What I’m thinking about: Two of the biggest data providers in housing just published opposite headlines about the same month…and the 366-year-old Latin motto I use to read past BOTH of them.
Realtor.com and Redfin just told opposite stories about the exact same housing market.
In the middle of our own 2-month, 30% reprice, I flagged Realtor.com’s June 2026 report (asking prices logging their steepest annual decline on record) and promised the full breakdown of the nuance behind that headline.
The lens for all of it comes from 1660, when the founders of the Royal Society (the oldest scientific society on Earth) picked a motto I think about constantly…nullius in verba. “Take nobody’s word for it.” Not the data providers’, not the AI’s, and not mine (everything below is fact-checkable, and I’d encourage exactly that).
Every report is selling something
I genuinely like the data behind Realtor.com, Redfin, and Zillow. They ingest raw MLS feeds, the highest-quality real estate data anyone could ask for.
Their revenue also rides on an active market…on-market listings and the agents representing them. Nobody there is fabricating numbers (incentives work more subtly than that, perhaps influencing which stat gets the headline and which gets buried in a methodology note). Wall Street runs the same playbook at larger scale…two months before Enron collapsed, 16 of the 17 sell-side analysts covering it still rated the stock a buy (a pre-reform era, when the analyst and investment arms of banks were more intertwined…and that regulation has since been abolished, so keep your eyes open FYI…SpaceX stock analysis anyone?).
So when a June 2026 housing report leads with “After Years of Waiting, Buyers Are Getting Their Summer,” my antenna goes up. Time to check the numbers underneath.
Relative vs. absolute…the first move I stole from medicine
Back in my medical training, EVERY drug study got the same interrogation. For example, a “30% reduction in disease incidence” sounds enormous…until you learn the absolute rate fell from 1 case per 1,000 people to 0.7. Relative numbers give you the direction, while absolute numbers give you the size of what actually happened.
Now apply that to the June report’s headline stat…pending sales up 3.7% year-over-year, the seventh straight month of growth. Certainly a bullish trend we’d all cheer for.
However, the absolute picture is brutal. Existing-home sales are running around 4.09M annualized (per NAR’s June report, down 2.4% from May), sitting at, and by some prints below, the trough of the Global Financial Crisis, and we’ve been flat on that bottom for ~4 years, where the GFC bounced within ~18 months. The country also holds ~38M more people than it did in 2008 (~342M now versus ~304M then), so per capita, this is the least active housing market in modern American history.
(Credit where it’s due, though. Realtor.com steel-manned its own bull case, showing contract cancellations at 6.9% of pending sales versus 7.3% a year ago…homes going under contract are staying there. That statistic belongs in the analysis.)
Another milestone deserves the same deeper look: “Time on market ends a 26-month streak” of year-over-year slowing…and the actual number is 53 days, identical to last June. Dig one layer down and the Northeast (-2 days) is doing all the lifting while the Midwest and West got slower. A streak ended…but the level didn’t move an inch.
Record declines AND record highs…both true at once
The two providers collide hardest on price. Realtor.com reports asking prices down 2.5% year-over-year, the steepest drop in their data’s history (which, notably, only reaches back ~10 years…no GFC, no early 2000s in the sample). Redfin, the same week, reports the median SALE price at $408,838…a record high.
Both are correct.
Remember from middle-school math that the median resists outliers (I still picture crossing numbers off both ends of the list until you land in the center). A handful of monster sales can’t drag it up the way they’d wreck an average. (Even the 8-figure homes are softening…NBA star Anthony Davis sold his Bel Air estate for $32M after listing at $39.9M, ~28% above market value, then closing ~20% below it ~11 months later, and after LA’s mansion tax, likely a net loss on his 2021 purchase.)
What CAN move a median is a change in which homes sell at all, and that’s exactly what’s happening. The cheaper the home, the less it’s transacting…the pricier the home, the more it is. Wealthier and older buyers are the ones still closing, which pulls the median sale price up while the median list price falls. San Francisco (+10.8%, riding the AI wealth machine) and West Palm Beach (+10.6%) lead sale-price gains, while Seattle (-5.3%) and San Jose (-4.0%) lead the declines…sometimes a single bay apart.
This is the K-shaped economy mapped directly onto housing. One dataset measures what sellers are asking, while the other measures what a shrinking, richer pool of buyers is actually paying.
