Ignoring Time to Improve Your Timing

Timing matters. It matters when you are cooking dinner, when you are trying to launch a spacecraft, and, of course, when you are trying to enter or exit a trade. Technical analysis is filled with references to timing for good reason. Yet there are situations in which the clock itself becomes a distraction. Why insist that every price observation must belong to a particular minute, hour, or day when the market can make an important move whenever it pleases?

The answer is simple. Price movement matters more than the clock. A major change in sentiment does not wait for the next hourly bar or the opening bell. News, events, supply, and demand can produce a meaningful price change at any moment. By removing time from the chart, you can sometimes remove a considerable amount of irrelevant information as well. The techniques in this chapter may look unusual when you first encounter them, but they can produce surprisingly useful results because they force the chart to pay attention to movement rather than elapsed time. For traders who are particularly concerned with risk, this ability to suppress meaningless activity can be especially attractive.

Focusing on Tick Bars

A tick bar is built around transactions rather than minutes or days. A new tick appears when another trade occurs at a different price, and the size of that trade is not what matters. It might represent one share, one contract, or a thousand contracts. The important fact is that the price changed. The chart therefore does not care how long it took for that change to occur. If trading activity is light, there may be long stretches in which very little appears on the chart, and the software will usually compress those empty periods.

You establish the number of transactions required to create a tick bar. Suppose you select 100 trades. The chart will wait for those trades to occur before recording the next qualifying price movement. A security that trades heavily can produce ticks rapidly, while a thinly traded security may take considerably longer. The result is a chart that emphasizes the actual movement of price rather than the amount of time that happened to pass while traders were doing very little.

That gives the tick bar one of its principal advantages. You are looking at direction, but you are also getting some sentiment information built into that direction because the movement has to be supported by actual transactions. Tick bars can eliminate a substantial amount of noise, although they cannot eliminate every misleading move. A random jump in price can still create a false impression, particularly when the underlying market is thin or unstable.

Your charting program will generally allow you to specify the number of trades required to validate a tick. The attraction of the method is that it attempts to expose supply and demand only when the activity becomes large enough to deserve your attention. A sustained series of rising ticks, for example, suggests persistent buying pressure. Large institutions such as hedge funds, mutual funds, and other professional money managers have the capital to generate substantial transaction activity, whereas individual traders generally cannot produce the same scale of order flow. In that sense, the tick chart incorporates a measure of volume into the price bar without displaying volume in the conventional way.

Tick bars can also be useful when trading volume becomes unusually quiet. When only a handful of trades are occurring, a conventional time-based chart continues to print bars simply because another five minutes or another hour has passed. That can make a quiet market appear more active than it really is. Volatility indicators are particularly vulnerable because they compare current prices with prices from earlier periods. A sideways market with very little actual activity can therefore produce an exaggerated volatility reading. A tick chart compresses that inactivity, reducing your exposure to the false signals and whipsaw trades that can result from treating every passage of time as meaningful.

Momentum indicators can benefit from the same treatment. On a normal time-based chart, a large price move may have occurred because of genuine trading pressure, or it may have resulted from a brief rumor that disappears almost as quickly as it arrived. A tick chart gives greater weight to the fact that transactions actually occurred at new prices. It does not make the signal infallible, but it gives you another way to separate meaningful movement from the endless background activity of the market.

Tick bars remain relatively uncommon. Many charting packages do not offer them, and traders who are accustomed to ordinary daily or intraday charts may never encounter them. If your software does provide tick charts, however, you can apply most of the same indicators and patterns that you use on conventional charts. The important difference is that the chart is organized around trading activity rather than the clock.

Narrowing the Focus to the Move Itself: Constant Range Bars

Constant range bars approach the same problem from a different direction. Instead of waiting for a specified number of trades, you establish a minimum high-low range and allow a new bar to appear only when that range has been achieved. Every completed bar therefore represents the same specified amount of price movement. At first this can look strange because you are accustomed to judging a bar partly by its height. With constant range bars, that information has already been fixed for you. You must therefore stop relying on the height of the bar, the placement of the high and low, and several other conventional bar-reading techniques.

The benefit is that insignificant movements are filtered out. Small price changes that might otherwise lure you into a trade simply do not qualify. What remains is the movement large enough to meet your chosen range. This makes trends easier to see and can also make reversals more obvious.

Suppose a security is trading at $10 and you establish a $1 range. A move from $10 to $10.50 does not create a completed bar because the required range has not been reached. When the price reaches $11, the condition has been satisfied and a bar is recorded. The resulting chart consists of bars representing the same specified range, with the close occurring at one end of the bar.

Constant range bars are sometimes called momentum bars, or Mbars, because they are particularly useful for examining momentum. Momentum is one of the few classes of indicators that can sometimes provide information ahead of price itself, which explains why traders have been interested in methods that isolate movement rather than time.

More than one analyst has claimed credit for the basic concept of constant range bars. The source attributes the copyright to technical designer Danton Long, who introduced his version in 2003, while acknowledging that similar ideas had appeared earlier and that later analysts developed refinements. Whatever the history of invention, the practical limitation is more important to the trader: constant range bars are available on only a limited number of platforms, including TD Ameritrade, TradeStation, and CQG, along with certain broker platforms.

Defining a Constant Range Bar

The constant range bar records a high-low movement only after the minimum range you establish has been reached. If you specify 10 points, a price movement that fails to cover those 10 points does not generate a new bar. The purpose is to remove the small fluctuations that clutter a conventional chart and leave you with price changes that are large enough to matter.

The time represented by one constant range bar can vary enormously. One bar might encompass several minutes, while another might require hours, days, or even weeks before the required movement occurs. This is the important feature of the method. You are not forced to participate simply because the calendar says that another period has passed. A new bar appears because the market has actually moved far enough to create one.

Other indicators can be applied to constant range charts, particularly support and resistance lines and channels. What you cannot do is assume that all of the ordinary bar-reading measurements remain meaningful, because the bars have been constructed under entirely different rules. If you want to investigate range bars further, there is plenty of material available online, including demonstrations on YouTube. Just remember that enthusiasm for a particular charting technique can quickly turn into salesmanship.

Establishing the Conditions for a Constant Range Bar

Several conditions must be satisfied before a new constant range bar is created. First, the market has to achieve the minimum high-low range that you specified. Each bar represents a block of price activity meeting that requirement. The price also has to move at least one tick beyond the previous bar before another bar can begin, regardless of how long that movement takes and regardless of the amount of volume involved. If the market simply drifts sideways within the same area, the chart can contain a large empty section, although the software will normally compress it.

There must also be a distinct movement outside the range represented by the previous bar. The purpose is to make sure that the next bar represents an actual change rather than a continuation of meaningless congestion. Finally, the close must occur at either the high or the low. This last requirement can be somewhat difficult to visualize, but it is consistent with the goal of identifying a meaningful change in momentum rather than ordinary back-and-forth price activity.

Consider a security with an average true range of 120 points. You might decide that a meaningful move should equal half of that amount, so you establish a constant range of 60 points. A new bar would then require the appropriate opening displacement from the previous bar, perhaps 20 points, together with a high-low range of at least 60 points. If the market takes several hours, several days, or even several weeks to satisfy those conditions, you simply wait. The chart has done its job by eliminating a large amount of noise and leaving you with price movement that meets your definition of significance. You have not spent that intervening period manufacturing trades in a sideways market and exposing yourself to unnecessary whipsaw losses.

Catching the Big Move: Point-and-Figure Charts

Point-and-figure charts, usually abbreviated P&F, have been used for more than a century and are available through most modern charting programs. The method is generally believed to have originated with floor traders in the late nineteenth century who needed a way to record meaningful price changes before electronic charting became available. Rather than attempting to capture every tiny fluctuation, P&F concentrated on the price movements that mattered.

That principle remains the heart of point-and-figure charting today. Time is removed from the chart, and only significant price movement is recorded. A significant movement is defined according to a specified amount above a recent high or below a recent low. Minor movements are ignored completely. They are not represented by a smaller mark or a faint line. They simply do not exist on the chart.

This gives P&F a peculiar but useful quality. A reversal is not recorded until the price has moved far enough in the opposite direction to satisfy your definition of a meaningful reversal. That can increase your confidence that a change in direction deserves attention. It also means that P&F charts look radically different from ordinary bar charts. Once you become familiar with the format, however, the simplicity can become remarkably appealing. P&F is often described as a computational method unique to financial prices, which captures something important about its character.

P&F is generally more appropriate for traders holding positions for weeks or months rather than minutes. Whether it belongs in your own trading process cannot be determined from theory alone. Apply it to the securities you actually trade. Examine how many trades it would have generated, how long positions would have been held, and what the resulting gains and losses would have looked like. The chart has to earn its place in your methodology.

Visualizing What Actually Matters

A conventional chart contains an enormous amount of information, and much of that information may be irrelevant. A daily chart prints another bar every day whether the security moved significantly or barely changed. P&F does not give the market credit for simply allowing another day to pass. If nothing important happened, nothing appears on the chart.

The result is a record of the useful price information with much of the filler removed. Breakouts, reversals, and other chart events often occur in response to real-world developments, which gives P&F an event-driven character. The chart records the consequence of the event rather than forcing you to watch every uneventful period that came before it.

Putting Each Move Into Its Own Column

A new P&F entry is made only when price moves sufficiently above the previous high or below the previous low. When the security is moving sideways and failing to establish a meaningful new extreme, there is nothing to record. The absence of an entry is itself useful information because it tells you that the market has not yet produced the movement required to alter the chart.

An X represents an upward movement, while an O represents a downward movement. The marks are arranged in columns, with each column representing one continuous directional move. A column of Xs therefore represents an advance, and a column of Os represents a decline. The chart alternates between these columns as the market makes sufficiently large moves in opposite directions.

The format initially appears foreign because the dates along the bottom no longer behave like the dates on a conventional chart. But after some exposure, the information becomes quite direct. A vertical column of Xs tells you that the market is advancing. A column of Os tells you that it is declining. You do not necessarily need to know whether that particular column took three days, three weeks, or three months to form because the method is deliberately designed to make elapsed time secondary.

Imagine that a security has a high of $9 today and you are beginning a new chart. You enter an X at the $9 level. If the next day's high reaches $10, another X is placed in the same column at $10. If the following day's high reaches $12, you add the appropriate Xs for the movement from $10 to $12. You continue adding Xs to that column for as long as the upward movement remains intact.

Eventually the advance ends. When the price falls sufficiently below the previous low to satisfy your reversal requirement, you stop adding Xs and begin a new column of Os. The first O is placed at the appropriate price level, and subsequent qualifying declines are recorded beneath it. The creation of that new column is important because it tells you that the market's price dynamics have changed enough to satisfy your reversal definition.

A P&F column can therefore represent virtually any amount of calendar time. On a daily data source, one column might represent two days, ten days, one hundred days, or some other period. The chart does not begin a new column merely because another day has passed. It begins one because the existing directional movement has ended and the market has moved sufficiently in the opposite direction.

This is why periods of inactivity simply disappear. Instead of showing ten days of sideways movement as ten nearly identical bars, P&F skips the period. The dates shown along the bottom can therefore jump forward by several days, weeks, or months. Those dates are included mainly for reference. In P&F analysis, the passage of time itself is not the information you are trying to capture.

Dealing With Box Size

The box size determines how much price movement is required before another X or O can be added. It is the minimum distance above the most recent high, represented by an X, or below the most recent low, represented by an O, that qualifies as a new entry. Getting this number right is one of the most important decisions in constructing a P&F chart.

Defining the Box Size

The horizontal axis compresses time while the vertical axis retains conventional price spacing. Suppose a security has recently reached a high of $10 and normally moves about 50 cents per day. You might establish a $1 box size. If the price then reaches a new high of $12, the movement is significant enough to record. It is four times the normal daily movement and a full $2 above the previous high, so you would enter two $1 boxes in the X column.

A move of only 98 cents does not qualify. It may look close, but close is not enough. The purpose of the box is to create a hard boundary between movement you consider meaningful and movement you consider noise.

Choosing the Right Box Size

The smaller the box, the more sensitive the chart becomes. You will see more detail and more small retracements, but you will also allow more noise into the picture. Increasing the box size filters out progressively more minor movement and gives you a broader view of the trend. A trader who is particularly concerned about missing smaller changes may prefer a smaller box, while someone interested in the larger trend may prefer a larger one.

