Plan Stress Test (Monte Carlo Simulation)
Runs the full strategic plan 100 times with randomized market return sequences to show the range of plausible outcomes. Measures plan resilience to sequence-of-returns risk — the danger that bad returns early in retirement cause permanent damage even if long-run averages hold.
How it works
The main projection uses a single fixed rate of return every year, which is optimistic — real markets don't deliver smooth returns. The stress test addresses this by running the entire strategic plan 100 times, each with a different sequence of annual returns drawn from a calibrated volatility model. The mean return matches your plan assumption; the spread reflects realistic year-to-year market volatility. Because every path runs through the full calculation engine (including all active strategies), auto-mode strategies like Structured Withdrawals and Roth Conversion adapt their amounts to each path's actual asset balances — this is not a simplified cash-flow approximation.
1. Generate Return Sequences
For each of the 100 paths, the model generates a sequence of annual returns using calibrated volatility around the plan's assumed mean return. The volatility factor (σ = 1.7 × mean return) produces realistic year-to-year variation — for example, a 7% mean return yields ~12% standard deviation, meaning individual years commonly range from -5% to +19%. A drift correction ensures the geometric mean across paths stays close to the assumed return. All results are deterministic (fixed random seed), so the same plan always produces the same stress test output.
σ = 1.7 × meanReturn | returnᵧ = meanReturn + (0.5 × σ²) + (zᵧ × σ)
Example: 7% mean return → σ ≈ 11.9%. One path might see -8% in year 1, +22% in year 2, +4% in year 3. Another sees +15%, +2%, -12%. Same average over time, very different outcomes.2. Run Full Projections Per Path
Each path runs through the complete calculation engine — the same two-pass architecture used for the main projection. All active strategies execute with amounts that adapt to that path's actual asset balances. Auto-mode Structured Withdrawals recalculate the income gap each year based on available balances. Roth Conversion amounts adjust to the tax space available given that path's income and bracket position. This means the stress test reflects how the plan actually responds under stress, not just whether assets survive a static withdrawal schedule.
Example: Path 37 has a -15% return in year 2. Auto-mode SW draws less from depleted accounts. Roth Conversion finds a larger tax bracket gap because income dropped. The stress test captures these interactions.3. Calculate Success Rate
A path 'succeeds' if the portfolio stays above the minimum balance floor throughout the entire projection — not just at the end. The floor is the sum of any Structured Withdrawals tier minimums (if configured); otherwise it defaults to zero. When no explicit floor is set, auto-mode SW gracefully degrades withdrawals as balances drop, so the success/failure threshold is whether the portfolio is fully depleted. The success rate is the percentage of 100 paths that never breach the floor.
Success Rate = (paths that never breach floor ÷ 100) × 100%
Example: 78 of 100 paths stay above the floor → 78% success rate. Color coding: green ≥ 60%, amber 40–59%, red < 40%.4. Compute Percentiles and Median
For each projection year, the model sorts all 100 path values and extracts percentile bands: 10th, 25th, 50th (median), 75th, and 90th. The median ending portfolio is the 50th percentile final-year value — it typically falls below the fixed-return projection due to sequence-of-returns drag. The outcome spread (10th to 90th percentile) shows how sensitive the plan is to market timing. A wide spread means outcomes depend heavily on the specific return sequence; a narrow spread means the plan is relatively robust.
Example: Ending portfolio: 10th percentile $420k, median $1.2M, 90th percentile $3.1M. Spread = $2.68M. The fixed-return projection showed $1.5M — the median is lower because bad early years are disproportionately harmful.5. Build the Fan Chart
The fan chart combines two visual layers. First, 10 named representative paths (hand-crafted scenarios like 'Early Crisis', 'Prolonged Bear', 'Strong Start', etc.) are drawn as individual lines that advisors can trace to understand specific narrative outcomes. Second, gradient bands show the full distribution from the 100 statistical paths — darker bands near the median, lighter bands toward the extremes. A solid line marks the median, and a dashed line marks the fixed-return projection for comparison.
Real-world context
Why Sequence of Returns Matters
Two retirees with identical 7% average returns over 20 years can have vastly different outcomes. A retiree who experiences -20%, -10% in years 1–2 and strong returns later may run out of money — while one who gets the same returns in reverse order ends up wealthy. This is because withdrawals in early down years lock in losses that compound forward. The stress test directly measures this risk by showing how many return sequences the plan can survive.
Interpreting the Success Rate
A 100% success rate usually means the withdrawal rate is very conservative — the client could likely afford to spend more or give more. A rate in the 70–85% range is generally considered solid for most clients. The key question is not 'is the rate high enough?' but 'does the client have flexibility to adjust if they land in an adverse path?' Clients with pension income, Social Security, or the ability to reduce discretionary spending have more flexibility and can tolerate lower success rates. Clients whose income depends entirely on portfolio withdrawals need higher success rates.
