A score you can
take apart.
StockScore converts six observable signals into category points, then makes every gap visible. The formula is deterministic and versioned. It does not predict returns or tell you what to buy.
See why a score changes
Hold liquidity, activity, distribution, stability, and disclosures fixed. Move only the token’s premium to a $100 underlying reference.
Price integrity: 30/30. 2.0% premium to underlying reference; 2% tolerance.
Price integrity
Compares a token quote with a suitable underlying reference or, for PreStocks, the provider mark. The first 2 percentage points of public-equity divergence or 5 points when the traditional market is closed are tolerated. Each additional point deducts 2.5 category points. A provider mark costs 4 points because it is not independent. A stale quote can cost 10 points. For Apple, a Pyth confidence interval above 0.5% can cost up to 5 points.
Liquidity quality
Uses USD depth of the single largest observed Solana pool. Score = 25 × min(1, log10(max(depth, 1)) / 6). This is an observable depth proxy, not an executable slippage quote. Pools are not summed to avoid double counting.
Market activity
Uses 24-hour USD volume and buy-plus-sell transaction counts from that same pool. Score = 9 × min(1, log10(max(volume, 1)) / 5) + 6 × min(1, transactions / 1,000).
Distribution risk
Uses the largest ten token accounts divided by mint supply from Solana RPC. Score = 10 − max(0, concentration − 30%) / 7, clamped to 0–10. Accounts are unadjusted: unidentified pool, issuer, and treasury accounts remain included and the UI warns about that limitation.
Volatility & stability
Uses absolute 24-hour price change reported for the observed pool. The first 2 points of movement are tolerated; each additional point subtracts 0.6 score points. Short-window divergence history is not backfilled, so this category is a limited stability proxy.
Issuer transparency
Awards 2 points for an official product page, 3 for backing disclosure, 3 for redemption disclosure, and 2 for terms. Evidence comes from official issuer or provider material. This measures disclosure availability; it does not audit collateral or legal enforceability.
Missing data & confidence
When a score appears
Missing inputs receive no category points. Known points are normalized to 100 only when all three conditions hold:
- At least 65 of 100 possible points are measurable.
- Both pricing and liquidity evidence are available.
- Confidence is medium or high.
Otherwise, the scorecard says “Insufficient data” and shows a broad possible range. Missing data never means zero risk.
How coverage grows
StockScore checks 30 public xStocks listings every five minutes, completing roughly one catalog pass every three hours. A score needs an observed market price, a reference or labeled issuer quote, pool depth, and enough other category evidence. Repeated scans cannot create missing market evidence.
Opening a scorecard checks sources. It then shows the latest complete committed observation until another complete observation is stored.
What the rating means
Ratings do not assess expected returns, company quality, or investment safety. Every observation has an ID, methodology version, source timestamps, category points, and a declared predecessor. The scorecard and history read the same record.
Confidence thresholds
High: at least 90 weighted points measurable, no input older than 24 hours, and an independent reference. Medium: at least 65 points measurable with no input older than 24 hours. All other cases are low. Provider-mark comparisons cannot earn high confidence.
Market hours & freshness
Open and closed markets
When entitled Apple prices are available, Pyth feed metadata identifies whether the underlying equity market is open. When it is closed, the token can keep trading while the equity reference is frozen. The scoring model therefore allows a wider 5% price gap. Without entitled prices, the Pyth-based comparison remains unavailable.
- Open market: an input older than 30 minutes reduces price integrity.
- Closed market: an input older than 18 hours reduces price integrity.
- Any input older than 24 hours lowers confidence.
What each timestamp represents
Every scored input names its provider and observation time. Entitled Pyth Apple prices use the feed publication time when available. PreStocks, DEX Screener, and Solana RPC use retrieval time because their responses do not provide a consistent source update timestamp.
Sources & limits
Price and market sources
- Pyth Pro REST ↗ supplies the three Apple comparison prices when the server key has feed access. Public Pyth Core metadata provides a separately labeled market-session fallback only.
- Non-Apple public-stock references use an owner-configured provider only when display rights are verified. Otherwise, an xStocks indicative quote can support a lower-confidence comparison with an observed Solana pool. The issuer API does not provide publication time or a per-response upstream source, and its quote may reflect onchain markets.
- PreStocks ↗ supplies its product prices, marks, valuations, mint addresses, and descriptions.
- DEX Screener ↗ supplies observed pool depth, volume, transactions, and 24-hour price change. Solana RPC ↗ supplies mint supply and largest token accounts.
How to read the evidence
Since v1.1.0, PreStocks price integrity compares the observed pool price with the official provider mark. Earlier v1.0.0 snapshots used the provider token price; history marks that methodology boundary because the scores are not directly comparable.
Pool depth is not an executable swap quote. Largest token accounts are not necessarily beneficial owners. Provider marks and indicative quotes are not independent equity references. Token rights can differ from underlying shares and from other tokens referencing the same company.
Charts use collected observations only, without historical backfill. API outages, market closures, and rate limits can leave a score incomplete.
StockScore provides market analytics, not investment advice. Scores reflect available data and may be incomplete or delayed.
Business model
Eligible noncustodial Jupiter swaps can carry a disclosed, configurable StockScore integrator fee. Planned products include StockScore Pro for monitoring, alerts, and deeper history; a paid score and risk-data API; and market-quality analytics for issuers and venues.
Payments, subscriptions, sponsorships, and commercial relationships cannot influence an asset score. The deterministic scoring code receives only sourced market and disclosure evidence.