The source gives the following approximate guidelines:

Security PriceBox Size$5 to $20$0.50$20 to $100$1$100 to $200$2$200 to $300$4$300 to $400$6

Do not automatically accept the default box size supplied by your charting software. A program may adjust the box to fit the screen and give you an arbitrary number such as $0.67. Box size is too important to leave entirely to a computer. There is another practical consideration as well. If other traders are using conventional round-number boxes such as $1, using an arbitrary number can make your chart behave differently from the charts being watched by other market participants.

Adding the Reversal Amount

The box size tells you how much movement is required to add another entry, but it does not tell you when to change from Xs to Os. For that you need a reversal amount. The traditional reversal requirement is three boxes.

Suppose your box size is $1 and the security is currently forming a rising column of Xs. The price must decline by $3 from the relevant high before you abandon the X column and begin a new column of Os. You can test different combinations of box size and reversal amount through historical data, but P&F practitioners commonly retain the three-box reversal while changing the box size according to the security.

Applying Patterns to P&F Charts

Patterns become surprisingly easy to recognize on P&F charts because much of the distracting price movement has already been removed. Support and resistance are particularly useful, but familiar formations such as double and triple tops and bottoms can also appear. The difference is that the P&F chart can sometimes identify the significance of these formations without requiring the same amount of waiting that conventional time-based charting requires.

Support and Resistance

P&F provides two broad ways of viewing support and resistance. One is horizontal and based on historical price levels. The other uses sloping lines similar to conventional support and resistance. Both can help identify breakouts and establish possible stop locations.

Horizontal Support and Resistance

The upper and lower boundaries formed by the ends of P&F columns often become natural areas of resistance and support. Several columns may repeatedly stop around the same ceiling or floor. When price finally moves through that level, the breakout becomes easier to identify because the historical boundary has already been established by the chart. On a conventional chart, you might have needed to search backward through the price history to discover that the level existed. P&F traders commonly mark these historical ceilings and floors because they represent places where supply has repeatedly overwhelmed demand or where demand has repeatedly stopped a decline.

Conventional Support and Resistance

You can also draw sloping support and resistance lines on a P&F chart. The basic interpretation is familiar. An upside breakout can provide a purchase signal, while a downside breakout can provide a sell signal. And once a support line has been broken, do not automatically erase it. Former support frequently becomes resistance when price returns to that area.

The 45-degree line is particularly useful in P&F charting. With only two columns, you can sometimes establish a projected support or resistance line when one column is one box higher or lower than the preceding column. An upward 45-degree line is called a bullish support line, while a downward 45-degree line is called a bearish resistance line. This technique can allow you to establish a meaningful line earlier than you might on an ordinary bar chart.

Double and Triple Tops and Bottoms

Double and triple tops occur when buying demand begins to weaken as price approaches a previous high. Bottoms are the corresponding formation on the downside. On a conventional time-based chart, traders often wait for additional confirmation before deciding whether such a pattern represents a reversal or merely a pause in the trend. That waiting period can be substantial.

P&F can shorten the process because the chart is already ignoring insignificant movement. An X column interrupted by rising O columns, for example, can create the conditions for what initially resembles a double top and then develops into a triple-top structure. A subsequent upside breakout can reveal that the apparent reversal pattern was actually becoming a continuation formation.

Projecting Prices After a Breakout

P&F chartists use box counts to estimate where price may travel after a breakout. The count can be made vertically or horizontally, although vertical projections tend to be used more successfully. In general, a purchase signal occurs when price exceeds the highest X in the previous X column, while a sell signal occurs when price moves beyond the lowest O in the previous O column.

Using a Vertical Price Projection

Suppose your security has just broken above a double or triple top. The obvious question is how far the new advance might travel. The same problem exists after a significant decline. If the security establishes a new low and begins recovering, you need some way to estimate the potential advance without assuming that the first pullback has ended the new trend.

P&F provides a projection method that is essentially a structured form of momentum measurement. Begin by locating the bottom of the last X column when you are dealing with an upside breakout. Count the number of boxes in that column. Suppose there are four. Multiply that number by the reversal amount, using the traditional three-box reversal:

4 × 3 = 12

Then multiply the result by the box size. If the box size is $1:

12 × $1 = $12

Finally, add that amount to the lowest price in the starting column. If the starting low was $10, the projected objective becomes:

$10 + $12 = $22

Do not mistake the projected objective for a promise. The price can fail to reach $22, or it can move well beyond it. The projection is a guide that must be placed within a larger risk-reward calculation. The potential reward is represented by the projected price, while the starting low may provide a reference for where an initial stop could be placed.

The calculation is reversed for a downside projection. Begin with the highest-high box before the decline begins, count the boxes in the down column, multiply by the reversal amount, and then multiply by the box size. The resulting amount gives you an estimate of how far the decline might extend.

Applying a Horizontal Projection

A horizontal count is used when the security has spent a substantial amount of time moving sideways and appears to be building a base. You will see alternating X and O columns, but instead of focusing on their individual movements, you look for the larger formation created by the entire congestion area.

Suppose the security has declined, begun forming a base, and now appears to be preparing for an upside breakout. First count the number of columns contained in the base, excluding the breakout column. If you count five columns and use the traditional three-box reversal, the calculation is:

5 × 3 = 15

Then add that product to the lowest low within the base. If the lowest price was $10, the projected target becomes:

$10 + $15 = $25

As with the vertical count, this is an estimate rather than a guarantee. The usefulness of the calculation comes from giving you a framework for judging potential reward rather than pretending that the market has an exact destination.

Combining P&F Techniques With Other Indicators

A P&F chart can make time disappear so effectively that you may forget just how much calendar time the chart actually represents. The greater usefulness of P&F often comes from combining it with other tools. You can use additional indicators either to produce an earlier signal or to confirm what the P&F chart is already suggesting. The complication is obvious: many other indicators are based on time. So how do you combine a time-less chart with time-based calculations?

For moving averages, one approach is to calculate the average of the price at the center of each P&F column. This gives you an average price for each reversal rather than for each calendar period. The parabolic stop and reverse indicator can also be applied in essentially the same way as it is applied to conventional charts. One useful feature of the parabolic SAR is that it tightens the stop as the momentum of the price movement begins to weaken.

Bollinger Bands can also be adapted to P&F analysis. If a persistent column of Xs keeps pressing against the upper band and occasionally moves through it, that behavior provides confirmation of the upward trend. If the following column of Os then crosses below the centerline, which is a simple moving average, you can expect the swing to continue toward the lower band.

Remember that Bollinger Bands widen when volatility increases and contract as volatility declines. A narrowing band often accompanies congestion, and congestion on a P&F chart appears as a series of short columns. Those short columns should not be trusted automatically to produce reliable buy or sell signals. When the P&F chart is producing small columns while the Bollinger Bands are simultaneously squeezing inward, the market is telling you that it has become indecisive. It is not establishing a meaningful trend, and there may be little reason to force a trade. Sometimes the most useful signal on the chart is the realization that you should go find another security.

Combining Techniques

Trading with one indicator is certainly better than trading with no indicator at all. Adding a second indicator can improve the process further, provided that the second indicator is measuring something genuinely different from the first. Its job is confirmation. This confirmation principle is one of the foundations of technical analysis because every indicator is going to be wrong on occasion. If one indicator can fail, it is reasonable to ask whether two independent methods can fail simultaneously. They can, of course, but the probability is generally lower when the two tools are not simply measuring the same underlying concept.

You can keep adding indicators to a chart, but more is not necessarily better. The benefit of a third or fourth indicator tends to diminish, and by the time you have five or six indicators competing for your attention, conflicting signals can become the dominant feature of the chart. In practice, it is useful to have at least two primary concepts, with one designated as the ruling indicator and the other serving as confirmation. Additional indicators can be useful for perspective, but they should not turn the trading decision into a committee meeting. My own chart may contain MACD as the primary indicator, along with a two-moving-average crossover, a standard error channel, Bollinger Bands, average true range, parabolic SAR, and ichimoku, all applied to a candlestick chart. At times I will add a support or resistance line or a recognizable pattern if it is obvious enough to demand attention. I also examine three time frames. That is a great deal of information, but it reflects decades of experience and produces a favorable gain-to-loss relationship in the situations where I use it. Every additional indicator makes the decision more complicated, but the additional complexity can be worthwhile if it reduces uncertainty and improves the relationship between gains and losses.

Adding a New Indicator: Introducing Complexity

A single indicator, such as the widely used 20-day moving average, can improve your results. But no indicator is right every time, so a second indicator, filter, or pattern can be used to confirm the first signal. Technical analysis has produced thousands of indicators and chart patterns, which means the possible combinations are almost endless. Once you start adding time requirements to the rules, the possibilities become enormous. A rule such as "buy 30 minutes after the open if x, y, and z occur" creates another layer of variables, and the number of potential systems can quickly reach the millions.

Complexity is not merely a nuisance. It changes the nature of the trading decision. Two indicators will frequently disagree because they are built to respond at different speeds. One will signal before the other. If you insist on waiting for confirmation, you may sacrifice some of the initial profit in exchange for a lower probability of acting on a false signal. That is the trade-off.

Imagine that price moves above its moving average and you buy. A momentum indicator then announces that the security is overbought. What do you do? The momentum indicator may be warning of a retracement, but you have no way of knowing whether that retracement will be a harmless pause or the beginning of a major reversal. The moving average cannot help much because it is inherently slower than price. You can choose to trust the momentum signal because it is more responsive, but momentum indicators can be wrong. The supposed retracement may never occur, and you may find yourself standing on the sidelines while the trend continues without you. There is no decision that eliminates the possibility of being wrong.

The purpose of combining indicators is not to create a perfect machine. It is to improve the odds that the next price movement will be interpreted correctly. That objective is useful, but it is never 100 percent reliable. The only dependable way to determine whether two or more indicators actually work well together is to test them together under explicit trading rules. Testing one indicator is relatively straightforward. Once several indicators are introduced, you have to decide which one rules, when confirmation occurs, how conflicting signals are handled, and what happens when a signal arrives late. At the very least, spend enough time watching the indicators together on actual charts to understand how they interact. The old advice to trade what you see on the chart remains sound.

Choosing Primary and Secondary Indicators

Many technical analysts recommend establishing one primary, or ruling, concept and treating everything else as secondary. I refer to the secondary tools as validator indicators because their purpose is to confirm or challenge the signal produced by the ruling concept.

Suppose your primary concept is a classic moving-average crossover and your validator is a momentum indicator. The moving average produces a buy signal, but momentum subsequently warns that the security is overbought. If you immediately obey momentum and abandon the moving-average signal, you have effectively allowed your secondary indicator to become the ruling indicator. Unless you have extensive experience demonstrating that momentum reliably leads price in this particular security, you should respect the original hierarchy and treat the momentum reading as confirmation rather than command. Alternatively, promote momentum to the ruling position and select another indicator to serve as the validator.

Every trader who combines indicators eventually encounters this problem. The ruling indicator says buy while the validator says sell, or the reverse. If you consistently find yourself following the validator instead of the ruling indicator, that is evidence that your system may have the wrong primary concept. Changing the ruling indicator creates another consequence that is easy to overlook. Your previous gain-loss statistics no longer apply because you have changed the rules that generated those statistics. Your stop and profit target may therefore be wrong as well. You need to select a ruling concept that fits your risk tolerance and then test the complete combination again before trusting the resulting expectancy.

When an Unexpected Validator Appears

Perhaps your normal toolkit consists of moving averages, momentum, and channels. You do not ordinarily use candlesticks, patterns, or ichimoku. Then one day a pattern appears so clearly that you cannot ignore it. If you can see it immediately without searching for it, there is a reasonable possibility that other traders watching the same security are seeing it too.

Suppose a hanging man candlestick suddenly appears and presents a substantial warning. Your indicators are still saying hold. Your stop is nowhere near the current price. Do you obey the system or do you respond to what the chart is plainly showing you? In many circumstances, the hanging man may deserve attention. But if you override the system, there is an accounting problem. That trade no longer belongs in the performance record of the original system. Your track record is supposed to measure what the system produces, not what instinct happened to tell you to do on one particular afternoon.

Assume that your trading method uses the Donchian 5/20-day moving-average crossover. You buy when the five-day average rises above the 20-day average and sell when it falls below it. A sell signal occurs, followed by a price bounce and a new buy signal, followed a few weeks later by another sell signal. You have a classic whipsaw.

