Methodology

Honest math for
retirement planning

Most retirement calculators assume markets return exactly 7% every year. That's not how markets work — and it leads to dangerously optimistic projections. Here's how we do it differently.

Why Monte Carlo simulation?

A fixed-rate calculator draws a straight line to your retirement. Monte Carlo draws thousands of possible futures — some great, some catastrophic — then tells you how often each kind happens.

50 simulated retirement paths
Same starting conditions, different market outcomes — this is the spread you're actually betting against.
Retirement354555657585Fund outlasts planPartially fundedFunds depleted
📈Sequence of returns risk
Retiring into a bear market is very different from retiring into a bull market, even if average returns are identical. Monte Carlo captures this.
🎲Each run is independent
Every simulation samples from a realistic return distribution each year. Bad years cluster — just like in real markets.
📊Success rate = real odds
The percentage shown is literally the fraction of simulated futures where you don't run out of money before your planning age.

Log-normal returns, not normal

A naive model draws returns from a normal (bell curve) distribution — which allows a portfolio to go negative from a single year. Real asset prices follow a log-normal distribution: returns are multiplicative, so a portfolio can fall toward zero but never past it.

✗ Naive: Normal distribution
return = μ + σ·Z
  • · Portfolio value can go negative in a single year
  • · Symmetric: equally likely to be very good vs. very bad
  • · Underestimates crash severity, overestimates recovery speed
✓ Ours: Log-normal distribution
return = exp(μlog + σ·Z) − 1
  • · Portfolio value is always ≥ 0
  • · Right-skewed: occasional large gains, bounded losses
  • · Better reflects how compounding actually works

Automatic glide path

Holding 80% stocks at age 60 is a very different risk profile than holding it at 40. The simulator gradually shifts your allocation from your current mix toward your retirement mix as you approach retirement — matching how target-date funds work.

Allocation over time (example: 80% → 60% stocks)
Stocks / Growth assets
Bonds / Stable assets

What we model

Retirement income is messier than a simple portfolio withdrawal. These are the real-world variables we account for in every simulation.

📉Stochastic inflation
Inflation is sampled independently each year from a normal distribution centered on your assumption. Your spending needs, Social Security, and healthcare all respond to their own inflation draws.
🏥Healthcare inflation
Medical costs historically outpace general inflation. Healthcare is tracked separately with its own higher inflation rate — defaulting to 6% — because ignoring it severely understates retirement costs.
🏛️Social Security COLA
Your SS benefit receives cost-of-living adjustments each year, sampled from the same inflation draw as the rest of the economy. Benefits are zero before your chosen start age.
📋IRS Required Minimum Distributions
At age 73, the IRS requires minimum withdrawals from traditional (non-Roth) accounts. We apply the Uniform Lifetime Table and force the RMD if it exceeds your planned withdrawal.
🔄Roth vs. traditional split
Roth withdrawals are tax-free. Traditional withdrawals are grossed up by your effective tax rate. The Roth fraction you set determines how your withdrawals are sourced and taxed.
🏠Home equity
Your home appreciates at a separate rate and your mortgage amortizes deterministically. Net equity is tracked at retirement and at end-of-plan, shown as part of total net worth.
📦Alternative assets
Crypto, gold, REITs, and other alternatives are modeled as a third asset class with independent log-normal returns and user-defined volatility — separate from your stock and bond allocations.
⚖️Contribution timing
We use the "midpoint method" — half your annual contribution is added before the annual return is applied, half after. This approximates monthly contributions without simulating 12 sub-steps per year.
🎯Three-way allocation
Stocks + alternatives + bonds sum to 100%. If you set 85% stocks and 15% Bitcoin, bonds get 0%. If the total would exceed 100%, the allocations are normalized proportionally.

Default assumptions

All values are editable. These defaults reflect broadly-used planning assumptions and can be changed to match your specific situation.

ParameterDefault
Stock return7.0%
Stock volatility15%
Bond return4.0%
Inflation2.5%
Healthcare inflation6.0%
Tax rate22%
Planning horizonAge 95

What we don't model

No simulator captures everything. Being honest about limitations helps you use the results correctly.

Tax optimization
The main simulator applies a flat effective tax rate. The Roth Conversion Optimizer models bracket-filling and RMD reduction, but tax-loss harvesting, state taxes, and full bracket management are not yet modeled.
Return correlations
Each asset class is sampled independently each year. In reality, stocks and bonds can become correlated during crises. Tail-risk scenarios may be slightly underestimated.
Variable spending
Many retirees naturally spend less in bad years. A fixed withdrawal assumption is more conservative than how people actually behave — which means our success rates may be understated.
Pension income
Defined-benefit pensions aren't modeled as a distinct input. Include any guaranteed pension income as part of your desired monthly income offset or as a Social Security equivalent.
Behavioral risk
Panic-selling in a crash, chasing returns, or deviating from your allocation are major failure modes the model can't account for. The simulation assumes you stay the course.
Future policy changes
Social Security solvency, tax law, healthcare policy, and RMD rules can all change. The model uses current rules as a baseline — plan for some uncertainty in these inputs.

Your data stays on your device

We built this with privacy as a default, not an afterthought.

🔒Stored locally only
Every number you enter — savings, age, income, portfolio — is saved in your browser's localStorage. It never leaves your device.
🚫No account required
We don't collect your email, create a profile, or ask you to sign up for anything. There's no server receiving your financial data.
🗑️Delete it anytime
Use the "Clear saved data" link in the footer to wipe everything instantly. Clearing your browser cache does the same.

The only external service this site uses is Google Analytics for anonymous page-view counts (no personal data, no financial data). You can block it with any ad blocker.

Everything in the toolkit

The retirement simulator is the core, with four more planning tools built into the same app — all free, no signup.

💼Portfolio Builder
Add your investment accounts, home, and debts to get a real net worth snapshot. Your Roth fraction, asset mix, and total savings flow automatically into the retirement simulator.
Build portfolio →
📊Retirement Simulator
Monte Carlo simulation across 500–5,000 futures. Models your full picture: stocks, bonds, crypto, Social Security, healthcare, RMDs, Roth/traditional split, and inflation.
Run simulation →
🏠Housing Calculator
Two tools in one: should you pay down your mortgage early or invest the extra money? And should you rent or buy in the first place? Both use Monte Carlo to compare outcomes.
Open Housing →
🏥Healthcare Estimator
Projects your ACA marketplace premiums year by year before 65, and Medicare IRMAA surcharges after. Shows exactly how your income level affects healthcare costs — including the ACA subsidy cliff.
Estimate costs →
🔄Roth Conversion Optimizer
Shows how much you can convert from a traditional IRA to Roth each year to fill low tax brackets, shrink future Required Minimum Distributions, and reduce estimated lifetime taxes.
Optimize conversions →

Learn the concepts

The numbers only mean something if you understand the ideas behind them. These guides use plain English and interactive visuals to explain the retirement concepts that drive your simulation results.

Run your own simulation

Free, no account required. Your data never leaves your browser.