Objective Support and Resistance: Calculated, Not Drawn

Ask five traders to mark support and resistance on the same chart and you'll get five different sets of lines. All five can defend their choices. That's the problem in one sentence: traditional support and resistance is a judgment call presented as a level, and two people making the same judgment call on the same data routinely disagree.

This article looks at why hand-drawn levels behave that way, and what a calculated alternative — an expected range derived from recent volatility — does differently. It's descriptive throughout: a calculated range describes where price has recently tended to trade. It does not predict direction, promise reversals, or generate signals, and nothing here is trade advice.

Why hand-drawn levels disagree

The classical idea is familiar: support is a price area where falling prices have previously paused, resistance where rising prices have. The concepts of supply and demand behind them are real. The weakness is in the method used to locate them — the human eye, applied in hindsight.

Every choice in drawing a level is discretionary. Wicks or candle bodies? The 15-minute chart or the daily? Two touches or three before a line "counts"? If price pushes through the line by a fraction and comes back, was the level broken or was that noise? Different, defensible answers to each question produce different lines — which is exactly why five traders produce five charts.

The common patch is to widen lines into "zones." That honestly acknowledges the imprecision, but it doesn't resolve it: a zone is a margin of error without a definition. How wide should it be? The answer changes with the trader, the timeframe, and the day — and a level whose width is undefined is hard to use consistently for anything, including deciding when it has failed.

There's also hindsight bias. On a historical chart, the turning points are obvious, and a line connecting them looks inevitable. But the line was drawn after the turns. Whether the same level would have been chosen in advance — before the bounces that now justify it — is a question a marked-up chart can't answer.

What "objective" means here

An objective level is one where the method, not the person, determines the output. Same instrument, same timeframe, same settings — same level, for every trader, every time. No interpretation step in the middle.

That's what a calculated expected range provides. Instead of asking "where has price bounced before?", it asks a narrower question: given how this market has recently been moving, how far from the prior close does a typical bar land? Oisigma's BTM measures the dispersion of recent returns — percentage moves, not raw price levels — and projects a band onto the chart: a normal range where price has usually traded, and a wider boundary crossed only on unusual days. Everything recalculates as each bar closes, so the structure adapts to current volatility instead of pointing at where the market used to be. The construction is walked through step by step on the How It Works page.

The band's coverage is also a checkable, calibrated claim rather than an impression. Historically, the inner band contained the next close about 71% of the time across 97 years of S&P 500 daily data, with similar figures across 40 instruments in five asset classes — past behavior is not a guarantee of future results. The full tables, including the misses, are on the Proof page. A hand-drawn level can't publish a number like that, because a level that differs by trader has no single track record to measure.

For traders who reach for band indicators instead of drawn lines, the construction still matters — bands built on raw price levels can lag and sit off-center in trends, which is one reason the comparison of band designs comes down to measured containment rather than appearance.

What an objective level does not promise

Here is the honest part, and it cuts both ways.

A calculated range removes the subjectivity of where the boundaries are. It does not upgrade the boundaries into predictions. The band is an expected range, not a claim that price reverses at its edges — a close near the lower boundary is a statement about how unusual the day is, not a forecast of a bounce. The center of the range carries almost no directional information, and calibration is an average property with published limits: the range runs too narrow in the first days of a fast crisis, and the outer band is slightly optimistic in the deep tails. Those limits are documented on the Proof page alongside the results.

So the trade-off is real: hand-drawn levels offer flexibility and narrative; a calculated range offers reproducibility and a measured record, with no story attached. Which properties matter more is a decision each trader makes for themselves.

How traders use a calculated range, descriptively

These are descriptions of use, not recommendations. Users commonly treat the band as an objective alternative to hand-drawn support and resistance — a consistent structural reference that reads the same on any market and timeframe, with no lines to maintain or redraw. Some use it to frame breakouts, pullbacks, and consolidations in one consistent yardstick; others anchor their own risk and invalidation levels to the range rather than to a round number. In every case the band supplies the same thing: a reproducible description of where "normal" currently sits. The decisions built on it remain the trader's own.

The takeaway

The weakness of traditional support and resistance was never the underlying idea — it's that hand-drawn levels change with the hand drawing them, which makes them impossible to test and inconsistent to use. A calculated expected range replaces that judgment call with a reproducible measurement and a published record of what it has and hasn't contained. Less romantic, more checkable.

Frequently asked questions

Does the range repaint or move after a bar has closed? No. Once a bar closes, its range and its markers are fixed and do not change afterwards. While the current bar is still forming, the live abnormal-move marker on that unfinished bar can update until it closes. This is the practical difference from a drawn level, which can be nudged after the fact to fit what price did next.

What settings does it have? One that matters: the length of the rolling window, which sets how much history the calculation looks back over. Shorter windows react faster to a change in conditions; longer windows are steadier and smooth out noise. No setting removes uncertainty or risk, and different symbols, timeframes, and window lengths all produce different structure.

Does it behave the same on intraday charts as on the daily? The construction is the same at any timeframe — it measures recent returns and projects a range from the prior close. The published calibration work covers daily bars in most detail, and also tested weekly and monthly bars, where coverage held up when the window length was kept constant and degraded when much shorter windows were used. The method and its tested scope are set out in the working paper.

Can it be used alongside hand-drawn levels rather than instead of them? Some traders do exactly that, keeping their own levels on the chart and treating the calculated range as a second, reproducible reference for how far a typical bar travels. The band does not validate or invalidate a drawn level; it answers a different question, about whether the current move is ordinary for this market.

If you'd like to see what an objective, self-updating range looks like on your own markets, Oisigma's BTM indicator is available for TradingView with a free 30-day trial (then $15/month, cancel anytime). Put it next to your own drawn levels and compare. Start your free trial →