Shelby County Foreclosure Report — Q3 2026

Jump to: key statistics · findings · quarter-by-quarter table · ZIP table · methodology · cite or reuse this report · download the data

Foreclosure filings in Shelby County jumped by more than half in the third quarter — 367, the most in any quarter since our records begin in January 2025, and 55% above Q2’s 237. Completed foreclosure sales rose 27% to 157; lender- and government-type buyers took half of the homes sold, their largest share in the five quarters tracked; and the median auction price slipped to $115,106. A “filing” here is an Appointment of Substitute Trustee, the document most often recorded first in a Tennessee foreclosure (it is also recorded when a lender replaces the trustee); a completed sale is a Substitute Trustee’s Deed. The Q3 total is 55% above Q2’s, which was the lowest quarter in our records, 35% above Q3 2025, and 8% above Q1 2026; through nine months, 2026 filings (945) are level with the same months of 2025 (942), and completed sales are 421 against 438.

Key statistics — Q3 2026 at a glance

The Q3 2026 figures below are computed from Shelby County Register of Deeds or County Trustee records (methodology); context figures are sourced where shown. All of it is free to republish with a link to this page (citation formats).

Findings

1. Filings rebounded to the highest quarterly total on our books

Shelby County recorded 367 new foreclosure filings in the third quarter of 2026 — about 4.0 a day, roughly one every six hours, up from 2.6 a day the quarter before. (Each filing is an “Appointment of Substitute Trustee,” the document most often recorded first in a Tennessee foreclosure.) That is 54.9% above the prior quarter and 35.4% above the same quarter last year, and it tops the previous high in our records, 346 in Q2 2025, by 6%. Monthly filings went from 81 in June to 132 in July, then 113 in August and 122 in September. The Register’s index was verified through October 2, 2026 at pull time and the data was pulled October 5, so September is fully counted.

One limit on the word “filing”: an appointment is also recorded when a lender replaces the trustee on a loan already in foreclosure, so these are counts of recorded paperwork, not necessarily of newly distressed households. The cross-referenced mortgage on each appointment lets us test how much that matters. Counting distinct mortgages rather than appointments, Q3 has 346 against Q2’s 224, a 54.5% increase, essentially the same as the 54.9% increase in appointments; 27 of Q3’s 346 mortgages had already been appointed on in an earlier quarter in our records. A small share of repeat paperwork does not explain the jump, though it cannot show which of the 346 loans are brand-new defaults.

Last quarter’s edition asked whether Q2’s drop to 237 was the start of relief or a pause. Q3 reversed the drop. Filings in Q3 alone equal 31% of the 1,185 recorded across the four quarters before it. Across the seven quarters in our records, quarterly filings have ranged from 237 to 367, which makes any single quarter a thin basis for a trend. The records show what was filed, not why.

A filing is also not a lost home: some of these loans are reinstated, modified, or paid off before any auction happens, which is why completed sales run far below filings in every quarter.

Facing foreclosure yourself? Free HUD-approved counseling: Homeowner’s HOPE Hotline, 888-995-4673. A filing does not mean the home is already lost.

2. Auction sales rose too, though not as fast as filings

157 homes went through auction to a recorded deed in Q3, up 26.6% from Q2’s 124 and 6.1% above Q3 2025’s 148 — the most since Q2 2025’s 171. Measured against the same quarter’s filings, completions ran at 42.8%, down from Q2’s 52.3% because filings jumped faster than sales. Last quarter’s edition said that if filings stayed low, the auction calendar should thin out in the second half of 2026. Filings did not stay low, and sales rose anyway. Sales follow filings with a lag that this report does not measure, so Q3’s filings may feed sales in later quarters, though loans also get reinstated, modified, or paid off along the way.

