Pokemon Cards as an Asset Class: What the Real Numbers Actually Say

I built the full Pokemon Cards module on this site to answer one question: is this a genuine, structurally growing market, or a very good story built around one insanely valuable card. That page is the reference version, deliberately without a personal verdict. This is the shorter version, with one.

The bull case, in one number

The Wall Street Journal, citing price tracker Card Ladder, reported its Pokemon collectible index up 3,821% cumulatively from 2004 to August 2025, against 483% for the S&P 500 over the same span. That is the number every bullish take on this topic eventually cites, and it is real. It is also doing a lot of work for one number: it is a single index, over a specific 21-year window, built from one tracker's own methodology. Change any of those three things and you get a different story, which is most of what this piece is actually about.

The production data really does back up "still growing"

This part is not in dispute. Cumulative Pokemon TCG cards printed worldwide crossed 85 billion in May 2026, and roughly 40% of every card ever printed came out in just the last three fiscal years.

Grading volume backs it up from the demand side: PSA is now grading around 90,000 cards a day, up from roughly 15,000/day in 2021, with a reported backlog of 10 million-plus cards. Trading-card submissions (Pokemon, Magic, Yu-Gi-Oh) are up 95% year-over-year and now outpace sports-card submissions more than 2-to-1 at PSA. That is not the shape of a fad cooling off. It is a demand curve still bending upward three decades in.

The card that is supposed to prove this is stable. It does not, quite.

The single most-tracked individual card in the hobby is the PSA 10, 1st Edition Base Set Charizard, and its real sale history is the honest test of "is this actually a boring, stable store of value."

That is not a boring asset. It is a card that went four years with no recorded sale at all, then round-tripped through a 40% drawdown, inside a longer uptrend that did eventually make a new high. Anyone selling this as "stable" is quietly skipping the middle of that table.

$30 in 1999

The comparison everyone eventually wants: what if you had put a small amount into a Charizard in 1999 instead of the index. $30 in the S&P 500 in January 1999, dividends reinvested, using Yahoo Finance's real adjusted-close data, is worth about $261 today, roughly 9x. The same $30 toward a Charizard that later graded a PSA 10 would be worth $550,000 today, over 18,000x.

That gap is real, and it is also the least representative comparison possible. It is the single most valuable specific outcome in the entire hobby, not a typical one, and the honest caveat matters here: a specific, dated PSA 10 sale record for this card does not exist anywhere from 1999 through mid-2017. $30 is a stated assumption about what a raw copy or a few packs cost at release, not a documented sale. Every point on this comparison from July 2017 onward is a real, publicly reported transaction. The 1999 entry point is not.

Nobody researching this professionally actually agrees

CNBC ran the WSJ number too, after AJ Scaramucci paid $16.49m for the only PSA 10 "Pikachu Illustrator" card and argued on air that collectibles are now a legitimate asset class. That is one buyer's argument after a record purchase, not a bank or an academic publishing research, and it should not be read as the latter.

The pushback is real and comes from people with no stake in the outcome. 24/7 Wall St. ran a piece arguing the "beats the S&P 500" framing is a category error, since a card produces no cash flow or dividend and comparing it to an equity index the way you would compare two stocks is methodologically off. A financial adviser quoted in that piece called it a "math crime." Two Northeastern University finance professors landed on opposite sides on record: one cautiously open (value drawn from scarcity and cultural appeal, similar to art), one openly skeptical, saying there is "no proven multi-decade outperformance track record."

The most rigorous work found on this, from the CAIA Association (a 2021 study by Andrew Keenan, CAIA, CFA, Credit Suisse), found buyers pay an average 1,487% premium for a PSA 10 over a PSA 9 of the identical sports card, and that PSA-10 portfolios returned about 42% annualized over the study window versus 35% for PSA-9. It is sports cards, not Pokemon, but it is the clearest quantitative proof that grading is most of what actually drives this market, which the Charizard case study above shows anecdotally.

