Auction Market Value Theory & Analytics

This chapter is about data and the utility of data for futures research. Before we look at the actual data, we must identify what we need; how the data will be used and how well conditioned the data are for our purpose. Data that are not right for a job will lead to costly errors. Our data should help us determine the market's structure: day structure from the Market Profile distribution, longer-term structure from the Overlay Demand Curve distribution.

Structure is a visual. It shows whether the market is well behaved, with a smooth distribution, or jagged and irregular. Half-hourly data have proven to be ideal for helping to determine market structure.The market's condition is also found from the distribution of the profile. Condition is what the market is doing, the state it is in: bracketing, trending, or too volatile to tell. Markets forming a single bell-shaped distribution are usually bracketing, i.e., in a trading range. Trending markets have their own distinctive form, combining several separate distributions.

Reference Points
A market's structure and condition determine its amenability to analysis. A great deal of market information and understanding can be obtained from bracketing markets. Value and the two bracket limits are three of the most important reference points. Others, of sometimes crucial importance, come from analysis of commercial behavior on the Liquidity Data Bank reports, from daily profile analysis, and from the measurement of internal trends in a bracket; there are several beyond even these. The better the reference points are defined, the better the market understanding and the better our risk control.

There are two principal types of value: "day value" Pertains only to the current trading day; value without the "day" modifier refers to longer- term, more general value. Day value is the value area in the Liquidity Data Bank, where value area actually has meaning only in balanced markets. Longer-term value is a similarly defined price (range) in a bracketing market and is harder to pin down in moving markets where value may lead or lag price. In either case, value is always a (limited) range of prices at any given time regardless of a market's condition. when we set out to calculate value we quickly come to the realization that value is intimately tied to sampling rates.

Value
Value is the key factor in responsive trade selection for the day trader. In a stable, bracketing market, market-offered opportunity occurs when price moves away from value. Oddly enough, value is (implicitly) used by the position trader as well. The position trader, looking for breakouts, is betting that value is changing and that price will not return to the bracket. Position traders who are trading an ongoing trend are betting that price has not reached value and, hence, that the trend will continue. Value is (1) A price (range) recognized as a fair estimate of monetary worth, arrived at via price reinforcement (sampling) over time. (2) The most frequently occurring price (region) within a period of price stability.

For example, tomatoes have cost $1.70/lb. the last six weeks. Price over time implies value, so tomatoes are worth $1.70/lb. Today at the market, tomatoes are $0.90 Per Pound. A bargain? Maybe. Price has certainly dropped, but value may have also dropped. we cannot tell from this one sample. Our experience, our sampling time, is too short.

The average time frame sets the limits on the required sampling rate to detect change. A critical question is the time rate of change of the phenomenon being sampled. That is, how fast it is changing. In order to measure the change, we must sample at a rate high enough to define the phenomenon. we must have a number of samples while the change is occurring. A market that moves substantially within a day or two will have to be sampled a number of times per day to define the change. A too frequent sampling rate (say, every minute) loads us down with meaningless data. If the sample time is too long (say just once per day) we are not able to define the change.

If we checked the tomato prices at the supermarket 20 times a day, we would gain no more information than by checking once a day. Our high sampling rate yields no additional information. Nor if we sampled tomato prices every three months would we gain much knowledge about the value of tomatoes, since we know from years of experience that the tomato value time frame is one to a few weeks.

Today's futures market time frame is now often in the range of 5 to 20 days. That is, prices change substantially within that length of time, on the average. Several years back the market's time frame was much longer, from one to several months. The change in average time frame has profound implications on technical market analysis.

Sampling Rates
The rise of technical analysis in the early 1970s was fueled by the newly emerging access to (mainframe) computers and the availability of daily summary data from the exchanges, published in newspapers. Eater, the invention of the personal computer and the rise of data services put analytic capabilities in the hands of many traders. The sampling rate, thus, was daily, and adequate for the times.

Sampling the futures market of today once per day, however, is usually inadequate to detect significant changes soon enough to do anything about them. A market with a time frame as short as five days cannot be defined by daily sampling; the sampling rate must be high enough to get inside" the market; five samples could not identify value. Many traders still use daily data because it is easy and cheap to acquire, and there is a great deal of software for processing this type of data.

