How much should I put in any one thing?Start here

There is no correct number, but there is a correct question: what happens to the whole if this one goes to zero?

This page cannot tell you a number, and anyone who gives you one without knowing anything about you is selling something. What it can do is show the arithmetic that any answer has to survive.

What happens if this one goes to zero?

That is the question, and it is worth answering literally rather than dismissing.

At 2% of the portfolio, a total loss costs you 2%. Annoying, recoverable, over within a normal year's movement.

At 20%, a total loss costs you 20%, and the remaining 80% now has to rise by 25% just to get back to where you started. At 50%, the rest has to double.

That asymmetry is the whole reason position size matters more than being right. The arithmetic of drawdowns is the page that goes through it properly, and it is short.

Why is size the decision that matters most?

Because everything else assumes you were right. Sizing is the only decision that is still doing useful work when you were wrong — which, across a lifetime of investing, is a large fraction of the time.

A portfolio of positions small enough that any one can fail without consequence can absorb being wrong repeatedly and still compound. A portfolio with one dominant position is a bet on not being wrong about that specific thing. People are far worse at knowing which of their views is the reliable one than they expect.

Does it matter that things move together?

Very much. Ten positions are not ten independent risks if they are ten technology shares, or ten banks, or ten funds that all hold the same index. In a bad week they behave like one position wearing ten hats.

The honest count is not how many holdings you have but how many distinct things can go wrong. Diversification explains why correlations rise exactly when you need them not to.

Does leverage change the answer?

It changes it completely. A leveraged position can cost you more than you put into it, and it can be closed out before the story finishes. So the question stops being "what if it goes to zero" and becomes "what if it falls 30% at the wrong moment" — a much lower bar, hit far more often.

Any position with borrowing in it, explicit or built into the product, has to be sized against the move that triggers a margin call rather than against total loss. Leverage covers the mechanism.

What about the money that is not invested at all?

It is part of the answer, and it is usually the part that decides whether a plan survives. The most common way people take a permanent loss is being forced to sell during a fall because they needed the money — not because they chose to.

Money you might need within a few years does not belong in something that can halve. Knowing where that line falls for you is worth more than any refinement of the split above it.

Is there a sensible rule of thumb?

Not a universal one, but a useful test. Write down the largest position as a percentage. Imagine it at zero tomorrow. If your reaction is "that would change my plans", it is too large — whatever the analysis said.

That is not a formula, and it is deliberately not one. The position sizing playbook and the portfolio practice page take it further.