Intraday Candlestick Trading

[brownsfan019 17]

During my time in the TL chat room this past week, a few of us discussed some candlestick trading on an intraday basis. Since the attendance in the chat room is minimal currently, I thought we could open a thread to discuss specifically trading on an intraday basis with candlesticks.

 

Let me preface this discussion with this - I've proven many times over that the smaller the timeframe, the less reliable a candlestick pattern becomes by ITSELF in isolation. In other words, a hammer that appears on a 1 minute chart is not a valid signal in my opinion simply b/c there is a hammer in front of you. Now, a hammer on a 15 minute chart could be a possible trade by itself.

 

Basically this thread is to toss some ideas out there on possible intraday trading techniques utilizing candlesticks in some fashion. Intraday meaning 1 minute (or lower) up to 1 trade per day.

 

Personally I have found that any type of minute timeframe (1-60) is probably going to require some sort of additional confirmation outside of just the candle pattern. Now, what that confirmation is could be any number of things. I've seen candle traders use pivots, fibs, oscillators, support/resistance areas, WRB's, moving averages, trend lines, etc. etc. to aid in the decision making process of using candlesticks in their trading.

 

Our good friend Walter has gotten the candlestick bug, so we'll see if he takes a few different twists and turns with the candlesticks or not.

 

Good trading!

 

PS

Go Browns! BIG Game Sunday!!

[MC 10]

I was sharing Walters enthusiasm and can't wait to see some more of your candle magic. Good stuff.

[westhilltrader 10]

Hi there everybody, I am the "westhilltrader" and new to this forum.

 

For two years now, I do mostly day-trading and only use 1 minute candlestick charts. They give the best representation of price evolution during any period, intraday or multi-day.

 

However, none of my critical decisions for entry or exit points is done based on candle formations. I only use indicators for that. I also belive that the 1 minute period is the best choice for day-trading. Going smaller than that adds too much noise to the indicators and to the price itself. Larger periods cannot keep up with the speed of data change and action response required in day-trading.

 

I do not have an expert view of the candles but, during my after market trade analysis, the classic candles or formations do not appear as reliably as required for a critical decision. So I stick to what has proven reliable to me

[mister ed 10]

Let me preface this discussion with this - I've proven many times over that the smaller the timeframe, the less reliable a candlestick pattern becomes by ITSELF in isolation. In other words, a hammer that appears on a 1 minute chart is not a valid signal in my opinion simply b/c there is a hammer in front of you.

 

Personally I have found that any type of minute timeframe (1-60) is probably going to require some sort of additional confirmation outside of just the candle pattern.

 

Regarding not viewing candlesticks in isolation, one of the major points raised in the VSA thread is to always look to the background, or the context, within which a "signal" occurs. Probably the post that summarises and highlights the importance of viewing the context within which a signal occurs is this one.

 

In "pure" VSA analysis the price bars do not have an open on them, the bars are simply HLC, with a volume histogram below. Now, candles of course require an opening price on them, and a lot of the traders on the VSA thread use candles anyway, so there is plenty of scope for applying candle "patterns"/bars with background/context, especially for intra-day trading.

[james_gsx 16]

I'd much rather use VBCs, I think candles are more valuable that way in the sense that you actually see the fight. I know I stole that from you brownsfan lol

 

I also think it depends on the type of day, is it trending or is it rangebound? If it's a trending day I like to use a larger VBC number and look for setups that will catch the bigger trend. For example if I see a hammer, and that is confirmed a MA crossover then I will enter the trade.

 

On a range bound day I use mostly support/resistance and other technical patterns that I have learned to read fairly quickly. This is a little more difficult and requires more screen time IMO. If I see the market trending towards resistance then a doji appears I will wait for confirmation, then when I get that confirmation I will go with the new trend. As Brownsfan said sometimes it helps to add something aside from just the candles and support and resistance lines. I came up with my own stochastic's that used the same integer I use for my VBC's. Not really sure if this works yet, but looking at historical charts it seems to work. I will have to play with it more in real time and decide if that's the direction I want to take. I also think it's important to point out that I use a smaller VBC setting for this.

 

For those that don't know how to choose a VBC setting all I did was found a simple integer (usually 3 or 4) and kept multiplying it until I found what worked best for me.

 

3 x 3 x 3 x 3 etc until I came up with 243 and 729

4 x 4 x 4 x 4 etc, it all depends on what works best with your trading style.

