Understanding Monte Carlo Analysis
Monte Carlo simulation runs your scenario thousands of times with randomized annual returns. The result isn’t a single prediction — it’s a range of possible outcomes. This page explains how to read and use that information.
What Monte Carlo Tells You
Section titled “What Monte Carlo Tells You”A deterministic projection says: “If returns are exactly X% every year, here’s your net worth.”
A Monte Carlo projection says: “Given your assumptions about volatility, here’s the range of outcomes you might see, and how likely each one is.”
The key outputs are:
- Probability of Success — what percentage of simulations never ran out of money
- The shaded band chart — showing the range from pessimistic (10th percentile) to optimistic (90th percentile)
- The median line — the middle-of-the-road outcome (50th percentile)
Reading the Band Chart
Section titled “Reading the Band Chart”The main Monte Carlo chart shows your net worth trajectory as shaded regions:
- Darkest band (center) — the 25th to 75th percentile range. Half of all simulated outcomes fell within this zone.
- Lighter bands — extend to the 10th and 90th percentiles. 80% of outcomes fall within the full shaded area.
- Solid line — the median (50th percentile) outcome.
A wide band means high uncertainty — small changes in returns produce very different outcomes. A narrow band means the outcome is relatively predictable.
Understanding Success and Failure
Section titled “Understanding Success and Failure”Success is defined by the failure_mode in your scenario’s configuration:
| Failure Mode | What Counts as Failure |
|---|---|
net_worth_below_zero |
Total net worth drops below zero |
liquid_depletion |
Available cash and investments are exhausted |
spending_shortfall |
Can’t meet the spending target |
preserve_home_equity |
Must sell the home to cover spending |
A 90% probability of success means that in 9 out of 10 simulations, the plan never triggered the failure condition. This does not mean there’s a 10% chance of going broke — it means that under 10% of the random return sequences tested, the plan didn’t hold up.
Key Metrics in the Simulation Tab
Section titled “Key Metrics in the Simulation Tab”| Metric | What It Means |
|---|---|
| Median Terminal NW | The middle outcome — half did better, half did worse |
| Worst-Decile Terminal NW | The 10th percentile outcome. If you’re planning conservatively, plan for this. |
| First Failure Year | The earliest year where any simulation failed. Tells you your risk window. |
| Peak Failure Rate | The highest percentage of simulations failing in a single year. A spike here identifies a vulnerable period. |
Tips for Using Monte Carlo
Section titled “Tips for Using Monte Carlo”- Start with deterministic — get your assumptions right in deterministic mode first
- Increase
num_runsfor smoother bands — the default is 1000, but 5000+ gives cleaner percentiles - Check the failure timing — if failures cluster in a specific year (e.g., right after retirement), that’s your risk period
- Use
failure_grace_period_months— a 6-month grace period prevents temporary dips from counting as failures - Compare scenarios in deterministic mode — Monte Carlo adds stochastic noise that can obscure scenario differences
When Not to Use Monte Carlo
Section titled “When Not to Use Monte Carlo”- Initial scenario design — use deterministic, it’s simpler and clearer
- Comparing two scenarios — use deterministic; stochastic noise can make small differences invisible
- Presenting to a partner — deterministic is easier to explain
Further Reading
Section titled “Further Reading”See the Render Modes page for configuration details and the Configuration Reference for the full [monte_carlo.success] schema.