At the other extreme, tick by tick sampling picks up mostly noise. The data are there, but each sample has a very small information content. The day trader who watches ticks is often misled by the random movement of the market. The increased sensitivity of more frequent samples is masked by the noise. . . the random fluctuations.

what is a reasonable sampling rate for futures? Daily is too seldom. Each tick is too frequent. How fast do the markets change? The market's rate of change determines how short or how long our measuring rate should be, but an actual, average rate of change is hard to measure.

Since we have the limits defined, a reasonable approach is to do "order of magnitude" estimates to zero in on the proper sampling rate. An order of magnitude is a factor of 10. If the whole day is too long, an order of magnitude less is one-tenth of a trading day, or about 30 minutes. If the tick-to-tick time is too short, we find an order of magnitude longer to be in the range of a few minutes. So, our best guess for a proper sampling rate is, without actual case studies, a few to 30 minutes.

Finding the ideal sampling rate that will gain the most knowledge about a market is still an open research problem. we know that we must define just what it is that we are looking for (not just value) and how to measure it. we want to be able to see significant features as they develop, rather than, as on daily bar charts, after they are over. we need a sampling rate that shows adequate detail without being swamped by too much data. And we can be sure that the best sampling rate for a market that is undergoing rapid change is not the best for one that is moving slowly. So, for most real-world situations, we seek the sampling rate that works well for a broad range of market conditions. That need pushes us in the direction of somewhat less rather than more sensitivity, indicating a sampling rate at the upper end of the time scale, around 15 to 30 minutes.

Fortunately, there is an extensive 30-minute database going back five or six years. It is incorporated in the Market Profile of the CISCO database and is available to CiSCO customers. This database is used to develop the Overlay Demand Curve as well as to identify value and the trading reference points associated with bracketing and trending markets.

VBPT p. 1-4

Value, the Key
Value is the key to market understanding. If the market is stable, i.e., bracketing, value is a definable quantity derived from price over time*. For bracketing markets where the sampling rate is adequate, value is the price (region) that wins the trading popularity contest.

Keep in mind that each bracket has a life that dates from the end of a trend or other non-bracket condition. Samples taken earlier will introduce error. A bracket's life can be one day (the day the trend ends), or many days. In such a case, value is determinable only since the end of trend; but within that period the bracket parameters are defined and trading risk can be estimated.

For markets where Price is not within a bracket, Price and value usually do not coincide. If a trend is underway, price can first lead value, then value catches up or even passes price, then price again leads, and so on until the trend is over. Here too, value defines the market condition, although value is much less well defined. If there is no bracket and no trend, either value is shifting often and unexpectedly or there is no consensus on value. Such situations often exist if there is a preponderance of outside, non-economic influences, as when central banks are intervening in the currency markets. Even in these cases, the absence of any sensible value tells much about the market condition.

Data Sources and Types
Futures data from actual trading are generated in two ways. Tick-by-tick information is sent from the trading floor as trades are made. Ticks are the raw data, since they include the reported floor activity (a tick is recorded for each price change), some of which may not be confirmed at clearing. At the end of the day, settlements are posted by the exchange: these, together with the open, high, low, and previous days volume and open interest, are reported as exchange official. These are the prices reported on the business pages of The Wall Street Journal and other newspapers.

Later in the day, the clearing process of trade matching, or reconciliation. turns up mismatches and other sorts of error. The final result of the clearing process is that every trade is matched on both the long and short side. These data are then reported to the public by the Chicago Board of Trade as the Liquidity Data Bank (LDB) and by the Chicago Mercantile Exchange as volume Data. Other exchanges do not publicly report reconciled trade information. LDB reports are extremely valuable to the trader. They contain actual trading volume at each price, and this volume is further broken down by the four types of trader involved (floor member, commercial, off-floor member, and all other). Incidentally, the actual cleared volume may be substantially different from the "exchange official". LDB data also contains a Market Profile showing the time (within half-hour periods) of the trade activity at each price. Profiles on the LDB are easily reduced to half-hour price bars.

Standard Data
Standard simply means multiple potential sources, guaranteeing a reliable supply. Most people only have access to standard data. If those data will not support the information needs of the job, their research project is doomed from the start.

The first step in technical research is collecting and understanding data. The proper data must be available and we must be able to use them. Availability can be a real problem; there are really only two standard types of data that are readily accessible.

