Capital Allocation

Introduction

Capital allocation is the process of deciding how much of a trading account should be exposed to each strategy, position, or risk source. Position sizing answers how large an individual trade should be. Capital allocation asks the broader question: how should the available risk budget be distributed across the whole portfolio?

A trader can have good individual setups, disciplined stops, and positive expectancy while still allocating capital poorly. Too much capital can be concentrated in one strategy, one market driver, one time horizon, or one correlated group of positions.

Professional capital allocation therefore connects Strategy Edge, Position Sizing, Correlation Risk, Concentration Risk, Drawdown, and Risk of Ruin.

1. Capital Allocation vs Position Sizing

Position sizing is primarily an individual-trade decision. Capital allocation is a portfolio decision.

Position sizing: How large should this trade be?

Capital allocation: How much of the total portfolio risk budget should this strategy or exposure receive?

A strategy can have an excellent setup but still receive a smaller allocation because the portfolio is already exposed to similar risk.

Allocation is therefore about priorities and interactions, not only trade quality.

2. Start With a Risk Budget

Capital allocation should begin with a predefined risk budget rather than with the number of attractive trades available.

Maximum risk for one trade

Maximum combined open risk

Maximum exposure to one strategy

Maximum exposure to one sector or common factor

Maximum portfolio drawdown before review

The exact limits are strategy- and trader-dependent. The important principle is that limits are defined before the portfolio becomes stressed.

3. Why Equal Allocation Is Not Always Equal Risk

Allocating the same dollar amount to every strategy does not necessarily create equal risk. Strategies can have different volatility, payoff distributions, drawdown behavior, and correlation with other strategies.

For example, two strategies each receive $20,000. If one regularly produces small fluctuations and the other experiences much larger swings, their contribution to portfolio risk can be very different.

Equal dollars and equal risk are not the same thing.

4. Risk-Based Allocation

A risk-based approach focuses on how much portfolio risk each strategy contributes rather than simply how much capital it receives.

A strategy with a large historical drawdown or high volatility may require less capital to keep its portfolio contribution within the desired range.

This does not mean historical volatility is a perfect risk measure. It is one input among several.

5. Strategy Edge and Allocation

Capital allocation should consider whether the underlying Strategy Edge is supported by evidence.

Strong evidence can justify consideration for meaningful allocation.

Weak or uncertain evidence should not automatically receive the same allocation as a well-tested strategy.

Positive historical performance alone does not prove future edge.

Allocation should reflect both evidence quality and portfolio interaction.

Capital allocation is therefore downstream from research. It cannot manufacture an edge that does not exist.

6. Correlation Changes Allocation Decisions

E-029 showed that strategies can be correlated even when their rules and instruments look different. Capital allocation must account for this.

Suppose three strategies each receive 15% of the risk budget. If all three depend on the same market condition, their combined exposure may be much larger than the labels suggest.

The allocation decision should therefore consider the group, not only each strategy independently.

7. Concentration Risk

Concentration occurs when too much of the portfolio depends on one source of outcome.

One strategy dominates the portfolio.

Several strategies depend on one sector.

Several positions react to one macro factor.

A large share of risk is concentrated in one market direction.

Several strategies can lose together during the same regime.

A portfolio can contain many positions and still be highly concentrated.

8. Allocation by Strategy Family

A useful practical approach is to group strategies into families based on their underlying behavior.

Trend-oriented

Mean-reversion

Momentum

Breakout

Event-sensitive

Short-biased

The labels are not enough by themselves. Actual return behavior should determine whether strategies really belong to the same risk family.

Once families are identified, the trader can set limits on the total risk allocated to each family.

9. Capital Allocation and Drawdown

Drawdown should be treated as a portfolio property. A strategy with a deep historical drawdown may require a smaller allocation if the trader wants to keep portfolio drawdowns within a predefined range.

The reverse can also happen: a strategy with moderate individual drawdown may become problematic when combined with several highly correlated strategies.

Therefore allocation should be evaluated using combined portfolio drawdown, not isolated strategy statistics alone.

10. Capital Allocation and Risk of Ruin

Risk of Ruin is affected by how much capital is exposed to losing sequences and how those losses interact.

If several strategies can lose together, allocating the maximum theoretical amount to each can create a portfolio-level survival problem.

A sensible allocation process therefore asks not only how much return a strategy might generate, but also what happens to the account during adverse joint outcomes.

11. R-Multiple as a Common Language

R-Multiple allows different strategies to be compared in a common unit based on predefined risk.

A strategy producing +0.3R per average trade and another producing +0.8R are not automatically comparable without considering frequency, distribution, drawdown, and sample size. But R makes their outcomes easier to place on the same analytical framework.

For allocation research, R helps separate the strategy's outcome characteristics from the absolute dollar size assigned to it.

12. Fixed Allocation vs Dynamic Allocation

Fixed allocation means the strategy receives a predefined share of the portfolio or risk budget. Dynamic allocation changes according to predefined rules.

Fixed allocation is simpler and easier to audit.

