Algorithmic trading
otherwise known as robots

purpose
In keeping with our philosophy of informing lone-traders of current trends you will find
    ○ an avalanche of material about algorithms in general.
    ○ some quality articles explaining what they do.
    ○ nothing about how they work.

The "majors" do not disclose their methods, which are constantly evolving, for fear of reverse engineering.


The SFE-SPI200 futures market is dominated by algorithmic trading systems.

The ability to recognize algorithmic activity is an essential skill for retail SPI200 traders.
During slow periods of intermittent activity, where the market maker simulates action with offset orders, filling the order book with spoof (tease) orders, a retail trader can execute an order, get filled, and find themselves under water in a short space of time. In these modern electronic times the need for "small" retail traders to learn the art of tape-reading has increased, not decreased. It's essential.

The skill is the recognition of the market-maker on both sides of the order-book. Before the electronic market, "locals" performed the function of market-makers. They've been replaced by a single computer system which can populate, de-populate and re-populate the order book in a nano-second. Re-arranging the order-book instantly.

lone-wanderer
The lone-trader must be able to detect when the market-maker is the sole occupant of both sides of the order book ("market depth"). During this time the market-maker will know when they have filled a lone-wanderer. During the night session when the spread is wide, 10 levels can be withdrawn from the order-book, pulling the rug out from underneath the lone-trader, testing their resolve.

If we were developing an electronic-algorithmic trading system, that's how it would be done.
See the screenshot below. We would know the total depth. We would know how many were ours.
If someone bought one off us at the front of the ask queue, we would pull our buy orders out, and sell 1 lot, 10 levels down, to see if the buyer had a stop entered.

 
electronic systems
A mechanical system is an automated set of rules operated to make a profit.

There are three type of algorithm systems
• A buy-side algorithm is an automated set of rules used for accumulation.
• A sell-side algorithm is an automated set of rules used for distribution.
• A dual-sided algorithm used by market-maker to either create or seek supply and demand.

Typically, a market maker operates both buy and sell processes, side-by-side, that are aware of one-another.

pinging
Pinging is searching for liquidity.
Orders are sent out, held for several seconds and then withdrawn.
In one reported example, in 60 seconds, 6000 orders were issued and withdrawn.
Only one successful fill was achieved.

robotics
Is there a solution?
Yes. It can be mastered.
With a long involvement in technical programming, robotics, and numerical-control systems we have a distinct advantage in this area in that the fashion is catching up with us. Or a rusty skill is coming back into fashion. We can see them. We can see it happen. There are imperfections in algorithmic systems. Our software solutions, detailed on later lessons, contain the kernel of these skills. Algorithmic trading systems are a modern application of old-style numerical-control-systems. Learn about them. Not hard.


tell-tales
The behaviour of algorithms is obvious once you know what you are looking for. The presence of an opposing robot is crystal clear to (a) another robot, and (b) institutional traders who are trading against robots. The tell-tales are there. Bigger and better than before.

If you think like a robot what would you expect when coming up against an opposing robot.
If you think like a market-maker what would you do when up against an opposing robot.

Algorithms are used by "majors" in

index rebalancing
arbitraging
market making

manifesting themselves in

sweep orders
iceberg orders
market depth dancing


some worthwhile reference material
We've done some of the research for you.

Financial Times Mandate - Algorithmic Trading
When the robots get it wrong - damage - and - opportunity

tell-tails

electronic order replenishment
order splitting, order scaling

  a tell-tail.

Scaled Orders. Just another tell-tail.
Some trading platforms feature scaled orders. An order of 100 is entered into the brokers system, but presented to the SFE system 10 lots at a time. As each block is done, it is electronically replenished, instantly, until the order is exhausted. Orders can be scaled so they are not evenly distributed. Observers can detect this activity by an order being hit and replaced at electronic speed. At the same price, repetitively. That's an alert a large order is being disguised. A high probability there are 3 orders being done at the same time. It can be seen if you look for it. Think this through from the perspective of an order of 300 split over 3 brokers. Why split it. Why scale it. Then put yourself on the other side. It's a game. Large scale "gaming tactics"


update October 2008
Over the past 4 years, the 5 major "prime broking houses" poured hundreds of millions of $ into the development of algorithmic systems and establishment of huge centralized computerized trading rooms. 3 have gone and the remaining 2 have become "banks" which means their "prop-shop" activities will be severely curtailed.

Their algorithms didn't help them. If anything, their unfailing belief in and reliance on algorithmic systems probably contributed to their demise.



the termite effect       sharks, sheep, sheep dogs, termites.
In the Market Depth Section it is stated commercial SPI traders never show their hand. Large orders appearing in the queue are placed there for a reason. What reason?. Should there be a reason?. Ask the right question.

Assume the market is going up. When a large order appears in the queue above the market, one of three responses can occur. (a) The market reverses, (b) comes to a stand still, or (c) goes straight through it. Usually, the market will trade up to but not touch the order, bouncing off it and retracing slightly. Then it has another look. And another. It will oscillate for some time, through a range of 5 to 6 points. Assume 5 points for the purpose of this discussion. During the oscillation process particular care needs to be exercised, examining the size of orders 5 points below. That's where the sheep dogs work. Sheep dogs are medium sized orders large enough to hold the market up. If the sheep dogs appear then look for a REVERSAL, otherwise look for a BREAKOUT.