 

I included a few charts. Two of them are 243vbc and the other two are 729vbc. One is for Friday the other is for Thursday, you can probably figure out which is range bound and which one is a trending day. What do you notice between the two? Frankly with that choppy market on Thursday that 729 chart would have chopped me up, but the 243vbc let me see more opportunities. Other people may see different things and completely disagree with me, but each person is different and has their own style of trading.

 

 

Right now I am working on learning Market Profile. I want to see if I can integrate Market Profile into my candlestick analysis. So if anyone has a lot of experience in MP or in both I think it would be beneficial if we could start a thread on that.

 

Personally I believe CA on it's own is worthless... and this comes from personal experience. But if you combine CA with other technical techniques such as VSA, support/resistance, other technical patterns, etc... then candlestick analysis can be very beneficial. For example would you buy/sell a doji in the middle of a tight range? No, but if the market was trending higher and you saw a defined doji against resistance whether that be a Fib number or a previous high then that would definitely resemble a much stronger warning.

 

I am definitely no expert with candlestick analysis, I can read a chart and tell you what the candle means and things like that. But I still have a lot to learn, and only screen time can get me there.

[james_gsx 16]

Oops, I forgot the charts so here you go.

 

I went with 243 and 729 because I like things to happen fast, for some of you that might be too fast or too slow. Notice that the contract is the YM. If you plug in a 243vbc into the ES it will look like splattered paint. Just multiply 729 by 3 once or twice until you get something that you can work with. If you don't like the integer number then just make one up, there isn't a magic number just go with whatever works for you.

[BlowFish 10]

Hope you dont mind if I tag along and play Devils Advocate from the "its the underlying price action that the candle represents that is important" point of view. I promise to behave.

[walterw 10]

Thanks Brown for this new thread ¡¡ it will be very interesting to see where we can get here...

 

 

My motivation here is to reduce my cronic scalping to a larger span but at the same time not so large... so it would be a more "larger scalping" aproach...

 

So far I am very new to candles, so I cant talk with authority on this and I would like to learn from the masters... what I see is that 3 min and 5 min candles can do "something" for my idea, key is to create a good context so this patterns on this small time frames could be more credible...

 

At this point I am paying a very close attention too dojis and hammers...

 

Thanks for this oportunity ¡¡ and let the interaction start ¡¡ cheers Walter.

[brownsfan019 17]

Hi there everybody, I am the "westhilltrader" and new to this forum.

 

For two years now, I do mostly day-trading and only use 1 minute candlestick charts. They give the best representation of price evolution during any period, intraday or multi-day.

 

However, none of my critical decisions for entry or exit points is done based on candle formations. I only use indicators for that. I also belive that the 1 minute period is the best choice for day-trading. Going smaller than that adds too much noise to the indicators and to the price itself. Larger periods cannot keep up with the speed of data change and action response required in day-trading.

 

I do not have an expert view of the candles but, during my after market trade analysis, the classic candles or formations do not appear as reliably as required for a critical decision. So I stick to what has proven reliable to me

 

west - this thread is to discuss candlestick trading. If you want to start a new thread to discuss how you trade on 1 minute charts with indicators, please do so! How you trade may work for YOU, but that does not mean it's an absolute by any means. I would argue that the 1 minute chart is full of NOISE and that it's not good for daytrading. Neither of us is right as trading is about finding what works for YOU. So please keep that in mind when posting. If you want to share some trading ideas with us, start a new thread and see where it goes. If not, please reserve your comments for threads where you are staying on topic. This thread, is about trading USING candlesticks, NOT ignoring them. Thanks.

 

Regarding not viewing candlesticks in isolation, one of the major points raised in the VSA thread is to always look to the background, or the context, within which a "signal" occurs. Probably the post that summarises and highlights the importance of viewing the context within which a signal occurs is this one.

 

In "pure" VSA analysis the price bars do not have an open on them, the bars are simply HLC, with a volume histogram below. Now, candles of course require an opening price on them, and a lot of the traders on the VSA thread use candles anyway, so there is plenty of scope for applying candle "patterns"/bars with background/context, especially for intra-day trading.

 

Good points Mr. Ed. I brought this up b/c I see too many people that have no idea how candles work saying - look at this hammer, it failed! Well, there's so much more that has to go on there it's not as easy as hammer hunting*.

 

Oops, I forgot the charts so here you go.

