One Piece Graded Cards: PSA Populations, Gem Rates & Premiums (September 2026)

Live PSA, BGS and CGC population data for the One Piece Card Game: 1,637,624 PSA slabs, an 87.2% gem rate — and new slabs grading at 83.0%, which is why that rate is falling.

The One Piece Card Game has 1,637,624 cards in PSA holders, 296,396 in BGS, 123,040 in CGC and 1,948 in SGC, across the 3,404 distinct graded entities Cardlogs tracks population for. According to Cardlogs population data, 87.2% of every graded One Piece card is a PSA 10 as of September 10, 2026 — but the roughly 105,000 slabs added in the last 30 days came back 10s only 83.0% of the time, and that is the number this edition is about.

Is the One Piece gem rate falling?

Yes — slowly, measurably, and for the first time visible in our own series. On a fixed cohort of 1,139 cards the stock PSA 10 rate moved from 87.68% to 87.31% between August 12 and September 10, 2026. A third of a point in a month sounds like nothing until you ask what has to be true for it to happen: the new slabs entering the population have to be grading worse than the ones already in it.

They are. We call that the marginal gem rate — PSA 10s added divided by total slabs added, over a fixed cohort — and it is the honest way to see a stock number that moves slowly by construction:

Take the 30-day figure as the signal and the 7-day one as what it is: a short window on a huge base, where one grader release moves the ratio fifteen points. The direction is the same in all three, and all three sit well below the 87.68% the stock started at.

This is not a One Piece story. Running the identical measurement on Riftbound over the same 30 days: +24.19% population on a 125-card cohort, marginal gem rate 84.6%, stock rate down 86.70% to 86.30%. Two games, different ages, different volumes, same shape. A graded population that is still growing fast is one whose average quality is being diluted by its own newest submissions — the easy, mint, straight-from-the-pack copies get sent first, and what follows is what was left in the binder.

A correction to our August edition

The last edition of this report said the opposite, and it said it from these same numbers: "The PSA 10 count grew at almost exactly the same rate as the total (+10.8% against +11.1% over 30 days), so the gem rate is holding as volume scales rather than degrading."

That inference was wrong in method, not in arithmetic. When a subset and its parent grow at nearly the same percentage, the marginal rate is not nearly the stock rate — it is levered. A 0.3-point gap between +10.8% and +11.1% implies a marginal gem rate of about 85.2% against a stock of 87.6%, a 2.4-point gap the percentages hid completely. The August edition had the evidence for a falling gem rate printed in its own growth table and read it as stability. A ratio's trend has to be computed as a ratio of the increments, never inferred from two growth percentages that look close.

How many One Piece cards have been graded?

2,059,008 in total across PSA, BGS, CGC and SGC, as of September 10, 2026. The split:

PSA holds 80% of the graded One Piece market. BGS takes 14% and earns it at the top end, where the four-subgrade-10 Black Label is the only slab in the hobby that reliably prices above a PSA 10. CGC is the fastest-moving of the three: 99,955 slabs in the August edition, 123,040 today, a 23% jump in two weeks against PSA's 8%.

The language split is the number most published figures miss:

454 Japanese printings carry nearly two thirds as many slabs as 2,549 English ones. That is submission concentration, not a bigger catalog: the Japanese graded market is a handful of chase printings graded in enormous volume, led by a single Luffy at 50,745 slabs. The English and Japanese gem rates have converged since August — 87.3% and 87.8% then, 87.2% and 87.3% now — which is the dilution above landing harder on the Japanese side, where the volume is.

That two-million total is the number worth sitting with. Under four years after the game launched, One Piece has produced more graded cards than most trading card games ever will:

All four rows run the same de-duplication and the same ladder screen described in the methodology, so they are directly comparable. One Piece's PSA population is 52 times Riftbound's — down from 61 times three weeks ago, which is Riftbound growing three times faster off a base a fiftieth the size — and 7.5% of vintage Pokemon's. And it still holds more BGS Black Labels than vintage Pokemon does, 12,962 against 7,892, on a population seven percent the size. The gap widened this edition. Black Labels are a modern-printing phenomenon, and a three-year-old game has produced more of them than sixty years of Pokemon.