THE BREAKOUT. If it is going to go through it, once that moment arrives, 1 of 2 things can occur. (a) the order is nibbled away in small chunks like termites, or (b) the large sell order will be pulled out. Once the termites start they don't stop. However whether it is pulled out, or taken out by the termites, the market will continue on through. Until the termites get to work, the market will oscillate, and the sheep dogs will not appear. So what are the sharks doing. They will feed during the oscillation process. If the commercial bid ask count is positive for this time period, they are buying, and, the sheep dogs will not appear. The termites will appear, eventually, after the sharks have had their fill.

THE REVERSAL. If it is not going to go through it then the large order will (a) never be touched, or (b) will be taken out entirely in 1 hit. If it is taken out suddenly in 1 hit the market rarely continues on up more than 1 or 2 points. If it is going to reverse (because it is a genuine top) it is likely that medium sized orders will appear 5 points below to keep the sheep penned in while the sharks get their fill. So what are the sharks doing. They will feed during the oscillation process. If the commercial bid ask count is negative for this time period, they are selling, and the sheep dogs will appear, and, will then disappear, after the sharks have had their fill. If the sheep dogs appear, then, the termites won't. And the large order will either (a) stay there, and not get touched or (b) get taken out in 1 hit. It's a probable top.

I have differentiated between a single hit and a nibble process. The single hit can be 1or 2 large hits. CAUTION: with low incidence, 1 or 2 hits can occur and the market continue on up. For the market going down, all of the above behaviour is true in reverse. Apologies for sharks feeding on sheep but it's easier than goldfish in a swimming pool with a shark and no dogs. If you read the battle of the elephants, you will realize the sharks feed on one another. Some sharks are just bigger than others. If the large order is ever pulled out it is usually a sign the sharks have had their fill and the market is going up. NB: Sharks must keep moving or they die.

All of this is null and void on ambush days, when anything can happen.

the above is achieved by use of the following orders in operation
iceberg orders
skyscraper orders
electronic scaled orders
automatic order replenishment


elephants = big hitters = smart money = major players
The term elephants is used to represent size and identify a group. The footprints are the signature. When watching elephants, remember, ½ of them will eventually be wrong. The secret is to identify the dominant group. The term "smart money" has been avoided as it denotes only the winning group, i.e. the ½ who will be right, and not the ½ who will be wrong. Whereas both groups leave their signature.


can you see the elephants
The eye only sees what the mind is prepared to comprehend. Henri L Bergson

battle of the elephants
Trading is a battle of elephants. They can't hide. A war of accumulation and distribution. It can't be hidden. When elephants fight, the grass gets trampled. Commercial traders constitute 80% of the total volume of the SPI. Because futures is a zero sum game, 80% of the activity becomes a battle of elephants trying to defeat one another. The remaining 20% represent the small traders. Commercial traders are elephants. They cant hide. They leave their footprints behind. Because of the psychological nature of sight, if you are not looking for elephants, you wont see elephants, and you wont see the footprints.

following the footprints
When elephants roam, follow the footprints (can you see the wind). Large elephants leave large footprints, small ones leave small footprints. When the market is quiet they disguise their presence by trading in small lots. They tread lightly. When the market is active they trade in larger lots. They tread heavily.

recognize the footprints
A computer can recognize the shape of an elephant. It can't identify a shadow. It can't identify the cause of a footprint. Shape, size or depth. Human perception can. Back to the grey-box solution.

Price is not a measure of size. What can be used to measure size?. Logically, it's the contract size of each tick.

the dancing elephant
can you see the wind
Yes. If you know what to look for. You can see the signs. In competition sailing it is necessary to "see" the wind. Wind vanes and telltales do not help when there is no wind. In calm conditions the sailor watches the surface of the water. Localized eddies of wind disturb and darken the surface area. As the wind moves, the disturbance, and dark shadow moves across the water. As the wind subsides, the surface disturbance subsides and the shadow disappears. The skill is to read where the wind is, its direction, and move toward it. Sailors know you cant predict the wind. After moving to the area of most disturbance (maximizing probability) the next gust often occurs in the location just vacated. Electronic devices cannot detect these phenomena. Wind vanes and telltales are useful only when there is wind.
 

can you see the gorilla
source - University of California, Santa Cruz and Monash University
If you focus your attention on a basketball game -- for example, on the number of times the white team possesses the ball -- you will be unlikely to notice the person in a gorilla-suit who walks across the court (Simons & Chabris, 1999). This is an example of what Mack and Rock (1998) call "inattention blindness" (IB).


the accumulation effect
Granville, an economist, was one of the first modern day American technical analysts. In his book "New Strategy for daily market timing for maximum Profit" 1963, said "accumulation cannot be hidden". It shows up in volume.

