Category: Technical Analysis

1. January Signals

One of the first indications that the Efficient Markets Hypothesis had flaws was the observation that in January the stock market appeared to have several consistent price patterns. The stock market cannot be considered efficient if every year the same pattern forms and can be traded profitably. For many years, this inconsistency with the hypotheses remained unanswered, but more recently the market’s January behavior has changed, likely because traders have taken advantage of the anomalies and dampened their profitability.

2. January Barometer

“As the Standard & Poor’s goes in January, so goes the year” is a basic contention proposed by Hirsch of Stock Trader’s Almanac. Statistical problems arise with this indicator, despite its popularity. First, if January is up, for example, the year starts with an advance and already has a leg up. Second, because the stock market has had a positive bias over the years, January usually is up as is the market for the year. When the periods and odds are broken down, the predictive value of the January barometer comes within normal statistical probabilities and is without predictive value.

3. January Effect

The January effect was a condition that lasted for many years. During January, small-cap stocks had a tendency to have abnormal strength. Although some analysts attributed this effect to investors timing trades for tax reasons in December and January, no one could explain why the January effect occurred in the magnitude it did. As a result, it became the darling of anti-random-walk theorists as an example of how the market can establish patterns that are profitable. In recent years, however, the January effect has not worked very well. Some say it has been arbitraged out of business.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Short-term traders sometimes practice what is called event trading. This is when either a news announcement is due, a surprise news announcement occurs, or a holiday is soon to occur. Holidays are seasonal and, thus, included here.

The major holidays during the year have shown to have recurring patterns. For example, some analysts have observed an Independence Day pattern. Stock market performance tends to be the strongest five days prior to Independence Day. Diminishing average performance generally occurs the five days following Independence Day. The sixth day after Independence Day is associated with strong stock market performance. Another example of a holiday pattern is that the two weeks before Thanksgiving seem always to be positive. However, as we have seen in the “Seasonal Patterns” section, November is generally one of the strongest months anyway. Thus, when we deduct for November’s habitual strength, the pre-Thanksgiving strength becomes more random and less useful. Is this type of information useful to the individual trader? The answer is up to the individual trader. In most cases, the observed pattern is well within expected statistically random results and is, thus, not helpful.

Another often-heard rule of thumb is to “Buy stocks on Monday and sell stocks on Friday.” However, monthly, weekly, and daily patterns have the same statistical problems as holidays. Any observed patterns, once adjusted for underlying trend and for randomness, show little in the way of consistent return. Testing any simple trading rule that has been popularized in current market conditions is important. In a recent study, Dahlquist (2009) finds a striking deterioration in the statistical significance and profitability of this trading strategy over the past three decades. In recent years, there is no evidence that Monday returns are any different from any other day of the week.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Funds that are currently in the financial market are undoubtedly important, but the availability of funds outside the financial market also plays an important role in determining market conditions. The value of and the liquidity of household financial assets are important considerations when measuring how much money households might move into the securities markets. Money supply measures and bank loan activity are also important when considering how much money is available for the financial markets.

1. Household Financial Assets

Households, like corporations and governments, have different kinds of assets. They have both physical assets, such as cars and houses, as well as financial assets, such as stocks, bonds, mutual funds, and banking accounts. Some of their financial assets are liquid and some are not. Liquid assets can be converted to cash quickly; cash, bank deposits, money market mutual funds, U.S. Treasury bonds, notes, and bills are liquid financial assets. Other financial assets cannot always be converted to cash quickly; stocks generally are considered to be more liquid than other financial assets, such as pension funds, retirement accounts, profit­sharing accounts, unincorporated business ownership, trusts, mortgages, and life insurance.

A ratio of liquid financial assets to total financial assets shows how “liquid” households are—in other words, how easily they can raise cash if they need it. Generally, the more liquid households are, the more able they are to invest in stocks. When household liquidity is high, it is, therefore, favorable for the stock market, whereas low liquidity is negative for the stock market.

The data presented in Figure 10.3 from Ned Davis Research, Inc. shows a substantial decline in household liquidity in the 1990s and more recently since 2010. This is one reason why consumer debt has risen so strongly and why households are unable to withstand a severe economic contraction. During a severe economic contraction, one major source of funds for cash-strapped households is the stock market. Thus, this indicator, although it does not give mechanical signals, does measure the potential funds for stock demand or supply.

The range in percent of financial assets need not change by a large amount to have an effect on stock prices. According to a Ned Davis Research, Inc. study, when the percent of household financial assets has risen to 29.4%, the stock market has averaged an annual gain of 16.1%. A decline in percent to 25.4% corresponded with only a 3.2% gain in stock prices.

2. Money Supply (M 1 & M2)

Expansion in the money supply is a measure of potential demand for stock, as well as other assets, and is, thus, a rough measure of the potential for business and stock market expansion. Increases in the money supply have been historically associated with increases in economic growth and productivity. Although this economic theory is straightforward, quantifying actual money supply growth is not as easy as it seems. Measuring the amount of money in the economy depends on knowing what money is. If I ask you how much money you have, you would certainly count the currency and coins in your wallet. However, you would also probably think about the money you have in your checking account. Then, there is also the money you have in a savings account or money market mutual fund. Thus, the definition of money, and what to include in a measurement of money, is not as straightforward as it would initially seem.

The Federal Reserve measures money supply in a number of ways. Table 10.1 shows the various measurements of money that the Federal Reserve uses. The categories are based on liquidity, or the ease in which the financial assets can be converted into cash.

M1, the narrowest definition of money, is a measure of the most liquid assets in the financial system. It includes currency and the various deposit accounts on which individuals can write checks. M2 is a slightly broader measure of money, which includes M1 and various forms of savings accounts. The additional components of M2 are slightly less liquid than the financial assets of M1. Additional money sources not included in M1 or M2 are institutional money funds (1,781.6), foreign demand deposits in U.S. banks (88.9), foreign time and savings deposits (206.7), IRA and Keogh accounts (656.1), and U.S. government deposits (209.7) as of April 27, 2015. Today, the most commonly quoted monetary aggregate is M2 because its movements appear to be most closely related to interest rates and economic growth.

Because M1 is the most liquid portion of M2, a ratio of the two measures the liquidity of M2 and thus the proportion of M2 that can be used immediately for purchase of stocks and other investments. The Ned Davis Research, Inc. chart in Figure 10.4 shows the year-to-year change in this ratio scaled to between 0 and 100. Ned Davis Research, Inc. concluded that when the change advances above 70, the S&P 500 index gains at a rate of 17.2% per year, whereas when the change is 36 or below, the index declines at an annual rate of 4.9%. The liquidity of M2, therefore, is a powerful indicator of stock market direction.

3. Money Velocity

The velocity of money is a measure of how fast money turns over in the economy. It is calculated as a ratio of personal income to M2 (or sometimes GDP to M2). Money velocity is related to inflation; the faster money circulates, the more pressure that exists on prices, and the more it serves as a leading indicator of long­term interest rates, again because it reflects inflationary pressure. In terms of being an indicator for the stock market, Ned Davis Research, Inc. found that when money velocity (a monthly figure) rises above its 13- month moving average, the stock market has advanced 4.8% per annum on average. When money velocity declined below its 13-month moving average, the stock market advanced 9.1% per annum. This relationship is shown in Figure 10.5. Clearly, inflationary pressures from increased money velocity put a damper on stock market prices.

4. Yield Curve

The yield curve is a graphical representation of the yield on bonds with various lengths of time to maturity. The yield curve changes over time as short- and long-term interest rates change due to Federal Reserve action or the marketplace. Figure 10.6 shows the 3-month to 30-year yield curve at three different dates (May 31, 2014, November 30, 2014, and May 31, 2015) to display how the curve can change over a year’s time. Yield curve information is also summarized as a ratio or difference between a short-term and long­term interest rate over time. Figure 10.7 shows the difference between the long-term Treasury bond yield and the three-month Treasury bill rate from September 1964 through May 2015.

Banks traditionally practice maturity intermediation, borrowing short-term funds and lending to corporations or individuals over longer periods. Thus, they are said to “borrow short and lend long.” Their profit depends on the spread between the cost of funds, the short-term interest rate, and the return from loaned funds, the long-term interest rate. As short-term interest rates rise, presumably from Federal Reserve policy action, and long-term interest rates remain steady or decline, the yield curve becomes flatter and the banks are unable to profit as much from the spread. The yield curve, therefore, is a crude measurement of bank potential profitability. Bank profitability affects interest rates, and interest rates affect the stock market. Thus, the yield curve is a forecaster of stock market direction, and, historically, it has had an acceptable record of anticipating major turns in the stock market.

The “normal” relationship between interest-bearing instruments of the same quality is for longer rates to be higher than shorter rates. This is presumably because over time there is a risk of inflation, default, and other economic problems, the risk for which the holder of the long-term interest rate instrument wants to be compensated. This results in a positively sloped yield curve when plotting a graph with the time to maturity on the horizontal axis and the interest rate on the vertical axis for these securities.

The yield curve becomes abnormally steep when long-term rates rise considerably higher relative to short­term rates than the historic average of around 200 basis points. Although this can occur because of an upward movement in the long-term rate or a downward movement in the short-term rate, it is usually caused by the Fed lowering short-term rates to expand economic activity.

At times, short-term interest rates rise above long-term interest rates and the yield curve becomes inverted.

This inverted yield curve usually has dire consequences for the economy because it curtails the incentive of banks and other lending institutions that borrow money at short-term rates to make loans at long-term rates. The Federal Reserve estimates that an inverted yield curve predicts recessions two to six quarters ahead. Figure 10.7 shows the history of the yield curve and how it tends to forecast the direction of the stock market. Ned Davis Research found that when the long-rates advanced 1.1 percentage points above the short-rates, the stock market advanced 11.4% per annum on average. Contrarily, when the yield curve inverted, the stock market declined on average 7.2% per annum.

5. Bank Liquidity

Generally, an increase in loan activity, the amount of loans being created and existing, is a sign of increased business activity. It can also be a sign of increased speculation or a particular period when banks, because the yield curve is so much in their favor, become less prudent in their lending policies. When loan demand increases, it puts upward pressure on interest rates. Conversely, a decrease in loan demand puts downward pressure on interest rates. One measure of bank speculation is the Liquidity Coverage Ratio (LCR). This ratio is the total of Category I financial assets, such as cash, gold bullion, and obligations of the U.S. government divided by the cash necessary to operate the bank for 30 days. Operation of the bank includes payroll, normal business expenses, payments on loans, and so on, that the bank incurs in its normal working mode. As the loan business gets hot, banks tend to loan right to the limit of their LCR. Thus, by following this estimate of bank liquidity, one can measure the risk of the banking system as well as the prospects for higher or lower interest rates. Ned Davis Research Inc., in Figure 10.8, displays a timing system based on the year-to-year change in bank liquidity that has shown excellent promise as a market timing device. It found that when the liquidity increases by 0.1 or more, the banks are becoming liquid and have the resources to lend. This is favorable for the economy and the stock market, which has gained 10.2% at an annual rate in such times. When the liquidity shrinks to -5.4 or below, the stock market declines at an annual rate of 4.8%. The liquidity year-to- year change is thus an excellent forecasting method for the stock market.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

The interest rate represents the price of borrowing funds. It is the cost that borrowers have to pay to use money; the higher the price, the less likely borrowers are to borrow. It is also the reward that lenders receive for letting someone borrow their money; investors compare this return with the expected return in other markets, such as the stock market, when they decide where to allocate their funds.