What is the PSA 10 gem rate for One Piece cards?

87.2%, drifting down about a third of a point a month for the reason above. Cardlogs computes the gem rate as PSA 10s divided by that card's total PSA population — the share of submissions that came back perfect.

Across every mainline English booster expansion, the whole range is nine points wide:

Age explains almost none of it. Romance Dawn, the launch set from December 2022, gems at 87.1% — better than Carrying on His Will from November 2025. Ignoring The Time of Battle, whose 2,254 slabs are still too few to trust, the best-grading set is 500 Years in the Future from summer 2024 and the worst is Pillars of Strength from summer 2023: one year apart in the release calendar and 7.2 points apart in gem rate. Whatever moves this number, it is a print-run property, not a shelf-life one.

The number that actually stands out is the cross-game one. On the same de-duplicated basis, vintage Pokemon returns a 46.4% PSA 10 rate against One Piece's 87.2%. A One Piece card sent to PSA comes back a 10 about seven times in eight; a vintage Pokemon card does it slightly less than half the time. That is the whole modern-versus-vintage grading argument in two numbers, and it is why "pop 500 PSA 10" means something very different in each game.

Which One Piece cards have the highest PSA populations?

Monkey.D.Luffy (ST21-014) (V.1), the Japanese printing, at 50,745 PSA submissions — more than the entire Riftbound game, twice over.

Nine of the twelve are Luffy. All twelve are promos or Japanese printings rather than English pack-pulls, and six are Japanese. The graded market for One Piece is not a market for the game's best cards — it is a market for its most distributed collectibles, and the two lists overlap far less than a price chart would suggest.

Row 4 needs naming honestly, because we got it wrong last time. That population record covers three catalog rows across two different card numbers — P-006 from Tournament Pack Vol. 1 and two ST10-006 printings — so no single card name describes it. The August edition printed it as "Monkey.D.Luffy - P-006 (Tournament Pack Vol. 1)", which is one member of the group presented as if it were the card. The population is real and correctly totalled; the label was not. It is the same hazard the premium screen below exists for, showing up in a table where we had not applied it.

Two other entries are worth calling out. The PSA Magazine Luffy at 8,773 submissions is a promo distributed by the grading company itself, about as clean a demonstration of induced supply as this hobby offers. And Monkey.D.Luffy (P-033) gems at 76%, the worst rate in the top twelve — nearly a quarter of the 13,074 people who sent one in got back a 9 or lower.

How fast is the One Piece graded population growing?

+8.55% in the 30 days to September 10, 2026, on a fixed cohort of 1,139 cards — about 105,000 new slabs. Measuring this correctly is not optional: a naive sum of nightly populations moves with how many cards were sampled that night, which invents trends in both directions.

According to Cardlogs population data, roughly 17,000 One Piece cards entered PSA holders in the single week to September 10, 2026. Growth is decelerating: the comparable 30-day figure in the August edition was +11.1%.

One screen is new this edition, and it is the reason the growth numbers above can be trusted. A fixed cohort that only checks a card was present on both endpoint dates still admits a card whose per-grade ladder went missing on one of them — and a naive sum reads a missing PSA 10 key as zero, which looks like a population collapsing. It happened here: Monkey.D.Luffy (037) (1st Anniversary Tournament) kept its 4,061 total on September 9 and 10 but lost its grade breakdown, and that single card alone dragged the measured 7-day marginal gem rate from 65.3% down to 47.1%. Every window in this article now requires both endpoints to carry a populated ladder, which drops 60 entities holding 6,339 slabs — 0.4% of the PSA total. The headline gem rate runs on the same basis; without it, it would read 86.9% instead of 87.2%.

Is it worth grading a One Piece card?

