How to Read the Ichimoku Cloud: Conversion Line, Base Line, Cloud and Lagging Span

Five lines and a cloud

The Ichimoku Cloud (Ichimoku Kinko Hyo) is an indicator created in Japan by Ichimoku Sanjin, whose real name was Goichi Hosoda, and it overlays five lines on the chart: the conversion line (Tenkan-sen), the base line (Kijun-sen), leading span A, leading span B and the lagging span (Chikou Span). The shaded area between the two leading spans is the cloud (Kumo). With so many lines it looks complicated, but the ingredients are simple. The four lines other than the lagging span are all combinations of 'the midpoint between the highest high and the lowest low over a set period', and the lagging span is just the close plotted further back.

Conversion and base lines: the midpoint of a period

The conversion line is the highest high plus the lowest low of the last 9 bars, divided by 2; the base line is the same calculation over 26 bars. Where a moving average averages every close in its period, these lines look at only two values: the highest and the lowest. So they move only when a new high or low appears or when an old extreme drops out of the period, and in between they stay flat, as in the figure. When the conversion line is above the base line, the midpoint of the last 9 bars is higher than the midpoint of the last 26, which is read much like short-term moving averages stacked above long-term ones.

Leading spans: values shifted 25 bars ahead

Leading span A is (conversion line + base line) ÷ 2, and leading span B is the midpoint between the highest high and the lowest low of the last 52 bars. Both are plotted ahead of the bar they were calculated on. This is often described as '26 days ahead', but that counts the current bar as 1, so TradingView and this site's Ichimoku tool actually plot them 25 bars ahead. Compared with programs that shift them by 26 bars, the cloud looks one bar out of step, but the values of the lines are the same. Despite the name 'leading', they are calculated from bars that have already passed, so they do not look into the future.

The cloud's position, thickness and twists

Three things are read from the cloud. First, whether price is above, inside or below it; inside the cloud is often interpreted as a stretch where the direction has not been settled. Next is thickness. The farther apart the two leading spans are, the thicker the cloud, and one interpretation treats a thick cloud as support or resistance that price has trouble getting through. Last is which line is on top. A cloud with leading span A on top is called a bullish cloud, and one with leading span B on top a bearish cloud; the point where the two cross and the cloud changes is called a twist, and it is watched as a possible point where the trend may change.

Lagging span: the close shifted 25 bars back

The lagging span involves no calculation. It is the current bar's close plotted 25 bars back, that is, at the 26th bar when the current bar is counted as 1. So the lagging span sitting above the price at that spot means the same thing as the current close being higher than the price 25 bars ago. This site's tool compares the current close with the close 25 bars ago. Because the line is pushed back, there is no lagging span over the last 25 bars of the chart, and the lagging span running into past price levels is sometimes interpreted as support or resistance.

Sanyaku Kouten and Sanyaku Gyakuten

When all three conditions below point up, the state is called Sanyaku Kouten, and when all three point down, Sanyaku Gyakuten. This site's Ichimoku tool judges it on the last closed bar. In the figure, the conversion line and lagging span conditions are met first, and all three come together on the bar where price clears the cloud. All three conditions are met only after price has already risen, so Sanyaku Kouten is closer to a confirmation that the market has tilted to one side, and is hard to treat as a starting signal.

• The conversion line is above the base line
• The close is above the cloud (higher than the top of the cloud plotted at the current bar)
• The lagging span is above the price 25 bars ago (current close > close 25 bars ago)
• All three reversed is Sanyaku Gyakuten; anything in between is mixed

The numbers 9, 26 and 52

9, 26 and 52 are explained as roughly a week and a half, a month and two months, based on a six-day week from the days when markets also traded on Saturdays. Applied as-is to crypto, which trades 24 hours a day, seven days a week, 26 daily bars are 26 days, not a month. So some people favor a setting that stretches the periods to 20, 60 and 120 with a shift of 30 (an actual 29-bar shift in this site's tool), while others keep 9, 26 and 52 because the default is what the most people watch. There is no verified conclusion that either one is better, so the realistic approach is to switch between the two settings in the tool and compare whether the verdicts match.

The limit: it reacts late

Every Ichimoku line is either a midpoint of past highs and lows or a close shifted back, so the conditions change only well after price has turned. Leading span B, a 52-bar midpoint, is especially slow. In sideways markets, price moves in and out of the cloud and the conversion and base lines cross often, so the verdict keeps changing. With so many lines, it is also easy to find at least one plausible reason in any situation, which is something to be careful about, and the verdict comes out differently for each bar length. How price moved after Sanyaku Kouten was not measured in this article.