Discussions on Order Flow/Volume | Page 7 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

How to flow with the order flow? 26 replies

Applying Order Flow & Volume Profiles To Futures Trading 121 replies

Reading Order Flow 19 replies

Order Flow 26 replies

Oanda Order Flow Box 5 replies

Attachments: Discussions on Order Flow/Volume
Exit Attachments

Discussions on Order Flow/Volume

  • #121
  • Nov 28, 2019 6:10pm Nov 28, 2019 6:10pm
Quoting Froggy
{quote} Don, ALWAYS family comes first and you shouldn't be worried about this thread. I for one (I am sure everyone else) will wait until you are ready to post again. Send you and your mum my best wishes.
That's very kind of you, thank you!! I just didn't want any of you to think I suddenly abandoned the thread after seeing such great enthusiasm and momentum. Again I am overwhelmed by your support, from all of you and very grateful. We'll talk again soon just as soon as I am able.

One final message before I go. I know many of you have this issue. When stressed in your personal life...STOP trading until you can settle back into your rythm. Otherwise you'll just give back gains that you worked so diligently to obtain. The market will always be there.

I most likely will not be in the market until after the New year.
@xxxx§|[::::::::::::::::>
  • #122
  • Edited Nov 29, 2019 1:34am Nov 28, 2019 7:55pm | Edited Nov 29, 2019 1:34am
Quoting DonPato
Hello again friends. I hope you have been enjoying your Thanksgiving holiday. I was all set to do so until my brother called with some unsettling news. My mother is in very grave condition and I must return to California. I will most likely not post for awhile, and any posts I do write will be sporatic at best. I will check with my cell phone and try to answer questions, as much as possible. I'm very sorry but family always comes first. I hope to post again soon (relatively speaking)
The Don, you will be sore missed, but family first is a standard procedure! When life is at risk, everything takes back stage. I don't even know if I should post this. I wrote a reply on text file to post. Let me post so I won't delete it mistakenly, you take your time to answer my friend. Maybe other members contribute.

I was racking my brains on the order flow delta charts you left us. I hereby have these to say, to clarify if I understand rightly. Bear with me, it's a long one.

# On Charts;
Your market structure explanation on aggressive/market and passive/limit orders shed more light on the construction of this charts.

1. So what we are seeing are aggressive/market orders' delta at each level. This does two things;
i. It shows trading at each price level.
ii. Because it is delta, unlike tick volume, it show TRUE dominance/neutrality of either sellers or buyers at each level.
iii. Because it is delta, there is a little blind spot here, but may not be so important. At quiet levels, I mean levels of low delta volume. It either means thin trading or due to delta being a net of bid/ask tick, it could mean they just cancelled each other out. Example at a low delta level, possible the highest volume could have traded there, but it was a close contest between buyers and sellers; example 200,000 aggressive buying LESS 199,935 passive selling net = 65 Delta = low delta volume print. Please correct me if this is wrong.

2. Aggressive/market order should produce expected result else that is unusual. Similar to Wyckoff Effort vs Result law. Meaning a swath of aggressive buying(green block) should produce higher prices, else that means there is a higher volume of passive selling liquidity above that level, at least for now, if there is another green block at the next bar yet same result, no higher prices, it is a confirmation of thick passive selling liquidity sitting above that level.

3. (Optional) This gives me an idea. If we swap the colors for their opposite counter parts, example
Every Green for Red
Every Red for Green
Every Aqua for Magenta
Every Magenta for Aqua

Then we have effectively constructed the passive order flow delta chart. Meaning passive order prints! Granted this does not give us the whole order book but this is "Traded" or "Activated" or "Executed" orders from the order book. Can we then guess construct roughly the order book? Is there even a need for this exercise? What can be a case of useful application?

4. Turning Points Conditions Combination;
a1. On climax, all/majority of participants have longed or shorted

AND

a2. Then late participants jumping on the band wagon slams into liquidity wall hard. The market balances. The current level of liquidity wall may be chewed through with another level just above that one, this causes sudden whipsaw aka pinbar, then the scale tips, I like to think of it as "order flow take-over", they get stopped out and the turning point is completed.
Note; in this case it is combat all the way till turning!

This a case that looks like this;
Attached Image


b1. On climax, all/majority of participants have longed or shorted

AND

b2. Disinterest or Lack of interest leading to drifting.
This is likely better illustrated by falling market, currently I think AUDNZD fits this case! Due to seller and buyers not eager to trade, price keeps drifting lower with occasional buyers trying to pick bottoms, then gets stopped out, basically not many people trading mostly the big money, then finally after several stop outs, bottom picking wiz-kids quits, volume dries up considerably, then, smart money sees opportunity to buy cheap, comes in, along with early sellers taking profit, market reverses up. But, since we did not have many trapped shorts that sold late and still holding hoping for lower prices, can this new move be sustained with so few committed to lose?