Median 2.20x the raw price on the 387 cards where the premium can be attributed cleanly, out of 3,404 in the report — down from 2.32x on 334 cards three weeks ago. The screens behind that number are strict and are explained in the next section. Here are the twenty-five biggest multiples:

That is the top twenty-five of 387 — unlike our Riftbound report, which publishes its complete list because it fits on a page, this one is a slice. Median 2.20x, mean 2.75x, range 0.75x to 14.52x, middle half between 1.65x and 3.34x.

Read the top two rows with their populations attached. Gorgon Sisters at 14.52x has thirteen PSA 10s in existence and passes corroboration on a single Cardmarket leg, with no eBay sold to check it against; Boa Hancock (Alternate Art) at 14.03x has thirty-nine and is corroborated on both legs. Neither is wrong, and neither is a rate to plan around — a 14x on a pop-13 card is one seller's ask meeting one buyer. The August edition's leader, Tony Tony.Chopper (Winner Pack 2024 Oct.-Dec.) at 12.07x, is gone from this table entirely: its Cardmarket price now reads $54.05 against a $26.98 eBay sold, a 2.0x disagreement, and our own venue-split screen removed it.

Six cards of 387 sell for the same or less graded than raw, worst first: St. Marcus Mars (Parallel) at 0.75x, where a $492.23 raw card is $368.00 in a PSA 10 holder, then Otama (Parallel) at 0.87x and Tony Tony.Chopper (Store Championship 2023) at 0.94x.

Scarcity is beginning to show a gradient — with a caveat. Sorting the same 387 cards by how many PSA 10s exist:

The scarcest bucket now leads the most abundant one by a full point, where three weeks ago it trailed it. Do not over-read that: the top bucket holds 30 cards against 275 in the bottom one, it moved from 2.50x to 3.11x on eleven new admissions, and the middle two buckets are still inverted. What is stable across both editions is the bottom row — a pop-400+ card medians at 2.10–2.20x whichever way the thin buckets wobble. Low population still signals low demand at least as often as it signals scarcity, which matters if you are paying up for a "pop 40" slab and expecting the number to do the work.

The practical read: at roughly $20–30 all-in per card for grading and shipping, a 2.20x median only clears its cost on cards already worth about $30–40 raw. Below that, an 87% chance of a 10 is not the question — the fee is. And note where the multiples cluster: alternate arts, parallels and event promos in the $20–$90 range, not the four-figure chase cards, whose raw prices already embed most of the condition premium.

Why we won't quote a premium for most One Piece cards

3,017 of the 3,404 entities in this report are excluded from the premium table. The buckets, in the order the screen applies them:

The third bucket is the One Piece-specific one, and it is worse here than in any game we have measured. One Piece prints the same card number many ways — a base common, an alternate art, a manga art, an SP, a gold SP, a tournament-winner version — and graders do not always separate them in a population report. 354 of our 3,404 graded entities cover more than one catalog printing, and 302 of those either span printings whose raw prices differ by more than 1.5x or share a card number with one that does.

"By more than 1.5x" understates it badly:

The consequence is unchanged from earlier editions and worth stating plainly: eleven of the twelve most-graded cards in this article are excluded from the premium table. The cards the world submits most are precisely the cards whose number also covers a bulk common, so their raw price and their slab price cannot be safely divided.

A caveat we cannot screen away, and it grew. 141 rows in our One Piece data carry a PSA 10 market price while holding no PSA 10 in their population ladder at all — up from 82 in the August edition — topped by a $29,100 serialized Roronoa Zoro whose PSA population is zero. A graded price with no graded population behind it is an asking price for an object that may not exist in that holder. Those rows are excluded from every table here, which is why the most-valuable-PSA-10 figures in this article are drawn only from cards with a real PSA 10 count: Monkey.D.Luffy - ST01-001 (Serial Number) at $61,851.66 across 1,319 PSA 10s, Monkey.D.Luffy - EB04-061 (Serial Number) at $45,600.00 across 643, and Monkey.D.Luffy (118) (Red Super Alternate Art) at $30,588.87 across 313. The EB04-061 figure was $54,100 three weeks ago — a 16% fall on a card whose population rose, which is the only kind of graded-price move we can evidence at this end of the market.

