Technical Analysis Doesn't Work

[rcanfiel]

Quote from JSSPMK:

Q: If you take a person that can successfully make trading decisions based on technical analyses and another that states that technical analyses doesn't work, then in that case what opinion can we be of a person that arrogantly disputes a fact by proving his personal inability? A: An idiot.

If there are reams of studies that TA doesn't work, then the problem is with the person who claims to make successful decisions based on TA. Thus you have the idiot backwards.

Technical analyses is not exact science, but a probability study based on personal research and in most cases theoretical, practical and financial hardships.

No. TA is a belief set, strongly held in spite of the available evidence.

 

[JSSPMK]

All I can say is - LOL.

something like religion is based on a belief, ability to use or not to be able to use technical analyses depends on hard facts, what you or others believe has nothing to do with another person's experience.

buy low, sell high - is part of technical analyses, you might not consider it to be or believe so, but it is.

 

[Thunderdog]

Quote from rcanfiel:

...In their recent paper entitled "Can Commodity Futures be Profitably Traded with Quantitative Market Timing Strategies?", Ben Marshall, Rochester Cahan and Jared Cahan investigate the effectiveness of 7,846 quantitative trading rules from five rule families (Filter, Moving Average, Support and Resistance, Channel Breakouts, and On-Balance Volume) for 15 kinds of commodity futures contracts...

Presumably, you understand that a complete trading method involves a combination of rules calibrated to work with one another as a unified whole. It does not involve just slapping together whatever happens to be in the refrigerator at the moment. Presumably, you understand that arriving at a usable system involves testing, tweaking, testing some more and then more adjustment until you arrive at something that is not overfitted and works on balance. As you might imagine, the effort is nuanced and is not measured solely by its quantity. (On the other hand, the production of academic papers is often fueled by the "publish or perish" rule. This is not to suggest that meaningful academic research is not performed, because that would be a foolish statement. However, sadly, there is no lack of academic "scholars" principally motivated by quantity of output, since research grants and other forms of compensation are often dictated by quantity of output.)

Are we to assume that the authors of the study you referenced took the kind of time and care it takes to develop a usable system, to arrive at their collection of 7,846 quantitative trading rules? That it involved a careful collection of criteria that were preceded by careful market study over an extended period of time? By way of example, to be able to prepare a gourmet meal requires more than just a list of its essential ingredients. It is a matter of how and when the ingredients are used that makes the difference. Merely pouring some, all, or even more of the essential ingredients in a vat and then stirring, will likely yield different results than those desired.

 

[QQQShort]

Quote from Thunderdog:

Presumably, you understand that a complete trading method involves a combination of rules calibrated to work with one another as a unified whole. It does not involve just slapping together whatever happens to be in the refrigerator at the moment. Presumably, you understand that arriving at a usable system involves testing, tweaking, testing some more and then more adjustment until you arrive at something that is not overfitted and works on balance. As you might imagine, the effort is nuanced and is not measured solely by its quantity.

Are we to assume that the authors of the study you referenced took the kind of time and care it takes to develop a usable system, to arrive at their collection of 7,846 quantitative trading rules? That it involved a careful collection of criteria that were preceded by careful market study over an extended period of time? By way of example, to be able to prepare a gourmet meal requires more than just a list of its essential ingredients. It is a matter of how and when the ingredients are used that makes the difference. Merely pouring some, all, or even more of the essential ingredients in a vat and stirring, will likely yield different results than those desired.

Well said.

 

[PitchBlack]

Quote from Thunderdog:

Presumably, you understand that a complete trading method involves a combination of rules calibrated to work with one another as a unified whole. It does not involve just slapping together whatever happens to be in the refrigerator at the moment. Presumably, you understand that arriving at a usable system involves testing, tweaking, testing some more and then more adjustment until you arrive at something that is not overfitted and works on balance. As you might imagine, the effort is nuanced and is not measured solely by its quantity.

Are we to assume that the authors of the study you referenced took the kind of time and care it takes to develop a usable system, to arrive at their collection of 7,846 quantitative trading rules? That it involved a careful collection of criteria that were preceded by careful market study over an extended period of time? By way of example, to be able to prepare a gourmet meal requires more than just a list of its essential ingredients. It is a matter of how and when the ingredients are used that makes the difference. Merely pouring some, all, or even more of the essential ingredients in a vat and stirring, will likely yield different results than those desired.

Indeed, an excellent ppsting.

It also clearly shows that people try to impress newbies by throwing huge numbers of tests on the tabel. But every professional knows that even small system development takes lots of time. The test as mentioned would take years and also a lot of professional and wel seasoned traders to get reliable conclusions.

It took me 10 years before the system, that i developped, made consistent profits. But i'm not as smart as the authors of the article.

Thunderdog shows more knowledge about trading than the authors. That's probably because he is trading instead of writing.

