16.1 The Meaning of Volatility to a Technician

Why volatility is a Level I topic

Volatility analysis sits inside the CMT Level I Advanced Techniques domain (26 percent of the exam). Section Ten of the 2026 Program Guide asks you to define volatility, contrast quiet and violent tapes, explain why technicians bother to measure it, and separate historical volatility (HV) from implied volatility (IV) and volatility skew. Candidates who treat volatility as a synonym for “bearish” miss items. A market can grind higher with small bars or collapse with small bars; it can also trend up with huge ranges. Volatility describes how much price is moving, not which way.

This chapter is independent OpenExamPrep teaching for CMT Level I. It uses the Program Guide learning objectives and standard technician constructions (Wilder, Bollinger, Keltner, Cboe VIX). It does not claim sponsorship, review, or partnership by the CMT Association or Cboe.

Define volatility

In statistics, volatility is the dispersion of returns, usually the standard deviation of percentage or log price changes over a chosen window. Annualizing a daily standard deviation by multiplying by the square root of the number of trading days in a year (commonly √252) puts that dispersion into the percent-per-year language option desks use.

A technician adds a market-structure reading on top of that statistic. Volatility is the size and speed of swings: bar range, gap size, how far price travels from a moving average, how quickly support or resistance is traversed, and how much “energy” is in the tape. Variance is the square of volatility; charts and quotes almost always show the square-root version (standard deviation or a range average), not variance itself.

Three traps sit in that definition:

• Volatility is not trend. A slow 20-day climb of 4 percent can print lower volatility than a two-day 4 percent round trip.
• Volatility is not the same as volume. Volume is participation. A quiet, low-volume drift can still have small ranges; a high-volume session can still close unchanged if buying and selling offset. Use both, but do not substitute one for the other.
• Volatility is scale-dependent. A $2 range on a $20 stock is 10 percent; a $2 range on a $200 stock is 1 percent. Compare assets in percent or ATR-relative terms, not raw points, unless you are trading a single contract you already know.

Low-volatility versus high-volatility price movement

A low-volatility asset typically shows tight daily (or bar) ranges, overlapping candles, few gaps, and modest percentage travel over the lookback you care about. Pullbacks are shallow in points. Oscillators look sleepy. Breakouts, when they finally appear, stand out because they escape a compressed box. Mean-reversion tactics can work inside that box until the box ends — and the box often ends suddenly.

A high-volatility asset shows wide ranges, long candles or bars, more gaps, and large percentage swings over the same lookback. Stops that were “reasonable” in last month’s tape get run. Indicators based on a fixed lookback (14-period RSI, 20-period bands) whip around because each bar is a bigger fraction of the window. Opportunity and risk both expand: the same position size produces larger P&L swings.

Worked contrast. Stock A closes a 20-session stretch with an average true range of 0.8 percent of price and no gap larger than 0.3 percent. Stock B, same 20 sessions, averages 3.5 percent ATR and prints three gaps of 2 percent or more. If both rose 6 percent over those 20 days, A was a quiet uptrend; B was a violent path to a similar net result. Position size, stop distance, and whether you fade a close outside a band must differ even though the net trend looks similar on a non-range chart.

Direction still matters for trade bias, but the volatility regime tells you whether you are in a coiling market or an expanding market. Many CMT items hinge on that distinction.

Why technicians track volatility

Technicians track volatility because it changes risk, tactics, and confirmation:

1. Position sizing and stops. A common rule is to risk a fixed fraction of capital per idea and set the stop a multiple of recent range (for example 1× or 2× ATR). When ATR doubles, size must drop or the dollar risk explodes. Ignoring the vol shift is how a “small” chart stop becomes a large loss.
2. Regime choice. Compression (a squeeze) often precedes expansion. Breakout methods have a better context after a long contraction; fading every touch of a band is more defensible when ranges are mean-reverting and there is no expansion underway.
3. Confirmation and divergence. A breakout on expanding range and volume is a different event from a breakout on dying range. A rally that makes new highs while ranges shrink can be a slow grind (still valid) or a warning that energy is fading — read it with trend and breadth, not as a standalone sell.
4. HV versus IV. Realized movement can lag or lead the options market. That gap is information: event risk, crash premium, or a market that has already calmed while options still price stress (or the reverse).
5. Cross-sectional comparison. Ranking stocks or sectors by volatility (or by ATR as a percent of price) helps you avoid treating a biotech name and a utility as interchangeable for stop placement.