1. Short-Term Interest Rates

The theory behind using short-term interest rates as stock market signals is based on two assumptions. First, interest-bearing investments are alternatives to stock investments. In other words, savers make choices about placing their investment funds in interest-bearing securities or the stock market. When interest rates are relatively high, the interest-bearing securities look relatively more enticing. Second, interest rates directly affect costs for corporations and, thus, corporate earnings. However, the interest rate itself is important because the expected rate of return in the stock market must be greater than the short-term interest rate for investors to invest. When interest rates begin to rise, while the stock market languishes, investment outlooks begin to change. On the corporate and personal level, rising rates translate into rising costs. Whether it is working capital borrowing or adjustable rate mortgage payments, increases in short-term rates have a negative effect on net income and, by extension, on confidence. Alternatively, when rates decline, costs also decline and confidence builds. Furthermore, and regardless of potential reasons, short-term interest rates correlate closely but inversely with stock market behavior. Knowing when a major switch in direction in short-term interest rates has occurred is important for a probable switch in stock market direction.

Federal Reserve policy indicators and short-term interest rates are generally very accurate, although often early, as a predictor of stock market direction. Figure 10.9 shows the relationship between short-term interest rate movements and the S&P 500. During the speculative bubble from 1998-2000 and its collapse into 2002, however, interest rates had little effect upon the market’s direction. Emotion had overcome logic, and the relationship between short-term interest rates and the market was abandoned for greed to make easy money and fear of missing the next upward wave in stock prices. Again, in the period between 2007 and 2009, when the stock market collapsed to a new 10-year low, lower short-term interest rates had no positive effect on the stock market. Financial jargon calls this behavior pushing on a string, when short-term rates do not cause incentives to buy stocks. The last time rates did not work was in the 1920s speculative bubble and early 1930s collapse. They were reliable for more than 50 years thereafter.

2. Long-Term Interest Rates (or Inversely, the Bond Market)

Long-term interest rates are related to, but not perfectly correlated to, short-term interest rates. The Federal Reserve effectively controls short-term interest rates through its various policy measures, but it does not have as tight control of the long-term market. When we speak of long-term interest rates, we speak of the bond market. Long-term interest rates and bond prices are inversely related. When long-term interest rates rise, bond prices fall, and when long-term interest rates fall, bond prices rise.

Relationships between the bond market and the stock market also exist. The relationships between bonds and stocks have much to do with the payout for security holders, coupon payments for bondholders, and dividends and earnings yield for stockholders. Generally, investors view bonds as long-term investments with a steady, fixed coupon return, whereas stocks are long-term investments with a variable, less predictable return. Both markets, however, can fluctuate widely.

For long-term outlooks (as opposed to long-term interest rates), therefore, it is important to know the historical relationship between these investments. As a rule, long-term bonds have tended to move in the same direction as the stock market. In other words, long-term interest rates have tended to move in the opposite direction from the stock market. As the bond market makes a major bottom, the stock market often makes a major bottom also. At tops, the bond market tends to lead the stock market and is, thus, very often, an early indicator of trouble ahead for the stock market. As in short-term interest rates, this relationship broke down during the period of the speculative bubble and collapse between 1998 and 2002 and between 2007 and 2009.

Prior to those periods, the relationship had been steady for more than 50 years and will likely return. See Figure 10.10 for the relationship between long-term Treasury bond yields and the S&P 500 over the past 50 years.

Ned Davis Research developed a simple trading rule for long-term interest rates. It is the number of points a three-week moving average changes, up or down. If the long-rate measure declines by 8.7% from a weekly peak, a buy signal is generated. On the other hand, a sell signal is generated when the long-term rate average advances 11.7% from a weekly low. The performance history of this simple method produced an average annual return of 9.8% in the S&P 500 over the past 50 years. This is compared with a buy-and-hold gain of only 6.6%.

3. Corporate Bond and Stock Market Yield Spread

If we consider the inverse of the P/E ratio for stocks as the stock yield—that is, the earnings return percentage of the stock’s market price, and the yield on average corporate bonds, the Moody’s Baa Bond index —in a ratio format, we should see how they are performing against each other and learn about which market, the stock or bond market, has the highest current return. In Figure 10.11, we show the Ned Davis Research, Inc. plot of this ratio and the signals it has generated for the stock market (and conversely for the bond market as the alternative investment). When the bond market is performing better than the stock market by 4.6 percentage points, the annualized gain in the stock market has been minus 4.4%. When the bond market has fallen behind the stock market by 3.4 percentage points, the annualized gain in the stock market has risen to plus 16.7%.

5. The Misery Indices

The economist Arthur Okum designed the misery index in the 1960s during the Johnson administration when inflation was a special concern. Inflation, coupled with high unemployment, resulted in what economists call stagflation. Okum created the index in an attempt to measure the social and economic cost of high inflation and high unemployment. A high Misery Index indicated that the combination of inflation and unemployment was high and that investors were experiencing a more stressful economic environment. Later, in 1999, Robert Barro of Harvard added an interest component and the spread, positive or negative, between the GDP actual and trend rate of growth to create the Barro Misery Index (BMI). The more common version today includes the interest rate, either prime rate or mortgage rate, but not the GDP trend information.

Figure 10.12 shows how the level of the misery index has related to the performance of the DJIA since 1966. This figure shows the results of buying the DJIA whenever misery index falls by 0.3 points and selling the DJIA whenever the index rises by 3.2 points. The trading accuracy of this system is 75% favorable, and its gain per annum is greater than the buy-and-hold gain per annum by 3.7%. Because the calculation of the misery index is easy and low cost, the profitable results of this trading system appear valuable.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

The Federal Reserve System (www.federalreserve.gov), often referred to as “the Fed,” is the independent federal organization that determines and implements monetary policy for the United States. The Federal Reserve’s policy regarding the money supply is the principal determinant of short-term interest rates. The Federal Reserve has three main tools for adjusting the money supply: changing the amount of reserves that banks are required to hold, changing the discount rate, and buying and selling U.S. Treasury and federal agency securities through its open market operations.

When the Federal Reserve purchases any kind of security, it adds money to the banking system. Because banks then have more cash in reserve, they are more likely to make loans, and interest rates tend to fall. Another way to look at this relationship is that the Fed’s purchase of bills and bonds represents increased demand for bills and bonds. As the demand for bills and bonds rises, bill prices rise and short- and long-term interest rates decline.

Fed selling of government securities has exactly the opposite impact. As money is used to pay for these securities, reserves are drained from the banking system. When there is less money in the banking system, banks can make fewer loans, and interest rates tend to rise. Fed selling of government securities represents an increased supply of securities in the marketplace, and, thus, the price of these securities falls. As bill prices fall, interest rates rise.

The Fed uses open market operations to reach a federal funds rate target. The federal funds rate is the interest rate at which banks borrow from each other. Although the federal funds rate is not set by the Federal Reserve, Fed action greatly impacts this rate. If the Federal Reserve makes open market purchases, banks will have more money in reserve. Fewer banks will need to borrow reserves from other banks, and more banks will have excess reserves that they want to lend. This, in turn, places downward pressure on the federal funds rate. When the Fed sells securities, bank reserves decrease, and as bank reserves decline, more banks want to borrow and fewer banks want to lend in the federal funds market. Thus, the federal funds rate will rise.

The Fed implemented expansionary monetary policy in the fall of 2007 by purchasing securities through its open market operations. By the fall of 2008, these open market operations had driven the federal funds rate down to almost zero. Worried about the lack of economic expansion with short-term interest rates near zero, the Fed decided to buy long-term bonds to add money to the financial system and lower long-term interest rates. This policy, called “Quantitative Easing,” or QE, has focused on flattening the yield curve and increasing the reserves banks hold in excess of their required reserves. In September 2008, before quantitative easing began, reserves in the banking system amounted to approximately $11 billion. By the end of 2008, reserves had ballooned to an amazing $860 billion. The amount of reserves has continued to expand, reaching $2.6 trillion by the spring of 2015.

The Federal Open Market Committee (FOMC) meets about every six weeks. At these meetings, the FOMC establishes a federal funds target rate. Although the minutes of each FOMC meeting are not made available to the general public until three weeks after the meeting, the FOMC announces its federal funds target rate in a statement released at the conclusion of the meeting. By implication, this federal funds rate target tells the public whether the Federal Reserve is pursuing a restrictive or expansionist policy. Because short-term interest rates are so important to business and the stock market, and because the Federal Reserve has such a large impact on short-term interest rates, it is important for the analyst to be aware of Fed policy and the subtle changes in that policy.

1. The Federal Reserve Valuation Model

Even though the Federal Reserve is in charge of monetary policy and is concerned with the overall health of the economy and the financial markets, the Federal Reserve seldom makes direct comments about the health of the stock market. However, economist Ed Yardeni (www.yardeni.com), in the back pages of a Fed report, discovered the Greenspan Model, also known as the Fed’s Stock Valuation Model. As shown in Figure 10.13, this model gives a general indication of whether the Federal Reserve sees the stock market as overvalued or undervalued. It is a valuation model that determines if the stock market is too high or low based on the stock market earnings yield relative to yield on the ten-year U.S. note, similar to the timing model in Figure 10.11. Although this is an extremely easy-to-calculate indicator of general market value, there are several criticisms of this model. The principal criticisms of the model are that (1) it is too simplistic, (2) a better correlation exists between actual reported earnings than estimated earnings, and (3) it doesn’t include inflation, an important factor in determining long-term interest rates.

2. Federal Funds

Federal funds, known as fed funds, are short-term, often overnight, loans made between banks. A bank with excess reserves can lend them to a bank that does not have enough reserves to meet its reserve requirements. The interest rate charged for this loan is negotiated between the banks. However, the Federal Reserve has significant influence over this rate through its open market operations. The Federal Open Market Committee sets a target range for the fed funds rate. Changes in the rate indicate potential changes in the Federal Reserve’s policy and thus are watched closely.

