by Ben Best
This article is a reflection of my specialized knowledge & interests in the use of money to make (or lose) money. I am not attempting or claiming to be encyclopedic. I express only knowledge & opinions of practices with which I am familiar -- and I describe what I have learned from my own experience. Investing & trading can involve facts, interpretations and intuition -- art & science -- as well as emotional involvement (fear, greed, joy, horror, caution, impatience and gambler's excitement).
Literally speaking investing refers to expenditures made for the purpose of making a profit, whereas trading refers to the process of buying & selling. But I use these words in the sense that stresses the connotation that investing refers to longer-term (months & years) buying & selling of securities, whereas trading refers to a more short-term (minutes to weeks) focus. But neither long-term nor short-term transactions should be conducted without cognizance of the context provided by the other.
Investing & trading can be done on the basis of fundamental or technical analysis (preferably both). Fundamental analysis dictates strategies based on underlying political & economic conditions and on the conditions of specific companies or governments. Technical analysis guides strategies based on charts & statistics. The "fundamentals" underlying technical analysis are mostly reflections of group psychology. Fundamental analysis should dominate long-term investing, whereas technical analysis is more important for short-term trading (but still not necessarily more important than fundamentals).
I have at times invested in precious metals, real estate, currencies, foreign indexes, stocks, bonds, options and futures, but have become more focused. I have a retirement investment portfolio that is mostly index-funds or fixed-income securities (bonds or T-bills). My long-term/intermediate-term portfolio can contain stocks, bonds, index-funds, T-bills and (occasionally) options. For short-term trading I use the futures market for indexes: S&P 500, DOW and NASDAQ 100.
(return to contents)
I believe that the most fundamental truth about the stock market is that the long-term trend is upward. Even before the bull market of the 1990s, annualized real returns (ie, above inflation) for common stocks in the United States was nearly 9% for the 62-year period following 1926. In the same period long-term & intermediate-term government bonds yielded 2% above inflation, while T-Bills did no better than match inflation. The reason for the high returns on common stock is that common stock represents the assets of businesses -- the creators of material wealth. The progress of civilization is associated with advances in technology, capital and managerial prowess that leads to growth of the productive assets of business.
Within the context of long-term economic growth ("secular" economic growth) there is a business cycle and the impact of political events such as wars. The bull market of the 1990s may have been due in large part to the "peace dividend" of reduced defense spending associated with the end of the Cold War.
The business cycle is not a level sine-wave, but is a wave within the upward-sloping long-term trend of economic growth. I believe that in a free market economy there would be no business cycle. I subscribe to the Austrian Theory of Business Cycles (link my article on the subject), which attributes periodic booms & busts to the control of money & to the manipulation of interest rates by government central bankers.
Artificially low interest rates & expansion of money supply leads to an inflationary boom. When central banks then artificially raise interest rates to fight the inflation, the stock market slumps and a recession follows -- causing the central bankers to artificially lower interest rates again to "stimulate the economy". Because the stock market anticipates economic conditions, stock market boom & bust generally precedes economic prosperity or recession by a period of 6 to 9 months. In cognizance of this, a prudent investor will invest in equities when the central bank is near a bottom in cutting interest rates -- and shift money from equities to bonds when the central bank is near the top of its interest rate tightening cycle. Bonds are especially profitable when interest rates are being cut, because not only do they provide shelter from stock market losses, but their price increases as the yield/interest-rate drops. But when interest rates stop dropping -- and are low -- the income from bonds will be poor (time to buy equities). Bonds are especially profitable when interest rates are being cut, because not only do they provide shelter from stock market losses, but their price increases as the yield (interest rate) drops. But when the recession is ending and interest rates stop dropping (and are low) the income from bonds will be poor (time to buy stocks). (An exception to this is so-called "junk bonds" which drop significantly in yield as economic conditions improve -- resulting in capital gain from price increases.)
