Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Auction Market Theory and Market Profile 22,206 replies

MarketProfile/Volume Profile/Auction Market Theory indi? 17 replies

ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies

Auction Market Value Analytics Charts & Data 14 replies

Analytics, recommendations - S&P 500, Euro and other 11 replies

Attachments: Auction Market Value Theory & Analytics
Exit Attachments

Auction Market Value Theory & Analytics

What is the meaning of all the two letter acronyms in the first column of the TFF Averages Table?
Quoting global3
What is the meaning of all the two letter acronyms in the first column of the TFF Averages Table?
Ticker symbols
https://web.archive.org/web/20150118...0/dailyda.html
Quoting thetail
{quote} The result of the calculation of the NET value (Direction + Activity) is asymmetrical for ascending vs descending markets. And it's not possible.
Could you clarify what you mean by this? When would they be asymmetrical for ascending vs descending markets? Thanks.

For me, I find it odd to assign +1 to TFF increasing while also assigning +1 to TPO total, VTY, & VA Range increasing. Unless perhaps I am misunderstanding how to interpret daily references, which largely reflect short term traders.
Quoting nonlinear
Primarily posts 569-574, along with some observations of activity reference point tallies throughout the thread (apparently assigning +1 to rising tff. My gut says assign -1 to rising tff). Comments like this, for example: "So in the short TF (daily profile) a greater TFF means price was facilitating two-way trade. That tells us what? That the short tf participants are happy, 'active' and in control." Many thanks for your thoughts on this. Please excuse me if I made blunder; this a complete retooling of my view on price action, well auction action...
My point of the posts to remind people not to confuse "how" Dalton uses the word "facilitate" in MOM as a way to interpret TFF by Jones.

"As we noted earlier, the best indication of trade facilitation is volume". MPM p 140
"Not surprisingly, volume is also the best measure of a market's ability to facilitate trade". MOM p 188
"Put simply, the greater the volume of transactions, the better trade is being facilitated". MOM

In that type of context, a MP traders perspective, the POC VA are where the most trade is being facilitated

"Finally, an "outside day" occurs when a day's value area overlaps the previous day's value area on both extremes, and represents greater trade facilitation. Much like a Neutral day," MOM p188-89

in MP nomenclature, neutral day is a balanced profile,, from that perspective the POC the area represents greater trade facilitation.

From a MP traders perspective, the POC VA are where the most trade is being facilitated. "How well the market is facilitating trade" in a MP context that can be taken as how well the market is facilitating 2 way trade.

My point of the posts to remind not to confuse "how" Dalton uses the word "facilitate" in MOM as a way to interpret TFF by Jones.
As I tell people over & over, This is not MP.

Quoting mzvega
"So in the short TF (daily profile) a greater TFF means price was facilitating two-way trade."
In VBPT, those who trade "Daily Profiles" are MP traders who employ MP concepts...
Probably not the best examples.

Cont'd
Markets are not efficient, rather they are effective - Jones
Quoting nonlinear
Primarily posts 569-574, along with some observations of activity reference point tallies throughout the thread (apparently assigning +1 to rising tff. My gut says assign -1 to rising tff). Comments like this, for example: "So in the short TF (daily profile) a greater TFF means price was facilitating two-way trade. That tells us what? That the short tf participants are happy, 'active' and in control." Many thanks for your thoughts on this. Please excuse me if I made blunder; this a complete retooling of my view on price action, well auction action...
You are absolutely correct about calculating the TFF ……...

Cisco provided different data packages depending on what type of trader you were. Since they were the first of only a handful licensed data vendors for cme LDB and market profile. Some tools & data packages could be used by both MP & Value analytics.

