Probability Analysis - Price Distributions

Probability Analysis

Option prices contain an embedded probability distribution of the underlying asset's future price. This risk-neutral density can be extracted using the Breeden-Litzenberger theorem: the second derivative of the call price with respect to strike equals the discounted risk-neutral probability density function evaluated at that strike. In practical terms, the options market is telling you exactly how it thinks the underlying will be distributed at every expiration — you just need the right mathematical tool to read it out.

This makes probability analysis fundamentally different from technical analysis or historical distribution fitting. Instead of assuming a normal or log-normal distribution and estimating its parameters from past data, you read the market's actual priced-in distribution directly. The shape can be skewed, fat-tailed, bimodal around binary events, or anything else the option chain supports. This is especially powerful around earnings, FDA decisions, and legal rulings where the implied distribution is materially different from anything historical data would suggest.

One practical caveat up front: the Breeden-Litzenberger identity is stated for European options and clean forwards. For American-style single-stock options with discrete dividends, the raw formula needs adjustment — either explicit dividend handling or conversion to the equivalent European forward price before inverting. The density shapes shown on index products are the cleanest case; single stocks with upcoming ex-dividend dates require more care.

How It Works

• Breeden-Litzenberger identity. ∂²C/∂K² = e^-rT × f(K), where C is the European call price, K is strike, r is the risk-free rate, T is time to expiration, and f(K) is the risk-neutral density at strike K. The second derivative of the call-price function with respect to strike recovers the pricing kernel's density — a model-free result under no-arbitrage assumptions.
• Surface fitting approach. Raw option chains are discrete and noisy, so directly finite-differencing the call-price curve produces unstable densities. Our platform fits an eSSVI volatility surface to market IVs, converts to a smooth call-price surface, and applies Breeden-Litzenberger analytically. When the eSSVI fit is unavailable, we fall back to a smoothed finite-difference estimate.
• Cumulative probability via delta. As a rough rule of thumb under driftless Brownian assumptions, the risk-neutral probability of finishing above strike K is approximated by N(d₂), the second term in the Black-Scholes formula. Call delta N(d₁) is close to this probability for short-dated ATM strikes, and |put delta| is the standard desk heuristic for P(S_T < K). These are approximations — the precise probability is N(d₂), which differs from N(d₁) by a σ√T adjustment that grows for long-dated or high-IV options.
• Probability cones. Derived from the implied distribution by integrating the density to build ±1σ, ±2σ bands at each point in time to expiration. These give a visual sense of how the market thinks the distribution widens as you look further out.
• Probability of touch versus probability at expiry. Delta approximates probability at expiration, not probability the price touches a strike along the way. Touch probabilities are roughly 2× at-expiry probabilities for OTM strikes under Brownian motion assumptions — important distinction for barrier-like strategies.

Risk-Neutral vs Real-World Probabilities

The probabilities extracted from option prices are risk-neutral, not actual statistical probabilities. Under risk-neutral measure, all assets drift at the risk-free rate and discounting uses the risk-free rate. Because investors are risk-averse, they pay a premium for assets (and options) that deliver payoffs in bad states — this premium inflates the risk-neutral probability of those bad states relative to their real-world frequency. The ratio of real-world to risk-neutral density is the pricing kernel or stochastic discount factor.

• Risk-neutral P(crash) > real-world P(crash). Crash protection is expensive because investors value it highly during calm regimes. The market-implied probability of a −20% SPX move in 30 days has historically run 2–4× the empirical frequency of such moves.
• Risk-neutral expected return = risk-free rate. Real-world expected return includes the equity risk premium, typically 4–7% annualized for broad equity indices. So a call's risk-neutral price uses a zero excess return, while a real-world fair value would reflect the equity risk premium.
• The variance risk premium (VRP). The difference between risk-neutral expected variance (priced into options) and realized variance (actually delivered by the underlying) is the VRP. It's the primary reason systematic premium-selling strategies have positive expected returns — sellers collect the risk-neutral expectation, buyers realize the real-world outcome, and the gap is the VRP.
• Directional tilt. The risk-neutral distribution for equity indices is typically left-skewed even when the real-world distribution is closer to symmetric — this reflects demand for downside protection (SPX put buying by asset managers) and creates the persistent volatility skew.

