THE FUNDAMENTAL PATTERN (Market Structure)

On Trends

Market movements can create a bewildering number of patterns and variations of patterns. Simple structures take on different meanings depending on context, and patterns sometimes develop and resolve in unexpected ways. Add to this the complexities of the interactions of multiple time frames and related markets, and we end up with a seemingly infinite number of possibilities. The human mind needs some kind of structure to help process information, so various solutions have been proposed by different schools of technical analysis. The classic chart patterns (head and shoulders, double top, pennant, flag, wedge, etc.) are one attempt at providing structure. Other solutions have been proposed, ranging from in-depth quantitative analysis to the various schools of wave and cycle analysis; these, and many other approaches, work for some traders. All of these offer valuable perspectives, but let us turn our attention to a simple, robust framework that focuses on the fundamental price structures created by price trends.

THE FUNDAMENTAL PATTERN

What I am presenting here is not new; I owe a tremendous debt to the authors and traders who have come before me. No significant intellectual construct emerges ex nihilo—anything of value rests on a foundation built by someone else. 

In the case of trend structure, Tony Plummer (2010) has written about the basic trend pattern, which he calls the “price pulse,” with clarity and in-depth. This simple, fundamental trend pattern is found at the core of even elaborate methodologies, like Gann or Elliott, when they are stripped to bare essentials. This pattern, very simply, is how markets move, and

it is a profoundly powerful concept. The fundamental pattern of market movement on all time frames is this: a movement in one direction, a countertrend retracement in the other direction, and another leg in the original direction. Visually, these movements are often labeled with letters, as in Figure 3.1. This is a very rough schematic structure, but it is a powerful pattern that repeats on all time frames. Whether we look at trends spanning decades or seconds, we will find this same impulse, retracement, impulse structure, though the psychological significance of the pattern will vary depending on the time frame. It is also reasonable to ask why this structure should exist at all, and there are several possible answers. Put yourself in the shoes of a trader who has to buy a significant amount of an asset in a very short time. Your primary objective is to get the order done without having a large impact on prices. How might this best be done? The best option in most cases is to buy a little bit, wait so that you don’t move the market too much, then buy some more, and repeat until the order is completely filled. This is how good execution traders work large orders. Smart traders (or smart algorithms) will judge how to plan their buying by the market’s response to their orders. In this way, they are judging the supply or the selling conviction hanging over the market. A single trader executing an order in a market otherwise composed of pure noise traders would create some variation of this price pattern by trying to fill a buy order through this natural process. There must be a firm theoretical foundation for anything in the market; you should be able to clearly articulate why something should be the way it is. Market prices are the result of buyers and sellers negotiating for prices, nothing more and nothing less. We do not need to invoke some mystical force to create the patterns we see in the markets; they are simply the result of buyers and sellers finding prices in a competitive environment. This is also why the fundamental structures in the markets have not changed since antiquity. Some of the earliest written records we have are price records from Phoenician merchants, which paint a story of price movements very similar to those we see today, albeit on a much slower scale.

Fractal Markets A fractal is a type of pattern in which the parts resemble the whole (see Mandelbrot and Hudson 2006). Markets are fractal in nature, meaning that essentially the same patterns appear on all time frames: Patterns on single-tick charts combine to form 1-minute bars. The patterns on the 1-minute bars are the same as those on the single-tick bars, and they combine to form the patterns on 5-minute bars. (The time frames are a convenient but unavoidably arbitrary structure.) This pattern building continues all the way to daily, weekly, monthly, and yearly patterns, which all contain essentially the same patterns. Figure 3.2 shows a schematic example of how this might play out in the market. Each trend leg is an impulse, retracement, impulse pattern, and each of those legs also breaks down into the same pattern on the lower time frame. Though only three levels are shown on the diagram, this structure theoretically extends down to the one-tick level. Furthermore, the entire large structure in Figure 3.2 could be a setup leg for a trend on the next higher time frame. This is not an abstract concept, but an important tool to understand market structure. No pattern exists in the vacuum of a single time frame. The market is like a set of Russian nesting matryoshka dolls—digging into one time frame will reveal similar structures nested on lower time frames, all the way down to the tick level. Understanding this structure is a key component of building intuition about markets. Traders usually focus on one specific time frame, but it is important to understand that the patterns of the lower time frame actually create the patterns on the trading time frame, and that the patterns on the trading time frame are influenced by evolving patterns on the next higher time frame. In practice, the interactions of structures on lower time frames are usually components of price action, while higher time frames are more likely to provide context or motivation for market structure patterns within the trading time frame. 

