WhitmanTrading

What Is a Line Break Chart?

Line break chart draws a new block only when the closing price exceeds the high or low of the previous blocks, so periods that set no new extreme produce nothing at all. Time is not on the axis, and the number of blocks a reversal must clear is the only setting.

A line break chart refuses to draw anything until price does something new. That refusal is the whole design, and it produces a chart that looks far more decisive than the market it describes.

How to read it

A price series redrawn as discrete blocks at new extremes.
A line break chart draws a block only on a new extreme. Illustrative chart - not real market data.

A new block appears only when the close exceeds a prior extreme. If today’s close is higher than the highest of the recent blocks, an up block is drawn. If it is lower than the lowest, a down block.

A steady series where most periods produce no block at all.
Time and volume are both discarded. Illustrative chart - not real market data.

Otherwise nothing is drawn. A session that closes inside the recent range produces no block, no gap in the chart, no record of any kind. Ten quiet days and one active day look identical.

A rising series with consecutive up blocks.
Three-line break is the common setting. Illustrative chart - not real market data.

The standard setting is three. An up block prints whenever a new high close appears. To reverse, price must break the low of the last three up blocks.

A falling series where a reversal clears three prior blocks.
A reversal needs to clear the last three blocks. Illustrative chart - not real market data.

So continuation is cheap and reversal is expensive. One new extreme extends the run; undoing it takes a move through three.

The filter and the lag are one thing

A choppy series where the chart barely changes.
So it filters noise by refusing to draw it. Illustrative chart - not real market data.

Nothing here removes noise cleverly. The chart simply declines to record anything below the threshold, which is the crudest possible filter and also the most honest one.

A slow series where the reversal confirms well after the turn.
And it confirms late by construction. Illustrative chart - not real market data.

The cost is arithmetic. Requiring three blocks to be cleared means the reversal signal arrives after price has already travelled the width of three blocks from the extreme. That distance is the price of the calm appearance, and it cannot be reduced without reducing the filtering.

A calm series with the filter and lag annotated together.
The filter and the lag are the same property. Illustrative chart - not real market data.

Anyone offering the smoothness without the delay is describing a chart of the past, where both the turn and its confirmation are already visible.

A worked example

Take this site’s shared series. Direction runs average 2.01 bars with a longest of 11. Median bar range is 0.493 and the ninetieth percentile is 1.101.

With an average run of two bars, most runs never produce three consecutive new extremes. So on the standard setting the great majority of the series generates one or two blocks and then stalls — the chart is mostly waiting.

When the 11-bar run arrived, it produced a long clean sequence. That is the chart working exactly as designed: silent through the noise, legible through the one move that mattered.

But the reversal at the end printed three blocks late. If each block spans roughly a median bar of 0.493, the confirmation arrived about 1.5 in price after the turn — roughly three times the typical bar’s entire range.

A falling series with a stop level marked.
A stop fills where the market is. Illustrative chart - not real market data.

That is the trade in one number. You get a chart with almost no false signals and you pay about three bars of movement for every genuine turn.

Where it belongs

On instruments and horizons where three bars of lag is tolerable. A position intended to run for months can afford to confirm late; a day trade cannot.

As a structure filter rather than an entry tool. Reading the block sequence to decide whether a market is trending, then timing the entry on an ordinary chart, uses each for what it is good at.

And not alongside time-based indicators. Moving averages, oscillators and anything with a period assume evenly spaced bars. A line break chart has no time axis, so pairing them produces a calculation whose inputs are not what the formula expects.

The original data

On this site’s shared series: direction runs average 2.01 bars, longest 11. Median bar range 0.493, ninetieth percentile 1.101. A round trip costs 0.0098, about 2% of the median bar.

The run distribution is what decides whether this chart is useful. With two-bar average runs, a three-block reversal requirement filters out nearly everything — which is the intent, and which also means the chart spends most of its life saying nothing.

A candlestick chart annotated with the cost of a round trip.
A round trip costs a share of a bar. Illustrative chart - not real market data.

And each signal still pays the full round trip. Fewer signals is genuinely cheaper in total cost — the one real advantage a heavy filter gives you that has nothing to do with accuracy.

A price series with volume shown beneath.
Volume and price measure different things. Illustrative chart - not real market data.

What the block count actually changes

One block is barely a filter at all. Every new closing extreme reverses the chart, which produces something close to a plain close-only series with extra steps.

Three is the convention, and conventions are not measurements. It became standard through use rather than through anybody demonstrating three is correct, and there is nothing special about it beyond familiarity.

Higher counts trade more delay for more silence. Five or seven blocks produces a chart that reverses almost never, which suits a very long holding period and is useless for anything shorter.

The setting has to be chosen before the data is examined. Picking the count that made last year look cleanest is fitting one dial to history, and a chart fitted that way describes the setting rather than the market. Tie it to the holding period you actually intend to trade and leave it alone.

When it fails

The characteristic failure is reading the clean history as a clean method. A line break chart of the past shows a small number of decisive sequences with no clutter, and the absence of false signals looks like accuracy. It is not accuracy — the failures were never drawn. Every whipsaw that did not clear the threshold is simply absent from the picture, so the chart is a record of the moves that worked with the ones that did not deleted.

A candlestick series with a gap through a level.
A gap skips the level entirely. Illustrative chart - not real market data.

A second failure is tuning the block count on historical data, which is a single-dial over-fit and the easiest one available on any chart type.

A third is losing track of elapsed time. Without a time axis it is genuinely hard to know whether a sequence took a week or a year.

A fourth is applying time-based indicators to it, where the period the formula assumes does not exist.

A declining series cut short at a decision point.
The block reversed. How far ago was the high? Illustrative chart - not real market data.

And a fifth is sizing as though the signal were early. It is late by design, so the stop has to sit further away than the chart’s tidiness suggests.

Kagi chart covers the other time-independent chart and its different reversal rule. Market trend covers the run lengths that decide whether the filter helps. And technical analysis covers the tradition both belong to.

What I actually do

Every noise filter is a delay wearing a disguise, and line break charts are the clearest example I know. The chart looks decisive because it only draws the moves that got big enough — and by the time a reversal block prints, the turn happened several sessions ago.

— Michael Whitman

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