Figure 10.14 shows a chart of the 14-month percent change in the fed funds rate and its corresponding signal for the stock market. Ned Davis Research, Inc. has found that a momentum decline of 18% is sufficient to generate a buy signal in the stock market, and a 9% or more rise in momentum is sufficient to sell the stock market. The signals have been 84% accurate since 1957.

3. Free Reserves

Free reserves are the amount of money banks have over that which they are required to keep for risk management that is free for lending. It is a figure that has been used in stock market timing for many years. After the quantitative easing by the Fed in 2008, excess reserves in the banking system swelled because the downward pressure from the purchase of bonds on the long-term interest rate squeezed the yield curve, making it unprofitable for banks to lend. Added to this concern was the introduction of the Dodd-Frank Bill that placed many undefined restrictions on the banking system. The net of all this was that banks held back from lending, and their excess cash from earnings remained in house, thus enlarging their excess reserves. Because there is a positive correlation between excess free reserves and the stock market, the stock market advanced strongly. The model in Figure 10.15 by Ned Davis Research, Inc. shows how increased liquidity in the banking system from excess free reserves coincided with a strong stock market. Whenever the three-month smoothed average of free reserves rose above 0.47, the stock market had an annualized gain of 12.0%, and when the free reserves declined below -0.34, the stock market declined an annualized 8.2%.

4. Three Steps and a Stumble and Two Tumbles and a Jump

In line with the desire to measure when the Federal Reserve is tightening credit, Edson Gould, a legendary technical analyst from the 1930s through the 1970s, developed a simple rule about Federal Reserve policy that has an excellent record of foretelling a stock market decline. The rule states that “whenever the Federal Reserve raises either the federal funds target rate, margin requirements, or reserve requirements three consecutive times without a decline, the stock market is likely to suffer a substantial, perhaps serious, setback” (Schade, 1991). This simple rule is still relevant. Although it tends to lead a market top, it is something that should not be disregarded. As shown in Figure 10.16, the rule has been followed by a median decline of 17%. Only two possible incorrect signals were given since 1915: the 1928 signal, prior to the 1929 crash, was possibly too early, and the 1978 signal was probably too late. Thus, this signal has an accuracy record of at least 89.0%.

The Two Tumbles and a Jump indicator was first mentioned in Fosback’s 1973 edition of Market Logic. It is essentially the opposite of Gould’s Three-Step rule. Although it uses changes in the fed funds target rate, margin requirements, and reserve requirements, it looks for two consecutive declines, or tumbles, in any of these policy variables. It has an excellent history of predicting the stock market rises. As Figure 10.14 demonstrates, the percentage of accuracy since 1915 is 84%, with some of the errors considered questionable.

One way to use these two interest rate indicators, which the evidence shows to be superior, is to take them as warnings of important changes in market direction. They are not necessarily strict signals by themselves. However, if a signal is acted upon, to avoid major loss from a false signal, the action must be reversed if the market turns in the opposite direction implied by the signal. For example, if the Three-Step rule gives a sell signal and the market begins to decline, a buy signal will be generated if either the market breaks above its earlier highs or a Two-Tumble signal occurs. The same protection can be used for the Two-Tumble rule in the opposite direction. Essentially, these are “stops” that would prevent any major loss from future incorrect signals, rare as they may be.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

According to exhaustive research on technical analysis by Lo and Hasanhodzic (2010), the earliest known recording of commodity prices with the intent to predict from them was in ancient Babylonia in the first millennium B.C. These recordings were diaries of traders and astronomers who were attempting to correlate astrology with price changes. By the fifth and sixth centuries A.D., price charts, similar to those used presently, were developed in China, Europe, and Japan. The Chinese were interested in cyclicality of prices; the Europeans were interested in astrology; and the Japanese developed the candlestick chart that is still in use today. The “opening of commodity exchanges in Western Europe (1561) and Japan (1654) provided the necessary environment for the development of the chart” (Shimizu, 1986, p. 14). At the time of these exchanges, freely traded markets had become sophisticated enough to produce multiple prices during a trading day and, thus, the requirement for recording the high, low, and close price of each commodity traded on the exchange. It was only natural that this information was portrayed in graphic form. By the 1830s, just before the invention of the ticker tape for the stock exchanges, several chart vendors in New York sold published charts on stock and commodity prices.

The information charted and the method of charting often flows from how the market being analyzed operates. For example, the current use of high and low in a bar chart is not feasible in those trading markets that settle a price only once a day.

Plausibly, the first type of chart was just a simple plot on paper of a number—either amount or price—and a date. In early Japan, for example, rice was traded by amount. Instead of a price per bag, it was the number of bags per price that was recorded by the famous rice trader Sokyu Honma in the 1750s. As markets began to trade more frequently during the day, the chart became more complex. A high and low price could be recorded, and eventually as multiple trades occurred, an open and close price could be added. Volume was recorded much later when more complete and public records were available. At first, witnesses located in the marketplace recorded prices. Eventually, markets became better organized, and prices and amounts were publicly available.

The growth of this business is of great monument to the stock exchange, for it is through the instant dissemination of the quotations made on its floor that the active and continuous interest in the markets is sustained. (Horace L. Hotchkiss, founder of the Gold Stock and Telegraph Company)

The invention of the ticker and the ticker tape revolutionized technical analysis and charting. Shortly after Thomas Edison invented a machine called the Edison Telegraph Printer to print messages from a telegraph, in 1867 Edward A. Calahan, an employee of the American Telegraph Company, invented the ticker tape. Eventually, it was improved upon and patented by Thomas Edison in 1871. This invention not only made conventional charting easier but also allowed for point and figure charting, because such charts required knowledge of every price at which an item had traded during the day. Without the ticker tape, the gathering of this information would have been difficult in markets with multiple trades during a day.

Box 11.2 W hat is a Tick er Tape?

A tick is any change, up or down, in the price of a stock (or any other traded security). Information regarding stock transactions is recorded on a ticker tape. By watching the ticker tape, an investor is able to keep abreast of changes in stock prices. The first ticker tape was developed in 1867, following the invention of the telegraph machine. This technology allowed information to be printed in a shorthand format on a narrow strip of paper, or tape. Messengers would run a circuit between the New York Stock Exchange (NYSE) trading floor to brokers’ offices, delivering tapes of the most recent transactions. Brokers would place an office near the NYSE because the closer they were to the trading floor, the more quickly they could get ticker tapes and the more up to date their information about recent stock transactions would be. Technology improved over the years, providing faster access to stock transaction data. In the hectic trading days of the late 1960s and early 1970s, the ticker could not keep up with trading, and there was a period when the markets were closed on Wednesdays to facilitate the clearing of trades. Not until 1996 did the real-time electronic ticker provide up-to-the-minute transaction data. Today, you will see immediate stock market transaction figures on TV news shows and Web sites. Although these figures are reported electronically, and the actual tape is no longer used, the name “ticker tape” is still used when a running list of trades is shown on TV or a quote machine.

When you are watching a ticker tape, you will see information recorded in a format such as this:

HPQ2K@23.l6-1.09

or

Ticker Symbolshares Traded@ Price Traded, Change Direction, Change Amount

The first information given is the ticker symbol; these unique characters identify a particular company’s stock. In the preceding example, HPQ indicates that this information is for the common stock of Hewlett-Packard Co. Next, the volume of shares traded appears. 2K indicates that 2,000 shares were traded. The price per share for the particular trade is then quoted. Next, an up or down triangle appears, indicating whether this trading price is higher or lower than the previous day’s closing price. Finally, the difference between the current trading price and the previous closing price is reported. Reading this ticker tape, we can tell that 2,000 shares of Hewlett-Packard stock just traded at $23.16 a share. We can also tell that this is $1.09 higher than the previous day’s closing price of $22.07.

Modern technology has greatly simplified the task of chart construction. Computer power has replaced much of the tedious human work. Now, even basic home computers have spreadsheet programs, such as Microsoft Excel, that can store daily stock price data and create a variety of charts used by technical analysts. In addition, other sophisticated software programs that are specifically designed for technical analysis are readily available. These programs not only plot charts and indicators or oscillators but also can test trading rules. Examples are AIQ, Amibroker (www.amibroker.com), eSignal (www.signal.com), High Growth Stock Investors (www.hgsi.com), Metastock (www.equis.com), Neuroshell Trader (www.neuroshell.com), Ninjatrader (www.ninjatrader.com), Thinkorswim (www.thinkorswim.com), Tradersstudio (www.tradersstudio.com), TradeStation (www.tradestation.com), Updata (www.updata.co.uk). and Wealth- Lab (www.wealth-lab.com). In addition to charting software, the Web hosts many charting sites, examples of which are StockCharts.com, www.bigcharts.com,finance.yahoo.com, and freestockcharts.com. Today, the technical analyst can focus much more time and attention on analysis and much less on chart construction.

Over the years, technicians developed several different approaches to chart construction. The four main categories of charting that we discuss in this chapter are line charts, bar charts, candlestick charts, point and figure charts, and other charts with no indication of time. Each approach has its own features, benefits, and drawbacks. Whichever method a technical analyst chooses to use, charts serve as the technical analyst’s road map. Charts give a quick and concise picture of past price action.

For example, look at Table 11.1. This table contains the daily reported prices for Apple Computer (AAPL) for the month of May 2015.

It is difficult to look at the 20 closing prices in this table and get an idea of whether the stock price trend is up, down, or sideways. Now look at Figure 11.1. This chart contains the same information as Table 11.1. Notice how much easier it is to process the information when it is provided in the picture form of Figure 11.1 rather than in tabular form. As the old saying goes, “A picture is worth a thousand words.” With just a glance at the chart, you have a road map of where prices have been; in a fraction of a second, you can easily spot the highest prices and lowest prices that occurred during the period. A chart quickly transforms a table of data into a clear visual representation of the information.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

To construct reliable charts, the technical analyst must be sure of the trustworthiness of the data. During a normal trading day, many errors appear on the tape, from which most data originates, that someone must screen out and adjust. Some trade reports show the wrong price or volume; some trades are in error and must be broken; and some trades occur out of order. When a price error occurs at a high or a low for the trading day, the error is especially troublesome because it affects daily calculations of averages and oscillators using highs and lows. It is, therefore, important that any data used for charting is extremely “clean” and reliable.

In addition to trading errors, other data errors can occur through stock splits, dividends, offerings, and distributions. In the commodities markets, because contracts have a settlement date at which trading halts, incorrect calculation of the time and price linkage between contracts may affect longer-term technical patterns and trends. Calculation of these linkages into time series data for a longer-term perspective is never precise. The results are called either “nearest future,” “perpetual series,” or “continuous series” and are provided by many data vendors and exchanges (see Box 11.3). There are other methods of joining the different contracts into a continuous series, but all have serious problems for the analyst wanting to test longer periods of data. Schwager (1996) recommends the continuous contract as the best method for such tests.

Box 11.3 Linked Contracts

For short-term testing and analysis of futures contracts, you can use the actual contract data.

