How many homes are over-assessed? We measured it in two states
We compared every property against its closest neighbours in Illinois and Connecticut. About 1 in 12 sits far enough above its peers to be worth an appeal, and the median gap is 22%.
Everyone with a tax bill wonders whether theirs is too high. The question is answerable, because uniformity is a legal standard rather than an opinion: similar properties in the same area should carry similar assessments per square foot. Where one does not, that is the case.
So we ran it. Every property we hold in four Illinois counties and all 169 Connecticut towns was compared against its closest peers — same town or township, same neighbourhood code where the county publishes one, similar building size, similar age — and we measured how far each sits above the peer median.
The result
Roughly one property in twelve sits far enough above its own peer group to be worth the filing. When one does, it is typically about a fifth above the neighbours — not a rounding error, and not a fortune either.
Why the agreement between the states matters
The two numbers were produced by almost nothing in common. Illinois assesses at 10% of market value in Cook County and a third elsewhere, then applies an equalisation factor. Connecticut assesses at 70% statewide under CGS 12-62a with no equaliser at all. Different rolls, different collection agencies, different revaluation cycles, different data formats, different peer groupings.
They landed one percentage point apart.
Three independent measurements, three methods, one answer around a fifth. That is the kind of convergence that makes a number worth trusting.
What the bar actually is
"Over-assessed" needs a definition or the figure means nothing. Ours has four conditions, and a property must clear all of them.
| Condition | Threshold | Why |
|---|---|---|
| Peer group size | ≥ 20 properties | Below that, the median is noise |
| Gap floor | ≥ 10% | Under 10% is inside normal valuation tolerance |
| Gap vs group noise | ≥ 0.75 × spread | A wide group needs a wider gap to mean anything |
| Upper bound | ≤ 100% | Above that it is a record error, not a valuation dispute |
The third condition is the one that does the real work. In a tight subdivision where every house is within 8% of the median, a 15% gap is a genuine outlier. In a mixed neighbourhood where the spread is 40%, the same 15% is unremarkable. A single fixed threshold would flood you with cases from the second kind of street and miss the real ones on the first.
What we throw away
The upper bound deserves an explanation, because discarding the biggest gaps looks like discarding the best cases. It is the opposite.
We found properties assessed 493%, 608% and 1,082% above their neighbours. Nobody is over-assessed elevenfold. Those are commercial parcels in a residential group, a condo whose building total was attached to one unit, or a square-footage field with a transposed digit. Taking a 1,082% gap to a hearing as a uniformity argument wastes everyone's afternoon; the fix is a correction request to the assessor's office, which is a different and usually much easier conversation.
Where the data itself failed
Two of the four Illinois counties are excluded from the figures above, and the reason is worth stating.
Lake County's building square-footage field, as published, has a median of 9,821 and a 90th percentile of 105,734. Those are lot sizes, not houses. Every uniformity calculation built on that column would be arithmetic on the wrong quantity — producing confident, precise, entirely meaningless results. DuPage failed the same check.
What it means for you
The prior is against you, and that is fine. About 92% of properties are assessed close enough to their neighbours that no uniformity argument exists. Checking costs nothing; assuming costs an evening.
If you have a case, expect about a fifth. Not the halving that appeal advertising implies. On a typical Cook County bill a 20% assessment reduction is somewhere around $1,300 a year — real money, modest expectations.
An absurd gap is good news, not a big case. If your property looks 400% above its neighbours, something in the county's file about your property is wrong. Ask for the property record card and check the square footage against your own survey. Correcting a record error is faster, more certain, and often worth more than any uniformity appeal.
Uniformity is not the only argument. It is simply the one that can be measured at scale from public data. A recent purchase below the county's implied value, a professional appraisal, or a condition problem the assessor has not seen are all valid and none of them appear in this analysis.
Sources
Cook County assessed values (uzyt-m557), Cook appeal decisions (y282-6ig3),
Connecticut statewide CAMA for 2024 and 2025 (pqrn-qghw, rny9-6ak2), and county
parcel services for Lake, DuPage and Peoria. All public, all downloadable, all checkable.