Why the Median Is Lower Than the Main Projection
The main projection uses a fixed return every year (e.g., 7% per year, every year). This overstates likely outcomes because it ignores the mathematical reality that volatility reduces compound growth. A portfolio earning +30% then -30% does not return to its starting value — it loses 9%. The stress test median captures this 'volatility drag', making it a more realistic central estimate than the fixed-return projection.
Using the Outcome Spread in Client Conversations
The 10th-to-90th percentile spread is a powerful communication tool. A narrow spread says: 'Your plan is relatively insulated from market timing — even unlucky sequences leave you in reasonable shape.' A wide spread says: 'Your outcome depends significantly on when good and bad years happen — let's build in more flexibility.' Advisors can use this to motivate strategies that narrow the spread (lower withdrawal rates, more diversification, Roth conversions that reduce future tax exposure).
Fixed Return vs. Stress Test — When to Use Each
The fixed-return projection (Assets Tab) is useful for explaining strategy mechanics — how Roth conversions shift tax brackets, how Structured Withdrawals manage draw order, etc. The stress test is useful for answering 'will this work?' It tests whether the strategy holds up under realistic market conditions, not just under the advisor's assumed return. Most planning conversations should reference both views.
What drives the result
Percentage of 100 simulated return paths where the portfolio stays above the minimum balance floor throughout the projection. Higher rates indicate greater plan resilience. Color coding: green (≥ 60%), amber (40–59%), red (< 40%). A success rate below 50% suggests the withdrawal rate or income target may be unsustainable under realistic market conditions.
78% success rate = 78 of 100 paths survived. 22 paths hit the floor at some point during the projection.
The 50th percentile final-year portfolio value across all 100 paths. Represents the most likely outcome under variable returns. Typically 10–30% below the fixed-return projection due to volatility drag. Compare this to the Assets Tab projection to understand how much sequence risk affects the plan.
Fixed-return projection: $1.5M ending value. Stress test median: $1.2M. The $300k difference reflects volatility drag and sequence-of-returns risk.
Width of the distribution between the 10th and 90th percentile ending values. Measures how sensitive the plan is to market timing. Wide spreads indicate high sensitivity; narrow spreads indicate robustness. Strategies that reduce portfolio concentration or shift assets to guaranteed income sources tend to narrow the spread.
10th percentile: $420k. 90th percentile: $3.1M. Spread: $2.68M. This client's outcome swings by $2.68M depending on return sequence.
Visual display combining 10 named scenario paths (individual lines) with gradient bands showing the full 100-path distribution. The solid line is the median; the dashed line is the fixed-return projection. Named paths help advisors tell stories: 'If markets crash early in retirement (like 2008), here's what happens.' Gradient bands show the probability envelope.
The 'Early Crisis' path shows a sharp drop in years 1–3 followed by recovery. The 'Prolonged Bear' path shows slow erosion over 5+ years. Both are visible as distinct lines against the gradient background.
Sets the mean return for all stress test paths. Higher mean returns improve the success rate and raise the median, but do not eliminate sequence risk. The stress test volatility is proportional to the mean (σ = 1.7 × mean), so higher assumed returns also produce wider outcome spreads.
6% mean → σ ≈ 10.2%, moderate spread. 8% mean → σ ≈ 13.6%, wider spread. Higher return assumption improves the median but increases the range of outcomes.
All active strategies run in every stress test path with adaptive amounts. Auto-mode strategies (Structured Withdrawals, Roth Conversion) recalculate per-path, making them particularly valuable under stress — they naturally reduce draws in down-market paths and increase Roth conversions when bracket space opens up. Adding or removing strategies changes the stress test results.
Plan with auto-mode SW shows 82% success rate. Same plan with fixed-rate SW shows 71% — auto-mode adapts to bad sequences by drawing less when balances drop.
Assumptions
- Mean return matches the plan assumption — the stress test varies the sequence, not the long-run average
- Volatility is calibrated at σ = 1.7 × mean return (e.g., 7% mean → ~12% standard deviation)
- Returns are deterministic (fixed random seed) — the same plan always produces identical results
- All active strategies run in each path with amounts that adapt to that path's balances
- 100 statistical paths provide stable percentile estimates and success rates
- The stress test always runs the strategic plan, not the base case
Limitations
- Returns are log-normally distributed — does not model fat tails, regime changes, or correlated crashes
- A single volatility parameter applies to all account types (no per-asset-class modeling)
- No correlation modeling between asset classes — all accounts experience the same return shock
- Does not model behavioral responses (e.g., advisor reducing withdrawals after a crash beyond what auto-mode does)
- 100 paths are sufficient for stable medians and success rates but may under-represent extreme tail events
- Does not account for inflation volatility — only investment return volatility is modeled
Related
This page explains how Stratum models this calculation. It is educational material for financial professionals, not tax or legal advice, and tax law changes. Verify current figures against primary IRS sources before relying on them with a client.