When prices move on news, an expert trader and an average trader have an equal chance at profits, since they have an equal chance at getting the news first. In fact, an average trader with above-average funding can improve his chances relative to the expert by subscribing to multiple news services. When one service reports breaking news, an average trader can respond with a quick buy or sell order (depending on the news), then take profits later by selling to or buying from lagging traders who get their news from a service that didn’t break the story that time. Price moves on news are external dynamics, and neither expert nor average traders can predict them (if they could, the “news” wouldn’t be news).
However, news doesn’t prompt most price moves. Trading prompts most price moves, and expert traders are experts because they have an intuition about where trading will take prices. Carnegie Mellon University’s FAST Lab Director, Dr. Sanjay Srivastava, has observed thousands of people learn how to trade. He reports that about one trader in fifty has this special intuition. Someone who’s never traded before will say “I’m going to hold on until the price reaches sixteen, and then sell at the peak.” The price will reach sixteen, the intuitive expert will sell, and the price will drop from there. There’s no rational reason for the price to peak at sixteen, but that’s what it does, and that intuition is what makes an expert trader. Every intuitive expert that Dr. Srivastava has helped place as a trader has earned at least US$100,000 in bonuses in his (or her) first year of trading.
Expert trader mental processes are a black box even to financial scientists such as Dr. Srivastava. However, their results are consistent and cannot be attributed to chance. Average traders also produce consistent results: eighty percent of all the new commodity futures trading accounts that open in any given year will close after going broke. Most new commodities exchange floor brokers also go broke. Since commodities trading is a zero- sum gain, all these losses go to the expert traders that survive the competition. Outside of futures contracts, trading is typically not a zero-sum game, but chief traders find that most of their hires cannot make enough profits to justify their capital allocations. The lucky chiefs hire on Dr. Srivastava’s recommendations.
Artificial neural nets (those that run on digital computers, as opposed to the real neural nets human traders are endowed with) are also black boxes, but they too have produced surprisingly consistent results. One of my students, Lic. Rosana Huerta, reports that forty out of forty Neuralstock(tm) nets she trained gave her sell recommendations on the same day just before an important market peak.
Like human expert traders, artificial neural nets predict price moves based on trading, not news. Their ability to do so sheds some light on human trader abilities, in the sense that scientists know neural net results stem from pattern recognition. The trading community in a given instrument can generate an internal market dynamic whose pattern an expert trader, human or artificial, extrapolates. Extrapolating lets the expert time her (or its) purchases and sales to profit from the price pattern.
Of course, when excited trading excites the trading community and traders generate patterns with big price moves, the expert traders in that community have their biggest opportunities. When a price bubble is big enough, even average traders will recognize it, and while some will stay away out of caution, others will jump in to extract a little bit of profit while the momentum is high. If their timing is right, they’ll compete with expert traders to establish a position in the stock, driving prices up even faster. If their timing is wrong, they’ll take over the positions that expert traders are abandoning. Either way, the expert traders win. Next week this series on matching expert traders to their biggest opportunities will continue.
Readers with questions or comments for Dr. White can call
011(525)595-6045, fax 011(525)683-5874, or email
white@profmexis.sar.net
Return to Mark White's Home Page