What do implied volatility and options skew actually tell you?

Definition

Implied volatility is the volatility number that makes an option’s market price consistent with a pricing model: given what buyers are actually paying, it is the amount of future movement being charged for. Computed across every listed strike and expiry, these numbers form the volatility surface. Its two famous cross-sections have names: skew is the pattern of implied volatility across strikes at one expiry — whether downside puts cost more, in volatility terms, than upside calls — and term structure is the pattern across expiries at comparable strikes. Cboe’s VIX family of products demonstrates the mature form of this machinery, distilling an options surface into tradable volatility indexes. Everything on the surface is a price. None of it is a prediction, and this guide reads it strictly as the former.

How it works

Skew records the relative price of downside protection and upside. When puts below the market carry higher implied volatility than calls above it, protection is expensive relative to upside — the well-documented, persistent state in equity indexes, where crash insurance is structurally bid. Crypto skew is less one-directional: depending on which tail is in demand, either downside or upside can be the richer side, and the sign can change as speculative call-buying or defensive put-buying takes over. The skew’s sign and steepness summarize which tail the market is currently paying more to cover — a reading of today’s prices, not a claim about what past regimes did.

Term structure records how implied volatility differs across expiries. Deribit's published DVOL methodology, for example, calculates near- and longer-term variance from option prices and interpolates between the two expiries. Around a known scheduled event, an expiry spanning the date can trade at higher implied volatility than a neighbouring pre-event expiry; that is a directly observable price difference, not a promise that it must collapse afterward or that the event will produce any particular move.

What the surface cannot do is issue probabilities about direction. Expensive puts do not mean a crash is likely; they mean crash protection is in demand — which may reflect hedging mandates, dealer inventory, one large structured trade or genuine fear, all indistinguishable in the price alone. Options prices can be manipulated less easily than thin spot markets but reflect risk premium as much as expectation: sellers charge extra for outcomes that would hurt them most, independent of likelihood. Reading skew as a forecast converts an insurance price list into a prophecy it never claimed to be.

Why it matters

The surface is the closest thing crypto has to a live, priced map of perceived risk. When CML analysis notes that implied volatility for expiries around a decision date has risen while adjacent expiries sit quiet, that is a factual, checkable statement about what the market is charging for that date — far more specific than sentiment surveys or social-media temperature, because someone is paying real premium behind every point of it.

It also disciplines event commentary. “The market expects a big move around the announcement” is empty as vibes but concrete as term structure: the implied move can be computed from the prices of options spanning the event and compared with what similar events actually delivered. Whether the priced move was too large or too small is only knowable afterward — which is precisely why the surface is evidence about pricing, not about outcomes.

Risks and misconceptions

The cardinal error — treating skew or implied volatility as a directional probability forecast — deserves its own warning. An elevated put skew has preceded crashes, rallies and nothing at all; hedging demand is not foreknowledge, and premium sellers are not oracles. Any commentary that converts a skew reading directly into “the market says X% chance of a crash” is manufacturing a probability the price does not contain, and CML’s own house rule is never to publish that construction.

The measurement caveats compound in crypto. Implied volatility is model-derived, so quoted figures depend on conventions the screen rarely shows. A surface inferred from sparse or wide-spread quotes can wobble without broad repricing, and one venue's settlement index or market-maker inventory can shape its own displayed surface. A skew data point without venue, tenor and liquidity context is decoration, not evidence.

Practical example

With a market-moving decision expected in about four weeks, options expiring in 30 days trade at 80 percent annualized implied volatility while those expiring in 7 days trade at 55 percent — the surface has priced the event window at a premium. Using the rule of thumb that an implied move ≈ annualized IV × sqrt(days/365), the 30-day expiry prices a move of about 0.80 × sqrt(30/365) ≈ 23 percent in either direction by expiry. The decision lands; the price moves 4 percent, and the 30-day implied volatility collapses toward 55 percent within hours. Holders of the event-window options watched their contracts lose value despite being directionally right, because the move realized was far smaller than the roughly 23 percent the surface had priced. The surface told everyone what the event cost to insure — it never said what the event would do.

What changes over time

Crypto volatility surfaces change as venues add underlyings and expiries, liquidity changes, and methodology evolves. Skew reflects the orders and inventories present on the measured venue, so there is no single “typical” crypto shape to assume in advance.

Read the surface beside its inputs and neighbours: the options primer for what the contracts are, realized volatility for what movement has actually been delivered, funding and basis for what the leveraged linear market is paying. When surface readings and positioning data point the same way, the case strengthens; when they diverge, the divergence itself is the finding — and the honest sentence about any of it names what is priced, not what is foretold.