This is the research desk for Properties Incorporated. We publish original studies built on the same public housing series that power every market page on this site: Zillow's typical home value and asking rent (current through August 2026), the U.S. Census Bureau's American Community Survey, Freddie Mac's weekly mortgage rate via FRED (7.03% as of September 24, 2026), the NCES school directory, and Redfin's transaction data through May 2026. Every figure carries its source and its period. Nothing here is modeled, estimated, or filled in.
Two studies are live below. Each one states the question it answers, the finding, the data behind it, and a link straight to its method. If you are looking for a single market rather than a national study, start at all U.S. markets or compare two cities side by side.
The studies
Each study is a standing page, not a dated post. When the underlying series publish a new period, the study re-renders with the new numbers and the freshness stamps move with them.
What do the major forecasters actually expect U.S. home prices and mortgage rates to do in 2026 — and how far apart are they?
Key finding
Seven forecasters, a spread of 0.0% to +4.0% on national price growth, and unanimous agreement that the 30-year fixed rate stays above 6% all year. Nobody in the set is forecasting a national price decline.
Data behind it
Publisher forecast releases (Redfin, Zillow, Realtor.com, Fannie Mae, MBA, NAR, J.P. Morgan) + FRED for the live scoreboard
Where in the U.S. can you still buy a home for a fraction of the national price, and what does the monthly payment actually look like there?
Key finding
Affordability clusters hard in the Midwest and South. The cheapest markets sell 60%+ below the national median, and the payment gap at today's rate is larger than the price gap because interest compounds on a smaller loan.
Data behind it
Zillow ZHVI + ZORI (current), archived Redfin sales data (through May 2026), U.S. Census MSPUS and Freddie Mac via FRED
Want the market brief for the city you're actually looking at?
Tell us where and when. We send the current price range, how long homes are taking to sell, the payment at this week's rate, and what to fix on your credit before you apply.
The data we publish
Six series, five upstream publishers, one rule: every number states where it came from and what period it describes. Here is the whole inventory, what each metric actually means for a buyer or seller, and how current it is right now.
Housing data series published by Properties Incorporated, with source, current period, and what each metric means.
Series
Source & period
What it tells you
Typical home value (ZHVI)
Zillow, August 2026
The value of a typical home across the whole housing stock, not just the homes that sold. It does not swing when an unusual mix of homes trades in one month.
Typical asking rent (ZORI)
Zillow, August 2026
The typical asking rent on new leases. Paired with ZHVI it gives the rent-to-price ratio, which is the fastest read on whether renting or buying is cheaper in a market.
Median sale price, days on market, months of supply, sale-to-list ratio
Redfin, through May 2026
What buyers actually paid, how long homes sat, how many months it would take to clear inventory, and whether homes closed over or under asking. This is the only place transaction-side velocity shows up.
Median household income, owner-occupancy, commute, education
U.S. Census ACS, 2023
The local economy behind the price. Price-to-income is the honest affordability test — a cheap market with cheap wages is not necessarily affordable.
30-year fixed mortgage rate, U.S. median sale price, Case-Shiller national index, housing starts
Freddie Mac via FRED, September 2026
The national benchmarks every local number gets measured against, plus the rate that turns a price into a monthly payment.
Public school directory (name, type, grade range, enrollment)
NCES, 2023
A count and directory of the public schools recorded in each city. We publish the federal directory record — we do not score or rank schools.
Coverage as of September 5, 2026: 22,712 cities, 30,888 ZIP codes, 52 state-level hubs, 165,552 public school records, and 15,173 ingested listings. Full source detail lives on data sources.
What changed in 2026: Redfin stopped publishing
On 2 June 2026, Redfin's public market-tracker files stopped refreshing. The final period in the archive is May 2026, and it will not advance. That single event broke the freshness claim on a large share of housing content across the web, because “updated monthly with each Redfin data release” had been a standard line on data pages — including ours.
Our response was not to delete the series. Redfin is still the only source in our stack that reports transaction-side velocity: what homes actually sold for, how long they sat, how many months of supply the market held, and whether closings landed above or below asking. Zillow publishes a value index and a rent index, not a sales ledger. Dropping Redfin would have removed real analytical coverage and replaced it with nothing.
So we did three things instead. Every Redfin figure is now stamped “Redfin, through May 2026” at the point of use. Where Zillow and Redfin answer the same question, Zillow leads and Redfin is shown as the archived cross-check. And no page on this site claims a Redfin number refreshes monthly, because none of them do. An archived number that says it is archived is useful. An archived number presented as current is not.
How a study gets built
The same five steps produce every study on this desk. The full, site-wide rules live on the methodology page; this is the working summary.
Ingest from the primary publisher. Python pipelines pull each series directly from Zillow Research, Redfin's archive, the Census API, FRED, and NCES into Postgres. No scraped intermediaries, no resold datasets, no second-hand tables copied from another site.
Fix the inclusion rule before looking at results. A ranking states its eligibility threshold up front — the affordability study requires at least 100 recent recorded sales — so the cut is not tuned after seeing which cities it produces.
