It's 'what' you trade that matters

You probably didn't know that the S&P 500 Index closed at an all-time high last Friday. You also were most likely unaware that the S&P had broken the 2000 level early this year and now approximates a lofty 2350. How can this be? The S&P 500 Index that I am describing is the un-weighted version of the index, tracked by Carl Swenlin in his excellent Decision Point site. Incredibly, the un-weighted version of the index has outperformed the traditional, weighted average by over 88% since November, 2000. Nor is the S&P the only index to display inferior performance as the result of weighting. The un-weighted NASDAQ 100 Index recently crossed the 4000 barrier, now standing at 4123--145% above its weighted equivalent.

One lesson I learned early in my investing and trading career, aided by well-known author and speculator Victor Niederhoffer--is that markets reward the assumption of risk. The seemingly safest strategies are rarely the ones that are best-rewarded: just look at the long-term rewards from stocks vs. bonds and cash, as beautifully detailed in the book Triumph of the Optimists by Elroy Dimson and colleagues. At a shorter term level, we can see how market returns following periods of out performance are dramatically lower than those following market weakness. (See my last article for one example of this phenomenon). The name stocks that represent the greatest weightings in the S&P 500 and NASDAQ 100 indices are the issues that most investors gravitate toward in their portfolios as safe blue chips. Yet these have significantly underperformed the lesser lights of those indices. Long-term charts reveal that they have underperformed the relatively anonymous universe of small and mid-cap issues as well. Since late 1999, the Dow Jones Industrial Average and S&P 500 weighted average of large cap stocks have lost ground, but the S&P Midcap and Small Cap Indexes have approximately doubled in value.

When you come to think of it, the decision to weight an index in one way or another--or the decision to weight it at all--merely reflects the preferences of those composing the indexes. The S&P 500 Index today is not at all similar to the index of the mid 1970s; many issues have entered and exited the universe, and sectors less weighted back then (finance) are more highly weighted now. Microsoft (MSFT | Quote | Chart | News | PowerRating) and Intel (INTC | Quote | Chart | News | PowerRating) are among the most actively traded and highly weighted of the stocks in the current index, but neither existed in the 1970s. In short, indices are the constructions of index makers, not timeless embodiments of Finance.

In spite of this elementary fact--and the fact that the most popular indices under perform their more unheralded counterparts--traders gravitate toward what is known. The e-Mini contract for the S&P 500 is consistently one of the most actively traded futures contracts, and the exchange-traded funds for the S&P 500 (SPY | Quote | Chart | News | PowerRating) and NASDAQ 100 (QQQQ | Quote | Chart | News | PowerRating) routinely head up the daily dollar volume lists among equities. This raises a fascinating question: Might traders and investors profit by constructing synthetic market indices that have the potential to outperform their favorite trading vehicles? If market indices are nothing more than constructions, why not construct our own to match our trading needs?

Clearly this is not a new idea. The major exchanges are always developing new products to meet the needs of traders. Most of these are designed to capture the equity performance of a particular country or market sector. In a recent talk I attended, Peter Steidlmayer--best known as a developer of the Market Profile graphic that arranges price and time in the shape of statistical distributions--encouraged listeners to boldly go where no index developers have gone before and create indices solely to maximize their tradability. Such an index might contain very different kinds of components, such as the U.S. Dollar, crude oil, 10-year Treasury Notes, and shares in the largest international companies. Constructing the index to meet the needs of daytraders might mean, for instance, weighting the components for their volatility--and then periodically re-weighting them so that the most volatile components always carry the most impact on the index price. The result would be a trading vehicle that has better trading properties than any of its individual elements.

Here's a practical illustration of the potential of synthetic indices. In my last article, I showed how the synthetic e-Conomy Index that I had created was a better predictor of future price changes in the QQQQ than the QQQQ itself. The synthetic index acted as a "lead fish" that guided the school of issues weighted in the NASDAQ 100. It turns out, however, that the e-Conomy Index is a better trading vehicle than the QQQQ in its own right. For instance, the previous article showed how the upper half of five-day returns in the QQQQ produce a worse next five-day return in the QQQQ (.26%) than the lower half of five-day returns (.36%). This is the classic pattern of strength following weakness and vice versa.

When we look at the e-Conomy Index since August, 2004, this pattern is greatly enhanced. The upper half of five-day returns in the e-Conomy Index return .57% in the next five days of the e-Conomy Index (93 occasions up, 60 down), but the lower half of five-day returns yield an eye-popping 2.23% over the next five days (121 occasions up, 33 down). If one wanted to trade reversal patterns following market weakness, the e-Conomy Index is simply a better vehicle than the QQQQ. It is not hard to imagine other synthetic indices that could outperform SPY and QQQQ with respect to reversing market strength--or reversing both strength and weakness.

The majority of trading advice in articles and books emphasize how traders should trade. Rarely do we see serious consideration to what traders should trade. The traditional trading vehicles are well-studied by the data miners; they are highly efficient as a result. Perhaps we will find that synthetic indices possess their own unique trading properties, opening the door to new sources of edge. That's my takeaway from Steidlmayer's work, and I think it's of critical significance to those of us in search of alpha.

Brett N. Steenbarger, Ph.D. is Associate Clinical Professor of Psychiatry and Behavioral Sciences at SUNY Upstate Medical University in Syracuse, NY and author of The Psychology of Trading (Wiley, 2003). As Director of Trader Development for Kingstree Trading, LLC in Chicago, he has mentored numerous professional traders and coordinated a training program for traders. An active trader of the stock indexes, Brett utilizes statistically-based pattern recognition for intraday trading. Brett does not offer commercial services to traders, but maintains an archive of articles and a trading blog at www.brettsteenbarger.com. He is currently writing a book on the topics of trader development and the enhancement of trader performance.