Open any pension calculator online and it will ask you one question: "What annual return do you expect?" Type in 7%, and it will show you a beautiful, straight line climbing to a satisfying number. Retire at 65 with £800,000. Done.
Except markets have never returned exactly 7% in any given year. Not once in recorded history. They have returned +32%, they have returned -38%, and everything in between — in an order that no one can predict. That single projected line is a fiction. A comforting one, but a fiction.
Monte Carlo simulation is what happens when you stop pretending the future is knowable and instead model the full range of futures that are statistically plausible — thousands of them simultaneously — including the ones where markets crash just as you retire. The difference between these two approaches is not cosmetic. It can be the difference between a retirement plan that holds and one that runs out of money 10 years early.
The method was born not in finance, but in physics — and in a card game played during a period of illness.
It was 1946. Stanislaw Ulam, a brilliant Polish-American mathematician working at Los Alamos, was recovering from encephalitis. Playing solitaire during his convalescence, he found himself wondering: what is the probability of winning a game of solitaire from a random shuffle? The theoretical calculation — accounting for every possible card arrangement — was computationally impossible. But Ulam realised there was a simpler approach: just play thousands of games and count the wins. The proportion would converge on the true probability.
His colleague John von Neumann immediately saw the potential for the far harder problems they were working on — neutron diffusion in nuclear weapons. Nicholas Metropolis coined the name "Monte Carlo" after the Monaco casino where Ulam's uncle famously gambled, since both the casino and the method relied fundamentally on the mathematics of randomness.
Today the same logic is applied to retirement planning. Instead of computing the one unknowable future, you run thousands of plausible ones and examine the distribution of outcomes.
Traditional retirement calculators assume markets deliver a steady, predictable return every single year. Monte Carlo simulation does the opposite — it treats each year's return as a random draw from a statistical distribution calibrated to how markets have actually behaved historically.
Run 5,000 simulations and you get 5,000 different possible retirement journeys. Some are good: markets cooperate, your pot grows strongly, you retire comfortably. Some are bad: early bear markets devastate your savings right when you start withdrawing. Most fall somewhere in between. The spread of all 5,000 outcomes is your honest picture of retirement risk.
That 74% success probability is the most important number. It tells you: in 74 out of 100 statistically plausible market histories, your money lasts the full 30 years. In 26 out of 100, it does not. That is something a compound interest calculator cannot tell you at all — it just shows you the median line and pretends the other 4,999 paths do not exist.
Here is the counterintuitive insight that makes Monte Carlo indispensable for retirement planning: the order in which market returns arrive matters just as much as the average return itself — especially when you are withdrawing money.
Consider two investors who both retire with £500,000, withdraw £25,000 per year, and both experience an average return of 4% over 30 years. The only difference: the sequence.
Investor A retires into a bull market. Returns of +18%, +22%, +15% in her first three years give her portfolio a large cushion. Even when poor years arrive later, the pot is big enough to absorb them. After 30 years she has approximately £800,000 remaining.
Investor B retires into a bear market. Returns of -25%, -18%, -12% in his first three years, combined with his withdrawals, devastate the portfolio. Each withdrawal takes a larger share of a shrinking pot. Even when good returns arrive later, there is too little capital left to compound meaningfully. Investor B runs out of money at year 22 — eight years early.
Same average return. Same starting amount. Same withdrawals. Completely different endings.
Most Monte Carlo tools on the internet are basic: enter a lump sum, pick a return, pick a horizon. They are better than compound interest calculators, but they are still missing most of what matters for real retirement planning.
Fintiq has built two lifecycle planners — one for UK investors, one for US investors — that model the full complexity of how retirement actually works.
The UK planner runs 5,000 Monte Carlo scenarios using a log-normal return model with Box-Muller transforms across three distinct phases of your financial life:
The outputs include a probability-of-success figure, a fan chart of all 5,000 paths, and a sensitivity table showing how small changes (retiring 2 years later, contributing £100 more per month) affect your probability of success.
The US planner uses the same Monte Carlo engine, adapted for the American retirement system:
The same 5,000-scenario engine runs across your full lifecycle, surfacing the probability that your specific combination of accounts, timing choices, and withdrawal needs holds up across all statistically plausible market histories.
Once you run the simulation, the key outputs are:
Probability of success — the percentage of 5,000 scenarios where your money lasts your full retirement horizon. Think of it as: "In what fraction of plausible futures does my plan work?" Target 80%+. Below 70% means the plan needs adjustment.
The fan chart — a visualisation of all 5,000 paths. The wide spread is not a bug — it is the honest reality of market uncertainty. The narrow band around the median is where most people end up. The thin tails above and below show the lucky and unlucky scenarios.
The sensitivity table (Pro unlock) — shows how your probability of success changes if you retire 2 years earlier or later, contribute more or less, or assume a different return. This is where the planner earns its keep: turning abstract uncertainty into concrete, actionable trade-offs.
| Success Probability | Interpretation | Action |
|---|---|---|
| 90%+ | Very robust — more resilient than you need | Consider retiring earlier or spending more |
| 80–90% | Strong plan — comfortable resilience | Monitor annually, no urgent changes |
| 70–80% | Acceptable — some vulnerability | Consider modest adjustments |
| 60–70% | Fragile — bad sequence could cause real problems | Increase contributions or delay retirement |
| Below 60% | High risk — needs significant revision | Material changes to plan required |
Most professional financial planners target 80–90% for clients. This does not mean the plan will definitely succeed — it means it survives the vast majority of statistically plausible market scenarios, including ones considerably worse than any historical period we have actually lived through.
Monte Carlo simulation is far superior to single-point forecasting, but it is not a crystal ball. Informed users should understand what it cannot do:
Use it as a stress-testing and decision-support tool, not as a prediction. The question it answers brilliantly is: "Is my plan robust enough to survive the kinds of bad luck that have occurred in the past?" That is a much more useful question than "What will my pot be worth in 2055?"
One of the most valuable things Monte Carlo simulation reveals is how sensitive your retirement outcome is to small, controllable decisions. Run the numbers and you will typically find:
None of these insights come from a compound interest calculator. They only emerge when you model uncertainty honestly.
Both planners are free to run. The full sensitivity analysis and PDF report unlock for £1.99 (UK) or $2.99 (US) — one-off, no subscription.
🇬🇧 UK Retirement Planner — Free →Models accumulation · bridge period · decumulation · State Pension / Social Security · ISA / SIPP / 401(k) / IRA