How it works

The peer group

Every parcel is compared against others on the same plat, of the same decade of construction, and of the same property class. The comparison is assessed value per square foot of building.

Property class is in the group for a reason that cost us a wrong answer first: without it, a commercial building sharing a plat with houses was measured against the houses and came out 978% above its peers. Douglas classes run from about $71 to $188 per square foot with median building sizes from 1,436 to 11,778 square feet, so mixing them produces numbers that look enormous and mean nothing.

Percentage alone is not enough

A parcel 40% above its peer median is a different claim in a tight group than in a scattered one. So every case also carries how many interquartile ranges it sits above the median — a measure of how unusual it is given how much that group varies. We only treat a case as filing-grade at 1.5 interquartile ranges or more, on top of 30% above the median.

Where a group has almost no spread, the ratio stops meaning anything — dividing by a near-zero spread once produced a parcel reported at 148,102 interquartile ranges. Groups with an interquartile range under $5 per square foot report nothing rather than something enormous.

Comparables, and how they are picked

Five named properties, chosen nearest the group median and within 30% of the subject’s building area — not the lowest-assessed five. That matters to whoever is on the other side of the table: a packet built from the cheapest comparables available is the first thing an assessor will take apart. Measured across every packet, the comparables sit a mean of 0.665% from their group median.

The equalized value is the group’s median per square foot applied to the subject’s area, not the median of those five. The named comparables illustrate the group; they do not set the figure.

We will not sell a packet with fewer than three named comparables, even though the underlying comparison still stands. The peer median is a measurement; a document you can put in front of a board needs evidence with addresses on it.

Where the method declines

• Build quality. Where a subject is graded above its comparables, part of the difference is explained by the building being better, and the document says so on its face rather than leaving a board to find it. In Douglas that check removes a fifth of otherwise-strong cases. Sarpy publishes no quality rating at all, so the check cannot run there.
• Part-built subdivisions. Groups whose median is under $50 per square foot average a year built of 2024: the assessor is holding land and partial improvements, not finished houses. A finished house cannot be equalized against them and the arithmetic cannot tell, so those groups are excluded.
• Agricultural land. Nebraska values greenbelt by agricultural use rather than market value, so assessed value per square foot of building does not describe how it was valued. Those parcels get a review, not a packet.

What the document is not

It is an equalization analysis from public records, and it says so on its face. It is not an appraisal, it states no opinion of market value, and nobody here is a licensed appraiser. The estimated saving is the share of assessed value removed applied to the tax actually billed — an estimate, not a determination.