How the assay works
Most rental calculators multiply rent by a rule of thumb and hand back one confident number. That number is almost always wrong, and it is wrong in a predictable direction: too optimistic.
Why a range instead of a number
Rent is uncertain. Repairs are uncertain. Vacancy is uncertain. Multiplying uncertain inputs and reporting a single figure to the dollar hides all of it. Abode Rock carries the uncertainty end to end and shows the band it produces, because the width of that band is the information you need to size a decision.
The three lines that sink first deals
A property can clear its mortgage every month and still lose money. These are the reserves left out of most back-of-envelope maths, and they are modelled here by default:
| Reserve | Modelled at | Why it exists |
|---|---|---|
| Capital reserve | 1.0–1.5% of price a year | Roofs, furnaces, and water heaters do not fail monthly, but they do fail. Averaged out, they are rent. |
| Vacancy allowance | 5–8% of gross rent | Twelve months of rent a year is a fiction. Tenants leave, and units sit empty. |
| Turnover and leasing | 0.4–0.8 months of rent | Paint, clean, list, screen. Every turnover costs real money before the next cheque arrives. |
How the downside figure is produced
Rent is drawn from a triangular distribution across its modelled band, and each expense line is drawn across its own. Four thousand draws are run, and the share landing below break-even becomes the chance of a negative month. The run is seeded from your inputs, so the same property always returns the same figure — this is a measurement, not a slot machine.
How the grade is weighted
Cash flow, cash-on-cash return, capitalisation rate, and downside risk are scored on curves calibrated to ordinary residential investment, then combined. The weighting shifts with the deal: a financed purchase leans on cash-on-cash, while an all-cash purchase leans on cap rate, because that is what each buyer is actually deciding.
The scale stops at 3 and 97. A 0 or a 100 would claim a certainty this model does not have.
What lowers confidence
- No connected comp feed — your rent estimate cannot be checked against the market.
- Few or scattered comparables — the band widens rather than the estimate hardening.
- Blank fields — named openly as gaps instead of filled with a plausible default.
- An estimated property tax — the largest local variable, and the easiest to get wrong.
What this is not
This is not an appraisal, and it is not advice. It is a disciplined estimate with its error stated. A range is a reason to look harder at a deal, never a substitute for a written quote, an inspection, and your own reading of the lease and the local market.