Why in the world should TA work ?!

[lolo55]

I need some help here please. Regarding market efficiency, particularly Technical Analysis and Price action. I really believe in TA and have observed that some patterns really has a positive expectancy (though not very big but big enough to beat random entry at least). My question is:
If anyone can easily recognize and trade the exact same pattern, then why in the world would anyone think he has an edge over the market ? I just can't get it, what am I missing here ?!

 

[tomorton]

Patterns are not the most important or helpful aspects of TA, they're not to be believed and traded off as if they were traffic lights. They may be handy guides for confirming what price action on the chart is already suggesting.

 

[timsk]

Hi lolo55,
Welcome to T2W.

I need some help here please. Regarding market efficiency, particularly Technical Analysis and Price action. I really believe in TA and have observed that some patterns really has a positive expectancy (though not very big but big enough to beat random entry at least). My question is:
If anyone can easily recognize and trade the exact same pattern, then why in the world would anyone think he has an edge over the market ? I just can't get it, what am I missing here ?!

When you say that you 'really believe in TA' - what do you mean by that? That's not a trick question and I'm not trying to be clever or funny - it's a serious question. TA has become an ill defined generic term that means different things to different people. For example, many traders regularly look at charts, but have little understanding of - or interest in - the basic tenets of Dow Theory. (If you're not sure of these yourself, this Sticky provides the briefest of introductions: Essentials Of Technical Analysis.)

Regarding your question, patterns that you may be able to observe do not in themselves provide a trader with an edge or a positive expectancy. There's nothing inherent in a chart that provides a trader with either of these: both are created by the trader himself in response to whatever it is that he thinks he observes in the chart. You could take a hundred traders who all understand and are able to identify a head 'n shoulders pattern - but very few of them will be able to trade it profitably and consistently over the medium to long term. If successful trading was simply about pattern recognition and nothing else - then most members of T2W would be fabulously wealthy!
Tim.

 

[barjon]

Hi lolo55,
............ then most members of T2W would be fabulously wealthy!......
Tim.

I thought they were

 

[tomorton]

I think the study of TA innately draws new traders towards unnecessary complexity. Its partly a result of our upbringing and all of our experiences in other fields. More knowledge must be better, more skills must be more efficient, more data must support better decisions, some training is better than none, etc. etc.

Forget all these innate assumptions in trading. They will help in a career as a GP but not in trading.

If you only know this you can start trading -
Successively rising (or falling) prices over a period of time make a trend. Trends tend to continue. Trends don't move in a straight line. Trends that have paused but not reversed will probably resume.

 

[dbphoenix]

I need some help here please. Regarding market efficiency, particularly Technical Analysis and Price action. I really believe in TA and have observed that some patterns really has a positive expectancy (though not very big but big enough to beat random entry at least). My question is:

If anyone can easily recognize and trade the exact same pattern, then why in the world would anyone think he has an edge over the market ? I just can't get it, what am I missing here ?!

Technical analysis is about behavior, particularly motivation: why are buyers and sellers transacting prices here and not there? Sometimes these transactions appear to form patterns, and since it's "easier" to look for patterns than to think about the whys of demand and supply, inexperienced traders look for the patterns. But the patterns are just bunnies in clouds. What matter are motive and desire.

 

[lolo55]

Patterns are not the most important or helpful aspects of TA, they're not to be believed and traded off as if they were traffic lights. They may be handy guides for confirming what price action on the chart is already suggesting.

For me any analyzing price action means nothing but detecting patterns that might work , not necessarily classical chart patterns.
What do you think ?

 

[lolo55]

Hi lolo55,
Welcome to T2W.

When you say that you 'really believe in TA' - what do you mean by that? That's not a trick question and I'm not trying to be clever or funny - it's a serious question. TA has become an ill defined generic term that means different things to different people. For example, many traders regularly look at charts, but have little understanding of - or interest in - the basic tenets of Dow Theory. (If you're not sure of these yourself, this Sticky provides the briefest of introductions: Essentials Of Technical Analysis.)

