are structural and involve participants to change trend to take me out.

Stochastics have a variety of uses when tied to the fractal structures of interest but their drawbacks (which you appear to know) means they are useful in some situations but not others. They are a useful tool in the arsenal.

Your use of MA's suggest that you primarily want to trade trends and seek quite a lot of confirmation. If what you trade trends a lot and in meaningful runs then fine because chop is where most money is lost.

If your rebuild is via cherry picking trends in markets that have meaningful runs you should be OK providing you are happy with the testing that you have done. Good Luck.

BTW if busting accounts is a badge of honour, I busted three and they were big by today's standards. Resolve, determination and loads of lateral thinking turned that around and rendered those losses as needless tuition expenses. I could have done with the help of an Expert then but none were around. Nothing changes does it

Fractal structure and stochastics

I think you better THINK before you ACT, but then again, you are one of the majority, so why would we expect anything else

It is obvious that you will continue to "follow" the market, when, The Expert and a few others follow the Money

Imagine, you actually think you have something to offer aspiring traders, and, after losing 3 times you would imagine that you would have at least cracked the basics by nowcheesy:

:smart:TE:smart:

 

[DionysusToast]

My presumptions as you put it are based on your many posts regarding back testing and tweaking. Your stance against TA must have a basis in something and there is little else that you have said to go on I have only assumptions to go on.

It would be easier to tell you how to fly a light aircraft or a helicopter because there are a limited number of variables and if you get it wrong the questions stop

If you understand the fractal nature of market structure and accept that the component development from ticks upwards affects different players in different ways then you may appreciate that in order for anything to work properly both a trader and whatever they use must be tuned to the players beneath and those above. In essence a trader must view and place themselves within the market structure that suits there requirements and with knowledge of the forces around them. I use five charts and a DOM running over two screens to see what's below and above. Trading is done off three other computers and screens.

Market structure does not fit computer friendly time frames but tick aggregation, constant volume bars (based on trades or ticks) can overcome time based charts. P&F does similar as does Renko. In other words data can be displayed as suits the user and the market dynamics and is not the stereotypical time bars\candlesticks or whatever. An innovation exists with the order book whereby much can be done with trades to create delta's off the bid\asks traded and these can display any look back period of positive or minus order flow which can in turn show where the only players that matter entered the market. If movers and shakers are long and are not selling then the fire they lit just needs a little more fuel from them if the little guys push price down. The fuel is the little guy's money as they are squeezed out.

This is rambling and probably not what you wanted. OK a bit of 101, is the market trending or ranging. If a clearly defined range is in place count the bars from top to bottom in a short time-frame. If the range provides enough points for a worthwhile risk\reward the use an oscillator with a setting that is slightly less than the bar amplitude of the range and surf away until it breaks out. Being within or close to the amplitude is essential. Early recognition of a range is also essential and most wait until its almost over before joining in. In small ranges it is best to sit it out because getting trampled in the carnage hurts.

If it breaks out and after a pullback shoots off into a trend then oscillators are no longer useful as they jam against top or bottom and give false signals. There are dozens of trend tools but one that will get a trader over minor pull backs should be chosen and if all else fails a well tuned SAR is often better than nothing. Of course a trend indicates following not feeding so contra trend is for the fleet of foot. Yet again in trends, picking timing is essential but even more essential is timing whatever is used to as an indicator within the timeframe. As a general rule 9 positive bars is in candlestick terms is a full belly so using a short fast indicator is preferable to something that starts back in the annals of history.

This is all fairly basic stuff and not even a microscopic view of what I use but I do operate within a total news vacuum and follow the money. The money moves markets and I suppose that my emphasis is on what they do far more than any indicator.

Pick holes in that In trading there is a correlation with squash. Looking upwards skill levels are incomprehensible to those below and each player must make transitions whether physical or tactical on their own. I used to coach but could never explain the inexplicable to those I trained, nor can anybody.

So, in essence, you are saying that textbook TA does not work in isolation which is exactly my point.

All of the above is all welll and good if your goal is obfuscation and attempts to make yourself seem smart to newbies and impress them with the number of screens you have. The part about 'you have to understand the fractal nature of the market' is another (albeit less oft stated) interweb truism.

Also - the part about you being so smart and that no-one below your level will understand is rather lame. It give you an out, when we dont need 'outs' - we need to seperate the wheat from the chaff.

So - why not just focus on one of the aspects - such as 'fractal nature' and we can discuss why you think that is the case and arguments for/against.

I am guessing this wont happen as not a single proponent of using Traditional TA versuse a holistic approach can explain why what they claim to use works.

 

[trendie]

Are we there yet?
After more than 600 posts, we are nowhere nearer a definitive conclusion.
Or, indeed, any conclusion. (I always thought it was impossible to prove a negative, so proving TA doesn't work is on a hiding to nothing.)

It seems to be a string of single peculiarities of specific situations which are then, translated into a (mistaken) general conclusion.
There is also the whiff of the needing to be right, and my-methods-better-then-your-method.

