← Back to Blog
Market Analysis

Richmond Had No Spring Peak in 2026. We Checked 366 Sales.

We lined up five dated scrapes of Richmond closings — 366 comparable suburban 4-bedroom sales from March to September 2026. Mix-adjusted, no month beat April by more than the noise. The seasonal premium sellers are told to wait for didn't show up.

September 28, 2026
8 min read
By Raam RVA Research Team
Market TimingSeasonalityPrice Per Square FootSellingBuyingData AnalysisMethodologyChesterfieldHenricoRVA-Wide

Richmond Had No Spring Peak in 2026. We Checked 366 Sales.

A 4-bedroom house in Midlothian or Glen Allen that closed in April sold for a median $219 per square foot. The same kind of house closing in late August sold for $221. In between, June and July came in at $217–$219, and late July into mid-August dipped to $214. Five windows, five months, a 3.2% spread from top to bottom — and none of it outside the margin of error.

That's not how the story is supposed to go. The standard advice is that spring is when buyers flood in and prices climb, and that sellers who miss it leave money on the table. National data still shows a version of that pattern this year. In Richmond's core suburban 4-bedroom market, the closing records don't.

Comparable sales analyzed
366
5 dated Zillow scrapes, closed 2026-03-30 → 09-04
Median $/sqft, April closings
$219
n=93
Median $/sqft, late Aug–early Sep
$221
n=34
Largest mix-adjusted move vs. April
+3.4%
95% CI −1.1% to +8.2% — not distinguishable from zero

VERIFIED — The Core Finding

Across 366 single-family sales (4+ bedrooms, 2+ baths, ≥2,250 sqft, lot ≥0.25 acre, $500K–$850K) inside a common Midlothian–Henrico–Mechanicsville map box, median price per square foot stayed between $214 and $221 in every closing window from March 30 to September 4, 2026. A regression controlling for size, lot, bedrooms, baths, and city found no window statistically different from April. Sources: data/sources/ Zillow sold scrapes dated 2026-04-27, 06-12, 07-12, 08-16, and 09-05.

What We Measured — and Why It Took Five Files

This site has been scraping Richmond-area closings from Zillow since spring. Each pull answered a different question at the time, so each used different search filters: one capped monthly payment at $6,000, another required four bedrooms and a quarter-acre lot, another covered any 3-bedroom home. Line those up naively and you'd be measuring our filter changes, not the market.

So we cut every file down to the homes all five searches would have caught: detached single-family, 4+ bedrooms, 2+ baths, at least 2,250 square feet, at least a quarter-acre lot, sold for $500,000 to $850,000, inside the smallest map box every scrape shared. That box runs roughly from Midlothian and Chesterfield north through western Henrico, Glen Allen, and Mechanicsville, and takes in the city's western edge. We removed any home that showed up in two overlapping pulls. That left 366 sales.

Closing windowApprox. contract month (ASSUMED)SalesMedian priceMedian sqftMedian $/sqft95% range
Mar 30 – Apr 24Late Feb – Mar93$625,0002,837$219$208–$232
Jun 5 – Jun 12Late Apr – early May30$655,0003,060$217$197–$225
Jun 12 – Jul 10May – early Jun120$650,4002,930$219$214–$226
Jul 17 – Aug 14Mid-Jun – mid-Jul89$635,0002,961$214$204–$222
Aug 24 – Sep 4Mid – late Jul34$700,0003,070$221$205–$231

Notice the median price column moves more than the $/sqft column. Late August's $700,000 median looks like a late-summer surge until you see those homes were also the biggest. Price per square foot, the number that controls for size, barely moved.

ASSUMED — Closing Dates Aren't Contract Dates

A closing records a deal that was struck weeks earlier. We assume a typical 30–45 day gap between ratified contract and closing, so April closings mostly reflect late-winter negotiations and late-August closings mostly reflect July contracts. The finding still holds under that translation: contracts written in late winter and contracts written in peak summer produced the same price per foot.

Size and Location Don't Explain It Away

Medians can hide a lot. If spring buyers happened to buy smaller houses in pricier ZIPs, the flat line could be a coincidence of mix. So we ran a regression: log sale price on log square footage, log lot size, bedrooms, baths, and city, with a separate term for each closing window. It explains about 46% of the variation in price. None of the windows came out different from April by a statistically detectable amount.

Window vs. April (mix-adjusted)Estimated difference95% confidence interval
Jun 5 – Jun 12+0.5%−4.0% to +5.3%
Jun 12 – Jul 10+1.9%−1.2% to +5.1%
Jul 17 – Aug 14−0.2%−3.5% to +3.3%
Aug 24 – Sep 4+3.4%−1.1% to +8.2%

Two more checks gave the same answer. Split the panel by side of the river and each side is flat on its own: Chesterfield-side cities (Midlothian, Moseley, North Chesterfield, Chesterfield) ran $206, $203, $211, $205, $203 per square foot across the five windows. Henrico, Hanover, and city-side ran $233, $229, $231, $224, $229. Then we loosened the filters to any 3+ bedroom home from 1,750 square feet and $200,000 to $1.2 million. Only three windows allow that cut, but it covers 661 sales: $219, $221, $220.