Dynamic allocation can respond to measured changes in risk or strategy behavior.

Dynamic allocation can also introduce additional model risk and overfitting.

Any dynamic rule should therefore be specified before it is tested.

Changing allocation reactively because a strategy has recently won or lost is not the same as having a disciplined dynamic allocation model.

13. Kelly as an Allocation Reference

The Kelly Criterion can provide a theoretical growth-oriented reference for allocation when its assumptions are appropriate. The E-026 and E-027 lessons showed why estimation error makes full Kelly sensitive.

In a portfolio context, single-strategy Kelly values should not simply be added together. Correlation, joint outcomes, and portfolio constraints matter.

Kelly can therefore be one analytical input, not the entire allocation system.

14. Capacity and Liquidity

A theoretically attractive allocation may not be executable at the desired size. Liquidity, spread, slippage, market impact, and available financing can limit practical capacity.

This is especially relevant when a strategy trades smaller or less liquid instruments.

Capital allocation must therefore answer two questions: “What would the model allocate?” and “What can actually be traded?”

15. Rebalancing

A portfolio's actual weights change as positions win, lose, open, and close. Rebalancing determines when and how those weights are brought back toward the intended allocation.

Calendar-based review

Threshold-based review

Risk-based review

Strategy-state review

Rebalancing should be governed by explicit rules. Otherwise, the trader can unintentionally increase risk after gains or reduce exposure after losses in an inconsistent way.

16. Expert Case Study — Equal Capital, Unequal Risk

A trader allocates $25,000 to each of four strategies. The portfolio looks perfectly balanced by capital.

Strategy A and B have similar behavior, while C and D are more independent. A later analysis shows that A and B frequently lose during the same market regime.

The dollar allocation was equal, but the risk contribution was not. The trader therefore evaluates allocation by strategy family and common drivers instead of dollar amounts alone.

17. Expert Case Study — Attractive Strategy, Limited Allocation

A newly researched strategy shows strong historical performance and positive Expectancy. However, its return series is highly correlated with an existing strategy that already consumes a significant part of the portfolio risk budget.

The new strategy can still be useful, but adding a full independent allocation would materially increase concentration.

The allocation decision is therefore not simply “Is the strategy profitable?” It is “What does adding this strategy do to the portfolio that already exists?”

18. Professional Capital Allocation Workflow

Define the total portfolio risk budget.

Define the maximum risk per trade.

Measure the Strategy Edge and evidence quality of each strategy.

Measure strategy Drawdown and outcome characteristics.

Review strategy correlation and common risk drivers.

Group correlated strategies into risk families.

Set maximum allocation or risk limits for each family.

Stress-test joint losses and portfolio Risk of Ruin.

Check practical liquidity and execution capacity.

Document the allocation rules and review triggers.

Reassess according to a predefined process rather than recent emotions.

19. Common Mistakes

Allocating equal dollars and assuming equal risk.

Ignoring Correlation Risk between strategies.

Allocating capital solely according to recent performance.

Giving every strategy its own maximum risk limit without a portfolio cap.

Ignoring Concentration Risk created by common factors.

Using Kelly numbers from small samples as if they were precise.

Ignoring liquidity and execution constraints.

Changing allocations after every winning or losing streak.

Optimizing allocation rules on the same data used to prove the strategy.

20. Expert Exercise — Build a Portfolio Allocation Plan

Create a hypothetical portfolio containing 5–8 strategies. For each strategy, record its evidence quality, average R outcome, Drawdown, volatility characteristics, correlation with the other strategies, and proposed risk allocation.

Set a total portfolio risk budget.

Set a maximum allocation for one strategy.

Group strategies by common behavior.

Set a limit for each group.

Create a stress scenario in which several correlated strategies lose together.

Check whether the portfolio remains inside the predefined Drawdown and survival limits.

Add one new strategy and evaluate its incremental portfolio effect.

Write the rule that determines when allocation is reviewed.

The objective is to move from “Which strategies do I like?” to “How should limited risk capital be distributed across the strategies I actually have?”

21. Knowledge Check

What is the difference between Position Sizing and Capital Allocation?

Why does equal dollar allocation not necessarily mean equal risk?

Why must Correlation Risk be considered when allocating capital?

What is Concentration Risk at the strategy level?

Why should portfolio Drawdown be measured on the combined portfolio?

How can Risk of Ruin influence allocation?

Why is R-Multiple useful when comparing strategies?

What are the main differences between fixed and dynamic allocation?

Why should liquidity be considered in capital allocation?

Why should allocation rules be defined before the portfolio is under stress?

Key takeaway: Capital allocation is the bridge between individual strategy quality and portfolio survival. The goal is not to give every strategy the same amount of capital. The goal is to distribute limited risk capital while accounting for Strategy Edge, Position Sizing, correlation, concentration, Drawdown, liquidity, and Risk of Ruin. A professional allocation process makes these trade-offs explicit and rule-based.

Copying, reproducing, adapting, distributing, or publicly displaying any part or all of the Academy’s textual content without the author’s prior written consent is prohibited.