> ## Documentation Index > Fetch the complete documentation index at: https://pegana.xyz/docs/llms.txt > Use this file to discover all available pages before exploring further. # Intrinsic vs market > Every Pegana signal is the gap between two numbers — what the asset's mechanism says it's worth, and what the market actually pays for it. The framing that makes peg monitoring tractable is **two numbers, side by side**: * **Intrinsic** — what the asset *should* be worth, per its own mechanism * **Market** — what you'd actually receive swapping it Every other concept in Pegana — discount, state, threshold, alert — is derived from the gap between these two. ## Why both, not one A monitor that reads only **market** misses the case where the market is the broken side. The [mSOL/Pyth incident](/docs/case-studies/msol-pyth-may-2022) is the canonical example: Marinade's on-chain redemption (intrinsic) stayed at 1.18 SOL while a thin Mango pool printed 0.85 SOL for four minutes. A market-only oracle would have said "mSOL just crashed." The right answer was "the market lied for 4 minutes; intrinsic held." A monitor that reads only **intrinsic** misses the [USDC/SVB case](/docs/case-studies/usdc-svb-march-2023): Circle's redemption claim was always $1, but the secondary market priced doubt at $0.87 for 36 hours. An intrinsic-only oracle would have said "USDC is fine." The right answer was "intrinsic still claims \$1, but the market won't pay it until the backstop arrives." Holding both numbers in the same signal — and publishing both — lets the consumer decide which side to act on for their use case. ## How each side is sourced **Intrinsic** comes from the asset's own mechanism. Each class has its own canonical source — Sanctum for LSTs, Pyth Redemption Rate for yield-bearing, on-chain decode for CDPs, Pyth FX for fiat-pegged. See [intrinsic value](/docs/methodology/intrinsic-value) for the full table. **Market** comes from Jupiter's routed quote, converted to USD via Pyth. We always swap *to* USDC and reprice USDC → USD via Pyth, so the numeraire is consistent across every asset and a USDC depeg doesn't contaminate other assets' spreads. See [market value](/docs/methodology/market-value). ## The two-numbers payload Every Pegana API response includes both numbers explicitly: ```json theme={"theme":"github-dark"} { "asset": "jitoSOL", "state": "PEGGED", "discount": -0.0034, "intrinsic_usd": 213.42, "market_usd": 212.69, "intrinsic_sol": 1.180, "market_sol": 1.176, "since": "2026-05-26T14:32:11Z", "confidence": "high" } ``` You receive the spread (`discount`) for fast checks and the underlying numbers for your own logic. For LSTs we also expose the SOL-denominated values — many integrators care about the SOL spread, not the USD spread. ## Sign convention ``` discount = 1 − market_usd / intrinsic_usd ``` * `discount > 0` → market trades **below** intrinsic (most common stress case) * `discount < 0` → market trades **above** intrinsic (less common; often the LST case where MEV-driven yield outpaces the published `sol_value`) * `|discount| < threshold` → `PEGGED` A common confusion: the dashboard shows `+1.20%` as "1.20% discount below intrinsic," *not* "market is 1.20% above intrinsic." Our positive sign convention matches the trading-floor convention where "discount" means "the market is offering this asset at a discount to its book value." ## When intrinsic and market converge by construction For a `PEGGED` fiat stable, intrinsic = $1 and market ≈ $1, so the spread hovers inside ±10 bps. That's the calm regime — and it's the regime where the two-numbers framing looks redundant. ("Just publish market — what could intrinsic add?") The framing pays off precisely when they diverge. Holding both numbers in the same schema means the divergence is **structural**, not an afterthought. Your downstream consumer never has to wonder which number to act on — both are there. ## Next How the spread gets turned into a state, with hysteresis and dwell timers.