According to the efficient market hypothesis neither fundamental analysis nor technical analysis can provide a strategy to "time" the market (ie, anticipate price-movements in real-time). The hypothesis claims that markets are so efficient that all information is already priced into the market -- with the consequence that the analyst randomly makes & loses money, but can never "beat" the market. Those who subscribe to the "random walk" or "efficient market" view of stock prices generally advocate a "buy & hold" approach to investing -- implicitly accepting the fundamental conclusion that the long-term trend is upward. Since stock-picking is considered futile, index-funds of the market as a whole is preferred. Carried to its logical conclusion the efficient markets hypothesis implies that there can never be such a thing as a business opportunity. This is patently false. Good entrepreneurs are superior forecasters who can earn superior return.
Many fundamental analysts use price/earnings (P/E) ratios as indicators of the value of individual stocks and to evaluate the market as a whole based on the summed P/E of stock indices. These so-called value investors are contrasted with growth investors who are more concerned with potential future sales of products than with current earnings. (For more on evaluating investments on the basis of earnings, see the price-to-earnings ratio section of my article Financial Statements in the "New Economy".)
Some analysts are chronic bulls (always predicting that the market will move up) and others are chronic bears (always predicting that the market will move down). A stopped clock is correct twice a day, but a chronic bull's predictions are correct more often than not. Chronic bears can also count on periodic attention & praise for their "foresight" -- when markets are down -- but this happens less often.
(return to contents)
Charles Dow could easily be called the "father of technical analysis". Dow's purpose in creating an index was to identify trends that could be used to guide investment. He used only closing prices to eliminate the effects of day-trading. In 1897 Dow created an 12-stock industrial average index separate from his 20-stock railroad index. Dow believed that for a trend to be meaningful the trend of the index of the companies that produce goods must be confirmed by the index of the companies that transport them.
As first editor of THE WALL STREET JOURNAL Charles Dow was able to expound his theories in that newspaper from the time it began publication in 1889 until the year of his death in 1902. Although Dow primarily charted stock prices to identify trends, he observed that trading volume normally expands on rallies and contracts on declines. He observed that contracting volume on a price advance can indicate that an upward trend is about to turn down. By one estimate, a simple application of Dow Theory to identify and invest in trends in the 90 years subsequent to the establishment of Dow's two indices would have resulted in 20 times the return of a "buy-and-hold" strategy. (For details concerning the Dow Jones Industrial Average and other American stock indices -- see my article The Major American Equity Indices.)
(return to contents)
Chartists are technicians who look for trend indicators in graphical charts, usually depicting price as a function of time, but often including information on intraday highs & lows as well as closing price. A chartist typically will want to buy into an uptrend, sell into a downtrend and stay out of the market for a sideways trend ("momentum investing"). Aside from price, the other variables most frequently charted are volume and breadth. Put/call ratio in the options market and open interest (number of unliquidated contracts at the end of the day) in the futures market can also be important indicators.
Volume refers to number or dollar value of transactions. Volume reinforces price trend signals when it is increasing and may signal an impending trend reversal when it is decreasing.
Breadth usually is measured by dividing the number of stocks advancing in price by the number of declining stocks (advance/decline). Toward the end of a bull market breadth typically declines while price continues to trend upward -- with blue-chip companies being the last to fall. Conversely, days of extremely high breadth are said to signal beginning of a bull market.
The Arms Index (devised by Richard Arms) is a breadth/volume indicator primarily used by day-traders. Also known as TRIN (short-term TRading INdex), the Arms Index is a ratio of a ratio:
Advancing Issues / Declining Issues ----------------------------------- Advancing Volume / Declining Volume
Price trends are usually drawn as straight lines, with the slope of the line sometimes described as the momentum. A price above-which stock or index seems to resist rising is called resistance, whereas a price below which a stock or index resists falling is called support. Some chartists search for support & resistance levels in their charts nearly as diligently as they look for trendlines.
Common patterns looked-for by chartists are triangles & rectangles that are wider than they are high. In the case of rectangles, after rising or falling significantly, a price may move up & down (oscillate) within a rectangular band (between support & resistance lines) for a while. Or, more significantly, the price may oscillate within a band of decreasing oscillations, forming a triangle. Suddenly, the price will increase or decrease, a breakout. When breakout from resistance or support occurs, trading volume increases and there tends to be a sharp movement of price in the direction of the breakout.