The first type is the official daily exchange summary data: open, high, low, close, volume, and open interest. The day's open (sometimes split with a high and low), high, low, and settlement prices are released shortly after the close of trading. (The volume and open interest are those of the previous day.) These are the data that appear in The wall Street Journal and other newspapers; the data can be purchased from a number of vendors. They are also delivered on many quote machines.

The other standard is the tick-by-tick time series from a quote machine, which is also available from several vendors. Ticker data is a standard because of its availability in computer-ready format for processing.

Monitor Addiction
On-line quote devices are everywhere. Powerful analytical programs take the quotes and form them into all sorts of data structures one-minute charts, oscillators, moving averages, and many, many more. Most of the displays are in color and the resulting graphs are striking. It is easy to get addicted to such devices, but they do not make it any less difficult to select profitable trades. The feeling that watching every tick keeps one abreast of a market is, for most of us, an illusion. Our minds integrate these mostly random price movements into structures we believe at the time to be meaningful. In the discussion below on sampling rates, we conclude that there is a minimum average time, on the order of minutes, that it takes a market to generate stable information.

Nonstandard Data
Non-standard data are so called because of their lack of general availability for history, for daily updates, or for both. The types we cover will likely become the standards of the future because they carry more and / or better information about the markets than the current two standards.

Market Profile
Market Profile is a way to organize the data over time. It was defined in Steidimayer and Koy (1986) as price activity recorded in relation to time in a bell-shaped curve. The graphic was developed at the Chicago Board of Trade in 1984. Market Profile uses half-hour data bars as primary input, keeping track of the time at each price tick.

An example of the raw data is the 30-minute grouping in Figure 1-1. Each price traded in each half-hour Period is counted as an event. There can be only one event at a given price per half-hour. One trade or many at that price within that half-hour is still a single event. Events are called TPOs in Market Profile analysis nomenclature, where TPO stands for Time Price Opportunity (or That Price Occurred'). A graphic is formed by collecting the events (TPOs) and plotting on a vertical price axis (see Figure 1-1 in the Introduction).

Balanced trading days show very little trading near the highest and lowest prices, quite a lot in the middle hence trading is distributed into a bell-shaped curve (see Figure 1-2, left side). The bell reflects a short-term (day) balance in the market. As the bell forms, the trader locates value (for that day). Then, day trades become attractive when price moves away from value. A number of quote-machine monitor programs can generate profiles, but history and daily update services are scarce.

Yet a third type of nonstandard data base is embedded in the Market Profile. The half-hour data bars used for the TPO distributions can as well be used to create a high-low database. The period's opening and closing prices are not carried. But, for many purposes, the half-hourly range alone is adequate. Figure 1-1 shows half-hour bars using the TPO letters, Figure 1-2 is the same day, same contract, posted in ticks.

Five-Minute Data
Between tick-by-tick and daily high-low, the profile data offer a middle ground. But we may need a finer subdivision of time. At CISCO we began collecting five-minute bar chart data in October 1991. Again, anyone with a quote machine could generate such data, but it is not generally available commercially.

Liquidity Data Bank
LDB reports come in three parts, the Volume Summary, the TPO Analysis, and the Futures Summary.The Volume Summary includes a Market Profile and the actual volume of trading at each price*, broken down by the volume percentage of each of(1) floor members (Cti1), (2) commercial clearing members (Cti2), (3) members filling orders for other members (Cti3), and (4) members filling orders for all others (public and other types of traders) (Cti4). Also included are the 'value area, defined as the price range that encompasses the central 70 percent of the trading volume and the total volume with Cti breakdowns. The TPO analysis table counts TPOs and locates them with reference to the center of value, calculates the TPO value area and finds the Trade Facilitation Factor (TPO's divided by the number of prices). The third part, the Futures summary lists the trading at the open, high, low, and close,the volume by type, and the related TPOs; with the same being done for the quadrants (the day's price range divided by four). An LDB report is shown in Figure 1-4. LDB data are covered in detail in Dalton, Jones, and Dalton (1990) and Jones (1991c), as well as in Chapter 2 of this volume.

The three parts of an LDB report.
1. Volume Summary (CBOT and CME). For each price traded

2. TPO Analysis. Calculated by CISCO on CBOT and CME.
TPOs are used to estimate the value area, to find the upper and lower activity counts, and to calculate the CISCO Trade Facilitation Factor (ratio of total TPOs to ticks [steps]). These data provide a quick measure of buying/selling pressure and whether the market is facilitating trade. CISCO has published research on the validity of TPO value areas, finding that they are an adequate substitute for volume- based ones. For non-CBOT profiles, TPO analysis is all we have.Of course, during the day, TPO analysis is very useful.

