Advanced tactics, and what the data actually shows
Most guides to Texas tax sales describe the process. This page describes the inventory. Every figure below is calculated from the 437,471 delinquent records and 3,781 auction listings on this site, recalculated each time the page is built. Where the data does not support a claim, it says so instead of guessing.
1. The debt is small before the auction, and large by the time it gets there
On the records side, delinquent taxes are usually a small fraction of what the property is worth. Across the counties that publish assessed values reliably, the middle half sit between 1.2% and 3.0% of assessed value. That is the number that makes people think tax sales are free money.
It is not, and the reason is visible in the auction data. By the time a property reaches an actual sale, the median county price has climbed to 48% of assessed value. The county's price is the minimum bid, which the judgment sets from the taxes, penalties, interest and costs owed; where a county published the back taxes themselves, that figure is used instead (27 listings, against 2,878 on the minimum bid). The cheap ones almost never arrive. The likely reason is simple, though this data cannot prove it on its own: an owner with real equity has options a tax sale removes, so the ones with the most to lose tend to find a way to pay. What reaches the courthouse steps is what is left.
2. Struck-off property is not the bargain it is usually called
Struck off means the property went to auction and nobody bid, so it can often be bought directly from the taxing entity without competition. That part is true, and it is why struck-off lists get recommended to beginners.
The part nobody mentions is why nobody bid. On this site's current inventory the median struck-off property is priced at 68% of its assessed value. Compare that to 28% for property with a scheduled sale date and 26% for future sales. Struck-off inventory is not overlooked. It is picked over. The room full of people who do this for a living already looked at it and passed.
| Status | Median county price as % of value | Listings |
|---|---|---|
| Scheduled for sale | 28% | 505 |
| Future sale | 26% | 743 |
| Resale | 20% | 106 |
| Struck off | 68% | 1,483 |
This does not make struck-off worthless. It makes it a sorting problem instead of a browsing problem. The range is wide: the median sits at 68% of assessed value, but the cheapest quarter of that inventory comes in under 38%. Two properties on the same county list can be very different purchases, and the list is not published in an order that tells you which is which.
Why any particular property went unsold is not something this data records, so anyone telling you the reason is guessing. What the data does support is the method. Sort the struck-off list by price against value rather than reading it in county order, and judge the cheap end on its own merits. Reading the list top to bottom as published is how people end up paying the most for the properties the professionals rejected first.
3. Roughly one in twenty auction properties is upside down
Of the auction listings carrying both figures, 172 carry a county price above what the county says the property is worth. There is no bid at which those make sense as a purchase, and they are mixed into the same lists as everything else with nothing marking them.
Before anything else, divide the county's price (the minimum bid, or the back taxes where the county published them) by the assessed value. Above 100% the property is underwater and there is no bid price that fixes it. That line is arithmetic, not opinion. Past that, use the inventory itself as the yardstick: the median listing sits at 48% and the median struck-off listing at 68%, so anything above those is expensive relative to what else is available right now. Both numbers are on every record on this site, so the check takes seconds.
4. Most delinquency is temporary, and the deepest cases are not the ones that sell
27% of the delinquent records on this site are in their first year. The large majority of those will be paid. A first-year delinquency is usually a late payment, a death in the family, an escrow error, or a refinance in progress, and it is the weakest signal on the site.
The obvious next thought is that the oldest debts must be the ones heading to auction. 45,195 records here are ten or more years delinquent, and it is tempting to treat that as the list to work. The auction data does not support it.
Among auction listings that publish a delinquency figure, the median is 4 years, and 60% fall between two and six years. Only 13% are ten years or more. Worse, of the deeply delinquent ones that did reach a sale, 44% were struck off, meaning nobody bid on them at all.
A property nobody has paid tax on in a decade usually has something structural holding it in place rather than pushing it toward the courthouse. Texas lets homeowners over 65 and disabled homeowners defer property tax indefinitely, with the balance accruing rather than triggering foreclosure. Unsettled estates and heirship disputes stall cases for years. And a county will not always spend legal money foreclosing on a low-value parcel. This database records the debt, not the reason, so it cannot tell you which applies to a given property. What it can tell you is that age alone is not the buying signal people assume it is.
Sample note: 683 of 3,781 auction listings publish a years-delinquent figure, so the percentages above describe that subset rather than every listing.
5. Where the debt runs heaviest
Median tax debt as a share of assessed value, by county. Only counties where at least 90% of records carry a published value are listed, because a ranking built on counties that publish values for two percent of their roll would be noise dressed as insight. A higher number means less room between what is owed and what the property is worth.
| County | Median debt as % of value | Records with a value |
|---|---|---|
| Bandera | 4.8% | 2,477 |
| Sabine | 4.4% | 1,688 |
| Nueces | 4.3% | 6,296 |
| Hood | 3.5% | 2,752 |
| Ector | 3.2% | 7,966 |
| Bexar | 3.0% | 33,916 |
| Tom Green | 2.3% | 2,250 |
| Gregg | 2.3% | 6,460 |
| Taylor | 2.2% | 1,311 |
| Travis | 2.1% | 6,725 |
| Brazos | 1.9% | 2,906 |
| Grayson | 1.9% | 4,036 |
| Grimes | 1.5% | 4,761 |
| Collin | 1.4% | 9,938 |
| Angelina | 1.4% | 1,915 |
| Waller | 1.4% | 1,793 |
| Denton | 1.3% | 8,972 |
| Midland | 1.2% | 10,429 |
| Tarrant | 1.1% | 60,262 |
| Williamson | 1.0% | 11,357 |
| Galveston | 0.7% | 9,811 |
| Brazoria | 0.3% | 8,170 |
| Cherokee | 0.1% | 2,608 |
Counties absent from this table are not absent from the site. They simply do not publish assessed values with their delinquent roll, so no honest ratio can be computed for them. Dallas County, for example, carries 20,203 records here but publishes a value on 2% of them.
6. What this data cannot tell you
Worth stating plainly, because most sites in this category imply otherwise.
- Assessed value is not market value. Texas appraisal districts assess for taxation, and the figure can sit well above or below what a property would actually sell for.
- Nothing here reflects condition. A house with a collapsed roof and a renovated house on the same street carry similar assessments and identical-looking records.
- No outcome data exists in this database. We can show what is owed and what is listed. We cannot tell you what anyone paid at auction or what they made afterward, and any site quoting you a typical return is quoting something it did not measure.
- Liens outside property tax do not appear here. Federal tax liens survive a tax sale, and a title search is the only way to find them.
Where to start
If you are new to this, the sequence that wastes the least time is to read the full lifecycle of a Texas tax sale first, so the vocabulary is familiar. Then pick one county, sort by debt against value, ignore everything above 60%, and look at what is left on the map. One county understood properly beats five counties skimmed.
Recalculated November 02, 2026 from live data. Figures move as counties publish new rolls.