We Analyzed 10,288 Private-Party Listings in One Week. Here's Which Marketplace Actually Delivers Under-Wholesale Cars.

We took one real dealer — we'll call him Julian — and his actual Hooptie feed: the Bay Area, centered on San Leandro with a 50-mile radius. Over seven days in July 2026, Hooptie scraped every private-party vehicle listing across every major marketplace it covers, deduplicated them, decoded the VINs, and scored each car against both retail and wholesale values.

Then we did the thing most "cheap cars online" claims skip: we filtered to clean titles only, and we were conservative about it.

Here's the full week, every marketplace, every stage of the funnel:

Read that top to bottom and you see how brutal real sourcing is. Of 10,288 listings scraped in a single week, only 4,337 — about 42% — were verified private sellers. The other 58% were dealers, reposts, and junk. Fewer than half of what's posted is even a real private-party car.

Of those 4,337 real private listings, 718 clean-title cars were priced under retail — worth a look. And just 40 were priced under wholesale: the actual steals. That's 0.4% of everything scraped, roughly one deal for every 257 listings.

34 of the 40 clean-title under-wholesale cars came from a single marketplace — the biggest one. That's 85% of every real deal in the metro that week.

Here's the part that surprised us. The top marketplace was 73% of the listings scraped but 85% of the deals — it over-indexes. It isn't winning only because it's bigger; its deal density is the highest too. On the top marketplace, roughly one in every 220 listings was a clean-title steal. On the #2 marketplace it was one in 400, and on the #3 one in 500.

So the top marketplace wins on both axes at once: the most listings, and the highest chance that any given listing is actually a deal. If you only have time to work one channel, the data says it's not close.

The #2 marketplace delivered 4 deals and the #3 delivered 2. That's not nothing — six clean under-wholesale cars you'd have missed entirely if you only watched the top one. Across a full year, that's roughly 300 extra shots at a deal. But as a primary channel, neither can carry an acquisition strategy on its own.

There's a structural reason the #2 marketplace is a fifth of the top one's volume: it started charging $5 per vehicle listing back in 2019, which pushed sellers toward the free marketplaces. It still has a loyal Bay Area base, but the migration is real and it isn't reversing.

The takeaway: watch them, don't live on them. The right move is to have them covered automatically so those six deals land in the same feed — free upside you never have to think about.

Both produced zero clean-title under-wholesale cars. Not because they're bad sites — because they're dealer-inventory marketplaces, not private-party pools. Of the 133 and 59 listings on the two retail-focused sites that week, almost all were priced at or above retail.

Their real value is as comp sources: they tell you what a car should retail for. They are not where you find a steal. Any acquisition tool that counts them as "deal sources" is padding a number — the data says they belong in your valuation workflow, not your sourcing one.

There's an obvious objection to any "cheap cars online" claim: sure, they're cheap because they're salvage or branded. We pre-empted it by filtering to clean titles only, conservatively — and the result went the other way.

With the salvage and branded cheapies stripped out, the top marketplace's share of the real deals held at 85%. These 34 cars are genuinely clean and genuinely under wholesale — not cheap-because-broken. That's real, defensible front-end margin, not a title-status trap.

1. 1If you only have time for one channel, it's the top marketplace — by a wide margin, on both volume and deal density.
2. 2Don't ignore the #2 and #3 marketplaces. They quietly added 6 deals — 15% of the total — that you'd otherwise never see. Covered automatically, they're free upside.
3. 3Use the retail sites for comps, not sourcing. Zero deals says it plainly.
4. 4The real lesson is the funnel. 40 deals hid inside 10,288 listings across five platforms. No human scrolling the marketplaces a few times a day finds a 0.4% needle. That's the entire case for automating it.

One honest caveat: this is one dealer, one metro, one week — a snapshot, not a national census. But the shape of it — one marketplace dominant, two others supporting, the retail sites near zero — matches what nearly every acquisition buyer we talk to already feels. The data just puts numbers on it.

Forty clean-title cars priced below wholesale, in a single metro, in a single week — 85% of them on a single marketplace, and all 40 surfaced automatically. The dealers winning on acquisition aren't scrolling harder than you. They're letting software work the whole funnel — 10,288 listings down to 40 deals — while they spend their time actually buying cars.