There are technical & psychological reasons why such patterns of trading occur. A technical reason for an increase in volume is that many trader use stops, ie, orders to automatically sell when a price falls to a certain level (when long) or to automatically buy when the price rises to a certain level (when short). (A long position is the more usual contract that makes money when price rises, whereas a short position is a contract that makes money when prices fall). It follows that a large upward or downward movement will trigger more stops than a small upward or downward movement -- hence volume increases.
Psychologically, a rectangle or triangle, especially when quite long in the sideways direction, is a trend that is unlikely to cause many traders to change their positions. A breakout, on the other hand, is a new trend that excites trader interest to "jump on the bandwagon" and reinforce the new movement. Moreover, if a large number of chartists are looking for the same patterns and trade on the basis of those patterns, the chart patterns become a self-fulfilling prophecy. This is particularly true during intraday trading where crowd psychology can be as important as fundamentals. (In the long run economic fundamentals are far more important.)
The most universally accepted of all chart patterns is the so-called head-and-shoulders pattern, consisting of a high (the head) between two lesser highs (the shoulders). The heaviest volume occurs on the first rally (the left shoulder), with lighter volume on the head and very light volume on the right shoulder. The final rally is a signal that prices are soon to go much lower -- reversing the bull-trend and establishing a bear-trend.
(return to contents)
Chartists don't always look for straight lines in charts, they may look for curves or, most commonly, cycles. One of the first and most famous of cycle chartists was the Russian economist Nikolai Kondratieff, who in 1926 delivered a paper based on prices & outputs going back to the 1780s. Kondratieff found a regular cycle lasting 50-54 years, a cycle also identified by English economist William Stanley Jevons -- as well as other economists who studied longer periods. Since Kondratieff's theory did not support the Marxist dogma of the collapse of capitalism, Kondratieff was removed as head of Moscow's Business Conditions Institute and sent to Siberia. Subsequent to his "discovery" there was a 70 year period without a worldwide depression, discrediting his theory.
A noteworthy cycle theorist in the United States was
Edward Dewey, who founded the Foundation for the Study of Cycles in
1940. Cycles of about 9-years were found for ozone levels,
tent caterpillars and wholesale prices. An 18-year real estate cycle was
identified. Industrial production was correlated with the 11-year sunspot
Among chartists, the most closely studied theoretician is Ralph Nelson (R.N.) Elliot, author of NATURE'S LAW -- THE SECRET OF THE UNIVERSE (1946). Applied to stock indices, Elliot Wave Theory predicts that prices follow a rhythm of a 5-wave advance (3-up & 2-down movements) followed by a 3-wave decline (2-down movements separated by one up movement). ("Wave" means a straight-line component a cycle, in Elliot's terminology.) As seen in the diagram, position (1) corresponds to 5 waves, position (2) to 8 waves, position (3) to 13 waves, position circle-1 to 21 waves and position circle-2 to 34 waves. To Elliot, the correspondence of these numbers to the Fibonacci series (0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ... -- where each number is the sum of the preceding two numbers) is no accident, but a fundamental law of the universe. The Great Pyramid of Giza had 3 edges visible from any one side, 5 surfaces, 8 edges and a total of 13 surfaces & edges. The division of each number in the Fibonacci series by the following number converges to a value of 0.618 with increasing numbers. The ratio of the elevation to the base of the Great Pyramid of Giza is 0.618 (about 62%). The most commonly used numbers for percentage retracements of price movements are 62%, 38% (100% - 62%) and 50%.
Like all forms of visual chartism, Elliot wave analysis involves
at least as much "art" (subjectivity) as "science"
(among those who take it seriously -- not me). Practitioners differ greatly
in their interpretations of what constitutes a wave and what is
"noise". The charge that visual chartism is largely a subjective
self-delusion does not explain why it continues to attract so many supporters
in reputable financial institutions. If chartism is entirely subjective, it
cannot be a self-fulfilling prophecy of crowd psychology (many people
responding the same to the same signals).
(return to contents)
In order to escape the subjectivity of visual chartism, many technical analysts have applied mathematical, statistical and computerized methods to determining price patterns. Linear regression analysis is used to determine trends. Trigonometric curve fitting or Fourier analysis is used to identify cycles. Mechanical systems are validated on historical data and then used to generate buy and sell signals on real-time data -- eliminating human subjectivity & emotion. In the late 1980s and early 1990s TARGET="_new"neural networks were frequently "trained" on market data, but developers failed to develop profitable systems and became disillusioned with this approach.
