← Back to the preview findings · See what you receive →

The 2026 St. Louis County Tax Sale: a bidder's guide

Sealed bids accepted from Friday 21 August · final deadline 5pm Wednesday 26 August 2026

Section 00The $15.5 million problem, and how to keep your share of it small

Across the six sales from 2020 through 2025, the people who won a contested parcel paid $15,468,650 more than the next-highest bid. Not more than the properties were worth — more than the sale required of them.

This sale is far older than that. Every superlative in this guide — largest, fewest, worst — is bounded by the six sales we hold end to end, and says nothing about the decades before them.

In the most recent sale the middle winner's share of that was $6,013. One buyer bid on 57% of that year's contested parcels and won 37% of them, so the figure to plan against is the $4,599 the typical winner paid with that buyer's envelopes set aside. Section 04 does that arithmetic.

Either way, the money comes back to nobody. Under RSMo 140.340 redemption interest runs on the outstanding amount the sale was held to collect, and it is paid by the owner catching up, not by the county. An overbid sits above that amount, so not one dollar of it earns a cent. Whether it comes back to you if the owner buys the parcel back is a question we do not answer — see section 07.

That is the half of the sale with a rival in it. The other half is quieter and it is not cheaper. Nearly half of every sale draws a single bidder, and those parcels have no runner-up to be measured against — only the county's own asking price. Measured against that, 228 of the 243 single-bidder parcels in the most recent sale sold for more than the county asked: $987,014 handed over by people bidding against nobody. Section 04 is where that sits.

This guide exists to make your share of that number as small as it can be.

It prices one thing, parcel by parcel: what it has taken to beat the other envelope. Not what a parcel is worth — what winning it has cost, at whatever odds you decide you want. Come in under that number and you are buying a smaller chance than you think. Go well over it and you are in the figure above. It does not appraise anything for you, and it does not tell you to buy. It reports what this market has paid.

The numbers come from the county's own bid sheets for the six sales from 2020 through 2025 — 2,735 parcel sales and 6,177 individual bids, winning and losing — matched to county parcel and assessment records, then applied to the 3,827 parcels on offer in 2026. The losing bids are the point. They are the half of the sale the county never publishes, and the losing bid is the number you actually have to beat.

One thing this cannot do, and no guide honestly can: get all of that $15.5 million back. Nobody bids the runner-up plus a dollar, because nobody can see the runner-up. What you can stop doing is guessing at the level.

How this sale works

Six rules run the whole thing. Skip them if you have bid here before.

• The bids are sealed. For each parcel you want, you write one number on a form and hand it in. Nobody sees anyone else's number until it is over. There is no bidding back and forth, and no second try.
• The county sets the floor. Every parcel carries an opening bid — the unpaid taxes plus interest, penalties and the collector's costs. You cannot bid less, and the county publishes the number before the sale.
• The highest bid wins, and every envelope you submit is live at the same time. Bid on six parcels and the county can hand you all six.
• Winning does not hand you the parcel. You get a certificate of purchase. The owner then has a window to buy the parcel back by repaying you with interest — but interest on the outstanding amount only, never on what you bid above it. That is redemption, and section 06 is about it.
• The window is one year on most parcels and ninety days on some. Which one you are looking at is published, and it is the first thing to check. Section 01 shows you where.
• If the owner does not redeem in time, you can apply for a collector's deed. From there, you should work with a real estate attorney.

One word about the numbers that follow. When this guide says typical, it means the middle of the record — half the parcels were above that figure and half below. It is not an average. A handful of enormous bids drag an average badly, and section 04 shows by how much.

Five figures puts you in the middle of this market, not the margins

Most people at this sale are not writing large checks. Across the six sales, half the bidders committed less than $19,703 in total, and a quarter committed less than $7,390.

Read that carefully, because it is a total and not a bid. Since every envelope is live at once, your budget is not a shopping list you pick from afterwards — it is a commitment to every parcel you enter, all at the same moment.

