To print article, click Print on your browser's File menu. Go back Posted 12/14/2000	sponsored by: The Speculator Turn a profit when the sky is falling Here's how to turn fears that the world is ending to your advantage. When volatility is high, investors run scared, and it’s a good time to buy. When it is low, investors get complacent, and it’s a good time to sell. By Victor Niederhoffer and Laurel Kenner If a serious question has been raised, whether it be in science or society (or stocks), then it is not enough merely to assert an answer. Statistics must be put on the table. -- Steve Stigler, "Statistics on the Table" Today, we are going to introduce a swing-trading system that has a reasonable chance of catching major moves in the broad market. But first, we'll print some of the withering responses to our article last week, which criticized the predictive value of candlestick charting. Money Plus. Easy online tools and free Bill Pay, too. While last Thursday's column drew equal amounts of applause and boos, the boos were particularly vivid. This letter from reader Andrea Kruse was representative: "I think it is horrible you are discouraging people to not use a system that could help them save or make money just because you do not know how to use it (and probably are too lazy to take the time to really learn how to use it)." And from a reader who identified himself as "GO3D:" "I'm so glad there are shortsighted, uneducated people like yourselves that write such nonsense. Keep up the bad work…it just makes it easier for me to make money from all of the 'fundamentalists' who will never make a dime in the market. No reply required." We have no doubt that Ms. Kruse, "GO3D" and our other critical correspondents are all fine practitioners of technical analysis, far superior at this exercise than we are, and that they have found the technique quite profitable for themselves. The big questions: Will they continue to find it so? Has their success been due to skill or luck? If the latter, to what extent would they have been successful if they had used a technique other than charting? The problem with patterns Richard Brealey, a past president of the European Finance Association and author of a standard university textbook on corporate finance, once tried to evaluate technical-analysis from a scientific point of view by comparing actual stock charts to charts whose data points were generated randomly. He reported in a finance journal that the charts looked so similar that professionals were unable to tell them apart.
While markets may not be completely efficient, they're pretty close, so any method that purports to find inefficiencies had better be good.	We are the first to admit that many variants of technical analysis serve a useful and profitable function, particularly when employed as an aid to visualizing where a stock has been. The problem is whether to assign patterns -- whether they be “head-and-shoulders tops,” “double bottoms” or “gravestone doji,” in the jargon -- some predictive power. Certainly, the degree of analytical rigor is not the key -- or else the world’s smartest academics would be considerably richer than they are. Unfortunately, those who have studied the subject with a reasonable degree of objectivity are fairly unanimous in concluding that many of the conventional techniques of technical prediction are not of great merit. Here's what Zvi Bodie's standard text, "Essentials of Investment", has to say: "It should be abundantly clear from our presentations that most of technical analysis is based on ideas totally at odds with the foundations of the efficient market hypothesis." While markets may not be completely efficient, they're pretty close, so any method that purports to find inefficiencies had better be good. We also believe it likely that for each technical analyst willing to share his or her successes, there are hundreds of others who have failed. Fortunately, one of them, the engineer and trader John Lamberg, was kind to enough to write us about his experience with the $60 coffee-table book by Steve Nison, "Japanese Candlestick Charting Techniques: A Contemporary Guide to the Ancient Investment Techniques of the Far East." "I have to admit I purchased a copy of Nison's book a few years ago. The ensuing losses were a sight to behold. Confidence lost was an even heavier toll. If others can profit from these techniques, then they have mastered what I can not." Chart challengers And Christopher Guida, who recommended Nison's book to us in the first place, had a confession to make: "My experience was not just similar -- it was exactly the same. I bought the book about four or five years ago, and was very revved-up over it. So, TC200 disk in one hand, and Nison's book in the other, the fool rushed in, and the bleeding commenced." There's a psychological, game theory aspect to the question, as brought out in a letter from reader Glenn Escovedo, who sent us a couple of samples from one of his favorite books, Harro von Senger's "The Book of Strategems": Fool the emperor and cross the sea. Trick the emperor into sailing across the sea by inviting him into a seaside house which is, in reality, a camouflaged ship. (Conceal the real objective; disguise the real course.) Openly repair the walkway, secretly march to Chencang. Hide your sophisticated intent behind an apparently innocuous action. "The Chinese have lived and died by deception for thousands of years," observed Mr. Escovedo. "Hard to believe they or the Japanese would take charts at face value." Yet these experiences must be balanced by the many able and wealthy practitioners we all know who swear that technical analysis, as they use it, is quite profitable. We believe that technical analysis works, if properly done. In fact, we have suggested viable methods that we have tested, both in our articles