What is a Monte Carlo Simulation?

A Monte Carlo simulation runs your financial plan through thousands of randomized market scenarios to estimate the probability it succeeds, instead of assuming a single fixed rate of return. Rather than telling you what will happen, it tells you how often a plan like yours works across a wide range of possible futures.

The name comes from the casino district in Monaco, a nod to the method’s reliance on randomness. It was developed by physicists working on the Manhattan Project in the 1940s and later spread into finance, where it became a standard way to model uncertainty in investment returns, retirement withdrawals, and long-term savings plans.

How a Monte Carlo Simulation Works

A single Monte Carlo run takes your plan (your starting balance, contributions, withdrawals, time horizon) and applies a random sequence of yearly returns drawn from a realistic distribution. One run might string together a strong first decade followed by a flat stretch. Another might open with a crash. Each run is one plausible lifetime for your portfolio, played out year by year until you either reach the end or run out of money.

The simulation repeats this hundreds or thousands of times, each with a different random path. When it finishes, it counts how many runs left you with money at the end and how many didn’t. That ratio is your probability of success, usually reported as a percentage.

The randomness is the point. Two plans with the same average return can behave very differently depending on the order those returns arrive in, which is why sequence of returns risk matters so much in early retirement. A Monte Carlo simulation captures that ordering effect directly, because every run is a distinct sequence rather than a smoothed average.

Reading a Chance-of-Success Result

Say your plan succeeds in 850 out of 1,000 simulations. That’s an 85% success rate: in 85% of the randomized futures the model tested, your money outlasted you, and in the other 15% it ran short before your plan’s end date.

An 85% result does not mean you have a 15% chance of going broke tomorrow, and it isn’t a prediction that a specific future will occur. It’s a statement about the plan’s resilience across the full spread of scenarios. A plan at 95% has more cushion against bad markets than one at 75%, even if both look identical under a single average-return projection.

The failed runs are often the most useful part. They tend to share a signature, usually a poor sequence of early returns or an inflation spike, and seeing what breaks the plan points you toward the fix: spending a little less, working a year longer, holding a larger cash buffer, or adjusting your withdrawal strategy.

Monte Carlo vs a Straight-Line Projection

A straight-line projection assumes one fixed return every year, say 7%, and draws a single smooth curve into the future. It’s easy to read and useful for a quick estimate, but it hides the risk that matters, because real markets never deliver the same return year after year.

Consider two portfolios that both average 7% over 30 years. One earns a steady 7% annually. The other loses 20% in its first two years, then recovers to reach the same average. If you’re retired and withdrawing from both, the second portfolio can run dry decades earlier, because you’re selling assets into a downturn while the average is still catching up. A straight-line projection shows these two as identical. A Monte Carlo simulation separates them, because it tests the bad sequences alongside the good ones and reports how often each type shows up.

Limitations of Monte Carlo Simulations

A Monte Carlo simulation is only as good as its assumptions. The return and inflation distributions you feed it shape every result, and if those inputs are unrealistic, a confident-looking success rate can be misleading. Small changes to assumed returns or volatility can swing the outcome by several percentage points.

Many implementations also assume returns are independent and normally distributed, which understates the real world’s tendency toward extreme events and multi-year trends. Markets have fatter tails and more persistent momentum than a simple bell curve suggests. Simulations that draw from actual historical return sequences, rather than a synthetic distribution, address part of this by preserving the patterns markets have actually produced.

Finally, a success rate is a probability, not a guarantee. A 90% plan can still land in the unlucky 10%, and no simulation can price in a job loss, a divorce, or a policy change that reshapes the assumptions. The value is in comparing plans and understanding your range of outcomes, not in treating any single number as certain.

You can run Chance of Success simulations in ProjectionLab using both randomized and historical-backtested returns, then adjust your spending, savings, or retirement date and watch how your probability of success responds.

Frequently Asked Questions

What is a good Monte Carlo success rate for retirement? Most planners treat 80% to 90% as a reasonable target, with many aiming near 85%. A rate that high leaves room for bad markets while acknowledging that chasing 100% usually means saving far more or spending far less than you need to. The right number depends on your flexibility: if you can cut spending in a downturn, you can comfortably plan around a lower success rate than someone with fixed, non-negotiable expenses.

How many simulations should you run? Most tools run somewhere between 1,000 and 10,000 simulations, and 1,000 is generally enough for a stable success rate. The quality of your return and inflation assumptions matters far more than the raw number of runs.

Why does a Monte Carlo simulation differ from a simple retirement calculator? A simple calculator applies one fixed return every year and produces a single ending balance, while a Monte Carlo simulation tests thousands of randomized return sequences and reports how often the plan survives. The simple calculator can’t see sequence of returns risk, so two plans it rates as identical may have very different real-world odds. That gap is exactly what a Monte Carlo retirement calculator is built to expose.

Can a Monte Carlo simulation predict the market? No. It doesn’t forecast returns or tell you what the market will do next year. It models a range of plausible futures based on the assumptions you give it, then reports how a plan like yours holds up across them. The output is a probability of success, not a prediction of any specific outcome.