Allocation Discipline: How I Think About Portfolio Construction
March 7, 2026
Most people think portfolio construction is about picking the right assets. It's not. It's about building a structure that survives being wrong, because you will be wrong, frequently, about individual positions.
The question isn't "what should I own?" It's "how much of each thing should I own, and what happens when my assumptions break?"
Start With Risk, Not Return
The first mistake in portfolio construction is starting with return targets. "I want 12% annualized" sounds reasonable until you realize the portfolio required to deliver that return has drawdown characteristics that would make most people sell at the worst possible moment.
I start with risk tolerance. Not the theoretical kind you mark on a questionnaire, but the actual, honest answer to: "How much can this portfolio drop before I do something irrational?" That number, the behavioral breaking point, is the real constraint.
Correlation Isn't Constant
Diversification works until it doesn't. Assets that appear uncorrelated during normal markets tend to correlate during crises, precisely when you need diversification most. This is one of the most dangerous assumptions in traditional portfolio theory.
I account for this by stress-testing portfolios under crisis correlations, not normal-market correlations. If a portfolio only works when markets behave normally, it's not actually diversified. It's diversified-looking.
The Role of Uncorrelated Assets
This is where precious metals, managed futures, and other alternative allocations earn their place. Not because they deliver the highest returns. They often don't. But because they behave differently during the periods when your equity portfolio is under maximum stress.
A 5-10% allocation to gold doesn't meaningfully improve your long-term returns. But it materially reduces your worst-case drawdowns. And reducing drawdowns is how you stay invested long enough to capture long-term returns in the first place.
Rebalancing Is the Discipline
The hardest part of portfolio construction isn't the initial allocation. It's maintaining it. Markets push your allocation off target constantly. Your equity allocation grows during bull markets (exactly when you should be trimming) and shrinks during bear markets (exactly when you should be adding).
Systematic rebalancing forces you to sell what's gone up and buy what's gone down. It's mechanically simple and emotionally brutal. But it's the closest thing to a free lunch in investing: a process that improves long-term returns while reducing risk, available to anyone with the discipline to follow through.
My Framework
I keep it simple. Define the risk budget. Allocate across truly different risk sources, not just different tickers that all move together. Include assets that earn their place during crises, not just during calm markets. Rebalance systematically, not emotionally. And review the whole structure quarterly to make sure the assumptions still hold.
That's it. No secret formula, no complex optimization. Just discipline applied consistently over time. In my experience, that beats cleverness every day of the week.