We publish the screen rather than the number it removes. It is the same discipline behind the corroboration rule in the monthly One Piece market watch.

Where this data comes from

Cardlogs aggregates prices nightly from TCGplayer, Cardmarket and eBay sold listings (via PriceCharting) across 12,524 tracked One Piece cards, plus PSA and BGS population data from GemRate and graded market prices covering PSA, BGS, CGC, SGC and TAG.

Population coverage is partial by design. 3,834 of those 12,524 rows carry a population record, which becomes 3,404 distinct graded entities after de-duplication. 976 of the 12,524 have a raw market price of $30 or more — so this is not a chase-card-only sample, but it does under-cover bulk commons, which are graded rarely and priced in cents.

Six screens run before any number above is published:

1. De-duplication on population identity. A graded population belongs to a population record, not to a catalog row, and one record routinely covers several printings of the same card. Every population figure counts each record once: 3,834 catalog rows collapse to 3,404 distinct graded entities, and the PSA total falls from 1,839,345 to 1,637,624. The same method runs in our Riftbound and Dragon Ball reports, so the cross-game table is directly comparable.
2. Ladder completeness. A record can carry a valid PSA total while its per-grade breakdown is momentarily absent. Every gem-rate and growth figure here requires a populated PSA 10 key, because summing a missing key as zero reads as a population collapsing. This drops 60 entities and 6,339 slabs, 0.4% of the PSA total, and moves the headline gem rate from 86.9% to 87.2%. The same screen applied to vintage Pokemon moves its rate from 45.6% to 46.4%, so the cross-game row needs it more than we do.
3. Prices are never deduplicated. Populations collapse across sibling printings; prices do not, because each printing has its own market. Every price in this article is read from the individual printing it belongs to.
4. Fixed-cohort growth. Population change is measured only across cards with readings on both endpoint dates, so a night that sampled fewer cards cannot masquerade as a decline — or, worse, a night that sampled fewer bigger cards inflate the growth rate. The marginal gem rate is PSA 10s added divided by total slabs added across that same cohort.
5. Cross-venue corroboration, both ways. A raw price is only quoted when Cardmarket (converted at the day's ECB rate, $1 = €0.86088) or the eBay sold median lands within 30% of the TCGplayer price, and — where both exist — when those two agree with each other to within 1.5x.
6. Sibling-print attribution. A graded premium is only quoted when the raw price and the slab plausibly describe the same object: the card's population record must resolve to that one printing, and no printing sharing its card number may sit more than 1.5x away in raw price without that resolution. 283 entities fail this.

Population figures are as of September 10, 2026; prices as of September 11, 2026. Graded prices come from a single source and cannot be cross-checked the way raw prices can — treat a graded price with a small population behind it as an estimate, not a quote.

We publish population data, not grading advice, and we do not sell slabs.

What to watch next

The marginal gem rate is now the number this report exists to track. If 83.0% is the new run rate, the headline gem rate reaches 86% around March 2027 without a single card grading differently — the stock simply gets averaged down. If it recovers above 87.7%, the August dip was a submission-mix artifact and we will say so. Either way the series is more informative than the level, and we will print it every edition.

Two other things move over the next quarter. The Time of Battle has 2,254 slabs across 22 printings today, up from 1,262 three weeks ago, and will be the first set we can watch gem from release rather than retroactively — its 92.6% is still on too little volume to believe. And the Black Label count is the one figure here that behaves like real scarcity: 12,962 across not quite four years, already 64% more than vintage Pokemon has produced in sixty.

If you are grading One Piece today, the conclusion is unglamorous and slightly worse than last month's. The gem rate is high enough that grading is barely a gamble, but it is drifting down, the populations are large enough that they are not doing the work your price assumes, and on anything under about $40 raw the fee eats the premium. The interesting money is in the $20–90 alternate arts and parallels, not in the cards everyone already slabs.

You can track raw, graded and sealed One Piece prices side by side — TCGplayer, Cardmarket and eBay sold, with PSA and BGS populations on every card page, English and Japanese printings alike — on Cardlogs, on the web and on iOS and Android.