Probably the most widely used mechanical indicator is the moving average, a method of smoothing data. In a 20-day moving average, the prices of the last 20 days are added and divided by 20. On the following day the oldest price is dropped and the newest price is added to the 20 days included in the average. Comparing a 200-day moving average with a 20-day moving average, a "buy" signal is generated when the 20-day curve crossed the 200-day curve from below. A "sell" signal is generated when the 20-day curve crosses the 200-day curve moving downward. In some systems more recent days are weighted more heavily (exponentially, in some cases) in the average than less recent days. Systems also differ in the time periods used in the short-term and longer-term averages (50-day is very common).
There is not limit to the number of ways by which to manipulate data to create indicators. Other examples include relative strength indicators, stochastic indicators, etc. Because I have little interest in such techniques I will say no more about them.
Although technicians believe that trends need to be identified & followed, they also believe that crowd psychology leads to oscillations about the trend due to panic overselling and enthusiastic overbuying. In contrast to trend indicators, oscillators are intended as indicators of overbuying & overselling -- contrarian indicators to sell when others have bought and to buy when others have sold. One such oscillator would be Bollinger Bands, which trace curves one or two standard deviations on either side of a moving average. Current prices crossing an upper Bollinger Band to the upside would be a "sell" signal, whereas crossing a lower Bollinger Band moving downward would be a "buy" signal.
(For more detailed online explanations of technicals such as oscillators, moving averages, candlesticks, etc. go to Technical Analysis 101)
(return to contents)
Seasonal patterns have long been noted for the stock market. December & January are usually described as the best months for stocks, followed by April & November. August & September are the worst months. March is typically good in the beginning and bad in the second half. October is the most volatile month.
Many such generalizations have been generated from many studies, covering quite different time periods. The reader is warned what a difference the choice of period can make. Over the century September has been the worst month for stocks, but since 1992 August has been the worst month. Eliminating 1987 from studies makes a great deal of difference in the average returns for October. The generalizations I give all apply to the United States stock market.
In Wall Street lore, if the market rises in the first 5 days of January, it is likely to continue to do so for the rest of the month. In the 1970 to 1981 period, small-capitalized stocks rose 8 times as much as large-capitalized stocks during those 5 days. A good January is a bullish indicator for the whole year. One possible explanation for the beginning of January effect might be that in recessionary times more investors are desperate to take profits at the beginning of the new tax year, whereas in expansionary times more investors are eager make their annual retirement contributions as early as possible -- and have the cash to do it.
In the 1912-1992 period there was a notable difference in stock market returns during each of the four years of a Presidential Election Cycle. In the year preceding an election, average returns were 11.0%, in comparison to 7.0% in an election year, 4.7% in a postelection year and 2.3% in a midterm year (6.3% overall average).
In the 1897-1964 period the day preceding a holiday showed an upward
movement more than two-thirds of the time, whereas the day following
a holiday showed upward movement only half the time. The day preceding
Christmas & New Years' showed upward movement in three-quarters of
cases. In the 1963-1982 period, returns on the day preceding Christmas,
New Years', Thanksgiving and Memorial Day were more than 7 times
as great as those on an average nonholiday. Strong but lesser effects were
seen for Good Friday and the Fourth of July. Although returns were high
for the days preceding Labor Day in those years, that period has been poor
Concerning weekdays, in the 1928 to 1982 period, Monday was
the only consistently down day (on average) of the week -- including during
the Depression. It has been my observation, however, that following a
week of selling, Mondays can be the occasion of a powerful rally. When
trading resumed following the September 11, 2001 attack on the World
Trade Center, there was strong selling every day of the week -- but a strong
rally on the following Monday.
The results of an analysis of weekday trading sequence patterns for the
S & P 500 during the 1986-1996 period is illustrated
in the diagram by sequences of Xs & Os. The results for
Monday, whether up or down, are designated by X. If Monday is up
& Tuesday is down, the pattern for Tuesday is XO -- and is
exactly the same if Monday is down & Tuesday is up. The numbers in the
boxes represent the % probability of each of the two possible outcomes
on subsequent days. Of some interest to me is the 62% probability of an
XXXO following an XXX, but a 62% probability of an
XXXXX following an XXXX. Elliot Wave Theorists might find
the precise percentage figure of interest.
