The Close of a Bar is Meaningless

[Trader1 10]

Do you think the big players give a damn about a bar chart,let alone sit around waiting for a bar to close?

 

YES! Definitely!

I understand your thinking and where you are getting at,

and I can agree about the wierdness about the whole thing and it seems a little strange indeed,

the fact is that it wouldn't matter on which timeframe you are looking at,

you have swing highs and lows on all the different timeframes don't you?

with your reasoning the highs and lows wouldn't matter either, but they do because they

indicate a turn in the market even if it's just a small one, a candle or a bar is based on time, why they have a very important factor is this, the highs and lows of a candle or bar creates support and resistance aswell as swing highs and lows even if that is psychological, and where a candle or bar closes the next one will open,

it is a great advantage if a candle would open above support or below resistance in the direction you would like to trade in, for that to happend, a candle must first close above or below support or resistance which is the highs and lows of the candles, this is a very important fact in price action, there are many traders only trading this. so the high low and close is very important and it does matter. Yes I have backtested it and this is mostly how I trade daily and Im profitable doing so.a breach of a high is very bullish and a breach of a low is very bearish but only if price can get the advantage to close the candle above or below a support resistance because then the next candle will open in your favor and price will have support in that direction, you also wouldn't have a clue what might happend if you don't wait for the close first. reading price action is an artform, practice practice:) yes you would take profit if a candle is closing above and below a high or low counter to your direction, so you would have to wait, because then price would have support buying counter move and you wouldn't want to be in on the trade anymore. this is all strange and it is not easy to understand, that is because this is all psychological. the market is all about psychology, a support and resistance line shouldn't to your theory have any significant meaning either, price go up up up and then suddenly stops and hangs around and going sideways, everyone knows there is no invissible line there stopping price, it's all psychological, the big bank traders are protecting this level because it is in their intrest doing so. yes they are definitely looking at highs and lows and closes.

Edited May 13, 2012 by Trader1

[MightyMouse 12]

Exactly who are you talking about here? Do you think the big players give a damn about a bar chart,let alone sit around waiting for a bar to close?

 

Mitsubishi,

 

If the above statement was not just a lead in to the rest of your awesome post, then I have to disagree. I do not feel that "big players" deserve to be held in high esteem. A small trader and a "big player" can enter the market at the same time and in the same direction. The small trader will lose money because he made an adhoc, uninformed decision to enter. A "Big Player" will enter the market as a result of careful analysis, meaning it was right and correct to enter even though he lost. If each lose, what is the difference? The difference is the level of arrogance and humility. The big player can be just as dumb as the dumb money small trader.

 

Big players make mistakes and the mistakes are frequently big and profitable for small traders to take advantage of. Long live the "Big Player".

 

Bar none, I enjoy your posts more than any others.

 

 

MM

[jackj 10]

lol... I could have written that very same post, Mitsubishi! And I'm probably about to prove it

And you definitely get cut some slack in your response to the ludicrous post (see, I can do it, too!) saying you're too emotional (right before he makes the seemingly emotional statement that the close DOES matter, whether you believe it or not, because it's the truth). Without, of course, providing any evidence of it being the "truth".

As he also says that "even the Open matters", I can't help but wonder if he's talking about daily bars?, since for any other timeframe the open is either the same or only a tick different from the prior close, so of course if one found the Close meaningful, they would find the Open of the next bar equally meaningful. But, I believe earlier in this thread it was acknowledged that indeed, the daily could be seen differently than other timeframes as it's an agreed upon time. Then again, he's referencing futures, and I don't know too many individual futures traders who are trading off of daily bars.

I think at this point everyone's said their view and covered all the bases; if some choose to use it as a criteria they can. It was good for them for these posts to point out (or attempt to, anyway) the fallacy (or at least problem) with using them. But a lot have found ways to trade more bizarre things successfully, apparently in spite of, not because of, those things (planets, etc.) To quote Stan Laurel, "You can lead a horse to water, but a pencil must be lead."

[SpearPointTrader 10]

According to your logic that time frame does not matter, they are all equally important, the implication is that every tick is important,if not,what is the smallest size wave i can "surf" ?. Though i call myself a technical trader,based on your statements,which you claim are non negotiable,i can sympathize with those who believe TA is pseudoscience at best and a crock of sh##t at worst.