3. Futures Summary . Futures summary data is released only by the CBOT. It shows the trading at the open, close, high, and low and within the four price quadrants. These data help in the evaluation of the market's response at those various price levels and shows which market participants are active at each point.



VBPT Chapter1 p. 4-11

*The volume Summary used in mp analysis includes a market profile graphic built from tpo's, which includes a report of actual volume traded at price. Two separate things. The MP graphic is not a graphic of volume at price.

*"value is a definable quantity derived from price over time"
That means that horizontal trend lines marked by candlesticks highs & lows does not measure or define "value" .
Value is a definable "quantity". That requires a "count" of something.

Market Understanding or Forecasting?

In an article 20 years ago, the author (Jones 1974a) said "what we are seeking is a method that tells us when to buy and when to sell with a reasonable probability of success", and later, "We feel that defining trends is a proper sphere of activity for the researcher" (Jones 1974b). The latter quote can be updated to: "defining market condition (both trending and bracketing) is the proper sphere of activity for the researcher or trader". Note the absence of "forecasting" in this commentary. We would all be delighted if we could know tomorrow's prices, so why is forecasting missing? Could it be that that is asking too much of our data?

Forecasting of events to come fascinates us. There is an element in human nature that makes us want to believe that the future can be divined. Apocalyptic writings of all sorts are avidly sought, whether Nostradamus or astrologers or economists or market prognosticators. All have one thing in common: Their predictions do not stand the test of time. The Wall Street Journal has a regular feature in which the experts market calls are compared to random selections ("a dart board"). We are not surprised that the dart board does as well as the experts. A report in the Babson Staff Letter, published by the David L. Babson & Co., investment managers, says in part, "we have frequently emphasized the folly of attempting to "predict" future moves in stock prices, except for the secular, long-term trend, which has been upward for over a century. Despite the clear lesson of history that stock market forecasting is nothing but a guessing game, thousands of investors, professional and amateur, still try to do it. The same is true of the bond market. . ." The Babson report carried a historical study of forecasts of leading economists, taken from the semi-annual Wall Street Journal survey and shown here as Table 1-3.


A recent article in The Wall Street Journal (January 21, 1993) reconfirms Table 1-1. The subtitle is "Going against Economists' views on Rates would Have Paid Off", and reports an updated 5 rights and 17 wrongs on economists semi-annual yields predictions. J. K. Galbraith was quoted as saying "we have two classes of forecasters: Those who don't know, and those who don't know they don't know". You are right on target if you have come to the conclusion that we do not expect our data to forecast the market.

Serial Correlation in Futures Markets
The lack of day-to-day serial correlation in auction markets is often the starting point for random market theory. Studies as early as Labys and Granger (1970) confirm the day-to-day randomness. Even in trending markets, the expectation of up price or down price for the next day is still about 50-50. Even odds in non-trending markets is easy to understand; in trending ones it is not so obvious. A market that is trending up or down does have an enhanced probability in the trend direction; it is just pretty well masked by the random component that is always present. For example, a trend of one cent per day (on average) is virtually lost in the average daily trading range of, say, ten cents. By grouping, however, the random component can be averaged out and the underlying price movement can be seen. We will illustrate this point with a trend in soybeans.

Selecting a Trend Period
The periodicity in a run of data is determined by locating peaks and valleys and dividing the length of time by the number of peaks. The number of peaks in a particular number of days depends in part on the number of days permitted between peaks a minimum of 7 days peak to-peak looks for short cycles (many peaks), while 20 days peak-to-peak gives a longer cycle (fewer Peaks). The 20-day periodicities in the November 1991 soybeans gives the results in Table 1-4.


Testing a Trend for Serial Correlation
There is one long downward trend from March 6, 1991, through July 10, 1991. This run covered 88 trading days (127 calendar days) or just over four months. The total price drop is 121 cents or 1.375 cents per trading day. The daily trading-range average over the 88 trading days is 8.17 cents, ranging between a high of 29.25 and a low of 3.25 cents. Table 1-5 has the prices and a list of higher/lower prices on the basis of 1 day, 2 days,. . . through 5 days.