Trading Applications

• Probability of profit (POP). For any defined-risk option position, compute the implied probability of expiring profitable by integrating the density over the profitable strike range. An iron condor's POP is the probability the underlying finishes between the short strikes; a credit spread's POP is the probability it finishes on the credit-favorable side of the short leg.
• Expected value analysis. Multiply each payoff outcome by its risk-neutral probability and sum. For a zero-cost trade at fair value, expected value equals the risk-free-rate discount on the expected payoff. Any deviation reflects your view that the market is mispriced relative to the real-world distribution.
• Tail risk assessment. See how much probability the market assigns to extreme moves. P(S_T < 0.9 × S₀) for a 30-day SPY option, for instance, tells you the market-implied probability of a >10% decline over the next month. Compare to your own base-rate expectation.
• Earnings trades. Extract the implied earnings-move distribution from weekly option prices by subtracting the expected continuous-vol contribution. Compare the shape (often bimodal for high-beta names) to historical earnings-move distributions to size positions and pick strikes.
• Strike selection with conviction. When you disagree with the implied distribution, trade the part of the distribution where your edge is largest. Believe the market is underpricing crash risk? Buy OTM puts where the density is most dispersed against your view. Believe the market overstates tails? Sell iron condors at exactly those tail strikes.

Real-World Context

Jackwerth (2000) showed that risk-neutral densities extracted from SPX options diverge systematically from realized return distributions — the market consistently overprices tails relative to empirical frequencies. This finding drives most of the academic literature on variance risk premia and explains why systematic option-selling strategies have historically produced positive risk-adjusted returns. On the other side, Bates (2000) and subsequent work on jump-diffusion models show that the risk-neutral jump intensity rises sharply around FOMC, earnings, and election events — the market prices discrete jump risk in ways that standard diffusion models cannot capture. Reading the implied density around these events is one of the highest-information-density exercises in options analysis.

Common Pitfalls and Limitations

• Risk-neutral ≠ real-world. The most common misinterpretation is treating implied probabilities as statistical forecasts. Delta of 0.30 means the risk-neutral probability of finishing ITM is about 30%, not that the market "thinks there's a 30% chance." The real-world probability is typically lower (for puts) or higher (for calls), depending on the equity risk premium.
• Requires a smooth, liquid chain. Noisy or illiquid chains produce unstable density estimates. If adjacent strikes show bid-ask spreads wider than the strike increment, the Breeden-Litzenberger second-derivative becomes dominated by quote noise rather than true pricing information. Use at-expiration listed strikes rather than interpolated far-OTM prints.
• Interpolation introduces assumptions. The distribution is model-free only in the continuous-strike limit; in practice, fitting a surface (SVI, SSVI, SABR) introduces the surface's assumptions into the density. Different parametric fits can produce materially different tail behavior, particularly beyond the last quoted strike.
• Dividend and early-exercise adjustments. The Breeden-Litzenberger identity assumes European options. For American-style single-stock options with discrete dividends, the identity requires adjustments (either explicit dividend handling or conversion to the equivalent European forward price). Ignoring this produces biased densities.
• Static snapshots miss path dependence. The implied density is for terminal price at expiration. Paths matter for barrier options, American-style early-exercise decisions, and rebalancing strategies. A 20% POP for an iron condor does not mean a 20% chance of holding the trade to expiration without breach — touch probabilities are higher, and many trades are managed or stopped out mid-life.

References & Further Reading

• Breeden, D. T. and Litzenberger, R. H. (1978). "Prices of State-Contingent Claims Implicit in Option Prices." Journal of Business, 51(4), 621–651.
• Jackwerth, J. C. (2000). "Recovering Risk Aversion from Option Prices and Realized Returns." Review of Financial Studies, 13(2), 433–451.
• Bates, D. S. (2000). "Post-'87 Crash Fears in the S&P 500 Futures Option Market." Journal of Econometrics, 94(1-2), 181–238.
• Figlewski, S. (2009). "Estimating the Implied Risk-Neutral Density for the U.S. Market Portfolio." In Volatility and Time Series Econometrics: Essays in Honor of Robert Engle, Oxford University Press.

Explore live probability data: SPY · /ES · BTC-USD

This section is part of the Options Analysis Suite Documentation. Explore the full Charts & Analytics hub for every options analytics view.