Example: An institution wants to buy 500 contracts at 3900.
There are two options. (a) market order or (b) limit order. Internal broker-exchange rules prevent single large limit orders sitting in the queue. A market order could be executed at that price, only if there was an accumulation of smaller orders aggregating to 500 on the offer. Which doesnt happen. .. Its done this way .. Even a part order to buy 200 would not be placed at 3900 as this would immediately signal to the market that an institution is in the market, and sellers would immediately retreat and the institution would not get filled. A single order for 500, buy, or sell will never be "seen". To accumulate a large number of contracts at a particular price of say 3900, the executing broker would place a large sell order of 100 at 3901, another 100 at 3902, another 100 at 3903. They would then place a buy order of 10 at 3900, and a buy order of 20 at 3899. The presence of the large sell orders will induce traders to sell at 3900, ie below the 3 large sell orders sitting in the sell queue. As soon as the 10 buy at 3900 is done, it is immediately replaced by another order of 10, and again, and again until filled. The constant replacement of 10 on the bid at 3900 is obvious. Like feeding the chooks. As soon as the buy order is filled, the sell orders at 3901, 3902, and 3903, not surprisingly, disappear. When this happens you know where the market is about to go.

the distribution effect

The reverse of the accumulation process. Done exactly the same way.

June 2003
Start with the premise the Australian market is dominated by a small number of large players. Logically they must be antagonists/opponents in the marketplace. We have seen behavioural evidence that can only lead to the conclusion the large players will and do act in collusion. This is not a marketplace populated by many groups and many individuals acting independently in their own self interest at all times. The fact is, the SFE had to take steps in the year 2001 to prevent market ramping and rorting. This had been going on for some years but the ASX and SFE only acted when a small upstart hurt the "majors" and the "majors" complained loud and long. The complaints had occurred many times before and nothing was done. But the one time the "majors" got hurt the noise was deafening. They had been doing it for years. Do leopards change their spots. So far no whistle blowers have appeared. No published swan songs have appeared in the financial press. In spite of the 2001 furor, the media was silent. No investigative research was ever published. It disappeared off the radar. Yet the authorities, and the media, pursued Rene Rivkin over an amount less than $1000.

references
Official ASX disciplinary report action by ASX against AOT Securities 16 Dec 2002
at www.asx.com.au/about/pdf/disciplinarycirculars/658c_02.pdf
The "Age" in Melbourne published the following article on 17 December 2002
at www.theage.com.au/cgi-bin/common/popupPrintArticle.pl?path=/articles/2002/12/16/1039656342953.html

australian asx market rigging articles
lawbookco.com.au/academic/FalseTrading Nomura Case
at www.lawbookco.com.au/academic/ccl-ezine/pdf/vol8issue1_FalseTrading.pdf

demonstrates the media is not part of the system of checks and balances and is not looking out for you.



the power of the few - a
the fine denim affair
better known as the waterhouse-rivkin law of relativity
For those who remember the fine cotton affair .. for those who don't it's on the web ..
When the ASX corporatized, Standard and Poors ("SAP") took over the ASX indices to provide transparency and eliminate conflicts of interest. SAP was to be the HANSARD of the indices. The independent reporter. On April 6 2004 NewsCorp ("NCP") announced it was relocating to US in October 2004. Initially SAP announced NCP could not be in 2 indexes at once. Capitalised at $70 billion NCP comprises 250+ points of the ASX200 index. Given the natural laws of the market place, (ie, thats the way the cookie crumbles, and, let the cards lay where they fall), and the approximate mechanics of the index, NCP should be replaced, in the index, by a company outside the top 200, worth at best, $½ billion or 5 index points. News of the potential removal of NCP from the index should have caused the December 2004 SPI contract to collapse 200 points. The contract did not even tremble. Not even a tremor. The fix was in. The fastest fix ever seen. Faster than fine cotton.

June 23 2004 SAP confirms its decision. Again the far contracts did not falter.

Standard & Poors 23 June 2004 Press Release NewsCorp.pdf states in such an event, SAP would, consistent with its general policies, co-ordinate index changes to limit unnecessary market volatility and the issue of dislocation.

What is SAP's definition of unnecessary and dislocation? When did SAP become responsible for managing market volatility?. Was it about the same time Hansard became responsible for managing the volatility of parliament?

Bryan Frith The Australian June 24 2004
The ASX was opposed to News falling out of its indices The SAP index committee has five representatives - three from SAP and two from ASX. If the talk is right the committee split on those lines - ASX wanted a compromise to be reached to allow News to stay in its indices, at least in some form, but SAP didn't want to create a precedent. Because it had the majority, SAP prevailed. However, SAP's decision to put News immediately in the NYSE S&P 500 index and to phase it out of the Australian indices over nine months, in four tranches, is thought to have satisfied most of the Australian institutional holders

Bryan Frith The Australian 29 June 2004
SAP originally intended NCP would be removed from ASX indices in three equal tranches over a three week period, and that was a concession because normally such changes would be done over a matter of days. SAP appears to be now backing away from that timetable, suggesting the phase-out may take place over a period of months rather than weeks, perhaps as much as six months. SAP said last week it would consult with market participants as to the appropriate level and period of the phase-out. It seems local institutions strongly favor the change taking place over as long a period as possible to ensure an orderly transition.

