The equation that tamed the market — building a portfolio that perfectly tracks an option.
The chef and the consistent dish
Imagine you are a chef — the Portfolio Manager — trying to maintain a perfectly consistent dish — the Option Value — while the prices of your ingredients — the Stock — change constantly. Your job is not to predict where prices will go. Your job is to continuously adjust the recipe so that no matter what the ingredient price does, your dish always tastes exactly the same. This is the central problem of Black-Scholes-Merton, and it has a precise mathematical solution.
The portfolio manager's job is continuous adjustment — not prediction. The recipe changes every instant so the dish always matches the option.
1 · The two-bin portfolio
The Black-Scholes-Merton framework starts by splitting total wealth X(t) into exactly two bins. One bin is risky and exciting. The other is boring and safe. Together they must replicate the option at every instant. The art is in deciding how much to put in each bin — and that decision changes continuously.
📦 Bin 1 — The Stock
Δ(t) · S(t)
The volatile part. You hold Δ(t) shares of the stock. This bin jumps around with the market. Δ(t) changes continuously as the option's price sensitivity changes.
🏦 Bin 2 — The Cash
X(t) − Δ(t)·S(t)
The safe part. Everything not in the stock sits in a bank account earning the risk-free rate r. This bin grows steadily and predictably — no randomness here.
The two bins always sum to X(t). As Δ(t) changes, money flows continuously between the volatile stock bin and the safe cash bin.
2 · The master equation — a three-line P&L statement
The change in total wealth dX(t) is driven by exactly three forces. Once you see the formula as a P&L statement with three line items, it stops being intimidating and starts being a recipe.
What the whole portfolio earns if everything were in the bank. The baseline return.
Δ(t)(α−r)S(t)dt
Risk Premium
The extra expected return for holding stocks instead of cash. (α−r) is the reward for bravery.
Δ(t)σS(t)dW(t)
Volatility
The random noise. Market wiggle multiplied by position size and stock volatility.
Term
Finance name
Layman meaning
rX(t)dt
Risk-Free Growth
What you'd earn leaving everything in the bank — guaranteed, boring, essential baseline.
Δ(α−r)S dt
Risk Premium
The "brave person's bonus." You hold stock instead of cash, so you expect extra return (α−r) for that risk.
Δσ S dW
Volatility Shock
The random shaking. Completely unpredictable. The part we are about to eliminate.
Notice the structure: two deterministic terms (dt) and one random term (dW). The entire strategy of Black-Scholes-Merton is to choose Δ(t) so that the dW term in the portfolio perfectly cancels the dW term in the option's value. When that cancellation happens, the randomness vanishes entirely.
3 · The magic of discounting — removing the bank interest
The text makes a brilliant simplifying move. Instead of tracking X(t) directly, it tracks the discounted wealth e^(−rt)·X(t). Multiplying by e^(−rt) is like putting on glasses that filter out inflation and bank interest. Once you do that, the only thing left that makes your portfolio move is the movement of the stock itself.
The Discounting Trick
dX(t) = rX dt + Δ(α−r)S dt + ΔσS dW
× e⁻ʳᵗ ↓
d(e⁻ʳᵗX(t)) = Δ(t) · d(e⁻ʳᵗS(t))
The risk-free growth term disappears entirely.
The risk premium shrinks into the stock's discounted movement. What remains: discounted portfolio moves only with discounted stock.
Only the discounted stock move survives→
After discounting, two terms vanish. Only the discounted stock's movement drives the discounted portfolio. The equation becomes beautifully simple.
This is not just mathematical tidying. Discounting has a deep economic meaning: it says that all returns above the risk-free rate must be justified by risk taken. When we work in discounted terms, the risk-free component is automatically subtracted from everything. What remains is the pure risk-return story — which is exactly what we need to price an option fairly.
4 · The perfect hedge — eliminating randomness
We now reach the pivotal insight. The goal is to find Δ(t) such that the portfolio X(t) perfectly replicates the option value c(t, S) at every moment. If the option goes up by $1, the portfolio goes up by $1. If the option goes down by $3, the portfolio goes down by $3. Exactly. Always.
