Almost every piece of advice about buying a used car rests on one hidden assumption: that you can look at similar listings and work out what the thing is worth.
For most of the country that assumption is false, and the numbers are more lopsided than they look from inside a big city.
| Market position | Listings |
|---|---|
| Largest metro | 16,376 |
| Median market | 424 |
| Bottom fifth | under 83 |
Sixteen markets out of 190 hold half of everything on sale. The median market carries 424 listings — total, across every make, model and year.
What 424 listings means in practice
Divide it the way a buyer actually shops. Suppose you want a midsize SUV, five to eight years old, under 120,000 miles, under fifteen thousand dollars.
In a market of 424 that filter returns somewhere between two and six cars. Not two or six of the model you had in mind — two to six that meet the whole description at all.
At that point the standard advice stops working. You cannot tell whether an asking price is high, because there is no distribution to compare it against. Three data points do not have a middle. And you have no negotiating position, because the seller knows exactly how thin the local supply is.
The trap in the price averages
Thin markets post higher average prices, and it is tempting to read that as small-town sellers charging more.
Mostly it is composition. Large metros accumulate cars over decades, so their pool skews old and high-mileage, which drags the average down. Smaller markets turn over faster and what is on offer at any moment is newer.
So the big-city discount is largely a discount on older cars, not on the same car. What the big market genuinely offers is not a lower price — it is twenty examples instead of two.
How to price a car in a thin market
Build the comparison from outside your area, then adjust. Pull listings for your exact model, year and mileage band from three or four large metros, and use that spread as the reference distribution.
The instinct to widen the specification instead — accept any year, any mileage, just get to ten listings — introduces far more error than the geography does. A three-year age gap moves the price more than a state line.
Then check what the local market pays for the same class of vehicle overall, and shift the reference by that much. It is rough, and it is a great deal better than negotiating against three listings.
When the drive is worth it
Travelling to a bigger market pays when the model you want is unusual, when you need a specific configuration, or when the local examples are all at the wrong end of the mileage range.
It pays less when you want something ordinary. A Camry is on sale everywhere, and the local price for one will not be far off the metro price once you account for the age difference.
Either way the remote purchase has one weakness: you get one visit, and the decision compresses into that visit. Doing the paperwork before you set off rather than after you arrive removes most of the risk — a free VIN check on the listing's VIN tells you whether the drive is worth making.
Frequently asked questions
Why are there so few used cars for sale in my area?
Because supply follows population density very steeply. Across 190 markets in our data the largest carries 16,376 listings while the median carries 424 — a factor of thirty-nine. Half of all listings sit in just sixteen metros, so most of the country is shopping in a genuinely thin market.
How many listings do you need to judge a fair price?
Enough of the same model, year and mileage band to see a distribution rather than a handful of numbers — realistically a dozen or more. In a market with 424 total listings across every model, a specific car you want may have two or three comparable examples, which is not enough to tell a good price from a lucky one.
Is it worth travelling to a bigger city to buy a car?
For choice, frequently yes. For price, less often than expected — large metros post lower averages mainly because their listings are older and higher-mileage, not because sellers ask less for the same vehicle. The real gain is having twenty examples to choose between instead of two.
How do I price a car when there are no comparable listings?
Widen the geography rather than the specification. Pull comparable listings from several large markets to build the distribution, then adjust for what your local market actually pays. Loosening the year or mileage instead introduces more error than the distance does.
Denis Kataev is a serial solopreneur and the founder of ForCar. With 15+ years in software engineering and 10 years in SEO, he builds data-driven products end to end — backed by a sharp eye for design. At ForCar he mines proprietary vehicle datasets, turning raw numbers into buying advice you can actually trust.
Keep reading
All articles →
9.5% of the Market Is Under $3,000. Here Is What You Get.
The bottom of the market is bigger than most buyers assume and far less random than it looks. The cars down there are strikingly consistent — same models, same age, same mileage — which tells you what actually survives to get there.
There Are 20 McLarens for Sale in America. There Are 71,161 Fords.
A used market is supposed to be a mirror of what people bought. Sort 413,888 listings by badge and the mirror turns out to have some very thin spots — and a spelling problem nobody talks about.
502 Used Cars Are Genuinely New. Another 664 Just Say So.
Craigslist lets the seller choose the condition from a dropdown. Nobody checks it against the odometer. We did, across every listing we hold, and the word "new" turns out to mean almost nothing.
Mark ForCar as a preferred source and our data shows up first in your Google results. One click, no account.