The Perpetuated Fraud of Candle Charting

Quote from mokwit:

The biggest problem with candlesticks is that it takes longer to unlearn them once you realise they are meaningless than it did to learn them in the first place.

Try counting all the Doji's etc on a stock that has had a long uptrend.

I use them for supply/demand visuals, but that's about it.

There are several dojis that are bullish about 58% of the time via a particular interval that occurs prior to the doji (use ET to search for the info). :cool:

Regardless, they can't be meaningless to you as you stated while at the same time you're using them for supply/demand visuals. Thus, they obvious do have some value to you regardless if the value is small. Kind'uv like TraderZones several years ago saying "everything involving TA" was useless while at the same time saying trendlines and s/r levels. However, after a little probing, his source of TA info was very bad TA.

Mark
 
Quote from NihabaAshi:

There are several dojis that are bullish about 58% of the time via a particular interval that occurs prior to the doji (use ET to search for the info). :cool:
And I'm convinced you have a delightful sense of humor 64% of the time. :)
 
Quote from Kassz007:

How is this different from your opposite viewpoint? You are basically saying the same thing:

"Of course it doesn't work. Evidence? Well, I believe it to be not true, so it musn't. You are wrong if you do not agree with me..."

Respectfully, this is why it's different:

Logically, "I believe it to be not true" is a very different (and much stronger) statement than "I don't believe it to be true".

It appears to me that TZ was asserting the latter.
 
TZ is a bitter loser, he doesn't deserve you or anyone else coming to his defense :)

Quote from Yossarian:

Respectfully, this is why it's different:

Logically, "I believe it to be not true" is a very different (and much stronger) statement than "I don't believe it to be true".

It appears to me that TZ was asserting the latter.
 
Quote from mokwit:

The biggest problem with candlesticks is that it takes longer to unlearn them once you realise they are meaningless than it did to learn them in the first place.

Try counting all the Doji's etc on a stock that has had a long uptrend.

I use them for supply/demand visuals, but that's about it.

Some time ago, I bought Nison's book on candlesticks and just could not get to grips with it. But prior to that whilst working at a bank I heard the term "Tombstone Doji" and in a very general, vague way it it seemed to capture the market mood (this was sometime between Bear and Lehman).

By default, I look at candlesticks when I want to see a chart, because they impart more info than a simple line graph. However, as Eli has astutely pointed out - intraday the open and close wicks are arbitrary.
 
I'm passing this on with permission because someone did the same for me a couple of months ago. Put a 3 line break format in 89, 177 and 277 tick candles with standard 10,3,3 stochs and a standard MACD. Pay attention to all 3 charts setting up together for an entry. If it doesn't make any sense, just consider it hot air. Good trading to all.
 
Quote from Fleming Snopes:

The original text on candlestick charting was Seiki Shimizu's "The Japanese Chart of Charts," translated to English by Gregory Nicholson. Therein it is clear that the original and still then current application of candlestick charting and theory was to the daily chart. Subsequent supposed applications of them to any other time frame are in my opinion simply fraudulent, a bald attempt to extract money from unwary traders for books, consulting and web memberships. This fraud is clear because only the daily chart can be said to have a clearly significant high, low, open and close. Ascribing any such significance to one hour or fifteen minute or five minute charts is ludicrous. But traders do it anyway because it is so easy to click on the time scale for the desired chart. I propose that with today's fast data providers like ESignal there is an alternative to slavish adherence to the practice of the herd: chart in a much faster time frame and compress it to appear like the timeframe you want. For example, I am viscerally comfortable with one minute charts. So to get the full shape of price history without the irrelevant distortion of candlesticks, I chart in one second and compress it, adding one minute time ticks for reference. Attached is an example of ten minutes from this morning. You will note that this inaugural post does not contain the text h.e.r.s.h.e.y, so it is unlikely that the ego-driven perpetually roving bots trying to extract significance from insignificance will disturb us here.


There is no dog in the clouds. Call the dog a doji or call it a Hymans-Sycophant-Oscillator

It don't really matter

what matters is collecting enough data and applying a strategy consistently enough using that data to make a few bucks on a purely statistical basis

ET is not a place to learn how to do this -ET is a place to break wind
 
Quote from Kassz007:

How is this different from your opposite viewpoint? You are basically saying the same thing:

"Of course it doesn't work. Evidence? Well, I believe it to be not true, so it musn't. You are wrong if you do not agree with me..."

The evidence against TA practiced by the masses is enormous. It has been tested by many, longterm and after costs, little value found.

The evidence for TA is usually belief and opinion.

If you do not grasp this, then you have your answer. If you grasp this, then you have your answer.
 
I wrote some code to markup the "common" candle patterns on charts. Aren't they wonderful? With so many to choose from, cherry picking a few to support any proposition one cares to put forward should be a snap.

More seriously, the code is also integrated into a stock screener with back testing. I could find no evidence that candle patterns provide any edge, either stand alone or in combination with other factors.
 

Attachments

Quote from Buy1Sell2:

...A trader simply needs to relate the action to the movements on the higher chart and then you have it...

Although I understand where you're coming from, what about the following.

For instance, what if you wish to trade a 5 minute chart's action, which is bullish, but which higher chart do you pick if?

30min bearish
60min bearish
240min bullish
Daily bulllish
Weekly bearish
Monthly consolidating

:)
 
Back
Top