This a case that looks like this;
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot3.png Size: 81 KB



This is my take, please correct where wrong.
  • #123
  • Nov 29, 2019 4:44am Nov 29, 2019 4:44am
Quoting DonPato
Hello again friends. I hope you have been enjoying your Thanksgiving holiday. I was all set to do so until my brother called with some unsettling news. My mother is in very grave condition and I must return to California. I will most likely not post for awhile, and any posts I do write will be sporatic at best. I will check with my cell phone and try to answer questions, as much as possible. I'm very sorry but family always comes first. I hope to post again soon (relatively speaking)
Sorry to hear about your mother Don. Wishing you all the best.
Money = Labour coupons
  • #124
  • Nov 29, 2019 4:57am Nov 29, 2019 4:57am
Nice thread!
I will join the conversation here next week
  • #125
  • Nov 29, 2019 6:17am Nov 29, 2019 6:17am
which is the best forex exchange to trade on
  • #126
  • Nov 29, 2019 6:34am Nov 29, 2019 6:34am
Quoting danasa
{quote} The Don, you will be sore missed, but family first is a standard procedure! When life is at risk, everything takes back stage. I don't even know if I should post this. I wrote a reply on text file to post. Let me post so I won't delete it mistakenly, you take your time to answer my friend. Maybe other members contribute. I was racking my brains on the order flow delta charts you left us. I hereby have these to say, to clarify if I understand rightly. Bear with me, it's a long one. # On Charts; Your market structure explanation on aggressive/market...
Which indicator changed the candle color? I am really excited to know more!
  • #127
  • Nov 29, 2019 7:39am Nov 29, 2019 7:39am
Quoting cindyrose143
which is the best forex exchange to trade on
The 'forex exchange' is more an overall name for the market,
if you are referring to brokers recommendation I would advise you to go for a broker with UK FCA license and with cheapest spreads and best trading conditions for you
  • #128
  • Nov 29, 2019 11:26pm Nov 29, 2019 11:26pm
Just in case anyone finds it useful as a condensed reference of DonPato main points.
Attached File(s)
File Type: pdf Discussion On Order Flow & Volume.pdf   1.0 MB | 1,259 downloads
File Type: docx Discussion On Order Flow & Volume.docx   901 KB | 930 downloads
Simplicity Is The Ultimate Sophistication!
  • #129
  • Nov 30, 2019 1:26am Nov 30, 2019 1:26am
Quoting Impenia
Just in case anyone finds it useful as a condensed reference of DonPato main points. {image} {file}
Good job! Thanks for taking time to put this together. The information DonPato has shared here is priceless!
Anyone struggling with trading should print this summary and read it over and over again until it clicks. Enough said..
  • #130
  • Dec 1, 2019 12:22am Dec 1, 2019 12:22am
This is one of the best "trading" videos of all time and fittingly belongs in one of the best threads of all time. I can't thank you enough DonPato.

Inserted Video


DonPato has motivated me to take another look, no, to take a better look, at the Market.

Attached Image (click to enlarge)
Click to Enlarge Name: Aggro 1.png Size: 301 KB
Without VSA, you're playing checkers while the Smart Money plays chess.
  • #131
  • Dec 2, 2019 12:53am Dec 2, 2019 12:53am
Quoting NeilWagner
{quote} Which indicator changed the candle color? I am really excited to know more!
Can't yet! but should it have a killer application, it is a matter of requesting this option from the developer of the indicator. No useful application, suggested by members here or the Don, no need to waste time coding it.
  • #132
  • Dec 2, 2019 3:34am Dec 2, 2019 3:34am
Interesting thread!

DonPato, have you reviewed how your volume profile indicator splits the volume in bullish and bearish parts or did you take its calculations for granted?

I had a shot at recreating the indicator in mt4 and the result looks very similar to the graphs shown in this thread, but I found it to be sensitive to the "base" timeframe chosen - in your case H2. My guess is that the calculation behind only considers the candlestick range and close but not the open. In this way the choice for the sampling period and frequency impact the profile's shape and colors quite a lot.

Putting it to the extreme:
If prices would follow a sine wave and you would apply OHLC sampling at its base frequency with the open/close each time at the top, you would have a series of bullish pin bars. Considering no variation in volume, I expect the volume profile indicator to show strong buy volume for those bars.
Now sampling at the bottoms of the sine, you would get a series of bearish pin bars and here I expect the indicator to turn red.
But ideally you would expect no buy or sell bias to be shown, as a sine wave represents a perfect ranging market and the underlying data had not changed.

