Data through September 2026. Updated quarterly.

This report covers 33,137 closed home sales inside Fort Worth city limits over the three years ending September 2026 — every arm's-length residential closing the MLS recorded, not a sample. The citywide median sale was $339,990 on 1,918 square feet, closing in 32 days at 97.0% of original list price.

I'm a Fort Worth broker and a Certified Residential Appraiser. Everything below is measured rather than asserted, and where a finding doesn't hold up I've said so.

The headline: Fort Worth is not one market, and the two halves are moving apart

Citywide, values are down about 0.70% a year once you control for house size, lot size and vintage. That is a real result, and on its own it is close to meaningless — because it averages together two parts of the city doing different things.

Measured ZIP code by ZIP code, across the 31 with enough volume to test:

Sixteen are declining at conventional statistical significance. Exactly one is rising.

Trend is the annual change in sale price controlling for house size, lot size and build year. "n.s." means not statistically distinguishable from zero.

The pattern is not subtle. The established central neighborhoods are flat. The outer ring is falling. Arlington Heights, Ridglea, Fairmount, Westcliff and Benbrook all sit within a percent or so of zero. Everman, southeast Fort Worth, Aledo, east Fort Worth and Justin are down three to five percent a year.

What actually predicts the decline

Three explanations suggest themselves, and they don't survive equally.

I tested all three against each ZIP's measured trend: how far it sits from downtown, its share of new construction, and its price level.

Builder supply is the explanation everyone reaches for, and it doesn't hold. The raw correlation looks convincing until you account for the fact that new construction is mostly built far out. Once distance is in the model, new-construction share stops explaining anything.

What survives is price level, and it's the strongest term. Cheaper ZIP codes are falling faster. That is an affordability result rather than a supply one: at the bottom of the market a rate move eats a much larger share of the payment, and the buyer pool thins accordingly.

Distance survives marginally, at roughly −0.13% per year for each mile from downtown. With only 31 ZIP codes that sits right at the edge of significance, so I'd call it suggestive rather than established.

The nine-mile line

This is the finding I'd put in front of anyone buying on the edge of the city.

Take every Fort Worth property that sold twice inside the three-year window — owner to owner, with builder sales stripped out and houses that were enlarged removed — and measure whether the second sale beat the first.

It is not a gradient. It is a cliff. Loss rates sit between 15% and 17% all the way out to nine miles, then roughly double and stay there. Median gain does the same thing — above 24% inside the line, under 2% beyond it.

Nine miles is approximately where Loop 820 runs, give or take by direction.

Put plainly: if you bought beyond about nine miles from downtown in the last three years and needed to sell, better than a third of people in your position took a loss. Inside the line, roughly one in six did.

Why the neighborhood pages on this site read differently

The neighborhood pages on this site divide almost perfectly along the line above, which is the clearest confirmation of it I can offer.

Every page inside the line reads flat. Every page outside it reads down. The four closest fall on one side of nine miles and the five furthest on the other, with nothing straddling it. I did not select these neighborhoods to produce that result — the close-in ones are where I have worked longest, the outer ones are where the transaction volume is — but the pattern is clean enough that I'd rather show it than describe it.

Every figure above is for the whole area. The Eagle Mountain page quotes −1.36% because it measures the non-waterfront market separately; including the 87 waterfront sales lifts it to the −1.03% shown here. The gap between those two numbers is not statistically significant, so I use the whole-area figure for comparison and the narrower one where the distinction matters.

What it is not is a gradient. Beltmill at 9.8 miles is already down 0.86%, barely past the line, while Presidio at 11.7 is down less. Distance predicts which side you land on far better than it predicts how far you fall.

It also means the reassuring version and the alarming version of the Fort Worth market are both true, depending on which half of the city you own in.

What I can't tell you is why distance is the variable that survives. The test above rules out new-construction share as the explanation once distance is accounted for. Whether distance stands in for commute, for vintage, for lot supply, or for something I haven't measured, I don't know — and I'd rather say so than pick the story that sounds best.

