Learn More about Candlestick Trading

[WmWaster]

Mark (NihabaAshi),

Although most of my candlestick books do discuss combining Japanese candlestick patterns with some form of western analysis...

There was nothing stated with any emphasis on that approach.

In fact, the bulk of the books only had one chapter towards the back that was devoted to the topic (small chapters in comparison to the other chapters) where the rest of the book was just about Japanese Candlestick patterns all by themselves.

That's the problem.
It should be the most important chapter! They should spend most of thier time on this approach, rather than naming each pattern & labeling them bullish/bearish/neutral; confimration/reversal.

Some of my books don't even mention it. It seems the author intends to teach people to interpret the candlestick straight away, which is very wrong in my opinion.

Thus, there seems to be three approaches:

1. Japanese Candlestick Patterns all by themselves.

Most go this route.

2. Western analysis or something as a confirmation tool for a Japanese Candlestick pattern signal.

3. Japanese Candlestick analysis as a confirmation tool to Western analysis or something else.

The bulk of my approach from the above is #3 although I do a little of #2.

I'm more on #3, little #2.

---------------------

Personally I do believe candlesticks should work. It's because, as stated previously, candlestick, if you see through, is merely price action. If you believe past & current price action (ie how it performs in price) have predicting values over the future, candlestick will always work, no doubt!

However you won't predict it's going to rise simply because it rose (white candle) yesterday, so i think the context does matter.

To interpret candlesticks (price actions) correctly, you need to consider:
- the current candlestick, including its:
-- shape (eg doji or hammer)
-- position (ie where it occurs)
-- size (eg there's difference between a white and a big white)
- other candlesticks around it, including their:
-- pattern (eg morning star)
-- position (ie where this pattern occurs)
-- size (the size of the pattern does matter)
- the trend of the current investment
- the market trend (market is king!)
- {any more to add? }

 

[WmWaster]

Quote from ktmexc20:

I code all my own.
Any pattern, candlestick or otherwise, can be coded. It's what pattern (or combination), how to apply it (them) and when. Thats where utilizing analyisis and synthesis comes into play. I believe the more knowledge one has in mathematics, the better. I'm always trying to improve my knowledge.

But that's the main point of argument in this discussion. How could you find a way to appy them in a profitable way?

As someone pointed out, it couldn't make use of candlestick patterns to gain.

So it's not really an issue whether the pattern can be coded. It's we're still not sure whether trading with the candlestick pattern is profitable at all.

After all, what did you add to make them work?

Note that you don't need to tell me the codes. I know it's your work too. I'm notact codes. But e just need some general but insightful ideas how to make the unprofitable candlestick patterns into profits

No, not a problem.
I started teaching my self programing with Python about 2 years ago after I realized that Wealth-Lab (take your pick... tradestation, ami-broker, etc...) was never gonna cut it for me.

Since my goals were a very high turn over rate, I started learning to code in "C" langauge last fall. But, that's for very heavy algorithmic lifting I'm doing.

I would recommend to anybody to learn how to code your own. Python is a fantastic language to learn. Not only for beginners, but very many experts prefer it too for it's productivity and flexibility. Python allows me to easily embed the C work I do.

I always wish I could code my ideas and build an automated trading system.

This should bring me towards another level - mony makinge even if I'm occupied (eg sleeping zzz...) How nice it is!

But I think it's better to buy a workable automated trading system, if any.

 

[Grob109]

Wm Waster and Mark bring up major thoughts that define and bound their interests and focus.

the 1, 2, and 3 as either choices or approaches are too narrow to represent the total aspects of ways people trade. I see additional substantive comments are made to round out what has to be one the table.

Coding the csticks as ststed does imply that valuing price action over time is an aspect of gaining a trading advantage.

I read Wmwasters list upside down as an assessment of the value of the items on the list.

1, 2, 3 are choices of 25% of the topics involved for making money. And they coincide with WmWasters's listing topically speaking.

What puts WmWaster ahead of Mark in reasoning is his preadth of consideration of using a tool to make money.

kt has the handle on knowing the "status" of the market through coding which would then be able to provide a "translated market price status continually. This is a venturing beyond the 25% of the others.

At last WmWaster spoke of his approaches purpose and why he wants the info not possible to get. By using his symmetric orientation (past to present to future (obtaining a prediction)), he further shows that he could be creating a delemma. Mark has one.

So the focus is on the second quarter of the trading process and kt advances a view on how to garner price action in the first quarter. Analysis is the second quarter and it depends upon the data sets obtained from the first quarter.

The first quarter, monitoring, is definitely not yielding a comprehensive data set. To get that complete set, Mark (after giving himslef permission) acquires the stuff he does (we do not know it on the detail levle of csticks price action part) that he spoke of.

So there is a giant fork in the road for Wm Waster at this point. He has a price act only data set; he is oriented to prediction; and he has some concerns about how to begin to get into the game of making money.

3/4 of making money happens after the monitoring data set is obtained and none of making money concerns itself with the future. When does this 3/4 occur? That can't be put on the table it seems.

