Why It Matters
Instead of plugging in one set of assumptions and calling it your projection, Monte Carlo simulation asks: what if vacancy were 5%? What if it were 15%? What if interest rates rose? It runs all those combinations — often 10,000 or more — simultaneously and tells you how often the deal makes money, breaks even, or loses. The result isn't a single number but a curve showing the likelihood of every possible outcome. When paired with a thorough discounted cash flow analysis, it transforms a static projection into a dynamic risk picture. That's the difference between knowing your expected return and knowing your probable return.
At a Glance
- Runs thousands of randomized input combinations through your deal model in seconds
- Outputs a probability distribution — e.g., "73% chance of positive cash flow"
- Captures correlated risks that sensitivity analysis misses
- Works best when you have historical data to inform your input ranges
- Common in institutional real estate underwriting; increasingly accessible to individual investors via spreadsheet add-ins
How It Works
Monte Carlo simulation starts by replacing fixed assumptions with probability distributions. Instead of saying "vacancy will be 8%," you define a range — say, 4% to 16% — and assign a distribution shape (often normal or triangular). You do this for every uncertain input: rent growth, cap rate at exit, repair costs, hold period. The model then randomly samples from each distribution, computes the output (cash-on-cash return, IRR, equity multiple), records the result, and repeats — typically 10,000 times.
After thousands of iterations, the results are aggregated into a probability distribution. You might find that your deal has a median IRR of 9.2%, a 10th-percentile IRR of 3.1%, and a 90th-percentile IRR of 15.8%. This tells you far more than a single-point projection ever could. You can also calculate the probability of a loss — something traditional underwriting never explicitly surfaces. Understanding that your deal has a 12% chance of delivering negative returns is exactly the kind of information that separates disciplined investors from hopeful ones. Thinking carefully about the opportunity cost of capital makes this probability especially meaningful.
The inputs you choose — and their ranges — matter as much as the math. Garbage in, garbage out still applies. If you set your rent growth distribution between 3% and 5% when the market has historically ranged from negative 8% to positive 12%, your simulation will be overconfident. The most rigorous use of Monte Carlo pairs historical data with expert judgment to set realistic bounds. You should also consider your weighted average cost of capital as a benchmark: if most simulation runs don't clear your hurdle rate, the deal fails the test regardless of what the base case says. Keep in mind that any sunk costs already spent — modeled via sunk cost thinking — should not influence which scenarios you test; only forward-looking cash flows matter.
Real-World Example
Bryce is analyzing a 12-unit apartment building listed at $1.4 million. His base case spreadsheet shows an 8.7% cash-on-cash return using 7% vacancy, $1,850 average rent, and a 6.2% exit cap rate. But Bryce has seen enough cycles to know that each of those numbers can move — a lot.
He builds a Monte Carlo model with vacancy ranging from 4% to 18%, rent between $1,650 and $2,050, and exit cap rates from 5.5% to 7.5%. After 10,000 iterations, the simulation shows a median return of 8.4%, a 10th-percentile return of 2.1%, and a 90th-percentile return of 14.9%. More importantly, it shows an 18% probability of negative cash flow in any given year. Bryce decides the deal is acceptable but negotiates the price down to $1.32 million — enough to shift that loss probability below 10%. He also tracks marginal cost assumptions on future capital expenditures within the model so that each repair scenario is independently randomized. The simulation turns a gut-feel decision into a quantified one.
Pros & Cons
- Reveals the full range of outcomes — not just best, worst, and base case
- Assigns explicit probabilities to both upside and downside scenarios
- Captures how multiple uncertain variables interact, which simple sensitivity tables miss
- Scales to any deal type: single-family, multifamily, commercial, development
- Increasingly accessible through Excel add-ins like @RISK and free Python libraries
- Output quality depends entirely on the accuracy of your input distributions
- Can create false confidence if ranges are set too narrow or too optimistically
- Requires more upfront modeling work than standard DCF projections
- Results can be difficult to communicate to partners or lenders unfamiliar with probability distributions
- Computationally intensive models may be slow without dedicated software
Watch Out
Don't treat the median output as your new "base case" and stop there. The median is useful, but the tails are where the real intelligence lives. A deal with a 9% median IRR and a 25% probability of IRR below 4% is very different from one with the same median and only a 5% probability of that outcome. Always examine the 10th and 90th percentile outputs — they tell you what you're actually signing up for in bad and good markets.
Correlated risks are the most common modeling mistake. Rising cap rates and falling rents often happen at the same time — in a downturn, both move against you simultaneously. If your simulation treats them as independent variables, it will underestimate your true downside. Build in correlations between inputs wherever historical data supports them. A 2008-style scenario that hits vacancy, rent, and exit cap rate at once will look far worse in a correlated model than in an uncorrelated one.
Monte Carlo is an analysis tool, not a decision-making machine. The simulation can tell you the probability distribution of financial outcomes, but it cannot capture regulatory changes, neighborhood transformation, or partnership risk. Use it to sharpen your underwriting, not to replace judgment. The best investors treat the output as one more input alongside market knowledge and experience — not as the final word. A strong simulation that clears your hurdle rate is a green light to keep digging, not a green light to close.
Ask an Investor
The Takeaway
Monte Carlo simulation is the most honest way to underwrite real estate risk. By running thousands of scenarios instead of one, it forces you to confront the full range of what could happen — not just what you hope will happen. It won't make a bad deal good, but it will make a good deal legible, and it will expose the risks hiding inside an optimistic base case. If you're managing more than a handful of properties or underwriting deals above seven figures, adding Monte Carlo to your toolkit is worth every hour it takes to set up.