For Pokemon specifically, peer-reviewed research basically does not exist yet. The one direct attempt is an undergraduate independent study at the College of Wooster (2024), using PriceCharting data from 2021-2023, and it found a portfolio of Pokemon cards returned -4.72% annualized over that window, underperforming the S&P 500. It is a real, named study, just not a peer-reviewed one, and a completely different three-year window than the WSJ's 21-year figure. Both numbers can be true at once. This is a young, volatile market where the answer depends heavily on which years you happen to measure, which is itself the finding.

Who actually owns this

Pokemon is owned in roughly equal thirds by Nintendo, Game Freak, and Creatures Inc, run day to day through their joint venture, The Pokemon Company. None of the three is separately investable. Only Nintendo (TYO: 7974 / OTC: NTDOY) is public, and Nintendo does not break out a Pokemon-specific revenue line, so even that exposure is real but blended, not a clean pure-play read.

The Pokemon Company's own most recent fiscal year (to February 2026) was a record: revenue of ¥531.4bn (~$3.34bn), up 29.3%; operating profit ¥144.0bn (~$904m), up 43.0%; net profit ¥120.1bn (~$754m), up 70.7%. Profit growing faster than revenue is exactly what a licensing business with low incremental cost is supposed to show.

The grading infrastructure is private too, and consolidating fast. PSA's parent was taken private in 2021 for roughly $853m by an investor group including Steve Cohen, and by 2022 had raised further capital at a $4.3bn valuation. Its main challenger, CGC, was majority-acquired by Blackstone's Tactical Opportunities arm. There is no clean public equity for the grading side of this trade either.

Retail investors have already tried to securitize this directly, and it is mostly gone badly. Otis fractionalized a Charizard at a $236,800 valuation, was acquired by Public.com in 2022, then liquidated, with investors reportedly taking a loss. Collectable, an SEC Reg A+ platform, discloses roughly 3.1% annualized returns since launch against an estimated 6-8% for the broader blue-chip card category over the same period, meaning the actual investable product underperformed the market it was built to track, after fees. Dibbs pivoted away from fractional ownership entirely in 2024. That is a real, market-tested answer to "can this be cleanly securitized," and so far the answer has mostly been no.

The comparison that should worry any bull

At the peak of the early-1990s baseball card craze, manufacturers printed an estimated 81 billion cards a year, about 325 for every person in America, chasing demand with no regard for scarcity. When the 1994 MLB strike broke collector confidence, the market never recovered. Revenue fell to roughly a seventh of its peak, and most of that era's cards are worth about a cent each today.

Being honest about the parallel: Pokemon's own print runs are accelerating too, the exact same "print more to chase demand" instinct that broke the sports-card market. What is different so far is that demand has kept pace instead of declining. A real structural stress test, a recession or a sharp drop in new-collector demand, is something this specific boom has not faced yet.

It is already splitting in two, not hypothetically

Everything above treats "is this a bubble" as a future question. Checking the most recent data changed that for me, because for one part of this market it is not a future question anymore. It is already happening, just not evenly.

Modern cards, released in the last two to three years, have corrected hard since late 2025. An Obsidian Flames Charizard fell from $126 to $79. A Prismatic Evolutions Umbreon SIR fell from $1,600 to $832, a 48% drop, in a matter of weeks. Elite Trainer Boxes are down about 25% (Phantasmal Flames booster boxes went from $305 to $275), and sealed product broadly is down 15-25% since March 2025. Independent trackers put modern singles down somewhere between 20% and 60% from their 2024-2025 peaks. Actual sales volume for booster and sealed product, not just asking prices, fell from an index reading of 410.5 in January 2026 to 270.77 by March, a real drop in how much of this stuff is actually changing hands.

