The Kelly Criterion
Optimal Capital Allocation, Growth, and Decision-Making Under Uncertainty
In financial markets, identifying an opportunity is only part of the problem.
Equally important is determining how much capital to allocate to that opportunity. The question of position sizing is often treated as secondary, yet it plays a central role in shaping long-term outcomes. Allocate too little, and potential gains are diluted. Allocate too much, and the risk of significant loss or ruin increases.
The Kelly Criterion provides a formal framework for addressing this problem.
Originally developed in the context of information theory and gambling, it determines the optimal fraction of capital to allocate to a favourable bet in order to maximise long-term growth. Its relevance extends directly to financial markets, where decisions are made under uncertainty and outcomes compound over time.
The Core Principle
The Kelly Criterion is based on a simple idea. Given a set of probabilistic outcomes, there exists an optimal level of capital allocation that maximises the long-term growth of wealth.
This approach differs from simply maximising expected return. Instead, it focuses on sustainable growth across repeated decisions, recognising the compounding nature of capital over time.
The framework determines position sizing by balancing three core factors: the probability of success, the probability of failure, and the magnitude of the potential payoff relative to the potential loss.
In this sense, the Kelly Criterion captures the relationship between probability, risk, reward, and capital allocation within a probabilistic decision-making framework.
Growth and Compounding
The central objective of the Kelly framework is to maximise the long-term growth rate of capital. This is achieved by considering the logarithm of wealth rather than its absolute value.
The logarithmic function reflects the impact of compounding. Losses and gains do not affect wealth symmetrically. A loss requires a larger subsequent gain to recover.
By maximising expected log growth, the Kelly Criterion balances return and risk in a way that accounts for this asymmetry. This makes it particularly suited to repeated decision-making environments.
The Trade-Off Between Risk and Growth
The Kelly Criterion embodies a fundamental trade-off.
Allocating more capital increases potential returns but also increases the risk of large losses. Allocating less capital reduces risk but limits growth. The optimal fraction lies between these extremes, at this level, the expected growth rate is maximised.
However, this optimum is sensitive to inputs. Small errors in estimating probabilities or payoffs can lead to suboptimal or even harmful allocations.
Sensitivity to Estimation
A critical aspect of the Kelly framework is its reliance on accurate inputs. The probability of success and the payoff ratio must be estimated.
In financial markets, these estimates are inherently uncertain, namely:
probabilities may be difficult to quantify
payoffs may vary across scenarios
conditions may change over time
This introduces model risk, as overestimating probability or payoff can lead to excessive allocation, increasing the risk of drawdown. This sensitivity highlights the importance of conservative estimation and robust analysis.
Fractional Kelly and Practical Application
In practice, many participants use a fraction of the Kelly allocation. This approach reduces sensitivity to estimation error and limits volatility.
Fractional Kelly maintains the structure of the framework while providing a margin of safety. It reflects the recognition that real-world conditions differ from theoretical assumptions.
Risk of Ruin and Drawdowns
While the Kelly Criterion maximises long-term growth, it does not eliminate short-term volatility.
Optimal allocations may still involve significant drawdowns. This introduces a practical consideration. Participants must be able to withstand fluctuations without being forced to exit positions.
The concept of risk of ruin remains relevant, as even if long-term growth is maximised, short-term losses can have irreversible consequences if capital is depleted. This reinforces the importance of aligning allocation with risk tolerance and constraints.
Time Horizon and Repetition
The effectiveness of the Kelly framework depends on repetition. It assumes that decisions are made repeatedly under similar conditions. In such environments, the law of large numbers allows the expected growth rate to emerge over time.
In financial markets, conditions are not always stable; opportunities may be infrequent or vary in characteristics. This affects the applicability of the framework, as it requires adaptation.
Integration with Market Dynamics
The Kelly Criterion does not operate in isolation.
It interacts with market dynamics, such as:
liquidity constraints may limit allocation
execution costs may affect payoffs
changing conditions may alter probabilities
These factors must be considered, as the theoretical allocation may differ from the practical allocation. Understanding this interaction is essential for effective application.
The MorMag Perspective
At MorMag, the Kelly Criterion is viewed as a foundational framework for capital allocation.
It provides a structured approach to balancing risk and growth. However, its application is adapted to the realities of financial markets.
This includes:
conservative estimation of probabilities
use of fractional allocation
integration with broader risk management
The framework is not applied mechanically. It is interpreted within the context of uncertainty, market conditions, and system dynamics.
From Bet Sizing to Decision Framework
The Kelly Criterion extends beyond simple bet sizing, it represents a way of thinking about decisions.
It emphasises:
probabilistic evaluation
alignment of allocation with conviction
consideration of long-term outcomes
This perspective integrates with broader approaches to market analysis.
Conclusion
The Kelly Criterion provides a rigorous framework for determining how much capital to allocate to uncertain opportunities.
By maximising expected logarithmic growth, it balances return and risk in a way that reflects the compounding nature of capital. While its theoretical formulation offers clarity, its practical application requires careful consideration of estimation, market conditions, and risk tolerance.
At MorMag, this framework informs a disciplined approach to capital allocation, integrating quantitative insight with an understanding of real-world constraints.
In financial markets, success is not defined solely by identifying opportunities. It is defined by allocating capital effectively in the presence of uncertainty.