The visual Graphic could be used for both MP & Value analytics traders
From a MP perspective the data provided compares one day's references to the previous day's data (greater or smaller) is all that is need for MP, the overlay data also can be used by different kinds of traders. However AMVA employs a "3 day rule" so to use the daily data from the visual graphic report for AMVA you were required to fill out a set up sheet by hand

Set up sheet by hand……….
https://web.archive.org/web/20150114054136/http://www.cisco-futures.com:80/amva_setup_sheet.html

This is when and why I first started tracking the data by hand with spreadsheets. Note there is no +1 or -1 only A,Q,U,D, There is no +1 or -1. You need first a prior knowledge of what of each of the references measures, and how it is used to determine if A more active, or Q quieter ,etc……

When one reads the instructions in the source material on "how to use the data" you learn how
How to tally and how to use the data.

When you see a picture of the data posted to the thread, it is used as an example, followed by a link that contains the instructions.

Later, Cisco created a single one page report, for AMVA containing all the data need for analysis. Rather than copy by hand data from different data packages, you needed only subscribe to one data package……… note the A,Q,U,D https://www.forexfactory.com/showthr...01#post9223101

Long story short, rather than 5.4 > 5.2 =+1 used by those who haven't read or followed the source material instructions on how the data is actually used, you took the time to read about TFF to understand what it is your actually calculating. Those who follow the source material as I do, know you answer the Active or Quieter first…..then you tally.

Learn what the reference actually means first….You are absolutely right that that it should be -1. You did the one thing I can't seem to get anyone to do….learn what each of the references actually means, how and why it is calculated first. Then use that data to organize a spreadsheet…..
What does it "mean" if TFF is increasing/growing (3 day rule) or quieter/or more active

The question isn't is one value greater than the other i.e. 5.4 > 5.2 =+1 or 5.2 < 5.4 =-1

The question is how well is the market facilitating trade? Not is one value greater than the other…...
"is the market facilitating more or less trade (greater or smaller)

If TFF is decreasing you have a more liquid market ,more activity A= +1
If TFF is increasing/growing you have a a less liquid, less activity quieter Q= -1

The answers to most questions are found by reading the source material........
Attached Image (click to enlarge)
Click to Enlarge Name: tff.JPG Size: 95 KB
Markets are not efficient, rather they are effective - Jones
Quoting nonlinear
{quote} Could you clarify what you mean by this? When would they be asymmetrical for ascending vs descending markets? Thanks. For me, I find it odd to assign +1 to TFF increasing while also assigning +1 to TPO total, VTY, & VA Range increasing. Unless perhaps I am misunderstanding how to interpret daily references, which largely reflect short term traders.
Yes, in fact sounds strange to me too ... but there is more... the asymmetrical result in calculation.

Assume that we use for direction calc these 7 references: Open, High, Low, Close, POC, VA_High, VA_Low
and for activity calc these 3 references: VA_Range, TPO_Total, VTY (excluding for the moment the TFF)

In a green candle day (ascending market) like the one on EUR/USD pair in the image below, for sure all the dir refs are increasing, so DIR=7
and probably, and anyway we assume that, all the 3 activity refs are increasing, so ACT=3
DIR+ACT=NET
7+3=10
NET=10

Go now on a perfectly mirrored descending market, the USD/EUR pair.
At the opposite now, for sure, all the dir refs are decreasing, so DIR= -7
but probably, and anyway we assume that, all the 3 activity refs are increasing (same as in the ascending case), so ACT=3
DIR+ACT=NET
-7+3=-4
NET=-4

So we have two different NET absolute measure (10 vs 4) of the same move. And, in my opinion, it's not correct ....

As noted by others, the correct procedure would change the sign in accordance with the sign derived from the calculation of the DIR references,
so if DIR is negative and NET is positive we assume NET as negative
NET=-7+(-(+3))=-10
So the equality of absolute values is respected.

The problem remains of what to do when DIR=0 and here we can discuss again ...

