Search for almost anything secondhand and you'll get two different kinds of numbers back: what people are currently asking for it, and what people actually paid for it. They are rarely the same, and the gap between them is where a lot of pricing mistakes come from.

Asking prices sit above what buyers actually pay

An asking price is a request, not a transaction. Sellers anchor high for obvious reasons — you can always lower a price, but you can't raise it after someone's already agreed to buy — so active listings skew upward from what the market will actually bear. Worse, a lot of what shows up in a search for "current listings" isn't current at all: listings that never sold and never got taken down, sitting at a price nobody wanted, for months. Average those in and the picture gets even further from reality. If you price an item by matching the highest active listing you can find, you're not pricing to the market, you're pricing to the most optimistic seller in it.

Completed sales are the ground truth

A completed, sold transaction is different in kind, not just degree. It means a real buyer looked at the item, the price, and the condition, and decided yes — and then paid. That's the only step in the whole process that involves an actual buyer voting with money. Everything upstream of that (a listing going up, a price being set, a description being written) is one person's guess. The sale is the only fact. If you want to know what something is worth, you want the list of what it has sold for recently, not the list of what people currently wish it were worth.

Median beats a cherry-picked outlier

Even inside completed sales, one number can mislead you. A single sale at a surprising price — a bidding war, a rare bundle, a buyer who overpaid — isn't representative, and neither is the single worst sale someone got stuck accepting. Look at the spread of recent sold prices for the same item in the same condition and take the middle, not the extreme. See how to find out what your stuff is worth for the full process of collecting and reading that spread.

Retail deserves its own lane

There's a third number worth keeping separate from both of the above: what a retailer charges for a brand-new one. It's tempting to blend retail into the resale picture, but they answer different questions — one is "what did people just like me actually pay secondhand," the other is "what does it cost to buy this new today." Blending them muddies both answers. Keeping retail in its own lane also surfaces something genuinely useful: sometimes a used item sells for more than a new one does. Across Price×Point scans, roughly one in six items resells above what a retailer charges for a new one — usually because the item is out of production, in short supply, or has become collectible. You'd never spot that pattern if resale and retail prices were averaged together.

How to read a price you find online

Before trusting any number, ask which kind it is. Is this what people are asking, or what people paid? Is it one data point or a spread of several? Is it for the exact item in the condition yours is actually in, or a rough approximation? A price that can't answer those questions is a guess wearing a dollar sign. A price that traces back to a handful of real completed sales, for the right item, in the right condition, is something you can actually rely on.