Jane Schulze The Australian August 07 2004
NCP's move to the US received a boost yesterday when index manager SAP provided a detailed timeline for any move. SAP had been widely criticized for fostering market uncertainty for failing to outline how the move would be managed. After consulting with fund managers, it has now unveiled a four-part transition process which will kick in if NCP shareholders approve the move later this year.

Wendy Frew SMH October 06 2004
In a surprise move, international stock index operator, MSCI announced it would remove NCP from the Australian indices, to the US indices, less than a week after the shareholder vote in Adelaide on 26 October should the relocation be approved.

the smoothing effect
Postponement of the effect of an event. An unacceptable, and not unknown practice in financial circles, known as smoothing. Postponement of expenses to next reporting session increases profit. Postponement of revenue to next reporting session decreases profit. Activities prevalent in the dot.com era the results of which are now well known. Symptomatic of the power and influence of the dominant few.

SAP has created a powerful and dangerous precedent. In the event of a substantial two-day "rout" in overseas markets with the potential to cause a drop in the Australian market of say 200 points would meet the very criteria SAP has established. Unnecessary volatility. Based on the above criteria, SAP would now be expected to step in, pull the levers, and "smooth" the impact over a 9 month period.

Footnote
26 October 2004 NewsCorp shareholders approve relocation to US. June 2005 SPI does not falter
24 Jun 2004 spread between December 2004 SPI and June 2005 SPI = +27 points
31 Aug 2004 spread between December 2004 SPI and June 2005 SPI = +20 points
25 Oct 2004 spread between December 2004 SPI and June 2005 SPI = +21 points
26 Oct 2004 spread between December 2004 SPI and June 2005 SPI = +21 points


Old saying - Follow the money trail. From the outset the heavy money was on one outcome. Evidenced by the heavy money staying on the table. Extreme confidence. The ensuing 6 months was merely the dance of the tarantulas.


demonstrates the inability to obtain simple answers to logical questions.


August 2004 inside the tent
On Friday 30 May 2004 between 10:00am and 14:00pm the SPI had an upward move of 14 points. Between 14:00 and 15:45 it fell nearly 15 points. The following week, the "market place" section of the AFR reported an email had circulated among brokers on that afternoon about the pending disposal of a $200 million portfolio. The sale took place at 16:05 in the after market auction. The cash index effect was only 4.40 points. The SPI recovered all its loss after 1610pm. And they got Rivkin for a mere $1000.


the power of the few - b
Michael West - Margin Call - The Australian - 19 February 2005
The ASX. What a profit result this week for the monopoly which is listed on itself and regulates itself. ASX is a fine company indeed but does it attend its own classes on market transparency? The latest ruminations about blacking out stockbroker numbers attaching to trades is a shocker. Retail punters should march, post haste, on Bridge Street because trading needs more transparency, not less. This is a blatant free kick to the powerful eight (merchant) banks that control 80 per cent of ASX turnover. Ban broker numbers (until three days are up, as is the proposal) and the funds management/principal punting/corporate advisory divisions of the ASX cartel member-owners will run amok. Um, perhaps this is why we heard this week that students of an ASX compliance course were advised that the term "Chinese walls" was being phased out in favor of "invisible barriers". A bit like schools changing blackboards to chalkboards.

The above is an example of Pareto's Law in action.


the power of the few - c
the amazing coincidence
Monday 21 February 2005 it is learnt the Sydney Futures Exchange will terminate its broker statistics service as at the end of February 2005. Broker Statistics is a live display of broker participation in the market. Showing broker code and quantities transacted. Gone.

Early in 2004 the Sydney Futures Exchange ceased live publication of Broker Statistics for Bonds and Bills. Reasons given for cessation, at that time, are identical to those outlined above in West's article on ASX broker identification. It's the old closing the stable door trick. A bit late after the horse has gone. As the saying goes, you don't know how good it was until you lose it.

Few know of it's existence. SFE doesn't advertise it. Vendors don't offer it. One solitary broker does. So who cares.

demonstrates the constant erosion of the system of checks and balances in the market place.



proprietary trading
smoke and mirrors - a bedtime story
source wikipedia
Proprietary trading is a term used in investment banking to describe when a bank trades securities with its own money as opposed to its customers' money, so as to make a profit for itself. Although investment banks are usually defined as businesses which assist other businesses in raising money in the capital markets, most investment banks make the majority of their profit from trading on their own behalf.
source investopedia
When a firm trades for direct gain instead of commissions. Essentially, the firm decides to profit from the market rather than from commissions and fees. Firms engaging in proprietary trading believe they have a competitive advantage that enables them to earn excess returns.
source - SMH - 20 May 2006
The ASX-ASE estimates 30 to 40 per cent of total stock market turnover on any given day is proprietary trading.
Merrill Lynch's proprietary desk is one of the most prominent in the local market, along with Credit Suisse.
Above estimate is considered an understatement made for public consumption. More accurate - about 60%

Goldman-Sachs proprietary trading desk reportedly accounts for 80% of total turnover on NYSE in the march 2009 quarter.



high finance in hightown - a true story
In the 1990's a stressed-out proprietary-desk trader (now long-gone) would periodically let slip a few gems. He worked for an Australian based broker that has since been swallowed up in the consolidation of later years. Call them HighTown.