Since a downtrend was Pre-selected, we have inserted a bias to the downside. Therefore ties those cases where the compared prices were the same-will be awarded to the H, or higher count. That results in Table 1-6.

Table 1-6 shows that the trend displays almost random choice between higher and lower prices on the" one-day" basis. The differences widen rapidly, however, with longer time frames. While there is no day-to-day (yesterday to today) serial correlation, even day to second day (day before yesterday to today) shows serial correlation in a trend.


Conclusion
We will demonstrate that the market's structure, its current condition, and its value can be developed from data sampled at the proper rate. The LDB offers a track of the commercial market capping (the setting of support and resistance points) and going with trends. That sort of information is now available during the trading day. Hourly LDBs allow public trend initiation to be observed as it happens. All these results ultimately become reference points for market trading management. The trader can have the information to select market positions intelligently and to minimize risk in doing so. Once that is done, winning becomes a question of probabilities. If you know what the market is doing, its current condition, and its current value, you have the basis for expecting certain behavior. Knowing just that much can give you a very large edge.

Having proper data is critical in understanding ones market.The sampling rate must be adequate to sustain the analysis. Much of the data used by traders today fails the test. New developments in data collection are the hope of the future. Today, many traders can generate distributions (profiles) in real time. As personal computer systems improve, both ticks and ancillary data such as the hourly LDBs can be received simultaneously The off-line position trader is gaining more and more access to nonstandard data, just the sort of information needed to be competitive in today's market.

VBPT Chapter1 p. 19-24

Data day of the trade

Data current

Still holding GU


NJ original entry


NJ added to position


EG

Hello mzvega

A few questions

Say you do your EOD analysis at NY close and its prime to buy

When do you actually place the trade?
How do you determine the exact entry price?
Do you make use of pending orders?

Let's NOT say "I do my EOD analysis at NY close and its "prime to buy" "

That would be misleading. Working off a faulty "assumption" that my EOD analysis produces some kind of "buy/sell" signals. There is never an occasion in which the data tells me the market is "primed to buy or sell". That falls in the realm of "prediction/forecasting", both non permitted practices.

Analysis always starts with "condition" first. As explained in VBPT in detail, my EOD analysis produces a list of bracketing pairs (as per Jones) in which I use for possible trade selection. Not every bracket is tradable. I then narrow down my list of potential trade opportunities by looking at the consistency of brackets. Determining where the bracket lies within the current "market unit", etc. It would be foolish to trade a break out when the market is testing for end of trend, or trading a responsive trade when the market is testing for new trend. This info isn't found on the chart, but in the data.

There is no exact entry price , I am not trading "prices" I'm trading changes in value. I am trading the "change" from one market condition to another. I consider it an "event". An event identifiable using objective quantifiable measures. An "event" isn't a predetermined location or price on the chart. An "event" doesn't exist until it "happens". I usually don't enter a trade until it's a objective quantifiable "fact". TPO analysis is the quantitative analysis of "price time events"

I've been asked the same question 99 different ways.
I wish I could tell you something different, but I can't. It's not any one "price" (or "candlestick pattern" for that matter) that defines or describes a change in value. It cant be measured, by measuring one price against another using price to price data (time series)

VBPT has a chapter called "trade selection"
"Trade selection in Value Based Power Trading begins with a survey of all the bracketing futures on the bracket screen. Potential trading candidates are sorted out by looking for futures with consistency, for those with several brackets, and for those with the requisite profit potential."

There are no buy/sell signals. Learn the requisite's for trade selection

I am not in the habit of using pending orders, but I have on rare occasion used them.

mzvega, I hope will not make anyone upset with my question: My understanding about trading is that price go up when there is no more sellers and goes down when there is no more buyers. Since value is function of price and time by changing price value will change over time... Somewhere was written: ...Price goes up to shut of buying and goes down to shut of selling...

Price change faster then value because it takes time to change value if price change...Price change when there is lack of either buyers or sellers..

Can you see those changes: lack of buyers and sellers in data you use for analysis?

Believe that my thinking doesn't deviate much of AMVTA. You were writing about supply and demand, day traders and other time traders in many posts but is too much for my old brain to pass all of it so I want to ask you if you can dedicate one post on that subject when you have time and willing to do so.

Thank you for your time.

Price is nothing more than an advertising mechanism............