3. Lender-type buyers took half of the homes sold at auction; investment entities fell to a quarter

Of the 157 homes sold at auction, 80 (51%) went to lender- or government-type grantees — the largest group, and the largest share in the five quarters this report tracks. 39 (25%) went to investment entities — LLCs, partnerships, and trusts — the smallest investor share in those five quarters (it was 40% in Q2 and 52% a year ago). 24 (15%) went to individuals, and 14 (9%) were ambiguous, deeds whose only indexed grantee carries a substitute-trustee marker (several follow lender-side sellers, so the lender-side share may be understated). Nine of the 80 lender-side deeds went to a single bank-held acquisition trust, RCAF Acquisition Trust, with U.S. Bank as trustee.

The 15% individual share is the same as Q2’s, but it is concentrated: 14 of the 24 individual-bucket deeds went to just two buyers (eight and six; the second also appears on two more deeds that sit in the ambiguous group, which would make that buyer’s count eight), nine went to nine different buyers, and one record is a misspelled trustee name that belongs in the ambiguous group.

Repeat investors came back in smaller numbers. Five investment entities bought two or more foreclosures apiece — 21 homes between them, more than half of the 39 investor purchases. Southeast Investors LLC led with eight, up from three in Q2; MHV1 LLC took six, Avalon Capital LLC three, and Memphis Investment Properties LLC and Butler Properties LLC two each. WZ Capital GP, which led Q2 with six, bought none this quarter.

A caution on that investor share. The buyer split is a name-based classification, and the Q3 investor bucket holds more look-alikes than earlier quarters: three homeowners’ or condominium associations, a nonprofit, six names that read as lenders, servicers or securitization trusts, and a law firm’s misspelled name. A stricter second screen would remove 11 of Q3’s 39 investment-entity deeds (leaving 28, or 18% of sales) and between 3 and 5 of every earlier quarter’s investment-entity deeds (for Q2, 4 of 50, leaving 37%). So the drop in the investor share shows up on either measure, but its size depends on the classifier.

Rule-based counts. One of the 24 Q3 individual-bucket records is a misspelled trustee name (see methodology). As-Is Home Buyer is not named as a grantee or grantor on any record counted here.

4. The median auction price slipped to $115,106

The median completed foreclosure sale in Q3 2026 was $115,106, across the 138 of 157 deeds that recorded a sale amount. That is 4.9% below Q2’s $121,001 and 12.7% below Q3 2025’s $131,806, and the second-lowest of the five quarters, behind only Q1 2026’s $107,050. Redfin’s Memphis median sale price was about $187,000 for the three months ending August 2026, which puts the typical auction price at about 61% of the citywide figure (Redfin’s figure covers the city of Memphis, while county auctions also include Bartlett, Collierville and other municipalities). As always, the records alone can’t say how much of that gap is condition and location and how much is auction discount.

5. Frayser returned to the top of the map, and East Memphis’s 38119 went from none to ten

Last quarter Midtown’s 38104 and Cordova’s 38016 tied for the most new filings. This quarter Frayser’s 38127 led. Among the 194 new filings whose property address could be mapped (see methodology), 38127 recorded 15, more than double its 7 in Q2. Southeast Memphis’s 38125 had 13 (up from 5), Southwest Memphis’s 38109 12, the University District’s 38111 11, and East Memphis’s 38119 10, up from none in Q2. Cordova’s 38016, the Orange Mound area’s 38114, Bartlett’s 38134, and Arlington’s 38002 followed with 9 each. Q2’s co-leader Midtown fell from 9 to 5, and South Memphis’s 38106 from 6 to 1. On the auction side, 38109 closed more foreclosure sales than any other ZIP (14 of the 144 mappable), with Frayser and the University District next at 10 each. Three cautions: about 47% of the quarter’s filings couldn’t be mapped to a ZIP, so this ranking is indicative rather than a census; these are raw counts, not rates, and bigger ZIPs have more homes to lose; and with total filings up 55%, most ZIPs rose — of the 33 ZIP codes with a mapped filing in either quarter, 19 recorded more filings than in Q2, 10 recorded fewer, and 4 were unchanged — so the rank and the share of mapped filings say more than the raw increase (Frayser’s share was 7.7% of mapped filings, up from 5.3%). At counts this small, a quarter-to-quarter reshuffle can be noise as easily as signal.

