How much should you actually put in? Questions, not a number

"Only invest what you can afford to lose" is repeated so often that it has stopped meaning anything. Nobody can afford to lose money — that is what money is. The phrase is trying to say something real, and the real version is: an amount whose complete loss would be annoying rather than consequential.

Six questions turn that into a figure. Answer them in order, honestly, and stop when one of them produces a clear answer — several of them are designed to end the exercise early.

The four questions that come before any amount

These are gates rather than inputs. If any of them lands badly, the answer to "how much" is not a number.

1. Do you have expensive debt? Credit-card debt or similar high-interest borrowing has a guaranteed cost. Paying it down is a certain return at that rate; buying a volatile asset while carrying it is borrowing at a known high rate to bet on an unknown one. This is the one place where the arithmetic is not really arguable.
2. Do you have money set aside for an emergency? If an unexpected bill would force you to sell, you have not bought an asset — you have bought a countdown. Forced selling at the worst moment is how paper losses become real ones.
3. Is this money needed within five years? A deposit for a house, tuition, a wedding, a planned move. Money with a date attached should not be somewhere that can fall by half and stay there for two years. It has done exactly that more than once.
4. Would you have to explain this to anyone? A partner, a family member, a joint account. Not a moral point — a practical one. Money that is not entirely yours to risk has a different answer, and financial secrecy between people who share finances damages more than the amount involved.

If all four are clear, the remaining two questions produce an amount.

The two questions that produce the number

What could vanish entirely without changing anything in your life? Not "without hurting" — hurting is fine. Without changing anything: not your rent, your food, your ability to handle a car repair, your sleep. For many people that number is smaller than they first guess, and writing it down before looking at any price is the point.

What would you do if it fell by seventy per cent next month? This is the real test, because it is a question about behaviour rather than arithmetic. Bitcoin has fallen more than seventy per cent from a peak on several occasions. If your honest answer is that you would panic, borrow to average down, or check the price hourly, then the amount is too large — regardless of what the first question said. A position that changes how you behave is too big, even if you could afford to lose it.

Why a percentage is the wrong unit for most people

The advice you will meet most often is a percentage of your portfolio — one per cent, five per cent, some range. It is not wrong so much as addressed to somebody else.

A percentage assumes a portfolio exists to take a percentage of. For a reader who has an emergency fund, no expensive debt, a pension arrangement and some investments, that assumption holds and the percentage does real work: it stops a single volatile holding from dominating everything else.

For a reader who has some savings, some debt, and no portfolio in the formal sense, the percentage is answering a question they have not reached. Five per cent of what? If the honest answer is "five per cent of my savings, which are also my emergency fund", the number has quietly recommended something the previous section ruled out.

An absolute amount is the better unit here, for a plain reason: you cannot lose a percentage. You lose an amount, and the amount is what has to be survivable. Deciding "no more than this figure, total, ever" is both easier to hold to and easier to check against your actual circumstances.

Buying it all at once, or spreading it out

Once the amount is settled, this is the next question, and the honest answer is that neither approach is superior in general.

Spreading purchases over time — a fixed sum at regular intervals — removes the risk of putting everything in on one unlucky day and removes the need to have an opinion about timing. Its cost is real: in a steadily rising market you buy at higher average prices than you would have by going in at the start.

Going in at once puts the whole amount to work immediately. That is better if the price rises from here and worse if it does not, and nobody knows which.

The reason to prefer spreading is not mathematical, it is behavioural. A plan of fixed amounts at fixed intervals is a plan you can follow while frightened, and following a mediocre plan beats abandoning a good one. Most people do not fail because they chose the wrong method; they fail because they stopped following whichever one they chose, at the worst moment, for reasons that felt compelling at the time.

One caveat worth stating: spreading purchases is not a hedge. If the price falls steadily for two years, buying every month means buying all the way down. It reduces timing risk, not market risk, and market risk is the one with no floor under it.

Two mistakes that come after the number

Setting the amount is only half of it. The two failures that follow are predictable enough to plan for.

Renegotiating during a rise. The amount you set calmly gets revised upward when the price is climbing and everyone is talking about it. That is the moment the original number was designed to protect you from, and it is exactly when it feels most conservative. The whole value of deciding in advance is refusing to reopen it.

Averaging down past the plan. Adding on the way down to lower your average price is defensible as a strategy and dangerous as a reaction. The difference is whether you wrote it down before the fall. If you are adding because it is cheaper than yesterday rather than because you planned to, you are increasing your exposure at the moment your judgement is worst — and the ways beginners lose everything are mostly this, with leverage on top.

What a defensible answer sounds like

Not "about five per cent". Something more like: I have no expensive debt and three months of expenses set aside. This money has no job in the next five years. If it went to zero I would be irritated for a week and nothing about my life would change. I would not add to it if it fell by half. My partner knows the number.

That is defensible because every clause is checkable and none of it depends on being right about the price. If you cannot construct that sentence yet, the honest conclusion may be that the answer for now is zero — and that is a legitimate result rather than a failure. Some people should not do this at all, and knowing which group you are in is worth more than any allocation rule.