Why Portfolio Construction Matters More Than Individual Trades

The most carefully analysed individual trade is still subject to the full volatility of whatever market it sits in. No amount of entry precision, no tight stop placement, no thorough research changes the fact that a single position is exposed to a single market’s full range of possible moves, including the ones that were not anticipated.

Portfolio construction determines the relationship between individual trade outcomes and account-level results. A well-constructed portfolio of positions produces more stable returns than any single position, not because the individual trades are better, but because the structure of how they are combined reduces the account’s dependence on any single outcome going the right way.

The aggregate performance of a trading account over time is determined far more by portfolio structure than by the quality of any individual trade selection. Yet most retail traders spend the majority of their analytical energy on individual trades and almost none on portfolio construction.

What Portfolio Construction Actually Means

Portfolio construction is the deliberate design of how multiple positions relate to each other in terms of risk, correlation, and aggregate exposure. A constructed portfolio is not a collection of trades that looked attractive individually. It is a set of positions whose combination produces properties that no individual trade could deliver alone: reduced dependence on any single market outcome, exposure to multiple independent return drivers, and a distribution of outcomes that is more stable than the sum of individual trade distributions.

The core tool of portfolio construction is diversification, but genuine diversification, not mere appearance. As covered in Correlation and Portfolio Risk in Multi-Asset Trading, positions in multiple instruments that share the same underlying driver are not a diversified portfolio. They are a concentrated position expressed through multiple instruments. A portfolio of long EUR/USD, long GBP/USD, short DXY, and long gold looks like four positions. If all four are driven primarily by dollar weakness, it is one position scaled up.

Genuine diversification requires positions whose return drivers are genuinely independent under the conditions most likely to produce large moves, specifically under stress, when correlations between risk assets tend to converge. The practical challenge is that true diversification within a leveraged trading context is harder to achieve than it appears and requires more active management than most retail traders apply.

The Kelly Criterion and Optimal Position Sizing Across a Portfolio

The Kelly Criterion is the mathematical formula that describes the theoretically optimal fraction of capital to allocate to a bet with a positive expected value to maximise long-run account growth. For a bet with probability p of winning and a reward-to-risk ratio of b, the Kelly fraction is (bp – (1-p)) / b.

A trade with a 55 per cent win rate and 1.5:1 reward-to-risk produces a Kelly fraction of approximately 23 per cent, meaning Kelly theory suggests risking 23 per cent of the account on that trade to maximise long-run growth. In practice, full Kelly is almost never used in professional trading because its assumptions about the certainty of probability estimates are unrealistic, and the resulting position sizes produce drawdowns that are psychologically and practically unsustainable.

Professional traders typically use fractional Kelly, sizing positions at 25-50 per cent of the full Kelly amount. The reduced sizing sacrifices some theoretical return in exchange for substantially reduced variance. A half-Kelly position on the trade above would risk approximately 11.5 per cent of the account, still aggressive by most standards but reflecting the Kelly logic of allocating more capital to higher-expected-value opportunities rather than flat-sizing across all trades.

The portfolio-level extension of this logic is that position sizes should vary in proportion to the estimated expected value of each trade, with higher-conviction or higher-expected-value opportunities receiving larger allocations and lower-conviction trades receiving smaller ones. Flat sizing across all trades, regardless of quality, treats a mediocre setup the same as an exceptional one, resulting in a systematic misallocation of the risk budget.

The Concept of Risk-Adjusted Return

Individual trade performance is typically assessed in terms of absolute return: how much did this trade make or lose? Portfolio performance in professional practice is assessed in terms of risk-adjusted return: how much return was generated per unit of risk taken?

The Sharpe ratio is the most widely used risk-adjusted return metric. It measures the portfolio’s excess return above the risk-free rate, divided by the portfolio’s standard deviation of returns. A Sharpe ratio above 1 indicates that the portfolio is generating more than 1 unit of return per unit of volatility. A Sharpe ratio below 0.5 suggests the portfolio is generating a modest return relative to its volatility.

