Secondary Market Pricing Transparency in Collectible Trading Card Games

Abstract

The collectible trading card game (TCG) secondary market — part of a broader trading card market estimated at roughly $50 billion — has consolidated rapidly since 2022. One corporate group (Collectors Holdings) now owns the three largest grading services and a leading graded-card analytics tool; another (eBay Inc.) owns the two largest online marketplaces for cards and a major auction house. Public pricing APIs have narrowed over the same period, and the pricing tools that remain face real accuracy, variant-identification, and usability problems that fall hardest on individual buyers and sellers.

This research builds and studies an independent, cross-platform pricing system that operates without privileged data access. As of October 2026 it catalogs roughly 596,000 cards across 67 game and language catalogs, tracks about 3.2 million priced condition/variant SKUs, and holds more than 44 million historical sale records from TCGplayer, eBay, and PriceCharting, plus Cardmarket listings and sales for European-language Pokémon cards. The resulting dataset makes it possible to measure, for the first time at this scale, how the same card is priced across platforms, conditions, languages, and regions — and where single-source “market prices” mislead.

Research Questions

1. Price dispersion: How much do identical cards vary in price across marketplaces, and how much of that variance do platform fees, seller type, and listing format explain?
2. Cross-region pricing: Do persistent differentials exist between U.S. (TCGplayer, eBay) and European (Cardmarket) prices for the same card, after currency, fees, and shipping?
3. Consolidation effects: Has vertical integration of grading, marketplace, and analytics services measurably changed price discovery since 2022?
4. Information asymmetry: To what extent does restricted access to transaction history advantage institutional sellers over individual buyers and sellers?
5. Condition and variant sensitivity: How do inconsistent condition terminology and variant identification across platforms affect pricing accuracy?

Market Context

Despite its size, the trading card market has no standardized valuation guide, no centralized listing service, and no consolidated transaction record — unlike secondary markets of comparable scale such as used automobiles or real estate.

Marketplace Consolidation

eBay acquired TCGplayer in 2022 ($295M) and Goldin Auctions in 2024. TCGplayer's public API was restricted after the acquisition, removing a data source that independent tools had relied on. In May 2026 GameStop made an unsolicited offer of roughly $56 billion for eBay, which eBay's board rejected as “neither credible nor attractive”; as of this update GameStop has continued to build its stake and the bid remains unresolved.

Grading Consolidation

Collectors Holdings, owner of PSA since 2021, acquired SGC in 2024 and announced its acquisition of Beckett in December 2025, placing the three largest grading services — and the analytics tool Card Ladder — under one owner. In April 2026 a proposed class action, Rasmussen v. Collectors Holdings (C.D. Cal., No. 8:26-cv-00897), alleged that these acquisitions gave the group roughly 80% of the U.S. card-grading market in violation of the Sherman and Clayton Acts. That share is the plaintiff's allegation, not a court finding; the defendants have moved to dismiss or compel arbitration. U.S. Representative Pat Ryan has separately asked the FTC to investigate; no formal FTC investigation has been publicly announced.

Independent alternatives remain: CGC, which expanded from comics into trading cards in 2021, and TAG, a smaller grader using photometric imaging. TCG and non-sport cards accounted for 16.8 million of the 26.6 million cards graded by major authenticators in 2025, surpassing sports cards for the first time.

The Social Commerce Gap

A meaningful share of TCG transactions happens through Facebook Marketplace, Facebook Groups, Discord, and local trading communities. Based on direct observation at trade events and in online communities, we estimate this share at 15–25%; that figure is qualitative and has not been validated. These transactions are absent from every existing pricing dataset, including ours.

Why Existing Tools Fall Short

The Pricing Complexity Problem

Consider the Charizard from Pokémon's Legendary Collection. It exists as a Holofoil, a Reverse Holofoil, and a Deck Exclusive Non-Holo, each with a substantially different value. Each variant has five raw condition grades and may also exist as a PSA, CGC, BGS, or TAG slab at grades 1–10. One card can occupy dozens of distinct price points, and many tools conflate them. Counterfeits add noise: sales of fakes are recorded alongside legitimate transactions and distort averages for tools that ingest completed sales unfiltered.

Observed Problems

• Unfiltered sales: In our testing of popular aggregators, lot sales were sometimes counted as single-card sales and graded and raw sales were sometimes conflated. For thinly traded cards, estimates diverged 20–40% from recent comparable sales.
• Variant mapping: We observed apparent misidentification (e.g., a Reverse Holofoil price attributed to the standard Holofoil) and stale prices persisting for weeks after the market moved.
• Speed: Pricing one card across eBay and TCGplayer by hand takes 3–5 minutes; a modest 500-card collection is 25–40 hours of work. At a trade show, a tool that cannot surface the right variant in seconds is not usable.
• Graded-only coverage: Graded-card guides and analytics (including Card Ladder, owned by Collectors Holdings) cover only the slabbed minority of the market and use catalog numbering that maps to no other platform.
• Undisclosed methodology: Most tools do not say how “market price” is computed — rolling average, lowest listing, or a proprietary weighting — so users cannot tell why two tools disagree.