 

Well, the smallest size wave would be up to you. Although I follow the waves on the 3 minute, 10 and 30 minute chart, I really use them to time entries on large waves I see on the daily charts. Most of the time, it's just to make sure the current, smaller wave, is going the same way when I jump in. It's an effort to minimize draw down time that works really well for me.

 

Sometimes, especially in choppier markets, I zoom down to the 60 minute, or even 30 minute, and use the 3 an 10 minute for my timing.

 

So, it's not about the size of the smallest wave, it's about knowing which one is the best to try and ride. That is something only experience can tell you.

 

 

Another way to answer your question would be to look at the potential wave ion question, and try and calculate if it will last a few bars. If so will you clear commissions with a profit?

 

A super small 1 tick wave will almost never be big enough to clear your commissions. The 10 minute on the other hand, very often is.

 

The 3 minute is "Ify" at best, short of a real power move. That is why I only use them as instruments of timing.

[SpearPointTrader 10]

YES! Definitely!

I understand your thinking and where you are getting at,

and I can agree about the wierdness about the whole thing and it seems a little strange indeed,

the fact is that it wouldn't matter on which timeframe you are looking at,

you have swing highs and lows on all the different timeframes don't you?

with your reasoning the highs and lows wouldn't matter either, but they do because they

indicate a turn in the market even if it's just a small one, a candle or a bar is based on time, why they have a very important factor is this, the highs and lows of a candle or bar creates support and resistance aswell as swing highs and lows even if that is psychological, and where a candle or bar closes the next one will open,

it is a great advantage if a candle would open above support or below resistance in the direction you would like to trade in, for that to happend, a candle must first close above or below support or resistance which is the highs and lows of the candles, this is a very important fact in price action, there are many traders only trading this. so the high low and close is very important and it does matter. Yes I have backtested it and this is mostly how I trade daily and Im profitable doing so.a breach of a high is very bullish and a breach of a low is very bearish but only if price can get the advantage to close the candle above or below a support resistance because then the next candle will open in your favor and price will have support in that direction, you also wouldn't have a clue what might happend if you don't wait for the close first. reading price action is an artform, practice practice:) yes you would take profit if a candle is closing above and below a high or low counter to your direction, so you would have to wait, because then price would have support buying counter move and you wouldn't want to be in on the trade anymore. this is all strange and it is not easy to understand, that is because this is all psychological. the market is all about psychology, a support and resistance line shouldn't to your theory have any significant meaning either, price go up up up and then suddenly stops and hangs around and going sideways, everyone knows there is no invissible line there stopping price, it's all psychological, the big bank traders are protecting this level because it is in their intrest doing so. yes they are definitely looking at highs and lows and closes.

 

Paragraphs man, paragraphs!!

 

Other than that, you are spot on!

[SpearPointTrader 10]

lol... I could have written that very same post, Mitsubishi! And I'm probably about to prove it

And you definitely get cut some slack in your response to the ludicrous post (see, I can do it, too!) saying you're too emotional (right before he makes the seemingly emotional statement that the close DOES matter, whether you believe it or not, because it's the truth). Without, of course, providing any evidence of it being the "truth".

As he also says that "even the Open matters", I can't help but wonder if he's talking about daily bars?,

 

Well, the daily chart does have extra significance, due to the stop, or break from trading that creates clearly defined sessions. However, where the bar opens on the intra day charts matters as well.

 

Many times I have watched a market on the 10 minutes chart, and it has a fully formed bar, where the price just sits there untill the open of the next bar. Then it starts jumping around and moving again, only to stall about 8 minutes in and pause till the new bar opens before movement occurs once more.

 

I have seen this so often over the years that I have concluded that I am not the only one trading like this. it has to be wide spread amongst the large professional market movers.

[joshdance 10]

the take away was, in your opinion, POC shifts were not statistically sound, I disagree.

 

I can't recall exactly what I said either, but it was more related to the mode of any statistical distribution. The mode is not calculated in such a way that it is defined as continuous, whereas the mean is, for example. So in your first picture, 2940 is the mode. At what volume, I don't know, but let's assume 1000 traded at 2940. Now let's assume that 950 have traded at 2922. If the market trades 45 more at 2922, 2940 is still the mode/POC. No shift yet. Now, if only 6 more trade at 2922, we will have a POC shift. Yet, if 6 more don't trade there, we won't. A statistical measure that is influenced so greatly by only a few in the data set is not "sound" or "robust" as a measure of value, IMO. What we have is a bimodal distribution, and both points may be significant (or not), but a tiny variation such as this does not suddenly affect how I view things. The mean is not this way, which is why the VWAP is a continuous, more "stable" measure of value for the distribution.