The focus here is just on the second quarter and that is the topic of analysis. More than candlesticks is required to do analysis.

Personally, I would recommned leading indicators. They involve other variable than price and they involve the two steps of using price (and other data) data in equations and getting signals from the representations of the equations.

I would also recommend using other direct market variables that are available to the same extent price is available as a variable. This is one giant step for all involved in posting so far.

Now the two big considerations. Analysis is not the cornerstone of trading to make money. Neither is its precursor, monitoring to get the data set. These two quarters are only the warm up drill for what a trader has to do to make money.

In trading a lot of people talk about trading as an entry exit game. You can observe, for example that most conversation on entries vastly outweighs exits. These both illustrate that most conversation and writing is just on the two moments of a trade that occur during a much longer period of time as the markets roll on.

Here in this thread some important considerations could be coming to the fore. Ones that are much more important than the usual palaver.

Monitoring get you a data set. This has not been gotten by the participants so far apparently from what they say. They are chatting about analysis, primarily.

The two more important quarters of consideration are those that follow analysis. None of these is the prediction that Wm Waster is looking for.

The vaccuum that is sought to be filled to make money is the decision making strategy that preceeds taking timely action. Action for ET is entering and exiting. It is not for people who are money makers. Action for money makers is different and it is a direct result of being able to have a comprehensive decision making system. A decision making system is what I refer to as a playbook.

Monitoring involves thinking and assembling data. analysis is thinking that involves processing the data into a market status report. You see Mark go get stuff from elsewhere to get the market status.

When you have the market status (think of having the ball and you are between plays). Do you look at the coach? Do you listen for a brief description in the form of an agreed upon generation breakdown that appears at the top of a page of a book that was given to you by the coach to memorize and learn so it could be used at practises and at games?

So analysis is stymied totally because there is nothing to do with the analysis results (the present stauts of the market).

Books that have been read and mentioned in this thred only talk about what is seen interms of bars and the top of the list WmWaster posted finally and recently (all monitoring jargon).

Analysis from complete data sets leads to decision making and I can state to you it does not matter one whit about whanever approach is used when it comes to decision making. In effect candlesticks have gone away during the lest two quarters of the trading process. Decision making only involves thinking (figuring out) what to do as a consequence of the current status.

You look in the play book to find out what to do. The play book has four sections.

Two small one page sections and two huge sections. The small sections are rarely used and the two huge section are so big that they are each divided into parts.

Each of the two large parts are what are used to "march on down the field".

So a person has to consider how to write a playbook if he is not going to buy one from a person who has one.

You can determine if a person has a playbook in a New York minute. Especially if you have a library of playbooks that you have acquired or written over 50 years.

One small section is entitled Entry and the other small section is entitled Exit.

Learn how to complete your data sets. This is still missing in this thread.

Learn how to do analysis with a complete data set. Can't do this without the complete data set.

Write a play book for making money. Get conscious of what you spend most of the day doing. Right now you are not even seeing the markets.

Use the play book to make money.

 

[jmmathieu]

Quote from WmWaster:

I realise candlestick pattern alone is useless, as I stated previously "canldestick pattern taken out of the context is notihng more than arbitrary".

But how about if you combine other factors into consideration, eg the location/position, the size of the pattern (some patterns have to take this into account), volume behind this pattern, and so on.

How about if you try to confirm candlestick patterns by other tools, like indicators?

That's what I'm trying to do.

No one in the planet ever try to do so?!
All authors still (stupidly) teach candlestick learners to indentify the canldestick & their patterns only?!
:eek:

This is a good point, and I have to admit I never did try this approach. My line of thinking was that if there seemed to be even a small edge I could do what you suggested to extract a greater one. Maybe I'll look a little further into this though. Thanks.

 

[dandxg]

I believe this point has been referred to, but I am going to bring it out and make it specific.

Trading candlesticks set ups require confirming price action, as Mark and others have stated. I trade and interpret this to mean for example.......... I have a doji or shooting star at the current high of the day in the 5 min e mini Russ what do I do? I wait for it to break the low of the doji or shooting star and then enter on a stop limit 1 tick below the low or if the DOM looks a little weak I wait for a 1 to 2 tick back up on a limit. I am also looking at my S and R lines for confirmation.

I am curious to hear feedback opinions from others how you would potentially trade this set up.

Hopefully I didn't hijack this thread. I just feel it needs a bit of specificity, that's all.

Dan

P.S. Mark, am I on track in my example on price action as you have referred to it? I believe I am, looking for feedback.

 

[NihabaAshi]

Quote from dandxg:

I believe this point has been referred to, but I am going to bring it out and make it specific.

Trading candlesticks set ups require confirming price action, as Mark and others have stated. I trade and interpret this to mean for example.......... I have a doji or shooting star at the current high of the day in the 5 min e mini Russ what do I do? I wait for it to break the low of the doji or shooting star and then enter on a stop limit 1 tick below the low or if the DOM looks a little weak I wait for a 1 to 2 tick back up on a limit. I am also looking at my S and R lines for confirmation.

