Auction Market Value Theory & Analytics

Appreciate the response. My vernacular is simply a synonym for "profiles", I as well do not trade them, but use the sampling of the "data" to make my trading decisions. Happy trading!

So where's the data? What data did you use to make an informed trading decision??? What was the basis for the trade? I don't "visualize" the data, I actually use the data. I don't see how we share the same "vision" of data. Where's the data used for this informed trading decision?

Thank you for your reply. Somehow the PM feature is disabled for my profile because I am a "Junior Member". Anyway, I really appreciate you taking the time to answer even though you are quite busy at the moment. I just wanted to let you know that I am not in a hurry here. Whenever you can find the time to answer is more than fine with me.

Have a nice weekend everybody.

MZ, I will apologize for somehow irking you with my post. It appears that what I have posted upsets you to some degree. I became intrigued by your thread because it is not often that I come across another trader who "seemingly" trades a similar style to my own. You are correct, perhaps we don't trade the same exact "vision" and part of my analysis I will disclose to no one. However, where we differ is that you will use a minimum of 3 days of data, where I will incorporate the most recent "value areas", into my analysis in conjunction with a variety of longer term composites, which I simply call multiple profiles. I have found these most recent, up to the minute value area formations also contain a vast amount of information about future events when the details are dissected correctly. You find no value in one single days frequency distribution, but I do. You choose to rely strictly on a series of mathematical and statistical calculations in your decisions, where I rely more on my own observations of the data, and how they are building over time to arrive at a more subjective view of market structure and participation, along with some very primitive math based models. In any event, however the final decision is reached to pull the trigger, we both seem to be successful at it, albeit from a slightly different angle. The data we are looking at for any market is more or less the same, however, unfortunately I cannot disclose my methods. As there is truly nothing for me to contribute to this thread, I will refrain from posting again, although I truly enjoyed reading your posts, and look forward to any additions you care to make. Happy trading.

I'm very interested. Not smart enough to comprehend, but very interested nonetheless.

Hi mzvega, your database and report is looking very neat indeed, you really set yourself out to learn programming just so to weed out any dependence from any platform; and gain more control and Independence. Are your data extracted from any source using API?

I did some trade entries last week; havent been trading for last 2 months though

EURUSD


GBBUSD


Both trades entry was last Friday and held on till now. Visually structure was rather similar but EUR metadata does gives a clearer indication of upward trajectory, but GBP nonetheless indicate a shift in direction to the upside and most importantly increase activity.

Yes, conflating AMVT with MP upsets me. When it happens, it usually means someone is bullshitting me about reading the thread. Your original post you used the word MP. I saw it before I could respond and you went back and removed the word MP. However, you're still implying we use the same MP methodology & concepts, somewhat differently.

Those who have read the thread or VBPT should know that the test for "serial correlation" started as a test using profiles to test the validity of MP methodology & concepts………

Students of Market Logic School and the Dalton group demonstrated a willingness to do the requisite homework and study to understand profile theory. Many students continued to have difficulty in applying it to the market. Our company, CISCO, had the impetus to encourage and help traders because we were a source of the CBOT data. We interviewed many traders who stopped buying our data because they could not make profile theory work satisfactorily. They reported two main problems: within the day, they were unable to project the day's action from the developing profile to its final form; and they were unsuccessful in predicting the next day's action from the previous day's profiles.
Pasted from <https://www.forexfactory.com/showthr...ed#post9132564>

The Watershed: No Day-to day Serial Correlation
At CISCO, we felt that a database with 10 or 20 times the information in the daily summary had to be useful for market analysis. We also believed(past tense) in the utility of the Market Profile approach. Consequently, we developed an algorithm to calculate day types. Profile theory argued for the significance of day types in predicting continuation. We felt that, once we could generate day types under computer control, we could develop enough information to find statistically the conditions under which the day type(s) could predict continuation.

Our study was negative. Surprisingly, we found little or no correlation between day type and continuation.This Point was the watershed in our research. It forced us to look outside Market Profile theory in our attempt to understand markets of longer time frame. It has long been known in futures research that markets have no day-to-day serial correlation (Labys and Granger 1970). That is to say, even in trending markets, the probability that the next day will be either up or down is about fifty-fifty. On a day basis, the market's normal fluctuation, the daily trading range, is much larger than any underlying trend, and hence hides it.
VBPT p. xix- xxii
Pasted from <https://www.forexfactory.com/showthr...ed#post9132564>

Believing that you are able to project the day's action from the developing (shape) profile to its final form. Follows that same underlying belief in "serial correlation", when evidence proves no such evidence exists There is no serial correlation between the developing/final form profiles and predicting continuation. Since AMVT is evidence based (an objective, scientific use of the data)., you need more than just a single test, is it just MP methodology or happenstance? You have to prove it applies to all market info. You see it in the source material it was tested on mp profiles first, then using futures data (VBPT), and I was able to demonstrate using candlesticks, because no such serial correlation exists.