Does it even matter if TA works of not?
Does the fact that some do indeed use TA successfully mean they are wrong, and their account growths over the years are illusory?
Does the fact that others don't use TA and make money any less successful?

I think in the race to get-one-over-the-last-guy we've lost the plot of this thread.

What is the point of this thread?
What is meant to prove, or disprove?

 

[jon1971]

Conservative cherry-picking for sometime to come, i'm currently trading to hone skills and with that in mind my primary objective is playing defensive. Thanks Traduk.

I think that with regard to the one-up-manship, I honestly don't think TA is a better method than any other, I just find that viewing the Market mathematically works for me, I'm comfortable with that.

I would add that (whilst we're talking about smoke & mirrors) the only constants are price and order flow, mostly everything else is open to manipulation & exploitation (IMHO).

 

[jon1971]

Ps. TE to think that I would place a trade based purely on structure & stochastics is a mistaken assumption on your part, these are just two elements of my method that could provide no more than 40% towards my trade total probability scoring system.

With JUST those odds, as i define them, I'd be better off playing black or red on roulette.

Hmm some light cloud, but likely to burn off by mid morning...

 

[Traduk]

So, in essence, you are saying that textbook TA does not work in isolation which is exactly my point.

All of the above is all welll and good if your goal is obfuscation and attempts to make yourself seem smart to newbies and impress them with the number of screens you have. The part about 'you have to understand the fractal nature of the market' is another (albeit less oft stated) interweb truism.

Also - the part about you being so smart and that no-one below your level will understand is rather lame. It give you an out, when we dont need 'outs' - we need to seperate the wheat from the chaff.

So - why not just focus on one of the aspects - such as 'fractal nature' and we can discuss why you think that is the case and arguments for/against.

I am guessing this wont happen as not a single proponent of using Traditional TA versuse a holistic approach can explain why what they claim to use works.

Many years ago I bought loads of cd's of tick data from various markets and to prove what I had been constructing on paper enlisted the aid of a programmer to run the data through progressive iterations using the filters I had created on paper. Every data set run displayed larger and larger levels of fractal development which is unsurprising but what was a revelation was that every development was limited in shape and dimension so that it fitted or was constrained by its place within the developing structural hierarchy.

Dow and Elliot where both right but Elliot was too clever with his rules many of which are not true within the natural development of fractal expansion.

Within this self organising chaotic structure the larger structure dictates what the structures below can do to conform with the larger development and that ripples down through every iteration. What frustrates people and wrecks a heck of a lot of TA is that all developmental stages are running concurrently and leads to how the hell do I make it work if the markets are up\down\up in three different time frames.

The answer is to understand where the iteration that you are trading fits in the overall scheme of things and using the governing factors tolerate the influences from below and above. From that standpoint I organise what I trade, how I present data, what I apply to it and how I trade.

I have no idea why this fractal development occurs in a self governing manner. I could put forward many theories but why bother when it has happened from at least back, in my case to the earliest data I bought and was seen by others way back in time. How to extract each iteration is the key and working out what to do with it unlocks the vault.

If you saw the runs against data you would be won over instantly but you ain't going to see it nor is anybody other than the two that did the work. I supplied all the theory and filter methodology and the programmer, programmed.

The underlying aggressive nature of some of your reply speaks volumes.

My kit is fairly good, offers multiple redundancy but is very unimpressive compared to many who have set ups that look like they are modelled on NASA. I have what I need, nothing more, nothing less but it ain't no big deal.

Did I come across as trying to project myself as smarter than others?. Sorry about that. I am obviously not smarter than you because it took me years to find my answers and you found yours in a 6 hour course. On a cost and time efficiency factor you're the one with the smarts.

I realise that I am proving you right and pointing out that canned stuff cannot work but some people might think and look at a chart and actually see that there is little point in running something that looks back too far into history and try to get within the timescale of focus. Some may ask the same questions as I did decades ago and ask themselves why something doesn't work, break it down and make it work with changes. Constructive discussion promotes thought but I don't think you have constructive desires.

 

[Traduk]

CQG, and I bet you used to use e-signal as well

You are confirming how stupid you really are:cheesy:

The Expert uses the best package available to man, his BRAIN

I think we will call you Tradmuk from now on

What a bunch of silly little "traders" pop up on these sitescheesy:

:smart:TE:smart:

Nope never used e-signal. Graduated up through some others but by-passed that one. That's one bet you lost.

We all use our brains. Using the computer analogy GIGO is a problem and garbage out warrants investigation. There are experts in that field but as a layman I suspect processor malfunction.

You may call me whatever you wish because I care not. You are using plural again (we) ... fertile or should that be futile imagination or fact?.

You are right. Silly people do pop up on sites. They are usually well tolerated and provide much amusement until the time that the village, missing its star attraction calls them back.

 

[Mysteron]

In the distant past, in order to understand TA, I read: Murphy, Technical Analysis of the Financial Markets and I was struck by the sheer multitude of indicators that could potentially be used for trading. The motivation behind studying TA was that I had lost money trading UK stocks and I thought that TA may possibly improve my performance as a swing trader where I hold positions from days to a couple of months.