The time-of-day effect illustrated shows that the strongest moves tend to occur in the first hour and last half-hour of trading. The chart indicates bullishness for the last half-hour, but it has been my experience that on down days the last half-hour is exceptionally bearish. The high or low of the day is most likely to occur in the first half-hour and second most likely to occur in the last half-hour.
An academic study of the effect of phases of the moon on stock returns in 48 countries found a 6.6% greater return on new moons than on full moons. The effect is about twice as great for emerging market countries than for G-7 countries (where people spend a greater proportion of time in artificial light) -- and is larger for small-cap than for large-cap stocks (suggesting a greater influence on individual investors than institutional investors). The influence has a sine-curve distribution through the entire lunar month, with maximum returns at the new moon (no moonlight) and minimum return at the full moon (full moonlight). Correlations between the moon & mood are sometimes associated with menstrual cycles or with light-induced release from the pineal body of the hormone melatonin (the light-sensitive hormone that controls seasonal rutting in animals). Another study reports New York Stock Exchange returns for the 1982-1997 period as being 4 times greater on perfectly sunny days than on perfectly cloudy days. (For an online review of the papers on sunny days & moonlight, go to How the sun and the moon move markets)
I often speculate on the reasons for the patterns I observe. I believe that patterns must have reasons and I believe that understanding those reasons can enhance predictive power (and forewarn of exceptions). For example, I suspect that Mondays are often down-days because many people have weekend expenses that leave them cash-strapped on Monday -- forcing sale of stock. A Monday pre-holiday might not be as good as another pre-holiday day.
The down-markets in September/October are often attributed to Mutual Funds selling their losers before the end of their fiscal year (usually the end of October). But I suspect that the demand for cash is high (motive to sell stock) because retailers need to invest in Christmas merchandise inventory. Likewise, most of the gains in December are usually seen in the days between Christmas and New Year's -- with retailers investing their profits from sales of merchandise into stock. This carries into January, which is also a time for large movements of money into pension funds. Managers of various kinds of funds may be more willing to risk potentially profitable investments at the beginning of a new year -- with the idea that there is plenty of time to compensate for mistakes. Down-years would not be expected to see as much of a "Santa Claus rally" in the last days of December because more people would be doing tax-loss selling in December and fewer would be doing profit-taking in January (although the incentives for profit-taking might be more urgent).
The expression "sell in May then go away" reflects the fact that most stock market gains occur in the October to May period, with more losses occurring June to September period is a good time to be invested in bonds. Also, buying very long-term Treasuries, munis or corporates when the yield curve is inverted -- or when interest rates have peaked -- is always a good strategy. The bonds will pay the highest interest for the longest time if held to maturity or can be sold at a good profit when interest rates have bottomed.
(For more details on seasonal and yearly patterns, see my article
Timing the Market. For details
on yield-curve interpretation and bond investing see my article
Fixed-Income Securities: Money-Markets
(return to contents)
Event trading can be subcategorized into trading on scheduled and unscheduled events. Anyone who watches intraday stock market indices graphed on time-series charts will have noticed the sudden & dramatic moves resulting from events. Some technical analysts claim they can ignore world events and can trade strictly on the basis of their charts or statistics, but they do so at their peril. On the other hand, making money on events is exceedingly difficult for anyone who is not a professional trader/broker.
The market indices are powerfully moved by almost every change in short-term interest rates by the US Federal Reserve (Central Bank). (For background on the mechanics of short-term interest-rate manipulation by central banks, see Money-Creation by Banks and TARGET="_new"A "Managed Economy" Under the Federal Reserve System.) The days, hours and minutes surrounding the regularly scheduled meetings of the Federal Open Market Committee (FOMC, "the Fed") are invariably times of high volatility. (See the FOMC "Schedule of Meetings, Statements, Minutes and Transcripts" for schedules). Low interest rates help make business more profitable and higher rates make business more costly.