 

Simply, price moves horizontally across the chart, Time is irrelevant, sort-of. What I look at at the opening bell is price not time,,, I don't think I'm alone with that. So during the "rotation of the Earth" when the Asian session begins to trade I capture that first tick of price to mark the event, not time. The current price is always highlighted on the top of the display and the volume profile begins to develop into the patterns we MP traders recognize, only rotated 90 degrees.

 

Thanks for posting these charts 5DAW. It seems though that you could just as easily watch the profile on its normal axis and call it "uneven breasts" or "alligator mouth" or "man with a big nose" and have basically the same thing. But if it helps to view it that way, there's no harm in rotating the monitor (except that it's hard to move the mouse up and down and watch it more left or right when your screen is this way, and that you have to learn to read numbers veritcally more quickly ).

 

It still seems that you have the more recent (there's the time variable) profile on top, hence you still have some notion of time here. Of the three variables to consider, only price can decrease; the other two can only increase (until you unveil that time machine you've been working on!) ...

 

I don't want to reveal much more of the secret sauce quite yet so I'll end by saying I believe, "The Close of a Bar is Meaningless," or should I say, "Bars are Meaningless."

This, I will drink to. But still, I do appreciate other points of view from those who disagree with me on this.

[joshdance 10]

Well, the daily chart does have extra significance, due to the stop, or break from trading that creates clearly defined sessions. However, where the bar opens on the intra day charts matters as well.

 

Many times I have watched a market on the 10 minutes chart, and it has a fully formed bar, where the price just sits there untill the open of the next bar. Then it starts jumping around and moving again, only to stall about 8 minutes in and pause till the new bar opens before movement occurs once more.

 

I have seen this so often over the years that I have concluded that I am not the only one trading like this. it has to be wide spread amongst the large professional market movers.

 

I agree of course with your opening sentence, and your last post about drilling down to various time frames. This simply hides or shows more detail. Anyone who uses a chart will have to make this decision on "zoom level" (and I also have my zoom levels), regardless of how he chooses to present the data.

 

But with your 10 minute chart example, certainly you would agree that this is purely anecdotal observation. Those with 15 minute charts might feel the same way you do, but logically it would conflict with your 10 minute anecdotal observation. And certainly it's objectively clear and provable that on the large majority of days, there are times when the market is always more active: at the RTH open and close, for example. But is 11:40 more significant than 11:45? According to you, yes. However, I have no reason to logically draw this conclusion, and I have no real evidence to affirm or deny this.

[karoshiman 10]

 

Well, the daily chart does have extra significance, due to the stop, or break from trading that creates clearly defined sessions.

 

...

 

 

 

I thought, it's because the vast majority of professional money (and, hence, the big volume) is traded at the daily open and close.

 

Well, maybe they trade at the daily open and close because of the "clearly defined sessions"... but anyway, what's the point in arguing about the reasons of their trades... all I need to know is that a lot of volume is traded during daily open and close... that's what gives them significance (by the way, I'm not talking about the ONE opening price at 3.30pm, but view the open more as an extended time period).

 

 

 

...

 

Many times I have watched a market on the 10 minutes chart, and it has a fully formed bar, where the price just sits there untill the open of the next bar. Then it starts jumping around and moving again, only to stall about 8 minutes in and pause till the new bar opens before movement occurs once more.

 

I have seen this so often over the years that I have concluded that I am not the only one trading like this. it has to be wide spread amongst the large professional market movers.

 

 

I've had similar observations on hourly bars, on volume bars and other stuff. The problem is, we tend to try to make sense of everything we see and so we observe certain "recurring" patterns. But what I've experienced is that if you start to analyze these observations with statistically significant amounts of data you'll become surprised of how few of your observations still stay valid.

[onesmith 10]

I can assure you there is real evidence to affirm or deny this.

 

is 11:40 more significant than 11:45? According to you, yes. However, I have no reason to logically draw this conclusion, and I have no real evidence to affirm or deny this.

[jackj 10]

I've had similar observations on hourly bars, on volume bars and other stuff. The problem is, we tend to try to make sense of everything we see and so we observe certain "recurring" patterns. But what I've experienced is that if you start to analyze these observations with statistically significant amounts of data you'll become surprised of how few of your observations still stay valid.