Quoting nealpek

Disliked

where I will incorporate the most recent "value areas", into my analysis in conjunction with a variety of longer term composites, which I simply call multiple profiles. I have found these most recent, up to the minute value area formations also contain a vast amount of information about future events when the details are dissected correctly. You find no value in one single days frequency distribution, but I do.

Ignored

Again AMVT there is no concept of Value Area. A gaussian distribution market descriptor is declared (value area is +/- 1 standard deviation) is a MP concept. Those who have read the thread should remember reading…….

"Remember, the basic premise of market profile, the foundation, is that the bell shaped price-volume curve forms a Gaussian distribution and value can be found (defined) as the first standard deviation of that price-volume distribution. If the price*time-volume curve is not bell shaped, obviously it is incorrect to apply gaussian statistical analyses. ……….

Comment:
The basic premise of market profile, it's foundation, is that the bell shaped price*time-volume curve from the LDB volume and the price-activity curve from TPO's forms Gaussian distributions. A Gaussian distribution has a well defined form, the bell shape. Two parameters (mean and standard deviation) define all points on the curve. Experimentally, for a Gaussian distribution, both the mean and standard deviation can be found from the data. The identical formula can be applied to non-gaussian data, but, of course, the interpretations are meaningless. One problem with market profile is that there is little, if any discussion about data validity; does the distribution at hand fit the bell shape model and what to do if the bell shape is not present that day or the day shows both bell parts and non-bell shaped areas. Tests of the quality of fit to a gaussian curve are available in the statistics literature (e.g. chi-square) but not mentioned in CBOT 1985, CBOT 1991 or MOM. A distribution may have a bell shape and still not be gaussian (i.e. frequency dist) That would mean that at the very least, value obtained is questionable.

If the price*time-volume curve is not bell shaped at all, it is obviously incorrect to apply gaussian statistical analyses. Since bell shaped curves occur in balanced markets (and probably only in balances), it is important to have tools to identify balance. None are offered in the Market Profile Manuals or MOM. "
Pasted from <https://www.forexfactory.com/showthr...an#post9505530>

"The common thread running through all our studies is validity of the data. If trading is to be a science, the data used in the analyses must be valid. The original Market Profile analyses, and still today, treats any calculated value area as a valid measure of value; with further analyses proceeding from that assumption. ……..Consequently, relying on simple reference points (point of control, value area) can give a totally misleading read of a market. If your data is not valid, how can you possibly succeed by following it?..............................."
Pasted from <https://www.forexfactory.com/showthr...an#post8682223>

The values resulting from gausian statistical analysis applied to non-gaussian data provides invalid data. How often are profiles "normally distributed", all you have to do is look at them, you can see, that rarely if ever is true.

"It forced us to look outside Market Profile theory" that means that the basic principles of MP methodology are base on a methodology not supported by evidence......

So when you come to this thread and post MP jargon, obviously without not reading it. You falsely give the impression which leads people to believe that there really is no need to read the thread or source material you. That falsely promulgates the idea that AMVT uses the same MP methodology applied differently to multiple profiles……

Anyway, it irks me when people conflate AMVT with MP, quite the opposite, AMVT is built on the foundation that both TA & MP methodology & concepts, are methods based on the premise of a faulty assumption... This year I have no intention of posting & reposting & reposting, the same stuff others have no interest in actually reading themselves. It really is pointless………..

Hey, Stranger

I ditched the database…..since you cant do the analysis from "within" a relational database. You have to import the data, then pull it out to work with it, then return it back to a database. It's more effective to read & write directly into excel using python.
You can use an folder of files as a database using excel……….

The only reason I continued using a database, was I could display multiple spreadsheets/tables in one place. Rather than open files, take screenshots , of several files and post them, it was easer to put the data in a database so I can show several files at once. Since I now have a single clean report, the need to store files in a database to show others my analysis is no longer needed.

