Start with one line from PSA's 2025 Fraud Report and let it sit: 181 submissions of the 1952 Topps Mickey Mantle #311, 112 of them deemed inauthentic. Sixty-nine came back genuine. That is a 61.9% failure rate — closer to two in three than one in two — among cards that someone cared enough about to pay a grading company to examine.
The trade coverage read that as a fraud story. It is a more uncomfortable story than that. Fraud is a crime-and-vigilance narrative with a satisfying moral: be careful out there. What these numbers actually threaten is arithmetic. Raw sales comps, "ungraded example" listings, and population-versus-print-run scarcity math all draw on a pool of raw cards whose composition nobody had measured. PSA has now put a number on part of it, and the number is bad enough to change what those figures mean.
The tally, stated cleanly
PSA's 2025 Fraud Report is a 32-page document and the company's first public fraud disclosure. As reported by Sports Collectors Digest, the card-level tables are the substantive part:
| Card | Submissions | Deemed counterfeit | Rate | | --- | --- | --- | --- | | 1909-11 T206 Honus Wagner | 13 | 12 | 92.3% | | 1952 Topps Mickey Mantle #311 | 181 | 112 | 61.9% | | 1984 Star Michael Jordan #101 | 360 | 96 | 26.7% | | 1986 Fleer Michael Jordan #57 | 3,285 | 821 | 25.0% |
Directionally, counterfeit and altered submissions rose 45.3% year over year. That growth is not evenly distributed: sports cards were up only 5.1%, while Pokémon submissions in this bucket rose 125%. The category split of counterfeits ran 56.3% TCG, 43.1% sports, 0.6% pop culture — and SCD makes the observation that matters more than the split itself, namely that it tracks PSA's global submission mix. The most-counterfeited subjects, in order: Charizard, Michael Jordan, Pikachu, Mickey Mantle, Gengar, Tom Brady, Rayquaza, Umbreon, Mewtwo, Ken Griffey Jr.
The number that led the coverage — more than $200 million — deserves less weight than it got. PSA's own wording is "projected market value of fraudulent collectibles intercepted," which means it prices items that were stopped before reaching the market. It is a modeled counterfactual, not realized loss, and PSA's valuation methodology has not been independently audited. Ryan Hoge, PSA's president of grading, went further on the Sports Cards Nonsense podcast: "It was a conservative number. If you looked at it on the extreme end, that number could be well north of a billion dollars." That extrapolation values intercepted items at PSA 10 levels. Treat it as a claim, not a finding.
One widely repeated figure — that counterfeits rose 250% as a share of total submissions — is reported inconsistently across outlets and we are not repeating it here. Anyone citing it should pull it from the PDF directly.
What a 61.9% rate is, and what it is not
This caveat belongs up front rather than buried, because most of what has been written about the Mantle figure overstates it.
181 is a count of submission events in 2025. PSA's own announcement language, as quoted by MLB Daily Dingers, is that the card was "sent to PSA 181 times." That is not 181 distinct cards, and it is emphatically not a random sample of extant raw '52 Mantles. Resubmissions can recur. More importantly, submitters self-select — and they select in the direction that inflates the rate. A collector who suspects a card is fake frequently submits it precisely to find out. A collector holding a card with clean provenance and no doubts may never submit at all.
So the honest concession: you cannot project 61.9% across all raw 1952 Mantles in existence. Nobody who cites this stat that way is reading it correctly, and PSA has not claimed otherwise.
Here is the narrower claim that survives the concession, and it still does real damage: the raw Mantles being actively presented into the market are majority fabricated. Cards submitted for grading are, almost by definition, cards in motion — being bought, sold, consigned, inherited and converted. That circulating pool is the population every raw comp is drawn from. The selection bias that ruins the universal claim is the same bias that sharpens the market-relevant one.
To be explicit about provenance: no third party has published a critique of PSA's sampling. The selection-bias reasoning above is our analysis, not a sourced finding.
The poisoned denominator
Take the three artifacts collectors actually price off and run a contaminated raw pool through each.
1. Raw sales comps. A completed-sale record does not know whether the card was real. Sold listings for raw examples average genuine and fake transactions indiscriminately, and the resulting "raw market price" settles at a fake-adjusted midpoint — below what an authentic example is worth and above what a counterfeit is worth. If a meaningful share of the raw transactions in a given card are fakes, the comp is not a noisy estimate of authentic value. It is an estimate of a mixture, and the mixture ratio is unknown and card-specific.
