Why Volatility Matters When Reading a Price Chart

Open any price chart and it seems to tell a complete story: where price started, where it went, how steep the climb or the fall was. What it does not show is the one number needed to judge any of that, which is how large this market's moves usually are. That number is volatility, and a chart without it is a picture with no scale bar.

This post explains why that missing scale matters when reading a chart, the three ways a chart quietly misleads without it, and where the Behavioral Transform Model (BTM) fits, including why it marks the unusual bars directly on the chart.

A chart shows distance, not significance

Most judgments made from a chart are judgments about size. Was that pullback deep? Is price stretched? Was yesterday's candle a big one? "Large" only means something relative to what is ordinary for that market at that time, and a chart has no built-in sense of ordinary. It shows how far price travelled and leaves the significance to the eye, which fills the gap with the height of the screen, the last few candles or memories of another market. None of those is a measurement. The standard way to measure ordinary is the dispersion of recent returns, which the guide to judging whether a move is normal covers in detail.

Three ways a chart misleads without a volatility scale

The axis rescales itself. Charting platforms fit the vertical axis to whatever is on screen. Zoom out to a year and a sharp week shrinks to a ripple; zoom into a quiet fortnight and ordinary noise fills the screen like a storm. Two traders looking at the same candle at different zoom levels can honestly disagree about whether it was large.

Distance means different things at different prices. A 5-point move in an instrument trading at 50 is a 10% move; the same 5 points at 500 is 1%. On a linear chart they look identical. This is why volatility is normally measured in returns rather than price; a band built on price levels inherits the problem, as the post on return space vs price space walks through.

Ordinary changes over time. A move that was routine during a stressed month can be rare in a calm one, for the same instrument. Turbulent periods have historically tended to bunch together, as have calm ones, which is the pattern described in what volatility clustering is. Any fixed yardstick, a set number of points or a level drawn by hand last quarter, goes stale as conditions shift. The same applies across markets: a daily move that is unremarkable in a cryptocurrency can be exceptional in a bond fund.

What a volatility reference on the chart changes

The fix for all three is the same: put a measure of recent volatility on the chart itself, drawn in price terms and updated bar by bar, so that every move is read against it rather than against the screen. Two requirements make such a reference worth trusting. It has to scale with returns, not price. And it has to be checked: if a band claims to describe where price usually trades, it is possible to count how often closes actually landed inside it. That counting is what the pillar post on calibration means by a calibrated range.

Where BTM fits

BTM is Oisigma's version of that reference, built as an indicator for TradingView. According to How It Works, it draws an expected-price reference line, an inner band for the normal range and a wider outer band for abnormal moves, computed from the standard deviation of percentage returns over a rolling window of about 60 bars, per the FAQ. Because the bands are computed from returns and fixed to the data, they mean the same thing at 50 as at 500 and keep the same relation to price at any zoom level; because the window rolls, the width follows volatility as it rises and falls.

The calibration has been published. On the Proof page, the inner band contained the next close 71.65% of the time on average across 40 instruments in 5 asset classes, with a standard deviation of about 2 percentage points between instruments, and the outer band about 94%. Past behavior is not a guarantee of future results. The band's width differed from market to market, while the share of closes it contained stayed close to the same, historically.

Why BTM marks the unusual bars on the chart

The fourth object BTM draws is a marker on any bar that closed outside the normal range. That is the volatility judgment made explicit, bar by bar, on the price chart itself, and its usefulness follows from the three problems above.

It answers the size question where the eye is already looking. Most chart reading is scanning. Without a marker, each candle has to be compared against the band, or against memory. With one, the unusual closes are labelled, independent of the zoom level, the price level or which chart was open five minutes earlier.

It judges each bar against the volatility of its own time. This is the property the eye cannot reproduce. Scrolling back through a chart, the axis is scaled to the largest moves on screen, so a sharp move from a calm stretch can look trivial next to a later crash. The marker was set against the range that applied when that bar closed, so a move that was genuinely unusual for its time stays marked, and a large-looking bar from a turbulent stretch stays unmarked if it was ordinary then.

It applies one rule on every chart. The same definition of unusual holds on an index, a currency pair or a cryptocurrency, and on daily or intraday bars. How It Works lists this among the common uses: keeping one consistent way of reading structure across markets.

It leaves a record that can be checked. Once a bar closes, its range and marker are fixed and never change afterward; only the live marker on the still-forming bar can update until the close. A chart's marker history is therefore a stable record rather than a shifting impression. It shows where unusual closes bunched together, which is volatility clustering made visible, and because a calibrated band is built so that closes outside it are the minority, a run of markers stands out against the background rather than blending into it.

What a marker does not do matters as much. It is a label applied after the close, not a call made in advance, and it says nothing about direction or about what the next bar will do. In How It Works' phrasing, "This is context, not a verdict." What each object on the chart describes, markers included, is set out in reading the expected range.

What the reference does not tell you

The center line is a reference level, not a directional forecast, and the band is an expected range, not a claim that price reverses at its edges.

Its limits are published too. Because the window learns from recent history, it trails sudden shocks: on crisis-onset days, inner-band containment fell to about 65% historically, a limit covered in when volatility bands fail. Some traders use the range to judge whether price is near balance or stretched, or to anchor risk levels to current conditions, as described in how traders anchor risk levels. The paper validates the range's calibration, not the profitability of any use; whether any use adds value after costs is an open question.

The takeaway

A price chart records distance. Reading it means judging size, and size is only meaningful against how much the market has recently been moving. A return-based, calibrated band puts that missing scale back on the chart, and BTM's markers apply it to every bar, each judged against the conditions of its own time.

Frequently asked questions

Is high volatility good or bad for trading? Neither, as a description. Volatility measures the typical size of recent moves, not their quality or direction, and high-volatility periods have historically come with wider expected ranges rather than any particular outcome. Whether a given approach suits calm or turbulent conditions is a question about that approach, and not one Oisigma has tested.

Does volatility tell you which way price will move? No. Volatility is measured on the size of moves in both directions, so on its own it carries no directional information. The working paper on oisigma.com/paper describes its subject as the size of typical next-bar moves, not their direction.

Does a log-scale chart solve the volatility problem? Partly. A logarithmic axis draws equal percentage moves as equal distances, which removes the price-level distortion described above. It does not show how volatile the market has recently been, or that volatility has changed over time, so the scale of an ordinary move still has to come from somewhere else.

How do you see volatility on a TradingView chart? TradingView includes several volatility tools, such as ATR, standard deviation and Bollinger Bands®, each measuring a slightly different quantity. The guide to adding a volatility indicator to TradingView covers the mechanics, and the comparison of volatility band indicators covers what each one publishes about its accuracy.

If you want to see what a calibrated volatility reference looks like on your own charts, BTM comes with a 30-day free trial, and you can start it here.