How to Check What a Japanese Pokémon Card Is Actually Worth

Someone quotes you a number for a Japanese card and you have no way to check it. The English-language sources that come up are mostly people who are also selling you something, and the Japanese sources that hold the real data are not translated.

How do I find out what a Japanese Pokémon card is actually worth?

Read three numbers: what a Japanese shop asks for the card, what the same shop pays to buy it back, and what raw copies actually closed at on Yahoo! Auctions in the last 180 days. When the buyback price and the raw closed average agree, that is the card’s real value. When they don’t, your search is catching the wrong object.

Below is that done, with live figures pulled on September 16, 2026.

What is a Japanese Charizard card worth right now?

There is no single answer, and that is the first useful fact. A keyword search for リザードン on yuyu-tei (遊々亭) on September 16, 2026 returned 126 priced entries, from ¥30 for a common in a Charizard-branded trainer deck to ¥998,000 for the 104/S-P Charizard VMAX promo. Fifty-five of those 126 showed zero stock.

Here is the shop’s own ask next to what it pays to buy the same card back, read off both lists that day and matched by collector number.

Two things fall out of that table immediately. The buyback share is not constant: it runs 60% to 70% on the expensive vintage and Mega prints and drops to 31% on a common modern SAR, which tells you the shop is far more willing to warehouse money in the scarce cards. And the most expensive rows are the out-of-stock ones. A shop with none in stock has no buyer correcting its number.

Why does the same card show a ¥22,000 price and an ¥88,000 price?

Because you are looking at two different objects. On Yahoo’s closed search, リザードンex 134/108 -まとめ -セット returned 276 ended sales on September 16, 2026 with an average of ¥61,742, a number matching no real transaction. Splitting it: 146 of those sales were PSA 10 slabs averaging ¥87,457, and raw copies averaged ¥34,092.

Adding -PSA -ARS -鑑定 -BGS to the same query left 123 sales, and the header block changed to 最安 ¥13,750 / 平均 ¥34,092 / 最高 ¥83,300. The graded query ran ¥65,000 to ¥295,500. Those are two separate markets sharing one card name.

Note what the sold rows do that the header cannot. The ¥22,000 sale drew 35 bids and the ¥29,500 sale drew 37, so those are contested, real clearing prices. The ¥33,000 and ¥36,500 rows had a single bid each, which usually means a fixed 即決 price someone accepted. Both are data, but only the first kind is the market arguing with itself.

Mercari Japan is the natural cross-check, since sold and live listings sit side by side and the ratio tells you how fast copies move. We could not read it programmatically on September 16, 2026: its results are built in the browser, and the raw page returned no listing rows. The mechanics of navigating it as an overseas buyer are in Mercari vs Yahoo! Auctions for buying cards.

The shop spread: two published numbers that bracket the market

A Japanese card shop publishes a price for selling you a single and a price for buying that same single back, both openly, for thousands of cards at a time. yuyu-tei is the cleanest example: /sell/poc/... is the retail side, headed 販売, and /buy/poc/... is the buyback side, headed 買取. The same card exists on both, keyed by the collector number, so you can match them without reading Japanese.

One thing to know before you read the 買取 page, because it will otherwise confuse you: many rows show two prices, one bold and one struck through. The struck-through figure is the shop’s previous buyback price. It is not a damaged-copy tier and it is not a second grade. Of the 119 リザードン rows on that list on September 16, 2026, 103 carried a struck-through figure, every one of them on a row the page marks with a price-rise badge, and in all 103 the bold figure was the higher of the two. Read the bold one.

Coverage varies by shop. Suruga-ya’s (駿河屋) search page answered us on September 16, 2026 with the site frame and no item rows, so we could not get a second shop’s number here, and its buyback search states that trading cards go through its かんたん買取 flow only. Card Labo splits the two sides across two domains: retail on c-labo-online.jp, buyback keyword-searchable on c-labo-kaitori.jp. Card Rush returned 403 to every request we made, so treat it as unavailable for research rather than assuming what it shows.

How fast do these numbers actually move?

Fast enough that a price you read six weeks ago is not a price, and the buyback page proves it on itself. yuyu-tei’s 買取 row for リザードンex 134/108, the worked example in this article, reads ¥35,000 with ¥15,000 struck through beside it. The shop has more than doubled what it pays for that card, and it publishes both numbers on the same line.

That is the strongest evidence on this page, because it is one source showing its own movement rather than two readings you have to date yourself. The same pattern ran down the whole list: メガリザードンXex 110/080 read ¥70,000 against a struck-through ¥35,000, 116/080 read ¥120,000 against ¥80,000, 139/108 read ¥6,500 against ¥4,500, and 331/190 read ¥2,000 against ¥1,000.

For a set-wide view, we published Storm Emeralda’s yuyu-tei numbers on August 3, 2026, three days after release. Re-reading the same two pages on September 16 gave a completely different set.

The chase card fell 66% in six weeks, and the shop went from having none to having four. That is the ordinary shape of a set’s first two months as packs get opened, and it is why every number on this page carries a date. Full release-week detail is in Storm Emeralda prices in Japan.

The Japanese you have to type, and what 美品 really means

You do not need to read Japanese to use these pages. You need about a dozen strings.

Condition vocabulary is where the number moves most, and it is worth being clear about what it is not. 美品 is a seller’s adjective with no defined tolerance, no third party behind it, and no relationship to a numeric grade. Two sellers can both write 美品 for cards a full grade apart. 傷有り and 初期傷 are more honest signals precisely because a seller gains nothing by writing them: in our 134/108 rows, the two listings carrying damage wording still cleared at ¥30,250 and ¥36,500, above the raw average, which tells you those particular buyers were pricing the card and not the wording.

If the card is valuable enough that the raw-versus-graded gap is worth chasing, the routes and costs are in PSA grading for Japanese cards.

Why is the JP number never your number?

Because it is a domestic price with domestic costs baked into both sides of it. Mercari’s fee page states a 10% 販売手数料 on the sale price plus a flat ¥200 bank withdrawal. Yahoo! Auctions charges the seller a 落札システム利用料 of 10% of the winning price, tax included, with ¥0 monthly and listing fees. Bidding and winning cost the Japanese buyer nothing. So a ¥34,000 clearing price leaves that seller about ¥30,600 before shipping.

Your side runs the other way. A Japanese domestic price excludes the proxy’s per-item fee, domestic shipping to a warehouse, international postage, payment charges, and whatever your customs regime does on arrival. That stack is why a JP sold price and an eBay or TCGplayer price for the same card are not comparable, and why the direction of the difference is structural rather than a bargain. The arithmetic, with a worked basket, is in is it cheaper to buy in Japan than the US.

What a finished reading looks like

Written out for the card in this article, as of September 16, 2026: Charizard ex SAR 134/108 from 黒炎の支配者 asks ¥49,800 at yuyu-tei with 2 copies in stock, buys back at ¥35,000, and raw copies cleared 123 times on Yahoo! Auctions in the last 180 days at an average of ¥34,092, in a band of ¥13,750 to ¥83,300. PSA 10 copies are a different market at ¥87,457 average. Before export costs.

That is four numbers, a count, and a date. It is the honest limit of what public Japanese data can tell you, and it is far more than the number someone quotes you with nothing behind it.