I always have a "working" version using the export data we currently have. However, this summer I got to familiarize with Oanda's API & a couple of 9wk programming courses

The version am working on uses Oanda API, calling the data is pretty simple,……I am stuck at programically determining "bracket no bracket". Until I can learn how to do that I have a user input box in which I answer the question…….I don't like it and I won't use it unless I have to. But, none the less it works until I acquire better programming skills……..

Beautiful trades!

Hi Mzvega,
you have no clue, how much what you share is appreciated by some in the know. You are one of the most valuable contributors on FF.
Thank you for that and keep posting

TRD

I have only one comment if I may from your post. The way you choose to display, crunch, reconfigure or sort any data set is infinite. The parameters or the way you've chosen to group your data to come up with buy or sell indicators, doesn't mean your groupings or crunching has exhausted all ways of looking at and using data. Conventional MP, frequency distributions, or any other form of data grouping basically contains the same information when looking at the same market. It is what you do with it, that determines an edge. You stated above that your statistical analysis shows one cannot determine direction or trade profitably out of a "cluster" of data from one day's sampling of data. I say you can! In fact I don't need a full day to complete in order to trade out of the "cluster" successfully. What I'm saying is that there are many ways to analyze data, by way of consecutive letters in classic MP, a series of only X's printing anytime price is used, or even looking at fixed period candles or bars. All of these are just different ways of viewing the same data. I agree that fixed period candles are not serially correlated(50/50) on their own, but a smaller grouping of candles, prints, x's, letters, or however you choose to view them are related. As the data continues to flow throughout a day or out of a cluster of used price levels, relationships of a different nature continue to build uncovering anomalies that can be exploited for profit. Any way you choose to carve up the data to perform analysis, doesn't change the sequence of the data or there relationship to each other.

Yes, I agree all those TA "methods" for analysis of a time series (OHLC data) are just different ways of viewing the same OHLC data. OHLC data is only used by people with an underlying belief in serial correlation, they believe they can infer actual meaning from the sequence/series of prices. That is a belief system, not supported by the market. Fooled by the data they use for analysis…….

All day trader's have a universal need to locate markets that can be read and understood. Many, many tools and training are available: technical analysis, candlestick charts, Gann, Fibonacci, pivot points, point-figure, etc. All are similar in that they apply some outside formulation/reasoning not supported by the market.

You fail to realize, this thread is about AMVT. The use of time series data (OHLC) is a non-permitted practice, not supported by AMVT because OHLC data assumes linear/serial correlation. And is misrepresentitive in it's display. AMVT only permits the use of "Market Generated Info"……….

"Market Generated Data (MGD) is the market’s transactional data organized so that price and price-over-time are displayed. The market variables are thus price and time. …...
Transactional data comes in the form of ‘ticks’. A tick is created at each price change. Trading at a given price will not generate a tick. A tick carries two pieces of information, price and time when it occurred. Throughout the trading day, there is a steady stream of ticks. The simplest way to present tick action is in a bar chart with equidistant times or in a point figure chart. Neither are ‘market generated data’ because MGD must display price and time".- Donald L. Jones CISCO-Futures
Pasted from <https://www.forexfactory.com/showthr...d#post10528862>

The market is reported as ticks, Price and the time in which it occurred (2D non-linear)
What happens next is the data is "subjectively" broken into TF's ( 1m, 5m, 15m, and 30m.etc.) depending on the viewer. This artificially skews the data. Each place where the data has been "subjectively" broken into TF's creates a artificial high , low, open, close. This is "subjective" info because it is biased by the observer's choice of TF. Already you have "flaws" in your data set. You are now using data "once removed" from its original market generated Price *Time format.

"Tick data does not have an open, high, low, or close quote it simply tells the previous price. The OHLC quotes occur only after tick data has been collected and coerced into interval data, such as one minute, one hour, one day, or any other selected duration"- the forex charting companion -Archer"
Pasted from <https://www.forexfactory.com/showthr...6#post10357556>

Now that the data has subjectively been broken into TF's. The data inside the timeframe(candlestick) is now "sorted" (rearranged) into a linear "series" of prices from high to low forming/constructing a bar/candle. Now you have skewed your data set once again. Remember the tick data didn't start out as a linear series. Now your data is "twice removed" from it's original context (market generated price *time format) Your candlestick TF's only look like a "series" of prices from high to low, because you manipulated the data to do so. This is not what the industry considers to be "Market Generated info"

"The difference between the current, "illogical" approach and the "logical" approach is the type of information used and how it is analyzed (e. g. , outside information related to the market versus market-generated information related to it).