I'm not criticizing your approach in any way, though was wondering about the math behind. If you found the indicator in its current state to be useful and reliable, I am more than willing to believe that it works because it works.
  • #133
  • Dec 2, 2019 6:30am Dec 2, 2019 6:30am
Quoting kennerU
Interesting thread! DonPato, have you reviewed how your volume profile indicator splits the volume in bullish and bearish parts or did you take its calculations for granted? I had a shot at recreating the indicator in mt4 and the result looks very similar to the graphs shown in this thread, but I found it to be sensitive to the "base" timeframe chosen - in your case H2. My guess is that the calculation behind only considers the candlestick range and close but not the open. In this way the choice for the sampling period and frequency impact the profile's...
Very interesting take...appreciated though nothing useful out yet, but very directional, nice angle.
  • #134
  • Dec 2, 2019 7:36am Dec 2, 2019 7:36am
Quoting kennerU
Interesting thread! DonPato, have you reviewed how your volume profile indicator splits the volume in bullish and bearish parts or did you take its calculations for granted? I had a shot at recreating the indicator in mt4 and the result looks very similar to the graphs shown in this thread, but I found it to be sensitive to the "base" timeframe chosen - in your case H2. My guess is that the calculation behind only considers the candlestick range and close but not the open. In this way the choice for the sampling period and frequency impact the profile's...
Hello all...I am in the hospital with my mom. I am answering on my cell phone so specific screenshots are not possible but I wanted to speak to the points kenneru brings up.

First I do not know the specific calculations done to separate out @bid and @ask. The proprietor of this code was unwilling to expound and rightfully so. However I have found the information it shows to be accurate (which ultimately depends on the data feed). The fact that the histogram doesn't show what one might expect to see on certain candles is exactly the point. The difference between seeing the order flow and watching the price response is why I use this. Again, it is a tool to help me better observe the order flow and watch how price responds.

If something doesn't seem "right" this usually means change is either coming or here and price will follow very soon. But the best way to know if something just seems off is to be intimately familiar with"normal".

Thus I would like to caution everyone again, to try NOT to fall into the trap of treating this as though it were some "superior" form of technical analysis. This trap has a wide mouth and once you're stuck in it you will become frustrated (as you have been with ever other form of TA). This is not a system with "signals" not can it be automated because the changes in order flow can only be detected by a well functioning human brain based on familiarity and good decision making. It is completely discretionary.

Think of it like a mother who takes a child's temperature because she "just knows" he/she is getting sick. She "knows" because she is so attuned to her child after carrying him/her in her own body. She can see very subtle changes that others can't and takes action long before anyone else can see anything is wrong.

You must become this familiar with the"normal" order flow that you can "see" or "sense" something is about to change and take action long before anyone else with their technical analysis can see anything happening with price.

So you see? It is not really relevant if the graph isn't quite what you expect. And in truth it is not what I expected either when I first saw it. But after observing this for awhile I have found it accurately shows the overall (Delta) flow which allows me to make better decisions that are far more accurate than the bar-by-bar delta extrapolation that I was using. While this also remains part of my way of evaluating the order flow the histogram running behind each price bar is far more informative.

BTW the 2 hr chart shown is irrelevant. I only look at each histogram (which is daily) and watch how price responded to the red/green indicators for net (delta) buying and selling. Does price rise during or immediately after the buying?...if not why not? Perhaps there is an area of deep liquidity...or perhaps those buys are really the smart money taking profit after a long move down.

Remember order flow/volume will often precede price movement if only just slightly.
@xxxx§|[::::::::::::::::>
  • #135
  • Dec 3, 2019 5:14am Dec 3, 2019 5:14am
Quoting DonPato
{quote} Hello all...I am in the hospital with my mom. I am answering on my cell phone so specific screenshots are not possible but I wanted to speak to the points kenneru brings up. First I do not know the specific calculations done to separate out @bid and @ask. The proprietor of this code was unwilling to expound and rightfully so. However I have found the information it shows to be accurate (which ultimately depends on the data feed). The fact that the histogram doesn't show what one might expect to see on certain candles is exactly the point....
Thank you for your response, I think it answers my question. Also much appreciated considering your situation.

Based on the resemblance, my guess is that the proprietor of the delta volume profile indicator used a formula that he mentions elsewhere on his website. In my opinion this calculation oversimplifies the situation and is not sufficiently robust. Note that I'm not claiming the same about its results.
In general, robustness is an important system aspect in many fields (engineering/programming/etc). An algorithm should respond in a - more or less - reproducible and reliable manner to the data you feed to it, regardless of arbitrary variations in sampling. If it doesn't then I don't trust it.
Currently I am trying to find something solid that treats the OHLC data in a more balanced way to get to volume delta.