Two portfolios with identical absolute returns over a period can have very different Sharpe ratios if one achieved those returns with consistent, moderate volatility, while the other did so with large swings. The second portfolio’s returns are less reliable: the same strategy applied over a different period might produce a materially different outcome because the large swings indicate a high dependence on specific outcomes going the right way.

Professional portfolio managers are evaluated on risk-adjusted returns rather than absolute returns precisely because absolute returns, without a risk context, are not a useful performance measure. A 30 per cent annual return achieved through 70 per cent drawdowns is a poor result. A 15 per cent annual return with maximum drawdowns of 8 per cent is excellent.

Retail traders who track only absolute P&L are missing the most informative part of their performance data.

Stress Testing the Portfolio

A constructed portfolio should be stress-tested against scenarios that represent the tail risks most likely to produce large simultaneous adverse moves across multiple positions.

The relevant scenarios for a multi-instrument retail CFD portfolio typically include a broad risk-off episode where equity indices fall simultaneously and risk-sensitive currencies weaken against safe havens, a sharp dollar strengthening event driven by a Federal Reserve surprise, a commodity supply shock driving oil and energy-related instruments sharply, and a geopolitical event with immediate market impact.

For each scenario, the question is: what is the approximate aggregate impact on the portfolio at current position sizes, and is that impact within the account’s risk tolerance?

The stress test does not require sophisticated modelling. A rough calculation based on the historical behaviour of each position type during similar scenarios, applied to the current position sizes, produces a meaningful estimate of the portfolio’s tail exposure. That estimate, reviewed before new positions are added, is the portfolio-level risk management check that most retail traders skip.

Why Individual Trade Quality Has Diminishing Returns

The relationship between trade selection quality and account performance is non-linear. Improving from a 45 per cent win rate to a 55 per cent win rate, all else equal, has a substantial impact on expected value. Improving from 55 to 60 per cent has a smaller impact, and improving further from there has an even smaller one, because the variance of outcomes in any probabilistic trading strategy is large enough that the noise of the outcome distribution increasingly dominates the difference in win rate.

Portfolio construction improvements, by contrast, can reduce portfolio volatility substantially regardless of the quality of individual trades. A well-constructed portfolio of average trades outperforms a poorly constructed portfolio of excellent trades over any extended period where concentrated adverse moves periodically impair the latter.

The implication is not that trade quality does not matter. There is a level of trade quality beyond which further refinement yields less return than equivalent effort applied to portfolio construction. Most retail traders are far below that level on portfolio construction while spending virtually all their energy on trade selection.

The Bottom Line

Portfolio construction is the design of how positions relate to one another in terms of risk, correlation, and aggregate exposure. A well-constructed portfolio reduces dependence on any single outcome, allocates risk in proportion to expected value, and produces a distribution of account-level returns that is more stable than individual trade results would suggest.

The analytical tools, correlation management, Kelly-proportional sizing, stress testing, and risk-adjusted performance measurement are all available without institutional infrastructure. What they require is the decision to evaluate performance and allocate risk at the portfolio level rather than the trade level. That decision, made consistently, is among the highest-leverage improvements available to any active trader.

Nicholas Spencer-Skeen is Senior Executive Officer at Lunaro Financial Services. He has spent over 35 years in the FX and derivatives communities, building operations for three major global institutions. He has served on the Futures Industry Clearing Committee and the London Clearing House user committee.

Disclaimer:

This material is a marketing communication and is provided for general information and educational purposes only. It does not take into account your personal circumstances, objectives or needs. Any opinions are those of the author at the time of writing and may change without notice. Nothing in this material constitutes (or should be construed as) financial, investment, legal, regulatory or tax advice, or a recommendation to engage in any investment activity. You should not rely on this material when making investment or trading decisions.