Vintage did the opposite over those same months. The Charizard case study above hit its own all-time high in December 2025. The Pikachu Illustrator sale mentioned earlier happened on 16 February 2026, a new world record for any Pokemon card, while modern cards were still falling. That is not a coincidence of timing, and I do not think it is really two separate stories either. It is the same mechanism producing opposite results. A modern chase card's supply can grow to meet demand, because The Pokemon Company can print more of that exact set, and that is exactly what has been happening. A 1st Edition Base Set card printed in 1999 cannot be reprinted. Its supply is fixed in a way no card from a 2024 set's supply is, no matter how hot that set gets.

My view

I do not think this is a fad, and I do not think it is a clean financial asset either. But by 2026 I also do not think "is Pokemon overvalued" has one answer, because it is not one market, and the data above makes that split too real to keep treating it as one question.

The modern segment was overvalued, and it is already proving that in real time rather than waiting for some future correction. I would expect that gap to widen, not close, through the rest of 2026 and into 2027. The Pokemon Company is still printing at record volume right now (85bn+ lifetime, roughly 40% of that in the last three fiscal years alone), so every new set adds fresh supply competing against everyone who bought a 2024-2025 chase card near the top and wants out. Vintage has the opposite setup, and I think it actually gets more scarce over time in a way most people do not account for: the surviving raw population of a 1999 card only shrinks as more of it gets handled, damaged, or lost before ever reaching a grader, so the PSA 10 population of a fixed vintage print run can really only go down, never up. That is a genuinely different supply curve from a modern card the company can reprint on demand, and it is the actual reason I do not expect vintage to follow modern down.

The production and grading data make a real case that this whole category is structurally different from a normal collectible wave: three decades of accelerating, not fading, demand, backed by genuine third-party authentication infrastructure rather than self-reported numbers. That part of the bull case is real, for vintage specifically. But the actual, computed academic returns on Pokemon cards range from -4.72% annualized (Wooster, 2021-2023) to numbers that beat the S&P 500 many times over (WSJ/Card Ladder, 2004-2025), and now I think I know why that gap is so wide: one of those studies was mostly measuring modern cards during a window that looks a lot like today's correction, and the other was dominated by the small number of ultra-rare, perfectly graded vintage cards that are still setting records. Almost everyone's actual holding is a modern card, not one of those.

If I were actually putting money into this, the CAIA finding is the one I would take most seriously: grading, not the card itself, is most of what drives return. A PSA 10 versus a PSA 9 of the identical card is a roughly 1,487% price gap on sports cards, and the vintage-versus-modern split above is that same idea at the level of an entire market segment, not just one card. The real skill this "investment" requires is not picking the right character or the right set. It is correctly telling, before you buy, whether you are holding something with genuinely fixed supply or something the company can print more of the moment demand asks for it, which is a distinction most people writing enthusiastically about this asset class right now are not making.

Sources

• The Hot Investment With a 3,000% Return — WSJ, via Securities Docket
• Scaramucci: collectibles are an asset class — CNBC
• Pokemon cards beat the S&P 500, but the math is a lie — 24/7 Wall St.
• Pokemon cards: should you invest? — Northeastern University
• The Price of Perfection — CAIA Association
• Trading card portfolio returns, 2021-2023 — College of Wooster independent study
• November 2025 grading volume recap — GemRate
• Card Ladder methodology
• Collectable returns, via MoneyMade
• The Pokemon Company FY26 financial results — Serkan Toto
• PSA/Collectors Universe take-private
• Collectors Holdings $4.3bn valuation — Sportico
• Blackstone acquires CGC's parent — press release
• Pokemon Card Market Crash 2026: What Really Happened — PokemonPriceTracker
• Pokemon Card Market Crash 2026: Signs & How to Prepare — PokemonPriceTracker
• Why Are Pokemon Card Prices Dropping? — Misprint
• Pokemon Card Market Trends 2026 — CardScan AI
• Logan Paul's Pikachu Illustrator sells for record $16.5M at auction — KSAT