Imagine two people — both careful savers, both sensible investors, both retiring at 65 with identical portfolios. They experience exactly the same annual returns over the next 20 years. The same numbers, just in a different order.
One of them runs out of money at 79. The other is still financially comfortable at 85.
This is not a hypothetical designed to alarm you. It is a mathematically demonstrable consequence of something called sequence of returns risk — and it is the single most underappreciated threat in retirement planning.
See how sequence risk plays out across 5,000 simulated versions of your retirement.
Start your free 30-day trialWhat sequence of returns risk actually means
During the accumulation phase of your financial life — the decades when you are working and saving — the order in which your investments return gains and losses barely matters. What matters is the average return over time. A bad year followed by a good year is roughly equivalent to a good year followed by a bad year, because you are not drawing money out. You are adding to the pot.
Retirement changes this completely.
The moment you start drawing down — taking money out of your portfolio to live on — the sequence in which returns arrive becomes critically important. A run of poor returns in the early years of retirement forces you to sell more units of your investments to generate the same income. Those sold units are then unavailable to benefit from the recovery when it eventually comes.
This is sequence of returns risk. It is not about average returns being lower than expected. It is about when the bad returns arrive.
A worked example — the same returns, opposite order
Consider two retirees — call them Alex and Sam. Both retire at 65 with £500,000. Both withdraw £25,000 per year. Both experience the same ten annual returns, just in reverse order.
Alex gets the good years first:
Sam gets exactly the same returns in reverse — bad years first, good years later.
Same starting pot. Same withdrawal rate. Same ten returns. A £194,000 difference after just ten years — purely because of the order.
Projected forward, Sam's portfolio is exhausted well before 80. Alex remains solvent into their mid-80s. A standard retirement calculator — which uses an average return assumption — would have shown both of them the same projection.
Why the early years matter most
The reason early retirement is so dangerous is compounding — working in reverse.
When you are accumulating, compounding works in your favour. Returns on returns build wealth over time. But when you are withdrawing, the maths flips. Poor early returns combined with ongoing withdrawals create a downward spiral that becomes increasingly difficult to escape.
Research consistently shows that the first five to ten years of retirement are the highest-risk window. A significant market downturn in this period — before your portfolio has had time to generate meaningful income — can permanently alter your retirement trajectory regardless of what happens afterwards.
This is why simply averaging good and bad scenarios gives a dangerously misleading picture. The average of a retirement that goes well and a retirement that goes badly is not a retirement that goes averagely. It is two completely different futures — and you will live in only one of them.
What makes sequence risk worse
Several factors amplify the damage from a bad sequence:
- High withdrawal rates. The more you draw relative to your portfolio, the more units you must sell in a downturn, and the less remains for the recovery. A withdrawal rate of 5% or above significantly increases vulnerability.
- Single asset class exposure. A portfolio concentrated in one asset class — equities only, for instance — experiences the full force of a market downturn with no offset. Diversification across asset classes with low correlation reduces the depth of drawdowns.
- No cash reserve. Without a cash buffer, every withdrawal during a downturn requires selling investments at depressed prices. A one to two year cash reserve allows you to wait out short-term volatility without crystallising losses.
- Inflexible spending. A fixed withdrawal plan with no capacity to reduce spending temporarily during bad markets accelerates portfolio depletion. Even modest flexibility — reducing withdrawals by 10–15% in down years — meaningfully improves outcomes.
What you can actually do about it
Sequence of returns risk cannot be eliminated — you cannot control when markets fall. But it can be managed. Here are the strategies that evidence supports:
1. Build a cash buffer
Hold one to two years of living expenses in cash or near-cash outside your investment portfolio. When markets fall, draw from the buffer rather than selling investments. When markets recover, replenish. This simple strategy breaks the forced-selling cycle that causes the most damage.
2. Use a flexible withdrawal strategy
Rather than drawing a fixed £X per year regardless of market conditions, build in a mechanism to draw slightly less in down years and slightly more in strong years. Even a 10% reduction in withdrawals during a significant downturn can add years to portfolio longevity.
3. Diversify across asset classes
A portfolio with some allocation to assets that behave differently from equities — bonds, property, alternatives — will typically experience shallower drawdowns. Shallower drawdowns mean less forced selling at the worst times.
4. Understand your personal sequence risk window
The risk is concentrated in your early retirement years. If you are still five or more years from retirement, you have time to build resilience — a cash buffer, a more diversified allocation, a flexible spending plan. If you are already retired, the priority shifts to protecting against further forced selling.
Why standard calculators miss this entirely
A standard retirement calculator assumes a fixed annual return — say 5% — every single year. In this model, sequence risk does not exist. The order of returns is irrelevant because the return is always the same.
This is not a minor simplification. It is a structural blindspot that causes these tools to systematically understate retirement risk for anyone who is drawing down a portfolio.
| Feature | Standard calculator | Monte Carlo (ODOpt) |
|---|---|---|
| Return assumption | Fixed single rate every year | Variable — drawn from realistic distributions |
| Sequence risk | Not modelled | Core of the model |
| Output | One projected balance | Probability distribution across 5,000 scenarios |
| Cash buffer | Not supported | Fully modelled |
| Withdrawal flexibility | Fixed only | Variable strategies supported |
| Downside visibility | None | 10th percentile shown alongside median |
Monte Carlo simulation runs your retirement across thousands of different return sequences — good, bad, mediocre, and everything in between. The result is not one answer but a probability landscape: in what percentage of realistic futures does your plan hold together?
That is a fundamentally different — and more honest — question than "assuming 5% returns every year, will I be okay?"
Run your retirement across 5,000 sequences — including the bad ones.
Start your free 30-day trial| Term | Meaning |
|---|---|
| Sequence of returns risk | The risk that poor investment returns early in retirement will permanently damage a portfolio, even if long-run average returns are acceptable. |
| Drawdown | The act of withdrawing money from an investment portfolio to fund living expenses in retirement. |
| Forced selling | Selling investments at depressed prices because income is needed, regardless of whether conditions are favourable for selling. |
| Cash buffer | A reserve of one to two years of living expenses held in cash, used to avoid selling investments during market downturns. |
| Withdrawal rate | The percentage of a portfolio withdrawn each year. A 4% withdrawal rate on a £500,000 portfolio is £20,000 per year. |
| Monte Carlo simulation | A modelling method that runs a scenario thousands of times with different random inputs — here, different sequences of annual market returns — to produce a probability distribution of outcomes. |