[suriNotes]

Pappo is correct with his assessment of Pattern logic and coding/testing patterns.
In my experience, patterns do have excellent reliability with great repeatable logic.
Many successful traders have implemented patterns/code logic to take advantage of them.
Most pattern success or failure depends on knowing the pattern logic with market
context and knowing how/where they are forming. Without market context logic,
testing patterns is a futile exercise. Programmers tend to miss this logic when testing
patterns. Also, individual mastering of patterns requires thorough market knowledge
and experience and mastery may be limited to just few patterns. Coding patterns may be
more difficult for even most seasoned programmers too.


Excellent trading ideas are discarded due to poor coding and poor back-testing methods.
I have seen this many times in my career as many brilliant programmers with great ideas
but with little or no trading knowledge discard their concepts as not profitable due to
novice back-testing methods/results. Also, each trading platform have various trade
execution logic and hence many Automated Trading Systems back/forward testing results
suffer from wrong/poor execution methods.


Here is an example of a recent Bloomberg Article where TA was slandered.

"Stock Charts Fail Forecast Test in Complete S&P Miss (Update3)" by Michael Tsang and Eric Martin

Stock Charts Fail Forecast Test in Complete S&P Miss (Update3) - Bloomberg.com

Tsang and Martin (journalists) with little TA and no real trading/investing/programming
experience published an article and slandered the TA and indicators due to their wrong
back-testing results... Bloomberg published the article as it was written by their journalists.
Many TA experts (Murphy, McClellen et. al.) have written their concern to Bloomberg.
The editors at Bloomberg have softly admitted but sadly it was never removed.

Regards,
Suri

[programmer]

For me the idea that a market is overbought or sold is absurd, yet people use stochastics like a "grail machine"....5,3,3 no 15,9,9....No 8,3,3.....No.....the Angle.....No the Cross of the stochs.....No.......the Macd with the stoch ...No...oh wait the 15 min must align with the 5 min.....No....wait the
Ema 34 must retrace before the CCI will give us a hook, oh yes , oh yes, oh yes, Meg Ryan in Harry met Sally now screams in the restaurant.......Eureka.. ........OH NO, There was news and the stochs of the 15m got tangled with the 5 and the leading index is at its monthly high plus the stoch is overbought, yes yes yes and now even Harry joins Meg ....oh no The fed killed us with the news, bloody hell maybe next time I will use a faster MA to compensate for the lack of feed accuracy due to my broker...oh wait that it, the feed is broken, I will buy a better feed : CQG(XXXX$), Bloomberg(xxxxxxxxx$) thats its , that was the problem the feed, oh wait its not the feed its the fact that I have to pea while trading, so I will read trading in the Zone again to distract my pain of trading through my Urinal glans directly unto the stoch 15,9,9 on the 3 Min which is the way to go, but wait my platform does not have Range Bars and everybody knows range bars are the way to go, unless the bottleneck is the CPU which is not strong enough to play DOOM and trade at the same time, so I will get a better CPU , oh wait, I have thunder storms so a great UPS is required....YES EUREKA ITS THE UPS GUYS, Much better than any technical analysis tool out there.


daily candle is it....lol.

[UrmaBlume]

Originally Posted by programmer »

For me the idea that a market is overbought or sold is absurd, yet people use stochastics like a "grail machine"....5,3,3 no 15,9,9....No 8,3,3.....No.....the Angle.....No the Cross of the stochs.....No.......the Macd with the stoch ...No...oh wait the 15 min must align with the 5 min.....No....wait the
Ema 34 must retrace before the CCI will give us a hook, oh yes , oh yes, oh yes, Meg Ryan in Harry met Sally now screams in the restaurant.......Eureka.. ........OH NO, There was news and the stochs of the 15m got tangled with the 5 and the leading index is at its monthly high plus the stoch is overbought, yes yes yes and now even Harry joins Meg ....oh no The fed killed us with the news, bloody hell maybe next time I will use a faster MA to compensate for the lack of feed accuracy due to my broker...oh wait that it, the feed is broken, I will buy a better feed : CQG(XXXX$), Bloomberg(xxxxxxxxx$) thats its , that was the problem the feed, oh wait its not the feed its the fact that I have to pea while trading, so I will read trading in the Zone again to distract my pain of trading through my Urinal glans directly unto the stoch 15,9,9 on the 3 Min which is the way to go, but wait my platform does not have Range Bars and everybody knows range bars are the way to go, unless the bottleneck is the CPU which is not strong enough to play DOOM and trade at the same time, so I will get a better CPU , oh wait, I have thunder storms so a great UPS is required....YES EUREKA ITS THE UPS GUYS, Much better than any technical analysis tool out there.


daily candle is it....lol.

Programmer,

You couldn't be more spot on. Thanks for the chuckle. I do a bit of training and have heard all of that buschwa again and again before the wannabees begin to learn. If it was that easy, you couldn't get a cab.

cheers

[zdo]

Originally Posted by Robert »

Can one rely on strictly chart patterns to trade the markets? For example: triangle, wedges, flags, pennants, etc...