But while watching the chart, you notice a double top. Pattern recognition was not part of your original toolkit, yet the pattern is so conspicuous that ignoring it would require you to pretend you cannot see what is directly in front of you. You research the double-top behavior and find the conventional rule that, after the confirmation level is broken, price can sometimes pull back above that level before resuming the decline. That pullback can offer a better opportunity to exit a long position or establish a short one. In this instance, following the double-top behavior saves you from the moving-average whipsaw.

Does this mean you should override your system whenever you notice something interesting? No. These situations should be unusual. A broad knowledge of technical methods can make you more flexible, but flexibility can easily become rationalization. After one successful double-top trade, you may suddenly begin seeing double tops everywhere. A hammer is no longer a warning because it has become a reason to do whatever you already wanted to do. Human beings are very good at turning the tool in their hand into the explanation for whatever they see. Remember the basic rule: no indicator works all the time.

Studying a Classic Combination

The selection of indicators can make a considerable difference in the behavior of a trading method. A simple moving-average crossover is a natural starting point because it is easy to understand and easy to test, but it has an obvious weakness. Moving averages are derived from past prices, so they necessarily lag the market. By the time a shorter moving average crosses above a longer one, a meaningful portion of the advance may already have occurred. The same problem appears on the way out. The crossover may not occur until after the price has already given back part of the move. You can attempt to compensate for this weakness by adding indicators that measure something different. If you want to anticipate an entry, add momentum. If you want to recognize that an advance may be running out of energy, add an overbought, oversold, or sentiment indicator.

This is a classic combination and a common starting point for newer technical traders because it offers a fairly straightforward way to improve the timing of an otherwise slow method. The moving average remains the basic trend-following mechanism, while the other indicators are given supporting jobs. You buy when the five-day moving average closes above the 20-day moving average, and you sell when price closes below the thinner five-day average. Momentum is then used to anticipate the entry rather than waiting for the moving averages to complete their crossover. In the example, the momentum indicator produces the buy signal three days earlier than the moving-average method alone. The relative-strength indicator provides the other side of the trade by identifying the move down from an overbought condition and producing an exit one day earlier.

Notice what has happened. The moving average has not become faster. Instead, you have added another method for interpreting the same price movement. The momentum indicator is looking for acceleration before the moving-average crossover becomes visible, while relative strength is watching for evidence that the advance has become stretched. Each indicator is being assigned a specific responsibility. That is a much more sensible way to combine indicators than simply putting every available oscillator on the chart and waiting for several of them to point in the same direction.

The numerical difference in the example is substantial. The moving-average method by itself produces a $2.51 gain over the three-week period. Adding momentum and relative strength increases the result to $3.94, an improvement of more than 50 percent. The underlying comparison is straightforward:

Method Buy Sell Profit

Moving average concept $64.35 $66.86 $2.51

With momentum and relative strength $63.38 $67.32 $3.94

Those numbers illustrate why traders become interested in combining techniques. An entry three days earlier and an exit one day earlier can make a surprisingly large difference in a short trade. A method that appears sluggish when viewed through one indicator can become much more responsive when another indicator is used as an early warning. This is the attraction of confirmation and validation. You are attempting to keep the reliability of a slower trend indicator while borrowing some of the speed of a momentum indicator.

But there is a catch. Real markets rarely provide such a cooperative sequence of signals. A momentum indicator can give you an early entry that turns out to be wrong. Relative strength can identify an overbought condition that remains overbought while price continues climbing. The moving average can then disagree with both of them. Suddenly you are no longer following one simple rule. You are deciding which indicator gets priority, how long you will tolerate disagreement, and what you will do when an early signal fails.

This is where combining indicators becomes more complicated than it first appears. Every additional indicator creates another decision point. Suppose momentum tells you to buy three days before the moving-average crossover, but the price immediately falls. The moving averages have not yet generated a sell signal because they never generated the formal buy signal in the first place. What do you do with the early position? You cannot simply wait for the moving-average system to tell you when to exit because the entire purpose of the momentum indicator was to get you into the trade before the moving averages responded.

A separate exit rule is therefore essential for an early entry. One practical solution is to use a stand-alone stop-loss rule. You might exit after losing a predetermined dollar amount or a specified percentage of the capital committed to the trade. The exact amount is a separate question, but the principle is important. Once you allow one indicator to override the normal entry rule, you must also decide what controls the trade if that early signal proves false. Otherwise you have created an entry without a corresponding exit.

The same problem occurs on the other side of the trade. Suppose relative strength becomes overbought while the price continues higher. If you automatically sell the moment the indicator reaches its threshold, you may leave a profitable trend far too early. If you ignore the indicator completely, however, you lose the very warning that justified adding it to the system. You therefore need to understand what the secondary indicator is supposed to accomplish. Is it an automatic exit? Is it merely a warning? Is it a signal to tighten the stop? Is it a reason to take partial profits? These are not the same decisions.

This distinction between a signal and a warning is easy to overlook. A secondary indicator does not necessarily have to replace the primary indicator. It can simply tell you to pay closer attention. If the five-day moving average remains above the 20-day average and price continues making higher highs, an overbought RSI reading does not automatically invalidate the trend. It may instead tell you that the reward remaining in the current wave is becoming smaller relative to the risk. Your response could be different depending on whether you are protecting an existing profit or deciding whether to initiate a new position.

Adding indicators, then, is not merely a matter of adding more information. It changes the structure of the trading method itself. You have more information, but you also have more choices. More choices can be useful when they are governed by clear rules. They can be destructive when they simply give you more opportunities to rationalize whatever decision you already want to make.

That is why the classic three-indicator combination is more valuable as a lesson than as a universal formula. The moving average identifies the basic trend. Momentum attempts to improve the timing of the entry. Relative strength attempts to improve the timing of the exit. The method becomes more responsive because each indicator is doing something different. But the improvement shown by one trade does not prove that the same improvement will occur on every trade. The moment you move from a clean historical example into live trading, you have to deal with false signals, conflicting indicators, stops, delayed confirmations, and the possibility that the market will behave differently from the example that convinced you to use the combination in the first place.

Trading Decisions Multiply Exponentially

Adding indicators is an arithmetic exercise, but the number of possible decisions grows much faster than the number of indicators. One indicator might give you a simple buy or sell rule. Add a second and you now have to decide whether the second indicator confirms the first, overrides it, or merely provides additional information. Add a third and the possible combinations increase again. The difficulty is not necessarily the number of lines on the chart. It is the number of situations you must decide how to handle.

Consider the early momentum signal in the example. It produced a profitable entry before the moving-average crossover, but an early momentum signal can just as easily be false. When that happens, the trader who entered early needs an exit rule that exists independently of the slower moving-average system. Otherwise the trader is trapped between two methods. The momentum indicator says get in, the moving average says nothing has happened yet, and the price is already moving against the position.

A stand-alone stop-loss rule solves part of this problem. You can decide in advance to exit after losing a predetermined dollar amount or a fixed percentage of the capital committed to the trade. The important point is that the rule exists before the trade becomes uncomfortable. Without such a rule, the temptation is to keep waiting for another indicator to rescue the original decision. Sometimes it will. Sometimes the loss will simply become larger while you wait.

The more indicators you add, the more carefully you must define their individual jobs. Otherwise, instead of building a stronger trading method, you are building a collection of opinions. A good combination should reduce uncertainty by giving each tool a distinct responsibility. If two indicators measure essentially the same thing, you may be counting the same evidence twice. If three indicators produce different answers, you need a hierarchy or decision rule before the disagreement occurs.

This is the hidden cost of complexity. The chart may look more sophisticated, but sophistication is not the same thing as effectiveness. Every new indicator should earn its place by contributing information that the existing indicators do not already provide. If it merely repeats an existing signal, it adds clutter. If it frequently contradicts the primary method, it may add confusion. And if you have no predetermined rule for resolving that contradiction, the final decision will probably be made emotionally.

The objective is therefore not to collect indicators. The objective is to construct a method in which the indicators work together without forcing you to make an endless series of subjective decisions. The fewer unnecessary decisions you have to make during an open trade, the easier it becomes to determine whether the method itself is working. Otherwise, when the trade succeeds, you will not know which part of the method deserves credit, and when it fails, you will not know which part needs to be changed.

Concepts Can Secretly Mirror One Another

The whole reason for combining indicators is to obtain independent confirmation. If two indicators are based on essentially the same principle, you are not really receiving two opinions. You are hearing the same opinion twice.

For example, adding one momentum indicator to another does not necessarily create independent confirmation. Technical writer Tushar Chande examined correlations among momentum concepts and found that the momentum indicator and the relative-strength version shown in the chart were more than 90 percent correlated. They were moving almost in lockstep.

That matters. A third indicator that measures the same phenomenon in slightly different clothing does not give you a genuinely new viewpoint. Before adding an indicator, ask what it is actually measuring. If the answer is the same thing your existing indicator measures, the apparent diversification may be an illusion.

Concepts Sometimes Clash

Indicators can also disagree in ways that leave you with no obvious answer. Suppose you are using a moving-average crossover to generate the primary signal and an average true range trailing stop to control risk. The moving averages produce a sell signal, and the ATR stop is subsequently hit. Should you exit?

Of course. A stop is not a suggestion. When your stop is reached, you exit.

The difficult question comes afterward. The stochastic oscillator may be telling you that the short position should be closed and a long position opened. The moving-average crossover provides no confirmation. Momentum is hovering around its centerline without delivering a decisive signal. MACD may have crossed upward, but the entire MACD reading remains below its bull-bear line. Something may be changing, but you do not yet know whether the change will become a trend reversal or merely another temporary fluctuation.

When several indicators disagree, you have two basic choices. You can stand aside and wait for the evidence to become clearer, or you can introduce another independent concept to determine where the current price sits within the larger structure. A linear regression channel, Bollinger Band, or ATR band can sometimes provide that broader context. Another useful approach is simply to count the evidence. How many indicators say buy? How many say sell? How many are neutral or too ambiguous to trust? You are looking for a preponderance of evidence, not proof beyond a reasonable doubt. Trading rarely provides courtroom-quality certainty. When the indicators diverge, waiting or following the side with the greater weight of evidence can be the more rational course.

Oddball Combinations Are Not Necessarily Odd

There is no official list of acceptable indicator combinations. If a particular combination works on the securities you trade and within the time frame you use, it is legitimate for your purposes. One trader, for example, uses the divergence between support and on-balance volume in a very short time frame and reports approximately 90 percent reliability. That does not make the method universally valid, but it illustrates the point. A combination does not have to look conventional to be useful.

Moving averages and momentum are common tools, but they are not merely beginner indicators. Other traders use pivot-point levels with candlestick interpretation and construct trades complete with stops around those observations. Others use Fibonacci counts to estimate where price stands within a wave. Even when traders recognize that Fibonacci and Gann retracement levels lack definitive statistical proof, they may still use them for practical purposes such as determining stop placement. Why? Because enough market participants know those levels that their behavior can make the levels relevant.

The KISS principle has a place here: keep it simple, stupid. Especially when you are still learning technical analysis, do not cover the chart with conditions designed to anticipate every possible circumstance. You need to know the expectancy of what you are doing. Covering the chart with exotic indicators may make the process more entertaining, but it can also make it much harder to determine which parts of the method are actually responsible for the results.

Sailing Into Outer Space

If you investigate trading systems and regimes that are offered for sale, you will encounter some methods that make conventional technical analysis look almost ordinary. Two examples are useful because they demonstrate how far traders can take the idea of combining signals.

The Conquistador

The Conquistador was originally designed for currency futures, although the concept can be applied to other securities. Its central idea is confirmation across three different time frames. Instead of confirming a signal with a completely different type of indicator, you examine the same basic price relationship from several time perspectives.

The system was devised by Bruce Babcock, one of the pioneers of technical analysis and author of The Dow Jones-Irwin Guide to Trading Systems. It was later refined by Nelson Freeburg, publisher of the former Formula Research newsletter.

The rules are straightforward. A purchase requires today's close to be above the 10-day moving average, today's 10-day moving average to be above the 10-day moving average from ten days earlier, and today's close to be above the close from 40 days earlier. The sell conditions reverse those relationships. Trailing stops are then applied using an average true range principle similar to the Chandelier exit discussed earlier.