In most cases the market reacts strongly (positively or negatively) to whatever decision is made on the day of the FOMC meeting. The market reactions to surprise intra-meeting actions by the Fed are even more dramatic. The surprise intra-meeting interest rate cut of 50-basis points (0.50%) on January 3, 2001 resulted in the biggest one-day stock market index jump (13%) in the history of the NASDAQ.
The market also responds strongly to certain economic reports of market conditions -- both because of what the reports indicate about the direction of the economy and because of what they indicate about future actions by the Fed. The strongest indicators of future Fed action are the unemployment report, the Consumer Price Index (CPI) and the manufacturing index (ISM). If CPI is unexpectedly high, the market would probably fall because the Fed may be more inclined to increase rates. If the ISM is unexpectedly low, the Fed may be more inclined to lower rates. The market may rise on the expectation of lower rates or may fall on the expectation of further deterioration of the economy.
The report giving the strongest & most perverse reaction, however, is the unemployment report. Increasing unemployment would be a sign of a deteriorating economy, but if unemployment unexpectedly increases the market will usually rally on this sign that the Fed has a greater likelihood of decreasing (or a lesser likelihood of increasing) interest rates. Only when the economy is seriously deteriorating and the Fed has already cut rates considerably will an increase in unemployment be interpreted negatively by the market.
The schedule of weekly reports/announcements can be found at http://biz.yahoo.com/c/e.html. Each report is "graded" (A-F) on the basis of how strongly large or unexpected changes can move the markets.
Corporate quarterly earnings announcements are regularly-scheduled market-moving events which begin in mid-January, April, July and October. The specific dates (and often times) of the announcements can be found at http://biz.yahoo.com/research/earncal/today.html. The biggest corporations, of course, are the biggest market movers. If Dell has unexpectedly negative results, for example, the whole PC sector (and semiconductor sector) will sell-off (usually over-sell). Unexpectedly positive results -- or even results not as negative as expected -- can have the opposite effect. Companies often try to keep expectations reasonably low (manipulate expectations), preferring (small) positive surprises when they make their announcements. (Cisco was masterly in beating expectations by exactly one penny per share for a large number of consecutive quarterly reports -- before the year 2001 and the "tech meltdown".) The weeks preceding the scheduled announcements is an unofficial "earnings warnings season" in which corporations having problems make unscheduled warnings -- usually negative reports intended to lower earnings expectations.
In my experience, attempting to trade on economic reports & corporate earnings reports is perilous. Often it is safer to stay out of the market in anticipation of a strong market-moving announcement. Moreover, the reaction to these announcements is greatly influenced by the condition of the market. The market is more likely to react strongly in the negative direction after a sequence of up-days and more likely to react positively after a sequence of down-days.
The adage "buy on the rumor, sell on the news" is amply demonstrated by the market, in my experience. I have often seen the market rise in anticipation of a near-certain scheduled positive announcement -- and then sell-off on the announcement. The best strategy for such a situation is to buy hours before the announcement, short-sell minutes before the announcement and then cover the short immediately following the announcement.
Unscheduled political or environmental events can have a devastating impact. On the announcement of Judge Jackson's anti-trust ruling against Microsoft the company lost $60 billion in market capitalization in a single day -- and the entire market (especially the NASDAQ) sold-off. The September 11, 2001 attack on the World Trade Center devastated the equity markets. Every investor & trader -- whether trading on technicals or fundamentals (or both) -- should never forget event risk: an earthquake or terrorist attack can always strike at any moment and send the market into a nosedive.
To keep a pulse of the market, I generally end every trading day by checking on:
(1) DOW Activity
(2) Sector movement
(3) US Treasuries movement
(4) NASDAQ COMPOSITE Daily Volume Leaders
(The NASDAQ 100 is elegantly shown at: Heatmap)
For current market news I prefer CBS MarketWatch
and for breaking market news I prefer Market News (Yahoo)
CNBC news television, the WALL STREET JOURNAL, BARRON'S and BUSINESS WEEK magazine are also helpful to keep on top of the market.
(return to contents)
I believe that there are three essential components to successful trading:
(1) Money management
(2) Risk management
(3) Forecasting ability
Money-management is absolutely essential. A trader should know how much his/her assets total at every moment. And (associated with risk management) a trader should know exactly how much money will be lost if everything goes completely wrong (as often happens). In short, always look both ways before entering the market -- look up and look down!!
