/ METHODOLOGY

How CollectIQ values cards

CollectIQ turns fragmented collectible-market data into research you can act on. This page explains what data we currently have, how we match it to your card, and how we express confidence in a valuation. The exact proprietary weighting formula is not disclosed.

Active listings vs. verified sold comps

There is a real difference between what a seller is asking and what a buyer actually paid. CollectIQ treats these as two different categories:

  • Active listings — currently for sale on eBay via the eBay Browse API. These are asking prices and typically overstate true market value on illiquid cards.
  • Verified sales — completed transactions with a confirmed sold price. CollectIQ was not approved for eBay's Marketplace Insights API, so we do not currently have direct access to eBay sold data. We are evaluating licensed third-party providers such as Card Ladder, CardSight, SportsCardsPro/PriceCharting, and JustTCG.
  • Auction results, accepted offers, index prices, price-guide values, and user-imported comps — additional categories that may appear as verified sold-comp providers are integrated.

We never present an active listing as a sale, "last sold," or completed transaction.

How comps are matched to a card

Every observation is matched against the card's core identifiers:

  • Card name (including native and English variants)
  • Set name and set year
  • Card number
  • Parallel / variant (holo, 1st Edition, shadowless, reverse, etc.)
  • Grader and grade tier — PSA 10 ≠ BGS 10 ≠ BGS 10 Black Label; CGC 10 ≠ CGC 10 Pristine
  • Language and printing (WOTC vs. Celebrations reprint, etc.)

Each observation gets a match tier — Exact, Strong, Moderate, Weak, or Ignore — and, where relevant, a plain-language "Why this comp?" explanation such as "Strong match: same set, card number, parallel, grader, and grade."

How valuation confidence is determined

Every valuation carries one of four confidence tiers:

  • High Confidence — multiple exact-match verified sales agree on a tight price range.
  • Moderate Confidence — some relevant data, but mostly active listings, few verified sales, or a wider spread.
  • Low Confidence — few relevant comps or comps disagree meaningfully.
  • Insufficient Data — we cannot responsibly value this card yet.

When there are no verified sales, confidence is capped at Moderate — we will not display High Confidence based purely on active asking prices. When confidence is Low or Insufficient, we suppress directional buy/sell recommendations.

How outliers are handled

We identify and down-weight or exclude observations that skew results — bulk lots, damaged copies, misgraded slabs, reprint parallels being sold as originals, and "buy it now" listings priced far outside the recent distribution. Excluded rows remain visible on the asset page with an IGNORE tag so you can see the raw data.

How price-guide data differs from transactions

Sources like TCGplayer's market price are aggregated index values — useful as a floor or sanity check but not a per-transaction sold price. We surface index/guide data separately from verified sales and factor it in at a lower weight.

The proprietary formula

The exact weighting formula that combines match quality, recency, sample size, verified-vs-active mix, and momentum into the CollectIQ Score is proprietary and not disclosed. This page documents the categories of inputs and the guardrails.

Data providers

  • Currently used: eBay Browse API (active listings), Pokémon TCG API + TCGdex (catalog + index prices), PSA Population (graded population).
  • Under evaluation for verified sold comps: Card Ladder, CardSight, SportsCardsPro / PriceCharting, JustTCG, and other licensed providers of historical-sales data.

Our internal schema stores each observation with a category (Active Listing, Verified Sale, Auction Result, Aggregated Price Estimate, Price Guide Value, User-Imported Comp), so adding a new sold-comp provider does not require redesigning the app.

Questions? Contact us.