From the August 01, 2007 issue of Futures Magazine • Subscribe!

Trading London and European hours


These examples show how, even in today’s supposedly liquid active marketplace, the domino effect and potential systemic risk trips one order into another forcing a major move. Although these appear to be aberrations, they can easily become a meltdown due to markets being linked via algorithmic auto-traded systems.

In the 1980s and early 1990s artificial intelligence (AI) became a fashionable following, as personal computers made even more significant inroads into market analysis. Multiple millions of dollars were spent on super computers to investigate whether markets were random walk or chaos or even fractal-like in their repetition.

However as the markets ultimately showed, this rendition of AI was no better at predicting price moves than centuries-old statistics, at best akin to trading using a simple linear best-fit regression line and at worst akin to trading highly optimized and curve-fit models.

Modern markets are, in effect, data warehouses. Programmers input known reference points (opens, highs, lows, closes) and automatic black-box systems generate buy and sell orders based on current market activity complying (or not) with past moves.

What has changed in the past 30-odd years is that now, instead of weekly charts, this analysis occurs in milliseconds and it does so automatically. Introduce the various types of momentum indicators, Fibonacci extensions, moving averages, the six or so standard candlestick formations, seasonality, Commitments of Traders report analysis and Elliott wave counts and there are a multitude of strategies waiting to be executed.

Of course, there is the argument that that these schools of thought are all different, and therefore affect the market in different ways at different times, and that is true most of the time. In the context of the liquidity mirage, however, they are all slaves to one occurrence: the big move.

What has become noticeable as technology has advanced in recent years is that markets are less likely to let you overstay your welcome. They tend to whoosh through several key price levels as algorithms roll though their various orders. Many charts show series of 10- to 14-day advances or declines, or even weeks for that matter, without the normal ebb and flow of two-way short-term (floor traders) vs. long-term speculators or hedgers.

In other words, the algorithms are taking over; and while — as in the S&P market condition in last October — they might not have a significant impact in the big picture, they are easily capable of disrupting retail accounts in terms of both risk and opportunity. Today, if you are not on board a move just before it starts, it is difficult to get involved afterward. And on the other side of the trade, afterward is too late to protect the bottom line.

<< Page 2 of 3 >>
comments powered by Disqus
Check out Futures Magazine - Polls on LockerDome on LockerDome