Section 01The market is pursuing parcels, not redemption interest

There are two ways to make money here. You pursue the parcel and take your certificate to deed, or the owner redeems — pays the back taxes with interest — and the certificate is paid off instead. Those are two very different numbers to write on the form, and the record shows which one this market is writing.

Redemption interest runs at 10% under RSMo 140.340, and it runs on the outstanding amount — the delinquent taxes and costs the sale was held to collect, which is what the opening bid represents. It is not paid by the county; it is paid by the owner catching up. And it is not paid on what you bid. Every dollar above the opening bid earns nothing and drags your return down. So a bid built to earn that interest has to hug the opening bid — and a bid that hugs the opening bid only wins where nobody else showed up.

That is not what happened. Of the 2,705 sold parcels the county published an opening bid for, only 21.7% went for a price within 25% of that bid. Call that the yield band: it is the zone where redemption interest is still worth having, and section 06 explains why the line falls there. Where a second bidder turned up, only 4.4% of parcels landed inside it.

Most parcels sold well past the point where redemption interest can carry the return. The majority of the bidders in this auction are paying for the parcel.

Which offering the parcel is in

A parcel that does not sell comes back to the next year's sale, and the third time it comes back the law treats it differently. The county publishes which of those you are looking at: the 2026 list carries one of two labels against every parcel — First/Sec on 3,611 of them, Third on 216. It is the first column you should read, because it decides how long the owner has to take the parcel back from you:

Those same two labels run through the sales we priced: 2,479 of the 2,735 parcels sold in 2020–2025 were first- or second-offering and 226 were third-offering. Every parcel that appears in a sale book matched exactly one of the two, so this is the county's own designation, not something we inferred. Thirty parcels from the 2025 sale appear in no sale book at all and carry no offering round; section 07 says what that does and does not change.

What the statute does not do is drop the floor. A third offering still requires a bid at least equal to the taxes, interest, penalties and costs — RSMo 140.250 says so in the same sentence that grants the ninety days. A third-offering parcel is not a cheaper parcel by law. It is a parcel that has already been offered twice and drawn nobody.

It is a thinner market, and the record is consistent about it — how crowded each book is, how far the winner had to go, and what that cost them:

The third row is the one to sit with. At the first and second offering the typical winner paid more than double what the sale asked; at the third offering, a quarter over the floor. The reason is not that those parcels are cheaper — the minimum-bid rule is the same. It is that there was nobody there to bid against.

All three rows favor a buyer who wants the parcel. None favors a buyer who wants the interest — and that is easy to get backwards, because a parcel that is cheap to win looks like the ideal place to earn redemption interest. It is the worst place in this sale to try. The owner has let the county advertise the property three times without paying, and the statute gives them ninety days rather than a year to change their mind. Section 06 makes that case and excludes the third offering from the yield discussion entirely.

One note on how to read the rest. Where a figure differs by offering, this guide says so and gives both. Where it says nothing, the figure covers all six sales together, and the two offerings are close enough that the record cannot tell them apart.

So: this guide is about winning the parcel — what it costs, which parcels cost less, and what being wrong costs in dollars. If you are the other kind of buyer, section 06 is written for you. Go there first if you like; nothing between here and it is wasted.

Section 02There are two kinds of parcel, and they cost different amounts to win

Every parcel in this sale belongs to one of two groups, and which group it is in decides what winning costs before you have looked at the parcel.

The line will look obvious to anyone who knows this market. There is north county, where bids are cheap, and there is everything else, where they are expensive. There is nothing subtle about it, and you do not have to take it on faith. The cheap half is entirely single-family houses — every one of them, in both of the sales we used to test the split — and 94% of it sits in four school districts: Ferguson-Florissant, Normandy, Riverview Gardens and Jennings.