The test period must be extremely short, however, because the liquid portion of any contract, the period when active trading occurs, is only a portion of the contract time span. Thus, the data available for realistic study is short, and to test a number of signals, it must be separated at short intervals.

Once the trading signal horizon exceeds an hour, however, one must calculate a linked contract to provide sufficient data for testing. There are three basic types of linked contracts: the nearest future, the perpetual, and the continuous. Each has its advantages and disadvantages.

The nearest future method is just a plot of each futures contract as it expires and is replaced by a new contract. Unfortunately, during the transition between old and new contract, a gap always exists at the rollover into the new contract. Thus, this method is useless for testing trading systems, even though the contracts plotted are historically correct. It is the least preferable method of analyzing longer-term futures price moves.

The perpetual contract (also known as the “constant-forward”), to avoid the nearest future rollover problem, uses a constant forward price—namely, the anticipated price some specific period ahead. This forward contract comes from the interest and foreign exchange markets where constant-forward contracts are traded. The adjustment to the futures markets assumes a price that adjusts for the time between a contract series that is current and one that is beyond the specific constant-forward period. The perpetual price then is the price of the current contract and the forward contract each weighted by the amount of time remaining in the hypothetical forward period. Thus, the perpetual price avoids the problem of the rollover by smoothing the two contract prices—one near and one far out—in a gradual manner as time progresses. The problem with this method is that it is not real. The actual prices recorded in the perpetual contract never occurred. It is, thus, not a suitable method for testing a technical trading system.

The third method, the continuous contract, is more realistic but is useless for calculating percentage changes over time. It adjusts for the premium difference between the current contract prices—the price of the previous contract and the contract into which the trader rolls his position at a specific rollover date, say 15 days before expiration—to avoid the trading bias that occurs as the contract nears its end. This continuous contract then carries the adjustment into the future. It reflects exactly what would have occurred to a portfolio that invested in the first contract and rolled over each contract at its rollover date. It is, thus, a realistic expression of the history of the futures contract and can be used for testing past data. It has two problems, however. One is that because the adjustments are additive, the continuous contract cannot be used for percentage returns; and two, the ending price of the continuous contract is not the same as the current price of the current contract. The continuous contract is the primary contract used for testing trading systems.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

In early markets, when the price for a security or good might be established only once or twice a day, the chart was extremely simple. It was merely a graph of closing prices connected by a line, sometimes directly connected, and sometimes connected by perpendicular lines. In Japan, this type of chart was called a “tome” chart from the word tomene, which means “close.” In the Western world, this type of chart is still used and is called a line chart.

When trades became more frequent, chart forms took two paths. The first and still most common style was borrowed from the bar graph or stick graph and portrayed the high and low with a “floating” vertical line not connected to the base line. These were called bar charts or vertical line charts. Interestingly, in Japan, where many chart types were developed, the original plotting of price data was from right to left rather than the now- universal method of plotting from left to right. This bar style then evolved into the candlestick chart, which uses the same information as the bar chart but has a more appealing appearance. The other path was called the price movement line, where prices were recorded as they occurred, and only the ones that deviated from earlier prices by a specified amount were graphed in a line. This method of price plotting is the forerunner to the modern point and figure chart.

Today, the three most common types of charts that record prices at given time intervals (such as hourly, daily, weekly, or monthly) are line charts, bar charts, and candlestick charts. Let us look at each of these charts and others to see how they differ.

1. Line Charts

Figure 11.2 is an example of a line chart. These simple charts provide information about two variables: price and time. In the figure, the price variable is the daily closing price for AAPL (Apple Computer). A line chart has price data on the vertical, or y, axis. On the horizontal, or x, axis, it has a time measure (hours, days, weeks, and so on).

Simple line charts are especially useful when studying long-term trends. Because line charts display summary statistics, they are often used when information about several different variables is being plotted in the same graph. For example, in Figure 11.3, three line charts are used to plot the weekly close of the Dow Jones Industrial Average, the S&P 500, and Nasdaq for the past two years. Journalists often use line charts to give the reader a quick, concise picture of the variables being discussed.

Figure 11.3 represents weekly data. Line charts, however, can be used to present data collected at any time interval. More frequent data collection will lead to a more detailed, but more cluttered, graphical presentation. Especially when studying long-term trends, these extra details muddy the picture and obscure basic trends. For example, compare Figures 11.4, 11.5, and 11.6. Each of these charts represents trading data for AAPL (Apple Computer) over the 22 months from July 2013 through May 2015. The first chart uses daily data, the second chart displays weekly data, and the third chart presents monthly data. See how broader, longer-term movements in the stock price are more easily discernible in the third graph, which relies on less frequent data observations.

2. Bar Charts

Although the line chart visually displays one piece of information for each time interval, a bar chart shows at least three pieces of information: the high, the low, and the closing price for each time interval. Some bar charts also contain a fourth piece of price information: the opening price. Each time interval (that is, day, week, or five-minutes) is represented by one bar.

Figure 11.7 is an example of a daily bar chart. Each bar represents one day’s price action. Just as with the line chart, price data is placed on the vertical axis, and time is measured on the horizontal axis. A vertical line shows the trading range for that day. The top of this vertical line represents the highest price at which the security traded on that day; the bottom of the bar represents the lowest trading price of the day. A longer line denotes a wider trading range during the day. Likewise, a short bar means that the spread between the highest price during the day and the lowest price during the day was small. A small tick mark on the right side of the bar indicates the closing price for the day. If the opening price for the day is recorded on the bar chart, it is represented by a small tick mark to the left side of the bar.

We see that the first bar in Figure 11.7 represents trading information for AAPL January 2, 2015. The lowest point of the bar is $107.35, which is the lowest price that a share of AAPL traded for on that day. The highest price anyone paid for a share of AAPL that day was $111.44, represented by the top of the bar. The difference between the high and low price in any bar is called the range. The opening price for AAPL was $111.39, represented by the left hash mark. The right hash mark at $109.33 represents the closing price.

We can glean a lot of information from a quick visual observation of the bar chart. For example, the bar for trading day 2 in Figure 11.7 closed lower than it opened, and it opened below the close of day 1, indicating the trend is downward. Another quick observation is that the bar for trading on day 5 has a lower high than the low of day 6. This space is called a gap. Thus, the bar chart makes it easy to spot changes in trend and price action from bar to bar, certainly more easily than the column of numbers in Table 11.1.

Just as with the line chart, bar charts can be constructed for various intervals of data collection. For example, Figure 11.8 presents a weekly bar chart for AAPL for the same period as Figure 11.7. When we use longer and longer intervals to gather information for our bar charts, we lose some of the details but will have a less cluttered chart that offers more of a broad-stroke picture of past price movement.

3. Candlestick Charts

As mentioned in Chapter 3, “History of Technical Analysis,” candlestick charts originated in Japan. This charting method was used as early as the mid-1600s to trade rice futures in the Japanese markets and continues to be the most popular form of technical analysis in Japan.

These techniques have been widely used in the Far East for many generations, but not until the publication of the book Japanese Candlestick Charting Techniques by Steve Nison in 1991 were Western traders introduced to candlestick charts. Before the publication of Nison’s book, few U.S. and European services offered candlestick charts. Today, almost every technical analysis software package and technical service offers candlestick charts. You can even create candlestick charts through the charting options in Excel.

Candlestick charts are similar to bar charts in their construction. Both charts use the high price, low price, and closing price, but candlestick charts always include the opening price. To construct a candlestick chart, the low and high prices are plotted on a thin bar, just as they would be for the bar chart we just discussed. A box is used to represent the opening and closing prices. To create this box, a horizontal mark is made at both the opening and closing prices; a rectangle is formed using these two horizontal marks. This rectangular box is called the real body of the candlestick. If the security closed at a higher price than it opened, the real body is white (gray in the charts here) or open. These white, or “open,” real body candlesticks indicate price advances from the opening. Conversely, if the closing price falls below the opening price, the real body of the candlestick is shaded black. These candlesticks with a “closed,” or black, real body designate price declines from the opening.

Figure 11.9 is a candlestick chart of daily prices from January 1 through May 29, 2015, the same as for Figure 11.7 and Figure 11.8, for AAPL. Much more colorful than the bar chart, the candlestick chart makes it immediately easy to spot days in which AAPL closed at a higher price than it opened. For example, the candle for the first trading day of the chart, January 2, has a black body, indicating that the stock closed at a lower price than it opened. The following day we see another black-bodied candlestick, indicating that the stock closed lower than it opened that day also. Not until the fourth day does a gray body appear showing that the close was higher than the opening.

As you can see in Figure 11.9, candlesticks come in a variety of shapes and sizes. If the real body of the candlestick is tall, the opening price and closing price were relatively far apart. Shorter real bodies indicate opening and closing prices that were similar. In the extreme, the real body can be so short that it is just a horizontal line, indicating that the opening and closing prices were identical.

The thin vertical bars, representing the price extremes of the trading session, are called the shadows. The shadow above the real body is called the upper shadow; the shadow below the real body is called the lower shadow. You can easily see how the candlestick chart got its name; many times, the real body will look like a candle and the upper shadow will look like the wick.

Individual candlesticks can take on a variety of interesting sizes. Some have long shadows; others have short shadows. Some have tall boxes; other have short boxes. The color of the box, the lengths of the boxes and shadows, and where the box sits relative to the shadows tell something about the trading that occurred over the time period represented by the candlestick. Looking at the first candlestick in Figure 11.9, we see that this candlestick has almost no upper shadow; this, along with the black body, indicates that AAPL opened near the
high of the day. If the opening price had been the highest price of the bar, then no upper shadow would exist. The third candlestick in Figure 11.9 has a relative small black body; this indicates that the closing price was below, but close to, the opening price; the upper and lower shadows indicate that the price range throughout the bar was much greater than the difference between the opening and closing prices. On some bars, the opening and closing prices represent the entire trading range for the day, resulting in no upper or lower shadow, as we see with the tenth candlestick.

Because candlestick charts contain all the information that a bar chart contains, all of the technical tools that are used with bar charts can also be used with candlestick charts. In addition, some technical tools rely on the color and size of individual candlesticks to signal trades. We study the trading techniques that are peculiar to candlestick charts in Chapter 17, “Short-Term Patterns.”

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Price units are generally plotted on the vertical axis of stock charts. The analyst must determine the scale, or the distance between these price units, to use. Generally, two types of scales are used: an arithmetic scale and a semi-logarithmic scale.

Arithmetic Scale

For all the charts we have examined so far, along with most technical analysts, we have used arithmetic (or linear) scales. A plot with an arithmetic scale shows the price units along the vertical scale at the same price intervals. For example, the vertical plot distance of a change of $1 to $2 would be the same as the plot distance from 10 to 11 or 100 to 101. In other words, using regular evenly divided grid paper, we plot each box vertically as the same dollar amount.