 

I went with 243 and 729 because I like things to happen fast, for some of you that might be too fast or too slow. Notice that the contract is the YM. If you plug in a 243vbc into the ES it will look like splattered paint. Just multiply 729 by 3 once or twice until you get something that you can work with. If you don't like the integer number then just make one up, there isn't a magic number just go with whatever works for you.

 

The key James in all of this is to find what works for you. Minute charts or VBC's can all work for candles, so it's taking the time to find what works for you.

 

Hope you dont mind if I tag along and play Devils Advocate from the "its the underlying price action that the candle represents that is important" point of view. I promise to behave.

 

BF - I am watching you. Make your point and then go home!

 

Thanks Brown for this new thread ¡¡ it will be very interesting to see where we can get here...

 

 

My motivation here is to reduce my cronic scalping to a larger span but at the same time not so large... so it would be a more "larger scalping" aproach...

 

So far I am very new to candles, so I cant talk with authority on this and I would like to learn from the masters... what I see is that 3 min and 5 min candles can do "something" for my idea, key is to create a good context so this patterns on this small time frames could be more credible...

 

At this point I am paying a very close attention too dojis and hammers...

 

Thanks for this oportunity ¡¡ and let the interaction start ¡¡ cheers Walter.

 

Walter - it really started with our interaction in the chat room.

 

I know you'll really put forth the time and effort into this, so that's why I am willing to discuss these ideas further. It's not much fun to start threads and see a couple responses. I would guess that if you get some ideas flowing, the threads and interest will follow as well.

 

* - the term 'hammer hunting' is originated and designed by brownsfan019. I like this term so much that I am claiming rights to it's use. If you like it, come up with your own idea.

[walterw 10]

Thanks Brown... I think that your 15 min chart really should be king here... I was really captivated with that chart and its performance... so lets start with the 15 min setups...

 

I would like to develop some exit strategies more near to my style probably on smaller timeframes, but setups on 15 min look the best as I see that from your experience (wich is very valuable) trying to setup on smaller timeframes could be a futile game...

 

My first question here... when a hammer doesnt make a mayor follow thru, is it possible to play in the oposite direction ? could that be a good setup ?

 

 

 

also, I am very enthusiastic to see how Nison uses "flips" as part of his S&R context strategies... cheers Walter.

[brownsfan019 17]

Thanks Brown... I think that your 15 min chart really should be king here... I was really captivated with that chart and its performance... so lets start with the 15 min setups...

 

I would like to develop some exit strategies more near to my style probably on smaller timeframes, but setups on 15 min look the best as I see that from your experience (wich is very valuable) trying to setup on smaller timeframes could be a futile game...

 

My first question here... when a hammer doesnt make a mayor follow thru, is it possible to play in the oposite direction ? could that be a good setup ?

 

also, I am very enthusiastic to see how Nison uses "flips" as part of his S&R context strategies... cheers Walter.

 

Walter,

I have not found much in terms of failed signals on larger timeframes, like the 15 minute. Doesn't mean it can't work, but for me, I've found sometimes a hammer's low is tested and then snaps back. In a scenario like that, you are going to create 2 losers.

 

The real key is how you exit. In your example above, that could have been a profitable trade depending on how you exited.

 

Personally, if I would have traded that, I would have been out 3 candles later on the red WRB. In this example, it was a timely exit.

 

So our discussion can lead to two new threads:

 

1) Failed signals turned to profitable ones (by reversing).

 

2) Timely exit strategies to keep most trades profitable and working for your trade.

[walterw 10]

ok sounds good, so we can explore some new grounds... would this exits be candle based ? or you are interested in exploring some other exit strategies...

 

cheers Walter.

[brownsfan019 17]

ok sounds good, so we can explore some new grounds... would this exits be candle based ? or you are interested in exploring some other exit strategies...

 

cheers Walter.

 

Walter,

I'd love to find the most efficient and effective way to exit trades, whatever that may be. Currently it is mainly based on visual WRB's as I've found these to be decent exits on the 15 minute chart. A WRB on the 15 is pretty strong in my opinion. Of course, no exit strategy is perfect (yet) and sometimes you exit 'too soon' or 'too late' in hindsight. Part of the game I guess.

[walterw 10]

Walter,

I'd love to find the most efficient and effective way to exit trades, whatever that may be. Currently it is mainly based on visual WRB's as I've found these to be decent exits on the 15 minute chart. A WRB on the 15 is pretty strong in my opinion. Of course, no exit strategy is perfect (yet) and sometimes you exit 'too soon' or 'too late' in hindsight. Part of the game I guess.