 

Beat me to it. Be careful of Confirmation Bias - look it up if you need to. Arguably a trader's worst enemy. Read Fooled by Randomness, too.

 

Now, PERHAPS these observations are legit, but your point, and the one that's been attempted to be made throughout this thread - how do we know WHICH timeframes to watch? - as well as what about different data feeds and computers not synched exactly the same, all lend a big question mark to any such observations and claims of relevance.

[joshdance 10]

I can assure you there is real evidence to affirm or deny this.

 

how do you define "significant," and how would you test and then collect this evidence?

[onesmith 10]

how do you define "significant," and how would you test and then collect this evidence?

 

I solved this in the past by asking myself the same questions. It wasn't difficult answering them.

[jackj 10]

I had a similar question (mine being, "so, what is the answer?"), but in looking at his profile, I see that he's a programmer, so I believe that he was simply speaking in general terms. That one can program and backtest it to see which is more predictive, but that he hasn't necessarily done so. I say that as he hadn't popped in previously, and even here didn't give an answer. Indeed, that would be one of the criteria - and people would differ on that - what constitutes significance? And whether either qualifies.

Perhaps he'll reply with his own answer that may differ, in which case obviously disregard mine, but in case he doesn't, I thought I'd add my .02...

[karoshiman 10]

Beat me to it. Be careful of Confirmation Bias - look it up if you need to. Arguably a trader's worst enemy. Read Fooled by Randomness, too.

 

Now, PERHAPS these observations are legit, but your point, and the one that's been attempted to be made throughout this thread - how do we know WHICH timeframes to watch? - as well as what about different data feeds and computers not synched exactly the same, all lend a big question mark to any such observations and claims of relevance.

 

 

Hm... that is exactly my point. Of course, one should be careful about confirmation bias.

 

My point is, if you are careful about confirmation bias in your statistical analysis of hard facts many of our initial observations will NOT be valid anymore. So, what I am saying is, we tend to assign significance to observations where there is none (I said in my first comment "...you'll become surprised of how few of your observations still stay valid"... maybe you did not notice the "few" in that sentence?).

 

Now, maybe I've misunderstood your point here...?

 

That's a good question of how we know which time frame we start our analysis with. At the end it is pretty much random, I guess. We are influenced by some sort of theory at the beginning of our journey, but from there we move on with our discovery of "how the markets work"... and it's good to question our theories with thoughts like the ones in this thread.

 

I do not focus on the close of bars, by the way, except for the daily... (... but I don't say, that it's not possible to build a profitable trading system on the basis of some intraday time frame closes.)

Edited May 14, 2012 by karoshiman

[jackj 10]

Sorry; my bad at being inarticulate. Indeed, I was acknowledging your point, and it seems that we're in agreement in all respects. I was intending to point the others, whom you were addressing, to the issue of Confirmation Bias and the Fooled by Randomness book. To quote another great one, Emily Latella (getting esoteric here)...nevermind...

[SpearPointTrader 10]

Beat me to it. Be careful of Confirmation Bias - look it up if you need to. Arguably a trader's worst enemy. Read Fooled by Randomness, too.

 

Now, PERHAPS these observations are legit, but your point, and the one that's been attempted to be made throughout this thread - how do we know WHICH timeframes to watch? - as well as what about different data feeds and computers not synched exactly the same, all lend a big question mark to any such observations and claims of relevance.

 

You have to learn to see the waves before they are complete. Then you have to pick the ones that are likely to be big enough for a decent profit to be extracted. Whatever time frame gives you that, is the one you want.

 

I don't choose, and use various time frames from the 3 minute, all the way up to the weekly.

[clmacdougall 11]

Time is Important

 

time Frame is not important

 

Time as in Duration is VERY IMPORTANT

 

a P&F chart has an X and Y axis

 

a Bar chart has ALSO an X and Y axis

 

Both charts move through Time. But differently !

 

As I said I think Time is very important..

 

Motorway

 

This most certainly is a non traditional way to approach P&F. Are you using time as a means to gauge momentum? Are you gauging time on each separate vertical wave or somehow on the horizontal. I would be thankful for any help you could offer Raymund.

P.S. I'm glad you got the book and are enjoying it!! Any new insights as a result of reading it?