2. "Ungraded example" as a listing category. The phrase does real rhetorical work: it implies a card that simply hasn't been through the process yet, with grading framed as a formality that converts value rather than tests existence. For high-counterfeit-rate cards, that implication is unearned. On the evidence in this report, "ungraded" for a '52 Mantle is not a state pending confirmation — it is the state in which most fakes are found.
3. Population-versus-print-run scarcity math. This is where the damage compounds. The standard argument compares graded population against estimated print run and treats the gap as latent supply — cards still out there, ungraded, that could enter the market. The graded population is the only clean number in that equation. It has been examined. The raw remainder has not, and PSA's tables are the first public evidence about what it contains. Secondary sources place the PSA-graded '52 Mantle #311 population in the neighborhood of 2,075 and the 1986 Fleer #57 population around 30,000; we were unable to confirm either against PSA's own population pages, which block automated retrieval, and we flag them as unverified rather than drop them silently. The structural point does not depend on the exact figures: the numerator is audited and the denominator never was.
Why Mantle is two-thirds and Jordan is a quarter
The tempting read is that Mantle counterfeits are better executed. The likelier explanation is structural, and it has nothing to do with counterfeiter skill.
The governing variable is genuine raw supply. The 1952 Topps high-number series — which #311 opens — was under-distributed and famously dumped rather than sold through. Authentic raw Mantles essentially do not surface; a "freshly discovered" one with no documented chain of custody is a priori suspect. The 1986 Fleer set, by contrast, was mass-produced. Genuine raw copies are everywhere, and they dilute the counterfeit share of anything submitted.
Layer the payoff ratio on top. Producing a convincing fake costs roughly the same for either card; the upside differs by orders of magnitude. Recent Mantle results include a $12.6M SGC 9.5 in August 2022, a $1,067,500 signed PSA 5.5 in August 2025, and a $945,500 PSA 1.5 in December 2025. Fakes concentrate where genuine supply is thinnest and the reward is fattest — which means a high counterfeit rate is itself a scarcity signal, and one of the few not generated by the people selling the scarce thing.
The counterfeiter economics Hoge describes point the same way. "We might find cards that are $100 to $300 cards," he told SCD's reporting. "If somebody can make a counterfeit version for $1 and get it into the market, they'll do that all day long." The Brady data supports it: his most-counterfeited cards are not his premium rookies but mid-tier issues — the 2000 SkyBox Impact #27 ($80–$120 raw, $1,377 in PSA 10) and the 2000 Pacific Crown Royale Rookie Royalty #2 ($70–$95 raw, $1,726 in PSA 10). The arbitrage being worked is the raw-to-slab gap, not absolute card value.
Where the fakes enter — and how much we actually know
Less than the hobby talks like it does. PSA's own framing is well sourced and worth quoting: counterfeiters pursue "modern cards that balance a recognizable name, strong resale value and simpler printing technology." That is a coherent target profile and it matches the Brady figures.
The specific channel claims that circulate alongside this report — estate lots, overseas print shops, particular marketplace vectors — were not substantiated by any primary source in our review. Open-market "reprint" listings, sold as reprints and later resold as originals, are the one channel with direct documentary support. Everything else in that list is hobby consensus, and we are labeling it as such rather than dressing it as fact.
There is one hard enforcement anchor: Anthony Curcio was convicted in the Southern District of New York in January 2026 on wire fraud conspiracy charges for forging PSA labels and serial numbers onto lower-grade cards. Note what that case is — label forgery, an attack on the slab rather than the card. It is a reminder that grading concentrates trust into an object that is itself worth counterfeiting.
The party publishing the data
This is not a disclaimer line. PSA is the sole source of these figures, the sole beneficiary of the trust narrative they generate, and the entity that converts raw-market anxiety into paid submissions. That is three roles in one, and none of the trade coverage reproducing the numbers has tested them adversarially. No independent audit or replication of the $200M methodology exists.
The financial exposure is concrete. Collectors Holdings owns PSA alongside a leading pricing platform, a vault and resale service, an instant-cash-offer program, and a lending arm collateralized by graded cards. Every one of those businesses appreciates as grading penetration rises and as raw cards convert to graded ones — which is exactly the behavior this report encourages.
