How Many Managers Are Too Many?

A Thought Experiment on Diversification in Multi-Manager Funds

A recurring question in fund design is simple on the surface, but subtle in substance:

Let’s strip away noise, selection skill, and real-world frictions, and run a clean thought experiment.

The Thought Experiment

Assume an idealized world:

1. Every manager (or pod) is on the efficient frontier

– same expected return

– same risk – return optimality

2. Managers are mutually uncorrelated

– correlation = 0

3. Total capital is fixed at X

We compare two portfolios:

• Portfolio A

Invest all X into one manager.

• Portfolio B

Invest X equally across infinitely many managers

(each gets X / N, with N → ∞).

What happens to the return curve?

What Stays the Same

The expected return of Portfolio B is simply the weighted average of the managers’ expected returns. Since they are all the same, Portfolio A and Portfolio B have identical expected returns.

Diversification does not dilute return in expectation.

What Changes Dramatically

Portfolio A: One Manager

• You experience the full volatility of a single return stream.

• Drawdowns are real, timing matters, and survival depends on that one path.

• The equity curve is noisy and regime-dependent.

Portfolio B: Infinitely Many Managers

• Individual fluctuations cancel out.

• Volatility shrinks as the number of managers grows.

• In the limit, randomness disappears.

The return curve converges to something close to a deterministic growth path.

Conclusion

Under the assumptions of this thought experiment, increasing the number of managers does not change expected return.

All managers share the same long-term return structure. Allocating capital across more of them neither creates nor dilutes return; it only changes how closely realised outcomes track that structure.

With one manager, realised performance reflects the full extent of deviation from the underlying return expectation. With many independent managers, these deviations are averaged across the cross-section, and realised returns converge more closely to the expected outcome.

In this idealised setting, adding managers is strictly beneficial: expected return is preserved, while uncertainty is reduced.

A Note on Value Investing

The framework above assumes the efficient frontier is given and shared by all participants. Under that assumption, diversification improves outcomes by reducing deviation around a common return structure.

By contrast, investors such as Warren Buffett start from a different premise: that the market’s efficient frontier is misdrawn. If risk is systematically overestimated and return underestimated for a small set of businesses, then concentrating capital is not a rejection of diversification, but a rational response to operating on a different frontier.

The distinction, therefore, is not between diversification and conviction, but between optimizing within a given frontier and believing one can consistently identify opportunities beyond it.

Back to Reality

1. Practical limits to the idealised model

Of course, the real world never satisfies the assumptions exactly.

The number of managers is finite.

Correlations are never truly zero.

Dependencies tend to rise during periods of stress.

Still, the direction of the result remains valid.

In practice, a fund can improve its overall behaviour by:

• Allocating to more managers of comparable, high quality

• Actively seeking low correlation across strategies, instruments, and decision processes

• Avoiding concentration in shared failure modes, even when expected returns appear similar

Multi-manager investing is therefore not about increasing return, but about realising the same return more reliably.

2. Exceptional alpha and portfolio choice

From a value investing perspective, some opportunities may appear to exhibit a risk – return profile far beyond the conventional efficient frontier.

In practice, however, such extreme alpha is rarely observable ex ante, often difficult to validate over short horizons, and even harder to allocate capital to with confidence at scale.

For a quantitative fund operating under uncertainty and capital constraints, it is therefore more robust to assume a given frontier and apply modern portfolio theory, rather than attempt to underwrite rare and potentially unscalable breakout outcomes.

How Many Managers Are Too Many? was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.