Why your Pokémon card sells for less than the price guide says

Say you looked your card up, a site told you a hundred dollars, and the best offer you have had is thirty-five. Nobody is trying to rob you. The two numbers are measuring different things.

Here are the reasons that gap opens up, roughly in the order they turn out to be the culprit, and what to do about each one.

A guide price and a sale are not the same measurement

Start with the thing underneath all seven reasons below. A price guide gives you a number computed from something: listings, past sales, a blend of the two, weighted somehow. A sale is one event, on one day, between two people, for one physical object.

The IRS, of all sources, has the cleanest statement of which one to trust. On fair market value, Publication 561 is flat about it: "the best evidence of FMV depends on actual transactions and not on some artificial estimate."

So when a guide and a sale disagree, the sale wins. The interesting question is why they disagreed, and it is usually one of the following.

It is a different card than the one you priced

This is the most common answer by a distance, and it is almost never obvious to the person it is happening to. One collector number can exist as a plain card, a reverse holofoil and a holofoil, and those sell for different amounts. The same Pokémon with the same artwork gets reprinted across sets. A promo with a letter-prefixed number is not the set card it looks like.

If the number you found came from a search on the card's name, assume it is wrong until you have checked the collector number, the set code and the print variant. Reading those off the card takes about ten seconds once you know where they are.

Condition, and the fact that yours is not mint

Guide prices generally describe a clean copy. The card that has been in a binder since 2016 has soft corners, and the one that lived in a deck has more than that. Whitening along an edge, a scuff across foil, a bend you can only see when you tilt it: each of those is a discount, and buyers looking at a photo will assume the worst about anything they cannot see clearly.

This one is not fixable, only accountable. When you pull your comps, keep the sales whose photos look like your card and drop the ones that do not.

The number you found belongs to a graded copy

A graded card is sealed in a slab with a number on the label. A raw card is the loose card in your hand. They are separate markets, and a guide that mixes them, or a search you ran that quietly included them, will read high.

It is worth being careful about how much higher, because the hobby repeats multipliers that nobody sources. The clearest published comparison we could find is not even Pokémon: writing for The Hockey News in August 2024, Chris Gawlik lined up sales of Connor Bedard rookie cards and found a PSA 9 averaging $347 against raw copies averaging $358, and on another card raw sales at $4,438 and $4,357 followed by a PSA 9 at $4,049 less than ten days later. The graded copies sold for less. The article's point is that the old rule of thumb, a PSA 9 being worth about a quarter more than the raw card, had stopped holding.

Grading also is not free or fast. Reporting in Sports Illustrated's collectibles section in August 2026, Brendan Darr put PSA's Regular service at $79.99 a card with estimated turnaround extended to 70 to 80 business days, after PSA paused its cheaper Value tiers in June 2026 to work down a backlog. Whatever the premium on a graded copy is for your card, the fee and the wait come out of it first.

The number is an average, and one sale was strange

Say a card sold nine times recently, mostly in the thirties and forties, and once for seventy-eight because two people wanted it on the same evening. The average of those nine sits well above the middle of them. The median does not move.

So if the guide you used publishes an average, one lucky sale is inside your number, and it is not coming to your card. Why the median is the one to take has the statistics and a picture of it happening.

The number came from listings, not sales

Anyone can list anything at any price. A page full of copies at ninety dollars tells you that several people would like ninety dollars. It tells you nothing about whether anyone pays it, and cards sit unsold for months at prices their owners are attached to.

Sold listings on eBay are the ordinary place to check instead, and there are two catches. The filter for them lives on the desktop advanced search form rather than in the filter rail most people use, and the public window runs back only 90 days, so a card that trades rarely can come back empty. How to pull sold listings, and what they leave out covers both.

Fees and shipping come out of your side, not theirs

A sold price is what the buyer paid, not what the seller kept, and eBay counts handling, shipping and tax inside the total its fee is charged on. So a card listed cheap with expensive postage and a card listed higher with free postage can be the same sale.

So a card that sold for forty dollars did not put forty dollars in anyone's pocket. Subtract the platform fee, the postage, the sleeve, the toploader and the envelope. On a cheap card that arithmetic is most of the price, which is the real reason a shop offers you so much less than the comps: they have to sell it again, through the same fees, and wait.

The card barely trades

Some cards sell every hour. Some sell twice a year. For the second kind, three months of sold data can be empty, and a guide number built on very little is a guess wearing a confident font.

• An empty sold search is not a zero. It means you have to look further out: auction house results, and older sales.
• A handful of sales is a hint. A dozen close together is a price. Always note how many sales your number came from, because that is most of how much to trust it.
• For anything above the cheap end, auction houses matter. Cards sell at Goldin and PWCC as well as on eBay, and for a higher value card those results are often the better evidence, because more than one bidder set them.

What to actually do

Nothing in the list above needs a subscription or a spreadsheet. In order:

1. 01Pin down the card: name, collector number, set code, print variant. Here is where each of those is printed.
2. 02Search sold listings rather than active ones, and take out the lots and the slabs.
3. 03Take the median of what is left and count how many sales it came from.
4. 04Check the cheapest live ask separately. It is a different question with a different use.
5. 05Subtract fees and postage before you decide what selling it is worth to you.

That is the whole method, and it is what looka automates. Hold the card up to your webcam and it identifies the card, then shows the median of recent sold listings from eBay, Goldin and PWCC alongside the lowest live ask on eBay, with the individual sales listed underneath so you can see what the median is made of. Free, no account, and it runs in the browser you have open.