WhitmanTrading

Candlesticks vs Tick Charts

A time-based candlestick closes when its period expires, so a dead hour and a frantic one each produce the same number of bars. A tick chart closes a bar after a fixed number of trades, so bar count tracks activity and quiet periods compress into fewer bars.

Every chart has to decide when to end a bar. Almost everybody uses the clock without ever treating that as a decision, and it is one — the clock has nothing to do with what the market is doing.

What each one is

A time-based candlestick closes when its period runs out. Five minutes, an hour, a day. Whether fifty trades or fifty thousand went through, the bar ends on schedule. Candlesticks covers what the bar shows.

A tick chart closes a bar after a fixed number of trades. Five hundred trades, then a new bar, regardless of how long that took. Tick charts covers the mechanism.

So one is paced by the clock and the other by participation. Whereas a time chart gives a dead hour the same twelve bars as a frantic one, a tick chart gives the dead hour one bar and the frantic hour sixty.

Where they differ

A candlestick series with evenly spaced bars including flat quiet ones.
Time bars: the same width whether anything happened or not. Illustrative chart - not real market data.

What a quiet period looks like. On a time chart it is a run of tiny flat bars that occupy as much screen as the open did. On a tick chart it is one bar, because that is how much activity there was.

A price series with bars that compress and expand with activity.
Tick bars: detail arrives where the trading is. Illustrative chart - not real market data.

Whether other people see the same chart. A one-hour candle is the same object for everybody, which is why hourly and daily levels are partly self-fulfilling. Your five-hundred-tick chart is yours alone, so a level on it is a level nobody else is watching.

A stretch of price where time-based and activity-based bars separate.
The same move, divided into bars two different ways. Illustrative chart - not real market data.

What the data requirement is. A time chart needs prices. A tick chart needs a trade count, which means a real tape — so it works on futures and listed stocks and does not work on spot foreign exchange, where the tick count is one broker’s feed rather than the market’s.

How indicator settings behave. A fourteen-period average on an hourly chart means fourteen hours. On a tick chart it means fourteen bars of five hundred trades, which is a different length of time on every one of them, so every lookback you are used to now measures something else.

Where they agree

A window of trending price drawn two ways showing the same move.
The same move appears on both, cut up differently. Illustrative chart - not real market data.

Both show real traded prices. Open, high, low and close are genuine on either, which is not true of every chart type.

Both leave you with the same market. Cutting the day into different bars does not change what price did; it changes where the detail sits.

Both fail to tell you whether there is a trend. Direction runs on this site’s shared series average 2.01 bars with a longest of 11, and neither bar rule fixes that.

And both cost the same per trade — 0.0098 a round trip here, about 2% of the median bar range of 0.493.

Which one to use

A range-bound stretch with many small time bars going nowhere.
A quiet stretch fills a time chart with bars that say nothing. Illustrative chart - not real market data.

Run time-based candles when levels matter. Session highs, daily closes and hourly levels are watched by a large number of people, and that shared attention is the reason they hold as often as they do.

A busy stretch of price with activity-based bars expanding into detail.
Where activity-based bars put the detail in the right place. Illustrative chart - not real market data.

Run tick charts when you are trading intraday and the clock is misleading you. If half your bars are lunchtime noise that look identical to real bars, a chart paced by trades removes that problem at the source rather than filtering it afterwards.

Run time charts when you need to compare notes with anybody. Every published level, every piece of analysis and every backtest you did not build yourself is on time bars.

And keep a time chart open regardless. The session structure — open, close, prior day’s range — only exists on one of these two.

Why the clock is an arbitrary rule

A candlestick chart annotated with the cost of a round trip.
Every trade costs a round trip however the bars are cut. Illustrative chart - not real market data.

Because nothing in the market resets on the hour. The five-minute bar exists because clocks do, and it means each bar contains a wildly variable amount of actual trading — so an indicator reading is computed over inputs that are not comparable to each other.

A section of a price series drawn without volume context.
Thin periods produce bars that look like data and are not. Illustrative chart - not real market data.

And because thin bars look exactly like real ones. A bar built from nine trades is drawn the same size as one built from nine thousand, and every pattern you read includes both without distinguishing them.

The original data

Of the 24,971 unique videos in the search corpus, no title compares these two directly. Tick charts appear in 55 titles at a median of 90,218 views across 45 channels. Candlesticks appear in 521, at a median of 5,236 across 397.

A candlestick series with several gaps, the largest of them marked.
An overnight gap exists on a time chart and not on a tick chart. Illustrative chart - not real market data.

Seventeen times the median audience on a tenth of the videos. Tick charts have one of the highest medians of any subject measured in this corpus, and candlesticks one of the lowest despite being the most-covered — the classic pattern of a saturated basic topic against an under-served specific one.

A stretch of price bars cut short at a decision point.
A breakout on the tick chart, nothing yet on the time chart. Act? Illustrative chart - not real market data.

On the chart above the tick chart is earlier and the time chart is shared. Earlier is worth something; shared is worth something else, and which you want depends on whether you are trading the move or the level.

When it fails

The characteristic failure is carrying indicator settings across unchanged. A twenty-period average on an hourly chart covers twenty hours every time. On a five-hundred-tick chart it covers twenty bars whose real duration might be four minutes at the open and four hours at lunch, so the same number is measuring a different window on every part of the day. Every level, threshold and lookback you had tuned on time bars is now measuring something else, and because the chart still looks familiar there is nothing to alert you that the settings stopped meaning what they meant.

A second failure is using tick charts on spot foreign exchange. There is no consolidated tape, so the tick count is one broker’s activity and your bars are shaped by your platform’s feed.

A third is expecting your tick levels to hold. Nobody else is looking at them.

A fourth is comparing a tick chart with published analysis, which is on time bars without exception.

And a fifth is losing the session structure. The open, the close and the prior day’s range are real and only one of these chart types shows them.

Candlesticks covers what a single bar reports. Tick charts covers activity-paced bars. And price action covers reading the bars themselves.

What I actually do

The strongest argument for tick charts is the one nobody makes: a five-minute bar at lunchtime and a five-minute bar at the open are the same width on your screen and completely different events. A tick chart at least makes each bar cost the same amount of activity.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.