Conflicts of Interest

Several services charge $3–50/month for data derived from publicly visible transactions, and many pricing tools also run marketplaces that earn 8–15% per sale, an incentive to steer users toward their own listings. Meanwhile the platforms that generate transaction data increasingly gate it: eBay's Marketplace Insights requires an established developer account with demonstrated volume, and TCGplayer's API now requires a partnership application. These processes structurally favor large partners over independent researchers and small vendors.

System & Data

Data Sources

Catalog Coverage

The secondary market does not respect platform or language boundaries: a Japanese-exclusive promo appears on U.S. marketplaces, and a German printing of an English card trades on Cardmarket at its own price. The system therefore maintains a product identity graph across 67 catalogs:

• Major English TCGs: Pokémon, Magic: The Gathering, Yu-Gi-Oh!, One Piece, Disney Lorcana, Digimon, Dragon Ball, Flesh and Blood, Weiss Schwarz, Star Wars, and more than 40 others.
• Pokémon in 12+ languages: Japanese, Korean, French, German, Italian, Spanish, Portuguese, Indonesian, Thai, Russian, and Traditional and Simplified Chinese.
• Vintage and specialty Pokémon: Topps, Bandai Carddass, Topsun, and Old Maid.
• Graded prices: calculated PSA, BGS, CGC, and TAG market prices for ~26K variants.

Pipeline

Data flows through a tiered database pipeline — landing, staging, test, production, and a serving tier — with every change promoted only after it passes the tier below. Cross-platform matching resolves each marketplace listing to one catalog variant using printed collector numbers, set identity, and visual verification.

Results are published through card-prices.com, which shows U.S. and European prices side by side, per condition and variant, for free and without registration.

Analysis Framework

• Cross-platform and cross-region price dispersion (coefficient of variation per variant and condition)
• Regression of price determinants: condition, rarity, platform, language, seller type, listing age
• Time-series analysis of prices around consolidation events
• Outlier and counterfeit-signal filtering of completed sales
• Market efficiency metrics: ask-to-sale spreads and price convergence rates

Platform Coverage

Pricing methods differ structurally by region: the U.S. relies on transaction-based marketplace prices (TCGplayer Market Price), Europe on trend-based aggregation (Cardmarket Trend Price), and Japan on posted buy/sell lists at major retailers. Comparing them requires collecting from each region directly.

Preliminary Findings

These observations come from building and operating the system; they have not yet been subjected to formal statistical testing and will be quantified in the paper.

• Price dispersion is the norm: Identical variants show substantial price differences across conditions and platforms; a single “market price” is insufficient for informed transactions.
• Non-linear condition sensitivity: The Near Mint to Lightly Played discount varies by value tier — modest for common cards, much steeper for high-value cards — which flat-percentage condition models miss.
• Variant identity drives error: A large share of pricing mistakes trace to misidentified variants (holo vs. reverse holo, printing, language) rather than to the price data itself. Matching on printed collector numbers resolves many of these.
• Raw sales need filtering: Completed-sale feeds contain lots, mislabeled grades, and implausible prices; unfiltered averages are unreliable for thinly traded cards.
• Single-source reliance: Vendors at in-person events typically consult one platform's price, missing cross-platform gaps that matter above roughly $10.

Barriers & Limitations

Barriers to Independent Research

1. API closure after consolidation. Public pricing APIs are restricted or removed after acquisitions, turning structured integration into unstructured extraction.
2. The access catch-22. API programs require applicants to show existing scale that independent researchers cannot reach without the data the API provides.
3. Automated-access defenses. Bot protection built to stop credential stuffing and inventory manipulation does not distinguish low-volume research collection, and often blocks cloud-hosted infrastructure outright.
4. Ephemeral data. Listings disappear when sold and no public transaction archive exists; any gap in collection is permanent.
5. Cost and expertise. Aggregating and normalizing cross-platform data is beyond the means of most individual buyers and sellers, producing a two-tier market.

Limitations

• Single researcher: Qualitative observations may carry observation bias; quantitative results will be reproducible from the dataset.
• Uneven coverage: Pokémon dominates sale volume; European coverage is Pokémon-only; Asian marketplaces are at the catalog stage.
• No social commerce data: The 15–25% estimate is unvalidated, and Facebook Marketplace data is not yet collected.
• Matching error: Cross-platform identity resolution is imperfect; unmatched or ambiguous listings are excluded rather than guessed.
• No external validation: Outputs have not yet been validated against an independent ground-truth dataset.

Data Access & Ethics

This research collects only publicly visible marketplace data — listing prices, completed-sale prices, product metadata, and the store or seller names platforms themselves display — without logging in. No private messages, account data, or non-public transaction details are collected.

