About Our Methodology

Excellent is built on a simple premise: better data leads to better decisions. We aggregate pricing from every major marketplace, validate it against actual sales, and surface the opportunities that matter.

Data Sources

Our pricing engine ingests data from multiple authoritative sources to build a consensus view of the market:

Graded Card Pricing

Real-time graded card pricing across all major grading companies. PSA, BGS, CGC, and SGC grade-specific values updated continuously.

Historical Market Data

Historical price data, chart analytics, and market trends. Coverage includes Japanese cards and sealed products. Grade-mapped pricing from raw through PSA 10.

Live Marketplace Listings

Live marketplace listings, market prices, and buylist data. Real-time availability and competitive pricing from verified sellers.

Completed Sales

Completed sales and sold listings. The ground truth for what cards actually sell for, not just what they are listed at.

No single source tells the whole story. By cross-referencing prices across marketplaces, we filter out outliers, detect stale listings, and provide a more accurate fair market value.

Multi-Source Consensus

Our pricing engine does not simply average prices. It applies a weighted consensus model:

  1. Source reliability weighting — Completed sales carry the highest weight as confirmed transactions. Listed prices are discounted.
  2. Recency bias — Recent data points are weighted more heavily. A sale from yesterday matters more than one from 60 days ago.
  3. Outlier filtering — Prices that deviate significantly from the cluster are flagged and excluded from the consensus calculation.
  4. Grade hierarchy validation — We verify that grade-specific prices follow logical ordering (PSA 10 > PSA 9 > PSA 8, etc). Contaminated data where this ordering breaks is automatically filtered.
Expected Value (EV) Model

The core of our investment analysis is the Expected Value model. For any raw card, we calculate the expected return from sending it to grading:

EV = Σ(P(grade) × Value(grade)) − Raw Price − Grading Cost

Where P(grade) is the probability of receiving each grade based on historical submission data.

The model accounts for:

  • Grade probability distribution — Based on PSA population reports and historical submission outcomes for each card and set.
  • Grade-specific market prices — What each graded version actually sells for, not what it is listed at.
  • Grading costs — Including submission fees, shipping, and insurance at current PSA/CGC/BGS service levels.
  • Time value — Grading turnaround times and the opportunity cost of capital tied up during the process.

A positive EV means the expected graded value exceeds the total cost of raw card + grading. We further refine this with a Sharpe-like ratio that adjusts for the variance in outcomes.

How Alerts Work

Our alert system monitors the market continuously and triggers notifications based on multiple criteria:

  • Price target alerts — When a card hits your specified buy or sell target.
  • EV threshold alerts — When a card crosses into positive EV territory for grading.
  • Arbitrage alerts — When the price difference between marketplaces creates an immediate opportunity.
  • Trend alerts — When a card shows significant momentum (3+ consecutive days of movement in one direction).

Premium members can set custom hunt criteria with specific conditions, and our system continuously scans for matching opportunities.

By the Numbers
125,601
Cards
18,098,634
Price Points
991
Sets
4
Languages

English, Japanese, Korean, and Chinese cards tracked. Updated daily.

See the data in action

Browse our free card database, or create an account for full access to price history, EV calculations, and alerts.

Browse Cards Create Free Account