Skip volatility and you will still see patterns, but you will mis-size them and mis-read whether a pattern is occurring in a quiet coil or a panic.

Historical volatility

Historical volatility (also called realized volatility when computed from past returns) is a backward-looking statistic. The usual construction is the sample standard deviation of close-to-close log returns over N days, then annualized. Shorter N (10 or 20 days) reacts faster; longer N (60 or 90 days) is smoother and slower to admit that the regime changed.

Close-to-close HV misses what happened between closes — the intraday range and any gap that reversed by the close. Range-based estimators (Parkinson using high-low, Garman-Klass using open-high-low-close) and Wilder’s Average True Range (covered in the next section) exist because technicians live on bars, not only on a return series. For Level I, remember the idea: HV describes what already happened at the sampling frequency you chose. It is not a promise about tomorrow, though volatility does cluster (busy days often follow busy days) and also mean-reverts over longer stretches.

Exam trap: do not substitute your practice-bank question count or a chart’s bar count for a published HV number. HV is computed from the price series. Another trap: a 20-day HV of 15 percent annualized is not “the stock moved 15 percent in 20 days.” It is a standardized dispersion figure.

Implied volatility

Implied volatility is forward-looking. It is the volatility implied by the current option premium given a pricing model and the other known inputs (spot or futures, strike, time, rates, dividends). Traders quote IV as an annualized percent, the same units as HV, so the two can be compared.

IV is a consensus residual. It absorbs genuine expected movement, demand and supply for options, model error, and the extra premium people pay for crash insurance. That is why IV often sits above HV in equity indexes (a volatility risk premium) and why IV can jump before the news print: the market is paying for a possible large move, not reporting a move that has already occurred.

For a technician who never trades options, IV still matters. It is an independent reading of expected range. A stock coiling on the chart while IV is elevated often means an event is nearby. A stock thrashing around while IV is subdued can mean the options market thinks the storm is already in the price — or that the options are cheap relative to realized chaos. Neither comparison is a mechanical buy/sell; both are context.

Volatility skew

If implied volatility were constant, every strike in an expiration would show the same IV. Real markets do not work that way. Volatility skew (sometimes discussed as a smirk or smile depending on shape) is the pattern of IV across strikes, and by extension the term structure of IV across expirations.

For U.S. equity indexes after 1987, the usual strike skew is reverse skew: lower strikes (out-of-the-money puts) carry higher IV than at-the-money options, which in turn often carry higher IV than out-of-the-money calls. Put protection is chronically in demand; a crash is a larger left-tail concern than a melt-up for a typical long-equity holder. Some commodity markets show the opposite forward skew when the feared squeeze is to the upside (short-supply grains, some energy complexes).

Worked sketch. Index at 4,800. The 30-day 4,800 straddle prices at 18 percent IV. The 4,400 put prices at 24 percent IV. The 5,200 call prices at 15 percent IV. That surface is reverse-skewed: the market charges more for downside convexity than for upside convexity. A technician reading only the ATM IV of 18 percent understates how expensive crash hedges are.

Calendar or term-structure skew is a different slice: near-dated IV versus longer-dated IV. A single event (earnings, FOMC, an election) can lift the front of the surface without rewriting the one-year tenor. Level I wants the definition and the equity-index pattern, not a full volatility-surface trading desk.

Keep the vocabulary tight for exam stems: define volatility as fluctuation magnitude; describe quiet versus violent movement; state why you track it (risk, regime, HV/IV comparison); define HV and IV; state that skew is the strike (and sometimes tenor) pattern of IV, typically richer on equity-index downside strikes.