As told ..

The principals of HighTown met every morning at 7:00am to review the overnight action in the overseas markets. The house had something like $50m of its own working capital to push around and get a return on. One of the top 10 ASX stocks, (big enough to push the market around), registered in Australia, and traded in London and New York. Being an Australian registered company, the London and New York markets for that stock were often thin. Frequently when the offices of HighTown opened at 7:00am there would be a number of overseas orders for large quantities of that one stock sitting in the the fax machine in-tray. Naturally the proprietary-desk derivatives trader wasn't supposed to know that - i.e. - Chinese walls. Knowing large orders in that one stock would move the market, the nature of the 7:00am meeting was to assess where best to trade the house-funds that day. Naturally if the large overseas orders were buy orders, the house activity for the morning was inclined to the buy side to give the market an extra shove up. If orders were sell orders, the house bias was to the downside, the boulder given an extra nudge down, with the house ambulance waiting at the bottom of the cliff.

An indication of the extent of proprietary trading can be seen on the last page of brokers recommendations and newsletters which have mandatory disclosure statements listing the stocks the house holds on its own behalf at the time of the recommendation. Disclosures relate only to stocks contained in the newsletter and do not represent the full extent of house holdings. Of course if recommendations are not issued, the broker is not obliged to declare their interests.
 

honkers and the house of the rising sun
Just prior to the Asian markets opening the fax machine again ran hot. Usually around 11:00am through 11:45am. Long-Gone PropTrader would express amazement at the size of the orders and dismay at the apparent suicide nature of them. That's where lunch-time surges came from.


program trading
a variation on tape reading - a simple clue

Program Trading can mean a number of things. Total turnover on the ASX-ASE is around $4b each day. It is estimated $3b of that is proprietary trading. It's not done all at once. It's managed in small parcels as discussed elsewhere in this site. Given the value and volume, it's unlikely screen operators sit in front of a screen all day picking away at small orders - mistakes can happen. It's mechanized. If you know what to look for it can be seen. It's regular. Progressive. Happens every day. On any of the top 200 stocks, regular, uneconomic transaction sizes are evidence of iceberg orders and proprietary trading. Brokers don't pay brokerage. read the Goldman Sachs article above.


source: Leon Gettler - age.com.au -March 24, 2007
ASIC has challenged the effectiveness of the so-called "Chinese wall" intended to separate Citigroup's proprietary trading desk, which trades in shares using the bank's own money, and Citigroup's corporate advisory services, which regularly deal with material information that could send a company's price up or down. ASIC's case challenges the way banks conduct proprietary trading, which is a big money spinner. In 2006 five of the biggest investment banks, Morgan Stanley, Goldman Sachs, Merrill Lynch, Lehman Brothers and Bear Stearns, generated $US61 billion ($A78.2 billion) from proprietary trading, about half their total revenue, and a 54 per cent increase over 2005.


Proprietary Brokers = Prime Brokers = Name Brokers
An article dated 20 June 2007 published by the SMH, states Westpac is looking to grow its online trading presence to compete against CommSec, which dominates the online share trading market with a 48 per cent share and Etrade, has 26 per cent. A total market share of 74% between the two retail majors. Allowing a 6% share to Westpac gives a total of 80%. That leaves a 20% remainder spread among the "name brokers".

A notable difference is Westpac, Commsec and Etrade are not proprietary trading houses as are the "name" brokers.

Prime Broker defined
List of Prime Brokers

It will be noted from the list of SFE Futures Brokers
•  Commsec, Etrade, and Westpac are not futures brokers.
•  The futures brokers are also (mainly) listed as "Prime Brokers".

imperator @ http://www.imperator.com.au/informationoutline~nocache~1~SubTopicDetailsID~1354.htm
There is no better illustration of proprietary trading than FX. MBL, an early leader in foreign exchange trading, would offer buy/sell prices to clients in US dollars and other major currencies. Miners being among the biggest users of this market. MBL’s ForeX desk did not link clients buying US dollars with clients selling (as a broker would) but rather took the other side of the client’s position, trading it in the market with other banks and proprietary operators with the purpose of making a profit. Today, of the billions of dollars traded daily in foreign exchange markets, only about 10% involves a "real" client such as an exporter. The other 90% is speculative (proprietary) trading by the banks.