 

[NihabaAshi]

Quote from Thunderdog:

Presumably, you understand that a complete trading method involves a combination of rules calibrated to work with one another as a unified whole. It does not involve just slapping together whatever happens to be in the refrigerator at the moment. Presumably, you understand that arriving at a usable system involves testing, tweaking, testing some more and then more adjustment until you arrive at something that is not overfitted and works on balance. As you might imagine, the effort is nuanced and is not measured solely by its quantity. (On the other hand, the production of academic papers is often fueled by the "publish or perish" rule. This is not to suggest that meaningful academic research is not performed, because that would be a foolish statement. However, sadly, there is no lack of academic "scholars" principally motivated by quantity of output, since research grants and other forms of compensation are often dictated by quantity of output.)

Are we to assume that the authors of the study you referenced took the kind of time and care it takes to develop a usable system, to arrive at their collection of 7,846 quantitative trading rules? That it involved a careful collection of criteria that were preceded by careful market study over an extended period of time? By way of example, to be able to prepare a gourmet meal requires more than just a list of its essential ingredients. It is a matter of how and when the ingredients are used that makes the difference. Merely pouring some, all, or even more of the essential ingredients in a vat and then stirring, will likely yield different results than those desired.

The best post of this entire thread.

Unfortunately, rcanfiel has excluded this type of TA from this discussion to concentrate on TA in its simplest form...

The TA stuff that's not being used by profitable traders as is.

Heck, throw in different trade management rules (stuff after entry) and the results will differ greatly than the results of someone else using another different type of trade management.

(Anybody notice how the trade management stuff is NEVER discussed in any detail by those he's been quoting).



Kind'uv reminds me of another system designer at ET that showed up one day in one of my threads to say a particular pattern he tested doesn't have a positive expectancy.

Yet, after many pm's of trying to get him to reveal his trade management rules...

He eventually gave in reluctantly to disclose his trade management rules that almost made me laugh and feel sorry for him upon examining his trade management info.

I simply told him take profit here via a particular exit strategy and convert the initial/stop loss protection into a profitable trailing stop when price does something in particular...

Guess what?

His results change from negative to positive and I haven't heard from him since that discussion.

Not even a thank you nor an admittance that there's more to trading than the entry signal.



Mark

 

[kut2k2]

Quote from Thunderdog:

Presumably, you understand that a complete trading method involves a combination of rules calibrated to work with one another as a unified whole. It does not involve just slapping together whatever happens to be in the refrigerator at the moment. Presumably, you understand that arriving at a usable system involves testing, tweaking, testing some more and then more adjustment until you arrive at something that is not overfitted and works on balance. As you might imagine, the effort is nuanced and is not measured solely by its quantity. (On the other hand, the production of academic papers is often fueled by the "publish or perish" rule. This is not to suggest that meaningful academic research is not performed, because that would be a foolish statement. However, sadly, there is no lack of academic "scholars" principally motivated by quantity of output, since research grants and other forms of compensation are often dictated by quantity of output.)

Are we to assume that the authors of the study you referenced took the kind of time and care it takes to develop a usable system, to arrive at their collection of 7,846 quantitative trading rules? That it involved a careful collection of criteria that were preceded by careful market study over an extended period of time? By way of example, to be able to prepare a gourmet meal requires more than just a list of its essential ingredients. It is a matter of how and when the ingredients are used that makes the difference. Merely pouring some, all, or even more of the essential ingredients in a vat and then stirring, will likely yield different results than those desired.

Outstanding post!

 

[kut2k2]

Quote from NihabaAshi:
(Anybody notice how the trade management stuff is NEVER discussed in any detail by those he's been quoting).



Mark

Oh yeah!

I often wonder if there's just some secret academic agreement to be all-in or all-out when they "test" those kludged "strategies." I also wonder how many of those guys have even heard of position sizing.

 

[SupremeTrader]

I think this senseless Thread has gone on for too long. If you use Voodoo Magic and it makes money for you, fine. If you use Technical Analysis and consistently make money with it, great. If you use your Intuition and make money all the time, just fantastic. If you just dream of trades at nights and they make money for you when you execute them, fantastic.

The bottom line is, are you making money with whatever system you are using?

If there are Professors or Research Gurus who are for or Against Technical Analysis, tell them to trade with their own money using their system or no system and show us their profits. If they cannot do this, then they should simply shut up and those for or against them should also keep quiet.

Trading is a business that encompasses the simultaneous culmination of self-management, portfolio management, money management, risk management and enormous discipline.

TA is just a miniscule aspect of it. And TA alone cannot make anyone money. Period. What the heck is this discussion about then?

 

[luckyluciano]

keep telling everyone TA doesn't work!!! but it sure does make me alot of money!!!

As for position sizing...like peter lynch says..."deversification is called DEWORSIFICATION!"