There is a hint of ichimoku thinking in the Conquistador, particularly in the way several time relationships are used together. The chief benefit of the three-time-frame confirmation is that it keeps you out of many sideways markets. You are not being tempted by a momentum indicator that claims a trend is developing. You wait until the trend has demonstrated itself through the price relationships. That strength is also the weakness of the method. You will remain out of the market much of the time. Because experts disagree about how much of the market is actually trending at any given moment, the three-time-frame approach can also serve as a way of estimating how frequently your chosen securities are in a tradable trend.

Following Waves With Relative Strength

Whether you accept formal wave theories such as Elliott Wave or not, price frequently moves in recognizable waves. Markets advance, retreat, consolidate, and advance again, often producing a rhythm that becomes easier to see after the fact. These waves commonly develop inside channels. A standard error channel, for example, can contain a series of advances and declines in which the peaks approach the upper boundary while the troughs move toward the lower boundary. The channel does not tell you exactly when a reversal will occur, but it gives you a framework for judging where a move is becoming extended.

The difficulty is determining whether a wave has actually reached an extreme or is merely passing through an extreme-looking area before continuing. This is where relative strength can add another dimension to the chart. If price is approaching the upper portion of a rising channel while RSI is simultaneously reaching a high reading, you have two pieces of evidence pointing toward an increasingly stretched advance. That does not automatically mean the security must reverse. A strong trend can remain overbought for a considerable period. What the combination does provide is a reason to become more cautious, tighten an exit plan, take partial profits, or wait for additional evidence before committing new money.

The reverse situation can be equally useful. When price approaches the lower boundary of a channel and RSI reaches an oversold condition, the decline may be approaching a point where the selling pressure is becoming exhausted. Again, oversold does not mean that the price must rise immediately. A security can remain oversold while it continues falling, just as it can remain overbought while continuing higher. The value comes from combining the oscillator reading with the location of price. An oversold RSI in the middle of a powerful downtrend means something different from an oversold RSI appearing as price reaches the lower boundary of a well-defined channel.

This is one reason channel analysis and relative strength can work together so well. The channel tells you where price is within its recent movement, while RSI gives you information about the speed and character of that movement. Price near the top of the channel with RSI near an extreme is a different circumstance from price near the top of the channel with RSI remaining in a neutral range. Likewise, a price decline that reaches the bottom of the channel while RSI refuses to make a new low can deserve closer inspection. The two tools are looking at different aspects of the same market behavior, which is exactly what you want when combining indicators.

Divergence can make the relationship even more interesting. Suppose price makes a higher high near the upper channel boundary, but RSI makes a lower high than it did during the previous wave. Price has moved farther, but the momentum behind that move has not kept pace. That is bearish divergence, and it can serve as an early warning that the current wave is losing force. The same principle applies on the downside. If price makes a lower low near the bottom of the channel while RSI forms a higher low, the decline is no longer being confirmed by the oscillator. The selling wave may still continue, but the failure of momentum to confirm the new price extreme gives you a reason to watch the next few bars carefully.

Sometimes a chart is unusually attractive because the waves and RSI line up so cleanly. Price reaches the upper channel boundary, RSI reaches an extreme, the next wave fails to make much progress, and eventually both the price structure and oscillator turn downward. Such charts are memorable precisely because they are not common. Real markets are considerably messier. Price may touch the channel boundary and continue through it. RSI may become overbought several days before the actual high. A divergence may appear and remain unresolved while price continues trending. The trader who expects every wave to terminate neatly at an oscillator extreme will eventually discover how expensive a tidy chart can be.

The same security may also spend considerable time moving sideways, which creates another problem for RSI and for oscillators generally. An oscillator evaluates today's price in relation to recent prices. When the market lacks direction, that relationship can change rapidly without representing any meaningful change in the underlying trend. RSI can move from one extreme to another as price simply bounces between support and resistance. Buy and sell signals can arrive in quick succession, producing the familiar whipsaw in which one signal is followed by another before either trade has had enough time to develop.

For this reason, the slope of the channel matters. RSI tends to be more useful when the price structure has a recognizable directional bias. In a rising channel, an overbought reading near the upper boundary can warn that the current advance is becoming extended. In a falling channel, an oversold reading near the lower boundary can identify a potentially exhausted decline. In a flat channel, however, the same RSI readings may simply be describing ordinary movement from one side of the trading range to the other. The indicator has not necessarily failed. The market is simply asking it to answer a question it was not designed to answer particularly well.

You should therefore resist the temptation to trade the oscillator by itself. First determine whether price is actually moving in waves and whether those waves have enough structure to be measured. Then examine where the current price sits within the channel. Only after that should you ask what RSI is doing. If the channel, price location, and oscillator are telling the same story, the signal deserves more attention. If they disagree, the disagreement itself is information. It may mean that the wave is stronger than expected, that the channel is being violated, or that the oscillator is producing one of the false signals that becomes common during sideways trading.

The best use of relative strength in this situation is not to predict every turning point. It is to help judge the quality of a wave that is already visible on the chart. Price provides the movement. The channel provides the boundaries. RSI helps you determine whether momentum is confirming that movement or beginning to fall behind it. When those pieces come together, an ordinary chart can become considerably easier to read. When they do not, you should be willing to wait. A wave does not become a reversal simply because an oscillator has reached an extreme.

Enhancing Gains With Selective Timing

There is another way to combine indicators. Instead of putting all your capital into a trade when the first signal appears, you can scale into the position as additional evidence arrives. Suppose you have $5,000 available and are watching five indicators. You might commit $2,000 when the first two indicators agree on a buy signal and then add another $1,000 as each of the remaining three indicators confirms the trade. Eventually the entire $5,000 is invested.

One common version of this technique is to add to a position after price tests support and then turns upward. As the evidence accumulates, you increase the size of the position. Once you receive the first indication that the trend is ending, you exit the entire position at once.

Scaling in while exiting all at once is controversial, although many trading gurus present it as a major secret of their success. Before adopting it, consider what those additional signals actually represent. The later signals are usually secondary validators because otherwise they would have appeared earlier. They may not be independent buy or sell signals at all. They simply make you more comfortable with the original decision. There is also a substantial bookkeeping problem. Once you enter at several different prices and in several different quantities, calculating the actual win-loss ratio and the resulting expectancy becomes more complicated. The arithmetic is not difficult, but it can become tedious very quickly.

Trading With Limited Expectancy: Semi-System, Setup, and Guerrilla Trading

A true system trader needs to know the expectancy of the system and needs to trade the same securities using the same indicators and rules that generated the historical gain-loss record. Systematic trading represents a highly disciplined form of trading used by sophisticated individual traders, small firms, and hedge funds. But not everyone has the time, patience, money, or computational resources required to build and maintain a fully tested system.

That does not mean discretionary trading is automatically useless. Instead of precise system expectancy, you may have to work with what can be called limited expectancy. You should still calculate expectancy from your trading history, even if the history contains trades in which you used indicators outside your normal toolkit or made judgments that were not part of the formal rules.

Those nontechnical judgments are discretionary decisions. A discretionary trader may select certain trades and reject others, add or remove indicators, incorporate fundamental information, or respond to circumstances that were never specified in the system. This creates the very exposure to emotion that systematic trading attempts to eliminate. You cannot backtest a truly discretionary decision because the rule is not defined precisely enough to reproduce. You can, however, calculate what actually happened over a sufficiently long trading history, even if that history is imperfect.

Discretionary trading can work because people learn from experience. Malcolm Gladwell, author of The Tipping Point and Outliers, popularized the idea that expertise requires approximately 10,000 hours of practice in a particular activity. Whether that exact number applies universally is less important here than the underlying concept. Experience can teach a trader to recognize circumstances that a rigid system did not anticipate. That is why occasionally adding an indicator or overriding a signal may produce a better trade. Your objective is ultimately to make money, not to construct the most aesthetically pleasing system imaginable.

Should you build a trading system? If your goal is reliable and consistent performance over a long period, a systematic approach can provide that structure. But you must be prepared to devote considerable time and computational resources to backtesting and to determining how much risk you are willing to accept. There are academics who spend years refining systems designed to account for every possible contingency and never place a live trade. Do not become so fascinated with the machine that you forget the purpose of building it.

A sensible starting point is a combination containing at least two indicators, especially for a beginning trader. Beyond that, you have several ways to blend systematic and discretionary behavior. Semi-system trading, setup trading, and guerrilla trading each occupy a different place on that spectrum.

Semi-System Discretionary Trading

In semi-system trading, you have a formal indicator system that produces buy and sell signals and has an associated historical gain-loss record. You nevertheless select which signals to take and may modify a trade using information outside the system. Fundamentals, unusual news, or a known upcoming event can influence the decision.

One reason traders cherry-pick is to avoid taking a weak signal when a major news event is approaching. Suppose experience tells you that an upcoming announcement is likely to produce an unusually large trading range. A normal stop may be so close that ordinary volatility will knock you out of the trade even if the underlying direction proves correct. You may therefore widen the stop temporarily in an attempt to survive the event.

Strict technical analysts generally object to this practice. Their argument is logical: if the system is not robust enough to handle the situation, add a better indicator or create a different rule rather than changing the rules after the trade begins. If you find yourself repeatedly modifying your system in the middle of trades, the cleaner solution may be to exit and redesign the system.

This leads to another problem in backtesting. Some software assumes that a system is always either long or short, although many programs allow a long-only setting. A conventional buy-sell system naturally carries a bias toward remaining invested. Proper risk management, however, often requires you to be completely out of the market. Being square means holding no position at all.

The problem becomes even more complicated when a stop is hit and the system later produces another signal in the same direction. This is the continuation-rule problem. Should the rules for reentering after a stop be identical to the rules for the initial entry, or should they be different because market conditions have changed? Traders and statisticians disagree about the answer. Giving yourself discretion here is one of the easiest ways for a system trader to drift gradually into discretionary trading.

Many discretionary traders probably began with the intention of following a pure system and then encountered circumstances the system had never addressed. Once enough exceptions accumulate, the system becomes increasingly dependent on judgment. That may also explain why some system designers never finish their work. There is always another contingency to account for.

Solving the Squaring Problem With Setups

Setup trading is a popular form of swing trading in which you identify a specific price configuration, enter when it occurs, and exit after the expected move. One extreme version involves trading a single candlestick, often a very large white candle produced by an announcement or rumor. Different eras have produced different favorites, from penny stocks to technology stocks to metals.

The important distinction is that a setup trader does not need to remain invested. Being square is the natural condition. Instead of maintaining a permanent buy-sell system, you wait for a particular configuration that historically or experientially tends to produce a favorable outcome.

A setup usually consists of a particular arrangement of bars accompanied by one or two confirming conditions, such as an indicator or recognizable pattern. One simple example is an opening-range breakout. Suppose the first 30 minutes of trading produce a range from $5 to $5.50. If price later reaches $5.75, you enter because the market has broken above the initial range.

The exit depends on the objective. You might remain in the position for several minutes or several hours, but the intention in this example is to close the trade during the same session. The reasoning is that a strong early movement creates additional buying pressure as price rises. You have a defined entry condition, but not necessarily a corresponding technical sell signal. Instead, you take your profit when your predetermined objective is reached.

Fund manager Toby Crabel developed this concept and discussed it in Day Trading With Short Term Price Patterns and Opening Range Breakout. He calls it the opening-range breakout, and it is also known as a volatility breakout. You can strengthen the setup by adding validators. The preceding bar might be an inside day or doji. The opening range might have been contracting over the previous three to ten days. Or the market might open with a gap from the prior session. Each condition provides another piece of evidence without changing the fundamental setup.

Why Setups Can Work

Setups generally attempt to enter much earlier in a move than indicator-based systems. A moving average has to wait for enough price history to change direction. A setup can identify the conditions that tend to precede the move and therefore get you involved sooner.

Setup traders often give their formations memorable names, such as pinball or coiled spring. Laurence Connors and Linda Raschke's Street Smarts: High Probability Short-Term Trading Strategies is one of the well-known books devoted to these approaches.

Do these setups work? When the setup has been correctly identified, price can behave as expected often enough to create an edge. The exact failure rate is difficult to establish from the source material, but experienced setup traders argue that the early entry itself provides an advantage. One major benefit is that you can remain completely outside the market until the conditions you want actually appear. No position means no position risk.

