Original research · Updated 2026-07-18

Which counties are most over-assessed?

We analyzed 6,929,104 homes across 32 counties in New Jersey and Texas, using each government's own public assessment roll. About 1,479,526 are assessed above their local norm, an estimated $3.7B a year in excess property tax.

The two methods are reported separately because their numbers are not on the same scale. The comparable-sales method flags a home assessed more than 5% above its matched neighbors; the uniformity method flags a home whose assessment ratio is more than 15% above its county median. They use different thresholds, different comparison sets, and different legal standards, so ranking a Texas county against a New Jersey county in one table would be a false comparison. Within each table the ranking is valid.

Method 1: assessment-ratio uniformity (New Jersey)

In these states the roll carries a statutory fraction of market value, so the test that matters is whether your home is assessed at a higher fraction than your neighbors. Because that ratio does not depend on what a home is worth, it stays a fair comparison across an entire county. Ranked by the share of homes assessed more than 15% above the county median ratio.

Over-assessment is expressed in market-value dollars. Median savings is the median flagged home's annual excess tax, which is lower than the average because the distribution is right-skewed, a handful of badly-assessed homes pull the mean up.

Method 2: neighborhood-matched comparable sales (Texas)

Texas assesses at 100% of market value, so every assessment ratio is 1.0 and the uniformity test above has nothing to measure. Instead each home is compared against its own matched neighbors, the method a Texas appraisal review board actually applies. Ranked by the share of homes assessed more than 5% above their comparables.

Full Texas detail, including per-ZIP breakdowns, is in the Texas Over-Assessment Study.

Methodology

Assessment-ratio uniformity. In New York, New Jersey and Colorado the assessed value on the roll is a statutory fraction of market value, not market value itself. That makes the assessment ratio (assessed value divided by market value) the meaningful test: every home in an assessing unit is supposed to be assessed at the same fraction. We computed that ratio for every residential parcel, took each county's median ratio as the local norm, and counted a home as over-assessed when its ratio sits more than 15% above that median, the "equal and uniform" standard an appeal actually turns on. Because the ratio is independent of what a home is worth, a $200,000 home and a $2 million home are directly comparable on it. A county is ranked only if it has at least 1,000 residential parcels and a tight internal ratio distribution (coefficient of variation below 0.35), which excludes counties spanning multiple assessing units on different statutory ratios. Homes more than 4x the county norm are treated as miscoded parcels and dropped rather than counted. Dollar figures apply each state's effective tax rate to the over-assessment expressed in market-value terms.

Comparable sales. We analyzed every residential property in 14 Texas counties using each county's 2026 public appraisal records, at true market value. For each home we found comparable homes in the same appraisal neighborhood, within 25% of its living area and 10 years of its build year (excluding mobile homes and tear-downs from the comparison), then compared its assessed value to the average of its closest comparables. A home is counted as over-assessed when its assessed value sits more than 5% above that comparable average, the same unequal-appraisal standard Texas appraisal review boards apply under Texas Tax Code 41.43(b)(3). Annual overpayment applies each county's effective tax rate to the assessed gap.

Limits. This study answers which counties skew high, not whether your specific home is over-assessed. County-level medians are a coarser instrument than the per-home comparison behind an actual appeal. Georgia is excluded despite our carrying roughly 592,000 Georgia parcels: its public roll ships no living-area figures, so matched comparables cannot be built for it, and its market-value basis makes the uniformity test inapplicable.

Data and reuse

The full dataset is public under CC BY 4.0 as overassessed-counties.json. Journalists and researchers are welcome to reuse it with attribution to AppealMyTax. Underlying records come from each county's or state's own published assessment roll; ratio-state figures were computed 2026-07-18, Texas figures June 2026.

Related: most overtaxed ZIP codes · Texas over-assessment study · Florida cap gap · is Ownwell worth it?