13,593 Cook County homes are assessed above their own neighbours, and the window to say so closes September 1

An analysis of the 2026 Cook County assessment roll across all seven townships currently open for appeal.

The finding

Across the seven Cook County townships whose Board of Review appeal window is open right now, 13,593 single-family and townhome parcels are assessed higher per square foot than the comparable homes around them, by a margin between 5 and 50 percent, on comparable groups tight enough to meet the assessment industry's own uniformity standard.

Together those parcels carry $134,842,277 in assessed value above what their own comparables indicate. At the Assessor's published 10x convention for residential property, that is roughly $1.35 billion of implied market value.

That is 22.7 percent of the parcels we were able to score, or a little better than one in five homes with usable data.

The median flagged home is assessed 9.8 percent above its comparable group. Taken separately, the median gap in dollars is $6,552 in assessed value, roughly $65,500 in implied market value. Those are two independent medians of the same list, not two measurements of one house.

The Board of Review stops accepting complaints for all seven townships on Tuesday, September 1, 2026. Evidence is due September 11.

The seven townships are not a sample. They are the whole open field.

This is the part most homeowners will not know.

The Cook County Board of Review publishes a schedule of which townships may file when. Its 2026 session schedule (PDF, dated July 31, 2026) lists exactly one group of townships as open:

Group 1: Evanston, New Trier, Norwood Park, Oak Park, River Forest, Riverside, Rogers Park. Open for filing complaint 8/3/2026. Closed for filing complaint 9/1/2026. Evidence submission deadline 9/11/2026.

Those seven are the entire set of Cook County townships currently able to file a Board of Review complaint. They are also, exactly, the seven townships in this study. No other township appears on the Board's open list.

Why an owner might reasonably believe the window is shut

Cook County runs two sequential appeal stages: first the Assessor's Office, then the Board of Review. They are separate filings with separate deadlines.

On the Cook County Assessor's Assessment and Appeal Calendar, the headline status label for all seven of these townships currently reads "Closed For Appeals." That label is accurate and refers to the Assessor's own window, which closed for these townships between May 26 and June 22, 2026. The Board of Review dates, which are open, appear one level down inside each township's collapsible "View Details" panel.

Source: Cook County Assessor's Office, Assessment and Appeal Calendar, retrieved August 20, 2026, and Cook County Board of Review 2026 session schedule.

An owner who checked the Assessor's calendar this week, saw "Closed For Appeals" and stopped reading would have no way of knowing that a second, live filing window closes in days. We make no claim about intent. We note only that the two-stage structure is easy to miss, and that the label a homeowner sees first says closed while the deadline that matters is open.

Township breakdown

A further 1,650 parcels show a gap but were held back into a secondary tier because the evidence is weaker. They are counted nowhere in the figures above. See limitations.

How the gaps are distributed

Most flagged homes are marginal cases, not outliers. Half sit in the smallest band.

Bands are half-open (5 to 10 means 5 up to but not including 10), except the last, which includes 50 exactly. They sum to 13,593.

Methodology in plain language

What an Illinois residential appeal actually argues

Illinois is a uniformity jurisdiction. A residential Board of Review complaint does not primarily argue that a home's market value is wrong. It argues that comparable properties are assessed lower than this one. The evidence of record is therefore the comparables' assessed values, not sale prices.

That is what this study measures, and it is the only thing it measures. Every figure here is in assessed value dollars. Nothing here is a market-value opinion or an appraisal.

The source data

Every value comes from Assessor - Assessed Values (dataset uzyt-m557), published by the Cook County Assessor's Office Data Department on the Cook County Open Data Portal. Building characteristics come from the Assessor's published characteristics file.

Every row in our output carries a value_roll_year column stating which roll its value came from. In these seven townships that value is 2026 on 100 percent of rows. Nothing is computed off a prior year.