1
Write the option's movement using Itô-Doeblin
Apply the formula from Chapter 11 to c(t, S). This gives dc = cₜ dt + cₛ dS + ½cₛₛ (dS)². After substituting dS = αS dt + σS dW and applying the multiplication table, the dW term in dc is: cₛ · σS · dW
2
Write the portfolio's movement from the master equation
From Section 2, the dW term in dX is: Δ(t) · σS · dW. The dt terms are different but will be handled separately.
3
Set the dW terms equal — and they cancel
For the portfolio to track the option, the random parts must be identical. So set: Δ(t) · σS · dW = cₛ · σS · dW. Cancel σS from both sides. The answer: Δ(t) = cₛ(t, S) — the Delta is the derivative of the option price with respect to stock price.
4
With dW gone, only dt terms remain — and they must equal r
After the random part cancels, what is left is a completely deterministic system. In a world with no randomness, any investment must return exactly the risk-free rate r. If it returned more — free money (arbitrage). If it returned less — nobody would invest. The requirement that both sides equal r creates the Black-Scholes PDE.
The key insight: set Δ = cₛ and the random dW terms cancel perfectly. What remains is deterministic — and must earn the risk-free rate by no-arbitrage.
5 · The no-arbitrage constraint — where the PDE is born
After the dW terms cancel, we are left with a purely deterministic equation. The final step uses the most powerful idea in all of financial economics: no-arbitrage. In a world with no randomness, if two investments have identical payoffs, they must have identical returns. Otherwise, a riskless profit would exist — and markets would eliminate it instantly.
⚡ The No-Arbitrage Argument ⚡
After setting Δ = cₛ, the dt terms must satisfy:
cₜ + rS·cₛ + ½σ²S²·cₛₛ = r·c
This is the Black-Scholes PDE — the Partial Differential Equation
that the option price c(t, S) must satisfy at every point in time and price.
Left side: how the option changes from Theta (time), Delta (price), and Gamma (curvature).
Right side: what a risk-free investment earns.
These must be equal. If not — someone would exploit the difference for free money.
The Black-Scholes PDE is the Itô-Doeblin formula applied to the option price, then constrained by no-arbitrage. Every term has a name traders use every day: cₜ is Theta (time decay), cₛ is Delta (price sensitivity), cₛₛ is Gamma (curvature). The PDE says these three forces, weighted by r and σ, must always balance out to the risk-free rate. This single equation determines the fair price of every option ever traded.
Try it — the replicating portfolio in action
Simulate a GBM stock path. A delta-hedging portfolio continuously adjusts Δ to track a European call option. Watch as the portfolio value (green) and the option's theoretical value (gold) track each other through the entire path — despite market randomness — because the dW terms cancel perfectly.
20%
5%
$100
Final stock price
—
Option value
—
Portfolio value
—
Hedge error
—
Gold = theoretical option price (BSM formula) | Green = replicating portfolio value | Blue = stock price path (scaled)
Chapter 13 — the architecture of the perfect hedge:
The two-bin portfolio: Wealth X(t) is split between Δ(t) shares of stock and X(t)−Δ(t)S(t) in cash earning rate r. The split changes continuously.
The master equation: dX = rX dt + Δ(α−r)S dt + ΔσS dW. Three terms: risk-free baseline, risk premium, random noise.
The discounting trick: Multiply by e^(−rt) and the risk-free growth vanishes. Only the discounted stock drives the discounted portfolio.
The perfect hedge: Set Δ(t) = cₛ(t,S) — the derivative of the option price with respect to stock price. The dW terms in portfolio and option cancel exactly. Randomness is eliminated.
The BSM PDE: cₜ + rS·cₛ + ½σ²S²·cₛₛ = rc. This is what remains after cancellation. It is the Itô-Doeblin formula constrained by no-arbitrage. It determines the fair price of every option.
What comes next: The Black-Scholes PDE has an explicit solution — the famous Black-Scholes formula. It tells you the exact fair price c(t,S) for a European call option as a closed-form expression involving the normal distribution. That formula is the next chapter.