Many setups are intraday, with entry and exit occurring on the same day. Others lead to genuine trends and therefore remain active for much longer. In those cases, the setup provides the initial entry while the eventual exit is determined by the trend, stop, or other established rule. Timothy Knight's High Probability Trade Setups provides additional examples of the approach.

Getting an Efficient Entry

A setup is essentially an attempt to recognize the conditions that accompany the beginning of a price move. When the setup is correctly identified, the security should begin moving in your direction without requiring a long confirmation period. This matters because the opening portion of a strong move can represent 25 percent or more of the total movement. That initial thrust is the setup impulse.

The central skill is therefore early recognition. A setup trader is generally not trying to capture every dollar of a major trend. The objective is to take a piece of the move while avoiding the need to remain invested through the entire sequence. That is quite different from the system trader, whose rules may keep the position open throughout a much larger trend.

Using Ruthless Exits

Efficient entries and ruthless exits are characteristic of setup trading. But quick exits do not mean reckless trading. Quite the opposite. Risk control is central to the method and can make setup trading suitable for traders who prefer not to maintain positions for long periods.

You must not give back accumulated profits simply because you are waiting through a retracement. Trend followers frequently accept retracements because they are attempting to capture the larger move. A setup trader has a different objective. You must also use stops without exception and keep them current. Good traders use stops. If you cannot obey a stop, setup trading is not appropriate for you, and the problem extends beyond setup trading.

When the setup works, move the stop upward as the position becomes profitable so that the trade continually earns a higher level of protection. Often the position remains open until price moves against you by a predetermined amount, commonly a dollar amount. Some setups develop and terminate so quickly that the initial stop is all you need. There is simply not enough time for a trailing stop to become useful.

Setup trading is hit and run.

Setup Drawbacks

Setup trading is often associated with day trading, although the two are not synonymous. You can run a systematic strategy on a daily basis, and you can hold a setup trade for much longer than one session. Setups can appear attractive to newcomers because they allow you to remain out of the market most of the time, but that advantage comes with a price.

The biggest problem is concentration. A system trader can often enter an order, place the protective stop and profit target, and walk away. A setup trader generally has to watch the market while the setup develops. Eventually software and AI may reduce this burden, but for now it can require substantial attention.

A setup you like may not appear every day or even every month. If you watch only a handful of securities, you may have to know a large number of setups to keep finding opportunities. If you prefer only one or two setups, you may have to monitor a very large universe of securities to find them.

There is another problem. If setups are the only technical concept you know, you may have no plan when none of them appear. Finding your favorite setups also requires scanning. The security producing the most beautiful technical setup may be one you would otherwise avoid completely on fundamental or valuation grounds.

Finally, setup trading often requires you to be available during market hours. Intense concentration and active execution are difficult if you have another full-time job. The freedom to remain out of the market comes at the cost of having to be ready when the market finally presents the opportunity.

Guerrilla Trading

Guerrilla trading takes setup trading and compresses it even further. The holding period becomes extremely short, the expected move becomes small, and the trader is attempting to extract a series of quick profits rather than capture a large portion of a major trend. Search for the concept and you will find everything from legitimate short-term trading techniques to exaggerated sales pitches, questionable claims, and plenty of material designed to make the activity sound more exciting than it really is. Strip away the marketing and the basic idea is fairly simple. A guerrilla trade may last only a few minutes, with a small profit objective that is repeated over and over throughout a trading session. A few cents or a handful of points on one trade may seem insignificant, but repeated opportunities can make those small movements meaningful.

The mathematics of the approach create an immediate problem. If your profit objective is very close to your entry, your stop will usually have to be relatively close as well. That means an ordinary fluctuation can stop you out before the expected move has a chance to develop. The market does not have to prove your analysis completely wrong. It only has to move against you far enough, for long enough, to hit the stop. This is one reason guerrilla trading can produce a high number of small losses even when the underlying method has an advantage. The trader is constantly balancing a small expected gain against the possibility of being removed from the trade by ordinary market noise.

This also makes execution unusually important. In a longer-term trade, being a few seconds late may not matter very much. In a guerrilla trade, a few seconds can change the entire trade. The entry price may move beyond the planned level, the available profit may disappear, or the price may reverse before the order is filled. A trader who intends to capture a move of only a few points cannot casually enter several points away from the original setup and assume the same mathematics still apply. The smaller the expected movement, the more important the relationship between entry, target, spread, slippage, and stop becomes.

Guerrilla trading is frequently concentrated around earnings announcements or scheduled economic news because those events can produce immediate and sometimes violent price movement. The trader is not necessarily interested in determining whether the news is fundamentally good or bad or whether it will change the company's long-term prospects. The immediate objective is narrower: capture the first reaction or one of the short-lived movements that follows it. A positive earnings surprise may send a stock sharply higher, but the stock may retreat almost immediately as traders take profits. It may then rise again as other buyers enter. The guerrilla trader is attempting to participate in those short bursts rather than predict where the stock will be several days or months later.

Prices rarely travel in a perfectly straight line after an important event. Even a powerful move can contain numerous smaller waves. Price jumps higher, pauses, retreats, finds buyers, advances again, and sometimes reverses completely. Those smaller movements create the opportunities. A guerrilla trader attempts to identify the short-term pattern inside the larger event rather than making a major commitment to the long-term direction.

This makes speed important, but speed alone is not enough. A trader who reacts instantly to every price fluctuation is not a guerrilla trader. That is simply uncontrolled activity. The approach still requires a defined setup, an entry condition, a profit objective, and a predetermined point at which the trade has failed. The trader must be able to act quickly when the setup appears and just as quickly abandon the position when the expected movement does not occur. There is very little room for hoping that a losing position will eventually recover.

The approach is therefore almost entirely hit and run. The trader watches short-term charts for particular candlesticks, breakouts, failed breakouts, momentum bursts, or other recognizable patterns and acts as soon as the required conditions develop. Long-term trend structure, broad economic analysis, and even longer-term cycles may have little relevance to the individual trade. What matters is the immediate probability that the developing pattern will produce a quick movement large enough to justify the risk being taken.

This does not mean that longer-term information is completely useless. A guerrilla trader may still benefit from knowing whether the market is generally rising or falling, whether a major support or resistance level is nearby, or whether an important announcement is about to occur. The difference is in how that information is used. A long-term trader might use the broader trend as the central reason for entering a position. A guerrilla trader may use it simply as background information while concentrating on the immediate setup.

Critics argue that guerrilla trading is not "real" technical analysis because it pays relatively little attention to trendedness and instead concentrates on immediate price opportunity. That criticism has some basis if technical analysis is defined primarily as the study of longer-term trends and major price structures. Guerrilla trading is certainly different from conventional trend following. But the distinction does not mean that the approach is necessarily devoid of technical analysis. A guerrilla trader can still use indicators, candlestick patterns, support and resistance, momentum, volume, disciplined entries, predetermined stops, and carefully defined setups.

The holding period is different. The need for discipline is not.

In fact, the short holding period can make discipline even more important. There is no time to negotiate with a losing position or repeatedly reinterpret the chart. A guerrilla trade is supposed to accomplish something quickly. If it does not, the reason for holding it begins to disappear. The trader therefore has to accept that many trades will produce tiny losses or tiny gains. The objective is not to make every trade profitable. The objective is to maintain a favorable relationship between the amount won on successful trades, the amount lost on unsuccessful trades, and the frequency with which the setup actually works.

That is also why transaction costs and execution quality can become much more important in guerrilla trading than in longer-term approaches. When the expected gain is small, costs consume a larger percentage of the potential profit. A method that looks attractive before accounting for commissions, spreads, slippage, and imperfect fills can become considerably less attractive once those costs are included. The trader must therefore judge the entire trade, not simply the chart pattern.

Guerrilla trading is ultimately a specialized form of short-term speculation. It can be highly disciplined, but it can also become highly destructive when speed is mistaken for skill. The trader who has a setup, a defined target, and a defined stop is following a method. The trader who enters simply because price is moving quickly is chasing. Those two behaviors can look remarkably similar from the outside, but they have very different consequences.

Judging Cycles and Waves

Market prices sometimes seem to move in repeating rhythms. Look at enough charts and you will eventually see a succession of rises and declines that resembles waves on the ocean. One advance is followed by a retreat, which is followed by another advance, and the process appears to repeat. Because the pattern seems to recur, analysts naturally begin calling it a cycle. Some cycles have an understandable connection to economic developments. Business activity expands and contracts. Inventories build and decline. Credit becomes available and then tightens. Investors become confident and later become fearful. Other proposed cycles are connected to numerical relationships, astronomical events, or organizing forces that have never been demonstrated convincingly.

That last qualification matters. A pattern that looks repetitive is not automatically evidence of a hidden cause. The human mind is extremely good at finding order in complicated information. Given enough price data, some patterns will appear simply because there are so many possible ways to divide the data into highs, lows, peaks, troughs, and intervals. The real question is whether the apparent cycle continues to provide useful information after it is identified and tested.

People have been searching for financial cycles for centuries, long before modern statistical methods made it possible to test many of these ideas systematically. The subject remains attractive because the possibility of identifying a recurring rhythm is enormously appealing. Imagine knowing that a market is approaching a period in which a major high or low is more likely to occur. Such knowledge would be valuable. The difficulty is that markets do not provide a clock on the wall telling you when the next cycle will begin.

You should therefore be willing to examine cycle theories without automatically accepting them, but you should also recognize that cycles are not required for technical analysis. Ordinary trend indicators, momentum indicators, volatility measures, support and resistance, and price patterns can provide useful trading information without requiring you to believe that some grand cycle governs the market. Cycle-based tools may still be useful when they identify a tendency that can be demonstrated through repeated observations, particularly if enough market participants are watching the same cycle and acting on it.

Defining a Cycle and a Wave

Cycles are everywhere in the natural world. Music produces repeating patterns, electrical energy oscillates, seasons repeat, and ocean tides rise and fall. The classic mathematical representation of a repeating oscillation is the sine wave. It rises, reaches a peak, falls, reaches a trough, and begins the process again. The regularity makes it an attractive model for thinking about cycles.

Financial markets are considerably less cooperative. Their waves are rarely smooth, evenly spaced, or perfectly repetitive. A market may advance rapidly, spend several weeks moving sideways, collapse suddenly, and then recover just as quickly. The next cycle, if one exists, may be shorter or longer than the previous one. Peaks may be separated by different amounts of time, and the size of the movements can vary dramatically.

A cycle consists of recurring waves, but the existence of waves does not necessarily prove that a cycle exists. You can identify a sequence of highs and lows on almost any sufficiently long chart. That does not mean those highs and lows were generated by a fixed periodic process. Natural waves are generally produced by identifiable physical forces. Financial waves are produced by people interacting with one another, and people can change their behavior when new information arrives.

A trader who was willing to buy a stock at $50 yesterday may refuse to buy it today after an earnings warning. An investor who was comfortable taking large risks during a bull market may become unwilling to own the same security after a major decline. A central-bank decision, geopolitical event, unexpected economic report, or corporate announcement can abruptly change the behavior of thousands or millions of participants. That makes financial cycles fundamentally different from the regular movement of a tide.

Human behavior can certainly repeat, however. Traders respond to fear and greed in recognizable ways. Large gains can encourage additional risk-taking. Large losses can produce defensive selling. Extended advances can attract new participants who do not want to miss the move, while extended declines can eventually produce capitulation. These recurring behaviors can create waves without requiring the market to obey a fixed mathematical cycle.

This is why comparisons between financial markets and natural cycles should be treated as useful metaphors rather than established physical laws. The metaphor helps you visualize the movement. It does not prove that the same mechanism is operating. A chart may resemble a sine wave without actually being generated by the mathematical process represented by that sine wave.

The distinction becomes especially important when a trader attempts to use a cycle to forecast the future. Describing the past is relatively easy. You can draw a line through previous peaks and troughs and identify a repeating interval. Predicting the next peak or trough is much harder. The cycle must continue after the point at which it was discovered, and that is where many attractive theories encounter their greatest difficulty.

Some Wild and Woolly History

The search for financial cycles has a long and sometimes unusual history. Analysts have spent decades examining economic data, stock prices, commodity prices, weather patterns, demographic changes, and other records in an attempt to discover recurring intervals. Some proposed relationships have been grounded in observable economic behavior, while others have attempted to connect market movements to forces far removed from conventional finance.

