Two worked examples, with the numbers the model produced.
Neither is a closed transaction. The single-family property is a real address that was really sourced and underwritten, and no offer was made on it. The multifamily deal is constructed. There is no such building, and the inputs it is built from are listed on the page.
1813 Linden Ave, Memphis, TN 38104
Single-family · 3 bed / 1 bath · 1,884 sqft · built 1892 · vacant
Viable as a disposition at or below $189,596. The margin tier would have allowed $226,996 and the deal was originally priced there; every simulated counterparty passed, because neither a flipper nor a financed landlord can reach that number. Still not viable as a levered rental hold: the asset yields 5.24% on total cost against 7.5% debt, so leverage subtracts, and that conclusion survives every sensitivity tested.
- No interior inspection. Condition is UNKNOWN on a house built in 1892, against an $80,078 repair estimate. The single largest uncertainty here.
- Comps run $136.61 to $252.39 per square foot. That dispersion is wider than an acquisition margin.
- Every expense line except the tax bill is a percentage convention, not a quote.
- The two counterparty screens land $985 apart. That convergence is why the ~$189,000 figure is treated as the clearing price rather than an artifact of either rule.
Illustrative. 40-unit garden-style apartments, Memphis, TN
40 units · 34,680 sqft · 1970s vintage · garden-style
Financeable at 1.355× coverage and a 9.96% debt yield. The interesting number is economic vacancy at 15.68% against 6.00% physical. The gap is loss to lease, non-revenue units, concessions and bad debt, and it is the figure a lender asks for.
Both workbooks are live-formula: change an assumption and everything downstream moves. Every output reconciles to the underwriting engine to the cent.