What I can't tell you is why. The obvious explanation is that the declines are where the building is, and the test above says that explanation stops working once distance is in the model. Distance is what survives. Whether distance is standing in for commute, for vintage, for lot supply, or for something I haven't measured, I don't know — and I'd rather say so than pick the story that sounds best.

What is not happening: there is no distress wave

The obvious next assumption is forced selling, and the data doesn't support it.

• Bank-owned, foreclosure, short-sale and probate listings run 0.46% on the outer ring against 0.83% in the center. The edge has fewer distressed listings than the core, not more.
• Outer-ring sellers are discounting less from original list — 2.73% against 3.33% inside.
• But they are paying more to close: 63.5% of outer-ring sales carry seller concessions against 52.3% inside.
• And they hold longer before reselling — 17 to 19 months against 10 to 12.

The picture that fits all four is not distress. It is people who cannot clear their basis, so most of them simply don't list — and the ones who must pay the buyer to leave, through concessions rather than price cuts.

Where the market actually is: 70% of it is under $400,000

Fort Worth is discussed as though it were a luxury market. It is not.

Seventy percent of Fort Worth trades below $400,000. Ninety-one percent below $600,000. Just 2.4% above a million.

The tax figure on a listing is the one number most likely to be wrong

This applies citywide and it costs buyers real money every month.

Effective tax rate, measured as the tax figure shown divided by the actual sale price, sorted by how old the house was when it sold:

A house sold in its first year shows a tax figure around one-fifth of what it becomes. That's the land-only assessment — the appraisal district hasn't added the finished building to the roll yet. By year two it has.

On a $345,000 new build, that is roughly $2,000 a year at closing becoming about $8,000 by year two — an extra $500 a month arriving in the second year, inside an escrow payment the buyer thought they understood.

The reverse trap sits on older homes. Texas caps how fast a homesteaded property's assessed value can rise, so a long-held house can be assessed far below market — and the cap comes off when it sells. The 1.84% on forty-year-old homes above is partly that effect.

In both directions, the number on the listing is the seller's tax bill, not yours.

What this means if you're buying

• Beyond about nine miles, treat a short holding period as a real risk. If it's a ten-year house, this doesn't apply. If a job change inside three years is plausible, price the entry accordingly.
• Ask for concessions. Nearly 59% of Fort Worth sales close with seller-paid money in them, and 63.5% on the outer ring.
• Get the assessed value, not the listed tax, and ask what it becomes without the seller's exemptions.
• Check for MUD and PID assessments on the specific parcel. They vary enormously and they don't always appear where you'd look.

What this means if you're selling

• Pricing to 2024 comparables is the most expensive mistake available in most of this city. Sixteen ZIP codes have moved down since then.
• Where you are matters more than it did. A central-neighborhood seller is in a flat market; an outer-ring seller is not, and the same strategy doesn't work in both.
• Expect to contribute. Budget for concessions as part of the plan rather than as a failure.

How this was measured

All figures come from NTREIS closed sales inside Fort Worth city limits, 19 September 2023 through 18 September 2026, reported in aggregate. Trends are annual coefficients from log-price models controlling for house size, lot size and build year; "significant" means p < 0.05. Repeat-sale analysis uses parcel numbers to match properties, excludes builder-to-owner legs and excludes houses whose square footage grew more than 5% between sales. Distance is straight-line from downtown Fort Worth.

Where a result didn't hold up under controls — new-construction share being the main one — I've reported that it didn't.

Questions about a specific property or neighborhood

A citywide report tells you the weather. It does not tell you what your house is worth, and the spread between neighborhoods in this city is wide enough that the average describes almost nobody.

What I can tell you is what the closed sales in your specific subdivision support, how wide the range around that is and what drives it, and what a property will cost you every month to hold once the assessment settles.

Call or text (682) 207-4310, or send me a note.

Terrence Lee Bilodeau is a Texas real estate broker (TREC #0686157) and a Certified Residential Appraiser with Compass RE TX, LLC. Market conditions change — confirm current figures for a specific property before relying on them. Nothing here is an appraisal or a valuation of any particular property.