One early and influential line of research examined long historical records and attempted to identify repeating patterns in everything from industrial production to financial markets. Proposed cycles included periods measured in months, several years, and even decades. Some researchers argued that these cycles operated independently of obvious economic forces such as inflation, interest rates, or business conditions. Others believed that the apparent cycles could ultimately be explained through human behavior or economic forces that were not immediately visible.

The attraction of this kind of research is obvious. If a pattern appears repeatedly over a long enough period, it is tempting to assume that something must be causing it. But the longer the historical record becomes, the more opportunities there are to find apparent relationships. A 40-month cycle, for example, may look impressive if enough historical turning points fall near that interval. The important question is whether the same relationship continues to hold when applied prospectively rather than being fitted to the historical record.

This problem appears repeatedly throughout cycle analysis. A researcher identifies a group of highs and lows, measures the distance between them, and discovers an apparent rhythm. The next step is to determine whether that rhythm survives outside the data from which it was originally discovered. If it does, the observation becomes more interesting. If it does not, the cycle may simply have been an artifact of the particular period being studied.

There is also nothing wrong with recognizing that different markets may have different rhythms. A commodity tied to an agricultural season may naturally behave differently from a technology stock. A market affected by inventory cycles may develop a recurring pattern that has little relevance to another market. A business tied to annual purchasing behavior may exhibit seasonality that is easier to explain than a supposed universal cycle operating across every financial asset.

This is where cycle analysis becomes most useful when approached cautiously. Instead of asking whether the entire market is governed by one hidden clock, ask whether a particular market has demonstrated a recurring tendency under particular conditions. That is a much smaller claim, and it is much easier to test.

The history of cycle analysis is filled with ideas ranging from sensible observations about supply, demand, and economic activity to theories involving numbers and forces that are much harder to demonstrate. You do not have to accept the more extraordinary explanations to learn from the more practical ones. The essential task is always the same: determine what the market has actually done, determine whether the behavior repeats often enough to matter, and then decide whether that information improves the trading process.

Starting With Economics

The economic cycle is much easier to understand than many of the more exotic cycle theories. Economies expand, reach periods of maximum activity, slow down, and sometimes enter recession before beginning another expansion. The process has a wave-like appearance around a longer-term growth trend. Economists have spent more than two centuries trying to identify and explain these recurring movements.

Among the commonly discussed economic cycles are the Kondratiev wave, generally described as lasting roughly 45 to 60 years; an infrastructure cycle of approximately 15 to 25 years; a business cycle that has been estimated at roughly 5 to 7 years or 7 to 10.5 years; and the inventory cycle, associated with Joseph Kitchin's work in 1927, which is generally placed around 40 months.

One of the more prominent modern approaches is the Minsky cycle, associated with economist Hyman Minsky and his book Stabilizing an Unstable Economy. Minsky's framework describes a progression in which increasing optimism encourages greater risk-taking and debt accumulation. Eventually the financial structure becomes unstable, confidence breaks, and the resulting contraction can require intervention from institutions such as central banks.

The expression "Minsky moment" became especially common after the financial crisis associated with the collapse of the real estate market in 2008. Larry Randall Wray's Why Minsky Matters provides an accessible discussion of Minsky's ideas and their relationship to market sentiment. The appeal of this framework is that it offers a logical connection between human behavior, financial conditions, debt, and economic instability.

That is very different from the more unconventional claim that economic cycles are somehow controlled by things such as sunspots or astrological events. Dewey and other cycle researchers have entertained such possibilities, arguing that economists sometimes overlook influences that cannot be explained through conventional economic variables. Whether those influences actually control financial prices is another question entirely.

Charles Kindleberger's Manias, Panics and Crashes: A History of Financial Crises provides another useful way to think about recurring market behavior. Kindleberger does not need to use the word cycle repeatedly to describe what is happening. His subject is the tendency of financial markets to pass through recurring patterns of speculation, excess, panic, and contraction.

The important practical point remains the same: you do not need cycle theory to conduct technical analysis. Your indicators can provide evidence about whether a trade has positive expectancy, meaning that the setup offers a reasonable chance of producing a gain. Those same tools can help establish when the position should be closed. Cycles are an additional framework, not a requirement.

Combining Market Metrics With Cycles

Robert Shiller, the Nobel Prize-winning economist, developed the cyclically adjusted price/earnings ratio for the S&P 500. The measure is commonly called CAPE or the Shiller P/E 10. It compares the index price with a ten-year moving average of inflation-adjusted earnings. A high CAPE has historically been associated with lower-than-average subsequent long-term returns, particularly over periods of ten years or more.

Analysts use CAPE to examine whether individual stocks or the broader market appear expensive or inexpensive relative to historical earnings. It has also been used to discuss periods of what former Federal Reserve chairman Alan Greenspan called "irrational exuberance," as well as periods when valuations appear unusually depressed.

The important point for this chapter is that Shiller's use of ten years was not an assertion that the stock market operates on a literal ten-year cycle. The ten-year earnings average was chosen as a way of smoothing the data. There are also substantial economic debates surrounding CAPE, including questions about the appropriate treatment of inflation and whether historical earnings are the right basis for evaluating current valuations.

CAPE belongs in a discussion of cycles because it illustrates something fundamental about human thinking. We are constantly trying to use the past to anticipate the future, and cyclical reasoning is one of the ways we do it. The desire to find repetition appears deeply embedded in how people interpret markets and nature. Sometimes the repetition is real and useful. Sometimes we are simply imposing a pattern on events that happen to line up.

Whether you eventually accept any particular cycle theory or reject all of them, knowing the major theories is useful. It makes it easier to evaluate the claims made by advisers, promotional material, and the endless stream of financial clickbait. And, frankly, cycle theories can be entertaining.

One problem is that there are far too many of them. They overlap, offset one another, and often operate on time scales too long to be practical for an ordinary trader. But consider the old story of the Rothschild family. In the nineteenth century, analysts working for the family reportedly plotted numerous cycles using historical data extending back hundreds of years. They looked for points where several independent cycles supposedly reached a top or bottom simultaneously. The confluence was then treated as a potential decision point.

The alleged Rothschild technique helped inspire a broader cycle-analysis industry beginning around 1912. That search for overlapping cycles never really disappeared. At least one Rothschild company has acknowledged continuing to use cycle analysis while keeping the precise methodology private.

Moving on to Magic Numbers

The cycle question eventually leads into a more difficult idea: perhaps certain numbers have a special relationship with the universe and, through some connection that is not immediately obvious, with financial prices. This is where cycle analysis begins to move away from ordinary statistical observation and toward theories involving recurring numerical relationships, geometry, and universal patterns. Some practitioners believe that certain numbers repeatedly appear in market movements because they reflect a deeper structure underlying both nature and human behavior.

There is no question that the universe contains repeating mathematical relationships. Mathematics appears throughout nature, astronomy, physics, and biological systems. We can identify recurring proportions and cycles in countless places. But recognizing a mathematical pattern in nature does not establish that the same pattern controls traders sitting in front of a stock chart. Those are two entirely different propositions. A mathematical relationship can exist without having any causal connection to financial markets.

This distinction is important because financial markets are produced by people. Traders react to earnings, interest rates, economic conditions, news, fear, greed, liquidity, institutional positioning, and thousands of other factors. If a particular numerical relationship repeatedly appears on a chart, the next question should be whether the relationship has a demonstrable connection to future price movement or whether it is simply a pattern that happens to fit historical data.

Other number-based approaches include Elliott Wave and systems built around mathematical properties such as pi. These approaches are sometimes described as involving "magic numbers" because particular numerical relationships are assigned unusual predictive importance. The terminology is not meant to suggest that every practitioner literally believes the numbers are magical. It reflects the difficulty of explaining why one particular set of numbers should possess greater authority over market behavior than countless other mathematical relationships.

That creates an uncomfortable fit with the statistical mindset normally associated with technical analysis. Technical analysis is supposed to ask whether an observation has a useful probability of repeating. Magic-number theories often begin with the assumption that a particular numerical relationship is inherently important and then look for evidence of it in the market. The difference may seem subtle, but it matters. One approach starts with market behavior and searches for a repeatable relationship. The other starts with the relationship and searches for market behavior that confirms it.

Using Cycles

Cycle analysis is considerably more complicated than applying an ordinary indicator. Give a group of traders a moving average and most will understand the basic idea immediately. Give the same group a collection of cycle theories and the discussion becomes much less orderly. There may be disagreement over the length of the cycle, the starting point, the proper phase, the time frame, and even what constitutes the beginning or end of a cycle.

Almost any cycle system can produce an attractive historical chart. The difficulty begins when you examine the adjustments required to make the cycle fit the price action. One analyst may use four days while another uses five. One may identify a 20-day cycle while another finds 22 days. A cycle may be shifted forward or backward to account for an unusual market movement. Exceptions may be introduced when the market does not behave as expected. Before long, a theory can become flexible enough to explain almost anything that has already happened.

That flexibility is both the attraction and the danger. A theory that can accommodate every possible outcome may appear remarkably accurate because it rarely has to admit that it was wrong. But a trading method becomes much more useful when its rules are defined before the outcome occurs. If you have to continually move the cycle to fit the latest price movement, you are no longer making a clean prediction. You are describing the market after the fact.

Cycle analysis also covers an unusually wide range of explanations. Some are grounded in conventional economic behavior. Business activity, inventories, credit conditions, construction, and investment can all move through recurring phases. Other theories reach into astronomy, numerical relationships, geometry, or forces that have not been demonstrated to control financial markets. These ideas should not all be treated as though they have the same evidentiary foundation.

At the same time, it would be too easy to dismiss every form of cyclical thinking simply because some versions are highly speculative. Traders have long noticed that particular securities, commodities, industries, and markets can develop recurring tendencies. Human behavior itself has repetitive elements. Fear follows large losses. Confidence grows during extended advances. Investors become more aggressive after successful trades and more defensive after losses. Companies increase spending during expansions and reduce it during contractions. These behaviors can create recurring market structures without requiring a mysterious universal force.

The practical question is therefore not whether every cycle theory is true. The more useful question is whether a particular cyclical observation provides information that improves a trading decision. If a cycle helps you anticipate when a market may become more active, identify a period in which a reversal is worth watching, or recognize that a particular seasonal tendency deserves attention, it may have practical value. But the trader should still demand evidence that the relationship is persistent and useful rather than simply impressive on a historical chart.

A complete examination of every cycle theory would be enormous, but the major approaches share one important characteristic. They attempt to explain why market movements appear to repeat. Some look to economics and supply and demand. Others look to numerical relationships. Others search for relationships between markets and astronomical or natural cycles. You do not have to accept all of them to understand why cycle analysis continues to attract traders.

The technology used to observe markets has changed dramatically, but the underlying participants have not changed nearly as much. Traders still respond to profits and losses. They still chase advances and fear declines. They still become optimistic after prices rise and pessimistic after prices fall. That repetitive human behavior alone is enough to produce some recurring structures. The challenge is separating those genuine behavioral tendencies from patterns that merely look impressive on a chart.

Cycling With Supply and Demand: The Pragmatic Mr. Wyckoff

You do not need a mysterious force of nature to explain why charts sometimes appear cyclical. A much more practical explanation is that markets move through recurring changes in supply and demand. When demand overwhelms available supply, prices rise. When supply overwhelms demand, prices fall. Between those two conditions, the market can spend considerable time absorbing buying and selling pressure before the next major movement begins.

This way of thinking produces a very different kind of cycle analysis. Instead of assuming that a market must turn because a particular number of days has passed, you watch what buyers and sellers are actually doing. Price, volume, trading ranges, breakouts, failed breakouts, and changes in the character of the bars can all provide evidence about which side is gaining control.

The framework can be thought of as four broad phases: accumulation, markup, distribution, and markdown. Accumulation occurs when demand begins to overcome supply after a period of weakness or decline. Price may begin producing higher highs and higher closes, although the transition can be gradual and difficult to recognize while it is occurring. The important development is that buyers are becoming increasingly willing to absorb the available supply.

Markup follows when that underlying demand becomes strong enough to push price out of its previous trading ranges. The security begins advancing, more traders notice the move, and the rising price itself can attract additional buyers. This is often the easiest phase to recognize because the trend has become visible. It is also the phase in which confidence can become excessive if the advance continues long enough.

