Disclosure Raw and ungraded, and ours alone. No grading company, owner or marketplace paid for it, previewed it or touched a word of it, and Cardboard Journal has no financial ties to any of them. Nobody is paying us to think this. We just do.
Ask a collector why a PSA 10 sells for more than the same card in a $15 slab, and the honest answer is that the market says so. Not because anyone has shown that PSA grades more accurately. Nobody has. There is no independent, widely accepted benchmark, no referee and no published comparative data that would let anyone prove PSA is more accurate than the smallest grader on our scorecards, or prove it isn't.
That is a remarkable thing to be true of a business whose entire product is certainty. And the companies with the money and the data to change it are the biggest ones in grading. They haven't.
Every grader marks its own homework
PSA writes the definition of a PSA 10. PSA applies it. PSA decides whether it applied it correctly. When a collector disagrees, the appeal goes to PSA. BGS, SGC and CGC work the same way, each with a scale of its own. A PSA 9 and a BGS 9 are two different claims about two different sets of rules, and there is no common reference to test either one against.
PSA's own pages don't see a problem. Its grading standards page says PSA's standards "have truly become the official standard for the most valuable cards in the hobby." Its service page says PSA has established a "universal grading standard for trading card collectors." Official according to whom? Universal by whose agreement? The title is self-awarded.
Read further and the standard admits it bends. PSA says the vast majority of grading is objective: measuring centering, finding print defects, stains and wrinkles. The rest, it says, is "somewhat subjective," and it defines that part with a question: "What will the market accept for this particular issue?" When a card sits on the line between two grades, usually over centering or eye appeal, its graders "reserve the right" to make a judgment call, and a small centering variance can be allowed "at the grader's sole discretion." PSA says this is rare. Judgment on its own isn't the problem; every grader uses it. The problem is what anchors it. On the borderline cards, the market trusts the grade, and the grade defers to the market. That is not a standard. It is a loop. And nobody outside PSA measures how consistently the call is made.
Two owners, no referee
This would matter less if the market were spread across many independent companies. It isn't. Collectors owns PSA, SGC and Beckett. In a December 2025 letter asking the Federal Trade Commission to investigate, Rep. Pat Ryan warned that the combination would control more than 80 percent of grading volume. That is his estimate, not a finding, and we have found no public response from the FTC. Blackstone's funds hold a majority of the group that owns CGC. Two investment-backed owners now control the biggest names in grading, and neither publishes independent, comparable evidence of how accurately its brands grade.
Ryan's concern was market power. The quieter problem is evidence. Concentration doesn't prove anyone grades badly. But it removes independent rivals with a commercial reason to test one another in public, in a hobby that has no common reference anyone could test against.
Bigger, faster, and harder to see
PSA has spent the last five years getting enormous. In April 2021 it bought Genamint, a machine-vision company whose software measures cards and helps spot alterations. Nat Turner, who leads Collectors, said the deal would let PSA "grade more trading cards faster while improving accuracy," and that "we're not eliminating humans from the grading process." In May 2026, Sports Business Journal reported that Collectors was putting $200 million into expanding PSA's operations. Hobby-press coverage said the plan covered new facilities, technology and about 1,000 jobs.
None of that shows grading got better or worse. Growth and machines show scale, not quality. But they raise questions PSA hasn't answered in public. How long does it take to train a grader, and how are graders calibrated against one another? Unpublished. What does the software contribute to a grade, and has PSA measured its effect on consistency? Unpublished. Is a card graded in 2026 judged the way it would have been in 2021? No public evidence says either way.
CGC faces the same questions. When Blackstone bought into its parent in 2021, the announcement said the money would help it add and train new employees and acquire new technology. We found no independent evidence of CGC's consistency published either.
Nobody has shown the big graders grade better
Here is the uncomfortable consequence. If no one measures accuracy against a common reference, then the public evidence can't say the giants grade better than the small companies on our scorecards. A $15 grader might be sloppier than PSA. It might be stricter. Gem rates won't settle it. They depend on which cards people choose to send to which grader and on what those cards are, so a low gem rate can mean a strict grader or a pile of weaker cards.
What the market pays for a PSA slab is real, and this site's own score is built on it. The premium is evidence of market confidence and resale demand. By itself, it isn't evidence of a more accurate grade. Our score measures commercial things: price, turnaround, ownership and whether the market honors a slab. We say on our method page that it does not measure accuracy, because there is no independent benchmark to measure it against. PSA does admit that part of grading is subjective. What none of the big graders publishes is evidence of how consistently that judgment is applied. They should.
They already share a scale, just not for cards
The most telling fact about card grading is what the same owners do elsewhere.