How a comparable group is built

• in the same Assessor neighborhood code (nbhd, which is scoped to a township)
• within ±25 percent of the subject's building square footage
• within ±10 years of the subject's year built
• assessed at $25,000 or more, with a recorded building area above zero
• not the subject itself

Candidates are ranked by closeness on size, age and property class, and the nearest ten are kept.

We then compute the subject's assessed value per square foot of building area and compare it to the median assessed value per square foot of its comparable group. The difference is the uniformity gap. The dollar figure is what the subject's assessed value would be at the comparable median, subtracted from its current assessed value.

The quality gate, and why it is there

A percentage gap alone is not evidence. It is a starting point that has to survive a test.

We hold no view, waterfront or location attribute for any parcel. The first build of this analysis put Sheridan Road and Lakeside Terrace, the Lake Michigan shoreline in Kenilworth and Winnetka, into 8 of the top 10 results by dollar gap. Those were lakefront houses being measured against inland houses. The gap was a missing attribute, not an assessment error.

What separates most of those cases is measurable without a view field: their comparable groups disagree with each other. Seven of those eight rows carried a comparable spread of 0.75 to 1.27 times the median, against a list-wide median of 0.29 times. The eighth, 381 Sheridan Road in Winnetka, had a spread of 0.55 and passed the gate. It is in the results. We flag it here because it is the clearest example of what the gate does not catch: a parcel can have a tight comparable group and still carry a physical attribute no public field records.

So a parcel is only included if its comparable group's coefficient of dispersion is 15 or below. COD is the standard IAAO uniformity statistic: the average absolute deviation from the median, expressed as a percentage of the median. The IAAO Standard on Ratio Studies holds 5 to 15 as the acceptable band for single-family residential. A subject whose own comparables disagree by more than that is not evidence of over-assessment. It is evidence of a heterogeneous neighbourhood.

The 13,593 parcels reported here have a median cohort COD of 6.3. Every one of them is at or below 15.

The inclusion rules, stated completely

1. Uniformity gap between 5 and 50 percent
2. At least 5 comparables found
3. Comparable cohort COD at or below 15
4. Property class in 202, 203, 204, 205, 206, 207, 208, 209, 210, 234, 278, 295, 299 (single-family and townhome)
5. Value roll year 2026

Below a 5 percent gap nothing is listed. That is inside the noise of any mass-appraisal model and not worth a filing fee.

Multi-family (class 211) and mixed commercial-residential (class 212) are excluded from every figure in this study. They are income property and are appealed on different evidence entirely. They are written to a separate file and counted nowhere here.

Verification against live county records

Thirty-five parcels, five per township, spread evenly through each list at a fixed stride, were re-checked against the Assessor's live endpoint on August 20, 2026. Each township file is sorted by dollar gap, and the stride starts at row 1, so five of the 35 are the largest gap in their township.

• 35 of 35 subject assessed values match the live 2026 roll exactly
• 35 of 35 building areas match the live characteristics file
• 105 of 105 sampled comparable PINs resolve to a live 2026 value

State that last one precisely: 105 comparables is three of the ten comparables on each of 35 parcels. That is about 0.26 percent of the 13,593 parcels and a smaller fraction of the comparable set. It establishes that the pipeline reads the county's values correctly. It is not a validation of all 13,593 rows, and we do not present it as one.

Worked examples, checkable in five minutes

Each of these can be verified by entering the PIN at the Assessor's property search. All figures are public record.

Each has ten comparables, all in the same property class, all in the same Assessor neighborhood.

What this study does not claim

This section is as important as the finding. Please read it before citing.

1. It does not claim any of these assessments is wrong. It claims each of these parcels is assessed above the median of its own comparable group by a defensible margin, on a comparable group tight enough to meet the IAAO uniformity standard. That is a prima facie uniformity argument. Whether it prevails is for the Board of Review to decide, and the Board sees attributes we do not.

