Objective research and reviews to aid investing decisions

Blog RSS Feed:



Guru Grades Guru Grades



Blog - Investing Notes

June 14, 2007 - Testing the QQQQ Crash Trade Trigger

Reader David Zaitzeff inquired about the usefulness of James Altucher's QQQQ Crash Trade Triggers, which signals:

BUY PowerShares QQQ (QQQQ) when the daily close is more than 1.5 standard deviations below the average low over the past ten (trading) days, and

SELL at the next close that is higher than the BUY price or at the close after 20 (trading) days, whichever comes first.

In testing this scheme, we assume that any BUY must clear with a SELL before executing another BUY (multiple concurrent positions not allowed). In other words, our mechanical investor is all in or all out of the market. We assume that all BUYs and SELLs occur at closing prices. We also assume round trip trading costs of $20 and an initial stake of $100,000 to calculate trade returns and cumulative returns. Using QQQQ daily low and closing prices for 3/10/99 (the earliest available) through 6/13/07, we find that:

The following table summarizes the results of the QQQQ Crash Trade Trigger scheme. Over the entire sample period, the scheme generates 66 trades with an impressive 97% positive return rate and an average return of 1.9% per trade. The average holding period for these trades is just a few trading days. The cumulative return across 8+ years is +240%, compared to -6% for buying and holding QQQQ.

However, the Crash Trade Trigger scheme achieves most of its results during the 2000-2002 bear market, during which overall market volatility is high. Its average trade return falls off for 2003-2007. As a sole trading policy (ignoring interest on cash while out of the market), it underperforms QQQQ buy-and-hold in four of the past five years (2007 is only partial). Across the entire sample, the scheme underperforms QQQQ buy-and-hold in five of nine years (1999 and 2007 are only partial). As would be expected, it seems to generate fewer buying signals during a bull market.

A test across multiple bull and bear cycles is needed to decide with confidence whether the QQQQ Crash Trade Trigger scheme beats buy-and-hold as a sole trading policy over the long haul.

Note that assuming a larger (smaller) initial stake would increase (decrease) the average trade returns and cumulative returns because trading costs would represent a lower (higher) percentage of the account. Note also that the QQQQ Crash Trade Trigger scheme produces all short-term capital gains, putting it at a disadvantage compared to buy-and-hold for taxable accounts.

In summary, the QQQQ Crash Trade Trigger scheme may work well in bear markets, but not so much in bull markets. Beware of small samples and favorable starting points.

For related research, see Blog Synthesis: Some Trading Indicators.

Disclaimer | Contact CXO
Design by Cavendo: Virginia Web Design Company and Search Engine Optimization
© 2004-2008 CXO Advisory Group LLC. All Rights Reserved.