To a player, a Pokémon card is worth what it does in a game. To a collector, it is worth what it is: how rare, how old, and above all what condition it is in. Two copies of the same card are not the same to a collector if one has a soft corner or an off-centre picture. Collectors needed someone they trusted to say which was which, and grading companies grew up to do it.
A grader takes a card, checks that it is genuine, scores its condition on a scale that tops out at a perfect ten, and seals it in a clear plastic case with the score printed on a label. From then on the card trades as a graded card, with its score attached. That simple service has become an industry of its own, and over the last few years Pokémon has become its largest customer.
A big share of a small slice
In 2025 the major grading companies handled 26.6m cards of every kind, sports cards included, according to GemRate, a firm that tracks grading labels. 16.1m of them were Pokémon, 60.5% of the total. PSA, the largest grader, says most of its trading-card work is “most of them Pokémon”.
Set that against how many cards TPC prints, though, and the picture inverts. All that grading covers about 0.15–0.18% of a single year’s print, on our estimate. The collector economy that dominates the headlines runs on a sliver of the cards.
A big share of a small slice
Pokémon’s share of all cards graded, against the share of a year’s print that is graded.
Pokémon leads the grading industry, yet nearly every Pokémon card printed is opened, played or kept without ever being graded.
Built from and .How fast it grew
The habit has grown quickly. On PSA’s figures alone, Pokémon grading went from 1.4m cards in 2021 to an estimated 9.5–10.5m in 2025, roughly 6.2–6.8× in four years. Across all graders it at least doubled between 2023 and 2025.
PSA’s Pokémon grading, 2021 to 2025
PSA only. The 2023 figure is a floor because its sources disagree, and 2025 is an estimate.
- Pokémon cards graded by PSA
- 2023: other sources say up to
The growth is real, but bigger multiples quoted elsewhere usually join a PSA-only count for an early year to an all-grader count for a later one.
Built from and .Show the numbers
| Year | Pokémon cards graded by PSA, m cards | 2023: other sources say up to, m cards | Source |
|---|---|---|---|
| 2021 | – | ||
| 2023 | |||
| 2025 | (9.5–10.5) | – |
Pokémon cards graded by PSA. 2021 is PSA data via Axios. 2023 is a floor (PSA’s recap: more than 4.4m). 2025 is an estimate of 9.5–10.5m: PSA reports more than 11m trading cards, “most” of them Pokémon.
2023: other sources say up to. Three figures disagree for 2023: more than 4.4m (PSA recap), 5.8m (43% of PSA’s 13.5m, our calculation) and 6.4m (Axios chart). None is averaged.
The numbers need care. Some widely quoted growth figures compare a PSA-only count for an early year with an all-grader count for a later one, which overstates the rise. The 2023 figures disagree among themselves, from 4.4m in a PSA recap to 6.4m in a press chart, so the chart treats that year as a floor.
The scarcity inside the score
What collectors pay for is not only the card but the score. A perfect ten is rare. Of 4,993 first-edition Base Set Charizards that PSA has graded, 124 received a ten. That is the kind of scarcity a print run cannot create and a grader can: thousands of copies of one card, and a few dozen at the top grade. In February 2026 one of those tens sold at auction for $954,800.
The most extreme case combines both kinds of scarcity. The Pikachu Illustrator card was never sold in packs: only 39 were given to winners of an illustration contest in 1998. A graded ten sold for $16.49m in February 2026, the record for any Pokémon card.
Modern sets build some scarcity into the pack itself. Secret cards went from a rounding error to a quarter of the cards in new English sets. Cards numbered beyond a set’s printed total, the illustration rares, ultra rares and gold cards collectors chase, make up 31% of the Paldea Evolved card list and 28% of Phantasmal Flames, but only about 2% of the cards people open: about 7 in a booster box. These rates come from collectors’ own opening counts; TPC publishes no odds.