2. Semi-Logarithmic Scale

Although the arithmetic scale is the scale most often used, sometimes adjustments need to be made, especially when observing long-term price movements. For example, compare Figure 11.10 with Figure 11.11.

Both these charts plot the daily price of Crude Oil Futures for July 1, 2014 through May 29, 2015. In July 2014, crude oil was trading around $100 a barrel; at that time, a $10 increase in price would represent approximately a 10% gain for the investor. By January 2015, the crude oil price had declined to about $50 a barrel. At that point, a $10 price increase represented about a 20% gain for an investor owning crude oil. On the arithmetic scale, in Figure 11.10, a $10 price movement is visually the same whether it is a move from $50 to $60 or a move from $100 to $110. This type of scale can be somewhat deceptive; a $10 move is much more significant to an investor if the price of a security is $50 than if the price of the security is $100. The logarithmic scale addresses this issue.

On the logarithmic scale, the vertical distance represents the same percentage change in price. Look at Figure 11.11. In this logarithmically scaled graph, the vertical distance between $50 and $60 is larger than that between $100 and $110. This vertical distance between $100 and $110 always represents a 10% increase in price rather than a particular dollar amount increase in price. The rule of thumb for when to use an arithmetic or logarithmic scale is that when the security’s price range over the period being investigated is greater than 20%, a logarithmic scale is more accurate and useful. As a rule, the truly long-term charts (more than a few years) should always be plotted on logarithmic scales.

Another important difference in the two scales is how a trend line appears. In the arithmetic chart, the downward trend line touching the two down arrows arrives at the bottom of the chart in February and slices through the first rally in crude oil from its first bottom giving a false buy signal. In the semi-logarithmic chart, the trend line drawn from the same locations arrives at the bottom in April and gives a true signal. Regardless of the signals, the location of the trend line in the two types of scales is different and can affect the analysis of the prospects for crude oil.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

The point and figure chart records price data using a different technique than line, bar, and candlestick charts. At first, it may appear that the construction of these charts is somewhat tedious. In addition, these kinds of charts are rarely published or discussed in the popular financial news. Because many of the point and figure charts are constructed using intraday trading data, use of these charts was historically limited to professional analysts who had access to intraday data. However, with some practice, you will see that point and figure chart construction is not that difficult and provides an interesting and accurate method of price analysis.

Point and figure charts account for price change only. Volume is excluded; and although time can be annotated on the chart, it is not integral to the chart. The original point and figure charts took prices directly from the tape as they were reported in the “Fitch Sheets” and by services prepared especially for point and figure plotting, such as Morgan, Rogers, and Roberts. Most of these services were put out of business by the computer and the use of the three-box reversal charts. The reading of each stock or commodity trade by trade was a laborious process. Today, few services provide the data to plot one-point reversal charts.

The origin of point and figure charts is unknown, but we know they were used at the time of Charles Dow around the late nineteenth century. Some have thought that “point” came from the direction of the entries on the chart, pointing either up or down, but more likely “point” refers to the location of the price plot, which at first was just a pencil-point mark. “Figure” comes from the ability to figure from the points the target price.

Construction of a point and figure chart is simple because only prices are used. Even then, only the prices that meet the “box” size and “reversal” size are included. Finally, the chart reflects the high and low of the period, whenever it is important. Many technicians believe that the high and low of the day are important numbers determined by supply and demand. In fact, they believe these numbers are more significant than the opening or closing prices, which occur at single, arbitrary moments in time.

As with all the charting methods, different analysts use variations of point and figure charting to best meet their particular needs. We begin our discussion of point and figure chart construction by looking at the oldest method, referred to as the one-point reversal point and figure method. Further information about these one-point reversal charts can be found in reprints of Alexander Whelan’s Study Helps in Point and Figure Technique.

1. One-Box (Point) Reversal

All point and figure charts are plotted on graph paper with squares that form a grid. There should be enough squares to include a significant period of trading activity. Early charts had a special outline around the rows ending in 0 and 5, just for clarity. As with the other types of charts, we will plot price on the vertical axis, but the bottom axis is not time scaled with the point and figure graph.

A point and figure chart is posted when a market price either reverses in direction by the amount of each square or continues in the same direction as the immediately preceding box. Figure 11.12 shows a one-box reversal point and figure chart of AAPL from April 28 through May 29, 2015. In this example, one box is equal to H point or dollar. A reversal requires the price to reverse in direction by H point, the amount of a box. Had the box been worth 3 points, the reversal would have required a 3-point reversal to reverse. A reversal is recorded in the column to the right of the operating column. For example, in AAPL, if the last price recorded in a column was in a rising column (an X) at 125 and the price declined to 124.50, the plot (an O) would be placed in the next column to the right at 124.50. To record a new entry, the price must have traveled through the imaginary line either above or below the box. Each price is recorded in this manner as it travels up and down the box scale, and eventually a series of patterns evolve that can be analyzed. Instead of declining to 124.50 in the example, if the price had risen to 125.50, a box size above the last plotted box, an X would be placed in the 125.50 box, the next higher box, because the trend is still upward. This could continue until a reversal from one of the boxes by H point occurred. If the price rises to 125.45, it is not recorded in the next higher box because it did not break the upper line of the box at 125.50. Likewise, if the price then declined H point from the 125.45 to 124.95, it would not be recorded because it did not break the lower bound of the next lower 124.5 box.

Box 11.4 How to Construct a Point and Figure Chart

The best way to learn to read a point and figure graph is to walk through an example of how this type of graph is constructed. Let us begin by taking a series of price changes in a stock of 43.95, 44.10, 44.3, 44.15, 44.5, 44.7, 44.9, 44.85, 44.95, 45.00, 45.05, 44.4, and 43.9.

Each square (now called a “box”) on the graph paper will represent one point in the price. In point and figure charting, the plot is made only when the actual price of the box is touched or traded through. In this example, 43 would not be plotted because the price never reached 43 exactly or traded through to below 43. Forty-four would be plotted because the price ran from 43.95 to 44.10, trading through 44.00. Thus, our first plot for the point and figure chart would be placing an X in the 44 box when the price of 44.10 is observed, resulting in a chart that looks like Plot 1 in Figure 11.13. For the next seven reported prices, no mark is made on the chart because all these trades are between 44 and 45. When the tenth price, 45.00, is observed, a second X is plotted because the price actually touched 45. This X is plotted in the 45 box in the same column, resulting in a chart that looks like Plot 2. We now know that this first column is recording an uptrend in the stock price.

As long as the observed prices range above 44 and below 46, no more marks are made on the graph. For example, the next prices recorded in our sample data are 45.05 and 44.4. Because neither the next higher number (46) nor the next lower number (44) has been reached, no mark is made to represent this price observation. These trades are considered “noise,” and the point and figure chart eliminates the plotting of this noise data.

It is not until the next price of 43.9 is observed that another mark is plotted. The price has now reversed downward through 44. Obviously, there is already an X at 44 in Column 1. Column 1 represented an uptrend in the price, and only price increases can be recorded in it. Therefore, we move to Column 2 and place an X at 44, as is shown in the figure below Plot 3. At this point, we don’t know whether the trend in Column 2 is upward or downward. The second posting in this column will tell us. If the price should now rise to 45 again, we would place an X at 45, and Column 2 would record rising prices. If the price should decline to 43, we would place an X at 43, and Column 2 would record falling prices.

Let us say that the price declines in a steady stream with no one-box reversals to 39.65, and then it rallies back in a steady stream to 43.15. This would be represented as Plot 4. A plot is made only in a new box when the price is trending in one direction and is then moved over and plotted in the next column when that price reverses by a box size and cannot be plotted in the same column. Remember that a particular column can record only price increases or price declines; in our example, Columns 1 and 3 represent price increases and Column 2 plots price declines.

2. Box Size

From this basic method of plotting prices come many variations. Box size can be expanded; in our example, the box size is expanded to two points and labeled 120, 122, 124, 126, and so on. Then a two-point change in direction would be required before prices moved to the next column. Increasing the size of the box reduces the amount of noise. Gradually, as the box size increases, the amount of price history becomes smaller and the chart becomes squeezed to the left, as fewer columns are necessary. The elimination of the noise makes the chart more useful to traders or investors interested in longer periods of time and activity. On the other hand, if a pattern appears to be developing in the longer-term chart, the box size can be reduced to give more detail near the potential longer-term change in direction. This more detailed view can give early signals based on what the longer-term pattern is forming.

3. Multibox Reversal

The other variable in a point and figure chart is the reversal amount. In our previous example, we used one point for both the box plot and the reversal amount. We could have expanded that reversal amount to three or five boxes, however. In other words, we could keep the one-point box scale but only record a reversal to the next column when the price reversed by three boxes. This also cuts down on the noise in price action and lengthens the time over which price action is recorded. Figure 11.14 shows a hypothetical increase in the number of boxes necessary for a reversal. Again, it reduces the noise and condenses the chart. One other attribute is that, unlike the one-box reversal, a complete data stream of prices is unnecessary, and the plot can be accomplished from data in the morning paper. It is for this reason that the three-box reversal became popular; it eliminated the tedious necessity of looking at every trade.

As an example, Figure 11.15 shows a one-point, three-box reversal chart of AAPL prices from January 1, 2014, through May 29, 2015. The plot itself is a little different from the pure one-point, one-box reversal chart. Xs are used for the column in which prices are rising, and Os are used for the column in which prices are declining. This gives an easier-to-read picture of the price history.

The three-box reversal method gained popularity when Abe Cohen and Earl Blumenthal of Chartcraft publicized it in the 1950s. More recently, Tom Dorsey has popularized this method in his book Point and Figure Charting, and the majority of point and figure analysts now use the three-box reversal method. Because the three-box reversal method is less concerned about small, intraday changes in prices, it is especially useful when daily summary (High, Low, & Close) price data is being used.

3. Time

In some point and figure charts, when a new price is first recorded for a new month, the first letter or the number of the month is plotted in place of the X or O. In other instances, the month is recorded at the bottom of the column in which a price is first recorded for that month. We can plot years, weeks, and days similarly, depending on how sensitive the chart may be to price changes. Both methods can be used concurrently. Often, when years and months are the principal periods recorded, the year will be plotted on the bottom of the column and the month number (1 for January, 2 for February, and so on) will be plotted instead of the X or O. Time is of little importance in point and figure chart analysis. In many cases, time is plotted only to see how long it takes for a formation or pattern to form.

