Anatomy of a candlestick

The four prices a candle keeps 

A chart cuts time into equal periods and draws one candle for each. Inside a single period, on a liquid instrument, thousands of separate prices can print. The candle keeps four. The first price to trade after the period begins is the open, the last before it ends is the close, and the extremes reached anywhere between them are the high and the low. Every other price is discarded, and the discarding is the point: a summary is defined by what it throws away.

Those four numbers are drawn in two parts. The rectangle spanning the open and the close is the body. The thin lines running out of each end of it reach to the high and the low, and are called wicks, or shadows, or tails depending on the tradition being read. The body measures the distance between the first price and the last one. The wicks measure how far past that pair the period travelled before coming back.

Which end of the body is the open is not marked on the drawing. It is carried by colour instead: a close above the open draws in the chart's up colour, conventionally hollow or green, and a close below it in the down colour, conventionally filled or red. That convention makes a wall of candles readable at a glance, and it is the only part of a candle that is not a price.

Two familiar shapes fall out of those definitions rather than being separate objects to learn. A period whose open and close are almost equal has almost no body and draws as a cross. A period that opened at one extreme and closed at the other has no room left for wicks and draws as a bare rectangle.

Building one from the prices 

The measurements are subtractions and nothing more. The body is the distance from the open to the close. The upper wick runs from the higher of those two up to the high, and the lower wick from the lower of them down to the low. Wick tip to wick tip, the high minus the low, is the range of the period.

Exchanging the open and the close leaves every measurement unchanged and flips the colour. A candle therefore carries two facts that are easy to run together and are not the same fact: how far the period travelled, which is the range, and where it finished relative to where it started, which is the body. A wide range with a small body and a narrow range with a full body describe very different periods.

What a long wick records 

A wick records one thing precisely. Price reached that level during the period and was not there at the end of it. Trades happened at the tip, someone bought and someone sold there, and by the close the market was somewhere else. That is the complete content of a wick, and it is a statement about the period's past tense, not about the period after it.

Length is meaningful only in comparison. A wick is long relative to the body of its own candle, and relative to the ranges the same instrument has been drawing recently. The same distance is a substantial excursion on a currency pair and an unremarkable one on an index. No absolute length makes a wick a long wick, which is why every description of one is comparative.

Candlestick traditions name the resulting shapes and have a vocabulary for what they depict. A long upper wick with a small body near the low of the period is conventionally described as rejection of the higher prices, and the inverted shape as rejection of the lower ones. The word describes the drawing, and the drawing is a pair of endpoints: the period visited those prices and did not close there. What it cannot establish is why, and practitioners disagree about whether the shape carries anything beyond the four numbers that draw it. One large order completed, a scheduled announcement and a thin moment in an illiquid hour all produce the identical picture.

What a single candle cannot support 

The most important limitation is that a candle does not record the order in which its own prices happened. The high and the low sit on the drawing with no indication of which came first. A period that fell to its low, turned, and climbed to its high before the close draws the same candle as a period that rose early, collapsed, and recovered at the last moment. Those are opposite descriptions of a market, indistinguishable at this resolution. The path is what the summary discarded, and it is not recoverable from the summary.

Three further absences follow from the same construction.

• How long price spent at any level. A tip touched once for a fraction of a second and a tip price sat at for most of the period draw the same line.
• How much traded. Volume is a separate series, drawn separately where an instrument has a reliable measure of it, and it is not encoded in the shape.
• Whether the period was calm or violent. A body is the distance between two endpoints, so a steady grind and a period that reached both repeatedly produce the same rectangle.

The claim one candle supports is narrow, and most of the trouble people have with candlesticks comes from stating it too broadly. It supports this: within this period, on this feed, trading opened here, reached at most this high and at least this low, and finished here. It supports nothing whatever about the period that follows. Anything beyond that sentence is interpretation added by the reader, and belongs to a tradition rather than to the drawing.

The same period on a different clock 

A period has to begin somewhere, and where it begins is a decision taken by whoever configured the chart rather than a property of the market. Currency markets trade continuously through the working week with no closing bell to anchor a day to, so a daily candle exists only because a clock was chosen and a boundary drawn on it. That choice is the session clock, and platforms, data vendors and exchanges make it differently.

Moving the boundary moves the open and the close immediately, because those two are defined by the boundary and by nothing else. The high and the low move too, less obviously, because a different window contains a different set of prices to take extremes from. The consequence is the one this lesson exists to make concrete: one market, one unchanged sequence of trades, two clocks, and two candles that can point in opposite directions.

Two consequences follow. A daily candle is comparable only with another daily candle cut on the same clock, so a shape read on one platform and looked up on another can genuinely differ without either being wrong. And the boundary is usually a server setting rather than the reader's own time zone. The platforms YAL runs, MetaTrader 5, each stamp their bars in the time zone their server is configured for, so establishing which clock a chart is drawn on is a prerequisite for reading it rather than a detail to sort out later.

The same reasoning scales in both directions. A weekly candle depends on which day the week is deemed to start, and the shorter the period, the fewer trades are summarised into the four prices, so the more the shape reflects the particular moments the boundary happened to fall between.

Where practitioners disagree 

The first disagreement is about how many of the four prices to keep. One tradition treats the close as the only price with a claim to significance, on the argument that it is the price at which positions were carried into the next period, and draws line charts accordingly. Another treats the open and the close as artefacts of an arbitrary boundary, as the section above demonstrated, and reads the high and the low as the only prices the market itself produced. Candlesticks keep all four and let the reader weigh them, which is a convenience rather than a resolution.

The second comes from the plumbing. Wicks in particular are feed dependent, because a brief extreme reached in one venue's book may or may not appear in another provider's aggregated stream, and a chart can only draw the prices it was sent. Two honest charts of the same instrument over the same period can therefore differ at the wick tips. How a quoted price is assembled in the first place is treated in the guide to quote feeds, and it is why a shape read down to the last fraction of a price is a fact about that chart rather than a fact about the market.