CSPA - ASX after market auction
The 16:10 market. Closing Single Price Auction "CSPA".
15:45 is a critical time for the SPI. To appreciate this we must visit the physical market of the ASX. The ASX conducts an aftermarket auction at 16:10 everyday. Periodically after-market auctions can move the ASX200 index up or down 10 points in one hit ie 11 pts 3 Nov 04. Not often but frequently enough. Back to the SPI. Interesting, each time a large move at 16:10 occurs, it is preceded by a surge in the SPI starting about 15:45. Be careful of unexplained moves in the SPI between 15:45 and 16:00 that are not accompanied by a corresponding move in the ASX200 cash index. i.e. the premium or discount suddenly expands. It is highly probable the corresponding move will materialize at 16:10. If one side of the market in any one of the top 10 stocks is particularly thin, the cash index can be moved significantly by an attack on the "thin" side of the market at 16:10 at lower than normal cost. If it feels like, looks like, and smells like insider trading it probably is. If the move is against you get out. You wont win. It will get worse after 1610pm. Attacking the thin side of the market.

On the last trading day of a quarter the pre-open period prior to the CSPA is extended by 5 minutes to 16:10 with the CSPA from 16:10-16:11. eight pages of explanation by ASX [ASX has removed this article] are required to explain the mathematics, without mention of its purpose. A basic business rule is the more complicated a scheme is, the more likely it will be abused. The CSPA is a forum for large trades only, where king elephants wrap up the day. Why would a large order be any more successful at this time, a 5 minute window, than any other time of the day, unless, by pre-arrangement, there is certainty it will be met. It may be argued to post a large order during the normal session, and leave it sitting there would lead to a "disorderly market". Knowing the CSPA is available enables large orders to be executed "elsewhere". Consequently the normal session is "thinner" that it would otherwise be. Which in turn creates the conditions conducive to an ambush. Hobson's choice. A disorderly market or an ambush.

It is understood (in broker circles) if one particular broker trades SPI futures aggressively during the "CSPA" pre-open, that particular broker will have been a heavy buyer/seller in the cash after market, with a flow on effect on the futures.
Footnotes:

November 2005. The 16:05 match was extended to 16:15. Don't know why. No explanation.
March      2006. The 16:15 match was reduced to 16:10. Don't know why. No explanation.
NB: Periodically, at calendar quarter endings, for one day, the match can be extended out to 16:20 hours.
NB: 2008 CSPA auction now at 16:10.

The ASX200 (XJO) index works hard to achieve a range of 16+ points on any day. In ½ point increments.
On 3 November 2004 the ASX200 surged 11 points in one hit, at 16:05, in the closing "CSPA"
On 9 December 2004 the ASX200 rose 11 points in two hits at 15:58. Whereas the SPI rose 13 points between 15:50 and 16:05. One broker was the only buyer. Both times. Same time, same broker, same action.
Intel travels faster than speed of light through paper walls. Arrives 10 minutes before leaving. Mental telepathy.

On Friday 30 May 2004 between 10:00 and 14:00 the SPI rose 14 points. Between 14:00 and 15:45 it fell 15 points. The following week, the "market place" section of the AFR reported an email had circulated among brokers on that afternoon about the pending disposal of a $200 million portfolio. The sale took place at 16:05 in the after market auction. The cash index effect was only 4.40 points. The SPI recovered all its loss after 16:10.

On 24 August 2005, the ASX200
Fell 2.5 points after the market closed at 16:00 on completion of the CSPA
Then fell a further 8.9 points sometime after 16:30 and after the CSPA

On 25 August 2005, the ASX200
Fell 3.6 points after the market closed at 16:00 on completion of the CSPA
The CSPA did not conclude until 16:20
Then fell a further 12.7 points sometime after 16:30 after (an extended) CSPA
Most news reporting systems did not report the fall until 07:30 the next day, 26 August 2005


No explanations. No announcements. Un-reported facts.

Trading Index Futures - a million different ways
Undressing the market. How the game is played. Who the opposition are. How they play. How they behave. Most trading sites offer specialized trading tools, which are only a part answer.

origins of this research
Had been trading for a few years, when a young guy in the office next door showed an interest in the trading screens. One day, as he was watching with me, I formed the view the market was going up. As I reached for the phone to place a buy order he commented that he thought it was going down. I withheld my order and watched. And he was right. It took 3 weeks to discover what it was he saw that I didn't. At first he couldn't articulate it. Eventually it came. It just so happened he was a serious blackjack player. So began an introduction to the concept of probability and the significance of high value and low value activity.


the futures game
Futures is a game of probability. A zero sum game. For every winner there is an equal and opposite loser. To win, someone else must lose. If you lose, someone else wins at your expense. So understand the context in which the game is played, together with the concept of probability.

quote:- The empires of the futures are the empires of the minds - Winston Churchill

the trading environment and its context
Factors to take into account are:

• No single organization dominates the US market. Perfect forces of supply and demand operate.
• The size of the Australian market is small by comparison
• Average total value of trading on the ASX is approximately $AUD 5 billion per day
• Most Australian broking houses trade on their own account and are active every day.
• The 2 largest brokers can each mobilize $AUD 3 billion on any one day.
• Any one of the largest brokers has the financial strength to ambush the market on any day.
• There are approximately 20 large institutions who actively trade, rather than just invest.
• Own-account brokers and institutions comprise 80% of the total volume on the ASX.
• Perfect forces of supply and demand do not operate in the Australian share market.
• Can the largest participant be active without leaving any trace of their activity.
• With the power to ambush the market why wouldn't a large broker do so if they can make money out of it.
• The aftermarket auction can move the ASX200 by up to 30 points in one hit at 1610pm.
• The aftermarket auction enables the big hitters to hide their hand. If they can, why wouldn't they.
• Australian based Hedge funds became part of the Australian landscape from January 2003

the trading bunker
Until the end of the tech boom it was common for many of the large insurance companies to operate trading rooms. These trading rooms would contain 5 people, be isolated from one another, and have a float of $30 million to trade with. Five groups per organization. At the end of each month the group which had performed the worst would be terminated. So the pressure was on to perform. They could trade any market any time. Just so long as they made money. It was ruthless and made for extreme volatility in the SPI. Most, but not all, of these trading bunkers have been disbanded. But it gives you some idea of what you are up against. As the saying goes don't stand in the way of a speeding train.

bunkers are back with a vengeance. 2004
It is understood that since hedge funds arrived, starting 2003, the trading bunker has returned. Same methods.


occam's razor
The preference for simple explanations is ancient. This principle, known as Occam's Razor, after William of Ockham (Occam), who, in the 14th century stated : "(Plurality should not be posited without necessity.)" or "keep things simple", or "simplicity is better". "Occam's Razor Proper" states if two models make equivalent predictions, the simpler is preferred. Jacob Eliosoff and Ernesto Posse at http://cgm.cs.mcgill.ca/~soss/cs644/projects/jacob/

Eliosoff & Posse's introduction is an excellent example of pattern recognition.

The rules of simplicity stated in Occam's razor appear to come from Ockham's work in comprehensive logic, based in turn on Aristolian Logic. See below.

In his book The Options Edge 1998, Willian R Gallacher, comments on the "impenetrable logic" of the famous Black-Scholes formula, and provides a reduced formula which he calls "Ockhams Equation". The book's flyleaf quote : "What can be done with fewer is done in vain with more". William of Ockham.


darwins blade
Quote by Dr. Darwin Minor Phd from Darwins Blade by Dan Simmons.
Darwins Blade states that "all things being equal, the simplest solution is usually stupidity" examines causes and effects after the event. (ie the simplest explanation for most acts of stupidity is stupidity). see Darwin Awards posthumously given to "those who improve our gene pool, by removing themselves from it" darwinawards.com
Occams "all things being equal, the simplest solution is usually the correct one" examines alternatives before the event.


logic
Logic is defined as the relationship between elements, and, between an element and it's set of objects, or events. Logic is concerned with what is true and how we can know whether something is true.
Paradox.

Aristotle (384-322 BC) the "father of logic" established a set of rules for deductive evaluation.
Abelard (1079-1142) and Ockham (1285-1349) extended Aristolian logic to comprehensive logic.
G W Leibniz (1646-1716) extended comprehensive logic to symbolic logic.
George Boole (1815 - 1864) founder of mathematical logic, developed Boolean logic.


Using trade classification symbols/mnemonics "a"=ask-hit ="buy", and "b"=bid-hit = "sell".
Then using the rules of logic ask the following questions in relation to an executed trade
1. does a single trade of 1b lot have the same weight as 20b lots at the same price
2. does a single trade of 1a lot have the same weight as 20a lots at the same price
3. does a single trade of 1b lot carry the same weight as 20a lots at the same price
4. Are 150a lots the same as 150b lots, at the same price.

Example - market bidding 130 @ 3900 and offering 20 @ 3901
Look at 3 sequential trades. 50b done at the bid, 50b done at the bid, then 1a done at the ask, 1 point higher.
Picture the three tick points on a chart. Bid price and ask price did not change. Price ticked up on the 3rd trade. Now consider the effects of graphics illusion and gestalt principle of continuity. Visually price is rising, while the underlying numbers are logically saying selling pressure exceeds buying pressure.


pareto's principle: the 80-20 rule
quantitative versus qualitative
The qualitative rule, the 80-20 rule, connection with auditing, surveys, sampling, computing and trading.

source - Arthur W. Hafner, PhD, at www.public.asu.edu/~dmuthua/pareto's_principle.html
Pareto's rule states that a small number of causes is responsible for a large percentage of the effect, in a ratio of about 20:80. In 1906, Vilfredo Pareto (1848-1923) an Italian economist, observed twenty percent of the people owned eighty percent of the country's accumulated wealth.

The Pareto Principle states a small subset ("vital few") affecting a common outcome tends to occur more frequently than the remainder ("useful many"). Pareto Curves graphically illustrate data concentration. Pareto Charts can be used to evaluate the relative importance of any subset.

