The two examples this time come from specialized investment newsletters that I receive. I am not going to specify the particular ones because I believe they are far from unique in this regard; they are just the ones I happen to see. It is natural to try to present the results of one’s advice and writing in the best light. In the current down markets doing so may require quite a bit of creativity. Over the past few years political pundits have used the term "spin" to describe politicians' attempts to make things look better than they really are. I believe that term applies equally well to many investment pundits and writers.

The first example comes from a newsletter that tests, analyzes, and develops formulas and models for trading stocks, mutual funds, and futures. To its credit, this newsletter revisits models from past issues a few years later. Normally, the reviews attempt to show that the model has been effective since the date of publication. The typical report shows a chart of some stock market index with arrows superimposed. The upward-pointing arrows indicate the buy signals and the down arrows show the sells. This technique is used by quite a few writers and system sellers and promoters. Of course, the desired impression is that the up arrows look as if they are near low points on the graph and the down arrows are near high points.

The first problem is the nature of the graph.(1) The particular example I am looking at shows a ten-year period, and many of the buy and sell signals are close to each other. That makes it quite difficult to tell how effective the signals were. It would be much better to present a table showing the signal dates and how the index changed between them. To their credit, many writers do this.

However, that is not my real objection in this specific case. Instead of using a market index in the graph, the author shows the weekly advance-decline line of the New York Stock Exchange, which is sometimes considered to be a measure of how stocks are doing in general. Many technical analysts use the trends of that line to make judgements about the likely direction of the stock market. That is all well and good, but the main problem is that there is no practical way to trade the advance-decline line. So even if the model’s signals are good for predicting the direction of the advance-decline line, that does not tell us whether the model would have been useful if we tried to use it to trade stocks or a market index such as the Dow, S&P 500, or Nasdaq 100, all of which can be traded easily. It makes me suspect that the model's signals applied to these indices would not have worked very well in the past two or three years.

The second example, from a newsletter that makes specific recommendations for trading mutual funds, might have made George Orwell, the author of 1984 (in which good is bad and black is white), proud. Despite the fact that its methods had shown substantial losses over the prior twelve months, the newsletter says "on a relative (emphasis in the original) basis our recent performance has been exceptional." The key word is, of course, "relative." In this case, relative to the Nasdaq Composite index. The newsletter goes on to say its "relative advantages over the Nasdaq for the past twelve months are among the best ever" and "on a relative basis, our long-term performance record is the best it has ever been when compared to our benchmarks."

To give credit where it is due, the newsletter makes clear the distinction between its absolute performance (sizeable losses over the past twelve months that are "among the worst ever" and "leave much to be desired") and performance relative to the Nasdaq Composite, which has been the stated benchmark all along. It also provides annual returns of its trading strategies, which are real-time I believe, for the past ten years, so one can compare them to the Nasdaq annual returns by researching the latter. However, the thrust of the article, in my opinion, is that the relative performance was outstanding.

Am I overreacting to all the crowing about great relative performance? Perhaps, but I don’t think so. I believe this is a fine example of "spin." In some prior years, notably 1998 and 1999, this newsletter’s methods made some nice profits, but trailed the gains in the Nasdaq by a considerable amount. In those years, there was not much emphasis on or discussion of relative performance, which was quite poor by the standards used in the recent issue.

Since the objective is to make money over the longer term, not lose less than someone else, I don’t think it is appropriate to emphasize performance relative to a measure that has gone down quite a bit when one has suffered large losses himself. One way such comparisons may turn up is in so-called "risk-adjusted"performance. This can be a valid and useful measure, but improperly or deceptively done in order to provide spin, it can also be an example of Wall Street gibberish. As always, evaluate what you hear and read about investing, including what I say and write, critically.


  1. For an excellent and delightful exposition of the right and wrong ways to present information graphically, see the book The Visual Display of Quantitative Information by Edward Tufte. His two related books, Envisioning Information and Visual Explanations, are also well worthwhile. They are published by Graphics Press, P.O. Box 430, Cheshire CT 06410, 800/822-2454. On-line sellers such as Amazon.com have them, and the books and additional information are also available from Tufte’s web site: "www.edwardtufte.com".

Back to previous page

Return to the Investing Page

Return to the Home Page