4. Arithmetic Scale

Scale becomes a problem in plotting point and figure charts, especially when the price rises or falls a significant distance. Obviously, when a stock is trading at $70, a one-point move is less significant than if it was trading at $7. Blumenthal first introduced the solution to this problem in three-box reversal charts. He suggested that the chart scale be one point per box for prices between $20 and $100, one-half point per box for prices between $5 and $19.50, one-quarter point per box for stocks trading below $5, and two points per box for stocks trading above $100. This scale has since become standard in most three-box reversal charts. However, depending on the behavior of the stock price, the scale can be adjusted, and of course, it is useless in the futures markets where prices are considerably different.

5. Logarithmic Scale

As in bar charts, when long periods of trading activity are plotted, distortion arises from the fact that most charts are plotted on an arithmetic scale. Low-price action does not look as active as high-price action. Logarithmic scale changes the plot to include percentage change rather than absolute price change. Thus, the low-price action in percentage terms may appear more variable than the high-price action, and it often is. To account for percentage change in point and figure charts, prices are converted into their logarithmic equivalent and plotted as a logarithmic number. This makes immediate interpretation difficult unless a table of logarithmic equivalents is immediately handy because most analysts cannot convert the logarithmic number into the actual price in their heads.

This scale, however, should be used only for long periods of price data in which considerable volatility has made an arithmetic scale meaningless. For most investment and trading, the arithmetic scale is not only just as useful, but also easier to read and to convert to actual prices.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Japanese Cloud charts are a relatively new method of interpreting price data. They use calculations similar to Western charts such as moving averages, but they also use such calculations as ranges and areas between plotted lines. While the charts themselves are easily read by Westerners, their interpretation follows a definite set of quite different rules. Cloud charts are therefore more of a technique than a different brand of charts.

We think of Japanese charting as being ancient and only recently recognized in the Western world. Cloud charts, however, are a relatively recent method developed by a team of seven people in the 1930’s and not made completely public until 1968 when the team leader Goichi Hosada (1898-1982) published them in a seven volume series called “Ichimoku Kinko Hyo,” literally translated as “At a glance, balance, bar chart” (Linton, 2010). Later, a more easily read Japanese book, Table of Equilibrium at a Glance, was written and published in 1996 by Hidenobu Sasaki (in Japanese). Finally, and the source for this book, David Linton, Senior Partner of Updata plc (www.updata.co.uk) authored the best book on cloud charts in English: Cloud Charts: Trading Success with the Ichimoku Technique (2010).

The cloud chart includes five lines in different configurations (see Figure 11.16). The periods for lines are based on 9, 26, and 52 bars. Originally the chart was daily, and the periods reflected the six days in a Japanese trading week: 9 for the days in one and one half trading weeks, 26 for days in a trading month, and 52 for two months. A question arises as to whether the periods should be adjusted in the U.S. markets for five trading days per week. Other periods have been used as well for the calculations, but most available charts and software use the standard Japanese periods.

The shortest period line (Tenkan Sen) is the average of the highest high and lowest low over the past 9 bars and is plotted without a time adjustment. The second longest period line (Kijun Sen) is the average of the highest high and lowest low over the past 26 bars and is also plotted without a time adjustment. This line is often used as an indication of support and resistance and as a stop order price. The third longest period line (Chikou Span) is the current closing price shifted backwards 26 bars from the present. The final two lines enclose the “cloud.” They are the A (Senkou Span A) and B (Senkou B) lines. The A line is calculated as the average of the Tenkan Sen and Kijen Sen plotted forward 26 bars, and the B line is calculated as the average of the highest high and lowest low over the past 52 bars plotted 26 bars forward. The space that exists between these two lines on a chart constitute the “cloud,” an area where no trades should be initiated because it represents a time of equilibrium rather than trend.

From various crossings of these lines trading signals are generated. The signals are based on an assessment of the entire cloud picture and are never used solely by themselves. To become familiar with the meaning and importance of the various crossings and their implications takes study but is likely worthwhile. The technique is similar to a moving average crossover system with additional constraints.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Beyond point and figure charts being independent of time, several chart methods—mostly of Japanese derivation—ignore time. These are rarely used in investment technical analysis because they are more suited for short-term trading. They include Kagi, Renko, and Line-Break charts.

1. Kagi Chart

Similar to a point and figure chart in philosophy, the Kagi chart is an old Japanese version of charting that is independent of time and only plots when a change of direction occurs by a specific amount. Figure 11.17 shows a representation of Apple Computer (AAPL) over one week with a reversal amount of 0.5 points. This chart uses only high and low prices but can include just closes, usually in a longer-term chart. Reversal amounts traditionally were percentages, which can still be used, but today points or an average true range (ATR) (see Chapter 13, “Breakouts, Stops, and Retracements”) are the most common.

The Kagi chart is posted only when a reversal by the amount specified is recorded from a high or low. The trend direction is recorded by thick, called yang lines, or thin, called yin lines, similar to the X’s and O’s on a point and figure chart. For example, in Figure 11.17, the first closing price was 131.60 on May 22. The price rallied from that prior day’s close to 132.92, at which point a vertical yang line was drawn from the previous close. Because the trend was upward from the previous close, the line is dark. A reversal of more than 0.5 points then occurred to 132.14. Nothing is plotted until there is an upward reversal that establishes the low. When that reversal upward occurs, a horizontal line is drawn from the low point to the next column, and a vertical line is drawn from that point in the new column upward for at least the reversal amount and perhaps more until it also reverses downward and the process is repeated. When a reversal breaks a prior high (called a shoulder) or low (called a waist), the line thickness changes and makes the change in direction more easily seen by eye. Most signals come from changes in line thickness as the price breaks out up or down. Trend lines can be drawn, and support and resistance are shown at the shoulders and waists. Head-and-shoulders patterns, called 3 Buddha patterns, are frequent and easily recognized.

2. Renko Chart

The Renko chart, also of Japanese origin, avoids time and changes direction by specific amounts. Unlike point and figure or Kagi, the plot is composed of “bricks” plotted on end with only one brick per column. The brick is a standard width, and the height is based on the reversal amount specified. For example, as shown in Figure 11.18, the chart of Apple Computer again uses a 0.5 reversal amount. The first plot, on the closing plot from the day before, is 129.20. Every subsequent brick is then plotted based on 0.5 intervals from that 129.20. The following day, the first move of 0.5 points was downward, causing a downward brick that extended to 128.70. The price may have declined further, but not more than another 0.5 or another brick would have been drawn. At that point, the price can either continue downward and be recorded for every 0.5 amount it declines, or it can rise 0.5 above the top of the brick at 129.20. If it rises 0.5 above that level to 129.2, another brick is placed in the next column to the right.

Trend lines can be plotted, as can moving averages. Some traders ignore one-block reversals and consider any larger reversal as a likely trend change. Short reversals are support and resistance levels, and the reversal interval can be either a percentage or an average true range.

3. Line-Break Chart (2 or 3 Lines)

The line-break chart is just another Japanese method of avoiding the time interval in price action and focusing only on the price action itself. Like the others in this category, the plot depends on a reversal in price. Rather than use a specific price amount to distinguish the reversal, the 2-Line, Line-Break chart uses the previous two lines for its reference point. A line is a block-type bar similar to that used in a candlestick chart as the body. The width of the line is constant, based on the chart, and the height is based on the price change relative to the previous lines. Only closing prices are used. Thus, in Figure 11.19, the first bar is the close of the first day relative to the close of the earlier day. It was downward from the previous close and thus is colored black and marked on the chart as a black line. The next day the market rallied and a white line for “up” is recorded. From now on, only a close that has exceeded two block lows can be recorded as a reversal. If it is not a reversal, it is either within the range of the two prior lines and is thus not recorded, or it is up from the last close and is thus recorded as another white line. The fourth line is a black line because it closed below the lows of the prior two up lines. The trend is now downward, and only a cross above the most recent 2-line highs can cause a reversal upward. Thus, each day may or may not have a line drawn in it depending on its close versus the previous two lines if in the opposite direction or in addition to the line in the same direction. A 3-line break chart is constructed in the same manner but must break three previous lines to record a trend reversal. Signals can occur from conventional, classic patterns, support or resistance breaks, or reverses in direction.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

By the end of this chapter, you should

• Know why identifying trends is paramount to profits in securities
• Be able to recognize an uptrend, a downtrend, and a trading range
• Understand the concept of support and resistance
• Be familiar with the major methods of determining trends
• Be familiar with the major signals that a trend is reversing

We are now entering into the more controversial aspects of technical analysis: the analysis of trends, and in Part IV, “Chart Pattern Analysis,” the analysis of patterns. This is the “fuzzy” aspect of technical analysis. Because the observations or rules are not specific, they discourage most students very quickly. Rules in technical analysis come from many observations by many traders and investors. In general, the rules have remained unchanged since Dow’s time, and in reading some of the old masters back in the 1930s, one sees the same observations today. The advent of the computer has sped the process and has often eliminated rules that are quantifiable but turn out to be unprofitable. However, the basics remain essentially the same. The markets have near, intermediate, and long-term trends. Patterns still form in much the same manner as 50 or 100 years ago, and analysts interpret them in the same manner as in the past. The details may be different, and perhaps the methods of profiting depend on various trade-offs between risk and reward, but still the analyst must use the rules and decide on the entry and exit points. The difficulty of profiting from technical analysis is not with the rules themselves but with their application.

In all the following chapters in this book, it is important to remember that the observations and statements we make derive from our observations and those of other practitioners of technical analysis. Most trends and patterns are not mechanical methods that can be easily programmed and tested on computers. Generally, they take long periods of practice to be fully utilized. One of the major criticisms of technical analysis is that it has yet to be thoroughly computerized and tested. As we saw in Part II, “Markets and Market Indicators,” numerous relationships have been tested in the past but then break down as the future unfolds. The only constants, it seems, are that trends occur and that they are the source of profit when recognized and properly used.

All analysts occasionally make statements that appear to be fact, but in many cases, they are statements based on subjective observations and should never be blindly relied upon without a thorough investigation. Our discussion of trend, support and resistance, and pattern nuances will show where they can be in error or where interpretation can be particularly difficult. Rules have developed over the years that will help with interpretation. Nevertheless, the student, when he can, should test and experiment. Nothing in technical analysis, or any other investment analysis approach, is foolproof. Indeed, it is surprising how much money is invested using fundamental and technical theories that have not been tested or, when they have, have proven to be unprofitable. Most professionals, who have spent their lifetime in the study and practice of technical analysis, will assert, “There is no easy, magic formula for wealth!” Do not, therefore, expect the following observations and rules to be an easy means to profit. Study, have patience, and study some more. We suggest that the student trade on paper, and finally with small amounts of money. There is no need to rush. The markets are always there.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Remember that profit with minimum capital risk in the securities markets is the sole objective, and technical analysis is an effective way to profit as well as to control risk. In earlier chapters, we emphasized the importance of determining and riding the directional trend in the security markets. The key to profiting in the securities market is to follow these three steps:

1. Determine, with minimum risk of error, when a trend has begun, at its earliest time and price.
2. Select and enter a position in the trend that is appropriate to the existing trend, regardless of direction (that is, trade with the trend—long in upward trends and short or in cash in downward trends).
3. Close those positions when the trend is ending.