• Collection accesses only publicly visible pages without authentication, consistent with federal case law (hiQ v. LinkedIn; Meta v. Bright Data)
• Official API access is preferred and pursued wherever it exists; collection of public pages is the fallback
• Collection is rate-limited to minimize load on platforms
• Research analyses and publications report aggregates and do not identify individual sellers
• No human-subjects data is involved; IRB review will be sought if a future phase requires it
• The lead researcher has completed the CITI Program Responsible Conduct of Research certification

European Data Protection

Collection from European marketplaces is limited to publicly visible listings and sales. Because public seller names on those platforms can identify individuals, this research is handled under the General Data Protection Regulation: seller names are used only to de-duplicate and validate observations, are never published in research outputs, and data subject requests under GDPR Articles 15–22 will be honored.

Legal Basis

In hiQ Labs, Inc. v. LinkedIn Corp. (9th Cir. 2022), the Ninth Circuit held that accessing publicly available data does not violate the Computer Fraud and Abuse Act. In Meta Platforms, Inc. v. Bright Data Ltd. (N.D. Cal. 2024), the court held that Meta's Terms of Service did not bar collection of publicly visible data by a party not logged in.

Research Timeline

Expected Contributions

• Empirical evidence on cross-platform and U.S.–Europe price dispersion in TCG secondary markets — to our knowledge, the first peer-reviewed study at this scale
• An open pricing tool giving consumers free, methodology-transparent market data
• A cross-platform identity graph linking the same card across marketplaces, languages, and catalogs
• Policy context for antitrust questions raised by vertically integrated grading, marketplace, and analytics firms
• A replicable method for studying price transparency in niche secondary markets with restricted data access
• Documented barriers to independent market research, to inform platform policy and research data access advocacy

Selected References

• G. A. Akerlof, "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism," The Quarterly Journal of Economics, vol. 84, no. 3, pp. 488–500, 1970.
• S. D. Levitt and C. Syverson, "Market Distortions When Agents Are Better Informed: The Value of Information in Real Estate Transactions," The Review of Economics and Statistics, vol. 90, no. 4, pp. 599–611, 2008.
• J. Heck et al., "Sales Characteristics of Pokémon Trading Cards: A Prospective One-Year Field Study," PLoS One, vol. 21, no. 3, Mar. 2026.
• E. M. Salonia, "Magic the Gathering: Economic Analysis of the Market of a Collectible Card Game," Working Paper, Jun. 2017.
• hiQ Labs, Inc. v. LinkedIn Corp., 31 F.4th 1180 (9th Cir. 2022).
• Meta Platforms, Inc. v. Bright Data Ltd., Case No. 3:23-cv-00077-EMC (N.D. Cal., Jan. 2024).
• Rasmussen v. Collectors Holdings, Inc., Case No. 8:26-cv-00897 (C.D. Cal., filed Apr. 2026).
• "Major Authenticators Graded More Than 26 Million Cards in 2025," cllct.com, Jan. 2026.

Affiliation & Contact

Card Research is an independent initiative led by Eric Schnelker, a student at Purdue University's Elmore Family School of Electrical and Computer Engineering. Eric holds B.S. degrees in Computer Science, Cybersecurity, and Mathematics from The University of Tulsa, with minors in Data Science and Music. The work draws on his direct experience as a collector and vendor in the secondary market.

Lead Researcher: Eric Schnelker
Affiliation: Student, Purdue University
Degrees: B.S. Computer Science, B.S. Cybersecurity, B.S. Mathematics — The University of Tulsa
Email: eric@card-research.org · eschnelk@purdue.edu
ORCID: 0009-0003-1288-9433
Pricing Tool: card-prices.com

Cite This Research

@misc{schnelker2026tcg,
  author = {Schnelker, Eric},
  title  = {Toward Pricing Transparency in Collectible Trading Card Secondary Markets: Challenges in Cross-Platform Data Aggregation Under Market Consolidation},
  year   = {2026},
  note   = {In preparation for IEEE submission},
  url    = {https://card-research.org}
}

Disclosures

Independence: This research is conducted on personal time with personal resources. The views expressed are the researcher's own and do not represent any employer or government entity. No employer or government funding, equipment, or non-public information is used.

Educational institution: This research is not sponsored, directed, or endorsed by Purdue University. The researcher's status as a Purdue student does not create an institutional relationship between Purdue and this project or Card Prices LLC.

Conflict of interest: E. Schnelker is the founder of Card Prices LLC, which operates card-prices.com, a pricing comparison tool in the market this research studies, and is developing a card scanning application. He also buys and sells cards as a vendor at trade events; that firsthand experience motivated this work.

Litigation: The researcher is not a party to, consultant for, or data provider to any litigation or regulatory inquiry referenced here. Legal proceedings are cited only as evidence of market conditions.

Acknowledgments

We acknowledge the use of Claude (Anthropic, Inc.) as an AI coding assistant for routine software engineering tasks such as boilerplate code generation, debugging, and deployment scripting. Research design, system architecture, data interpretation, analysis methodology, and written conclusions are the work of the lead researcher.