On Friday, June 27, 2008, the Dow Jones Industrial Average closed 19.9% below its October 2007 high. The generally accepted definition of the start of a bear market in equities is a 20% drop, which the Dow hit a few days later and the S&P 500 hit a week after that. This is an excellent time to begin a discussion of algorithmic trading.
During that Friday’s trade, the Nasdaq 100, after an early break, recovered and closed high. The S&P 500, after an early break, closed low. The euro, 30-year Treasury bonds and gold all closed high. Crude oil, after an early rally, closed mid-range. Equities were weak on the day. Bonds, the euro, gold and crude oil were strong. To someone who trades or invests in multiple markets, June 27 was just another day. There were opportunities to go long and go short. There was price movement. There was liquidity. It was orderly and active. It was a good day to trade.
The next day, in the business section of the local Saturday paper, there was an article on the stock market. It was decidedly dark; the storyline was that falling prices are bad. On the same page, there was another negative article about energy prices. The storyline was that rising prices are bad. For those of us who traded the short side of equities and the long side of crude oil the previous day, it might have created a tinge of guilt. Then, hopefully, we came to our senses.
One purpose of free markets is to achieve a balance among buyers and sellers, between production, supply, demand and distribution. The basic mechanism is price exploration. Each time you are in the markets, buying or selling, you are helping the system work. Traders will not succeed by feeling guilty or by waiting to be rescued. We need to learn how to succeed in any market conditions. Thanks to the predominant application of algorithmic trading on an institutional level, and the volume, liquidity and order flow that has followed, that is becoming easier to do.
AS THE MARKETS TURN
As a trader or an investor, the most basic thing you can learn is not to fight a market. If a market is falling, you should situate yourself to benefit, or at least protect yourself from the downtrend. If a market is rising, get long or protect yourself from the uptrend. Be market and price neutral. This is where algorithmic trading comes into play.
Large players dominate algorithmic trading, while small players are learning to navigate the algorithmic world. Why was the S&P 500 weak on Friday, June 27? Why were bonds, the euro, gold and crude oil strong? The answer to both is the same: Large, successful players were doing their job. They were moving (rotating) money between markets. They were market/price neutral. “Mirror opposites” offers a graphical representation of this relationship.
At a conceptual level, algorithmic, or algo for short, trading is a way to move money, on any scale, into or out of or between markets. If you don’t think you need to know more about algo trading, think about this: The bulk of the trades on U.S. equity exchanges are now algo-driven. Or, for an anecdotal idea of how big this concept is, simply do an Internet search on “algo trading.” The good news is you can start using algo concepts to your own advantage.
An algorithm is a “procedure for solving a mathematical problem,” according to Merriam Webster’s Collegiate Dictionary. One good definition of computer-based algo trading, from the research firm Tower Group, is “placing a buy or sell order of a defined quantity into a quantitative model that automatically generates the timing of orders and the size of orders based on goals specified by the parameters and constraints of the algorithm.” Large players use sophisticated computer systems, but small players have choices too. You can even use algo concepts by watching the markets in the right way with your naked eye.
WALK THROUGH TIME
A complete embrace of algo trading may require a different perspective on the markets. Visualize particles moving in a liquid or in a gas. How, in terms of physics, do these particles move? Is it possible to map and predict their motion with relative accuracy? Financial markets are like a liquid; there’s a reason the term “liquidity” is used to describe the ability for market participants to move into and out of a position. The prices of what you can buy and sell are like particles moving in a liquid. How, in terms of financial markets, do prices move? The work of five individuals can help us better grasp this concept.
Australian botanist Robert Brown (1773–1858) is the namesake of Brownian Motion, also known as Particle Theory. Whichever name you use, Brownian Motion or Particle Theory, the concept involves a mathematical model or models, an algorithm, which attempts to map and predict “random” particulate movement.
French mathematician Louis Jean-Baptiste Alphonse Bachelier (1870–1946) in 1900 published his Ph.D. thesis, “The Theory of Speculation.” He took on the task of modeling Brownian Motion with regard to financial markets. He appreciated the similarity between particles in a liquid and the prices of financial instruments in a market space.
Then, there’s Richard Davoud Donchian (1905–1993), who became convinced that it was possible to extrapolate price behavior whether markets are rising or falling. In 1949, he started the first publicly managed futures fund. He popularized what was at the time a radical idea. He devised a way of using price-based rules to successfully determine when to enter and exit a long-term trend. Before Donchian, market analysis was market analysis. After Donchian, market analysis was split into fundamental analysis, based on economic factors, and technical analysis, based on price alone.
The American economics professor Burton Malkiel is the next critical player in this evolution. In 1973 Malkiel published “A Random Walk Down Wall Street.” Malkiel is an opponent of market timing and rotation. He is a proponent of buy-and-hold. Malkiel’s studies convinced him that it is impossible for any one participant to predict where a price may move next. He postulated that market participants collectively and necessarily create an efficient market in which price movement becomes effectively random (see “Forget about it”). However, instead of halting minds like Bachelier and Donchian, Malkiel in many ways encourages them.
Peter Williston Shore in 1994 devised a quantum algorithm that was exponentially faster than the best currently in use. Its invention heightened interest in quantum computers and helped to galvanize algo trading.
Then, there are the quants, short for “quantitative analysts.” Quants began using computers to analyze market behavior to an unprecedented degree. At first, these analysts were considered oddities in the financial world. Soon, they became masters. It’s no surprise that their rise has paralleled the growing power and sophistication of computers.
Now algo traders are riding the computerization Tsunami. The first quants primarily analyzed. The first algo traders primarily executed. Institutions, hedge funds, trading desks and brokers needed, first, a way to break large orders into smaller anonymous orders, and, second, a way to manage complex bundles of orders across a variety of markets with minimal market disruption at minimal execution cost.
The task required some serious algorithms (see “Hidden agenda”). Algo traders rose to the challenge. It is a short step from order-flow maximization to the realization that trading strategies themselves can be ramped into higher gear via computers and automation (see “Slice and dice”).
The term “quants” now is used less than the term “algo traders,” with “algo trading” also becoming more popular. These words have become catchall terms for sophisticated computer-based analysis and execution regardless of the method of analysis. Indeed, the line between fundamental and technical analysis seems like a throwback.
Now, consider this landscape from the perspective of the individual trader. Do the markets in general seem more volatile these days? It is bewildering to see how often a market or stock closes sharply higher one day then sharply lower the next (see “Going to extremes”). It would be much simpler to buy-and-hold your way to success. But the world is changing. Will you be prepared to detect and ride the next shift, whether it comes in the next minute, hour, day or week? You will be if you dig deeper into the concepts of algo trading. The algo genie is out of its bottle, and don’t expect it to go back inside anytime soon.
In the next part of this article, we will review some of the automated algo solutions available to independent traders and present an algo trading concept called net flow value.
Richard L. Muehlberg uses linear regression channels and intermarket analysis to day-trade his own account. He publishes a day-trading diary on his Web site www.DayTradingWithLinesInTheSky.com. E-mail him at firstname.lastname@example.org .