What this chart shows
A standard projection uses a single fixed growth rate and produces one line. Real investment returns are not fixed — they vary significantly year to year, and the order in which those returns occur can dramatically affect how long a retirement fund lasts.
The Monte Carlo fan chart addresses this by running thousands of simulations, each with a different randomly generated sequence of returns, and plotting the distribution of outcomes. The result is a fan shape that widens over time, showing where outcomes converge in early years and diverge in later years.
This chart is available on premium scenarios.
FutureClear generates random annual returns using a log-normal distribution anchored to the growth rate assumption. This distribution is widely used in financial modelling because it produces only positive values in any given year, reflects the skewed nature of equity returns, and calibrates well to historical market return distributions over long periods.
For each simulation run, the engine generates a random return sequence for the full projection period and runs the complete year-by-year projection using those returns. After all runs complete, the results are ranked by outcome at each point in time and aggregated into percentile bands.
How to read it
The horizontal axis is time (years or age). The vertical axis is total asset value. The coloured bands represent the following percentiles:
- p10 (inner lower bound): 90% of simulated runs produced a better outcome than this at that point in time. This represents a poor return environment — consistently below-average returns.
- p25: 75% of runs produced a better outcome. Still below average, but less severe than p10.
- p50 (median — the middle line): Half of runs were better, half were worse. This is the median outcome, not the expected value.
- p75: 75% of runs produced a worse outcome. An above-average return environment.
- p90 (outer upper bound): 90% of runs produced a worse outcome. A very favourable return environment.
The width of the fan at any point in time represents the spread of uncertainty. The fan typically widens over longer time horizons because small annual differences in returns compound into large differences over many years.
What percentile bands mean
Percentile bands describe the distribution of simulated outcomes, not the likelihood of a specific event.
The p10 band does not mean "there is a 10% probability of running out of money." It means "10% of simulated return paths produced a worse result than this line at this age."
A more useful way to read the chart:
- If the p10 band stays above zero throughout the projection: The scenario produces positive fund values even in 90% of poor-return simulated paths. This indicates a high degree of robustness to return variability.
- If the p25 or median band reaches zero before the projection end: Shortfalls occur in average and below-average return environments. The scenario may be sensitive to normal market variability.
- If only the p10 band dips below zero: Only the poorest 10% of simulated paths produce a shortfall.
Sequence of returns risk
One of the most important things the fan chart captures is sequence of returns risk — the risk that poor investment returns occur in the early years of retirement, when the portfolio is at its largest and most vulnerable.
A scenario where returns are poor in the first five years of drawdown depletes the portfolio substantially faster than a scenario where the same average returns are distributed evenly across 30 years. This holds true even if total returns over the period are identical.
The lower percentile bands (p10, p25) typically represent scenarios where early retirement returns are poor. If these bands decline steeply in the first decade of retirement, the projection has meaningful sequence-of-returns sensitivity.
What it does not tell you
Monte Carlo results are based on statistical assumptions about future returns calibrated to historical patterns. They do not account for correlation between asset classes or structural economic changes. The model does not predict actual future markets.
The simulation uses a log-normal distribution to generate random returns. This is the industry standard for consumer retirement tools and is well-calibrated for central estimates — the p25 to p75 range matches historical return data closely.
However, log-normal returns understate the likelihood of extreme outcomes at both tails. Observed equity markets show heavier tails than the model predicts — extreme events beyond two to three standard deviations occur more frequently in historical data than a log-normal distribution implies. This is a known property documented by the Institute and Faculty of Actuaries (IFoA).
In practice, this means the p10 and p90 bands may not fully capture how severe or how favourable the most extreme outcomes could be. The central range of the fan chart (p25 to p75) remains reliable. The outer bands indicate the direction and relative magnitude of extreme outcomes rather than precise boundaries.
Professional adviser tools use more complex stochastic models — regime-switching volatility, ARCH-type processes, historical bootstrapping — that better capture tail behaviour. These models are designed for regulated financial advice contexts.
The fan chart is a diagnostic and stress-testing tool. It shows where a projection is most sensitive to return variability. It is not a guarantee of any outcome.