 

 

Yes I read your entire wrb thread yesterday, found it smart...

 

One idea that came to my mind was that wrb`s are followed eventually by a small candle, that small candle that follows wrb`s as stated by Nison would first suggest a pause, but not necesarily a reverse on direction, as the trend may resume and continue... so instead of outing at the close of the wrb , my thought was that maybe trailing the extremes of this small candles that follow the wrb could be a way to stay inside... just ideas Brown, very little experience on the subject... only some cronic sick research adiction here.. cheers Walter.

[westhilltrader 10]

west - this thread is to discuss candlestick trading. If you want to start a new thread to discuss how you trade on 1 minute charts with indicators, please do so! How you trade may work for YOU, but that does not mean it's an absolute by any means. I would argue that the 1 minute chart is full of NOISE and that it's not good for daytrading. Neither of us is right as trading is about finding what works for YOU. So please keep that in mind when posting. If you want to share some trading ideas with us, start a new thread and see where it goes. If not, please reserve your comments for threads where you are staying on topic. This thread, is about trading USING candlesticks, NOT ignoring them. Thanks.

 

 

Thanks for the comment, Brown.

Here are a few points I would still like to make before moving to another thread:

1. One must listen to both the PROs and the CONs on any topic to get a fair idea whether it is worth spending the time on. There are many beginers here who should hear the CON's and concentrate on learning first the most basic trading methods. As they become more sophisticated, as you seem to be, they can learn and play the pure candles.

2. I do trade with candlestick charts, I wouldn't do without them. The price action which they represent is all I use along with my indicators to arive to critical decisions.

3. The period selection for the intraday trading is market specific. I am trading equities (Toronto Stock Exchange) which, compared with some future charts I see attached here (see James_gsx), are moving much faster. I also

studied Nasdaq and that market moves even faster. Trades on these markets are computer driven and, sometimes, an intraday rally may not last more than five minutes. Any period larger than a minute on these markets would miss half the action. On occasion I feel I coould do with smaller periods but my software is limited down to 1 minute.

 

Good luck with your thread

[brownsfan019 17]

Yes I read your entire wrb thread yesterday, found it smart...

 

One idea that came to my mind was that wrb`s are followed eventually by a small candle, that small candle that follows wrb`s as stated by Nison would first suggest a pause, but not necesarily a reverse on direction, as the trend may resume and continue... so instead of outing at the close of the wrb , my thought was that maybe trailing the extremes of this small candles that follow the wrb could be a way to stay inside... just ideas Brown, very little experience on the subject... only some cronic sick research adiction here.. cheers Walter.

 

I agree Walter on your points. Here's what you will see on the 15 minute however - many WRB's appear in your favor and the next 15 minute candle produces a snap back or complete retracement of previous WRB. So for me, I want out at the high/low of the JUST FINISHED WRB, not the one forming that goes against my position. Again, just my observation on the 15 minute. A lower timeframe, like the 3 or 5 minute that you have been looking at, could be very,very different. I struggled with WRB's on lower VBC's so it could be a function of the tiemframe setting.

[BlowFish 10]

....I struggled with WRB's on lower VBC's so it could be a function of the tiemframe setting.

 

The culprit is the VBC'c. From statistical analysis of the ES and from observation of other instruments, the range of a bar (in a constant time) is highly correlated to the volume of the bar. What this means is a 'WRB' is much much more likely to show as 3 white soldiers / 3 black crows on VBC's.

[brownsfan019 17]

The culprit is the VBC'c. From statistical analysis of the ES and from observation of other instruments, the range of a bar (in a constant time) is highly correlated to the volume of the bar. What this means is a 'WRB' is much much more likely to show as 3 white soldiers / 3 black crows on VBC's.

 

I disagree BF b/c if you load up a 50,000 VBC that is not true at all. The culprit was a LOWER setting VBC. Just like a 1 or 3 minute chart as well. WRB's are subject to some discretion the lower the timeframe setting, whether minute based, volume based, tick based, etc.

[BlowFish 10]

OK we are making different points I think. The statistics I took (not just looking at a bunch of charts) where from 1 minute 5 minute 15 minute 1 hour 1 day and 1 week charts. These showed absolutely conclusively for of all of those sampling periods that greater volume resulted in greater range.

