An empirical investigation into pricing dynamics, market consolidation effects, and information asymmetry across the $50B+ collectible trading card secondary market.
The collectible trading card game (TCG) secondary market has undergone rapid consolidation since 2022, with a single corporate entity now controlling the dominant grading services (PSA, Beckett, SGC), the largest online marketplace (eBay), the primary TCG-specific marketplace (TCGplayer), a major auction house (Goldin), and pricing analytics tools (TeraPeak). This vertical integration creates systemic information asymmetry: the entities that set market prices also control access to pricing data.
This research investigates how consolidation affects pricing transparency by collecting and analyzing transaction data across multiple marketplace platforms. By aggregating publicly available pricing signals from diverse sources — including major card marketplaces, general auction platforms, and social commerce channels — we quantify price dispersion, identify arbitrage patterns, and measure the consumer impact of information gatekeeping in collectible markets.
The collectible trading card market reached an estimated $50.1 billion in 2025, driven by sustained interest in Pokémon, Magic: The Gathering, and Yu-Gi-Oh! products. Despite this market size, pricing transparency remains remarkably poor compared to other secondary markets of similar scale (e.g., used automobiles, real estate).
Since eBay's acquisition of TCGplayer in October 2022 ($295M), followed by Goldin Auctions in April 2024, and the pending GameStop acquisition of eBay (May 2026, $56B unsolicited bid), the market faces an unprecedented concentration of control:
A single corporate structure now controls or has announced intent to control: the three largest grading services, the two largest online marketplaces for trading cards, the largest collectible auction house, AI-based authentication technology, vault storage services, pricing analytics tools, and potentially 1,600+ retail storefronts.
Critically, TCGplayer's public API was restricted shortly after the eBay acquisition, limiting third-party access to pricing data that was previously available. This creates a structural barrier to independent pricing tools and research.
An estimated 15-25% of TCG secondary market transactions occur through social commerce channels — primarily Facebook Marketplace, Facebook Groups, Discord servers, and local trading communities. These transactions are almost entirely absent from existing pricing datasets, creating a systematic blind spot in market analysis.
Facebook Marketplace is of particular interest because it represents the largest informal marketplace for collectibles, operates without the fee structures that inflate formal marketplace prices (typically 10-15% seller fees), and captures local/regional pricing signals absent from national platforms. Understanding how prices in these informal channels compare to formal marketplaces is essential for accurate market modeling.
Before describing the system we built, it is important to understand why it was necessary. Several pricing tools exist for the TCG secondary market, and each has significant limitations that we have encountered directly as market participants.
Consider the Charizard from Pokémon's Legendary Collection. It exists in three variants: Holofoil, Reverse Holofoil, and a Deck Exclusive Non-Holo — each with substantially different market values. Most collectors don't know the Deck Exclusive exists, and pricing tools frequently conflate the three. Each variant then has five raw condition grades (NM, LP, MP, HP, DMG), and may also exist as a graded slab from PSA, CGC, or BGS at grades 1 through 10. A single card can occupy dozens of distinct price points, and a buyer or seller must identify the exact intersection of variant, condition, and grading status to arrive at an accurate value.
Compounding this, counterfeit cards pollute marketplace transaction data. Sales of fake cards — which most buyers and many sellers cannot reliably identify — are recorded alongside legitimate transactions, distorting price averages. Pricing tools that ingest completed sales without filtering for authenticity are incorporating noise from a counterfeit market that has grown significantly in recent years.
Nearly all pricing tools lack transparency in how "market price" is calculated. Some use rolling averages, others use lowest current listings, others use proprietary weighted formulas — and most do not disclose their methodology. When two tools report different prices for the same card, a user has no way to determine which is more accurate or why they differ.
Beyond technical inadequacy, the pricing tool landscape has structural conflicts of interest. Several services charge $3–50/month for access to data derived from publicly visible transactions, with limited transparency about what portion covers infrastructure versus profit margin. More critically, the platforms that generate the underlying transaction data are now consolidated under the same entity that controls grading, creating a vertically integrated system where the party that determines a card's condition grade also controls the marketplace where it is sold and the pricing data used to value it. Vendors and collectors across the market widely recognize this conflict.
We collect publicly available transaction and listing data from multiple marketplace platforms using a distributed collection infrastructure. Data sources include:
| Source | Data Type | Volume | Status |
|---|---|---|---|
| TCG-specific marketplaces | Listings, recent sales, market prices | ~35,000 products × 7 conditions | Active |
| General auction platforms | Completed sales, active listings | Historical + real-time | Pending API |
| Social commerce channels | Local listings, asking prices, time-to-sale | Regional + national sample | Planned |
| Auction houses | Realized prices, lot compositions | High-value items (>$100) | Planned |
Collected data is normalized into a unified schema that accounts for condition grading differences across platforms, variant/printing distinctions, and temporal price movements. Statistical analysis includes:
The research platform operates on cloud infrastructure with a distributed architecture across multiple compute instances. Data flows through a six-stage cloud database cascade:
The frontend research tool (comp.game-enthusiasts.com) provides public access to aggregated pricing comparisons. The entire stack operates on free-tier cloud resources by design.
The system catalogs products across multiple collectible card game ecosystems and languages, extending beyond the three major English-language TCGs:
| Catalog | Scope | Products |
|---|---|---|
| English TCG | Pokémon, Magic: The Gathering, Yu-Gi-Oh! | ~35,000 |
| English Printings | Variant/condition combinations per product | ~245,000 |
| Japanese TCG | Japanese Pokémon sets | ~18,000 |
| Topps | Vintage Pokémon (TV series era) | ~2,400 |
| Carddass / Bandai | Bandai-produced Pokémon cards | ~3,200 |
| Multi-language TCG | Korean, Chinese (Trad/Simp), and others | ~14,000 |
| Cross-platform sales | Fuzzy-matched transactions with visual verification | ongoing |
This multi-catalog scope is necessary because the secondary market does not respect platform boundaries. A Japanese-exclusive promo may appear on an English-language marketplace; a Topps card from 1999 may be listed alongside modern TCG singles. Accurate cross-platform pricing requires a product identity graph spanning all of these catalogs.
This research collects only publicly available data — listing prices, completed sale prices, and product metadata visible to any marketplace user. No private user data, personal information, or non-public transaction details are collected or stored.
A note on data collection methods: This research strongly prefers official API access over web scraping. Structured API access is more reliable, less burdensome on platform infrastructure, and produces higher-quality data. However, several major marketplace platforms have restricted or eliminated public API access to pricing data in recent years — in some cases immediately following corporate acquisitions. Where official API channels exist, we actively pursue formal data access agreements. Web scraping of publicly visible data is used only as a last resort when no API alternative is available.
This is an independent research initiative led by Eric Schnelker, an Operations Analyst for US Transportation Command (TRANSCOM) at Scott Air Force Base, Illinois. Eric previously served as a Computer Scientist at Tinker Air Force Base, Oklahoma, where he graduated from the Palace Acquire Program — the Air Force's civilian acquisition development program. Eric is currently pursuing a Master of Science in Software Engineering at Purdue University's Elmore Family School of Electrical and Computer Engineering. The research infrastructure is independently operated, drawing on the lead researcher's direct experience as a collector and vendor in the secondary market to inform data collection methodology and analysis.
Lead Researcher: Eric Schnelker
Position: Operations Analyst, US TRANSCOM · Scott AFB, IL (transitioning from Computer Scientist, Tinker AFB)
Program: MSECE in Software Engineering, Purdue University
Email: eschnelk@purdue.edu
Research Platform: game-enthusiasts.com