How Concentrated Should Your Portfolio Be? Position Sizing

The Disagreement Nobody Resolves

Index investors hold thousands of securities. Some concentrated value investors hold fewer than ten. Both have produced good outcomes, and the disagreement persists because position sizing is not primarily a mathematical question — it is a question about what you know and what you can tolerate.

Diversification Does Exactly One Thing

Precision matters here, because diversification is usually defended with claims it cannot support. It does not raise expected returns; removing idiosyncratic risk carries no return premium.

What it does is narrow the distribution of outcomes. A concentrated portfolio has a much wider spread — a greater chance of beating the market substantially, and a greater chance of trailing it badly. A diversified portfolio collapses toward the market's own result.

That is the trade. Not higher returns for lower diversification, but wider dispersion in both directions.

Buffett's Position

Buffett has called diversification protection against ignorance, and has said it makes little sense for those who know what they are doing. He and Munger ran concentrated portfolios for decades, at times with the majority of Berkshire's equity portfolio in a handful of names.

The qualifier is doing all the work. Protection against ignorance implies you know something specific. Most concentrated portfolios are not informed but unexamined: ten positions chosen without comparison against alternatives is not conviction, it is neglect.

What Actually Determines Your Number

Knowledge per position. Size should scale with how well you understand the business. Twenty positions you cannot explain is worse than eight you can.

Correlation. Holdings that fail for the same reason are effectively one position. Six lenders with the same loan book is not six positions.

Drawdown tolerance. Specifically, the drawdown you can hold without selling. A portfolio that forces action at the bottom converts volatility into realised loss. This constraint binds earlier than most people expect, because tolerance measured in calm markets misrepresents tolerance measured in a real decline.

Time horizon. Concentration requires years. Selling a concentrated position into weakness removes the mechanism that was supposed to make it work.

What else you own. Concentration in a portfolio representing all your investable assets is a different decision from concentration in one sleeve.

Practical Starting Points

Rather than a target number, size on conviction and cap the downside:

Core positions — the largest allocations, reserved for businesses whose economics you can describe and whose next ten years you can argue for

The rest — smaller, sized so that any single thesis being wrong is survivable

A cap per position — most concentrated portfolios cap individual holdings well below the level where one bad outcome becomes unrecoverable

Never exceed what you can actually track. Whatever your count, it should not exceed the number of businesses you can follow through a full earnings cycle.

The Two Failure Modes

Too concentrated without knowledge. The consequences become permanent rather than temporary. A position sized beyond what you can absorb converts a recoverable mistake into an unrecoverable one.

Too diversified without purpose. Over-diversification produces a portfolio that cannot beat anything, held by someone who has stopped forming views. Any position too small to matter is a position you have no reason to hold.

The Usable Answer

Concentrate where you have genuine knowledge and no forced selling; diversify where you have neither. If you cannot articulate why one position is larger than another, that is not a portfolio, it is an accident — and it should be sized down until it is no longer one.

---

This article references Warren Buffett's published remarks describing diversification as protection against ignorance, and Berkshire Hathaway's historically concentrated equity portfolio. It is educational material, not investment advice.