 

My point is simply this - WRB's are less frequent on constant volume charts with a similar number of bars to a constant time chart. This is true regardless of 'time frame'. (e.g. comparing daily to N hundred thousand constant volume charts)

 

They also 'mean' something different (obviously). One means that a candle has gone 'a long way' in a fixed period of time. The other means that price has gone a 'long way' on a fixed number of contracts.

 

Do you still use constant vol btw?

 

Anyway leaving constant volume charts aside I'm Interested that you believe lower time frame charts have less clear signals. I know you are using 15 minute ES charts I wonder if you consider there is a 'cut off point' where its all noise? 10 minutes seem pretty safe to me and 5 minutes are OK except when volatility is real low (and you probably want to be sidelined anyway). Mind you if volatility is high 3 minutes seem to get by. Just wondered what your thoughts are?

 

Cheers.

[brownsfan019 17]

OK we are making different points I think. The statistics I took (not just looking at a bunch of charts) where from 1 minute 5 minute 15 minute 1 hour 1 day and 1 week charts. These showed absolutely conclusively for of all of those sampling periods that greater volume resulted in greater range.

 

My point is simply this - WRB's are less frequent on constant volume charts with a similar number of bars to a constant time chart. This is true regardless of 'time frame'. (e.g. comparing daily to N hundred thousand constant volume charts)

 

They also 'mean' something different (obviously). One means that a candle has gone 'a long way' in a fixed period of time. The other means that price has gone a 'long way' on a fixed number of contracts.

 

Do you still use constant vol btw?

 

Gotcha BF. Currently just using the 15 min chart setup. That's it, fairly simple. Sometimes it's a matter of KISS.

 

Anyway leaving constant volume charts aside I'm Interested that you believe lower time frame charts have less clear signals. I know you are using 15 minute ES charts I wonder if you consider there is a 'cut off point' where its all noise? 10 minutes seem pretty safe to me and 5 minutes are OK except when volatility is real low (and you probably want to be sidelined anyway). Mind you if volatility is high 3 minutes seem to get by. Just wondered what your thoughts are?

 

Cheers.

 

Cutoffs for using candlestick patterns is a toss up b/c as I mentioned earlier, the smaller the timeframe, the need for some sort of additional confirmation is stronger. So, one could trade on the 1 minute chart using candlesticks AND XYZ. In that instance, the 1 minute is just fine. Now if primarily using candlestick analysis for your decision making, a longer timeframe is good. I happen to like the 15 minute chart (currently). I will watch other timeframes for comparisons but the 15 is providing some nice signals with good risk/reward potential.

[nycdweller 10]

I currently watch the NQ. Is there a big difference between a 10 minute chart and a 15 minute chart in terms of candlestick analysis? Thanks.

[brownsfan019 17]

I currently watch the NQ. Is there a big difference between a 10 minute chart and a 15 minute chart in terms of candlestick analysis? Thanks.

 

There can be a difference nyc and sometimes it's larger than other days. You just have to test different settings to see what works for you.

[BlowFish 10]

Of course we have talked about this in the past These are the important decisions - what size moves you are trying to capture, how much time you want to sit in front of the PC. etc. etc. Every bit as important as whether you are going to trade Dojis, Dunnigan or the DOM also one of the things you have absolute control over as a trader!

 

I know a chap who trades houly bars (predominantly on bar close). He goes outside does some work and then just comes in now and then to see how things are panning out. - Rambling off topic.

[monad 10]

Another neat trick is learn to blend candles of in your head,

ie. a WRB up, squat and a down bar on 5min chart, now take the open of the WRB, high and low of the 3 bars and close of the down bar,

 

Now you have a 15min candle , either a doji or an inverted hammer/dragonfly.

 

Same process again to gauge what is on the 30min.

 

My 2 cents. helps me anyway

[walterw 10]

I enjoyed today following this 5 min candles on the ES....

 

Now I thought that adding a little bit of context would actually make the trades a little more confident... this are "ideas" I want to share here...

 

For example adding some ketners 9/1 atr simple and a macd 12/26 could give some interesting context of trend and exhaustions as well...

 

Basicly this is the way Keltners can be used :

 

 

 

here is some examples of how this very nice candle patterns happen on some very intersting keltners contexts :

 

 

 

The macd could give some trend bias actually corresponding to higher time frames as it has a quite high input...

 

those are some ideas... please tell me what you think... cheers Walter.