What a projection can and can't tell you

Every calculator draws a smooth line. Here's how to read one without being misled by it.

Any investment projection, including the one on this site, asks you for a growth rate and then draws a smooth curve. Reality is neither smooth nor knowable in advance.

That doesn't make projections useless. It makes them useful for a specific purpose, and misleading if used for a different one.

What growth rate to use

There's no correct answer, but there are reasonable ranges and a common mistake.

The mistake is using a headline market return. Long-run equity returns look impressive, and then charges come off, and you don't hold 100% equities, and inflation erodes what's left.

Work in real terms. Use a rate net of inflation, and read the output as today's money. A projection showing £800,000 in thirty years means nothing if you can't picture what £800,000 buys in 2056. A projection in today's money is directly comparable with what things cost now.

Deduct your charges. If your funds and platform cost you 0.8% a year, your growth assumption should be 0.8% lower than whatever you think the underlying investments will do. See What you actually pay.

Match the rate to the mix. A cautious portfolio with a large bond holding shouldn't be modelled at the same rate as an all-equity one. Cash should be modelled below inflation in real terms, because that's what it has usually done.

Be conservative. A projection that turns out pessimistic leaves you with more than you planned. One that turns out optimistic leaves you short at the point when you have no time to fix it.

Run it more than once

The single most useful habit with any projection is to run it three times: a rate you'd consider pessimistic, one you'd consider reasonable, and one you'd consider good.

What you learn is not the answer. It's how sensitive the answer is.

If your plan works at the pessimistic rate, you have genuine margin. If it only works at the optimistic one, you don't have a plan, you have a hope. That distinction is the most valuable output of any calculator and it's invisible if you only run it once.

Small changes in assumptions produce large changes over long periods, because they compound. A one percentage point difference over thirty years is not a small difference — which is exactly why a single projection shouldn't be trusted as a forecast.

The smooth line is the biggest lie

Markets don't deliver 5% a year. They deliver 22%, then minus 14%, then 3%, then 11%, then minus 6%, and the average happens to be 5%.

Over a long accumulation period the destination is often close to what the smooth line suggested. The journey is not, and the journey is what makes people abandon the plan. See Time in the market.

One reassuring thing about falls while you're saving

This is worth knowing because it's the opposite of what most people assume.

If you're contributing monthly, a market fall part-way through your saving years isn't the disaster it feels like. Your regular contributions buy more units at lower prices, and when the recovery comes those cheap units participate in it. Someone in their thirties should arguably want a poor decade followed by a good one rather than the reverse.

The reverse is true once you're drawing income. Then a fall means selling units to fund withdrawals, permanently, and the timing matters enormously. That asymmetry is covered in How much can you take without running out?.

So the same market fall is roughly good news at 35 and genuinely bad news at 65. Projections rarely make that distinction, and it's one of the more important things to hold in your head.

What projections are actually good for

Comparing options. Contributing £300 a month against £400. Retiring at 60 against 63. Paying 0.4% in charges against 1.1%. The relative answer is far more reliable than the absolute one, because both scenarios share the same assumptions.

Finding out whether you're roughly on track, in the sense of whether the gap is small, large or enormous.

Showing the cost of waiting. The difference between starting now and starting in five years is usually stark, and seeing it is more persuasive than being told it.

Testing your sensitivity to assumptions, as above.

What they're not good for

Predicting a number. Nobody knows what markets will do, and a figure carried to the nearest pound thirty years out is false precision.

Being run once and filed. Circumstances change, contributions change, and rules change. A projection is worth redoing annually, and the update is the useful part.

Deciding anything irreversible on their own, particularly around retirement, where the sums are large and the mistakes permanent.

How to use ours

The Investment Growth Calculator lets you model several pots at once with different wrappers, contributions and rates, which is closer to how most people's money actually sits than a single-pot projection.

Run it in real terms. Run it three times with different rates. Then look at the spread rather than the number.

This article is for general education only and isn't personal advice. Projections are illustrations, not forecasts, and past returns are not a guide to future ones.