Independent analysis, September 2026
We re-ran the math on Philadelphia's 2027 assessments. It fails two of the three industry accuracy tests.
The City published its methodology for the 2027 reassessment and reported that its values meet professional standards. We checked, using the same public records the City publishes: 14,506 single-family sales from the final 12 months of the City's own model window, compared against the new 2027 assessed values. Citywide accuracy misses the industry benchmark, and the misses are not evenly spread: the less your home is worth, the more likely it is over-assessed.
14,506
recent single-family sales analyzed, from the City's own records
54%
of those homes are assessed for 2027 above their own recent sale price
156%
median 2027 assessment, as a share of sale price, for homes that sold under $134k
2 of 3
industry accuracy standards failed citywide (COD and PRD)
The test the industry uses
Assessors worldwide grade mass appraisals with a ratio study: divide each property's assessed value by its actual sale price, then look at the spread. The International Association of Assessing Officers (IAAO) sets the passing grades. The City's methodology paper says its values pass. Our independent run of the same test, on unadjusted public sales records, says otherwise.
| Measure | What it captures | IAAO standard | City reports | Our result | |
|---|---|---|---|---|---|
| Median ratio | Overall level | 0.90 to 1.10 | 0.95 to 1.02 | 1.014 | Pass |
| COD | Uniformity: how far typical homes stray from the median | 15 or less | Under 15 | 28.7 | Fail |
| PRD | Regressivity: cheap homes vs. expensive homes | 0.98 to 1.03 | 0.98 to 1.03 | 1.165 | Fail |
| PRB | Regressivity, regression-based | -0.05 to +0.05 | Not reported | -0.229 | Fail |
The cheaper the home, the higher the assessment ratio
Group the same sales by price and the pattern is unmistakable. Homes that sold for under $134,000, the bottom fifth of the market, carry a median 2027 assessment of 156% of their own sale price. The most expensive fifth sits at 97%. A dashed line marks 100%, where an accurate assessment should cluster.
Median assessment as % of sale price
2,902 sales
2,901 sales
2,928 sales
2,894 sales
2,881 sales
Share of homes assessed above their own sale price
An assessment above the price a home actually just sold for is the single clearest appeal signal there is. For the cheapest fifth of the market, it is the norm, not the exception.
Where the over-assessment concentrates
The geography repeats what the City Controller and Community Legal Services found in earlier cycles: North, West, and Southwest Philadelphia carry ratios far above 100%, while several higher-priced neighborhoods sit at or below it. Median ratio by ZIP, sales from the study window (minimum 50 sales).
Assessed above recent sale prices
Assessed at or below recent sale prices
And the 2027 values moved up on top of it
Across all 462,893 single-family properties, 71% saw their assessed value increase for 2027, with a citywide median increase of 4.2%. The largest median jumps: Kensington 19134 (+11.4%), Manayunk 19127 (+10.8%), Chestnut Hill 19118 (+10.6%), Roxborough 19128 (+10.2%).
Methodology, honestly stated
Data: the City's public property and assessment records (opa_properties_public and the assessments table, phl.carto.com), pulled September 12, 2026. Universe: single-family residential.
Primary sample: arm's-length-priced recorded sales from July 1, 2024 through June 30, 2025, the final 12 months of the City's own model window, with recorded prices of $25,000 or more, joined to certified 2027 market values. Ratios outside 0.1 to 10 were dropped as data errors (360 of 28,910 records).
What we did not do: individually validate each sale as arm's length, or time-adjust prices to the valuation date. The City's own study does both, which partly explains why its statistics look better. Public records do not let outsiders replicate those adjustments, a transparency problem the City Controller has flagged before.
Why the finding survives anyway: we re-ran everything at price floors of $50,000, $75,000, and $100,000. Excluding ever more of the low-priced sales most likely to be non-market transfers, the cheapest tier still shows a median ratio of 1.28 or higher and at least 79% of homes assessed above their own sale price, and citywide PRD never falls inside the standard.
Fairness note: independent reviews credit the City's assessment office with real improvement over the past decade, and our median ratio result is consistent with an accurate overall level. The failures we measure are about uniformity and regressivity: who is over-assessed, not whether the average is right.
| Minimum sale price | Citywide COD | Citywide PRD | Under-$134k median ratio | Under-$134k share over-assessed |
|---|---|---|---|---|
| $25,000+ | 28.7 | 1.165 | 156% | 88.6% |
| $50,000+ | 23.0 | 1.110 | 144% | 86.5% |
| $75,000+ | 19.2 | 1.074 | 135% | 82.1% |
| $100,000+ | 16.8 | 1.053 | 128% | 78.9% |
Sources
- City of Philadelphia, OPA Tax Year 2027 Mass Appraisal Valuation Methodology (June 2026)
- City Controller, The Accuracy and Fairness of Philadelphia's Property Assessments
- Community Legal Services and Reinvestment Fund, systemic bias in Philadelphia property assessments
- Pew Charitable Trusts, How Property Is Taxed in Philadelphia
- Federal Reserve Bank of Philadelphia, Assessment Frequency and Equity of the Property Tax (WP 21-43)
- IAAO, Standard on Ratio Studies
What this means for your house
A citywide pattern does not decide any single appeal; your own numbers do. But it tells you the odds: if your home is worth less than about $220,000, or sits in one of the ZIP codes above, the data says the coin is weighted against you, and checking costs nothing. The Board of Revision of Taxes must receive appeals for the 2027 assessment by October 5, 2026.
About us, stated plainly: Appeal Squad is a paid service. We prepare BRT appeals for a flat $49.99, never a percentage of savings, and the free check tells you honestly when the numbers do not support an appeal. This analysis is our own, based entirely on public data; we are not affiliated with the City of Philadelphia. It describes the assessment roll in aggregate and is not an appraisal of any individual property.