Trading risk

What this book actually teaches

Trading Risk (2004) by Kenneth L. Grant is one of the few books written specifically for portfolio risk management in an active trading context — not for buy-and-hold investors, not for financial engineers building derivatives books, but for portfolio managers, prop traders, and risk officers who need a practical framework for managing risk across a book of active positions in real time. Grant spent years as a risk manager at large hedge funds and commodity trading advisors, and the book reflects direct experience with the problems that occur when risk management is treated as a compliance function rather than a performance driver.

The central argument is that most trading operations underinvest in risk management relative to the return it generates. Grant frames position sizing and risk allocation not as constraints on return but as the primary mechanism through which return is optimized — the Kelly criterion logic applied to portfolio management, where finding the optimal position size for a given strategy's risk-reward profile generates more return per unit of risk than improving entry and exit signals marginally. This reframing — risk management as alpha source rather than alpha limiter — is the book's core intellectual contribution.

The practical content covers the measurement toolkit that trading risk managers use: value at risk (VaR) and its limitations, expected shortfall, stress testing methodologies, and the specific risk decomposition approaches that allow a manager to understand how much of total portfolio risk comes from individual positions, sectors, factors, and correlation regimes. Grant is clear about VaR's well-documented failures — it assumes normal distributions and recent correlation structures, both of which break down in exactly the market conditions when risk measurement matters most — and covers more robust alternatives.

The performance analytics chapters are among the most useful in the book for practicing managers. Grant covers Sharpe ratio calculation and its limitations (its denominator rewards strategies that suppress volatility through illiquid positions or option writing without capturing actual risk), the Sortino ratio and downside deviation as alternatives, and the maximum drawdown and recovery analysis that tell a different story about strategy robustness than return-based metrics alone. The discussion of how to distinguish skill from luck in a trading track record — and the statistical power required to do so with confidence — is more rigorous than most trading texts manage.

The position sizing methodology draws on Kelly and fractional Kelly approaches, with realistic discussion of why full Kelly sizing is too aggressive for most practical trading operations and how the trade-off between growth rate and drawdown depth shifts as the Kelly fraction is reduced. The correlation section covers the regime-dependency of correlations — how positions that appear uncorrelated in normal markets can move together sharply in stress scenarios — and the practical implications for portfolio construction.

The weaknesses are primarily scope-related. The book is written for professionals managing substantial portfolios; retail traders managing personal accounts will find some of the infrastructure requirements (risk system integration, real-time position data, scenario analysis across hundreds of positions) impractical to implement at small scale. The VaR and expected shortfall material, while accurately presented, requires quantitative background to use well.

For risk managers, portfolio managers at trading firms, and serious independent traders who want a rigorous professional framework for thinking about risk allocation, position sizing, and performance attribution, this remains one of the best specialized texts in the field.

Read next

About Kenneth L Grant

Read more from Kenneth L Grant and explore the full bibliography on ClearValue Books.