 

All the best, Cory

[clmacdougall 11]

Do not look at the charts as just a form of breakout method.

 

Look at it in the same terms as market profile.

 

Look at The Horizontal Formations especially..

 

Your charts will seem to adapt to market time.

But they are not adapting .. They are market time.

 

Do not focus just on the static aspects of the chart

But the changes in activity as well.

 

Watching the chart FLOW and Build structure is the best teacher.

 

Motorway

 

I've got to ask if the structure you speak of is in the traditional sense of P&F patterns or something quite personal you've discovered.

Thanks, Cory

[Shanghai 10]

.. there's also this to consider: you and I could both be looking at five minute charts, but if my five minutes are staggered by, say, just one minute, then I'll be looking at a rather different chart to you. If price moved a lot in that 'sixth' minute, then your five minute candle will be a nice long one, and I'll have some crummy little doji.

Please forgive my ignorance. How in the world could this happen? I don't understant :doh:. What do you mean by staggered? Or having two 5min different charts?

 

Also,

...if you test for a particular method, then stick with that for consistency.I find it funny when someone says it works on a 5min bar, but does not on a ten minute bar.....to me it does not make much sense.....anyone else know?

I think it could make sense only if you buy and sell within that 5min period? You could be winning a 5min bar but get stopped out in the next 5min if that is a micro period strategy I guess.

[joshdance 10]

Please forgive my ignorance. How in the world could this happen? I don't understant :doh:. What do you mean by staggered? Or having two 5min different charts?

 

Like the charts I attached in this post:

http://www.traderslaboratory.com/forums/technical-analysis/12081-close-bar-meaningless-6.html#post139955

[SIUYA 11]

I think it could make sense only if you buy and sell within that 5min period? You could be winning a 5min bar but get stopped out in the next 5min if that is a micro period strategy I guess.

 

exactly. IMHO trends occur in the time frame you are looking at, so trade those and apply the setups and entries and exits that apply to that time frame.

While its fine to maybe look at smaller time frames to try and improve/fine tune the entries, you dont want to fall into the trap of confusing yourself by saying, on the daily chart its a donwtrend, the 60 minute its an uptrend, on the 5 min chart its a downtrend.

 

Also If you have a system that works while testing on the 5min charts, works on the 10 min charts, but does not work on the 6,7,8,8.5 min charts then maybe its not a great system (dependant on slippage and commissions etc). Hence the requirement to remain consistent.

[equtrader 10]

Also If you have a system that works while testing on the 5min charts, works on the 10 min charts, but does not work on the 6,7,8,8.5 min charts then maybe its not a great system (dependant on slippage and commissions etc). Hence the requirement to remain consistent.

 

I think everyone who is system trading or planning to develop systems should read this blog and try the simulation in the link that is provided in it:

 

Fooled by Randomness Through Selection Bias | Price Action Lab Blog

 

When I tried it I screamed "Noooooooo...":)

[zdo 20]

...If you have a system that works while testing on the 5min charts, works on the 10 min charts, but does not work on the 6,7,8,8.5 min charts then maybe its not a great system .

 

… not disagreeing in general…In a nice clean statistical sense with systems using trading’s characteristic 'robust' patterns, this is accurate…

... however, to point out one of the important exceptions….

> when method aligns with cycles (and also tests for presence of satisfactory ‘cyclicity’) it becomes blatantly obvious which timeframes outperform and which timeframes are victims of phase cancellation, etc.

…not thinking in probabilities… thinking in cycles

 

…hardly thinking at all… after being gone from home and forums for all but a few days in the last two months *, it may not be wise to go flailing at SIUYA first post back…and haven’t read the link equtrader posted but most likely this post diverges from its point too…

 

 

* (and fortunately finding the derecho didn’t lay the place on its side)

 

See ya'll next week

[Patuca 13]

if close is meaningless then all is meaningless. what is support resistence based on? patterns? staggered time frames? how would that affect support resistance? high , low, open, all meaningless. TF all meaningless. mathematical calculations all meaningless... because...different staggered time frames BS...fib meaningless..... gaps meaningless...all kinds of charts..meaningless...staggered TF invalidates all measurements....computer sync invalidates all measurements....no real reference points...all is meaningless....fruitless..a waste of time...send your money to me...i will give it some meaning on daily and 5 minute charts.....

 

this is all BS....time to get on my moto and ride away...leaving the land of dipsticks...adios....