I pulled all 35,324 Davidson County listings from the last 12 months and cut them by
year built, which is a split the usual market reports never do. They stop at single
family, townhouse and condo. That tells you nothing about whether to build or renovate on
a given street.
Here is what 26 zip codes look like when you separate new construction from the
pre-2000 stock a flipper actually buys.
First, the number that breaks the model
| Cohort | Listings | Failed | DOM (sold) | SP/OLP | Took a price cut | Months supply |
|---|---|---|---|---|---|---|
| New construction, 2023+ | 10,125 | 60.3% | 14 | 100.0% | 22.6% | 5.7 |
| Built 2000 to 2022 | 10,314 | 45.3% | 26 | 97.0% | 37.4% | 5.8 |
| Flip pool, pre-2000 | 14,129 | 40.0% | 17 | 96.8% | 34.3% | 3.8 |
New construction fails at 60.3% while closing at 100.0% of original list price, with
only 22.6% of listings ever recording a price cut. That looks impossible until you see the
mechanism.
Builders do not discount. They cancel the listing and relist at the number they
want. A homeowner who misprices leaves a trail of reductions. A builder leaves a
cancelled listing and a fresh MLS number. So a real share of that 60% is listing churn
rather than unsold houses, and it means two things if you underwrite here:
- Sale-to-list ratio is close to useless as a softness signal for new construction. It
will read near 100% no matter how dead the submarket is. - Judge new-build demand by months of supply instead. That number is
not gameable by relisting.
Where renovating beats building
In ten zip codes the older stock earns more per square foot than a new build does. In
six of them it also clears faster, which is the combination that actually decides the
trade.
Flip markets
| Zip | Neighborhood | $/sqft spread | Resale supply | New-build supply | Resale failed | New-build failed |
|---|---|---|---|---|---|---|
| 37216 | Inglewood | +36 | 2.7 | 4.3 | 36.9% | 57.0% |
| 37209 | Sylvan Park / The Nations | +35 | 3.1 | 6.4 | 41.1% | 61.7% |
| 37206 | East Nashville | +27 | 2.3 | 5.5 | 35.5% | 59.7% |
| 37138 | Old Hickory | +12 | 4.0 | 6.9 | 37.4% | 44.3% |
| 37210 | Wedgewood-Houston | +11 | 4.8 | 7.0 | 39.2% | 47.0% |
| 37207 | North Nashville | +4 | 3.3 | 6.4 | 45.8% | 64.8% |
Inglewood is the widest gap in the county. Pre-2000 stock there earns $36 more per foot
than new construction and moves in 2.7 months of supply against 4.3 for new. East Nashville
is the cleanest version of the same story: $27 more per foot, 2.3 months against 5.5.
If you are weighing a teardown against a renovation in those six zips, the exit data
says renovate.
Build markets
| Zip | Neighborhood | $/sqft spread | Resale supply | New-build supply | Resale failed | New-build failed |
|---|---|---|---|---|---|---|
| 37204 | Berry Hill / Melrose | -93 | 4.5 | 4.3 | 40.2% | 55.1% |
| 37215 | Green Hills / Forest Hills | -141 | 4.7 | 6.1 | 39.7% | 54.0% |
| 37027 | Brentwood edge | -180 | 3.1 | 6.5 | 34.8% | 60.1% |
| 37205 | Belle Meade / West Meade | -188 | 4.4 | 6.3 | 39.3% | 59.4% |
Demand only. Whether these can be built at a profit is the next section, and it changes the answer in several zips — read it before you price a lot.
The reverse is just as clear. Belle Meade pays $188 more per foot for new, the
Brentwood edge $180, Green Hills $141. These are not flip markets. The buyer is paying for
new and the older stock does not carry the price.
Where I would be careful right now
| Zip | Neighborhood | $/sqft spread | Resale supply | New-build supply | Resale failed | New-build failed |
|---|---|---|---|---|---|---|
| 37220 | Oak Hill | -31 | 3.7 | 12.0 | 31.9% | 40.4% |
| 37208 | Germantown / Hope Gardens | -41 | 6.1 | 10.6 | 50.4% | 61.1% |
| 37218 | Bordeaux | -52 | 5.6 | 10.1 | 50.6% | 69.6% |
| 37203 | Midtown / Music Row | -341 | 9.0 | 21.8 | 48.3% | 62.0% |
Midtown is carrying 21.8 months of unsold new-construction supply. Oak
Hill is at 12.0, Germantown 10.6, Bordeaux 10.1. Those are not opinions about the
neighborhoods, they are counts of finished product already sitting in front of whatever
you deliver next.
Germantown deserves a specific flag on the renovation side too: pre-2000 stock there
closed at 93.2% of original list, the worst ratio of any cohort in any zip in this
dataset, on 6.1 months of supply. That trade is currently being priced wrong.
Does it pencil? The cost side
Everything above is exit-side: what a finished house sells for and how fast. None of it tells you whether the thing can be built for less than it sells for, which is the question that actually kills most hot-build-zip calls.
So here is the cost side. Each row takes that zip’s median new-construction sale and median square footage and builds it at the cost tier that product falls in: $160/sqft for low-end product, $210 to $255/sqft for high-end above $900K resale. Plus 15% soft costs, 7% selling costs, against a 20% target margin.
