A coordinated chart framework

Ichimoku Kinko Hyo places several related reference lines on the price chart. The aim is to organise trend, shorter and longer range midpoints, and price relationships across time in one view. The familiar cloud is the shaded area between two leading spans.

A common interpretation gives an upward bias when price is above the cloud, a downward bias below it and an uncertain or congested state within it. That is a descriptive framework, not a guarantee that a cloud edge will hold as support or resistance.

All five components ultimately come from the same price history. Agreement among them can create a consistent rule set, but it does not supply five independent sources of evidence. Start with the standard settings so the display is understandable, then evaluate any proposed change through a documented test rather than adjusting it to fit an attractive historical trend.

The five components and their exact meaning

Using conventional 9, 26 and 52 periods, define a range midpoint as (highest high + lowest low)/2 over the specified window. This is not the arithmetic mean of all closing prices.

• Tenkan-sen: midpoint of the latest 9 candles.
• Kijun-sen: midpoint of the latest 26 candles.
• Senkou Span A: average of current Tenkan and Kijun, plotted 26 candles forward.
• Senkou Span B: midpoint of the latest 52 candles, plotted 26 candles forward.
• Chikou Span: the current close plotted 26 candles backward.

The cloud lies between the two Senkou spans, regardless of which is higher. A thicker cloud indicates a larger separation between these calculated midpoints; it is not an independently measured probability of support. MetaQuotes describes the components and shifts.

Avoid the most important Ichimoku backtest error

Keep the calculation time separate from the plot location. Let A(t) and B(t) be the unshifted leading-span values calculated using data through completed candle t. The cloud displayed directly above candle t consists of A(t−26) and B(t−26). The cloud calculated now is displayed at t+26. Its forward position does not mean it contains future market data.

Chikou creates the opposite trap. The close at t is drawn at t−26, so a completed historical chart can display a later price at an earlier horizontal location. Reading the Chikou value plotted at historical time t can therefore introduce the close from t+26 into a decision supposedly made at t.

A safe, explicit comparison available at time t is close(t) versus high(t−26) or low(t−26). All those observations already exist. If implementing buffer access, verify the displacement handling against timestamps and the official iIchimoku example instead of guessing which array index matches the visible cloud. Official iIchimoku reference.

A reproducible H4 cloud-breakout study

The following educational framework uses EUR/USD H4 and 9/26/52 settings. Load at least 78 prior candles before evaluating the cloud, with additional history for other indicators and implementation warm-up.

1. At completed candle t, require close(t) above the top of the cloud displayed at t. Require close(t−1) at or below the top of the cloud displayed at t−1. This defines a fresh completed upward crossing.
2. Require Tenkan(t) above Kijun(t), unshifted A(t) above B(t), and close(t) above high(t−26). These conditions are all assessed with data available at t.
3. Enter at the next available ask. Skip if spread exceeds 2 pips or the fill is more than 0.2 SMA-ATR14 above the signal close.
4. Place the stop 0.25 signal ATR below the lower of Kijun(t) and the bottom of the cloud displayed at t. Freeze it after entry and set a target at twice actual entry-to-stop distance.
5. Exit earlier at the next quote after a completed close below Kijun, or after 24 completed H4 candles. Use the earliest stop, target or rule-based exit.
6. Hold one position and budget at most 0.5% of equity for the planned loss including costs. Require a fresh completed cloud crossing before another entry.

Research the mirrored short specification separately. These filters define an experiment, not a proven system.

Worked component and trade calculations

Suppose the latest nine-candle high and low are 1.1060 and 1.1000, giving Tenkan 1.1030. The 26-candle extremes of 1.1080 and 1.0940 give Kijun 1.1010. Unshifted Span A is therefore 1.1020. If the 52-candle high and low are 1.1120 and 1.0880, unshifted Span B is 1.1000. Those two spans will be plotted 26 candles ahead.

Separately, assume the cloud displayed at the signal candle has top 1.1020 and bottom 1.0990, derived from calculations made 26 bars earlier. Price closes at 1.1040 after meeting the crossing and other filters, then fills at 1.1042. With ATR14 at 0.0040, the stop becomes 1.0980: 10 pips below the lower cloud edge.

Price risk is 62 pips, and the 2R target is 1.1166. A USD 50 risk budget and USD 10 pip value per standard EUR/USD lot allow roughly 0.0806 lots before costs. Reduce and round down for the actual fee schedule and volume increment.

Regimes, costs and false confidence

Sideways price action can cross a thin cloud repeatedly. A very thick cloud may delay a breakout signal and require a wide stop. Sudden policy changes can overwhelm every component because the framework only summarises earlier prices. Several coloured confirmations should not conceal that shared limitation.

Backtest with the exact time shifts and a chronological holdout period. Compare the full specification against a simpler price or moving-average rule. Inspect how much each filter reduces the sample and whether any apparent improvement survives nearby settings.

Include bid/ask spreads, commission, slippage and overnight financing. H4 positions may remain open for multiple rollover events, so gross chart distance can differ from net return. Resolve stop/target order using suitable tick data or conservative assumptions. Report net expectancy, drawdown, losing streaks and time in market, and keep losing examples as prominently in the journal as clean textbook trends.

Display and verify Ichimoku in MT4 or MT5

Add Ichimoku Kinko Hyo from the desktop trend-indicator list and enter Tenkan 9, Kijun 26 and Senkou Span B 52. Use distinct line colours and a translucent cloud so candles remain readable. Leave enough chart space to the right to inspect the forward projection.

Use the Data Window and dated candle positions to distinguish the cloud at the current price location from the cloud generated by current calculations. In automation, validate both against a small table of hand-calculated values before running a large test.

Record the platform, symbol, candle timezone and settings with the study. A custom indicator that removes or changes the shifts is a different display and may require different access rules. Treat the shifted illustrations as a way of organising known information, not as a forecast drawn ahead of the market.