Eventually the balance can change again. During distribution, existing holders begin attempting to sell while new buyers become less willing to pay progressively higher prices. Price may still appear strong, but the character of the movement begins to change. Advances become less efficient, resistance becomes more important, and large amounts of trading can occur without producing much additional upside progress.

Markdown is the final phase of the basic cycle. Supply has overwhelmed demand, support begins to fail, and the security moves lower. What previously looked like temporary weakness can become a sustained decline as more holders decide to sell. The process can then eventually create the conditions for another accumulation phase, beginning the cycle again.

The exact duration of any of these phases is unknown. A market can accumulate for months or years, advance rapidly, distribute gradually, or collapse in a matter of weeks. That uncertainty is one reason stops remain necessary. A cyclical framework can help you understand the stage of a market, but it cannot tell you precisely when one phase will end and another will begin.

This approach also provides a useful alternative to purely numerical cycle theories. If a market is behaving differently, you can change your interpretation because the evidence has changed. You do not have to insist that the market is still in the same phase simply because a predetermined amount of time has passed. Supply and demand provide something observable. The trader can watch the evidence develop rather than waiting for a theoretical clock to reach zero.

Finding Universal Harmony: Hurst's Magic Numbers

Another approach assumes that several cycles can operate simultaneously. Instead of searching for one perfect cycle, the trader looks for a collection of shorter and longer rhythms that overlap. A short cycle may influence the daily movement while a much longer cycle determines the broader direction. The result is a market that can appear confusing when viewed through only one time frame but more organized when several time frames are considered together.

One proposed framework identified approximately 20 natural harmonic wavelengths, ranging from periods measured in minutes to cycles extending for many years. Among the suggested periods were roughly 60 minutes, 160 minutes, one day, five days, 40 days, and a much longer cycle of approximately 17.93 years. The fact that similar long-term periods appear in several different cycle theories is interesting, although similarity alone does not establish that a universal cycle exists.

The underlying idea is that markets may contain several waves at the same time. A trader studying a 20-day movement would therefore not look only at the 20-day cycle. The trader might also examine longer cycles such as 40 days, 80 days, 20 weeks, 40 weeks, and approximately 18 months. The slopes of these longer cycles could then be considered together to determine whether they are generally rising or falling.

The appeal of this approach is easy to understand. A short-term cycle can point upward while a longer cycle is declining, creating a very different trading environment from one in which both are rising. When several cycles turn in the same general direction at approximately the same time, the trader may regard that agreement as a form of confluence. When they point in opposite directions, the market may be more difficult to trade.

But confluence should not be confused with certainty. If enough cycles are available, some of them will eventually line up simply by chance. The more cycles you introduce, the easier it becomes to find a combination that appears meaningful. You therefore have the same problem encountered throughout cycle analysis: the method must be tested before the pattern is accepted as useful.

The idea of multiple cycles does, however, offer one valuable lesson that applies far beyond cycle theory. Markets exist simultaneously on several time frames. A five-day movement can be a small retracement inside a 40-day trend, while that 40-day trend can itself be part of an even larger movement. A trader who ignores the longer structure may misinterpret a short-term reversal as a major trend change. Conversely, a trader who focuses only on the long-term trend may miss a short-term opportunity.

That is the practical value of examining cycles, even if you remain skeptical about the numbers themselves. Different rhythms can exist in the same market, and movements on one time frame can influence how another time frame should be interpreted. You do not need to believe that the market is governed by a hidden universal clock to recognize that price has structure across multiple horizons.

The important discipline is to avoid turning that structure into a rigid prediction machine. A cycle can tell you when to pay attention. It cannot guarantee what you will find when you look. Price still has to confirm the idea.

Looking to the Moon and the Stars

Some cycle theories leave economics altogether and turn toward astronomy. People have studied the relationship between celestial movements and human behavior for thousands of years, so it should not surprise you that financial analysts have attempted to apply the same reasoning to markets.

Examining the Lunar Cycle

There are lunar theories that attribute market behavior to the phases or movements of the moon. One claim is that the Chinese lunar calendar successfully anticipated major U.S. market crashes in 1929, 1987, and 2008. Such a claim immediately raises another question: what about major crashes elsewhere and at other times that do not fit the proposed pattern?

A somewhat more practical observation concerns the length of the lunar cycle when weekends and holidays are removed from financial-market calculations. Under that adjustment, the moon's cycle can appear to be approximately 20 trading days, which is close to the accounting month of about 21 trading days. Some academic research has reported evidence for approximately 20-day cycles in certain securities, with observed ranges around 19 to 23 days.

The existence of such a correlation does not settle the question of why it occurs. A recurring number in two unrelated systems may represent a genuine relationship, a behavioral effect, or simply coincidence.

Saros Cycle and Eclipses

Another lunar concept involves the Saros cycle. The Saros describes a recurring astronomical relationship associated with eclipses and has a period of approximately 18 years, 11 days, and 8 hours. That happens to be reasonably close to other proposed long-term market cycles, including the approximately 17.6-year Balenthiran cycle and Hurst's 17.93-year cycle.

Eclipses have also been associated with several notable financial events, including the Tulip Mania collapse in 1637, the dollar crisis in 1979, and the emerging-markets crash in 1997. But eclipses occur several times each year somewhere on Earth. If every eclipse were capable of producing a major financial event, the relationship would be considerably easier to establish.

Adding More Celestial Bodies

One particularly unusual cycle system is the Delta Phenomenon, promoted by Welles Wilder in The Delta Phenomenon or The Hidden Order in All Markets. Wilder was a major figure in technical analysis and developed several familiar tools, including the relative strength index, average true range, average directional index, and parabolic stop-and-reverse indicator.

The Delta Phenomenon attempts to identify turning points by combining cycles operating across daily, weekly, monthly, annual, and approximately 19-year periods. According to the theory, the hidden symmetry becomes apparent when several lunar and solar cycles converge, supposedly producing a repeating period of 235 months, or roughly 19 years and several hours.

Despite Wilder's reputation, this particular approach has never achieved the same practical acceptance as his more conventional indicators. The methodology remains difficult to reproduce reliably. I cannot make it work either. The other problem is practical: a 19-year cycle is a very long time horizon for most traders.

Including the Sun

Sunspots provide another potential source of cyclical analysis. Individual sunspots can persist from days to months, while the number of visible sunspots varies according to an approximately 11-year solar cycle. Historical records allow researchers to chart sunspot activity for centuries.

Attempts to connect sunspots with economic and stock market activity date back to at least the eighteenth century. One proposed relationship is that increasing sunspot activity raises human excitability and contributes to market rallies. Analysts in the 1930s and again in 1965 reported correlations between increasing sunspot activity and U.S. stock market performance, including an observation that stock market peaks tended to occur before sunspot-cycle maxima.

Figuring Out What May Be Wrong With Astronomy Cycle Theories

Astronomy is not automatically an absurd place to look for market relationships. Human beings are affected by the physical environment, and the financial system exists within the same universe as everything else. The difficulty comes when a handful of historical coincidences are treated as evidence of a reliable causal mechanism.

There may be dozens of instances where a market event appears to line up with an astronomical event. But there have been millions of astronomical events. The relevant question is not whether you can find coincidences. You can. The question is whether the relationship survives statistical testing when all the other events are included. Correlation, as always, does not establish causation.

There is another practical problem. Many astronomical cycle theories make predictions on time scales of ten, eleven, seventeen, nineteen, or even eighty years. Ask yourself what practical trading decision can be made from a forecast that says a major top or bottom should occur sometime within one of those intervals. Even if the underlying relationship were real, the time horizon creates enormous difficulties for anyone trying to trade it.

Following the Earth's Axis: Seasonality and Calendar Effects

Seasonality is a much more practical form of cyclic analysis because it provides enough historical repetitions to permit meaningful statistical examination. Heating oil demand increases as winter approaches in colder regions. Agricultural commodities follow planting and harvesting schedules. These are genuine seasonal forces, and prices can respond to them.

Seasonality simply refers to recurring price behavior associated with the time of year. Equities and financial futures can also display seasonal tendencies, although the pattern is not guaranteed to repeat every year. The difference between a genuine seasonal effect and a coincidental historical pattern is something you have to test rather than assume.

Differentiating Between Seasonality and Calendar Effects

The term seasonality was traditionally associated with agricultural commodities, while calendar effects was more often used for equities. The distinction has become less important, and the terms are now frequently used interchangeably.

You can investigate the seasonal characteristics of individual securities using specialized services such as Seasonax, Seasonality.ai, Equity Clock, StockCharts, and TradingView. Some services also use algorithmic or AI-based methods to identify securities whose historical price behavior has shown a strong seasonal relationship.

The Most Well-Known Calendar Effects

The U.S. stock market has one famous calendar rule: "Sell in May and go away." The idea comes from research by Yale Hirsch and Jeffrey Hirsch, who examined the relationship between stock market performance and different periods of the calendar in the Stock Trader's Almanac.

The approach is commonly known as the best six months rule. The Hirsches found that much of the historical gain in the S&P 500 occurred between November 1 and April 30. The rule was not profitable every year, but the historical tendency was reported across much of the period beginning in 1950.

The mechanical version is simple. Hold stocks through the end of April, move the money into U.S. Treasury bills for the summer and fall, and return to stocks on November 1. A historical hypothetical calculation combining this seasonal rule with adjustments using MACD produced a dramatic increase from an initial $10,000 stake in 1950 to $3,753,209 by April 2025.

But this needs to be kept in perspective. The result is hypothetical. Nobody actually followed the exact strategy continuously from 1950 through 2025 under the same assumptions. The strategy also leaves the stock market for approximately half the year, which removes market exposure but also removes the opportunity to participate in rallies occurring during the supposedly weak period.

Markets do not consult the calendar before moving. A powerful rally does not stop simply because May arrives. The U.S. market rallied during the summer and fall of 2025 after beginning its advance in April. A trader following the seasonal rule would have been outside the market for that portion of the move. This is the fundamental tradeoff with calendar effects. A historical tendency can exist without becoming a command to ignore what price is doing in real time.

Examining Big-Picture Cycle Theories

The grand theories of market cycles deserve separate treatment because they are widespread, influential, and controversial. None of the major long-term cycle theories discussed here has been established by statistical evidence as a universal law. At the same time, the absence of proof does not mean every cyclical observation is automatically false.

Some cycle schools contain mystical elements and claims about hidden forces that have not been demonstrated. A supply-and-demand explanation is considerably easier to understand because it begins with observable market behavior. The alternative idea, that traders are merely instruments responding unknowingly to a hidden universal law, requires a much larger leap.

Still, some cycle theories appear to work in some circumstances, and many technical traders are reluctant to dismiss them completely. Edward R. Dewey, introduced earlier, was among the most prolific writers on the subject. His books included Cycles: The Science of Prediction and Cycles: The Mysterious Forces that Trigger Events. The titles themselves tell you something about the difficulty of the subject.

Dewey's work included a proposed 40-month stock market cycle. The Foundation for the Study of Cycles, which Dewey founded in 1941, later reported a 40.68-month cycle across the historical record of U.S. stock prices available to it through the 1960s.

There is an obvious practical question here. Even if a 40-month cycle exists, are today's traders prepared to hold a position for 40 months? And even more important, can you identify the beginning of the cycle reliably enough to make the information useful?

Shining a Spotlight on the Magnificent Mr. Gann

It is almost impossible to study market cycles without encountering W. D. Gann, who lived from 1878 to 1955. Gann wrote extensively about trading and technical analysis, but his interests went considerably beyond conventional chart analysis. He also wrote a religious work, The Magic Word, and a novel, The Tunnel Thru the Air, which contained predictions that some readers regard as remarkably accurate.

Gann is best known for works such as How to Make Profits Trading in Commodities and Forty-Five Years in Wall Street. Some of the material attributed to him is firmly rooted in chart analysis and remains influential. Other portions involve astrology, geometry, and numerical relationships that many modern analysts would regard as unsupported.

Gann covered virtually every theme discussed in this chapter. He identified commodity cycles of 20, 30, 45, 60, 84, and 90 years. He incorporated astronomy, numerical relationships, and geometric constructions into his analysis. Only a small part of that enormous body of work can be examined here.

