This is a university course in the derivatives pricing (Black-Scholes, hedging, term structure models) that underpins the options business, not a system for finding trades. It teaches you the math professional market makers use to price and hedge risk, and it is honest that the math is a model, not the truth.
The one idea
Wilmott's whole project is to build option pricing from one idea: if you can hedge away the randomness in an option's payoff by holding the right amount of the underlying stock, then the option must be worth exactly what it costs to run that hedge, or arbitrage is free money. Chapter by chapter he builds toward the Black-Scholes formula for a call, Call = SN(d1) minus Ee^(-r(T-t))N(d2), and shows the same machinery pricing puts, binaries, barriers, and interest rate products.
The one chapter written for a trader rather than a pricing desk is chapter 10, on delta hedging when your forecast of volatility disagrees with the market's. Wilmott shows, with the actual mark-to-market math, that hedging a mispriced option with a delta built on your own (actual) volatility locks in a guaranteed profit by expiry but a bumpy, path-dependent ride to get there, while hedging with the market's (implied) volatility gives a smoother, day-by-day certain profit whose total size depends on the path the stock takes. Neither is wrong, they are different bets on the same mispricing.
Threaded through the book is a second theme: money management. Chapter 20 uses blackjack, not because trading is gambling, but because it is the cleanest place to show the Kelly criterion in numbers: bet a fraction f of your bankroll, and the fraction that maximizes long-run growth rate is f* = mean edge divided by variance (mu over sigma squared). Bet more than double that and your expected return goes negative even though the underlying game still favors you.
What it actually teaches
The book has no single trading method, but its core techniques stack in a fixed logical order that Wilmott follows chapter by chapter.
MODEL THE RANDOMNESS
Before you can price anything, assume the asset follows a lognormal random walk with a drift and a volatility, dS = mu S dt + sigma S dX. Chapter 4 shows this is a modeling choice, not an observed fact, by comparing it against real return data.
BUILD THE HEDGE
Chapter 3's binomial model and chapter 6's Black-Scholes derivation both rest on the same trick: hold delta shares of stock against one short option so that the combined position has no first-order exposure to the stock's next move. Whatever is left over must earn the risk-free rate, or there is arbitrage.
SOLVE FOR THE PRICE
Turning the hedging argument into a partial differential equation and solving it for a call gives Call = SN(d1) - Ee^(-r(T-t))N(d2), with d1 = [ln(S/E) + (r + sigma^2/2)(T-t)] / (sigma * sqrt(T-t)) and d2 = d1 - sigma*sqrt(T-t). For an at-the-money forward option there is a fast approximation Wilmott credits to Brenner and Subrahmanyam (1994): Call is about 0.4 * S * sigma * sqrt(T-t).
TRADE THE GAP BETWEEN ACTUAL AND IMPLIED VOLATILITY
Chapter 10's worked scenario: implied volatility priced into an option is 20%, you believe actual volatility will be 30%. Buy the option and delta hedge daily. Hedge with actual (30%) and your mark-to-market P&L is jumpy but the total profit by expiry is locked at V(S,t;30%) minus V(S,t;20%). Hedge with implied (20%) instead and the daily P&L is deterministic and always positive, given by half of (actual squared minus implied squared) times S squared times gamma times dt, but the total by expiry is path-dependent.
SIZE THE BET
Once you have an edge, chapter 20's Kelly criterion tells you how much to risk: f* = mu / sigma^2 of the bet's return distribution, which maximizes expected long-run growth. Wilmott's own example uses mu = 0.01, sigma = 1 to plot the growth curve; at twice the Kelly fraction expected growth is zero, so most practitioners run 'half Kelly' and give up about 25% of expected growth to cut volatility in half.
What it looks like Monday morning
A reader who works a trading or structuring desk walks away able to build a Black-Scholes pricer from scratch, read a delta or vega number and know exactly what it means, and see why a market maker's P&L on a hedged book depends on which volatility they hedged with. That is a genuine, checkable skill, not a vague takeaway.
A discretionary or systematic trader without a derivatives book gets less to actually do differently. The one piece that travels is money management: size positions as a fraction of edge over variance (Kelly), and default to half that fraction rather than the full number, because the full Kelly fraction has zero margin for a wrong probability estimate.
The more people following the same trend as you, the more money you will lose.Paul Wilmott, Paul Wilmott Introduces Quantitative Finance
Where it fails
- It is an academic textbook, and says so. The preface states the book is written for advanced undergraduate, MBA and taught graduate courses, and that it is a deliberately shortened version of the professional-level Paul Wilmott on Quantitative Finance (PWOQF2). An appendix spends five pages listing the roughly 60 additional chapters (stochastic volatility, jump diffusion, transaction costs, static hedging, crash modeling) you only get by buying the bigger book. Anyone shopping for a trading edge on the strength of the title is buying the classroom edition of a much longer professional text.
- Its own chapter on chart-based forecasting rejects most of what it describes. Appendix B walks through candlesticks, point and figure charts, Elliott waves, Fibonacci retracement levels and Gann lines, then Wilmott closes it by saying he plotted every technical indicator early in his career and 'was not very successful at it,' and that the only correlation he actually believes in is hemline length against the economy. The chapter is a tour of methods the author does not use.
- The Black-Scholes machinery that fills most of the book assumes constant, known volatility, which chapter 10 itself shows is false. Wilmott's own VIX-versus-realized-SPX-volatility chart in chapter 10 shows implied volatility running with a mean near 20% against realized volatility with a mean near 15% over the sample studied, a persistent, non-constant gap the base model cannot represent. The book flags the defect but the fix (stochastic volatility, jump diffusion, the models that actually handle it) is reserved for the professional edition, not this one.
- Nothing here is a retail trading system. The delta-hedging edge in chapter 10 assumes you already have an options book, real-time hedging infrastructure, borrow access, and a volatility forecast good enough to be right more often than the market. Chapter 20's Kelly sizing assumes you already know your edge's mean and variance precisely, which for a discretionary trader you almost never do; the book itself warns that betting the full Kelly fraction on a mis-estimated edge is how the strategy blows up.
- The 2nd edition (2007) is already an older snapshot of a fast-moving field. Numerical methods, credit derivatives (chapters 23-24, written pre-2008) and volatility modeling have all moved on substantially since publication. The book predates the post-2008 credit crisis literature on model risk and counterparty risk that reshaped how those chapters' subject matter is actually used on trading desks today.
Who it is for
Buy it if
You are heading into a derivatives, structuring or quant desk role, or a CQF-style course, and need the Black-Scholes and hedging math built up from first principles with real formulas you can check against Excel. It also earns its place for the chapter 10 delta-hedging argument and the Kelly criterion chapter, both of which are genuinely useful outside pure derivatives work.
Skip it if
You want a system to find or size trades in stocks or futures without touching options math. Skip it too if you already own or plan to buy Wilmott's full Paul Wilmott on Quantitative Finance, since this is explicitly the trimmed classroom version of that book and you would be paying twice for overlapping material.
As an Amazon Associate I earn from qualifying purchases. It costs you nothing extra, and it does not change what gets recommended or what the verdict says.