And then there is TFF reference too, wich sign we assign to it if its value is increasing (so there is less activity)?
Attached Image (click to enlarge)
Click to Enlarge Name: EURUSDEUR.png Size: 20 KB
Hi friends and traders,

its good to see a discussion in the thread. And its important to keep the thread alive.
First we must clarify that we have 2 questions here:

1. Why is the rising TFF calculated with +1 in the act table?
2. Why is the net flow (act+dir) calculated in that fashion without paying attention to the dir (+/-)?

My answer for the first question:
The rising TFF is calc with +1 in the act table, because of the choosen strategy (breakout strategy). You dont want to trade a market, thats already underway. You want to trade a market thats still in a pause (to trade the break and limit your risk). But you can adjust that formula, as long as you know the refs. For example you can simply program a voting table (vote for balance or vote for break). But the point here is, you have to know the condition of the market to trade it accurately.

My answer to the second question:
The net flow calculation in ciscos example is suited for a UP-breakout strategy (for example 20 day trending up). After reading Daltons great work "Markets in Profile" I learned, that it can be very tricky to trade against a trend... And the way cisco uses its net flow data could save you from doing that/falling into a trap.
You can adjust the formula. I have done that too and it works. How? Use 2 formulas (one for trending up - one for trending down). If you only change the math sign, you will mess up the whole idea. By the way Dalton invented a similar indi. You can find it somewhere in his books.


greetings
c
Quoting dljonesFan
My answer for the first question:
The rising TFF is calc with +1 in the act table, because of the choosen strategy (breakout strategy).

My answer to the second question:
The net flow calculation in ciscos example is suited for a UP-breakout strategy
...
Hi dljonesFan, thanks for your inputs.

Partly I don't agree with your observations.

Point 1:
If we consider the DTE Flow Table as a predictive tool for a BO then probably assigning +1 for increasing TFF values is the correct way, (increasing TFF value is what I look when I search extreme value of TFF as explained in the other post).
But if we consider flow as picture of the "power" of the market, once we had a direction (DIR) and our car is moving in that direction, the ACT is like a turbo boost and adds power in the same direction ... so, if DIR is positive, increasing VA_Range and TPO_Total and VTY are positive too for that direction (more power...), but increasing TFF is like a brake ... slowing and pausing the move so maybe is better use -1.

Point 2:
Here I totally disagree. Better you don't have an only up strategy, ( and don't think that's the way on cisco) especially on futures and particularly on currencies, it's a non-sense (maybe on stocks you can have).
Forex pairs are ratios, one currency against another currency, so, on EUR/USD for example, what is long from an EUR point of view is the opposite short from an USD point of view, same "intensity", opposite "direction".
So having 2 formulas (if you intend 2 calculation that gives 2 different absolute values) is incorrect in my opinion, because in this yes case you can fall in a trap believing for example that EUR/USD NET= 10 has bigger probability to go UP then the USD/EUR (or 1/EURUSD) NET=-4 to go down.
No, they have the same identical probability because it's the same thing, same movement, same forces.
Quoting thetail
{quote} Hi dljonesFan, thanks for your inputs. Partly I don't agree with your observations. Point 1: If we consider the DTE Flow Table as a predictive tool for a BO then probably assigning +1 for increasing TFF values is the correct way, (increasing TFF value is what I look when I search extreme value of TFF as explained in the other post). But if we consider flow as picture of the "power" of the market, once we had a direction (DIR) and our car is moving in that direction, the ACT is like a turbo boost and adds power in the same direction ... so,...
Hi tail,

I must say Iam surprised by your answer....

Point 1: You cant predict the markets... You dont understand my point...

Point 2: Do you realy belive, that I dont know that markets can break out or down? You dont understand my point.