In the current state, almost all transactional information, information derived from newspapers, retrieval services, and even live quotes disseminated by vendors, falls into the classification of "information outside the market". (Most people are surprised that live quotes are not included in the other category, market-generated information. But upon examination, live market quotes are not formulated into a statistically measurable and logically sound arrangement of data which provides information. ) In other words, transactional information is either presented in tabular, numerical form or else reorganized and presented in graphic chart form". -J.Peter.Steidlmayer Markets and Market logic
Pasted from <https://www.forexfactory.com/showthr...d#post10525163>

Attached Image (click to enlarge)

Minutes before the 30m close, prices are being used below the open closer to the lows………look at the data…..price feed/tick data is not a series of prices that move sequentially, from high to low, 1 pip at a time. Look at your own tick data, price does not come in/move in a series of sequential, consecutive prices (1.01, 1.02, 1.03, 1.04, 1.05 or 1.05, 1.04, 1.03, 1.02) from high to low or low to high. Price does not move pip by pip. Different prices are used at different times. They are not used one by one in a series, one price at a time, pip by pip in a series of h-l/l-h. No such series/sequence exists in how the data is reported.

Again, look at the tick data, each price and the time in which occurred is independent and not serially correlated/related to the previous tick/price. No such linear "series" or "relationship" exists.

If what you believe is true, If price does move in a series from H-L or L-H, minutes before the candle close, the tick data should be printing prices at the highs. But its NOT, The tick data shows prices being used/traded on the lows. OHLC data (candlesticks) are not accurate and misrepresentative of current market activity.

Look at the example the "candlestick" representation of the same data tick data. A candle closing above it's open makes it a green candlestick, A candle closing on the high. For those who believe in serial correlation, and believe price moves pip by pip one price at time in a series would think price is moving & trading higher(bullish). When the actual data being reported at the time of the close is trading at the lows. The candlesticks/ohlc data is a gross misrepresentation of the actual prices being used. . You are right, you can analyze OHLC an infinite amount of ways…... no mater how you slice & dice it, garbage in = garbage out.

What sequence/series? What relationship? The artificial H-L series used to coerce tick data into a computer ready format( OHLC) for charting? You're fooled by your own data……..
NO, your choice of data, and how you choose to view that data leads you to "Believe" prices are serially correlated, you believe you can analyze and infer something meaningful from a series or sequence of prices. You see serial & linear relationships where none exist…..
Verbatim……... I said what? Use my exact words…….because I don’t talk in "ambiguities" like the word "cluster". A "cluster" F#$% of misrepresentative grouping candlesticks? wtf is a "cluster" of data? Patterns/letters on the chart? Do you not realize this thread is about quantitative analysis? The data used here for analysis doesn't live on the chart, you can't see or measure "non linear" relationships using a linear display (time series). AMVT data isn't derived from a series of calculations using "OHLC" data. The use of timeseries data (OHLC/candlesticks, series of letters) is a non-permitted practice. What ever it is you say you can do,
can you do it somewhere else……..

Your talking about data & methodologies unrelated to the thread. We're talking about 2 completely different animals…….

If you want to discuss items related to the topic of the thread, fine but don't waste my time talking in ambiguous & airy fairy market terms rather than facts. Speaking in ambiguous terms rather than facts is a B.S. way to ensure your argument can't be proved or disproved…...

But when you analyze the flow in the market condition report you are using OHLC references and MP concepts ... so the use of OHLC data is a permitted practice.
Or not?