Of course you can see lack of buyer or seller participation. Auction Market Value Analytics measures those changes over time (data).
http://www.forexfactory.com/showthre...21#post7460221
http://www.forexfactory.com/showthre...90#post7233590
http://www.forexfactory.com/showthre...09#post7233309
http://www.forexfactory.com/showthre...48#post9022648

Clear and concise.

Forgive my assumptions, I am still shedding the "popular" thought process that has been ingrained over the years.

I was asking because I wanted to deduct the money management part of it, mainly Risk & Stop Loss (if need be). You know popular risk calculations has more to do with "exact entry price & therefore timing".

Judging from the trades you've shared, I think you are a longer time frame market participant, coupled with the fact that your experience makes this quite an intuitive process it might be difficult to understand how a novice trader can't "see" these things.

If you have the time. It would be wonderful if you shed light on the Money Management side of things, I think MM is directly tied to the type of trading style.

How does one tie an adequate MM to AMVT? (As beginners we would make mistakes in interpreting the data, a way to mitigate risk and losses during this process is welcome.)

Also since AMVT considers changes in value and this occurs over periods usually longer than a day, does this restrict AMVT to longer-time frame trading?

PS: As usual please ignore if the answers are already stated clearly in any of the books. Thanks

Hi Sinqua,

"Also since AMVT considers changes in value and this occurs over periods usually longer than a day, does this restrict AMVT to longer-time frame trading?"

No, every trade starts as a day/intradaytrade. How long do you hold is your choice (what the data tells you).

Greetings
C

Ah I see, thanks dljonsFan

Hello,

Please check your pm.

Thanks!

When you read & learn what a "bracket" is you also learn that the bracket references are used to define risk. When you learn about the volatility reference (VTY) you learn that in addition it can also be used to help define risk. i.e. setting a stop 1 or 2 times the VTY number keeps your stop loss outside the normal "fluctuation envelope" (VBPT).

From your perspective, yes I strive to position my trades with the LTF. But not all trades end up that way. I trade what the market offers. But from an AMVT perspective I am a considered short time frame trader. "Time frame" is one of 2 things, the intraday(daily) or every "other" timeframe (not 1,5,15,30etc.). We are either talking about the "minor auction" or the "major auction"

I can still be an intraday trader, but I trade/position myself with OTF/LTF. The point of trading the change from one condition (bracketing) to trend (not bracketing) is getting into the trend early. You can hold on until the data tells you the trend has ended or exit end of day and position yourself to re-enter the following day, trading with the flow of the OTF/LTF. Not every trade is a multi day trade. Sometimes it's just a day trade. Sometimes my decision on how long I hold depends on "rollover". OTF/LTF are more inclined to hold positions when they are gaining interest on their trade. Fees & rates influence trading decisions. Whether I hold for a day or longer I still need to trade with the flow of traders that drive & dominate trends (LTF). Even if you are not a LTF participant, as a short time frame trader, there is value in knowing where they position themselves.

Understanding the auction market process you need a full understanding on how auction markets operate, knowledge of each kind of market participant, the different roles they play in the market, pricing & fee structures

The logic of trading along with the LTF may be better explained in my following post...

"The Trading Arena: Exchange Environment, Regulatory Agencies
Trading futures is much like any other career choice. First there is a basic, theoretical body of knowledge to master. A part of that is the applied learning, the nuts and bolts that one gets during an apprenticeship (doctors and lawyers practice and learn, so must futures traders). Futures traders study their models, but most fail to learn enough about the exchanges and the regulatory services.

The Exchange: A Private Organization
Members on the exchange trade for a living. Floor members either trade for themselves or fill paper for others (your brokerage firm, for instance); or they do both, trade for themselves and fill paper. Commercial members trade only for their institutions. A membership can cost a half-million dollars. There is money to be made on the floor. The unwary are liable to get plucked. Your job as an outsider, is to pay for services received only, to make sure you are treated properly within the rules and to generally protect yourself. The system tends to pit you against your broker--in the case of an error, who pays? Know the rules!

The Exchange: Path of a Trade
The longest path your order can take is: you =>broker =>trading desk =>floor =>pit =>floor =>trading desk => broker =>you.
An order passing through so many hands has an opportunity to get altered or lost, especially in trying times such as 'fast' markets.

The shortest path for your trade is:
you =>floor =>pit =>floor =>you This path is called going through the 'arb' desk; you often get your fill while still on the phone.

