a million permutations
Combining the principles detailed in Rules of Logic, Occam's Law of Simplicity, Paretos Principle, Law of Large Numbers, Laws of Probability, and the trading rules, the methodology is simple. Select the "the vital few", ignore the "useful many" (jobbing), establish a method of classifying them into buying or selling, (laws of supply and demand) then assess which is the dominant side. Provides something manageable to work with. It identifies where the weight of money is going.

This gives rise to the vacuum effect. When the buying or selling stops. Price retreats into the trailing vacuum. The slipstream. Where most of the errors occur.

COMMERCIALS - the major players

Commercial trader
A commercial trader is any trader who deals in large quantities of contracts.
Commercial traders can trade in small lots and large lots.
Commercial traders behave differently on light volume days compared to heavy volume days.
Commercial traders behave differently when entering a trade compared to exiting a trade.

Commercial trade
A Commercial Trade is any single trade of a large quantity of contracts.
On large volume days, approximately twenty-five percent (25%) of all trades done, and, up to eighty percent (80%) of all contracts traded, will be done in commercial sizes i.e. 25% of total trades may comprise up to 80% of total volume.
If a sudden move occurs in the price of the SPI it is important to know what the commercials were doing just prior to the move. Did they initiate the move. Are they continuing to participate in the move.
There are two order queues: the bid (buy) queue and the ask or offer (sell) queue. The bid queue will always be at least 1 point lower than the ask queue. For a commercial trade to occur, there must be a reasonable number of contracts available in one of the queues.

fragmentation - data reporting
The SFE reports trades as component parts. Where 20 lots of 1 are in the queue, and a commercial trader hits them in a single trade, the SFE system reports 20 separate trades of 1 lot each. Where 1 lot of 20 is in the queue and 20x1 lot sellers hit simultaneously, the SFE system reports 20 separate trades of 1 lot each. In essence the SFE sells kitset furniture in kitset form. It needs to be assembled to be useful.
Both the SFE and ASX trading systems report trades as fragmented parts. Fragmentation is defined as separating something into particles. It is further defined as the disintegration of the behaviour of related items, being critical to our studies of logic, analysis, technical indicators, and visual perception. It has a significant effect on line and tick charts, indicators, and course of sales. Important if using Pareto graphs or curves. And Pareto's classification of items according to their relative importance. It has no effect on point and figure or time based bar charts.

Price charts based on time bars using the following data sets (a) fragmented, (b) defragmented, and (c) selection based on Pareto principles, are identical. However indicators produce different results.

The fragmentation of the data stream by the SFE is equivalent to a nuclear breeder reactor producing more material than it consumes. The questions we have are: Is this intentional, or given the resources of both organizations is it an example of the technology dictating the reality together with reverse engineering of the Pareto phenomena. Data fragmentation, broker split orders, client scaled orders, create a smoke screen, disguising trading activity. We at Camron overcome this effect by defragmenting the data.

Over the course of a full day the order of fragmentation is about 2/1. In actual terms the true rate is about 3/1. As stated elsewhere elephants cannot hide.

The implications should be considered in relation to information compression, pattern recognition, Pareto's rules, relative importance, tick charts, item relationship and thus technical indicators. defines "quantum" as the smallest physical quantity that can exist independently. By reporting trades in this way the SFE takes quantum physics to its limit, inducing a "quantum effect" as discussed under "observation and quantum theories"

We have examined data feeds from the exchange transmitting the dow mini, russel mini and S&P500 e-mini. The same is true with these data feeds.

Ask the right question(s)
why do Institutions spread large orders over 3 brokers
why do trading platforms provide "iceberg" order facilities.

Fragmentation started in 2000 when electronic trading began. ASX and SFE data specifications don't disclose it. These observations were released in 2001. For the world to see. web-search for "defragmentation spi200" or "fragmentation spi200". Six years later, we are still the only site. No-one else bothers about it. One charting technician, using "tick charts" with fixed, constant number of ticks per bar, responded by doubling the number of ticks per bar, assuming uniformity of behaviour. Prime-brokers defeated that by releasing icebergs, in varying quantities, in varying time frames. Ensuring one cluster of ticks will never be the same as a previous cluster.

commercial bid ask count - example
The following image displays the commercial bid ask count for the SPI200.
Y axis is price. X axis is time. Yellow columns are lunchtime. Blue row is opening price.
Bright figures indicate point of column change.
The value in each cell is the bid ask count at each price-time point.
SPI opens at 6432, falls to 6428 in the first column, then rises to 6435 in second column. The bid ask count starts at 0, falls to -2, rises to +9. In column 3 the SPI fell 13 points while the bid ask count stays positive at +5, indicating the commercials are not enthusiastically participating in the fall. The lunch time columns indicate a typical lunchtime ambush.