Do you think one can make money in the markets in the long run relying only on chart patterns? Thanks

Robert,

Practical answer: Yes.
You can trade ‘classic’ chart patterns successfully
> by placing a pattern in context / background. Others have described this well here and other threads.
> by being ok with sloppiness in their formation. Btw often on the right side of patterns, the obvious, ‘prettier’, more symmetrical patterns don’t seem to fair as well as the misshapen, ugly, unattractive ones. They get faded more?
> by implementing active position management, knowing where your pattern failure point is and ruthlessly stopping out there.
> by not expecting them to work. (rephrasing the preceding one in terms of attitude)
> by being patient, precise, and demanding with your entry placement.
> by developing skills at sizing proper scaling out– some before target, some at target, and some (as many as possible) left on for runners…
> by sizing correctly – this one might should be first!

Originally Posted by Robert »

Can one rely on strictly chart patterns to trade the markets? For example: triangle, wedges, flags, pennants, etc...

Do you think one can make money in the markets in the long run relying only on chart patterns? Thanks

BS (but better?) answer:
This is why the Zen master, when asked the nature of the Buddha, beats the student's head with a stick. Our words cannot convey participation or the realization to which participation can lead.
Here’s one for you – can you trade (except randomly) for an hour, a day, a week, or in any instance(s) - completely without the use of pattern?
The real question is - which patterns are YOURS? !!!!!!!!!!!!!!!

PS In addition to Suri’s book, see Trading Commodity Futures with Classical Chart Patterns by Peter Brandt
Trading Commodity Futures with Classical Chart Patterns by Peter Brandt, ISBN: at the Global Investor Bookshop

[programmer]

This is a very wise anecdot from an Israeli Tech trader(Ob1..) freind of mine :

" A guy goes to a psychiatrist,
The doc says what's wrong?
The guy says I am seeing "body-parts" on charts.
The doc says what do you mean body parts?
The guy says I am seeing head and shoulders and some Bottoms on charts.
The doc says , that is bad, and gives him some Meds.

After two weeks the guy comes back to the doc.

The doc says, well do you still see body parts on charts?
The guys says, no, no more body parts.
The Doc says, do you see anything else?
The guy says, well I do see some butterflies and flags but not that often.
The doc says, its bad and gives him some more meds.

After two weeks the guy returns to the doc and says, Finally I don't see
anything on my charts, just charts.
The doc says' thats great, and gives him the Rorschach test, and tells him
well what do you see in the inkbolts?
The guy says I see complex scientifically derived algorithms..
The doc says thats great, maybe you should write a book on that.

A year later the guy publishes a book entitled" Body parts on charts", alas
nobody buys it, because they think the guy is a lunatic. So, the guy
changes the title to "Patterns on charts". Now, everybody buys the book,
and everybody is seeing butterflies, heads, shoulders, double bottoms,
tripe heads, even some claim that after reading the book they could swear
they saw the Monica/Clinton's Stain on the ES chart, which proved to be
an amazing reversal pattern.

Anyway, the book becomes a best seller along with titles like:

A GUIDE TO DATING ETIQUETTE , by Mike Tyson

FRENCH WAR HEROES. by Jacques Chirac

MY BEAUTY SECRETS, by Janet Reno

MY LITTLE BOOK OF PERSONAL HYGIENE,by Osama Bin Laden

AMELIA EARHART'S GUIDE TO THE PACIFIC

and THE AMISH PHONE DIRECTORY


Anyway the joke is on me because I bought the book...

[zdo]

Trad&#1 01; Cha& #114;t P&#97 ;tter ns L&#1 05;ke a  Pro

Maybe Suri will have time to step up and toot his own horn...

[crabbydog]

Originally Posted by pappo »

Chart patterns are reliable-- Mastering them is another subject....so too is Trading them profitably .

The problem that many have with patterns or their claim that chart patterns are not reliable is usually linked inexperience of understanding Price action or other variables like not enough time to master a pattern...to truely know it like the back of your hand and that takes time.

Part of the mastery is realizing that many variables influence any particular chart pattern that is trying to form. Because of this fact Programmatic back testing via a code usually will miss the value and the true reliability of chart patterns--In contrast a trader that has mastered his favorit chart patterns knows exactly when to trade them and when to let them go by. He has learned to be aware of the influencing variables and understands their impact on what he is trading , thus he can make appropriate adjustments on the fly.
Asking a Code to do the same thing is near impractical, unless you have a team of top notch programmers at your disposal and even then I dont think you would get the same results....some things just cant be communicated through code.

Bulkowski has been successfully trading patterns for years. I came across his article after another trader I respect made a similar comment as something he had observed in trading. Its not that they don't work, but their reliability is tailing off. the other trader also mentioned that he had found increasing cases of patterns apparently failing but of you monitored them they would resume later than expected.

Chart patterns are essentially representations of people's beliefs about the market, these beliefs can be manipulated by the smart money operators.