The column that matters is margin with free land. That is the profit left if the dirt costs nothing. A zip that cannot clear 20% when the land is free cannot be rescued by a cheap lot. It is not a build market at any land basis.
| Neighborhood | Median new sale | Cost tier | Margin, free land | Max cost/sqft | Call |
|---|---|---|---|---|---|
| Midtown / Music Row | $800K | $160 | 61% | $370 | PENCILS |
| Belle Meade / West Meade | $3.30M | $210-$255 | 52% to 43% | $373 | PENCILS |
| Berry Hill / Melrose | $2.48M | $210-$255 | 47% to 37% | $331 | PENCILS |
| Brentwood edge | $2.80M | $210-$255 | 46% to 36% | $324 | PENCILS |
| Green Hills / Forest Hills | $2.48M | $210-$255 | 43% to 32% | $306 | PENCILS |
| Crieve Hall / South Nashville | $850K | $160 | 41% | $224 | PENCILS |
| Germantown / Hope Gardens | $678K | $160 | 35% | $200 | PENCILS |
| South Nashville / Industrial | $550K | $160 | 33% | $194 | PENCILS |
| Sylvan Park / The Nations | $754K | $160 | 33% | $194 | PENCILS |
| Inglewood | $668K | $160 | 33% | $193 | PENCILS |
| Hillsboro Village / Vanderbilt | $1.05M | $210-$255 | 32% to 19% | $252 | COST SENSITIVE |
| East Nashville | $994K | $210-$255 | 26% to 12% | $230 | COST SENSITIVE |
| Oak Hill | $2.01M | $210-$255 | 25% to 10% | $226 | COST SENSITIVE |
| North Nashville | $512K | $160 | 23% | $167 | THIN |
| Bordeaux | $483K | $160 | 23% | $166 | THIN |
| Whites Creek | $582K | $160 | 20% | $160 | THIN |
| Donelson / Airport | $555K | $160 | 17% | $155 | NO MARGIN |
| Old Hickory | $477K | $160 | 17% | $154 | NO MARGIN |
| Hermitage | $470K | $160 | 9% | $140 | NO MARGIN |
| Antioch | $432K | $160 | 7% | $137 | NO MARGIN |
| Bellevue | $985K | $210-$255 | 7% to -11% | $179 | NO MARGIN |
| Madison | $387K | $160 | 5% | $132 | NO MARGIN |
| Goodlettsville | $324K | $160 | 3% | $130 | NO MARGIN |
7 of these zips cannot be built at a profit right now, even with free land: Donelson / Airport (17%), Old Hickory (17%), Hermitage (9%), Antioch (7%), Bellevue (7%), Madison (5%), Goodlettsville (3%).
That is the part worth sitting with. Antioch and Goodlettsville have the two best new-construction absorption rates in the county, 1.8 and 1.7 months of supply. On demand data alone they look like the obvious places to build. The median new house in Antioch sells for $447K at 2,164 square feet, which costs about $398K to build and sell at builder grade. That is roughly $18K of profit with the land free. The lot is not free.
Cost sensitive: Hillsboro Village / Vanderbilt (clears at $210/sqft, fails at $255), East Nashville (clears at $210/sqft, fails at $255), Oak Hill (clears at $210/sqft, fails at $255). Your real cost decides these, not the market.
Max cost per square foot is the number to carry around. It is the hard cost that still leaves 20% with free land, so you can hold it against your own build number for whatever product that zip actually demands and get an answer in one step.
One caveat on the tier split: spec drives cost, and resale price is only a proxy for spec. A big plain house just over $900K builds nearer $160 than $255. Bellevue is the live example in that table, so check the resale per foot before you trust a high-tier verdict on a low per-foot row.
What resale price to aim at
The same band behaves differently depending on what you are delivering. These are exit
prices, not budgets.
| Target resale | New: failed | New: supply | Reno: failed | Reno: supply |
|---|---|---|---|---|
| Under $300K | 52.1% | 4.2 | 38.5% | 4.1 |
| $300-400K | 62.1% | 3.5 | 39.6% | 3.5 |
| $400-600K | 61.6% | 4.8 | 42.1% | 3.9 |
| $600-800K | 67.2% | 8.1 | 39.9% | 4.0 |
| $800K-$1.1M | 55.0% | 6.2 | 36.2% | 3.4 |
| $1.1M+ | 56.7% | 8.2 | 43.8% | 4.8 |
Read down a column, not across. Comparing the two failure columns head to head would
hand renovation every band, for the relisting reason above rather than for any reason
about demand.
Within new construction, the worst place to land right now is $600K to
$800K: 67.2% failure on 8.1 months of supply, worse than the bands on either side
of it. The best is $800K to $1.1M. Within renovation, $800K to $1.1M is also the strongest
band at 36.2% and 3.4 months, and $1.1M+ is the weakest.
New condo product is its own warning: 18.6 months of supply and a 63.9% failure rate.
How to use this
Three filters, in this order. Is there already a pile of unsold product in your exit
band and vintage, measured in months of supply. Does the spread favor the vintage you are
planning to deliver. And only then, does your land and construction budget clear the
number the data says you will actually exit at.
The submarket failure rates matter as much as the price comps. I wrote up the
county-wide version of that in
the piece on why 46.9% of Nashville listings never sold, including
why the days-on-market figure everyone quotes is calculated only on the homes that closed.
I run this analysis monthly across all 26 Davidson zip codes. If you build or flip here
and want the full flipbook, including the per-zip build and flip profiles with bed, bath,
square footage and garage detail, call or text
615-436-8003 and I will send it over.
Method
New construction means year built 2023 or later. The flip pool means built before 2000.
Failure means the listing ended Expired, Cancelled or Withdrawn rather than Closed, as a
share of all listings in that cohort. Months of supply is active listings for that vintage
divided by its 12-month sales rate. Zips with fewer than 25 listings on either side are
omitted. Davidson County only, 35,324 listings, 12 months ending August 3, 2026, source
RealTracs MLS. I compiled and ran it myself.
One caveat worth repeating: because builders relist rather than reduce, some sellers in
every cohort appear more than once under different MLS numbers. This measures listing
attempts, not unique houses.