Significant are the rows at 6426,6427,6428 after lunch. The commercial count rises from 4 to 25 and 3 to 31, while price doesn't, indicating the commercials are buying at these levels.
Reaction, price rises to 6438.

can you see 3 dimensions
Close examination of this graphic will discover 3 dimensions instead of a conventional 2 dimension price chart.

dgas trade - the don't-give-a-shoot-trade
When the big hitters attack the big sitters.
the trades that drive the market - elephants on the move

Under the rules of supply and demand, price equilibrium occurs when buying and selling are equal.
When buying exceeds selling, price will rise. When selling exceeds buying, price will fall.

When sellers outweigh buyers, buyers retreat to absorb excess supply. Sales occur on the "bid" side.
When buyers outweigh sellers, sellers retreat to absorb excess demand. Sales occur on the "ask" side.

In the following examples buying and selling are not equal.
Trades go against the weight of numbers and defy the rules of supply and demand. Why?

You can't see this happening in a standard colour coded course of sales screen.
Particularly in a fast market

dgas bid ask count - counting the elephants
The DGAS count is a sub-set of the commercial bid-ask count. Defined above.
Only counted when a dgas trade occurs.
When the dgas count exceeds 10 (or minus 10) and leads the ba count
there is little doubt where price is going.

dgas results - sfe spi200
Note - this is a different screen compared to the bacount screen above
The dgas were disinterested at the open, a buyer became interested at 5085 through 5094 driving the net count up to 23, then capitulated at 5075, when the selling started. The worst the count got to was minus 9 at 5069 after price had fallen 40 odd points, indicating the commercials were not driving the price down, merely following the cash asx200 index down. Tentative buying started below 5060 down to 5040 the low. Indicating shorts taking profits. They certainly were not selling. Buying strength after lunch speaks for itself.


ambush - attacking the thin side of the market
The market is constantly monitored to determine which is the thin side. ie
The side that can be moved the greatest number of points, for the least cost.
Must be imbalanced to occur. If both sides are evenly balanced it wont happen.

how it's done

For the purpose of this example assume the following
market depth for National Australia Bank

bid qty

bid price

ask qty

ask price











NAB last sale price was $30.00
NAB fair value is $30.00 per share
Note the absence of heavy sellers below 32.00.
A situation that happens occasionally and does not have to reflect fair value.
There are several organizations who can muster $1billion in funds quite easily, at any time.

Objective (a)
to raise the price of NAB $2.00 to $32.00 will require the purchase of
20,000 shares lifting the price to $31.00, for an outlay of $620,000, or
20,001 shares lifting the price to $32.00, for an outlay of $620,032
for a lift in price of $2.00 with a corresponding lift in the index of 20 points.
If done just before or during the ASX closing 'CSPA', the index will be stranded high.
The SPI future has little choice but to match the ramp.

Objective (b)
to reduce the price of NAB $1.00 to $29.00 will require the short selling of
100,000 shares @ $30.00 for an outlay of $3,000,000 to clear out the 1st level.
Then 1 additional share @ $29.00 to get to the second level.
for a fall in price of $1.00 with a corresponding fall in the index of 10 points.
If done just before or during the ASX closing 'CSPA', the index will be stranded low.

An outlay of $600,000 produces a lift in price of $2.00 and approx 20 index points
An outlay of $3,000,000 produces a fall in price of $1.00 and approx 10 index points

Consider the possibility an organization continuously runs software monitoring this situation, looking for extreme thinness on one side of the market, which can be attacked, where the cost of the attack is less than the gain in the futures.

Think it's not happening?.
Why do ambushes happen on holidays more often than not.
The possibility one side of the market will be on holiday is high.
The possibility both sides of the market may be on holiday is low.
Large 4:00 pm gyrations are occurring more frequently now, beginning 2005.

The profit on 5,000 contracts @ 20 points @ $25 is $2½ million at a cost of $620,000
And they still own the shares, which can be liquidated.
It wont be a loss. Maximum loss = 20,000 x $2 = $40,000 + finance costs if any.
If liquidated before T3 and T5 come around, there is no cash outlay at all.

The hypothetical example used to demonstrate the point is one dimensional, with a single stock, with extreme price separation. In practice that wont happen. The top 5 stocks are monitored, over a narrower price range, for a situation that achieves the same price-cost effect i.e. lift 5 stocks 4 points each. The software required is extremely simple. The opportunity for these circumstances to occur is during the dislocated Australian holiday season. NSW and VIC celebrate same holidays on different days i.e. labor day. One state closes, the other remains open. As during the spring racing carnival - The Melbourne Cup - when VIC closes and NSW stays open. When some cats are away the mice should play carefully.

Additional information