Security Analysis : Sixth Edition, Foreword by Warren Buffett: Additional Aspects of Security Analysis. Discrepencies Between Price and Value

positive argument could be made in favor of “technical study.” But suchprecise signals seem to occur only at wide intervals, and in the meantime human impatience plus the exigencies of the chart reader’s profession impel him to draw more frequent conclusions from less convincing data.
    4.
Other Theoretical and Practical Weaknesses
. The appeal of chart reading to the stock-market trader is something like that of a patent medicine to an incurable invalid. The stock speculator does suffer, in fact, from a well-nigh incurable ailment. The cure he seeks, however, is not abstinence from speculation but profits. Despite all experience, he persuades himself that these can be made and retained; he grasps greedily and uncritically at every plausible means to this end.
    The plausibility of chart reading, in our opinion, derives largely from its insistence on the sound gambling maxim that losses should be cut short and profits allowed to run. This principle usually prevents sudden large losses, and at times it permits a large profit to be taken. The results are likely to be better, therefore, than those produced by the haphazard following of “market tips.” Traders, noticing this advantage, are certain that by developing the technique of chart reading farther they will so increase its reliability as to assure themselves continued profits.
    But in this conclusion there lurks a double fallacy. Many players at roulette follow a similar system, which limits their losses at any one session and permits them at times to realize a substantial gain. But in the end they always find that the aggregate of small losses exceeds the few large profits. (This must be so, since the mathematical odds against them are inexorable over a period of time.) The same is true of the stock trader, who will find that the expense of trading weights the dice heavily against him. A second difficulty is that, as the methods of chart reading gain in popularity, the amount of the loss taken in unprofitable trades tends to increase and the profits also tend to diminish. For as more and more people, following the same system, receive the signal to buy at about the same time, the result of this competitive buying must be that a higher average price is paid by the group. Conversely, when this larger group decides to sell out at the same time, either to cut short a loss or to protect a profit, the effect must again be that a lower average price is received. (The growth in the use of “stop-loss orders,” formerly a helpful technical device of the trader, had this very effect of detracting greatly from their value as a protective measure.)
    The more intelligent chart students recognize these theoretical weaknesses, we believe, and take the view that market forecasting is an
art
thatrequires talent, judgment, intuition and other personal qualities. They admit that no rules of procedure can be laid down, the automatic following of which will insure success. Hence the widespread tendency in Wall Street circles towards a composite or eclectic approach, in which a very thorough study of the market’s performance is projected against the general economic background, and the whole is subjected to the appraisal of experienced judgment.
    The Second Type of Mechanical Forecasting. Before considering the significance of this injection of the judgment factor, let us pass on to the other type of mechanical forecasting, which is based upon factors outside of the market itself. As far as the general market is concerned, the usual procedure is to construct indices representing various economic factors,
e.g.
, money rates, carloadings, steel production, and to deduce impending changes in the market from an observation of a recent change in these indices. 3 One of the earliest methods of the kind, and a very simple one, was based upon the percentage of blast furnaces in operation.
    3 These indices may also be plotted on charts, in which case the forecasting takes on the aspect of chart reading.
Examples
: The A, B, and C lines of the
Harvard Economic Service
which were published in weekly letters from Jan. 3, 1922, to Dec. 26, 1931 (since continued through 1939 at less frequent intervals in
The Review of Economic Statistics
); also the single composite Index Line in the “Investment Timing Service” offered by Independence Fund of North America, Inc., in 1939.
    This theory was developed by Col. Leonard P. Ayres of the Cleveland Trust Company and ran to the effect that