skills/ca-property-tax-appeal/references/comparable-sales-method.md

# Comparable-sales method

## Selection order

Rank candidates by the combined fit of:

1. Sale timing relative to the jurisdiction's valuation date.
2. Same property type and legally relevant use.
3. Same neighborhood, subdivision, school/market area, or other recognized submarket.
4. Geographic proximity when market-area boundaries are unavailable.
5. Similar above-grade living area, generally starting within about 15–25% and widening only with disclosure.
6. Similar lot size, age, quality, condition, bed/bath utility, parking, view, waterfront, basement, pool, ADU, and renovation status.

These are search heuristics, not legal thresholds. Replace them with jurisdiction-specific standards when official guidance exists.

## Timing

Anchor research to the valuation date, not today's date. Prefer closed sales before or around that date. Include later sales only when locally permissible and explain their reduced relevance. If market prices changed materially across the period, do not apply a time adjustment without credible local evidence.

For California assessment appeals, a board may not consider a comparable sale occurring more than 90 days after the valuation date. Treat the 90th calendar day as inclusive and the 91st day as excluded. A sale may predate the valuation date, but closer sales are generally stronger evidence. For a Proposition 8 decline-in-value appeal, the valuation date is normally January 1 of the appeal year; do not substitute the notice date or today's date.

## Verification

For each comparable record:

- Exact address.
- Closed sale price and recording/closing date.
- Source URL and access date.
- Property type, living area, lot, year built, beds/baths, and condition when known.
- Distance or same-market-area rationale.
- Arm's-length status or a clear `unknown` label.
- Material differences from the subject.

Cross-check secondary portal facts against official records where possible. If two sources disagree, show both and do not silently choose one.

## Required table

| Property | Sale date | Sale price | Living area | Price/area | Lot | Year built | Beds/baths | Distance/area | Material differences | Source class |
|---|---:|---:|---:|---:|---:|---:|---:|---|---|---|

Add source links in the property or source-class cell. Put `unknown` rather than guessing.

## Analysis guardrails

- Do not average all sales mechanically.
- Median price per square foot may be a descriptive check, not the valuation conclusion.
- Do not use listing price as sale price.
- For California, do not include a sale more than 90 days after the valuation date in the evidentiary set, even as a low-weight comparable.
- Do not use an automated valuation model as a substitute for comparable evidence.
- Do not make a dollar adjustment for a pool, bathroom, view, condition, garage, or square footage without market-derived support.
- Give more weight to the most comparable verified transactions and explain the weighting qualitatively.
- Include credible counterexamples; advocacy is stronger when limitations are candid.

## Confidence

- `Strong`: at least three closely matched, verified arm's-length sales near the valuation date, consistent subject facts, and little contradictory evidence.
- `Moderate`: useful verified sales exist but material differences, timing, or record gaps remain.
- `Weak`: sparse, secondary, distant, or materially dissimilar sales dominate.
- `Insufficient`: the subject cannot be matched confidently, sale data cannot be verified, or the relevant valuation standard is unknown.