Demystifying Ichimoku Analysis

©2006, Reprinted with permission of Currency Trader Magazine (www.currencytradermag.com)

It’s been around awhile, but Ichimoku analysis seems to have become an increasingly talked-about subject among technicians, especially in the forex market. For better or worse, traders are often attracted to exotic or complex (or complex-sounding) methodologies, presumably under the assumption that something difficult to understand must be useful.

In the case of Ichimoku, a charting technique that combines trend-following and support-resistance components, some exotic terms (to non-Japanese ears, that is) mask common charting tools that are relatively easy to understand and analyze.

The Ichimoku Kinko Hyo charting technique was reportedly developed by Goichi Hosoda, a Japanese newspaper writer, prior to World War II, although he did not publish the method until 1968. The phrase loosely translates to “oneglance balance chart” or “equilibrium chart at-a-glance technique.” Most traders refer to the method simply as “Ichimoku.”

The technique combines trend-following tools similar to moving averages with other calculations that define supposed support and resistance areas. After describing the general precepts of Ichimoku, we will perform some simple tests to better understand how this methodology functions and whether it offers unique benefits.

Ichimoku components

Figure 1 is a daily chart of the EUR/USD pair with the Ichimoku study, which consists of five lines: tenkan sen (signal line), kijun sen (base line), chikou span (lagging span), senkou span A, and senkou span B. Table 1 provides explanations of each of the lines labeled in Figure 1.

Click here for Table 1: Ichimoku Line Calculations

Figure 1 is an example of a strong uptrend. First, the shorter- term tenkan sen, which is the “signal line,” is rising and is above the kijun sen, which is the longer-term “base line.” In addition, both these lines are above the shaded cloud area, or kumo, that lies between senkou span A and senkou span B. Finally, the chikou span, which is the closing price plotted 26 bars back, is climbing. (The chikou span basically provides a convenient way to contrast today’s closing price to the closing price 26 bars ago by comparing the end point of the chikou span to the close of the bar directly above or below it.)

Basically, the tenkan sen represents a short-term moving average, the kijun sen represents an intermediate-term moving average, and the kumo functions as a longerterm moving average, or support- resistance zone.

A downtrend is considered to be in effect when the tenkan sen is below the kijun sen and both lines are below the kumo. In Figure 2, at point A, the tenkan sen crosses below the kijun sen, signaling a downtrend. According to Ichimoku analysis, the significance of this downside crossover was supported by the fact that both lines (and price) were below the kumo. In addition, the chikou span was below the tenkan sen and kijun sen when the crossover occurred (B). Look to the endpoint of the chikou span and compare it’s placement relative to the closing price directly above/below it on the chart. In other words, the chikou span is comparing today’s closing price to the closing price 26 bars ago. If the chikou span is above the closing price 26 bars ago, buyers are in charge. If the chikou span is below the closing price 26-bars ago, sellers are dominating.

The kijun sen can be thought of as the line defining the primary trend monitored by the approach. In Figure 3, the price action crisscrosses the flat kijun sen from mid- March to mid-April and remains mostly within the cloud (kumo). Here, the kumo functions as support. The chart shows how price dipped into the kumo in late March while the kijun sen was moving sideways just above the kumo. The kijun sen and the kumo intersected around the second week of April, but as price began to advance the tenkan sen crossed above the kijun sen, signaling the beginning of an uptrend.

Figure 4 is a 60-minute chart of the EUR/USD pair. In this example, the kumo acts as resistance. The kijun sen works its way lower across the chart, reflecting the downtrend. The tenkan sen crossed above the kijun sen at point A, but the market could not penetrate the kumo more than once (point B). Price subsequently dropped below the kijun sen, and then the tenkan sen crossed below the kijun sen, signaling the renewed downtrend.

Analyzing the approach

In reviewing the Ichimoku technique, there are obvious similarities between it and classic moving average crossover techniques, in which trend moves are signaled when either price crosses above or below a moving average or a shorter-term moving average crosses above or below a longer-term moving average. The tenkan sen and kijun sen lines are simply different ways to calculate shortand intermediate-term trend indicators. Instead of using the average price over the past nine and 26 bars, they use the midpoint of the range (highest high minus lowest low) over the past nine and 26 bars, respectively.

Figure 5 is the 60-minute chart of the EUR/USD pair comparing the kijun sen (green line) to a 26-bar simple moving average (red line). At the left side of the chart, price, moving average, and the kijun sen are all declining. At point A the market is moving sideways and both the moving average and the kijun sen turn flat. At point B, the kijun sen turns up one step while the moving average is heading down. The kijun sen moves horizontally after the one upturn, and the moving average turns back up again. Then at point C price turns down, the kijun sen quickly follows, and the moving average rolls over again. In this case, the direction of the steps by the kijun sen appears to be a slightly better representation of the trend than the moving average.

Testing the lines

To better understand how the tenkan sen and kijun sen lines compare to moving averages, we tested simple trade signals using both tools. The Ichimoku signals consisted of going long when the tenkan sen line crossed above the kijun sen line and going short when the tenkan sen line crossed below the kijun sen line. Similarly, the moving average strategy went long when a shorter-term average crossed above a longer-term average and went short on the opposite condition.

The first test compared a nine- and 26-day simple moving average crossover to a crossover of the standard (nine-day) tanken sen and (26-day) kijun sen lines in the daily euro currency futures (EC) from Feb. 9, 2004 to Feb. 9, 2006. Table 2 summarizes the results. The two approaches were very similar. The moving average crossover technique produced two more trades overall and higher winning percentages, but the end results were comparable. Figure 6 shows charts with signals from both strategies.

One aspect of the Ichimoku technique is that it uses defined look-back periods — nine and 26 days for the tenkan sen and kijun sen lines. As is the case with any indicator, the choice of look-back period will in large part dictate the results. No one look-back period will work in all conditions or markets. To expand on the first test, a second test was conducted using 20- and 60-day periods for the moving averages and the tenkan sen and kijun sen lines. This time, the strategies were tested on daily U.S. dollar/Japanese yen rate (USD/JPY) prices from Feb. 8, 2001 to Feb. 9, 2006. Figure 7 shows some of these trade signals and the test results are shown in Table 3.

Again, the number of trades is roughly the same and the Ichimoku signals have lower winning percentages (and in this case, lose much more money).

These tests are not intended to prove the relative merit of one technique over the other, but merely to illustrate their similarities. There is no reason to suppose the number of days used in one technique is directly transferable to the other, although the tests suggest lines using the same number of days are roughly comparable in both techniques. The Ichimoku line crossovers and the moving average crossover have the same function and produce similar results. More comprehensive testing might uncover interesting aspects of the Ichimoku lines, but there is no reason to think of them as anything other than moving average-type signals. Most of the available Ichimoku literature on the Internet consists of generalizations and unsubstantiated claims about the technique’s usefulness.

No need for subjectivity

There’s no need to ascribe any special properties or abilities to the Ichimoku lines, and certainly no reason to accept any subjective interpretations of what they represent or how useful they are. (For example, signals are sometimes described as strong, weak, or normal, depending on whether crossovers occur above, below, or within the kumo, among other factors.) Testing should determine that, and the tests here offer a departure point for more analysis. Similar quantification and study could be conducted of the senkou lines, which are the longer-term support-resistance pieces of the Ichimoku puzzle.