Maybe you dont understand me because of my english. I will have a look again tomorrow.

greetings
C
Quoting dljonesFan
{quote} Hi tail, I must say Iam surprised by your answer.... Point 1: You cant predict the markets... You dont understand my point... Point 2: Do you realy belive, that I dont know that markets can break out or down? You dont understand my point. Maybe you dont understand me because of my english. I will have a look again tomorrow. greetings C
1. Oh yes obviously, I know, no one can predict, we are talking about odds. Predict was a contraction, estrapolated from cisco web site

The collective posting is the 'Market Flow Table'. The goal of the Flow Table is to predict the market's most likely flow direction for the coming day.
...
The sum of the two sets of measures is the 'NET'. A positive NET indicates an upward flow and predicts higher prices. Conversely, a negative NET flow predicts lower prices.


2.

Maybe we can't understand each other because of OUR english ... it's not easy ...
Question 1 (TFF): I dont care how do you use your net tables. Its important that you know what the tff is about. And I think you already know that. Dont get me wrong. If you belive that the mathematical sign should be negative or positive and that suits your personal strategy, then change it. The important fact here is, that you know and understand the current condition and can apply successfully a strategy. We all have our own tables.

Question 2: Point here is that you can have counter auctions in an uptrend AND in an downtrend. And its possible to show that with the right calc of the net flow. Iam sure that the net flow examples from cisco are all about uptrend data.
thetail, I see what you mean, I agree with that approach if you are combining dir and act tallies for a net value. Of course, one need not combine them unless you really want/need a net value. I am trying not to use a value for my net. For example, if dir is higher and act is greater, and then flow is up; if dir is lower and act is smaller, then flow is neutral, etc. Good discussion here, thank you, including dljones and mzvega.
Quoting nonlinear
thetail, I see what you mean, I agree with that approach if you are combining dir and act tallies for a net value. Of course, one need not combine them unless you really want/need a net value. I am trying not to use a value for my net. For example, if dir is higher and act is greater, and then flow is up; if dir is lower and act is smaller, then flow is neutral, etc. Good discussion here, thank you, including dljones and mzvega.
Yes, using only dir and act - thats also a way to deal with the problem. But keep the bigger picture in mind (trending,bracketing,...).

greetings
C

Quoting thetail
{quote} Yes, in fact sounds strange to me too ... but there is more... the asymmetrical result in calculation. Assume that we use for direction calc these 7 references: Open, High, Low, Close, POC, VA_High, VA_Low and for activity calc these 3 references: VA_Range, TPO_Total, VTY (excluding for the moment the TFF) In a green candle day (ascending market) like the one on EUR/USD pair in the image below, for sure all the dir refs are increasing, so DIR=7 and probably, and anyway we assume that, all the 3 activity refs are increasing, so ACT=3 DIR+ACT=NET...
@Tail:

That is what I tried to explain and already showed months ago. We must change the math sign IF THERE IS AN DOWNTREND in the longer timeframe. NOT because of the DIR of one day... But my english was not good enough...

"As noted by others, the correct procedure would change the sign in accordance with the sign derived from the calculation of the DIR references,
so if DIR is negative and NET is positive we assume NET as negative
NET=-7+(-(+3))=-10
So the equality of absolute values is respected."

Or Jones was wrong. And the dir of the day makes the math sign... But that would be such an obvious error - I cannot believe that he would made an error like that.

"We first selected a set of Visual Graphics reference points for directional (price higher (+) or lower (-)) measurement. We selected another set for activity (more active (+) or less active (-). The activity set is an 'assist' for the directional. That is, if price is flowing up (or down) and activity is increasing, the flow is strengthened. The sum of the two sets of measures is the 'NET'. A positive NET indicates an upward flow and predicts higher prices. Conversely, a negative NET flow predicts lower prices." (Flow Analysis Studies, Jones 2009)

I dont use the net flow sum. Because of all that uncertainty. MzVega can you help us out there?
Quoting nonlinear
thetail, I see what you mean, I agree with that approach if you are combining dir and act tallies for a net value. Of course, one need not combine them unless you really want/need a net value. I am trying not to use a value for my net. For example, if dir is higher and act is greater, and then flow is up; if dir is lower and act is smaller, then flow is neutral, etc. Good discussion here, thank you, including dljones and mzvega.
Yes i prefer to have a value, a DIR value, an ACT value and a NET value like the DTE flow table.
Anyway if DIR is +7 or DIR is HIGHER it's the same thing, the Higher is calculated too, if you have 5Up and 3Down you have a 2Up so DIR=HIGHER
But in my opinion you have less information, less granularity and possibility to use those numbers for other studies, think at a mean or standard deviation calculation; obviously this not imply that is a disadvantage or a problem, depends if you needs or not as you say.