Secret cards went from a rounding error to a quarter of the cards in new English sets
Share of cards in regular English sets numbered past the printed set total, by era, from Bulbapedia set lists.
Before Sun & Moon secret cards were a handful per set. Since Scarlet & Violet they are about a quarter of the list.
Built from .Chase cards fill a big part of the list, a small part of the packs
Cards numbered beyond the printed set total: % of the card list against % of cards opened.
- Share of the card list
- Share of cards opened
Show the numbers
| Set | Share of the card list, % | Share of cards opened, % | Source |
|---|---|---|---|
| Paldea Evolved (2023) | |||
| Phantasmal Flames (2025) |
Share of the card list. Cards numbered beyond the set’s printed total, as a share of all the cards in the set. A count of different cards, not a pull rate.
Share of cards opened. Expected chase cards per pack divided by the 10 cards in a pack (the energy and code card excluded), from community pull-rate studies whose methods are not stated. TPC publishes no odds.
Which creatures get graded
Grading also shows which creatures collectors care about, and it gives two different answers depending on whether you count cards or prices. By count, on an all-time PSA table compiled by a vendor, Pikachu leads with 2.47m graded cards and Charizard follows with 2.34m. Together they are more than half of the top ten.
Who gets graded
All-time PSA-graded cards by creature, top ten, vendor estimate.
Pikachu and Charizard dominate the count. Behind them the list is long, with Eevee and its evolutions prominent.
Built from .Show the numbers
| Pokémon | Cards graded by PSA, all time to January 2026, m cards | Source |
|---|---|---|
| Pikachu | ||
| Charizard | ||
| Mew | ||
| Mewtwo | ||
| Umbreon | ||
| Eevee | ||
| Blastoise | ||
| Venusaur | ||
| Zapdos | ||
| Gengar |
Cards graded by PSA, all time to January 2026. Estimates from a grading-data vendor (PSA populations via PokéWallet/GemRate). Pikachu and Charizard are 56.3% of these ten.
By price, Charizard wins clearly. In the same February 2026 auction, the first-edition Charizard ten sold for 6.9–12.6× the Blastoise and Venusaur from the same set. A first-edition Pikachu from that set, also a ten, fetched a median of $19,700 across two sales that month in PSA’s own record, so Charizard fetched about 48.5× as much, and about forty times in December 2025, when Heritage sold another Charizard ten. Part of that gap is the card rather than the creature: that Pikachu is a common card in the set, while the three starters are shiny holographic ones (see Characters).
Who earns from it
Grading is a fee business, priced per card and by speed. PSA charges $59.99 a card for its standard service, more for faster ones. PSA no longer publishes its accounts, so its Pokémon grading income can only be modelled, at about $167–333m a year. The graded cards then trade on marketplaces. eBay told investors its Pokémon sales “grew in the triple digits year over year” in 2025, and that it was “expecting a deceleration in Pokemon growth” in 2026.
None of this reaches The Pokémon Company. TPC sells the pack at its list price and does not raise it when demand runs hot. The grading fee, the auction premium and the marketplace’s cut all belong to others.
What the queue can and cannot tell us
Because grading is so closely watched, its volume is often read as a gauge of collector demand. It is a useful one, with three limits. It measures the graders’ capacity as much as demand: PSA’s backlog peaked at about 7.5 months of its own output in 2026, so the counts are a floor. It mixes old cards with new, so it cannot say whether current sets are what collectors want. In 2026 grading volumes rose while the prices of that year’s sealed sets fell. And it attracts fakes. PSA’s own fraud report says counterfeit Pokémon submissions rose 125% in 2025.
What it adds up to
Grading is where the collector side of Pokémon becomes an industry: a score that turns one copy of a card into something rarer than its print run, a queue that has grown several times over, and fees, auctions and marketplaces built on top. It runs on a tiny share of the cards, and none of its money reaches TPC. For TPC its value is indirect: it keeps adult collectors engaged with the cards.