The SPI displays exactly those characteristics. Strange that.
20% of trades executed comprise 80% of total volume / value.
80% of total volume arises from 20% of executed trades
Approximately 30,000 contracts (lots) traded each day
Approximately 12,000 trades are executed each day, an average of 3 lots per trade.
Approximately   2,000 trades comprise 10,000 lots, an average of 5 lots per trade
The remaining 10,000 trades comprise 20,000 lots, or 2 lot per trade. A lot of smoke and noise.


audits, surveys, polls, and quality control
Are activities which use random sampling on the basis the selection is representative of the whole. In auditing and quality control, if members of the sample are unusual, greater focus is placed on those items and the area of examination expanded and intensified. In auditing, if 20% of an organizations financial transactions represent 80% of the value, greater emphasis is placed on sampling within the 20% high value, low quantity, transactions. An organization with $100 million assets and $20 million turnover, main focus is on verification of assets. An organization with $20 million assets and $100 million turnover, main focus is on verification of revenue. While sampling remains random it is directed. Toward a specific class. If abnormalities are detected within the sample, enquiry escalates, and a full examination conducted.

With SPI data we can do better than that. Don't need to sample. The 20% identify themselves. They don't hide. Just buried among the main data set.

computers, the lot, the many, or the few
A computer is a logic device. Computers perform repetitive tasks well. Computers cannot perform the human skill of perception. A computer is best used performing quantitative tasks of extracting the 20% from the main body of data. Enabling the trader to use the grey-box-system and perform the qualitative decision making task using the smaller 20% data set. The 80% can be down-weighted (ignored) as stated in Law of Large Numbers below. By focusing on the high quality 20%, time and effort are reduced by 80%, and performance is enhanced.

Is the SPI any different?. Ask this question. Does a single trade of 1 lot carry the same weight as a single trade of 20 lots?. Logically, considerably less weight. A single trade of 1 lot will not move price. A single trade of 20 lots can. It is not necessary to examine 3000 items of data. Single lot trades occur progressively during the day. High value trades occur periodically and in clusters. Observation can be restricted to the action within the smaller number of high value trades. Single lots can move price only when a sustained program, feeding one lot at a time, is undertaken over a short period of time. In which case all trades will hit the same side i.e. sweeping the market.


law of large numbers
source - Professor Richard G Lipsey. Phd. Economics. Queens University UK.
A law relating to a large quantity of numbers, as distinct from numbers containing large values. Successful predictions about the behaviour of large groups are made possible by the statistical "law" of large numbers. Roughly stated this law asserts that random (abnormal) movements in a large number of individual items tend to offset one another. The larger the group, the more likely the neutralizing effect.
In other words a large number of small items can be ignored or discarded.


revolving door
Trading is like a revolving door. You see the direction. You see the speed. The trick is to judge the speed, enter at the right time, move in sync with it, and exit at the right time. Get it wrong and you're stuck inside, come out facing the wrong way, or get trapped when the power goes off.


vacuum effect   =   zig-zag effect
when the selling stops

Assume a down move. Reverse for an up move.

An implicit (charting) assumption is each price point has a uniform (constant) value i.e. the force of a 5 point rise is equal to the force of a 5 point fall. That's wrong. It's a natural visual illusion. The majors know it and take advantage of it.

Price can change for a number of reasons. A large price fall is often the result of an aggressive sell action by one or two brokers. It's never continuous. They stop and start. When the program is complete, the pressure ceases. Leaving a vacuum in its wake. And price immediately rises back into the vacuum. The magnitude of the retreat is usually, 1 unit of the root of the controlzone. If the move was swift and unexpectedly large, the magnitude of the retreat will be, two units of the root of the controlzone. A retreat into the vacuum is identified by an absolute lack of commercial participation in the rise. If they are participating, the fall is over.

The fall is due to selling pressure exceeding buying pressure. Once the sellers have completed their program, they are (obviously) not selling any more. The selling pressure stops cold. The buyers who took the other side were sitters. They're not chasing. Thus they are not the cause of any subsequent price rise. They're buyers at lower levels (below the cessation point), not higher levels. It is the jobbers who now step in and play in the vacuum.


reversal versus pullback
Understand the section above and identify the difference between a reversal and a pullback.


markets fall for 3 reasons
economic recession
profits recession
an economic event


trading rules
Basic themes extracted from the site in the form of a set of rules in order of priority
with a comparative list of the common rules found elsewhere (on the net).


 

trading plans
The fallacy of having "a" trading plan is having just one.
It is more realistic to have 365 trading plans. One for each day.
The challenge is - which one do you use.
Same as the challenge of selecting the right tool for the day.


hardware v software v wetware
source - Greg Iles - Dark Matter - 2003
"The human brain is slow in terms of computing speed. But it is massively parallel. For processing, it is capable of making 100 trillion calculations simultaneously. Together with the equivalent of twelve hundred terabytes of memory".

the surfing plan
The best explanation of a plan is "the surfer" who goes surfing. Goes down to the sea. Checks the weather conditions. Is the surf up?. Is the breeze on-shore or off-shore?. If the conditions are right, jumps on the board and paddles out 300 meters. And waits. Waits for the right wave. There are many waves. The key is to pick the right wave. Don't want to pick one too soon or too late. The choice of wave comes from seasoned experience. Not a surfing plan. A surfer does not go out surfing with a surfing plan in mind. The conditions of the moment on the day will dictate the play. Depending on the conditions, what was a good wave yesterday might well be a poor wave today.