Four prices, four questions
| Price | Question it answers | Typical limitation |
|---|---|---|
| Underlying equity price | Where is the listed company’s share trading? | The token may not be redeemable into that share, may trade outside market hours or may embed fees and legal risk. |
| Issuer NAV or mark | What value does the issuer calculate for its product? | It may rely on the issuer’s assumptions and may not be executable. |
| Oracle price | What value is a smart contract instructed to use? | It can be stale, manipulated, wrongly mapped or disconnected from available exit liquidity. |
| Pool / venue price | What did recent buyers and sellers pay? | A small trade can set the displayed price in a shallow pool; larger trades move it. |
The underlying share closes at $100. An issuer marks its token at $100.20 after fees. A thin pool last traded at $94. An oracle still reports $100. All four observations may be real. A lender liquidating $500,000 cares most about the amount the market can absorb—not the prettiest number.
Stocks and blockchains keep different hours
US equity markets close overnight, on weekends and on holidays. Solana keeps trading. A token price can move while the primary share market is closed, or the token can keep repeating Friday’s equity close until Monday. A premium or discount during those hours may represent information, fear, low liquidity or simply incompatible timestamps.
Comparisons must therefore include publication time, market-open status and source. A fresh token trade compared with a day-old reference is not a clean arbitrage signal.
Oracle design determines failure mode
An oracle can use the underlying share, the token’s own venues, issuer NAV or a composite. Each choice solves one problem and creates another. Underlying-share feeds are deep but may ignore token-specific redemption and credit risk. Token-market feeds reflect the actual product but can be manipulated when liquidity is thin. Issuer marks know the product terms but are not independent.
For lending, the protocol also needs rules for staleness, confidence intervals, market closure, circuit breakers and what happens when sources disagree. Mapping the wrong ticker or decimal can be as dangerous as manipulation.
A correct oracle can still produce a bad liquidation
Suppose an oracle values collateral at $1 million and the lending protocol liquidates at 80% loan-to-value. If the actual pool can absorb only $50,000 near the quoted price, the lender cannot recover $1 million merely because the oracle says it is there.
This is the distinction between valuation and exit quality. Price impact, pool concentration, venue access, transfer restrictions and direct redemption all affect realizable value after default.
Questions to ask about any displayed price
- Which instrument does this number price: the share, token, note or issuer NAV?
- When was it observed, and was the underlying market open?
- Is the source independent of the issuer?
- What happens when the feed is stale or sources disagree?
- How much can actually be traded before price impact becomes material?
- Does the lender have access to the venue or redemption rail behind the price?
How RWA Sonar shows it
The analytics table names the reference source and keeps it separate from observed token trades. It shows premium/discount, venue spread, liquidity and trade age rather than presenting one “price” as complete. Health rules flag stale or wide disagreement; unknown coverage remains visible.
Takeaway
An oracle tells a program what number to use. It does not promise that the number is fresh, correct for this legal instrument, or realizable at the size that matters.