6. Why a foreclosure cluster is more than a real-estate statistic

This report measures filings and sales, not what follows them, and it has no data on which of these homes are vacant. But research has measured what follows. A study of geocoded Pittsburgh foreclosure and crime data published in the Journal of Urban Economics (Cui & Walsh, 2015; NBER Working Paper 20593) found that a foreclosure by itself did not raise neighborhood crime — but once the foreclosed home sat vacant, violent crime in its immediate vicinity rose roughly 19%, and the effect grew the longer the property stayed empty. In a quarter when filings jumped by more than half — and lenders and investment entities together took roughly three of every four homes that reached auction — what happens to those homes after the sale matters as much as the sale itself.

7. Foundation Legal Group is back on top, and Q2 was the odd quarter for several firms

Foreclosing lenders typically appoint a substitute trustee — usually a specialist law firm — to run the process. Foundation Legal Group was named in 41 of the 367 filings this quarter, back on top after a dip to 12 in Q2. Padgett Law Group, Q2’s leader at 37, slipped to 31, while Rubin Lublin, LLG Trustee TN and Clear Recon all rose. Nestor Solutions of Tennessee had 13 in Q1 and none in Q2; it was named in 20 this quarter. AVT Title Services, which led Q1 with 43, stayed low. Q2 looks like the exception, not the trend, for several firms.

The five busiest trustee firms in Q3 — Foundation Legal, Rubin Lublin, Padgett, LLG Trustee TN, and Clear Recon — were named in 164 of the quarter’s 367 filings (45%), against 44% in Q2 and 35% in Q1 (counting each filing once; the Q2 edition’s 42% for Q1 summed firm counts, and some Q1 filings name two firms). Trustee assignments may follow servicer contracts, so a shuffle like this can say more about which lenders’ loans are foreclosing than about the firms themselves.

8. The tax-sale queue shrank as the next sale was advertised

Mortgage foreclosure isn’t the only way Shelby County properties are sold; the tax-sale list covers any parcel with delinquent property taxes, not only homes. As of October 5, 2026, the County Trustee’s published tax-sale list — properties with delinquent property taxes; the parcel-level list can be downloaded from that page — held 1,266 parcels, all assigned to Tax Sale 2302 (scheduled April 13, 2027). On July 6, 2026 the list held 2,490 parcels: 640 for Tax Sale 2301, 1,444 for Tax Sale 2302, and 406 still listed under an April 2026 sale that had already run. The like-for-like change is Tax Sale 2302, down 178 parcels (12.3%) from 1,444, as parcels were paid or otherwise removed. Tax Sale 2301, scheduled for October 27, 2026, no longer appears in the list; the Trustee’s schedule shows it was advertised on August 26; the Trustee’s schedule page does not give the advertised count, so we do not report one (the list showed 640 parcels assigned to it on July 6). Tax sale is a separate process from every number above, and the parcel counts are not comparable with the mortgage counts. How it works and what a homeowner can still do is in our tax-sale explainer.

Quarter by quarter

Completed sales in a quarter generally stem from filings made in earlier quarters, so the ratio column compares same-quarter totals — it is not the share of those particular filings that ended in a sale. Some filings never reach sale: loans are reinstated, modified, paid off, or sales are postponed. Q4 2025 shows 136 completed sales, one more than the Q2 edition printed: the Register added one deed recorded in December 2025 to its index after our July pull (the median is unchanged; the deed went to a lender-side buyer, so that bucket reads 63, not 62). The report’s records begin January 2025; Q1 2025 (325 filings, 119 completed sales) and Q2 2025 (346 filings, 171 completed sales) predate the five-quarter window above; the 2025 filings are visible in the monthly chart below. The Q2 edition printed 120 for Q1 2025 completed sales; that was a transcription error, and the dated index count, unchanged across every pull, is 119 (the raw January 2025 export holds 120 rows because one record carries a null date).

Where Q3 2026 foreclosure activity concentrated

Top 10 ZIP codes by initiations, among records with a resolvable property address (38135 and 38122 tie the last row at 8; both appear in the full table below).

ZIP resolved for 194 of 367 initiations and 144 of 157 completed sales; see methodology.