Trending is simple in concept, but it is difficult in practice. Almost all successful mechanical trading systems that have made millions for their investors have been based on the simple concept of jumping on a trend and riding it to its inevitable end. We discuss the principles behind some of these methods later in this chapter and in following chapters.

The principal caveat, however, in technical analysis, as mentioned previously, is that although the trend concept is easy to understand, its application is difficult largely because the determination of trend and trend reversal is, in many instances, a subjective decision that depends on one’s skill and experience in the securities markets and one’s ability to control one’s own emotions. Practice and mental anguish are the background of any successful technical analyst. The most expensive education in the world is likely the money lost in incorrect, sloppy, and undisciplined decisions. All market participants make mistakes, but the regimented professionals correct theirs quickly.

When we arrive at Chapter 15, “Bar Chart Patterns,” and discuss the various price patterns that have been observed, we will note that almost all patterns are a combination of trend lines—up, down, or sideways. It is, therefore, imperative first to understand trends and trend lines. In addition, all patterns are used to either confirm that a longer trend is still in control or warn that such a trend is changing. Patterns are, therefore, not trade signals explicitly in themselves, but they are the means of taking advantage of the underlying and, perhaps, changing trends.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Trends define a direction in prices. When we refer to a trend, we describe a directional trend, one of rising or falling prices from which a profit can be generated with a trend-following method. We refer to a sideways trend as a trading range or neutral area. These are the recognized terms for describing different types of trends. Trend-following techniques work poorly in nontrending markets. Most technicians prefer to use price oscillators and trade from outer bound of the range to the opposite bound when dealing with such patterns.

In the next several chapters, we look at prices from the positive, advancing perspective. By that, we mean that when we discuss trends per se, we assume an upward trend (an uptrend). In most cases, the description and rules of a downtrend are exactly opposite from those of an uptrend. It does not make sense to duplicate every statement for both trend directions. Likewise, when we discuss support and resistance, we discuss support and make the assumption, unless otherwise noted, that resistance is the exact opposite but in an opposing direction. We do this for readability and because most investors prefer to look at rising prices anyway, even though there is no rational reason for doing so.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

As we learned in Chapter 6, “Dow Theory,” Charles Dow was one of the first of the modern technicians to write about the fact that stock market prices trade in trends. Virtually all items that trade in free, liquid markets trade in trends. As noted by Dow, investors or traders must concentrate on the time horizon most favorable to their circumstances.

Trends are fractal in that their behavior is the same regardless of the period. Minute-to-minute trends behave exactly like day-to-day trends with only minor differences because of the understandable variation in liquidity over the shorter periods. Dow suggested there were three principal time horizons—the primary, the intermediate, and the minor—that he likened to tides, waves, and ripples. In fact, there are considerably more trend periods. Dow focused on the first two because he apparently believed no one could analyze the ripples. Today, some technicians recognize considerably more trends than Dow observed, but then he did not have the advantage of a computer that could track prices trade by trade.

Dow’s final, and perhaps most important, observation was that, by their very nature, trends tend to continue rather than reverse. If it were otherwise, first, there would not be a trend, and second, the trend could not be used for profit. This seems like a silly and perhaps too obvious statement, but it underlies almost everything the technician assumes when looking for the beginning or end of trends. It also vexes the academic theoretician who believes that price changes are random.

Any particular trend is influenced by its next larger and next smaller trend. For example, in Figure 12.1, we can see a well-defined uptrend in the stock of AAPL (Apple Computer). It is not a straight line upward, however. Within the rising trend are many smaller trends, both down and up, and if we look more closely, there are even smaller down- and uptrends within these. This is the fractal nature of trends. Notice also that the next set of trends below the long uptrend have larger rises and smaller declines. This is the effect the larger trend is having on the smaller trends. It is why the analyst, when studying any particular length trend, must be aware of the next longer and shorter trend directions. The longer trends will influence the strength of the trend of interest, and the shorter trends will often give early signs of turning in the longer. By definition, short-term trends reverse before medium-term, and medium-term trends reverse before long-term.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

As we know from basic economics, supply and demand establishes the price of any good. It is no different in the securities markets. Supply and demand, when sellers and buyers agree on a trade, determines prices. What does price, and especially price change, tell us? Presumably, if a substantial number of transactions occur at one price, the price is telling us that supply and demand are in a temporary equilibrium and that both buyers and sellers are satisfied. Of course, in the financial markets, long-term equilibrium is rarely reached. Prices constantly change, if only by miniscule amounts, as they move toward a theoretical equilibrium. They can oscillate in small increments or large; they can go up or down; or they can do both. Whatever the price movement, it is ultimately determined by the expectations and power of the buyers and sellers. If broad expectations are for a higher price in the future but with little or no capital to act, prices will remain as they are or even decline. Of course, expectations change, as does the power to act. Nothing is perfectly stable or constant in the markets.

When prices travel in a trend, called trending, they remain headed in one direction, and they tell us that there is an imbalance of demand and supply. Some will incorrectly say that there are more buyers than sellers or vice versa. However, in every transaction, there are an equal number of shares transacted, and, thus, there is always temporary equilibrium between buyers and sellers at that instant in time. What makes a trend is the power of the buyers or sellers—do they have enough stock or money?—and the aggressiveness or anxiousness of buyers and sellers—do they have specific information or deductions, rational or irrational, or are emotions of fear or greed propelling their action?

We know from behavioral studies that, psychologically, a positive feedback mechanism in our minds tends collectively to sustain a trend. In an uptrend, for example, buyers who have profited tend to continue being buyers, and new buyers, seeing what they have missed, also buy. The price trend continues upward. Eventually, over a longer period, prices revert to some kind of mean or value, but meanwhile, they trend up, down, or sideways. If, for example, prices are gradually rising, then buyers must have stronger positive expectations and be willing and able to place more money in the security. Contrarily, if prices are declining, sellers must have stronger negative expectations and larger positions to sell. The price trend, thus, tells us the amount of power, aggressiveness, and anxiousness there is in the marketplace to buy or sell each security. To the technical analyst, the basis for the expectations—and there are many—as well as the source of power, money, or stock, is largely irrelevant. The anticipating and “riding” a trend in prices, as long as it continues, is the way the technician profits.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Of course, the trend is never a straight line. Then it would be too easy to tell that it had reversed. Instead, the trend is a direction rather than a line. Many doubting participants often accompany this direction. Sometimes, it can be arbitrageurs who bet against the trend or who are merely trading the spread. More likely, it is investors or traders running out of money or stock or just holding back for a little while, hoping that the price will retrace back toward their orders. In other words, the security price oscillates back and forth in smaller trends along its travel in the larger trend. This makes determining when that larger trend is reversing a difficult decision because any signs of reversal may only be for smaller trends within the larger trend. Additionally, securities occasionally “rest” during a trend and move sideways as the earlier rise or fall is “digested” by all the different players. The psychology of what causes these spurts, stops, and retracements is an interesting study by itself, but again it is irrelevant for our present discussion. We simply want to know what our trend of interest is and whether there are signs of it ending or changing direction.

Peaks and Troughs

What is the simplest way to look at prices and determine the trend? The easiest is to look for peaks and troughs within a series of price oscillations. If the peaks tend to be higher than the earlier peaks, and the troughs tend to be higher than the previous troughs, the trend must be upward. As you see in Figure 12.2, it is that simple. If peaks and troughs are lower than previously, the trend must be downward. If the peaks and troughs are scattered, the trend is undeterminable, and if the peaks and troughs occur at the same relative levels, the trend must be a trading range.

It is much easier to look at price trends on a chart. As the previous chapter demonstrated, a table of data makes recognition of price order of any sort difficult. Most technicians use either bar or candlestick charts to draw lines representing trends. There are many ways to do this that are discussed later in the chapter. For now, let us begin with discussing sideways trends because they display very clearly an important technical concept called support and resistance.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.

Trading ranges (or sideways trends) occur when peaks and troughs appear roughly at similar levels (see Figure 12.3). The peaks cluster at a certain price level, and the troughs cluster at a certain price level below the peaks. The configuration usually occurs after a larger trend has come to a temporary halt. A trading range also is called a consolidation or congestion area or a rectangle formation. Charles Dow called small lateral patterns a line formation, and using it in the Dow Jones Averages had specific rules by which the averages had to abide for that designation. William Hamilton, Dow’s successor editor at the Wall Street Journal, thought the line formation was the only price formation with any predictive power.

1. What Is Support and Resistance?

When prices have been rising and then reverse downward, the highest point in the rise, the peak, is referred to as a resistance point, a level at which the advance has met with selling “resistance.” It is the level at which sellers are as powerful and aggressive as the buyers and halt the advance. When the sellers (supply) become more powerful and aggressive than the buyers (demand), the result is a subsequent price decline from the peak. A resistance level becomes a resistance zone when more than one resistance level occurs at roughly the same price. Prices rarely rise and stop at the same level. A single, high-volume price peak often defines a resistance point. Even then, however, because the high volume, especially if it is preceded by a sharp price rise, is a sign of speculation and emotion, and location within the price bar where large sellers actually begin to enter the market is unclear.

A support point is the opposite of a resistance point in that it is a single trough. At the support point, buyers become as powerful or aggressive as the sellers and halt a price decline (see Figure 12.4).

The concept of support and resistance presumes that in the future prices will stop at these recorded levels or zones and that they represent a remembered psychological barrier for prices. The zones will carry through time and become barriers to future price action. Not only will the zones carry through time, but once they are broken through, they will switch functions. Previous support will become resistance, and previous resistance will become support.

2. Why Do Support and Resistance Occur?

Have you ever bought a stock, watched it decline in price, and yearned to sell out for what you paid for it? Have you ever sold a stock, watched it go up after you had sold it, and wished you had the opportunity to buy it again? Well, you are not alone. These are common human reactions, and they show up on the stock charts by creating support and resistance. (Jiler, 1962)

Let us look at the presumed psychology behind a support level and see why it might carry into the future.

There is little question that a price trough is a point at which buyers overwhelmed sellers. In Figure 12.5, AAPL peaked at $133.60 on February 24, a potential resistance level, and then declined touching $121.63 on March 12, at which point it reversed upward but failed to reach the old supply level. Subsequently, the stock price fell back to $122.60 within a dollar of the earlier support on March 26, rallied to $134.54 within $0.86 of the first peak and resistance level, and fell again to the support level and halted at $123.36, only $0.76 from the previous trough. We now have two, well-defined resistance points and three support points: $133.60 and 135.54 resistance points and $121.63, $122.60, and $123.36 support points.

