Choose the allocation you can hold, not the one that optimises

The theory of choosing a starting allocation is well covered elsewhere. This is about the version you meet in practice.

The gap between the model and the holder

Portfolio models describe what an investment would have returned to someone who held it without interruption. That is a strong assumption, and studies of investor behaviour have generally found that real returns lag fund returns because of poorly timed entries and exits. An allocation is therefore only as good as your ability to keep it during the year you most want to leave.

Any comparison of two allocations that ignores this is comparing hypothetical holders, not people.

Optimisation is fragile in ways that are easy to miss

Optimisers are highly sensitive to their inputs, and the inputs are estimates of future returns that nobody has. Small changes to assumptions can produce very different "optimal" portfolios, which tells you how much confidence the output deserves. This is not an argument against thinking carefully; it is an argument against treating precise outputs as precise knowledge.

A broadly sensible allocation held for decades sits well inside the margin of error of any optimisation.

Test tolerance in money, not percentages

Ask what a forty per cent fall would look like as an absolute amount of your own money, then say the number out loud. People routinely accept a percentage they would refuse in currency, because the percentage is abstract.

If the number makes you uncomfortable now, in calm conditions, it will make you sell later. That discomfort is information about your allocation rather than a character flaw to be overcome.

Cautious and kept beats aggressive and abandoned

A portfolio with a lower equity share produces smaller declines, and smaller declines are more survivable. The cost is expected return, and that cost is real and worth stating plainly rather than glossing. It is nonetheless a smaller cost than selling at the bottom and returning after a recovery has already happened.

Choosing the allocation you would still be holding in the worst year you can imagine is a defensible way to decide.

You can move up later

Risk appetite tends to be discovered through experience, and someone who has held a real decline knows something they could not know before. Increasing equity exposure after living through a fall is a well-grounded adjustment made from evidence. Decreasing it during a fall is the same adjustment made at the worst possible price.

Put simply, building in the expectation of upward revision removes the pressure to get it exactly right at the start.

What nobody can tell you

There is no allocation that is correct in general, because the answer depends on your horizon, your other resources and your own behaviour. Rules of thumb linking age to equity share are convenient rather than evidenced, and they ignore capacity for loss entirely. Anything more specific than the principles here is a matter for regulated advice in your own jurisdiction.

In practice, what is general is the criterion: pick the one you would still hold on the worst day.

The takeaway

The best portfolio on paper is worthless if you sell it. Pick the one you would keep.

The version you keep doing is the version that works.