Regardless of how you enter your order, your trade is filled by a "floor member" who takes the opposite side. If your order eventually ends up offset by another public trader, that side too will have passed through the hands of a "floor trader". Floor traders do about sixty percent of a day's volume.

The Exchange: Filling Your Order
Orders are filled on the floor by open outcry, the auction process. Your order might not get filled at your price because there may not have been enough contracts offered at that price to fill all takers. Getting a worse price is called 'slip'. Normally, you will get a fill near your price because of exchange rules which govern the way floor traders do business with the public (members are held to these rules). However, your price may be hit and still you get an 'unable'. This can happen if you have an order for a price or better. Say it turns out that your price was the low of the day and you were buying. When the low was hit very few contracts were traded and yours was not one of them. Or it was a 'fast' market with violent price action. Then the brokers are 'not held'. Even then, there are rules your broker must follow. He must not fill someone ahead of you, if he has your order first.

The Exchange: The Initial Balance
On many occasions, most of the day's range will be covered during the first hour after the open. This comes in part because of the orders that came in after the close the previous day (resting orders). These are in the floor traders 'book'. The floor finds a way to trade as many of these resting orders as possible, by rotating price. It is said that the floor 'controls' the market during the initial balance. Obviously, you would like to avoid getting stopped out by a non-economic process. Whenever possible, you should place your orders after the open, and ideally, after the initial balance." - Jones

We (public/retail) are not floor traders. Our job as retail traders is not to take the other side of trade for the sake of providing liquidity to the market. The pricing & fee structure for "floor traders" allows them to profit from short time frame moves.

Steidlmayer & Dalton were both floor traders (members) Let's put this in the proper context..........

Quoting Jones........

"Advisory: The following material is primarily of use to day traders. Standard day trading using some sort of pattern recognition has generally not proved a successful way to trade. Typically, the trader is competing with the floor in picking profitable short time frame moves. The difficulty, in part, is: that the off-floor, non-member trader trading is delayed relative to the floor. That delay, plus the cost factor (members trade for $2, non-members for $20) creates large odds, a house advantage that is hard to beat.

Floor Member
Can trade in 2 seconds
Pays $2 commission
Can be profitable with $20 gain

Public Trader
Takes 2 minutes to trade
Pays $20 commission
Can only be profitable with $100 gain

It is the thesis of AMVA that the Public trader needs an edge, one that comes from superior knowledge. The Public day trader is helped by knowing what the commercials are doing, where balance exists, the market's volatility structure and other reference points from the Trader Control database. This is superior knowledge, but is it enough to overcome the huge odds in favor of the "floor members"?

Generally, no! We believe the Public day trader must be much more selective than his floor counterpart. The way we see to beat the odds is to not compete head to head. Trade the market structure, use market condition as a guide: trade breakouts, trade responsively. These are positioned trades that require the market to offer an opportunity. There are far fewer such trades than one can find through pattern recognition. But the odds are better. "- Jones

the retail trader is "competing" with the floor in picking profitable short time frame moves. The pricing & fee structure of being a "member" allows them to be profitable trading short intraday moves. If we were to trade the exact same trades as a floor trader who ended the day profitable, head to head, due to fees & pricing structure, we as non-members can end the day with a negative balance. Trying to compete head to head for the same short timeframe moves as floor traders is a losing strategy over time due to the retail fees & pricing structure. In order to beat the odds as a Public/Retail you cant employ the same methods of short timeframe trading used by floor traders with a house advantage

Different market participants, have different roles in the market. The floor traders job was to provide liquidity by taking the other side of your trade.

Four Types of Trading on the Floor: The Participants

Floor Members: These are the 'middlemen'. They do about sixty percent
of volume. They seek a well defined fair price, and hence are happiest
with balanced markets. Known as 'short timeframe traders', they are in
control during the initial balance, or approximately the first hour of
the trading day. Floor members end the day with no open trades. Floor
members are identified as Cti1 in the Liquidity Data Bank Report

Commercial Members: These are the 'businessmen' on the floor. They trade
for their institutions. They know value better than any other type of
trader. They are in control in balanced markets, where they buffer the
edges of the distribution. Commercials end the day either hedged or out
of the market. Commercial members are identified as Cti2 in the
Liquidity Data Bank Report

Off Floor Members: Members trading from screens outside the floor.
Otherwise very similar to Floor Members. Off floor members are identified
as Cti3 in the Liquidity Data Bank Report

