Volatility is a measure of how much an asset's price swings. Not whether it goes up or down — how much it moves, in either direction. A stock that gains 1% every single day and one that alternates between +5% and -5% can end up in the same place, but the second is far more volatile.
The standard yardstick is realized volatility, also called historical volatility. You take a series of past daily returns, compute their standard deviation — a statistic that captures how spread out the returns are around their average — and then annualize it. Annualizing means scaling the daily number up to a yearly figure, conventionally by multiplying by the square root of 252, the number of trading days in a year.
The result reads as a percentage. A calm large-cap index might run at 12% annualized volatility. A single tech stock might run at 40%. During the 2008 crisis, the S&P 500's realized volatility briefly exceeded 80%. Same market, wildly different weather.
Realized volatility looks backward. Implied volatility looks forward. It's the level of future volatility that option prices are betting on right now.
Here's the logic. An option is a contract that pays off if a stock moves past a set price. The more the underlying stock tends to swing, the more likely a big payoff becomes, so options on jumpy stocks cost more. Run that reasoning in reverse: if you can see what an option costs, you can back out how much movement the market expects. That backed-out number is implied volatility.
A persistent quirk shows up in the data: implied volatility has usually sat a bit above the volatility that later actually happens. That gap is called the volatility risk premium. One reading is that options function partly as insurance, and insurance sellers demand to be paid for carrying the risk of rare disasters. That premium is real in the historical record — but like every backtested pattern, it's paid for in occasional violent losses, and it can shrink as more money chases it.
The VIX is the most famous volatility number in the world, often called the fear index. Conceptually, it's the implied volatility of the S&P 500 over the next 30 days, expressed as an annualized percentage.
It isn't taken from one option. The exchange that publishes it blends the prices of a whole strip of S&P 500 options — puts and calls across a wide range of strike prices — into a single weighted average. Using many strikes makes the index capture the market's full distribution of expected outcomes, including the fear of crashes, not just typical wiggles. That's why the VIX jumps hardest when investors rush to buy protective puts, which are options that pay off in a selloff.
Some intuition for the levels: a VIX of 16 implies daily S&P moves of roughly 1%. Readings below 12 mark unusually calm markets. Readings above 30 mark stress. The VIX closed above 80 in both November 2008 and March 2020. One more thing: you can't buy the VIX itself. You can only trade futures and products built on it, and those behave differently from the index — a detail that has hurt a lot of people.
Returns are hard to predict. Volatility is easier, because of a robust empirical fact: volatility clusters. Calm days tend to follow calm days, and wild days tend to follow wild days. A 3% drop on Tuesday makes a big move on Wednesday far more likely, in either direction.
Why? Uncertainty doesn't resolve instantly. A shock — a rate surprise, a war, a banking scare — takes weeks to digest, and while it's being digested, everyone trades nervously. Forced selling adds fuel: leveraged investors who lose money must reduce positions, which moves prices, which triggers more reduction.
Clustering is why quants model volatility as a slow-moving regime rather than a coin flip, and it's the basis of volatility targeting: sizing positions so that portfolio risk stays roughly constant, which means trimming exposure when markets get wild and adding it back when they calm down. It's also why position sizing rules based on recent volatility aren't just guesswork — they lean on one of the few market patterns that has stayed reasonably stable across decades and countries. Stable, though, is not the same as guaranteed.
Because implied volatility usually runs above realized volatility, selling volatility — through options or VIX-linked products — has often produced steady small gains. Traders call this picking up pennies in front of a steamroller. The pennies are real. So is the steamroller.
February 5, 2018 is the textbook case. Short-volatility trades had printed smooth profits for years, and products that bet against the VIX had become popular with everyday investors. That day the S&P 500 fell about 4% — painful but hardly historic. The VIX, however, more than doubled in a single session, its largest one-day percentage jump on record. One popular inverse-VIX product, XIV, lost more than 90% of its value overnight and was shut down by its issuer days later. Roughly $2 billion evaporated, much of it from people who thought they'd found a money machine.
The mechanics made it worse. Short-volatility products had to buy VIX futures as volatility rose to stay hedged, and that buying pushed volatility higher still — a feedback loop. The lesson generalizes: strategies that sell insurance show beautiful backtests right up until the claim comes due. Years of pennies can vanish in an afternoon.
You don't have to trade volatility to profit from understanding it. Three practical uses stand out.
First, treat it as a risk speedometer. Position sizes that feel fine at 12% market volatility are oversized at 40%. Scaling exposure down as measured volatility rises is a defensible, testable discipline — you can run it against decades of history and see how it would have behaved in 2008, 2020, and 2022 alike.
Second, treat the VIX as context, not a signal. A high VIX tells you insurance is expensive and fear is elevated. It doesn't tell you the bottom is in. Extremely high readings have historically coincided with major lows more often than not, but the sample of true panics is tiny, and small samples make overconfident rules.
Third, respect the asymmetry. Long-volatility positions bleed slowly and occasionally pay off hugely; short-volatility positions pay steadily and occasionally lose catastrophically. Neither is free money. If a volatility strategy backtests too smoothly, assume the steamroller simply isn't in your sample yet — and that any edge in the data has likely decayed since it was discovered.
Previous: Risk Parity and Portfolio Weighting Schemes · Financial glossary
← Back to all investing concepts