You will hear it everywhere: implied volatility is the market's forecast of future realized volatility. This framing is intuitive, easy to repeat, and mostly wrong.

Implied volatility is the $\sigma$ you back out of an option price using Black-Scholes (or Black-76, depending on the instrument). You observe the price, you know the other inputs, and you solve for the one unknown. It is a translation of price into volatility units. Nothing more.

Price vs. prediction

The distinction matters. A price reflects supply and demand at a given moment. It includes hedging flows, dealer positioning, structural demand for protection, and liquidity premiums. None of these are forecasts.

When a market maker on Deribit sells a put to a trader hedging a leveraged long, the resulting IV does not represent what the market maker thinks BTC volatility will be. It represents the price at which that trade clears, given the market maker's inventory, the current skew, and how much gamma they are willing to carry.

Empirically, IV is a biased estimator of realized volatility. On average, options are priced above what volatility actually turns out to be. This is the variance risk premium:

$$\text{VRP} = \sigma_{\text{implied}} - \sigma_{\text{realized}} > 0 \text{ (on average)}$$

If IV were a forecast, that premium would not persist. Option sellers systematically earn a return for bearing the risk that volatility spikes. The premium is compensation, not forecasting error.

What I observed in BTC options

Trading 0-DTE puts on Deribit made this concrete for me. ATM implied volatility might sit at 52%, but puts 5-7% out of the money routinely show IVs of 65% to over 100%. That is not the market forecasting 100% annualized volatility for a specific strike. It is the market pricing the fat-tailed risk that BTC could gap through that level in the remaining hours, multiplied by the structural demand for downside protection.

The skew tells you about positioning and risk aversion, not about expected magnitude of movement.

Why it matters for trading

If you treat IV as a forecast, you look at 60% IV and conclude the market expects the underlying to move roughly 60% annualized. When realized volatility comes in at 45%, you think the market was wrong. But the market was not forecasting. It was pricing.

The reframing changes which trades you take. Instead of asking "is IV too high relative to where I think vol will land?", you ask "does the premium embedded in IV compensate me for the risk?". One is a prediction game. The other is a pricing game. My hedging framework entry checklist is built around the second question: not whether IV is "right" or "wrong", but whether the cost of protection at current IV levels produces an acceptable risk-reward profile.

What IV actually tells you

IV tells you how much uncertainty the market is pricing into the option right now. It reflects the collective willingness of participants to buy and sell protection at a given level. When IV rises, it might mean participants expect more volatility. Or it might mean demand for protection increased. Or sellers pulled back. Or a structural flow is pushing prices. Reading IV as a forecast ignores all of this context.

The better mental model: IV is to expected volatility what a house price is to the cost of building the house. Related, but shaped by forces that have nothing to do with the underlying fundamentals.