Ask a sophisticated allocator what keeps them up at night and sleeve sizing will come up before long. It sounds administrative — how much capital goes to each sub-manager — but it quietly determines whether your diversification is real or theatrical. Get it wrong and you're paying for six strategies while running the risk of two.
The direct answer is this: optimal sleeve sizing is not about equal weighting, conviction ranking, or splitting assets by Sharpe ratio alone. It requires simultaneously solving for strategy capacity, pairwise correlation, and each manager's marginal contribution to total portfolio risk. Ignore any one of those three and your allocation is a guess dressed up in a spreadsheet.
Think of it like adding musicians to a band. A second drummer doesn't double your rhythm section's output — it creates noise. Two trend-following managers trading the same futures markets are the same problem. Their correlation compresses diversification benefit dramatically, and sizing them as if they were independent overstates how much risk reduction you're actually buying.
Capacity constraints add another layer of honest brutality. A small-cap equity manager running a high-alpha strategy might genuinely degrade their own returns past $200 million AUM — market impact erodes edge as position sizes grow relative to daily liquidity. Allocating $500 million to that sleeve doesn't generate 2.5x the alpha. It destroys it. Institutional allocators using the framework described in the risk parity literature often cap sleeves at a fraction of a strategy's estimated capacity ceiling, then redistribute the remainder toward lower-correlation alternatives. Marginal risk contribution — the change in total portfolio volatility for each additional dollar allocated to a sleeve — becomes the governing metric. You're essentially solving an optimisation where the objective isn't maximum expected return but maximum risk-adjusted diversification per dollar deployed. This connects directly to concepts explored in modern portfolio theory, though practitioners typically apply more robust covariance estimation methods than classical mean-variance to avoid the garbage-in problem. Dynamic rebalancing rules matter too — as strategies drift or correlations shift regimes, sleeve weights that were optimal twelve months ago can quietly become concentrated bets, a nuance detailed extensively in research on portfolio management best practice.
The practical takeaway you can use today: pull your current sleeve weights and run a simple correlation matrix across manager returns. If two sleeves share correlation above 0.65, you likely don't have two strategies — you have one strategy with higher fees. Start there.
Sleeve sizing done properly is risk budgeting with manners — every manager gets exactly the influence on total portfolio risk they've earned, no more.
This content is for educational purposes only and does not constitute financial product advice. Past performance is not indicative of future results. Profit Logic Ltd (ACN 688 669 936) accepts no responsibility for errors or omissions in this content or anywhere on this website. Always seek advice from a licensed financial adviser before making investment decisions.