How simulation turns random draws into a price
Instead of solving an integral on paper, Monte-Carlo pricing rolls the dice many times and averages the result. Under the risk-neutral measure the underlying is assumed to follow a lognormal random walk, so each trial draws a standard normal shock Z and projects one possible price at expiry, ST. The option’s payoff is computed for that terminal price, the payoffs from every trial are averaged, and the mean is discounted at the risk-free rate. By the law of large numbers that discounted average converges to the same fair value theBlack-Scholes calculatorcomputes analytically — Monte-Carlo simply estimates the integral by sampling rather than evaluating it in closed form.
The simulation
ST = S · exp( (r − q − σ²/2)·T + σ·√T·Z ), Z ~ N(0,1)
Price = e−rT · mean( payoff(ST) )
Standard error ∝ 1 / √N
where S = spot, K = strike, T = years to expiry, σ = volatility, r = risk-free rate,q = dividend yield, Z is a standard normal draw, andN is the number of simulated paths. The payoff ismax(ST − K, 0) for a call andmax(K − ST, 0) for a put.
What makes this calculator different
- It shows the convergence. The estimate is plotted as paths increase, with a confidence band around it, so you can watch the price settle onto its true value instead of just reading a single number.
- It reports the standard error honestly. Every estimate comes with its uncertainty, making clear that simulation noise is real and that precision costs paths at a 1/√N rate.
- It uses antithetic variates. Each random draw is paired with its mirror image to cut the variance of the estimate, so you reach a given accuracy with fewer paths.
- It reuses the same tested core as the Black-Scholes page.The risk-neutral inputs and discounting share the same vetted code, so the simulated price and the closed-form price are directly comparable.
Frequently asked questions
What is Monte-Carlo option pricing and how does it work?+
Monte-Carlo pricing values an option by simulation rather than by a formula. Under the risk-neutral measure the underlying is assumed to follow geometric Brownian motion, so each trial draws a random shock and projects one possible terminal price at expiry. For every simulated terminal price the option payoff is computed, and the average of those payoffs is discounted back to today at the risk-free rate. Run enough trials and that discounted average converges to the option’s fair value — the same number a closed-form model would give for a European call or put.
What is the simulated price-path formula?+
Each terminal price is generated from the lognormal solution of geometric Brownian motion: S_T = S·exp((r − q − σ²/2)·T + σ·√T·Z), where Z is a standard normal draw, Z ~ N(0,1). The (r − q) term is the risk-neutral drift, the −σ²/2 term is the Itô correction that keeps the expected price unbiased, and σ·√T·Z injects the random diffusion. Because we only need the price at expiry for a European payoff, the path can be taken in a single jump rather than stepped through time. The payoff is then max(S_T − K, 0) for a call or max(K − S_T, 0) for a put.
Why does Monte-Carlo converge to the Black-Scholes price?+
Both methods are evaluating the same thing — the discounted risk-neutral expectation of the payoff. Black-Scholes computes that expectation analytically by integrating the lognormal distribution; Monte-Carlo estimates the very same integral by random sampling. By the law of large numbers the sample mean of the discounted payoffs converges to the true expectation as the number of paths grows. So with enough trials the simulated price settles onto the Black-Scholes value, and you can watch the estimate tighten around it as paths increase.
What is standard error and why does it shrink with 1/√N?+
The standard error is the estimated uncertainty of the simulated price — roughly the standard deviation of the payoffs divided by the square root of the number of paths, N. It tells you how far the Monte-Carlo estimate is likely to sit from the true value, which is why this tool reports it alongside the price. Because the error scales with 1/√N, it falls slowly: to halve the standard error you must quadruple the number of paths, and to cut it tenfold you need a hundred times as many. That square-root law is the fundamental cost of Monte-Carlo precision.
When is Monte-Carlo actually needed instead of a formula?+
For a plain European call or put, Black-Scholes gives an exact answer instantly, so simulation is purely educational there. Monte-Carlo earns its place when no closed form exists: path-dependent options such as Asian (average-price), barrier, or lookback options, where the payoff depends on the whole price trajectory, and multi-asset options such as baskets or rainbows whose joint distribution is intractable analytically. In those cases you simulate the full path or correlated paths and average the payoffs. The single-step European case shown here is the foundation those harder problems build on.
Disclaimer: This calculator is foreducation and illustration only. Monte-Carlo estimates carry simulation noise and will vary slightly between runs; the output is not a tradeable quote and real option prices will differ. Nothing here is investment, tax, or trading advice.