Every number on this page is a direct read of the CBCS Population Report, pulled on July 30, 2026. Nothing is modeled, smoothed, or estimated. Where the census is ambiguous, this piece says so rather than papering over it. The full row-by-row data lives on the census page; this is what it means.
Collectors argue about manga scarcity constantly, and almost always from vibes. Everyone knows a high-grade One Piece Vol. 1 is hard. Nobody could tell you how hard, because the population data was locked inside a lookup tool nobody had ever read end to end. So we read it.
CBCS stopped accepting submissions on April 17, 2026. That turns its population report from a moving target into something rarer: a closed set. The numbers here will not grow. What was graded is what will forever have been graded under this label, which means this census isn't a snapshot that ages. It's a final count. That alone made it worth capturing before the report disappeared behind a defunct service.
Across 25 grails the census holds 3,982 slabs. That is the entire graded universe for the most important first-print manga in the hobby. Here is what surprised us.
01Nobody has a perfect one
Not a single grail we checked has a CBCS 10. Not one has a 9.9 either. Across all 3,982 books, the highest grade that exists is 9.8, and even 9.8 is genuinely scarce: 134 of them total, spread across the whole set. For most of the twenty-five, no 9.8 exists at all; the census caps at 9.6, or lower.
To be precise, because precision is the point: CBCS 10s do exist in manga generally. They just don't exist on these books. The debut volumes that matter most, the ones people actually chase, top out well short of perfect. Dragon Ball Vol. 1, with 346 slabs on file, caps at 9.4. Akira caps at 9.4. Bleach, 231 graded, never breaks 9.6.
This reframes what a "high-grade" manga key even is. In the American comic world, a 9.8 is a strong copy but hardly rare; population reports run to thousands. Here, a 9.8 first print is often a low-double-digit or single-digit object, and the grade above it is empty. The ceiling is lower and harder than the market prices it, because the market has been guessing.
02The One Piece number almost nobody believes
One Piece Vol. 1 is the blue chip, and its census reflects that: 1,038 first-print slabs, more than any other book on the list by a wide margin. You would assume that with a thousand-plus graded copies, high grades would be reasonably available.
They are not. Of those 1,038 copies, exactly 2 are 9.8s. No 9.9. No 10. Two copies, in the entire graded world, sit at the top of the most important manga key there is. Everything else grades lower, most of it far lower.
That is the single most valuable line in this whole census, and it's the kind of fact that only falls out of reading the data rather than sampling it. A book can be simultaneously abundant in graded population and almost nonexistent at the top grade. Volume and scarcity are not the same axis. One Piece has a deep census and a razor-thin peak, and those two facts together describe the book far better than either does alone.
03The top of the market is a few dozen books wide
Here's where the census overturns intuition hardest. The books with the most 9.8s are not the old, iconic keys. They're the recent ones. Chainsaw Man leads the entire set with 33 copies at 9.8. Demon Slayer has 31. Jujutsu Kaisen, 28. My Hero Academia, 22. Attack on Titan, 13.
The pattern is clean and it makes physical sense once you see it. Modern books were printed in the smartphone era, bought by collectors who already knew to bag and board them, and sent to grading fresh. Older keys, the 1990s and early-2000s debuts, survived as reading copies that got loved, shelved, and worn. So the recent books have the pristine survivors, and the vintage books have the history. Chainsaw Man has 45% of its census at 9.6 or better. One Piece has 2%.
Which means the "top of the market," the entire supply of genuinely high-grade first-print manga, is not a broad tier. It's roughly a few hundred books total at 9.6-or-better across every grail combined, and only 134 at 9.8. That is a smaller high-grade universe than a single popular American comic issue. The whole graded manga apex would fit in one shelf.
04The rarities hiding in plain sight
Some of these census entries are so thin they barely exist. Hunter x Hunter Vol. 1, a first print of one of the most respected shonen series ever made, has a single graded copy on file. One. Spy x Family Vol. 1, also one. Fruits Basket, three. Battle Angel Alita, three.
These aren't obscure titles. They're major series with tiny graded footprints, either because the raw books are genuinely hard to source in Japan or because almost nobody has bothered to grade them yet. Both explanations point the same direction: the graded market for these is embryonic, and the first person to build a high-grade run of them is building something the census says does not currently exist.
05The part where the data fights back
A census is only as honest as its willingness to show its own seams, so here are ours.
The label vocabulary is inconsistent. CBCS splits each book into printings, which is exactly what you want, since a first printing and a fortieth are different objects. But the labels aren't standardized. Fullmetal Alchemist Vol. 1 appears under both a "1st Edition" entry and a separate "1st Printing" entry, terms that may or may not mean the same physical book. Uzumaki has the same problem. We logged these separately and flagged them rather than merging them on a guess, because merging would have manufactured false precision. Anyone using this data to value a book needs to know the census itself isn't always sure what it's counting.
One-Punch Man has no Japanese entry at all. The census shows English Viz printings and foreign editions, but no Shueisha first print. We don't know why, and we're not going to invent a reason. The Japanese first printing certainly exists as a book; it simply isn't in this census as a graded entry, whether because none was ever submitted, because it's filed somewhere we couldn't surface, or for some other reason. An honest absence is still data. It tells you the graded market for that specific book is, as far as this report shows, empty.
Grading is not survival. This census counts what was graded, not what exists. A book with one slab isn't necessarily a book with one surviving copy; it's a book almost nobody has paid to certify. The census measures the intersection of scarcity and collector attention, which is a different thing from raw rarity. Reading it as a pure survival count would be a mistake, and it's the mistake this data most easily invites.
06Why this is worth having
Disclosure, as always: I deal, and a hobby that takes census data seriously is good for people like me. Weigh it accordingly.
With that said, here's what a closed, fully-read census gives you that vibes never could. It tells you the ceiling on these books is a 9.8 and the ceiling is thin. It tells you One Piece's apparent abundance hides a two-copy peak. It tells you the pristine end of the market belongs to recent books while the vintage keys are defined by their scarcity of high grades, not their abundance of them. And it tells you which respected series have graded populations you could count on one hand.
None of that is opinion. It's just what the report says, read carefully, before it froze. The market has been pricing these books off assumptions for years. Now there's a number to check the assumptions against, and on the most important book of all, the number is two.
This is the first full read of the closed CBCS manga census. The complete data is public and free on the census page, every figure labelled by how it was obtained. If you find something in it we didn't, that's the point. Corrections and additions welcome.