I am curious to hear feedback opinions from others how you would potentially trade this set up.

Hopefully I didn't hijack this thread. I just feel it needs a bit of specificity, that's all.

Dan

P.S. Mark, am I on track in my example on price action as you have referred to it? I believe I am, looking for feedback.

Hi Dandxg,

I myself would need more info about the price action involving the doji or Shooting Star.

Also, I don't know what you meant by break the low.

Do you mean the first time it pushes below that low or do you mean its a dark candlestick line that has completed its interval below that low.

If you meant the latter...is it the first interval after that doji low or is it the 2nd, 3rd or what interval after that Doji.

Also, what's occurring in the price action prior to the doji.

What's the reason behind that doji as in the topic of the day.

What's the trend analysis so that you can make adjustments in your trade management after entry.

In addition, you talk about s/r levels.

However, the only thing you say is that your looking at them.

How's the price action of the doji suppose to behave at your s/r level for you to decide it merits a trade position.

All the above has impact on the reliability of the pattern and its results.

Simply, unless your question is rhetorical...its impossible to answer especially since you posted no chart example.

Thus, your going in the right direction in using s/r levels for confirmation but the info you gave is not enough.

I'm not good at hypothetical trade situations...

I need to see a chart example because on a 1 to 10 scale of the info I need to make a trade decision...

The info you gave is about a 3.

Mark
(a.k.a. NihabaAshi) Japanese Candlestick term

 

[dandxg]

I understand Mark, I will get a good chart, when I presents itself, and post it then. Thanks for the answer anyway. I now understand that was vague, looking at it for an outsider's point of view.

Dan

 

[hcour]

It seems to be what I exactly want.

It's because a single candlestick or its pattern may mean differently in different contexts, but no book explains this in details.

So what books/(online) resources about Wyckoff, Morris, Edwards, Magee should I read?

Thanks.

Well E&M wrote the standard textbook for patterns, "Technical Analysis of Stock Trends". John Murphy also has a similar, more updated, just-as-comprehensive book, "Technical Analysis of the Financial Markets", which I picked up in a used book store for a few bucks. Check Amazon or Ebay or whatever for used versions of these books.

As for Wyckoff, there's a good inexpensive introductory book called "Charting the Stock Market: The Wyckoff Method" ($15), but the real bible of Wyckoff is the course he wrote about 100 yrs ago, presently offered by the Stock Market Institute. It's expensive (about $800) but you really don't need the whole thing. The main section is Pt 2 and cost about $200 if I remember correctly. Steep price but it covers every single aspect of trading. If you're really interested I'd suggest you join the Yahoo Wyckoff group and look in the files section for all kinds of information, and ask your Wyckoff-specific questions there, to see if it's something you wish to pursue:

http://finance.groups.yahoo.com/group/Wyckoff-SMI/

Harold

 

[ktmexc20]

Quote from NihabaAshi:
Hi kt,
Your the first trader I've met trading Japanese Candlestick patterns profitably via computer codes.
My next question...are you applying your profitable Japanese Candlestick pattern computer codes to trading stocks, futures or what???
The reason why I've asked you these questions in the recent messages is that there's been many threads here at ET where someone has either asked me or asked someone else about knowing any one doing such and being profitable in their Japanese Candlestick pattern trading...
The answer was always...
I didn't know anyone and/or it couldn't be done because I failed at it and I've never met anyone that has succeeded at it.
By the way, I had used TradeStation easylanguage.
Now that I do...I've saved this discussion for pointing traders in the future that have an interest in how to get started in profitably trading Japanese Candlestick patterns via computer codes.
Once again...thanks for the info.
Mark
(a.k.a. NihabaAshi Japanese Candlestick term

Hi again Mark,

I think what I was saying is that imo any possible pattern can be put to code. By this I'm implying that any type or amount of "discretion" needed, can be anticipated, included, and therefore automated.

I haven't said I trade candlestick patterns, sorry. I'm not saying that any particular cs patterns aren't profitable... I just don't know. Hell, I don't even know what a "bar" is. Well, I guess I should rather say I dislike discretized data analysis, even though it's somewhat necessary in the low level environment of a computer. But I personally prefer to try to not discretize by time when possible... just a hang-up I've got

kt

 

[NihabaAshi]

Quote from ktmexc20:

Hi again Mark,

I think what I was saying is that imo any possible pattern can be put to code. By this I'm implying that any type or amount of "discretion" needed, can be anticipated, included, and therefore automated.

I haven't said I trade candlestick patterns, sorry. I'm not saying that any particular cs patterns aren't profitable... I just don't know. Hell, I don't even know what a "bar" is. Well, I guess I should rather say I dislike discretized data analysis, even though it's somewhat necessary in the low level environment of a computer. But I personally prefer to try to not discretize by time when possible... just a hang-up I've got

kt

Hi kt,

Thanks for the clarification.

Mark
(a.k.a. NihabaAshi) Japanese Candlestick term