TREND STRUCTURE 

We now turn to the structure of trends in a little more depth: how they start, how they unfold, and how they come to an end. The presence of a trend suggests an imbalance of buying and selling pressure; it is this imbalance that actually drives the price change of the trend, and the trend will eventually end when the market finds equilibrium at a new price level. It is extremely important to be able to read the trend structure and to know what patterns support a continued imbalance or what patterns indicate that the trend might be coming to an end. 

Using Indicators 

This section will also use the indicators that I have found most useful in actual trading: modified Keltner channels set 2.25 multiples of the Average True Range (ATR) around a 20-period exponential moving average, and a modified moving average convergence/divergence (MACD). These tools are examined in detail in Chapter 7 and Appendix B, but the focus here will be on actually using them to delineate market structure in a discretionary context. If you prefer a different set of indicators, you will find that these same concepts can be adapted to a wide range of tools without loss of generality. For instance, most of the Keltner concepts can also apply to Bollinger bands or even to simple, static percentage bands around a moving average. Most momentum indicators (e.g., rate of change [ROC], Commodity Channel Index [CCI], standard MACD, or some applications of the stochastic oscillator) can be used more or less like the modified MACD.

Impulse and Momentum

The first leg (AB) of the basic structure is often called a momentum move or an impulse move. In an uptrend, this is a relatively sharp advance driven by buying pressure (demand) overcoming existing selling pressure (supply) and creating a lack of liquidity on one side of the market. After a market has been locked in an extended trading range with no clear momentum, the emergence of a sharp momentum move penetrating one side of the range is often a sign that there is a new imbalance of buying or selling pressure. Figure 3.3 shows an example of a new trend emerging in the 30-year Treasury bond futures. Notice a few things about this pattern: first, it is not subtle. Even someone with no chart-reading experience would recognize that the right side of the chart shows a break in the existing pattern. Next, notice that prices were able to penetrate through the upper channel, though the last bar of the chart is somewhat suggestive of a short-term climax. Last, the fast line of the MACD registered a new momentum high. (This idea will be explored in Chapter 7, but essentially, the MACD made a new high reading relative to its recent values.) Figure 3.4 shows an example of a downtrend emerging in the intraday Standard & Poor’s (S&P) 500 futures. In this case, a large downward-closing 5-minute bar penetrates the lower channel, and spikes the MACD to a significant new low. This brings up an important idiosyncrasy of the MACD: the indicator is calculated from moving averages, which smooth and lag prices, so the indicator also lags price movements. It is important to understand this and to understand the picture you would have seen at the time. (Lay a piece of paper over the chart to hide everything to the right of the price spike, and

then slide it to the right to reveal one bar at a time to see the response of the indicator.) Also, notice that both of these cases are very clear, and that the indicator is somewhat redundant because the information is already clearly visible in the price structure. Impulse moves drive trends. As long as each trend leg extends in a momentum move approximately consistent with previous moves, the probabilities favor buying the next pullback for another trend high, or, in the case of a downtrend, shorting the pullbacks for another run at the lows. What you do not want to see in a trend is the emergence of sharp contratrend momentum on one of the pullbacks. Figure 3.5 shows an example in which very sharp downward momentum emerged on a pullback in Sugar futures. Notice that this pullback completely broke the established trend pattern, and spiked the MACD to a sharp new low, which is not consistent with an intact uptrend. Though it is important to understand the subtle clues the market gives us, sometimes the most important ones are very obvious. There is no point trying to trade with the uptrend after an event like this; it is safe to assume the trend is broken until further notice. Remember, the normal pattern in an established trend is that each setup leg is a momentum move, and the subsequent extensions also function as momentum moves setting

up the next trend leg. Economics 101 tells us that increased demand in a market will lift prices as the market tries to find a new market-clearing price; this description of trend movement is one of the mechanisms through which that adjustment happens. Eventually, higher prices will bring enough sellers into the market that a new, perhaps temporary, equilibrium is achieved. Astute technical traders can usually see clues to this process in the market tape. The most important patterns are: new momentum highs or lows, subsequent trend legs making similar new impulse moves, and the absence of strong countertrend momentum on pullbacks.