Anyway NET calculation or not the asymmetry remain and for me this is crazy.
Quoting dljonesFan
{quote} Yes, using only dir and act - thats also a way to deal with the problem. But keep the bigger picture in mind (trending,bracketing,...). greetings C {quote} @Tail: That is what I tried to explain and already showed months ago. We must change the math sign IF THERE IS AN DOWNTREND in the longer timeframe. NOT because of the DIR of one day... But my english was not good enough... "As noted by others, the correct procedure would change the sign in accordance with the sign derived from the calculation of the DIR references, so if DIR is...
Oh well ... we agree on the formula, and that is good, the asimmetry is gone
But we don't agree on the timeframe ... why not apply the math change for 1 day ... who knows when a downtrend starts (multiple day down)?
The flow is calculated day bay day, today i don't know if I'm in a downtrend or not ... maybe i know that 3 o 4 days in the future ... but today flow data are old at that moment.
The 3D flow is a "short term indicator", i can't wait a 20D overlay change to assign the right sign to the ACT refereneces.
Quoting dljonesFan
{quote} " Or Jones was wrong. And the dir of the day makes the math sign... But that would be such an obvious error - I cannot believe that he would made an error like that.

Me too ... but who knows .. things happens...
Other nick take the same validation test on same data that Jones used, with "classic" Jones method take a 88.37% validation, with the sign change the validation is 93.33% .... not a big difference but changing the sign gives better results..

https://web.archive.org/web/20110518..._example2.html
ok, last post for day .. I think ..

On the AMVA FLOW ..... These are the things that drives me crazy ...

https://web.archive.org/web/20110518..._sp060407.html
TPOs count: decreasing (428, 361, 376): Market activity decreasing, quieter

https://web.archive.org/web/20110518...p_sheetmp.html
TPOs count: decreasing (200, 229, 159): market activity decreasing, quieter

https://web.archive.org/web/20110518...analytics.html
TPOs count: increasing (246, 404, 307): Market activity increasing

https://web.archive.org/web/20110518...planation.html
# TPOs - Total 209 216 196 ==> Q

So in a more compact version from the top to the bottom we have:
* (D2 = 2 days ago, D1= 1 day ago, D0 = last trading day(EOD))

D2 > D1 < D0 gives decreasing
D2 < D1 > D0 gives decreasing
D2 < D1 > D0 gives increasing
D2 < D1 > D0 gives decreasing

Someone is able to explain the logic here? A logic that a computer can use ...
Thank you
Tail, it looks like cisco is taking the average of the last three days and comparing it to the oldest day to give an estimate of increasing or decreasing. I don't know if that is prudent or not, but that is what it looks like to me...
Quoting nonlinear
Tail, it looks like cisco is taking the average of the last three days and comparing it to the oldest day to give an estimate of increasing or decreasing. I don't know if that is prudent or not, but that is what it looks like to me...
WOW! great.... thanks nonlinear for the input, it seems is working on that example, I have averaged anything but not these ahahaha
Going to check the logic with other references too
Quoting thetail
{quote} Me too ... but who knows .. things happens... Other nick take the same validation test on same data that Jones used, with "classic" Jones method take a 88.37% validation, with the sign change the validation is 93.33% .... not a big difference but changing the sign gives better results.. https://web.archive.org/web/20110518..._example2.html
Hey tail,

who is "other nick"?

greets
C
0 traders viewing now