on MSN MoneyCentral and in the body of Vic's works over the past 40 years. Our own trading also relies on patterns based on opens, closes, highs and lows. The difference is that we vary our reliance based on past performance. And most important of all, we measure what it is that triggers a trade, starting with its exact defining characteristic and ending with an estimate of the expectation, risk and uncertainty of the trade and how this varies with respect to duration. The value of volatility Now, to the basic question that our editor is always asking us: How do you make money in the market today? Observe that the Nasdaq ($COMPX) rallied 22% in the seven days following Nov. 30, when it looked like the world was ending. Now all the bulls are coming out of the corral, led by Judas goats who have been bearish since Nasdaq 400 and now see a rally of 25% in the cards. Here's a technique to round up some short-term profits next time it looks like the end of the world has come. It is based on buying good stocks in the market when the implied volatility in the market is high and selling when it is low. The idea is that when volatility is high, investors tend to be frightened, and it's a good time to buy. When it is low, investors tend to be complacent, and it's a good time to sell. As a proxy for volatility, we use the Market Volatility Index ($VIX.X), the average of implied volatilities of short-term, at-the-money index options on the Chicago Board Options Exchange. We looked back to the beginning of the 1990s, and decided to start with Oct. 27, 1997, the first time in several years that the Volatility Index topped 30. We found that buying the S&P 500 ($INX) at the Volatility Index's first close above 30 and selling at its first close below 25 would have resulted in an average move of +3.7%. There were nine completed trades, with an average holding period of one month, and one loss. Trading General Electric (GE, news, msgs) on those dates produced an average move of up 7.6%, with no losses. Trading Intel (INTC, news, msgs) had an average move of up 2.7%, with three losing trades (the worst was down 14%). If a trader waited to sell until the Volatility Index closed below 20, there were only five round trips, with no losses. The average holding period was about five months. The average moves were S&P 500, up 15%; GE, up 21%; Intel, up 20%. VIX Swing System     S&P 500     GE     INTC       Date VIX Buy Sell % Chg Buy Sell % Chg Buy Sell % Chg # of days 10/27/1997 39.96 876.99     20.64     18.69       12/1/1997 24.88   974.77 11.1%   24.67 19.5%   20.38 9.0% 35.00 12/24/1997 30.47 932.70   23.69   17.56     12/30/1997 24.93   970.84 4.1%   24.79 4.7%   17.92 2.0% 6.00 1/9/1998 34.46 927.69     24.14     17.97       1/14/1998 24.21   927.94 0.0%   24.79 2.7%   18.86 5.0% 5.00 8/4/1998 33.1 1072.12     28.48     20.48       11/5/1998 24.8   1133.85 5.8%   30.27 6.3%   23.50 14.7% 93.00 12/14/1998 32.47 1141.20     28.89     27.89       12/18/1998 24.66   1188.03 4.1%   32.33 11.9%   30.00 7.6% 4.00 1/13/1999 31.26 1234.40     32.19     34.75       3/8/1999 24.98   1282.73 3.9%   34.94 8.5%   29.91 -13.9% 54.00 9/23/1999 30.28 1280.41     38.92     38.75       10/6/1999 23.53   1325.40 3.5%   41.23 5.9%   38.47 -0.7% 13.00 10/15/1999 31.48 1247.41     38.54     35.44       10/21/1999 24.77   1283.61 2.9%   41.14 6.7%   35.84 1.1% 6.00 4/5/2000 30.59 1487.37     51.06     64.94       6/1/2000 24.74   1448.81 -2.6%   52.25 2.3%   64.84 -0.1% 57.00 10/11/2000 30.95 1364.59   56.56     35.38       Average       3.7%     7.6%     2.7% 30.3 Buy at first close above 30 on the VIX; sell at first close below 25. The "Buy Above 30, Sell Below 25" system shows nice results for the S&P 500 and General Electric. Speaking broadly, for the S&P 500 the move during the time that you were long would have accounted for the entire rise in the averages from 875 to 1,400 during the period. Yet you were only in the market one-half of the days.
Recent Articles • Why candlestick charts don't work, 12/7/00 • A swingin' Gestalt theory to rock your stocks, 11/30/00 • Place a bet on strategic long-shot investing, 11/16/00 more...	However, for important questions like this, it is imperative, in the words of Steve Stigler, to put the statistics on the table. "Evidence must be provided, and that evidence should be accompanied by an assessment of its own reliability….Methods for measuring and expressing the uncertainty in the conclusion and conventions for settling issues such as how much uncertainty is too much are also necessary." Regrettably, caution is appropriate here, because the uncertainty surrounding the superior performance of the system here is too much for us to clearly differentiate it from randomness. The one-point-a-day edge you get from your longs is statistically small relative to the 15-point average variability of daily moves in the S&P 500. May we suggest, however, that if there are any words in the English language for a degree of uncertainty 500 times greater in magnitude than caution, they would be appropriate for those who rely on charts, candlesticks and all the other associated mumbo-jumbo of technical analysis without putting their statistics on the table. As always, we are humble as to our own wisdom and will respond to all reader critiques, even those that hone in our weaknesses as saliently as those recorded. At the time of publication, the authors owned no stocks or funds mentioned in this article. MSN Money's editorial goal is to provide a forum for personal finance and investment ideas. Our articles, columns, message board posts and other features should not be construed as investment advice, nor does their appearance imply an endorsement by Microsoft of any specific security or trading strategy. An investor's best course of action must be based on individual circumstances.