The spring wave that never formed
In a normal year, the spring wave is enormous…from 2013 to 2019, sales at the June peak averaged ~75% higher than the winter trough, and February to March alone typically jumps ~33%. This year, the wave barely formed (only about a 9% seasonal bump). NAR just logged the slowest March since 2009 (3.98M annualized, DOWN 3.6% from February, in what should be the seasonal liftoff), making 2026 the fourth dud spring in a row. When even the spring selling season can’t move this market, the absolute-number story from earlier gets its confirmation.
What the quieter numbers say
(By quieter, I mean not leading the headlines.) But a LOUD number by implication is that active inventory hit 1.1M in June 2026, up 92.2% since June 2022. That near-doubling in four years is, to me, the single most telling stat in the report…though it still sits ~11% below typical pre-pandemic levels (contributing to a vicious cycle of not enough sellers being open to listing their home, and buyers priced out from purchasing).
Months of supply looks tame at 4.6 for existing homes per NAR’s June report, but that figure leans on a large percentage of owners locked into ~3% mortgages who aren’t incentivized to sell.
New construction, the segment I treat as the leading indicator, is carrying 9.3 months of supply per Census data released this week (a normal range runs ~4-6), with land-banked lots stacked even deeper behind it (the permitted-but-not-yet-started pipeline is at an all-time high). Builders can’t wait out a market the way a locked-in homeowner can, which is why I watch them (and how they adjust pricing) first.
75% of the country is priced out
The demand side goes beyond any single monthly report. Per a Bankrate analysis (published Dec 2025, updated this January 2026), the typical American household earns ~$80K, while affording the median-priced US home takes ~$113K. The result…over 75% of homes on the market are unaffordable to the typical household, when a balanced market would put that figure closer to 50%.
Metro by metro, the map is stark. Pittsburgh (54.6% of listings affordable on a local median income) and St. Louis (49.7%) are essentially the only balanced major markets left. Then it falls off a cliff…San Francisco 7.3%, New York 4.9%, Boston 4.8%, San Diego 1.6%, Los Angeles 0.5%, and Miami at 0.4%, where roughly 1 in 250 listings is affordable to a typical local household.
The consequences show up in who’s actually buying. First-time buyers made up just 21% of purchases (the lowest share since NAR began tracking in 1981), and the median first-time buyer is now 40 years old. (The new federal ban on institutional investors owning 350+ homes, which became law by default on July 11, 2026 might help affordability at the margins, but large investors are a small slice of total ownership, so I’d expect this ban to take years to show up in actual pricing, if it ever meaningfully does.)
The dream, and the math
Even priced out, Americans haven’t stopped wanting in. Per Prof G Markets, 56% still say owning a home represents the American Dream, and 93% believe a home is a better investment than stocks.
The math currently disagrees. A Moody’s comparison ran an average renter (investing the monthly difference in the market) against an average homeowner across 30 years, and the renter finished ahead ~$2.8M to ~$1.6M (pre-tax, with plenty of assumptions doing work in there). My own stance hasn’t changed…a primary residence is shelter and stability, and I don’t count my own house as an investment asset (besides the ability to utilize a HELOC).
For the (non-speculative) land investors reading this, that distinction is the whole business. We buy dirt to sell it, underwritten on comps and absolute demand…never on the belief that prices simply go up.
Read it like a statistician
You could paste any of these reports into an AI model and ask it to strip out the spin…I do a version of this constantly with my Learning Accelerator skill (a custom workflow I built to compress long-form material into decision-ready notes). Just remember the model is drinking from the same PR wires the spin arrived through, so it needs your guidance on the incentives at play and your judgment on what’s missing. The critical-thinking muscle stays yours.
The habits are simple…ask who benefits from the framing, convert every relative number into an absolute one, and check how a metric is defined before comparing it to anything. The operators doing that right now are reading a market that is genuinely brutal YET understandable (the data quantity and quality have never been higher), which is precisely the setup where discipline and liquidity get paid.
366 years later, the Royal Society’s founding rule is still the whole game. Nullius in verba…take nobody’s word for it. Not realtor.com’s, not Redfin’s, and not the AI’s.
If you’re an experienced operator with routine deal flow looking for a capital partner that underwrites the data as hard as the dirt, reach out via the Serious Land Capital URL in my author bio below. We write checks from $50K+, we close 100% of the deals we commit to, and our national underwriting was built by digging for the real story in every market we’ve ever entered.