VERIFIED — A Steady North-of-the-River Premium

The one gap that held all season wasn't between months. It was between sides of the river. Comparable homes in Henrico, Glen Allen, Mechanicsville, and the western city sold for $19 to $27 more per square foot than those in Midlothian and North Chesterfield in every window, from April through early September.

What 'Flat' Does and Doesn't Mean

Flat doesn't mean nothing happened. It means that for this segment, no move got bigger than a few percent, and the data can't tell which way the small moves went. The widest interval, late August, runs from −1.1% to +8.2%. On a 3,000-square-foot house, that's a range of roughly $650,000 to $710,000 at April's median price per foot. A real seasonal swing smaller than that could exist and we wouldn't see it.

It also doesn't mean the national pattern is wrong. S&P Cotality's Case-Shiller national index rose 0.4% in June before seasonal adjustment and 0.1% after it. The seasonal factor was still pushing prices up nationally this summer. Richmond's broader numbers don't point to a stall either: Central Virginia Regional MLS reported a July single-family median of $460,000, up 2.2% from a year earlier, with 1.8 months of supply. Those are metro-wide figures across every price point. Ours is one slice, and that slice didn't follow the calendar.

Where This Doesn't Apply

This is the upper-middle suburban 4-bedroom market: $500K–$850K, quarter-acre-plus lots, the Midlothian–Henrico–Mechanicsville band. It says nothing about condos, townhouses, starter homes under $500K, the city core (the Fan, Church Hill, Northside), or luxury homes above $850K. Those segments may be more seasonal. We couldn't test them with comparable filters across all five scrapes.

The Data Limits We're Putting Up Front

Three caveats, in order of how much they matter. First, the five scrapes used different search settings. We kept only the homes every setting would have caught, but the scrapes can still differ in ways a filter can't see, like how Zillow sorted results before the scraper stopped. Second, the two smallest windows (early June, n=30; late August, n=34) carry the widest error bars. Third, our automated refresh failed this week, so the newest closings in this piece are from September 4, 2026. We'll re-run the panel when fresh September and October data lands and publish the result either way.

What This Means If You're Buying or Selling This Fall

If you're a seller in this segment who missed spring, the 2026 closing data doesn't show a penalty for listing now. The houses that closed in late August sold for the same price per square foot as the ones that closed in April. Price against recent comparable sales, not against a calendar. If you're a buyer who has been waiting for a fall discount, this data doesn't show one either. What you might find in fall is less competition, not a lower price per foot.

Either way, the more reliable number this season was geography: $19 to $27 a square foot between the north and south sides of the river, steady from April to September. For more on reading that number carefully, see our look at what a bigger sample did to our June price-per-square-foot findings, the July 2026 Market Pulse, and why assessed value is a poor stand-in for market value.


Data: 366 single-family sales closing 2026-03-30 through 2026-09-04, from five Zillow sold scrapes in this repository: data/sources/sf-rva-area-sold-30d-2026-04-27.csv, rva-recent-sold-2026-06-12.csv, rva-recent-sold-2026-07-12.csv, rva-sf-4bd-2250sf-sold-30d-2026-08-16.csv, and rva-recent-sold-2026-09-05.csv. Harmonized filter: 4+ bd, 2+ ba, ≥2,250 sqft, lot ≥10,890 sqft, $500K–$850K, lat 37.385–37.722 / lon −77.727 to −77.396, de-duplicated by Zillow ID. Medians with 5,000-draw bootstrap 95% ranges; hedonic OLS with heteroskedasticity-robust (HC1) standard errors. Robustness slice: 3+ bd, ≥1,750 sqft, $200K–$1.2M, lot ≥1,000 sqft (n=661, June 12, July 12, and September 5 scrapes). External context: S&P Cotality Case-Shiller June 2026 release as reported by Calculated Risk (2026-08-25); CVR MLS July 2026 figures as reported by Providence Hill Real Estate. Reproduction script: docs/editorial/newsroom/2026-W39/analysis.py.

About the Author

Raam RVA Research Team · Investigative Analysis

Guide updates

Get corridor updates by email

Pick one or more relocation guides. We email when that guide is meaningfully updated, plus a monthly market digest scoped to its ZIPs. Prefer feeds? Atom.

Corridors

Confirmation link per corridor. Monthly ZIP digest + update emails. One-click unsubscribe.