How the verdict is calculated.
RealVerdict scores every deal on the same six dimensions, against the same thresholds, using the same data sources. No black box. Here's exactly what happens between "paste an address" and "STRONG BUY."
1 · Where the inputs come from
Every input on the verdict page has a source badge. We do not guess. If a number is unavailable from a primary source, we use a transparent fallback and label it as such.
| Field | Source |
|---|---|
| Purchase price, beds/baths/sqft, year built, last sale | RentCast public records → Zillow listing fallback |
| AVM (estimated value) | RentCast AVM |
| Estimated market rent | RentCast AVM rent estimate, cross-checked against active rent comps |
| Property tax | RentCast assessor records → state-level effective rate fallback |
| Insurance | State-level annual averages by property type |
| 30-yr mortgage rate | FRED weekly Freddie Mac PMMS |
| Long-run appreciation | FHFA House Price Index for the metro (CBSA) |
| Flood zone | FEMA NFHL |
| Sale & rent comparables | RentCast within a 1-mile radius, beds/baths-matched |
2 · The math
Standard rental analysis, with one important detail: every line item is real, not assumed. We don't silently set vacancy to 0% to make a deal look better.
NOI = (Gross Rent + Other Income) × (1 − Vacancy %) − Property Tax − Insurance − HOA − Utilities − Maintenance − Property Mgmt − CapEx Reserve
Cap Rate = NOI / (Purchase Price + Rehab)
Monthly Cash Flow = (NOI / 12) − Monthly Mortgage Payment
DSCR = NOI / Annual Debt Service
CoC = Year-1 Cash Flow / Total Cash Invested (Down Payment + Closing + Rehab)
Break-even % = (Operating Expenses + Debt Service) / Gross Rent
Internal rate of return on the year-by-year cash flow series, including projected sale proceeds at the end of the hold period.
3 · The verdict tiers
The rubric awards points across six categories — cash-on-cash (12pts max), cap rate (15), DSCR (15), IRR (22), GRM (12), and break-even occupancy (12). Total possible is 100. We add a small long-term "appreciation rescue" modifier that softens the penalty when a deal is cash-flow-negative year 1 but the IRR + equity growth still make the math work over the hold.
| Tier | Score | What it means |
|---|---|---|
| STRONG BUY | 75 – 100 | Income comfortably covers debt, returns clear the bar, healthy margin for error. |
| GOOD | 55 – 74 | Numbers pencil out. A few metrics below ideal, nothing disqualifying. |
| BORDERLINE | 35 – 54 | Marginal. Returns are modest and the deal needs things to go right. |
| PASS | 15 – 34 | Cash flow, leverage, or both work against you. |
| AVOID | 0 – 14 | Projected to lose money or leave you dangerously exposed. |
4 · The walk-away price
This is the number no other rental analyzer gives you. For each tier, we run a binary search over the purchase price holding every other input constant (rent, rate, taxes, expenses) and find the highest price at which the verdict still earns that tier or better.
The result is a card showing: "Highest price for STRONG BUY: $X. For GOOD: $Y. For BORDERLINE: $Z." You walk into negotiations knowing exactly where the deal stops working.
We omit a tier if it's unreachable at any price ≥ $1,000 — for example, an all-cash deal at a great rent will never have a walk-away ceiling at the AVOID tier, because no purchase price that small is realistic.
5 · Stress tests
Five scenarios that all happened in the last 24 months. Each one re-runs the full verdict against the same property with one variable shocked.
- Mortgage rate up 1 percentage point
- Rent down 10%
- Vacancy doubles (e.g. 5% → 10%)
- Insurance up 30%
- Property tax reassessed up 25%
The stress test panel highlights which scenarios push the verdict from BUY to PASS, and which the deal absorbs comfortably.
6 · Comparables
Pro users get a comps tab on every verdict. We pull recent sales and active rentals within a 1-mile radius from RentCast, filtered by beds/baths and (when known) property type. The statistics shown — median, low/high, sample size — are computed per-deal, not pre-baked. If we don't have enough comps to be statistically meaningful (fewer than 3), we tell you instead of guessing.
We use the median rent comp to cross-check the rent assumption you (or RentCast's AVM) provided. If your projected rent is materially above the comp median, we surface a "rent looks optimistic" warning before the verdict.
What we don't do (yet)
- Short-term rental modeling. Verdicts assume long-term lease income only.
- BRRRR / refi-out math. The walk-away price assumes a single purchase at the listed terms.
- Tax depreciation and personal income tax effects. Returns are pre-tax.
- Local rent control or eviction-moratorium adjustments. We use national stress assumptions.
Each of these is on the roadmap. If your deal hinges on one, treat the verdict as a directional read, not an oracle, and verify with a CPA and local counsel.