Position sizing: turning conviction into a number

Ask most people how they decide what to invest in and they’ll talk for an hour about which stock. Ask them how they decide how much to put in and the answer gets vague fast — “it felt right,” “a few thousand,” “whatever I had spare.” That’s backwards. Picking the right company is hard and uncertain. Sizing the position is arithmetic, and it quietly controls more of your result than the pick itself.

Here’s the uncomfortable math: a brilliant thesis in a 2% position and a mediocre thesis in a 20% position — the second one moves your portfolio far more. Your outcomes are dominated by your biggest positions, whether you sized them deliberately or by accident. So sizing deserves at least as much thought as selection, and it almost never gets it.

Why does position size matter as much as stock selection?

Because your largest positions dominate your result, and a large loss is hard to recover from. Two forces make position size decisive:

Concentration cuts both ways. A position big enough to make you rich when you’re right is big enough to hurt badly when you’re wrong. The size is the risk. You can be correct about a company and still damage your portfolio by holding too much of it into a drawdown you couldn’t sit through.

Ruin is permanent; gains are not. Lose 50% and you need a 100% gain just to get back to even. This asymmetry is why survival comes first. A sizing rule’s real job isn’t to maximize the good case — it’s to make sure no single position can take you out of the game. You can’t compound if you’ve blown up.

How do you calculate position size?

With a rule: a fixed cap per name, the loss you would accept worked back from your exit, or a fraction of Kelly. There’s no single correct number, but there are disciplined frameworks that beat gut feel.

Fixed fractional. Cap every position at some percentage of the portfolio — say no single name above 5%. Crude, but it enforces diversification automatically and it’s almost impossible to get catastrophically wrong. For a lot of people this alone is a large upgrade over vibes.

Risk-based sizing. Instead of sizing the position, size the loss you’d accept. Decide you’ll risk 1% of your portfolio on an idea, then work backwards from where you’d admit you were wrong (your stop). If your exit is 20% below entry, then risking 1% of the portfolio means the position is 5% of it. This ties size to how wrong you can be, which is the variable that actually matters.

The Kelly criterion. Kelly is the formula that maximizes the long-run growth rate given your edge and odds; John Kelly derived it at Bell Labs in “A New Interpretation of Information Rate” (Bell System Technical Journal, 1956). Its lesson is more useful than its precise output: bet more when your edge is larger, less when it’s smaller, and never bet the amount that risks ruin. In practice, many users deliberately bet a fraction of what full Kelly suggests — “half-Kelly” or less — because full Kelly is brutally volatile and assumes you know your edge exactly, which you never do. Kelly is a compass, not a GPS.

What are the most common position-sizing mistakes?

Sizing by excitement, ignoring correlation and letting winners drift unnoticed.

• Sizing by excitement. The most exciting idea gets the biggest check. But excitement and edge are not the same thing — often they’re inversely related.
• Ignoring correlation. Five 5% positions that all move together aren’t diversified; they’re one 25% bet wearing five name tags. Size to your true exposure, not your ticker count.
• Letting winners silently become the whole portfolio. A position that runs can quietly grow into an outsized share of your book. That might be fine — but it should be a decision you noticed, not a drift you didn’t.
• Never writing the number down. If you can’t state your position size as a percentage and the reason for it, you didn’t size — you guessed.

What should you decide before buying a position?

Its size as a percentage, why, and the most it can cost you. Before you buy, finish two sentences: “This position is ___% of my portfolio, because ___.” and “If I’m wrong, the most it costs me is ___.” If you can answer both, you’ve thought about size like it matters. If you can’t, no amount of being right about the company will save a position you accidentally made too big — or too small to matter.

Tradune’s position-size calculator turns this into a number in a few seconds — give it your risk tolerance and your exit, and it works out the size that keeps a single wrong idea from doing outsized damage. Inside the app, rebalancing suggestions flag when a position has drifted away from its target weight, so size stays a decision rather than an accident. As always, this is educational — a framework for thinking about risk, not advice on any specific position.