A valuation is a best guess with a range around it. Run the range through a simulation and you can ask how often a price actually gets reached. We ran the live model on our home page against a $1.4m asking price: about 1 in 10, and the yield decided most of it.
The method in plain words
Income property is often valued by capitalisation. Take the rent, take off vacancy, and divide by the yield the market is paying for that kind of property. A higher yield means a lower value, and a lower yield means a higher value.
For example, using the middle figures in our model: $86,000 of rent with 3% vacancy leaves $83,420. At a 6.25% yield, that is $1,334,720. At 5.75%, the same income is worth about $1,450,783. Nothing about the building has changed, and the yield alone moves the value by about $116,000.
The model calls the rent input "net rent". Before you use the figure, check that it is net of the outgoings the owner pays.
Why a range beats one number
A single valuation of $1,334,720 against a $1.4m asking price looks like a clear answer. The asking price is about 4.9% above it. That is not the whole picture. Each input is a best guess, and the answer moves with each one. A range shows how far.
What the simulation does
- Each input gets three values: low, likely and high. The model draws from a curve that peaks at the likely value and thins out toward the ends.
- It draws rent, vacancy and yield together, works out a price, and repeats that 20,000 times.
- It counts the share of runs that reach the asking price.
The "give or take" figure is the simulation's own noise. It is not the uncertainty in your ranges. If a range is wrong, the answer is wrong, and the plus-or-minus will not show it.
What we ran
We used the figures on the home page: rent of $82,000, $86,000 and $90,000 a year; yield of 5.75%, 6.25% and 7.00%; vacancy of 0%, 3% and 8%; and an asking price of $1.4m. Run 20,000 times with seed 4417, the model gives a 9.93% chance of reaching $1.4m, give or take 0.20%. That is about 1 in 10.
Ten seeds in all, including 4417, gave answers between 9.52% and 10.06%, so the answer is stable at this scale. The median run came in at $1.32m. Only about one run in ten reached the asking price.
The yield decides it
Each row below holds one input at one end of its range, and the other two stay as ranges.
| Input | Chance of $1.4m at one end | Chance of $1.4m at the other end |
|---|---|---|
| Yield | 90.68% at 5.75% | 0.00% at 7.00% |
| Rent | 0.03% at $82,000 | 40.41% at $90,000 |
| Vacancy | 30.39% at 0% | 0.07% at 8% |
Yield moves the answer most, then rent, then vacancy. With rent at $86,000 and vacancy at 3%, the price is reached at a yield of about 5.96% or lower. Above that, it isn't.
Put every input at its likely value and the price is not reached at all. The middle case gives $1,334,720 at 6.25%. That is the trap of a single number. It looks like an answer, and it answers a narrower question than the one you are asking.
The ranges matter as much as the model
The model is only as good as the ranges you feed it. Narrow the yield range to 6.00%, 6.25% and 6.50%, and the chance of reaching $1.4m falls to 3.15%. Widen it to 5.50%, 6.25% and 7.50%, and the chance rises to 17.68%. Narrow the rent range to $84,000, $86,000 and $88,000, and the chance is 8.49%.
The median barely moves in those runs. It stays between about $1.32m and $1.33m. What changes is the tail: how low the yield could go, and how high the rent could go. A seller who wants the asking price to look reachable has a reason to widen the yield range. A buyer has a reason to narrow it. Ask who set the ranges, and why.
Running the same question with 100,000 runs gives 9.88%, give or take 0.09%. More runs make the simulation tighter. They do not change the inputs, and the inputs are where the real uncertainty sits.
How to use this on a real price
- Ask for ranges. Low, likely and high for rent, vacancy and yield, each with a reason.
- Name the input you are least sure of. In this example, the yield matters most.
- Mark what is confirmed and what is a best guess. Rent on a signed lease is confirmed. A yield taken from one sale is a best guess.
- Re-run it when a comparable sale lands, and change one range at a time to see what moves.
The model is on the home page, under the odds section. You can run your own numbers there, and each answer shows its seed, so the same question gives the same answer. If the building is leased, the rent figure comes from the lease, so the dates in it matter too. Our piece on commercial lease option dates covers those.
Command is a private system that sets itself up around one business. It watches dates like these and the news that touches your clients, and tells you early. Give us one past deal and we'll show you the day you could have known. → First look
This is general information, not legal or financial advice. Check your own lease and get advice for your situation.
Sources
- Live odds model, on the home page at /#odds. Figures and results above were re-run on 15 October 2026 with 20,000 runs and seed 4417, using the same inputs the page uses. The capitalisation arithmetic is shown in the text. We could not retrieve a primary definition to cite, so none is used.