FinisdomFinisdom
Log inGet access
Advanced track
AdvancedLesson 26

When the optimizer fools you

When the optimizer fools you

Math that finds the “best” mix can backfire, because it trusts shaky estimates too much.

In short

An optimizer picks the mix with the best past trade-off. But it leans hard on guesses about future returns that are really just noise — so it often picks extreme, fragile mixes.

A well-known critique called this the “optimization enigma”: the math is elegant, yet it piles money into whatever looked best by luck, then crumbles in the future.[1]

Strikingly, studies found that a plain equal split across holdings often matched or beat fancy optimized mixes once real-world estimate errors were included.[2]

Like tailoring a suit to a photo taken in a funhouse mirror. The fit looks perfect on the distorted image — and terrible in real life.

The fixes are humility: add sensible limits, lean on steadier inputs, spread bets, and don’t chase the last decimal of “optimal.”

Where these numbers come from

Finisdom’s Allocation Explorer shows the frontier but also offers simple, sturdy mixes like equal-weight and risk-parity — and uses honest walk-forward testing.

Compare sturdy mixes in the ExplorerPart of the Finisdom app — sign in to open it.

Check your understanding

Why can an optimizer pick a fragile mix?

Sources & further reading

  1. 1.Richard Michaud (1989) The Markowitz Optimization Enigma: Is “Optimized” Optimal? — Financial Analysts Journal — Showed optimizers over-trust noisy inputs and pick fragile extremes.
  2. 2.DeMiguel, Garlappi & Uppal (2009) Optimal Versus Naive Diversification — Review of Financial Studies — Found a simple equal split often beat optimized portfolios out of sample.

Related

The efficient frontierRisk parity: balance the risk, not the cashHow backtests lie

Tripped up by a word? Look it up in the glossary.

Learning only — not investment advice.