Collectors owns PCGS. Blackstone's group owns NGC. They are the two dominant coin graders, and both grade on the 70-point scale that grew out of one William Sheldon devised in 1949, the scale the American Numismatic Association adopted for its published grading standards in 1977. A shared scale isn't a shared verdict: the two grade independently, and they can disagree about the same coin. But a coin grade is written in a vocabulary neither company owns alone. Blackstone's PMG grades paper money on a 70-point scale it says is derived from Sheldon's.
Comics and diamonds show other routes. CGC says its comic-book standard was derived from the Overstreet grading criteria and first built from a consensus of grades given by a nationwide pool of dealers and collectors. GIA, which created the 4Cs, grades diamonds by its own system, but as a nonprofit that doesn't buy, sell or appraise them.
None of those systems removes judgment, and none is perfect. What they show is that other markets have found ways to give grades a shared vocabulary, published criteria, or a grader with no stake in the sale. Card grading has no widely adopted cross-company benchmark at all. The same owners grade coins on a shared scale and cards on scales of their own. It is worth asking them why.
What happens when nobody checks
Collectors has owned Wata, the video game grader, since July 2021. In 2022, buyers filed a class action alleging that Wata helped inflate the retro game market, failed to disclose ties to a co-founder of Heritage Auctions, and misstated its turnaround times. Wata and Heritage denied the allegations. The case moved to federal court in Colorado, and in March 2026 it was dismissed without prejudice after the plaintiffs failed to respond to a court order. No court ruled on whether the allegations were true.
We are not saying they were, or that the case says anything about how cards are graded. It is a small illustration of why outside scrutiny matters. When a grader is the only authority on its own grades, claims like these are hard for anyone to test either way, and this one ended without being tested.
Who could grade the graders
In 1893, Chicago's fire insurers sent a young inspector named William Henry Merrill to look over the electrical wiring at the World's Fair. Electricity was new, faulty wiring was starting fires, and the insurers were the ones paying for them. So they stopped taking anyone's word that a product was safe. The next year, with a little money from two insurance groups and $350 of equipment, Merrill opened a lab above a fire station and started testing products himself. It became Underwriters Laboratories. The UL mark on the plug next to you is there because the people with money at risk paid for a check the manufacturers didn't control.
Card grading is still waiting for its insurers. The tools aren't a mystery. Coins have a second-opinion sticker. Auditors got an auditor of their own after Enron. Credit-rating agencies register with the SEC. Every version comes down to two things cards don't have: a written standard no single grader owns, and an outside check that sends the same cards through every grader, blind, and publishes what comes back.
That takes someone with money and clout, and a reason to want grades right rather than high. Marketplaces and auction houses earn more when grades come back higher, and a body the graders fund inherits their conflict. Insurers, lenders and the biggest buyers are the ones who lose when a grade is wrong. Somebody in that group has to step up, the way the insurers did in Chicago.
If PSA is the standard, prove it
PSA says its grading is "the official standard." Then it should be the first to prove it. A company that is genuinely the best at something tends to say so with numbers. PSA has the most cards, the most data and the most money in grading, and if it is as good as its prices suggest, it has the most to gain from an audit. Instead, it banks on being the default.
Five questions would settle it, and PSA could start on every one of them tomorrow.
| The question | What would answer it |
|---|---|
| Would the same card get the same grade twice? | Results from blind resubmissions, with sample sizes |
| Do your own graders agree with each other? | How often two graders give the same card the same grade |
| Do you agree with anyone else? | Blind tests that send the same cards to other graders |
| Agree against what? | The reference set grades are checked against, and who checks it |
| What did the machines change? | Consistency before and after Genamint |
We could find no published answer to any of them, from PSA or anyone else. The same questions stand for SGC, Beckett, CGC and every grader that wants to be trusted. But the leader goes first. Being the biggest is not the same as being the standard, and the only way to show the difference is to let someone check.
Until someone checks
So here's the awkward part, the one we can't argue our way out of: send your best card to PSA anyway.
You'd be a fool not to. PSA owns the market. Its label is the one buyers search for, and its slab usually sells for the most. Picking a stricter or cheaper grader on principle is a fine way to feel right and get paid less, and this site's own scores say as much.
Just know what you're buying. Not a better grade, or at least not one anyone has shown. You're buying a reputation PSA hasn't proved it deserves, because it has never had to. It is the most valuable opinion in the hobby, and nobody outside the company checks it. A PSA 10 is worth the most because everybody agrees it's worth the most. That's a perfectly good reason to buy one, and a terrible reason to call it a standard.
A reputation is only as good as the last time somebody looked. Somebody should.