I know of too many cases of traders who have lost everything on a single trade. It is only possible to lose everything on a single trade by risking everything on a single trade. A trader may make big money a few times by doing this, but as a policy it is a recipe for disaster. Conservative money managers will never risk more than 5% of total capital on a single trade. The market is very efficient in transferring money from the impatient investor to the patient investor.
Concerning risk-management, protective stops are a powerful tool for self-discipline in daytrading. Stops should never be below round numbers on long positions and above round numbers on short positions -- to reduce the chance of being stopped-out at the same time as other investors (who have a fondness for round numbers). Round numbered stops typically increase slippage, ie, extra loss beyond the stop due to a bad fill of the stop order. Fast price movements increase slippage.
Stops are also a way to get yourself cut to pieces if they are too close to your current position. The market must have some slack to make excursions up and down on the way to your desired destination. More often than not, when I am worrying about what my stop should be, it is a sign that I should exit my position (sell-out if long, buy-out if short). I always prefer to be in control of when I exit my position. Stops have probably cost me more money than they have saved me because of sudden & temporary spikes in price. (There was a significant exception.)
Excessive confidence in one's forecasting ability is dangerous. A good forecaster who is right more often than wrong, who always knows how much money she/he has available and who does not risk too much money on any one trade is most likely to succeed.
(return to contents)
Day-trading is very risky -- and the odds are stacked against the amateur day-trader. Over 80% of day-traders in stocks lose money in a bull market -- a much higher percentage in a sideways or bear market. The on-line day-trading companies were devastated by the 2000/2001 bear market. The futures market is even more risky. The very few of those who enter the futures market with less than $10,000 remain in the market for a full year.
Even without trading on margin, a trader automatically loses money on every trade because of commissions & fees, bid/ask price-spread and (if not on Globex) a small gouge from the broker (personal observation/opinion). A high volume of trading in which the trader is right on half the trade volume automatically means a loss.
There is a great temptation to buy when there is a sudden run-up in price and sell when there is a sudden fall. (Sudden falls are more frequently sudden and rises more frequently gradual, in my opinion.) Generally this is a bad policy. In fact, there are many contrarians who sell on run-ups and buy on run-downs -- and advise strongly against adding to a position that is currently proving profitable.
My most conservative form of daytrading is to buy on a low where the market has sold-off strongly for a couple of days and has been relatively flat for a couple of hours. I prefer to do this when I have enough margin to hold the position overnight, if necessary (not day-trading, literally, but a good time for entering a position).
Less conservatively, I will look for what I expect will what I call a monotonic-upday, ie, a day when the price movement is primarily upward with little downward excursion. This may be a Monday after a period of deep selling or the Friday after American Thanksgiving. Although it is normally bad practice to add to a profitable position, it can be conservative if I feel confident of my forecast because my risk only increases in increments -- and if I am right I am still limiting my loss with stops. For example, I would enter the market with an S&P 500 contract on the open (9:30am), with a 4-point or 6-point stop (allowing for the fact that the first half-hour of trading is the most volatile). If the market moves up 2 or 3 points I will buy another contract and raise my stop. 10 o'clock is a very volatile time, so if I can get beyond then I would narrow my stop to 3 points and continue the process. If I have patience to last-out the day I may attempt to do so, but normally the afternoons can be saggy & risky, so it is most prudent to sell-out earlier -- even though the last hour or half-hour often shows a final up-surge.
My most aggressive day-trading is on the day of the FOMC meetings. Typically the market rises in the morning of those meetings. At the time of the 2:15 pm announcement of the interest rate move (and bias) the market is usually very volatile (after being very unvolatile shortly before). It is usually most prudent to say on the sidelines for the first 15 minutes, at least. By 3 pm a trend is usually established that generally lasts at least until 3:45 pm. Buying or selling to follow that trend can be very profitable. Of course, if you are heavily margined and an earthquake hits you can be in deep trouble if you are not watching CNBC and not ready to exit the position at a moment's notice.
I try to day-trade only rarely, when I believe the opportunity is exceptional. Oddly, I normally take a position in the futures market where I have ample margin to cover the most extreme of market movements -- buy-&-hold or sell-&-hold -- not what you normally see in futures traders.