Gann developed techniques involving spirals, angles, and geometric charts. His spiral chart used the equinoxes and solstices as reference points, while his hexagonal chart incorporated planetary positions. These tools are generally not included in ordinary charting software. Traders interested in them have to obtain specialized Gann software or construct the charts themselves.

Applying Core Gann Concepts

The Gann concepts most commonly found in standard trading software are the Gann angle and Gann fan. A Gann angle connects price movement with the passage of time. The best-known relationship is the 1x1 angle, in which one unit of price change occurs over one unit of time, traditionally represented by a 45-degree line.

This gives you an interesting reference point when examining trendline slopes. A relatively flat trend, perhaps around 25 degrees, can indicate a weak trend, while an extremely steep slope, such as 70 degrees, may represent a move that is difficult to sustain and therefore vulnerable to correction. The 45-degree slope is treated as the middle ground, the Goldilocks slope of Gann analysis.

Once you become familiar with the 1x1 concept, it becomes difficult not to notice the slopes of ordinary support and resistance lines. That can be useful even if you never adopt Gann's broader theories. Looking carefully at the steepness of a trend can force you to examine whether other indicators support the same interpretation.

Gann's 45-degree angle served as the starting point for a series of additional lines arranged at specified intervals, creating what became known as the Gann fan. The various lines represented alternative support and resistance levels. Other geometric techniques developed later, including Andrews pitchforks and speed resistance lines, resemble some aspects of this approach.

Celebrating Gann's 50 Percent Retracement Rule

One of Gann's most enduring contributions is the 50 percent retracement rule. Gann observed that prices frequently retraced approximately half of a preceding move before resuming the original trend.

Suppose a security advances from $10 to $30. The crowd eventually decides that the price has gone far enough and begins selling. A retracement follows. Under Gann's rule, the decline might stop around $20, exactly halfway between the original low and high.

Gann considered the 50 percent level particularly important, but he also regarded it as a danger zone. A retracement can stop there and allow the original trend to resume, or it can continue through the midpoint and develop into a genuine reversal. If the trend does resume from the 50 percent area, Gann's framework suggests that the previous high should eventually be exceeded, giving you a natural minimum objective.

Other halfway-type retracements can also be observed, including 25 percent and 12.5 percent levels. Statisticians have not established that these levels occur more frequently than random chance would produce. That does not necessarily make them useless, however, because there is a practical difficulty in testing the idea.

Before a computer can test a retracement rule, you have to tell it exactly what constitutes the beginning and end of the original move and exactly where the retracement stops. Those definitions are not always obvious. Two analysts can look at the same chart and identify different starting points, ending points, and retracement bottoms. Change the definitions and you change the results.

Consequently, you will encounter studies showing that actual retracements frequently fall within a range around 50 percent rather than landing precisely on 50 percent. A range such as 45 to 55 percent may capture more observations than an exact 50 percent requirement. There is also a possible self-fulfilling element. If enough traders know about the 50 percent level, they may place orders there, helping to create the very behavior they are attempting to observe.

There is a final irony to using the 50 percent level as an exit. If the rule works and the price stops declining at that midpoint before resuming the original trend, selling precisely there means you may be exiting just as the better reentry opportunity appears. Returning to the example, a decline from $30 to $20 followed by a renewed advance means the trader who sold at $20 has surrendered the opportunity to participate in the move above $30.

Embracing the Most Popular Wave Idea: The Elliott Wave

Ralph Nelson Elliott proposed that stock market behavior displayed recurring patterns similar to the Fibonacci sequence. He observed that price movements often appeared in waves and developed a forecasting methodology around that observation. The resulting Elliott Wave theory became one of the most widely discussed forms of cyclical analysis.

An internet search produces an enormous amount of Elliott Wave material, including specialized websites, books, videos, and educational courses. The popularity of the theory is itself worth noting because a methodology followed by a large number of traders can influence market behavior regardless of whether its underlying numerical explanation is correct.

Connecting the Elliott Wave to Fibonacci

The numerical foundation of Elliott Wave theory is the Fibonacci sequence. Leonardo Fibonacci, the thirteenth-century Italian mathematician, is credited with bringing the sequence to European mathematics, although the underlying sequence had been known earlier by Indian mathematicians.

The sequence begins 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, and continues indefinitely. After the initial terms, each number is produced by adding the two preceding numbers. Ratios between successive numbers approach approximately 1.618, or its inverse, 0.618. The 1.618 relationship is commonly called the golden ratio and was discussed by Euclid centuries earlier.

Examples of apparent golden-ratio relationships have been identified in flowers, ferns, sunflowers, seashells, hurricanes, whirlpools, and other natural structures. Similar claims have been made about human architecture and art, including the Parthenon and the proportions used by artists such as Cézanne.

But claims about the golden ratio in nature need to be handled carefully. Some of the famous examples do not survive close measurement. The chambered nautilus is often presented as a perfect illustration, yet measured growth ratios in actual nautilus shells are reported in the range of approximately 1.24 to 1.43 rather than 1.618.

Looking Closer at the Elliott Wave

The basic Elliott Wave model divides market movement into two broad phases, impulse and correction. An impulse wave carries prices in the direction of the dominant trend. In the classic five-wave structure, three of those waves travel with the larger trend while two move against it. The countertrend waves are not necessarily signs that the entire trend has failed. They are part of the structure itself, providing the pauses, retracements, and temporary setbacks that occur as a larger move develops.

A correction has a three-part structure in the basic Elliott model, with two waves moving against the larger trend and one moving back in the direction of that trend. The important idea is not simply that prices move up and down, but that smaller movements can be contained within larger movements. A bull market can therefore produce a five-wave advance, followed eventually by a much larger corrective decline. What looks like a minor pullback on one chart may turn out to be only one portion of a much larger wave when viewed from a longer time frame.

You will often find three recognizable waves within an advance, but actual market behavior is rarely as orderly as the textbook diagram suggests. Prices do not know that they are supposed to stop after the third wave. A market can produce several visible advances, interrupted by several declines, before the larger movement is complete. This is an important distinction because the three-wave concept is best treated as a model for organizing what you see rather than as a rigid rule that every market must obey. If you begin counting waves simply to make the chart conform to the theory, you have reversed the proper process. The chart should determine the interpretation, not the other way around.

Corrective waves create an even greater problem. Even experienced Elliott Wave practitioners acknowledge that corrections can be difficult to identify and classify. An obvious impulse movement may be relatively easy to recognize after it has developed, but a correction can take many forms and may change character while it is unfolding. A decline that initially appears to be a temporary retracement can continue farther than expected. Once that happens, the assumption that it is merely a correction becomes increasingly dangerous. The market may not be correcting the trend at all. It may be changing the trend.

This is where Elliott Wave analysis requires discipline. One of the worst mistakes you can make is forcing every countertrend movement into an Elliott label simply because you have already decided what the larger pattern is supposed to be. If you identify a decline as a correction and prices continue falling, you cannot keep moving the wave labels around indefinitely to preserve the original interpretation. At some point the evidence has to take precedence over the theory. A correction that continues to violate the structure of the larger trend may eventually prove to be a genuine reversal.

Retracement measurements create another area of uncertainty. In one series of waves, for example, corrections might measure 69.2 percent, 35.7 percent, and 78 percent of the preceding movement. Those numbers are interesting because they can be compared with familiar Gann and Fibonacci retracement levels, but none of them corresponds precisely to one of the standard percentages. A trader might decide that one of the measurements is close enough to a recognized level to be useful, but that judgment introduces discretion into the process. How close is close enough? Ten percent? Five percent? Two percent? Once the answer depends on interpretation, the supposed precision of the numerical relationship begins to disappear.

The same question should be asked about Fibonacci relationships generally. Why should one numerical sequence have exclusive authority over the behavior of millions of people making financial decisions? Fibonacci numbers are certainly fascinating. The sequence produces a recognizable mathematical relationship, and the ratio between successive numbers approaches approximately 1.618. That does not automatically establish a connection between the sequence and financial markets, however. Nature contains countless numerical relationships. Even numbers produce 2, 4, 6, and 8. Prime numbers produce 3, 5, 7, 11, 13, 17, and so forth. Pi is one of the most important constants in mathematics. Euclid devoted considerable attention to geometric relationships involving the pentagon, yet that does not require us to conclude that five is a hidden organizing force in the stock market.

The scientific objection is therefore fairly simple. There is no obvious reason that one particular self-replicating numerical sequence should dictate the behavior of a crowd of human beings who have different objectives, different information, different time horizons, and different amounts of capital. A numerical relationship can be interesting without being a causal law. It can describe something that happened without explaining why it happened. This distinction matters because financial markets contain an enormous number of patterns that can appear meaningful after the fact.

There is also a practical problem with wave analysis. Once you know that several different wave counts are possible, it becomes tempting to select the interpretation that best fits the market after the movement has already occurred. A successful wave count can look remarkably convincing when viewed backward. The real test comes while the market is still moving and the future waves have not yet revealed themselves. If several interpretations remain possible, you must recognize that uncertainty rather than pretending that the correct count is obvious.

That does not make Elliott Wave useless. A framework can be useful without being a scientific law. Wave analysis encourages you to think about trends as structures containing advances and retracements rather than as straight lines. It can also give you a language for describing where price appears to be within a larger movement. The danger begins when the labels become more important than the price itself. You should be willing to abandon a wave interpretation when the market stops behaving according to it.

Elliott Wave nevertheless became enormously popular, and its popularity was reinforced by several highly publicized calls made by Robert Prechter. In 1982, Prechter called for a major bull market, which was followed by a substantial advance. He later called for a market top shortly before the 1987 crash. Those events attracted enormous attention because they appeared to demonstrate the predictive power of Elliott Wave at precisely the moments when investors were most interested in knowing what would happen next.

Prechter's work helped bring Elliott Wave analysis to a much larger audience. His book with A. J. Frost, The Elliott Wave Principle, was first published in 1978 and was revised in 2024, and it remains one of the best-known works associated with the theory. The history is worth knowing, but the important lesson is broader than any individual forecast. A few highly visible successes can make a theory famous, while the difficult question is how consistently the method performs across all of its applications. That is the question you should keep asking with any market theory, whether it involves Elliott waves, Fibonacci ratios, Gann numbers, or an ordinary technical indicator.

The useful approach is to treat the Elliott model as another way of organizing market behavior. Use it to examine the structure of advances and corrections, to identify areas where a move appears extended, and to think about alternative interpretations of the same price action. But do not let a wave count become a substitute for evidence. The market gets the final vote.

Validating Retracements

Gann and Elliott Wave can be useful when you are looking for a broader framework to organize what you see on a chart. If you are uncertain whether a decline represents an ordinary retracement or the beginning of a reversal, charting software allows you to place Gann and Fibonacci levels on the price series and examine how the movement behaves around them.

There is an immediate problem, however. The two systems use different numbers. Put Gann's principal retracement levels of 12.5, 25, 50, and 75 percent on a chart alongside Fibonacci levels such as approximately 23.5, 38, and 62 percent, and you can quickly cover the chart with horizontal lines. Add intermediate levels and you eventually create so many potential targets that the next retracement is almost certain to come close to one of them.

Traders sometimes use the midpoint between two Fibonacci levels as another target. There is also a reported tendency in foreign exchange markets for prices that have exceeded the 62 percent retracement level to bounce in the opposite direction for several periods. The source reports this behavior as being approximately 90 percent reliable in FX. The more important point is not that the number is magical. It is that traders in the foreign exchange market actually know these levels and use them.

That distinction matters. If thousands of traders are drawing Fibonacci levels on their charts, including professional money managers, those levels can become relevant because traders act on them. The number itself does not possess supernatural power. The behavior of the participants using the number can give it practical importance.

Be skeptical of anyone who shows you only the retracements that worked. An adviser can select the levels that produced impressive reactions and conveniently omit all the instances in which the same technique failed. If you are evaluating a trading adviser or methodology, the relevant evidence is a consistent record across many trades and market conditions, not a collection of attractive examples.

Gann and Elliott Wave are ultimately interpretive techniques. Statisticians are right to be skeptical of unsupported claims about magic numbers, but that skepticism should not blind you to the fact that traders actually use these numbers. You do not have to believe that Fibonacci ratios are written into the structure of the universe, or that markets follow a hidden cyclical law, to recognize that the expectations of other traders can affect the price you are watching. Sometimes understanding what the crowd believes is more useful than deciding whether the crowd's explanation is literally true.