MZ, once again the failure to bring my point across is my fault by way of the language or terms selected. The raw data used in your analysis looks very much like common "time and sales" data. I hope this is correct(sweating). What I am saying is although the common candle or any fixed period analysis uses an arbitrary start and finish, over a period of time of data collection, at some point, one can say for "x" periods of data collection, the highest price recorded in the time price data i am observing is "y" and the lowest price recorded so far in the time price series I am observing is "z". Forget about clusters. Although common ways of looking at data, like candles, are manipulated, there are still peaks and valleys that manifest themselves as highs or lows on a pretty candle chart, but can I take the leap and say if a high over the last 3 hours is observed on a conventional chart of 111.00 on U/J, for example, that somewhere in your raw data, a rotation has occurred pushing up to 111.00 forming an interim high in your continuously flowing and non periodic data? Is this a correct statement? If it is, then in a large sampling of your continuous data, there will be many subsets of shorter term peaks and valleys within your data. Some of those subsets of "highs and lows" within your data are more important or more significant then others. Again a subset high in your data of 111.00 might also appear on a fixed period representation of the data depending on which period is used. I have found although I cannot describe it as eloquently as I would like, that without the use of any statistical or mathematical modeling, a select number of highs and lows within your data, are more significant than others, and that these more important highs or lows which can also be observed on a fixed period data set are related to each other. I believe discovering which are related is the key!! Whether this is acceptable in amvt is left to you. But regardless of whether I use fixed period data representation, or a continuous flow of raw data like you do, I can still extract the same info required to make my trading decision although the trigger is simply displayed in a different format. Again, the data doesn't change, but how you utilize it is what matters. All of my decisions are still based on probabilities, but the probability doesn't change in any way based on the format I choose to view the data. Some representations are simply easier to view than others....I understand all of your determinants are computer generated, and mine are subjectively visible. Perhaps one day I will ask someone to code what I "see", as I lack the programming skills required to do so. No disrespect of any kind intended, but amvt is but one way of viewing the same data flowing across all of our screens. If it works for you and you are satisfied with the results, you are way ahead of most, and I tip my hat to you for having the fortitude and perseverance in discovering an edge that suits your character and personality. Unfortunately I cannot prove any of this, as simply showing you one example will give away not "THE" edge to trading, but "MY" own.
I will stop here and hope you can forgive the intrusion into your thread. Please disregard any prior posts. Peace!

GbpChf
Long entry on the vertical line on the BO of 20,15,10,5,3 brackets
In my opinion 1.3390 is a dangerous possible false BO target level.
If the entry is good possible target at 1.3421 and 1.3490.
Stop below the mid of the 5D bracket.

False BO level has shown little resistance, good
Reached first target, out with half position at 1.3433.
Go with the remaining half to last target at 1.3490 or stop profit at 1.3387.

EOD, time to close the last half position at 1.3454.

No I don’t use time & sales data
If your talking about the picture in my previous post, That data has nothing to do with my analysis………. AMVT doesn't use raw data for analysis …….. That image was used to illustrate a point , why AMVT does not allow the use of time series data, because OHLC data is invalid as it is misrepresentive in it's display

Despite evidence to the contrary, you still have this underlying belief the market is linear & serial correlated. You can convince yourself that although the data you use for analysis is "misrepresentive" in it's display, it makes these pretty little shapes of valid info.

Garbage in = Garbage out. Invalid data doesn't magically produce valid results
Charting is not synonymous with analysis

OK, and what exactly does that tell you about the condition of the market right now? Does it mean the market is trending? Does it tell you how well the market is facilitating trade? Does it tell you were you are within the market unit? Maybe if there is an imbalance of prices being used?........Other than just 2 prices and their relative (absolute), locations (positions) on the chart, what exactly does that info tell you about the market currently?

No, not correct, I don’t analyze raw market data, or time series data (tick by tick)……….

You missed the point of my previous post, it's the very point/reason why AMVT does not use timeseries data(OHLC)/or tick by tick data…….No AMVT does not use "the same OHLC data flowing across all of (y)our screens", that's another assumption caused to your inability read a thread before posting to it…….

……..why don’t you actually read the thread, then rephrase what ever argument it is you're trying to make. Your trying to justify your use of data "misrepresentive" in it's display, by implying we analyze the same data………….

Many Day Traders do Technical Research.
There are at least 100,000 copies of database management/technical research programs in use. The users have databases, some striking charting abilities and the capability for all sorts of optimization on moving averages, oscillators, ticks, etc. It is a good business for data providers, but, some 90+% of public traders lose. Why? Some claim they are victimized by the brokerage industry, others have bad luck, yet others practice poor money management, etc., etc. No doubt there is some truth to these claims. But the biggest problem is that the wannabe traders are working from a faulty premise. The premise is that technical analysis (TA) can tell one where to trade, i.e. solve the trading equation. The fatal flaw is in the data. TA uses open, high, low, close, volume and open interest. The prices (o, h, l, c) fit into the TA equations, but oddly, have little relevance to solving "the trading equation". The market responds to change in value, not change in price. Price is an inexact marker of value. A major problem with TA is that two of the four daily prices (high and low) are quickly rejected by the market and hence are not representative of the day's value. With at least half of the price inputs questionable, is it any mystery that calculations from these data are also of questionable validity? This point is explored in some detail ……………...