We can assume there are potential buyers between $121.63 and $123.36 because

1. In the next sell-off, those who sold short at the $134 level will be covering because they have seen that the price halted its earlier decline at about $122 and do not want to take the risk that it will rally again to $134 and wipe out their profits.
2. Those who had been watching the stock but did not buy it at $122 earlier will be satisfied that the decline to $122 is back to where they earlier had wanted to purchase it but “missed it.”
3. Those who sold the stock at the low of $122, when it declined from $134.00, saw the price immediately rise thereafter and wish to reenter a position at the price they sold it earlier.

Notice that none of these players is using a fundamental or other informational reason for buying the stock at $122. The reasons are purely psychological, but they are strong reasons by themselves. The presumption for technical analysts is that $122 has now become a support zone and that prices will stop declining at that level in the future. The presumption is that the more frequently prices halt at a zone, the stronger and more important that zone will likely be in the future.

A resistance zone will likely now also exist at $134 for similar reasons because sellers want to sell at that price: sellers who missed $134 before, sellers who bought at $134 and want their money back, and sellers who want to short the stock at $134 where it halted earlier. Support and resistance zones, therefore, are price levels where supply and demand reach equilibrium for unusual but persistent psychological reasons.

3. What About Round Numbers?

Ironically, when prices reach round numbers, the tendency to buy and sell increases. People think in terms of round numbers. Otherwise, why would Walmart sell a shirt for $29.95 rather than $30? They know people subconsciously will associate with the “29” and will believe they are getting a $29 shirt rather than a $30 one. People think in terms of round numbers and act accordingly in the securities markets as well. The current problem with the concept of round numbers is that knowledge of that tendency is widespread. From the standpoint of entering orders then, it is best to determine entry and exit points based on the technical situation rather than worry about round numbers.

4. How Are Important Reversal Points Determined?

The more important the reversal point, the more important the support or resistance level. There are a number of ways to identify a significant reversal point. Let us look at some of them.

In this section, we focus primarily on how to determine significant troughs and support levels. Of course, significant peaks and resistance levels would be determined in the same fashion, only in the opposite direction.

5. DeMark or Williams Method

Tom DeMark and Larry Williams each have a method of determining a reversal point by using the number of bars (in a bar chart) on either side of a suspected reversal point. In a low bar, for example, the analyst may look for two bars with higher lows directly on either side of the suspected trough bar. The number of bars on either side can be increased to boost the importance of the trough, but the number of troughs will be sacrificed. The higher the number of confirming lows necessary, the more important but less common the trough.

As an example, look at Figure 12.6. Each of the two-bar lows and highs is marked with an arrow. Point (a) is not a trough because it does not have at least two bars on either side of it with higher lows. Likewise, Point (b) is not a peak. It does not have two bars on either side of it with lower highs. Is Point (c) a trough? We do not know because we can’t see if there are two bars to the right of the low to judge Point (c).

6. Percentage Method

Another method of identifying significant troughs is by deciding beforehand how much the price should decline into and rally from the trough. A percentage is usually used. Using 1%, for example, any time the price declines more than 1%, makes a low, and then rallies more than 1% would define a 1% trough. The larger the percentage used, the more important but less frequent the reversal point.

7. Gann Two-Day Swing Method

W.D. Gann’s swing method is similar to the DeMark’s or Williams’s method. To find a support point, or trough, a low bar is identified. Once the low bar is identified, the two following trading days are observed. If these two days have higher highs than the low bar, the low bar is a support point. Originally, Gann used the three following trading days to determine a support point, but more recently, it has been switched to two days (Krausz, 1998). Likewise, a resistance point is defined as any high bar during an uptrend that is followed by two successive bars with lower lows. Figure 12.7 is identical to Figure 12.6 except that the reversal points are determined using the Gann rule. The difference between the two charts is that at Point (a), (b), (c), and (d), the reversal points as defined by the Gann rule do not occur at the DeMark/Williams reversal points. The reasons are that the days of the actual reversal points were not followed by the required two successive days. Thus, by Gann’s rule, the reversal may not occur on the actual high or low bar as at (a) and (e) in Figure 12.7.

8. High Volume Method

Very large volume can also identify a significant reversal point. High volume indicates that larger than usual activity occurred on that trading day. Figure 12.8 shows a one-day reversal on high volume at a high, creating a significant reversal point and resistance level.

Figure 12.8 illustrates a one-day reversal. One- or two-day patterns can occur at peaks or troughs. When these occur on high volume, they usually signify important reversal points. Because these formations usually occur at a stage of high emotion, they signify either a panic or a speculative bubble. As such, the actual price level at which the reversal took place is not identifiable on a large bar chart. Sometimes intraday action must be inspected to see at just what price level the majority of buying and selling occurred.

In Figure 12.8, the large spike in volume on April 28, 2015 occurred with a pattern called an outside reversal day, where daily range is large, the high is greater than the previous high, the low is lower than the previous low, and the close is near the low. These patterns can occur without the large volume, but when large volume is present, they are an excellent signal that the price rise has reached a speculative peak and will be a strong resistance in the future.

9. How Are Support and Resistance Zones Drawn?

To construct a support (or resistance) zone, simply draw a horizontal line through each significant trough (or peak) into the future. These lines can be drawn through the respective bar lows or, as Jiler (1962) suggests, using the bar’s close because this is what most investors read in the paper. These lines should also be extended into the past to see if earlier price declines stopped at the same price level. Where these horizontal lines bunch together, sometimes overlapping at the same price level is a support or resistance zone. This zone is usually stronger the more horizontal lines there are within it. In other words, the more times the price level has halted previous advances or supported previous declines, the stronger will be the resistance or support in the future. Because all previous significant troughs have not likely occurred at exactly the same price level, an area called a zone is constructed between the highest and lowest horizontal line. This defines the actual support or resistance area clearly.

FIGURE 12.9 Support and resistance zones (Nasdaq Composite Index (daily: October 7, 2014–March
24, 2015)

If a horizontal line is by itself with no other horizontal lines close to it, it is likely an independent support or resistance line. Such lines, unless accompanied by extraordinary volume, usually do not have the same strength in the future that a combination of horizontal lines might have within a zone.

In the future, prices will tend to halt at these zones, and occasionally at a single line. Prices will often enter the zone but will not break out of the outer horizontal line of the zone. If they do break that level, we have what is called a “breakout” that has important consequences. A price break above the resistance zone implies that sellers are satiated at that level and buyers are anxious. See the upward breakout in Figure 12.9 at the resistance zone in February. In this instance, the break left a vacuum of sellers, and the buyers, at least at that price, controlled the stock. If there is another resistance zone at some distance above the current broken zone, prices will generally trade up to that next higher zone. Thus, a resistance zone in an advancing market can become a price objective once a lower resistance zone is broken. Resistance zones exist at all horizons—day, week, or even minute-to-minute. Some traders only trade stocks or futures—especially e-mini futures— between these zones often on an intraday basis between very short-term support and resistance zones.

What we have said about resistance zones is equally applicable to support zones. Horizontal lines at significant troughs will show the existence of these zones, and extended into the future, they will become zones of support to price decline. In Figure 12.9, for example, the previous resistance zone became a support zone at Point X when the correction after the upward breakout halted at that same level. In addition, as time goes on, the importance of past horizontal lines diminishes for both support and resistance zones. More recent price reversals are more important. Human memory fades quickly.

10. How Do Analysts Use Trading Ranges?

Getting back to our earlier introduction of sideways trends, a trading range, as shown in Figure 12.9, is a price level where both support and resistance zones are relatively close together, and prices “bounce” between them until finally breaking out in one or the other direction. Some traders will trade the “bounces” between support and resistance, but this is usually dangerous and requires low operating costs and constant attention (Schwager, 1996). The most profitable and reliable way to use the trading range is to practice “breakout” trading. Let us look at each of these strategies a little more closely.

11. Range Trading

Trading within a range is difficult. Although many books suggest it as a strategy, it is almost impossible for the nonprofessional to profit through range trading. First, it is difficult to recognize that prices are trading in a range until after a considerable amount of trading and time has passed. It is, therefore, largely in retrospect that the opportunities are recognized. In addition, operating costs, such as commissions and slippage, must be small and execution efficient, or else any potential profit will be overwhelmed by transactions costs. Because the bounds of a trading range are often zones rather than specific price levels, the point at which an execution order, either buy or sell, should be placed is indefinite. Finally, the location for a protective stop-loss order to prevent a breakout from ruining trading profits is difficult to determine. By the time all these costs and execution levels are recognized, the potential for profit has diminished considerably, making any profit versus risk unlikely. Thus, most traders stay away from trading within a trading range and instead wait for the inevitable breakout and beginning of a trend.

The one exception to range trading is channel trading. A channel is a trading range tipped at an angle such that it trends upward or downward. Trend lines define the bounds of the channel just as support and resistance lines define the trading range. One can trade these channels back and forth but only in the direction of the channel trend. In other words, if the channel is trending upward, only long positions are taken at the lower bounds and sold at the upper bounds, but no short position is taken contrary to the channel trend. As mentioned earlier, upward trends have longer upward subtrends and shorter downward subtrends. To a certain extent, depending on the slope of the channel trend, the subtrends in the direction of the trend reduce the difficulty seen in trading ranges.

12. Breakout T rading

Breakout trading is as old as technical analysis and likely the most successful. Remember that a trading range is somewhat like a battleground, where the buyers and sellers are warring for dominance. Most chart patterns are combinations of trend lines and, thus, are battlegrounds also. Before the battle is over, it is almost impossible to determine who will win. It is usually wiser, and more profitable, to wait rather than to guess. Once, however, prices break out of the trading range, the investor has information about who has won the war. If the breakout is to the upside, buyers are driving price up; if the breakout occurs to the downside, sellers are overwhelming buyers. Trading on this breakout is probably the most profitable and reliable strategy for the investor faced with a trading range or pattern.

Breakout trading can be used in many different ways other than for just trading ranges. One of the most famous is the Donchian breakout method, also called the “four-week breakout system,” originated by Richard Donchian and later improved upon by Richard Dennis. It still appears to work. A recent study by Active Trader Magazine (Kurczek and Knapp, 2003) indicates that even though the method is popular and has been widely known for many years, it still produces profits, especially in the commodities futures markets. Its calculation is absurdly simple. Buy when the highest high over the past four weeks is broken, and sell when the lowest low over the past four weeks is broken. A “stop-and-reverse” strategy requires a position, long or short, at all times.

A breakout is a powerful signal. It indicates that the balance between demand and supply has been settled, usually violently, and, thus, is an indication of the initiation or continuation of a directional trend. The subject of how to assess breakouts is covered in the next chapter.

Source: Kirkpatrick II Charles D., Dahlquist Julie R. (2015), Technical Analysis: The Complete Resource for Financial Market Technicians, FT Press; 3rd edition.