But it is not "north county." It is single-family in north county, and the distinction is absolute. Of the 451 parcels sold in those four districts, 405 are single-family and every one of them landed in the cheap half. The other 46 — 23 vacant lots, 9 multi-family buildings, 7 duplexes, 5 commercial parcels, 2 industrial — landed in the expensive half, all 46 of them, without a single exception. The expensive half is everything else. A vacant lot in Riverview Gardens does not price like the house next door to it. It prices like the expensive half, because that is the market it is in.

Run that rule across the 2026 offering and it takes in 1,956 of the 3,827 parcels — half the sale, and it misses none of what the scoring puts in the cheap half. Those 1,956 are houses, and they are in one place:

How we know the line holds

A rule that only describes the past is worth nothing at a sale you have not bid in yet. So we built the split on the earlier sales and then tested it on sales it had never seen: 2023 priced from the 2021 and 2022 records, 2024 priced from 2022 and 2023, and 2025 priced from 2023 and 2024. That is 1,417 parcels sorted by a rule that had never met them. We also tried finer cuts — three tiers, then five. On none of the three years does the record carry them. It carries two, and it carries them consistently:

Different sales, different source years. Across the three, the expensive half moves eight tenths of a percentage point and the cheap half five.

The third column is the one that makes this worth printing. 2023 and 2024 were both scored inside a rising market, so a split that replicated across those two could have been describing the trend rather than the market. August 2025 broke the trend: the winner's number fell for the first time in this record, and the typical gap over the runner-up fell with it — section 03 takes apart how much of that was the bidding and how much was the parcels. The two halves came through it intact, and the finer cuts failed there the same way they failed everywhere else. The line is structure, not momentum — which is the only reason it is worth carrying into a sale nobody has bid in yet.

Across the two test years the printed prices are drawn from, the two figures come out at 10.7% and 17.8%, and those are the numbers the rest of this section uses. Put 2025 in the same pool and they move to 10.9% and 17.5% — a smaller step than the one between the columns above. We have not repriced on it, deliberately. Every price in this section and every parcel in the interactive guide is quoted off one 864-parcel ladder, and moving the report without the guide would put a number on this page that nothing in your spreadsheet was priced against.

What the two halves cost

They are not cheap and expensive parcels. They are cheap and expensive to win, and a parcel can be one without being the other. Take two parcels the county appraises at exactly $100,000 — one in each half. Here is what it has cost to win them, stack against stack:

The same 50/50 shot costs $7,100 more on one parcel than the other, and the county values them identically. Nothing about the buildings has to differ. That gap is the other bidder, and it can be seen coming.

Both figures are quoted as a share of what the county says the parcel is worth — 10.7% and 17.8% for a coin flip — so they scale with appraised value. A $60,000 parcel in the cheap half is a coin flip at about $6,400.

But do not carry those two percentages to a parcel appraised over $134,000. The two halves are the first cut, not the last. Appraised value folds the market again two subsections down, and it moves the cheap half hard: a cheap-half parcel over $134,000 is a coin flip at 17.0% of appraised, not 10.7%. On a $160,000 parcel in Ferguson-Florissant that is writing $17,120 where the market asks $27,200 — a $10,080 underbid on one envelope, and a parcel you do not win.

Read the two halves to learn which market you are in, then read your value band before you write a number. Every chance in this guide is quoted at the same five levels — 50%, 66.6%, 75%, 85% and 90% — so a rung means the same probability everywhere. It does not mean the same price. The price is per group, and the four-group figure below is where it lives.

Three things to take from the two halves

Two things on that curve. A bid of 10% of appraised value wins about 44% of the cheap half and only about 21% of the expensive half. The same number is a fair shot in one group and a long shot in the other.

And the top of the curve is reachable. About 30% of appraised value takes better than four parcels in five anywhere in the sale, and that is still a 70% discount to the county's own valuation. Certainty is for sale here, but it is expensive — section 04 puts a price on it.

← Back to the preview findings · See what you receive → · mosaicsyndicate.com

We strive to exceed your expectations with honesty and transparency.