When I do day-trade futures indices I find it essential to continually monitor charts of the indexes (to follow trends) and the numbers themselves. To watch the numbers, I use:
Major U.S. Indices (Yahoo)
whereas for charts I use:
DOW Day Chart (Yahoo)
NASDAQ 100 Day Chart (Yahoo)
S & P 500 Day Chart (Yahoo)
(return to contents)
Holding positions in the market can be extremely emotionally involving. And, of course, the larger the position (the more at risk) the greater the emotional involvement. Every uptick causes joy and every downtick causes grief. It is enough to cause otherwise intelligent people to spend hours staring mesmerized at changing numbers.
The craving for emotional involvement may be the same as gambler's excitement. But I am not a gambler and I hate gambling -- I know the odds are against me. I never buy lottery tickets and when I go to a conference in Nevada, the small amount of money I have put into slot machines was motivated by curiosity rather than by an expectation of winning.
I don't randomly trade. When I commit money in the future's market I always do so on the basis of some theory I have about how the market will move and why. Nonetheless, some of those theories can be pretty flimsy, so I try to make the amount of money I commit correspond to my intensity of certainty (knowing that I am often wrong). I am right more often than wrong -- but have been terribly wrong too many times. I continue to learn, but continue to find new mistakes I can make.
A professional investor friend of mine has said that many people lose money in the market because being in the market is motivated by reasons other than making money or avoiding losses -- such as talking to their friend about trading or the emotional involvement of trading. "When in doubt, stay out" is good advice. I experience great grief when I am out of the market and am watching the market make big moves that could have been making money for me. I wonder at times if that emotional grief isn't greater than the grief I experience when I have a position and am losing money.
A run-up in price inspires optimism (an emotional buy-signal) even though it is bad policy to buy into a run-up (similar to selling into a sudden price decline). (Contrarians are wired backwards.) Like many traders I have great difficulty taking a loss from a losing position. For some people this difficulty is associated with too much ego-involvement in their trading decision. They took a position on the basis of some belief and are unwilling to admit having made an error. In my case, however, it is more that I cannot truly accept the fact that I have already taken a loss -- as if my unrealized loss is not real and that by holding the position the market may move in my favor. Moreover, I reason like a contrarian that if the market has already moved so far in one direction a reversal is more likely (when prices have fallen, buyers are attracted and vice versa). When I realize that I have walked into a propeller and am being cut to pieces, I have all the rationality of a deer who stares transfixed into the headlights of an on-rushing car. The worst position is to be feeling sick with grief and to be passively/helplessly hoping the price will make a move for the better. Such emotions are a sign of poor money management -- too much has been risked.
Programmed or mechanical trading would remove the emotion from decision-making, but it would not remove the emotion from profitting or losing. Futures traders often blame their failures on lack of self-discipline. By this theory, one should not enter a position without having a well-defined exit price -- and one should have the discipline to act on the plan (or use stops to enforce the decision).
I do have non-monetary, emotional incentives for my involvement in the market -- and I don't believe this is a bad thing. Years ago I made a conscious decision that I did not want to drink alcohol at parties as a means of reducing my anxiety. I wanted to experience my emotions fully -- and to learn to deal with them. I can say similar things about trading. Also, I am one of those people for whom market behavior is a fascinating puzzle -- and I am arrogant enough to believe that I have the capacity to solve that puzzle. I go from mistake to mistake, thinking that I am approaching the point where I will run out of mistakes and consistently make more money than I lose.
I believe that I have learned a great deal about myself by trading -- and I believe that I have grown as a person in learning to control my impulsiveness and impatience. Unlike Keynes, who arrogantly proclaimed that "The market can be irrational longer than I can stay solvent", I have learned humility in the face of my many false forecasts. Life & survival is a process of risk management. I believe the market is teaching me wisdom & good judgement that has helped me (and, hopefully, others) in managing many many areas of life. I still have lots to learn. I still become too giddy & reckless when successful -- and dysfunctional from despair (or desperate & reckless) when I fail. The learning is on a very deep level of personality -- much deeper than factual knowledge because in many cases I already know the mistakes but have not gain enough mastery over myself so as to not make them.
(return to contents)