Effectiveness of Candlestick bars...

S

Shraga18

I've been backtesting a (preliminary) profitable system based on volume breakouts. Having looked at the close of candlestick bars relative to their high/low for a long time to gauge the bullishness/bearishness of the market, I figured it made sense to add a filter to the system based on said conditions. I tried 2 different variations:

1. close>=h-(range/3)

in order to only enter longs if the bar closed at least 2/3ds up the range. (VV for shorts)

2. close>h-(range/2)

in order to avoid entering a long off a bearish dragonfly-type bar. (VV for shorts)

In both cases my profit went down DRASTICALLY. Which got me to thinking, is using candlebars just another delusion we use to complicate our trading? As we know, the market always has retracements in a trend, so there's always a 50-50 chance that that is the reason for the shape of the bar. Thinking of it in another way, very often, if the bar would have closed just a few seconds earlier or later, the bar could have been bullish instead of bearish or vice versa. Does all this mean that there is absolutely no edge in candlestick bars? I imagine there are those here who do use them profitably (NihaB) comes to mind, so I'd be interested in hearing your opinions!

Shraga
 
Quote from Shraga18:

I've been backtesting a (preliminary) profitable system based on volume breakouts. Having looked at the close of candlestick bars relative to their high/low for a long time to gauge the bullishness/bearishness of the market, I figured it made sense to add a filter to the system based on said conditions. I tried 2 different variations:

1. close>=h-(range/3)

in order to only enter longs if the bar closed at least 2/3ds up the range. (VV for shorts)

2. close>h-(range/2)

in order to avoid entering a long off a bearish dragonfly-type bar. (VV for shorts)

In both cases my profit went down DRASTICALLY. Which got me to thinking, is using candlebars just another delusion we use to complicate our trading? As we know, the market always has retracements in a trend, so there's always a 50-50 chance that that is the reason for the shape of the bar. Thinking of it in another way, very often, if the bar would have closed just a few seconds earlier or later, the bar could have been bullish instead of bearish or vice versa. Does all this mean that there is absolutely no edge in candlestick bars? I imagine there are those here who do use them profitably (NihaB) comes to mind, so I'd be interested in hearing your opinions!

Shraga
More...

Assuming your talking about Candlestick charting and not Candlestick patterns...

It's a preference on what you need to visually analyze the charts.

In fact, most traders I know that use candlestick charts...

They are not using candlestick patterns for their methodology. (ex. MACD, VWAP, CCI, Elliott Wave, Moving Averages et cetera).

Some traders prefer the following...bar charts, candlestick charts, line charts and many other types of charts.

It also depends on your method.

For example, someone using a macd indicator based method...probably sees no difference between a bar chart and a candlestick chart.

However, someone that uses a method that's dependent upon the size of a intervals...probably prefers candlestick charts.

Also, there are lots of past threads on this topic of Bar Charts versus Candlestick Charts et cetera.

Yet, if your talking about Candlestick patterns.

I'm not surprise you have saw poor results in looking at just the Close relative to the High/Low of an interval.

Try looking at the Open and Close relative to the High/Low.

Also, be careful with the range involving candlestick charts.

There's the range involving the difference between High/Low.

Then there's the range involving the difference between the Open and Close.

Like I said before in many other past discussions about this topic...

There is NO EDGE in using either one interval patterns or one interval filters involving Japanese Candlesticks.

Yet, it does make for easier backtesting but why bother when we know what the results are going to be especially after discussing the same thing every few months...year after year at EliteTrader.com...

Poor results.

Simply, don't use candlestick patterns as a surrogate mother to analyze the price action.

You should already understand the price action prior to the appearance of any trade signal.

If you don't understand the price action, you will be disappointed in Candlestick pattern analysis and disappointed in any other type of pattern analysis that has nothing to do with Japanese Candlestick analysis.

P.S. Japanese Candlesticks are not my primary methodology. I only use them as a confirmation tool to what I should already know about the price action.

I just like talking about them because its an easy topic to discuss. :cool:

Mark
(a.k.a. NihabaAshi) Japanese Candlestick term
 
Quote from bighog:

I was thinking candlesticks were used to light candles.

Fooled again. ....:p
More...

At least you weren't trading your misunderstanding.

:D

Mark
(a.k.a. NihabaAshi) Japanese Candlestick term
 
Mark, yes, good thing. :D

Truth be known, i have never read much about the sticks simply because bar charts have served my charting just fine. I use opening prints on new bars like a "DECISION MOMENT" in TA.

EXAMPLE: take a 5 minute ES chart and say you are using a moving avg cross of the 4 period crossing the 9 period as a signal. The main bugaboo with using crosses as entry is within the period itself as one decides to take the signal is WHEN to take the signal. Take the first cross? Take the second cross? Drives you crazy...........so i made a decision (discretion is always allowed to override ) to take first tick of new bar.

Stick to your knitting i say. If sticks work for others, fine. I will leave them in the closet and if lights go out i will use them. In the mean time, lets make some bacon today and be up late tonight as election returns roll in.
 
Quote from Shraga18:

I've been backtesting a (preliminary) profitable system based on volume breakouts. Having looked at the close of candlestick bars relative to their high/low for a long time to gauge the bullishness/bearishness of the market, I figured it made sense to add a filter to the system based on said conditions. I tried 2 different variations:

1. close>=h-(range/3)

in order to only enter longs if the bar closed at least 2/3ds up the range. (VV for shorts)

2. close>h-(range/2)

in order to avoid entering a long off a bearish dragonfly-type bar. (VV for shorts)

In both cases my profit went down DRASTICALLY. Which got me to thinking, is using candlebars just another delusion we use to complicate our trading? As we know, the market always has retracements in a trend, so there's always a 50-50 chance that that is the reason for the shape of the bar. Thinking of it in another way, very often, if the bar would have closed just a few seconds earlier or later, the bar could have been bullish instead of bearish or vice versa. Does all this mean that there is absolutely no edge in candlestick bars? I imagine there are those here who do use them profitably (NihaB) comes to mind, so I'd be interested in hearing your opinions!

Shraga
More...

Sometimes, they work every time :p
 
Never liked candlesticks that much, lately ive been overlaying the median price in wide "point" form over the bars, i like it as an alternative.
 
You must log in or register to reply here.
Back
Top