About the "sub"-timeframe of 2hr, I am not sure if it is completely irrelevant. I'm aware of the histograms being drawn on a daily interval and those are the ones to look at, however to calculate the delta volume for each of the price levels, I'm pretty sure it uses the 2hr volume/OHLC data. The website's description mentions this and you also clearly see how the histogram bars react to the bodies and wicks of the 2hr candlesticks.

Don't get me wrong, I think the approach of order flow analysis has great potential. It seems to bring more value than standard technical indicators. I do realize I would still require a lot of practice before being able to properly read order flow from the delta volume profiles.

It's just that I do not like it that using the alleged formula, my indicator (and I expect yours as well) shows such difference in daily histograms while varying the underlying timeframe. Perhaps I can show some screenshots later on to clarify what I'm talking about.
Also my previous comment wasn't so clear, in a sense that I now read it as if I said to be expecting certain behavior from the indicator. In fact I was not, I only wrote "expect" because I did not want to make claims based on assumptions about how it would behave since I do not have the indicator.
  • #136
  • Dec 3, 2019 6:49am Dec 3, 2019 6:49am
Kenneru... thank you for your response. I completely agree with your points and confidence in the tools you utilize is a must. I look forward to seeing your screen shots. I would explore them myself if I had more time. Perhaps in a few more days I will be able to set up my mobile office and can more thoroughly understand the points you bring up. I would like to emphasize however that what I'm using now (the configuration) has greatly improved my accuracy. I think this can be noted by reviewing my thread on volume.

I look forward to seeing what you can show us. Thank you for your time.
@xxxx§|[::::::::::::::::>
  • #137
  • Dec 4, 2019 1:47pm Dec 4, 2019 1:47pm
This is what the indicator with my latest calculations looks like for EurUsd.

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSD-ECNH1.png Size: 55 KB


And here the indicator in its base form, of which I expect it would approximate your graphs for the same timeframe (1hr):

Attached Image (click to enlarge)
Click to Enlarge Name: EURUSD-ECNH1_.png Size: 55 KB


I find the 1st one less chaotic, 'cleaner' actually. Still not so happy about the timeframe variation (not shown here yet).
Note that I do not let the wicks contribute to the delta volume, as for me they represent zero net price movement. I find them distorting the view in the original indicator.

And below my take on EurNzd, what do you think? After-the-fact is of course easy but I hope I am making sense out of it.

Attached Image (click to enlarge)
Click to Enlarge Name: EURNZD-ECNH1_.png Size: 67 KB
  • #138
  • Dec 4, 2019 4:21pm Dec 4, 2019 4:21pm
Hello again KennerU. Forgive me (still on my cell). However I would like to point out to you why I think your 2nd posted screenshot shot is a more accurate view of the order flow.

On a bearish candle that breaks below the opening range (marked below with the number 022). You can see some buying that was overcome by selling when the range broke. This is not present on the same candle in your first screenshot. As well in the large bullish candle that breaks higher you can clearly see the buying tapering off. This is typical of sudden large moves as the liquidity dries up. I feel this reflects a more accurate picture of how the order flow is functioning.

As for the difference between the same price level on different time frames, this would simply be the result of more order flow over the same area that is seen on a larger time frame. Again I see this same thing in futures. A small time frame divides the order flow into numerous bars, each with a slightly different order flow/volume profile and would look markedly different when all combined together in a single (larger) time candle.

Adjusting the calculation to fit your idea of what it "should" look like may have other unintended consequences.
@xxxx§|[::::::::::::::::>
  • #139
  • Dec 5, 2019 12:07am Dec 5, 2019 12:07am
Hello, Don
could you give an example that "order flow/volume will often precede price movement if only just slightly"?
How do I feel that the order flow and price changes are consistent？
just briefly talk if it is not convenient to sent pic on cell..
learning...
  • #140
  • Dec 5, 2019 7:08am Dec 5, 2019 7:08am
Quoting hefei
Hello, Don could you give an example that "order flow/volume will often precede price movement if only just slightly"? How do I feel that the order flow and price changes are consistent？ just briefly talk if it is not convenient to sent pic on cell..
Hello Hefei...there is actually a very good example of it on the hourly screenshot provided by KennerU above.

On his second screenshot...on the candles marked 015, 014. You can see buying (in green) start coming in strongly about 4 candles before price moved up in the "whoosh" higher. These represent the subtle changes in order flow that I was speaking about. While these candles are (in my view) not actionable - meaning I would not enter long or short on any of them the are showing me that change is coming and indeed that was the case.

My entry level based on the following two order flow candles would have been at level 1.10000 and 1.0925. you can take a look on that chart and see where those line up and then see how that would work going forward.
@xxxx§|[::::::::::::::::>
3 traders viewing now