The company is also, per Value Added Resource, defending three federal suits filed in 2026: Funk v. Collectors Universe (D. Md., 1:26-cv-02933, filed July 28, 2026, alleging RICO and consumer protection violations); Lichtman v. Collectors Universe (C.D. Cal., 8:26-cv-02048, filed July 21, 2026, damages claimed over $5M, alleging undergrading and "population control" to protect insider-owned cards); and Rasmussen v. Collectors Holdings (C.D. Cal., 8:26-cv-00897, filed April 2026, an antitrust challenge to the SGC and Beckett acquisitions). Collectors calls the Lichtman claims "without merit" and states that PSA would "never compromise our grading independence." These are allegations, untested, and should be read as such.
PSA's stated controls are substantial and worth acknowledging: more than 80,000 hours of specialized training across the grading organization in 2025, graders completing PSA Grader University plus 13,000+ supervised evaluations before working independently, AI imaging models trained on four-plus years of pre- and post-grading images, dimensional analysis for trimming, and software alerts for known counterfeits. One piece of that framing has weakened, though — the brand-protection team's "cross-industry" intelligence sharing across PSA, Beckett and SGC now describes three graders under a single owner.
And then there is cert #00000001. PSA's first-ever certificate was issued to a T206 Honus Wagner that Bill Mastro later admitted, in a federal plea agreement, to having trimmed. It was graded NM-MT 8. The certificate has never been withdrawn. The same company now publishes a 92.3% Wagner counterfeit rate.
What a submitter can actually do
Nothing here argues for generic caution. Three things follow directly from the data:
Invert the default on thin-supply cards. For a raw '52 Mantle offered without a documented provenance chain, fake is the base rate, not the edge case. Treat it as such and require the seller to move you off it. The same logic scales down to any issue where genuine raw supply is structurally near zero.
Understand that grading is the authentication test. There is no cheaper substitute for the dimensional analysis and imaging comparison PSA runs. But it is a paid test, and its price is rising: PSA's backlog reportedly ran about 10 million cards in May 2026 and 12.4 million by late July, and in June 2026 the company suspended Value-tier submissions and raised its minimum submission price to $74.99. That cost floor pushes more transactions into the raw market — the precise channel this report identifies as compromised. The economics of the remedy make the disease worse at the low end.
Know the raw-channel option and who runs it. eBay's Authenticity Guarantee now covers raw as well as graded single cards at $200 and up, a threshold lowered from $250. PSA performs that multi-point inspection — cards are checked for authenticity but not graded. Genuinely useful, and one more point in the transaction path occupied by the same firm publishing the fraud data.
The reframe
The 2025 Fraud Report's real contribution is not the $200 million. That figure is modeled, unaudited, and structurally unfalsifiable. The contribution is four denominators — 13, 181, 360, 3,285 — and the failure rates attached to them.
Comps, listings and scarcity arguments have been computed for years off a raw pool nobody had measured. The first party to measure any part of it is the one with the most to gain from what the measurement says. That doesn't make the tables wrong. It does mean the right response isn't vigilance. It's discounting raw-market numbers to reflect a denominator we now know is contaminated by an unknown amount — and asking for the same tables from someone who doesn't also sell the cure.
Related reading
Sources
-
PSA — 2025 Fraud Report (official landing page)
-
PSA 2025 Fraud Report — full PDF (32 pages)
-
Sports Collectors Digest — Fraud, counterfeit trading cards rampant in hobby, PSA report shows
-
Sports Illustrated — $200M in Fake Cards: PSA's New Report Reveals a Growing Threat
-
MLB Daily Dingers — 62% of 1952 Mickey Mantle Cards Submitted to PSA in 2025 Were Deemed Fake
-
Value Added Resource — PSA, Collectors Face Growing Legal Challenges Over Card Grading Power
Note: This article contains AI-assisted content and has been reviewed in our editorial workflow.
DISCLAIMER: PureGrail articles are for informational and entertainment purposes only. Nothing on this site constitutes financial, investment, or legal advice. Collectibles are speculative assets and values can decrease significantly. Always conduct your own research before buying or selling. Past price performance does not indicate future results.