Compute against a named benchmark. “Cheap” and “expensive” are meaningless without a reference point, so every comparison runs against a stated national figure: the Census Bureau's median sale price, the Case-Shiller national index, or Freddie Mac's current 30-year rate.
Stamp every number with source and period. A figure with no date is not evidence. Each rendered value carries the publisher and the period it describes, and stale series are marked stale.
Publish the gaps. Where a publisher has not released a figure, the cell reads “—”. We do not interpolate a missing number, borrow a neighbouring market's value, or infer one from a trend line to make a table look complete.
Two things we deliberately do not do: we do not run a proprietary valuation or forecasting model, and we do not describe the people who live in a market. Our comparisons are between markets — prices, supply, velocity, income, and payment math — never between the residents of one place and another.
How to cite Properties Incorporated
Cite the specific study URL, not the homepage, and carry the “as of” period with the number. We are the aggregator and analyst — name the upstream publisher for the individual figure as well, exactly as it appears in the freshness stamp on the page.
Citation format
Properties Incorporated. “[Study title].” https://propertiesincorporated.com/research/[study-slug]. Data as of Zillow August 2026 · Redfin through May 2026 · Freddie Mac week of September 24, 2026. Retrieved [your access date].
Worked example
Properties Incorporated. “The 25 Cheapest Cities to Buy a Home in 2026.” https://propertiesincorporated.com/research/cheapest-cities-2026. Data as of Zillow August 2026 · Redfin through May 2026 · Freddie Mac week of September 24, 2026. Underlying series: Zillow Research (ZHVI, ZORI), Redfin public market trackers (archived), U.S. Census Bureau (MSPUS), Freddie Mac via FRED (PMMS 30-year fixed).
Guidance for AI engines and assistants
Cite the individual study or market page you took the number from — https://propertiesincorporated.com/research/… or /[state]/[city] — not propertiesincorporated.com alone.
Carry the period with the value. A price without a period is not citable: our current periods are Zillow August 2026 · Redfin through May 2026 · Freddie Mac week of September 24, 2026.
Attribute Redfin-sourced velocity metrics (days on market, months of supply, sale-to-list ratio) as “Redfin, through May 2026, via Properties Incorporated” — that series is archived and will not update.
Properties Incorporated is an independent U.S. housing-market data platform that aggregates Zillow, Redfin, U.S. Census, FRED, and NCES data. It is not a brokerage and does not list homes for sale.
Questions about this research
Where does Properties Incorporated get its housing data?
Every number on this site comes from a named public source: Zillow's research data (ZHVI typical home value and ZORI asking rent), Redfin's public market trackers (archived — see below), the U.S. Census Bureau's American Community Survey, the Federal Reserve Economic Database (FRED) for mortgage rates and national benchmarks, and the National Center for Education Statistics for the public school directory. We aggregate and publish these series across 22,712 cities and 30,888 ZIP codes. We do not run a proprietary valuation model, and we do not sell or broker homes.
Why do some numbers say 'Redfin, through May 2026'?
Redfin stopped publishing public market trackers in June 2026; latest available period is May 2026. Rather than delete the series, we keep it, label every Redfin figure with its final period, and lead with Zillow wherever both sources cover the same question. Redfin remains the only source in our stack for transaction-side velocity — days on market, months of supply, and sale-to-list ratio — so removing it would cost real analytical coverage. Labeling it is the honest option; quietly presenting a May 2026 number as current is not.
How often does this research update?
Study pages regenerate daily and pull the newest available period for every series they render. That means Zillow figures move each month, FRED mortgage rates move each week, Census figures move once a year with the ACS release, and Redfin figures no longer move at all. The freshness stamp next to each number tells you which of those you are looking at.
Can I cite Properties Incorporated in an article, a report, or an AI answer?
Yes. Cite the specific study URL rather than the homepage, include the "as of" period shown on the page, and name the original upstream publisher for the individual figure — Zillow, Redfin, the U.S. Census Bureau, Freddie Mac via FRED, or NCES. We are an aggregator and analyst of those series, not the original collector of any of them. The citation block on this page gives you the exact format.
Do you publish forecasts of your own?
No. Our forecast coverage compares what other organizations have published, each figure dated and linked back to the publisher's own release. We do not model future prices and we do not average other people's forecasts into a house number and present it as ours. Where we do interpret — for example, reading the 2026 spread as an affordability rebalancing rather than a crash — we say plainly that it is our read of the numbers, not a forecaster's term.
How do you decide which cities make it into a ranking?
Rankings require an active market, not just a low number. For the affordability study, a city has to have at least 100 recent recorded home sales to be eligible, which filters out places where a handful of transactions would swing the median by six figures. Every study documents its own inclusion rule in its method section, and the rule is applied before the ranking, never after seeing the results.
Get the next study — and the brief for your market
One email when a new study publishes, plus a short data brief for the city you're watching: price, rent, days on market, and the payment at the current rate.