At the beginning of 2009, in order to evaluate some aspects of TA, I decided to model a mechanical trading system using variations on Bollinger bands, MACD, MACD histogram, RSI and slow stochastics. The model had parameters to represent the overbought and oversold thresholds for RSI and slow stochastics, the bollinger band sigma factor, the bollinger band period, stoploss as a % factor of buy price, profit factor as a % of buy price at which point a trailing stop loss would be implemented, trailing stoploss % of peak price reached. A moving average of the share price slope was used to determine if a stock was bullish or bearish. The model had 14 parameters and the various indicators could be enabled or disabled depending on the strategy I was investigating. Each indicator could generate a buy or sell signal and if they occurred sufficiently close together in time then an overall buy or sell signal was used to initiate or end a trade. The model used historic share price data of UK stocks downloaded from Google finance and only used daily closing price data. The reason for that is because I work full time and can only make trading decisions using data available at the end of each day. The model was coded in Fortran and Matlab used to plot results. The model would simulate the effect of the trading strategy over the share price data over a period of 5 years and each trade always invested £5000. At the end of the 5 year period the total profit or loss was used as a measure of trading performance. The model had an optimisation facility that would adjust the parameter values to maximse the profit. The duration of the trades lasted from a few days to a few weeks and was determined by the model.

At first I used the model to generate buy/sell indications for stocks that I was interested in buying and my live trading was successful. I made a profit over the whole of last year, not a fortune but many times the return than if I'd left my money in a savings account. I wondered to what extent the model was actually helping me, so the model was modified (to allow forward testing and back testing) to optimise trading performance over a sliding window of 250 days over the 5 year share price history. For each day in the 5 year history the trading parameters were optimised over the preceding 250 days and those parameters were used in the trading strategy to generate a buy/sell/nothing signal for that day. That was repeated for each successive day over the entire 5 year history. The trading performance was again evaluated by the total profit achieved. This modification was tested on a range of stocks with various combinations of the indicators. Various modifications to the model were made to improve it. A forgetting factor was used in the optimisation process so that the most recent trades in the sliding window would have greater weight. The Bollinger band fixed period of 20 days was discarded in favour of a variable period chosen in relation to the largest period in the share price data over the 250 day sliding window. That period was determined from the amplitude spectrum of the share price data (my professional background is in control systems and signal processing so modelling, simulation and data analysis is routine). As a result the model could adapt to changing conditions represented by the share price data, ie bullish, bearish or sideways.

So what was my conclusion following this effort to model a mechanical trading system using TA alone?

Answer: TA on its own does not work!

At least that is true for the set of indicators I looked at in the model when applied to UK stocks.

That is my conclusion despite the fact that I actually made a profit. Having thought about this I feel that I was using the model as additional information (or a crutch maybe) as an aid to decide whether to trade or not and ultimately it was my own ability to assess a price chart which led to success. Then again maybe it was because during much of 2009 the markets were recovering so it was easy for a novice to profit in a benign trading environment. I rarely use my model now and my trading of UK shares so far in 2010 has still been successful but less so than in 2009, but then February was a bad month due to the debt problems in Greece.

Now that my personal quest for the Holy Grail of Indiators has ended I still look at basic indicators (MACD histogram, RSI, Stochastics, Bollinger) as an initial screening of stocks worthy to buy but I am now more inclined to look at volume, price and support/resistence to make the ultimate decision to trade. Volume spread analysis (VSA) and it's inferences of how smart money is operating has grabbed my interest in recent weeks.

As far as FA is concerened I limit that to a look at the financial calander of a company to determine if there are any forthcoming events that may affect share price and I also find out what other investors concerns are about a company.

I have only traded UK shares and have had a modicum of success after my anguish at loosing a few years ago. I've tried a demo account for CFD trading from Ample Interactive Investor but was suprised to find that their trading platform does not show volume on the tick charts so a trading strategy using VSA isn't possible. My attempts at CFD trading (on weekday evenings after work) have been a dismal failure so far but fortunately its only involved play money on a demo account.

 

[Mr. Charts]

Good post, Mysteron.

 

[trader_dante]

The reason trader's fail has nothing to do with using TA whatsoever and anyone that tells you that is a complete idiot. If you want to increase your chances of failing as a trader when you are starting out, you can start by risking 2% of your account and use stops. That should keep you slowly bleeding until you give up.

 

[temporarilylight]

I get a feeling that 'no stops' is gonna be the next 'thing' on t2w.
These things work in cycles as to what's popular:
Something like:
No stops - tight stops - Price action - TA is useless - FA - FA is useless - Pinbars - VSA -
No stops - tight stops - Price action etc etc

PS - I like DT. I think he's saved this thread from being a total 100% failure, and if anyones 'made me think' its probably him (although im still pro TA )

Having said that, is price action TA for example?? We need to al be on the same page to advance discussion i think