AMVT uses value as a basis for market analysis.
Value, not prices, are the proper research medium. Usually, the high and low prices are rejected by the market while the open and close may or may not carry valid information. Another technical quantity, the volume, has problems. A day's volume is a combination of trade, spreads and out-trades. The spreads and out-trades vary from day to day sometimes as much as 30 percent of the total. Research on irrelevant numbers cannot be expected to provide valid information. Putting faith in marginal calculations is an important part of the reason that 90+% of the public loses.
https://www.forexfactory.com/showthr...67#post7487367 Post #6

While there is no theory competing with AMT for market analysis, a great deal of effort continues to go into Technical Analysis (TA). TA is price based market research focused on 'indicators'. There are hundreds of books on TA techniques and well over 100,000 traders have databases and TA analysis programs. How does TA differ from AMT? The difference is simple but profound. TA uses price based analysis (all prices treated equally), AMT is based on value (all prices are not equally valued). If TA were an effective methodology for the trader, it would be easy to say that TA and AMT were just two different ways of looking at the market. However, most traders do use TA and it is estimated that over 90 percent of futures traders lose (recent surveys of equities day traders appear to show the same result). At the least, AMT has a sound theoretical foundation, whereas TA does not. With AMT, one can look inside the market to divine what is happening and why, with TA, one cannot.
The difference between TA and AMT can be illustrated easily……… page #4 of thread

In any field of endeavor, preparation shows up in the results. Nowhere is this more true than in futures trading. The average public speculator has little knowledge of the structure of the futures market and the regulations that govern. It shows: the odds against the outside trader are large; reportedly some eighty five percent of outsiders fail. In our experience, the failures are characterized by a lack of understanding of how the market functions. Trading is a profession. It requires preparation and practice. Ignorance of the market environment usually leads to abuse of leverage (overtrading). Most (public) traders want to trade now and learn later, implicitly relying on luck to stay out of trouble.

In the long run, it is not luck that makes winners. Winners understand the markets and how to find and use an edge.

Not only does most of the public not really understand how the markets work, but their whole approach is wrong. They seek out a methodology for market prediction and inevitably are drawn into 'Technical Analysis'. They find themselves developing databases, buying technical analysis software and losing, losing, losing. Something is wrong when so many intelligent people find it impossible to master trading. The root of the problem very likely lies in the methodology, Technical Analysis. The input data (ticks, and daily prices) are not amenable to tracking value and value changes. For that one needs Auction Market Data.

Auction Market Analysis differs greatly from Technical Analysis. The difference is simple but profound: Technical Analysis deals with price; Auction Market Analysis works with value. If you know value and it's history you can know in depth what the market is doing (market condition). If you know only price and it's history you can see where the market has been, but you know little about it's current condition. Value is in-depth knowledge, price is superficial.

Oftentimes Technical Analysis uses exactly the wrong information: in the popular candlesticks display the body is made of the open and close while spikes on the top and bottom are the high and low. Highs and lows are used for decision making. But we know from the data that the highs and lows are the excess points that rarely have much to do with value! The same argument can be made against Pivot Points and most other technical analysis methodologies. They use price, not value, to analyze markets. There are well over 100,000 traders who maintain price databases capable of optimizing trading models based on price. There are a lot of people competing against each other using a shaky premise for trading decisions. Most lose!

The basic, underlying theme of Auction Market Analysis is that if you understand your market and the roles of the participants, you will have the information upon which to base rational, low-risk trading decisions. You can know what a market is doing by monitoring the 'market generated' data it develops. Market generated data combines price, time and participation How well you unwind this information from the data determines how much you can learn about the market. AMVT's primary function is to teach you how to 'read' the market generated information.......... page #4 of thread

Auction Market Value Theory (2005). Auction markets are complex with many variables (reference points) and there is feedback between reference points. This complicates the development of trading models. Overly simplified methodologies, like indicators, are doomed. The holistic or pattern recognition approach is deviled by the almost infinite variations of patterns one has to learn and recognize (Mastery Theory, again). Having a theoretical formulation, even for a complex market, simplifies the problem: Some market friendly solutions become available. Possibly more importantly, theory helps in avoiding proposed solutions that are untenable. For example, much of technical analysis relies on moving averages. The run-pause nature of markets clearly shows that one run-pause case does not have the same elapsed time as the next case, or the next, or the next. But a moving average assumes an underlying cycle. Observation of market-units (the run-pause times), show no underlying periodicity. Smoothing via a moving average introduces errors (fitting the data to a non-existent, governing time period).

Coercing, non-linear data into a artificial time series, produces the same type of errors in analysis......................

since you dont use OHLC data for analysis , you don't use it for buy/sell either so that means you use data from report:
I assume that your sell price was approximately at 0.8812. which parameter(s) from report give you that price to sell? thank you.