2. It does not predict any outcome, refund or tax saving. We do not model tax rates, exemptions, equalization factors or levies. Assessed value is not a tax bill.

3. It is not a complete census of over-assessment. The condominium gap is real and it is large. A comparable grid cannot be built without a recorded building area, and Cook County records many condominiums without one. Coverage of building square footage by township:

Read that plainly: the Rogers Park and Evanston results cover single-family and townhome stock well, and condominium stock badly. The missing parcels are silently absent, not counted as unaffected. The true Rogers Park figure is unknown and is very likely higher than the 1,045 reported here. This is a limitation of the published characteristics data, not a modelling choice, and it is not fixable from the Assessor's own file.

4. It holds no view, waterfront, corner-lot, condition or renovation attribute. The COD gate is a proxy for this, and a good one, but it is a proxy. A parcel can pass the gate and still have a physical reason for a higher assessment that no public dataset records.

5. The 1,650 secondary-tier parcels are excluded from every headline number. Those are parcels showing a gap that is extreme (above 50 percent), thin (only 3 or 4 comparables) or scattered (COD above 15). They are published separately and are not part of the 13,593 or the $134.8 million.

6. This is horizontal equity, not vertical equity. Existing academic work on Cook County, notably by Christopher Berry at the University of Chicago Harris School, examines whether lower-value homes are assessed at higher effective rates than higher-value homes. That is a different question. This study asks only whether a given home is assessed in line with homes very much like it, nearby. The two findings are complementary and neither tests the other.

7. AppealMyTax has a commercial interest in property tax appeals. We sell comparable-property data and evidence packets. We do not file appeals, we do not represent owners, and we take no percentage of any outcome. The methodology, inclusion rules, source dataset and verification counts are published above precisely so that this finding can be checked by someone who assumes we are motivated. Please check it.

Download the data

The full per-parcel lists are published below as static CSV files, ungated, under CC BY 4.0. No email, no form, no conditions. Questions and corrections: john@revxl.net.

• All townships — Tier A (13,593 parcels) (5.3 MB)
• All townships — Tier B / secondary (1,650 parcels) (0.6 MB)
• Multi-family & mixed-use (classes 211/212, excluded from all figures) (0.4 MB)
• Evanston — Tier A (1.0 MB)
• New Trier — Tier A (1.7 MB)
• Norwood Park — Tier A (0.6 MB)
• Oak Park — Tier A (1.0 MB)
• River Forest — Tier A (0.3 MB)
• Riverside — Tier A (0.4 MB)
• Rogers Park — Tier A (0.4 MB)
• Data dictionary and build notes (6 KB)
• Live-roll verification record (35 parcels, Aug 20, 2026) (8 KB)

One note for anyone auditing the per-parcel files: the published comp_cohort_cod and uniformity_gap_pct columns are rounded to one decimal place for display. Tiering uses the unrounded values. A handful of secondary-tier rows therefore print as 15.0 or 50.0 and look as though they should have qualified. They did not; their true values sit just above the threshold. If you are checking the inclusion rules against the CSV, check against that.

Reproducing this

• Source dataset: Cook County Open Data Portal, uzyt-m557
• Deadline source: Cook County Board of Review, 2026 session dates
• Uniformity standard: IAAO Standard on Ratio Studies
• Underlying per-parcel data: the CSV files above, free to download and reuse with attribution.

Every number in this study is derived from a public dataset using rules stated in full above. If you find an error, tell us and we will correct it in place with a dated note.

Corrections

None to date. This section stays here whether or not it is ever needed.

Attribution

Free to quote, reproduce and republish. Please cite as:

AppealMyTax, "13,593 Cook County homes are assessed above their own neighbours" (August 20, 2026).

For homeowners in these seven townships

You can file a Board of Review complaint yourself, for free, at appeals.cookcountyboardofreview.com. You do not need to pay anyone, and you do not need us. The deadline is September 1, 2026, and evidence is due September 11, 2026.