Regarding your question, patterns that you may be able to observe do not in themselves provide a trader with an edge or a positive expectancy. There's nothing inherent in a chart that provides a trader with either of these: both are created by the trader himself in response to whatever it is that he thinks he observes in the chart. You could take a hundred traders who all understand and are able to identify a head 'n shoulders pattern - but very few of them will be able to trade it profitably and consistently over the medium to long term. If successful trading was simply about pattern recognition and nothing else - then most members of T2W would be fabulously wealthy!
Tim.

I believe that price action contain some valuable information about balance of supply and demand, order flow created by more informed traders or market makers. I believe that price action shows the path of least resistance.

Sorry, but I tried to understand what do you mean but you lost me!
What I understand that plain vanilla price action can't provide an edge, but if used in combination with other distinct information it may work , Am I understanding right ?!

 

[lolo55]

If you only know this you can start trading -
Successively rising (or falling) prices over a period of time make a trend. Trends tend to continue. Trends don't move in a straight line. Trends that have paused but not reversed will probably resume.

This is very helping, thanks!
So you are saying that plain vanilla price action works, right ?

 

[lolo55]

Technical analysis is about behavior, particularly motivation: why are buyers and sellers transacting prices here and not there? Sometimes these transactions appear to form patterns, and since it's "easier" to look for patterns than to think about the whys of demand and supply, inexperienced traders look for the patterns. But the patterns are just bunnies in clouds. What matter are motive and desire.

I understand, so if i have a hypothesis that some pattern occurs and repeats because of underlying behavior of traders , maybe a panic, or greed , or gambling or rapped in a losing position, or whatever, does this make a difference. Still It's a pattern in price action. Knowing the reasons behind or not doesn't make a difference. What do you think ?

 

[barjon]

I understand, so if i have a hypothesis that some pattern occurs and repeats because of underlying behavior of traders , maybe a panic, or greed , or gambling or rapped in a losing position, or whatever, does this make a difference. Still It's a pattern in price action. Knowing the reasons behind or not doesn't make a difference. What do you think ?

The patterns you see may result from some sort of repeated behaviour or they may not. Understanding why they may be occurring is helpful.

Try charting a series of coin tosses and you'll see a lot of those self same patterns occuring, but sure as hell that's not going to help in determining what is more likely on the next toss.

 

[dbphoenix]

I understand, so if i have a hypothesis that some pattern occurs and repeats because of underlying behavior of traders , maybe a panic, or greed , or gambling or rapped in a losing position, or whatever, does this make a difference. Still It's a pattern in price action. Knowing the reasons behind or not doesn't make a difference. What do you think ?

If it made no difference, then the "head and shoulders pattern" would "work" every time. But it doesn't, largely because most of those who look for it and trade it don't understand why it's being formed. It's the behavior that determines whether or not the "pattern" will unfold in the expected way, not the pattern itself.

If following patterns were all there was to successful trading, the success rate would be many times higher than it is. Who can't follow a pattern?

 

[lolo55]

Still no answer to my question:
If anyone can easily know about and follow the same strategy; being pure price action or PA as a component in a larger but simple strategy, would not competition ruin the edge ?!

 

[barjon]

Still no answer to my question:
If anyone can easily know about and follow the same strategy; being pure price action or PA as a component in a larger but simple strategy, would not competition ruin the edge ?!

Your question is based on a false premise about patterns as has been discussed.

 

[forker]

The problem is that any given pattern has zero correlation with other seemingly mirror patterns. Market conditions are forever changing and no 2 intervals are the same (different players, different circumstances, different sentiment). So what I am saying is that any given price structure ultimately doesn't signal a given outcome. It's more a case of perceived patterns by the observer having significance when in fact it doesn't. This is the inevitable cost of TA

 

[dbphoenix]

Still no answer to my question:
If anyone can easily know about and follow the same strategy; being pure price action or PA as a component in a larger but simple strategy, would not competition ruin the edge ?!

That's the first time you've posed the question in that form, and the answer is no.

First, no two people are going to interpret the behavior behind the pattern in the same way due partly to the fact that no two instruments are traded exactly the same way.