The Public: (members trading for): Speculators, funds and all others
outside the first three types. (this is us as retail traders) This type takes both long and short term
positions in the market. Usually long timeframe traders, they dominate
the trends and drive prices away from the center (value) in balanced
markets. The Public ends the day with positions open, creating demand.
Members trading for the public are identified as Cti4 in the Liquidity
Data Bank Report

The Public is what Dalton refers to as OTF "other timeframe participants". We as Public/Retail traders, are the "other timeframe participants"

Why do OTF/LTF participants dominate trends, and drive markets? Because it is profitable. In order compete in the market against short timeframe traders with a house advantage, we as public/retail traders have to make more gains per trade due to the retail pricing structure .

Most Retail traders dont understand how Auction Markets operate. When Dalton wrote his book describing the auction market process, he did from the perspective of a floor trader with a house advantage. When describing OTF he almost describes them in a way that gives an impression it's an "Us vs. Them" mentality. Rightfully so, when the job of the floor trader is to provide liquidity to the market. The pricing structure allows for it. We as retail traders, are not Floor Traders. Why do you think "Big Money" drives the market and dominates trends, rather than dominate balanced markets? Because thats where the money is made. "Big Money" knows they/you can't make big money trading short/intraday timeframe moves. They are the same class of trader we are and have the same disadvantage when trying to compete head to head with floor traders with a house advantage, for short time frame moves.

Now lets jump into the here and now, floor traders are a thing of Past due to online trading. Most floor trading no longer exists. No overnight resting orders, there is no reason to rotate price first hour (IB) filling overnight resting orders. IB concept needs a market that has a daily open & close.

IB is a non-economic process used by floor traders who rotate price in order to fill resting overnight orders. A 24h FX market has no overnight resting orders.

In the past it was the floor trader who provided liquidity to the market by taking the other side of trades. They no longer exist. Now the intraday & day traders are providing the liquidity. The majority of day traders, with their indicators, misinformation, axioms , and lack of market understanding are virtually on the other side of every trade. They provide the fuel & liquidity for the OTF participants.

With the house advantage and fee structure of a floor trader, there was an opportunity to make a lot of money on small moves by taking the other side of the trade. Do the same thing with retail account and pricing structure, and trade for trade, those profits made by the floor trader, result in losses to those with a retail account over time. It's not complex, its math.

80-90% of retail traders lose. Maybe because greed allows them to believe they can make huge amounts of profits trading short timeframe moves if they can just find the right indicator settings, or learn to trade like professional floor traders. As long as retail traders believe that they can make huge profits from short timeframe moves using TA, rather than learning the way auction markets actually operate there will always be someone on the other side of the trade.

The retail pricing structure makes profiting from short term market moves like a "floor trader" difficult and a losing strategy over time....

mzvega, thank you very much for explaining markets, OTF...

Dear Jer21,

ROTATION INDEX PRINCIPLE in pic:



Q="quadrant of close"-> quadrant of the whole range or range/4= 4 equal ranges within the whole range. Measured from day 1 to 4 (for example).
Daily Rotations=daily ranges from high to low, running through the wohle range (and the quadrants). Measured only day by day.

Thats the rotation principle. You can take 4 to 8 days if you like (Jones doese that). Its about the daily rotations movement in the wohle range. If the daily rotation (daily range) moves into a quadrant than 1 goes into your matrix for that timesegment and location (quadrant1or2or3or4). From there you calc your matrix. Thats the whole thing. Now all you have to do is to bring that into a formula.

What is it good for? If you want to trade a break of the 5 day bracket, than its wise to track the developement within the 5,10,15,20 day brackets, because trends start at smaller ranges. First the daily will break, than the 3 day and so on... The rotations can show you a shift from horizontal to vertical development within the brackets. A shift from horizontal to vertical development is a sign of conviction or change from balance to imbalance that might lead to an breakout of the 5 day balance. Jones created an way to calculate that shift, so for some traders it might be easier to read, because its clearly presented by numbers.

Greetings
C

Thanks for taking the time to break it down. Only knew about the 1,2..xVTY (from a previous discussion with Evak I think) in the thread. Very informative.

decided to close GU & EG to take profits and decided to re enter both trades. Yesterday's EOD analysis continued to support both trades....

Still holding both GU & EG as the data continues to suport both trades