Second, few people understand what "price action" is, instead trading bars or candles or patterns or even indicators. Those who trade the ladder probably come closest, but there are disadvantages to adopting this type of trading which make it unacceptable to many. And it isn't necessary to resort to the ladder in order to trade price. One can trade price using a daily chart.

Third, no two traders will have the same risk tolerance, the same goals, the same tactics, the same management protocols and so on. Nor will everybody be trading the same instrument in the same interval. One's edge lay only partly in the trading plan. If one can't or doesn't follow the plan, it becomes irrelevant and the edge evaporates.

 

[timsk]

. . . Sorry, but I tried to understand what do you mean but you lost me! What I understand that plain vanilla price action can't provide an edge, but if used in combination with other distinct information it may work, Am I understanding right ?!

Hi lolo55,
I'll try and clarify my comments.

A trader's 'edge' and positive expectancy are a by-product of a carefully thought out and tested trading plan. A part of that plan may involve identifying a particular chart pattern, e.g. head and shoulders (H&S). It's virtually impossible to say that H&S patterns in isolation 'work' X% of the time because there are so many other factors involved, not least the points at which the trade is opened, closed and where the stop loss set etc. Even if these are known and the trader does a mechanical backtest which, for the sake of argument, the result indicates that the pattern 'works' a mere' 10% of the time - this is still largely meaningless. The reason being that if the trader is able to discern why it doesn't work 90% of the time and only trades the 10% of patterns that do work, then he'll have a 100% winning methodology. But, to do that, the trader will have to look beyond the mere H&S pattern itself to examine the forces at play that create it in the first place, And that's the point that dbphoenix and barjon are trying to impress upon you.
Tim.

 

[tomorton]

For me any analyzing price action means nothing but detecting patterns that might work , not necessarily classical chart patterns.
What do you think ?

Well, as far as using price action's identifiable and repeating behaviour is concerned I suppose these could be termed patterns and traded off. Though the trade for me is safest based on the behaviour, not the pattern the behaviour depicts.

What I was getting at is the classic patterns like head-and-shoulders, double tops, bull flags and the like, are not enough for successful trading on their own for most people. I think there is a thing called the pattern trader but very few people seem to stay profitable this way.

Maybe pattern trading was successful at one time in history, some older books seem to suggest so. But the problem there is that most of the older classic trading books were written for US investors in a decades-long bull market. So surely not entirely applicable today.

 

[dbphoenix]

Well, as far as using price action's identifiable and repeating behaviour is concerned I suppose these could be termed patterns and traded off. Though the trade for me is safest based on the behaviour, not the pattern the behaviour depicts.

What I was getting at is the classic patterns like head-and-shoulders, double tops, bull flags and the like, are not enough for successful trading on their own for most people. I think there is a thing called the pattern trader but very few people seem to stay profitable this way.

Maybe pattern trading was successful at one time in history, some older books seem to suggest so. But the problem there is that most of the older classic trading books were written for US investors in a decades-long bull market. So surely not entirely applicable today.

Depends on how old the older books are. Schabacker, who started all this, wrote

When we become quite familiar with stock charts we shall find ourselves looking for various pictures and patterns formed by our charts, but if we are to be complete masters of our study and get the fullest benefits from our own analysis it is important that we do not entirely lose sight of the fundamental basis for the formation of those pictures and patterns.

That fundamental basis is in actual stock market trading, and actual stock market trading is the result of individual actions by many thousands of people, based in turn upon their own hopes, fears, anticipations, knowledge or lack of knowledge, necessities and plans. It is the danger of losing sight of this human element in stock charts that we must guard against, and since this human element is basic it may be wise to fit it into the foundations of our study at the very outset. (1933)​

 

[seekingTruth16]

The patterns you see may result from some sort of repeated behaviour or they may not. Understanding why they may be occurring is helpful.

Try charting a series of coin tosses and you'll see a lot of those self same patterns occuring, but sure as hell that's not going to help in determining what is more likely on the next toss.

If the coin is biased, then it might be possible to plot it, and detect that bias. The stronger the bias, the easier to detect.

So as a trader either one assumes the market has no bias in either direction, and in that case, TA would be pointless (and maybe trading too!), or there is a bias, and TA might be able to detect it.