Methodology
How KYSC scores stablecoins. Every number on this site traces back to the framework below. Deterministic, reproducible, open.
Contents
Overview
KYSC produces a 0–100 trust score for each stablecoin using a 7-component weighted composite. Each component captures a distinct dimension of stablecoin risk. Components are scored independently on a 0–100 scale, then combined using published weights to produce the final trust score.
Dual-track scoring distinguishes centralized, fiat-backed designs (Track A) from decentralized, crypto-collateralized designs (Track B). Each track applies evaluation criteria appropriate to its architecture — you cannot penalize a decentralized protocol for lacking a bank custodian, nor credit a centralized issuer for on-chain governance it does not have.
Scores are further adjusted for jurisdiction-specific regulatory compliance, and can be viewed through 5 named profiles that re-weight components for different use cases (payments, treasury, DeFi, institutional, or default).
Trust Statement
KYSC never charges issuers — zero conflicts of interest.
We do not accept payment from stablecoin issuers for scoring, listing, or favorable treatment. Our scoring engine is deterministic: the same inputs always produce the same outputs. No manual overrides, no editorial discretion, no sponsored placements. Revenue comes from institutional data subscriptions and API access — never from the entities we evaluate.
Component Weights
The 7 components and their default weights. Weights sum to 100%. Named profiles (below) apply overrides for specific use cases.
| Component | Weight | Sub-Scores | Focus |
|---|---|---|---|
| Backing Quality | 30% | 6 | Measures the quality and verifiability of reserves backing a stablecoin. |
| Execution Cost | 18% | 5 | Assesses how easily the stablecoin can be traded or redeemed at par. |
| Operational Controls | 12% | 9 | Evaluates the organizational and legal structure protecting holders. |
| On-Chain Health | 12% | 5 | Real-time signal of operational health derived from on-chain data. |
| Operating History | 12% | 4 | Time-weighted evidence of reliability. |
| Yield Sustainability | 9% | 3 | For yield-bearing stablecoins, evaluates whether the yield source is sustainable and transparent. |
| Adoption Signal | 7% | 2 | Measures ecosystem integration depth and institutional confidence. |
| Total | 100% | 34 |
Scoring Components
Each component is scored independently on a 0–100 scale using the sub-scores listed below. Click any component name above to jump directly to its section.
Backing Quality
6 sub-scoresMeasures the quality and verifiability of reserves backing a stablecoin. Covers reserve ratio adequacy, composition quality (cash vs. treasuries vs. commercial paper vs. crypto), attestation rigor and frequency, auditor reputation, custodian arrangements, and banking concentration risk.
Sub-Scores
How fully collateralized the coin is. 1:1 backing = baseline. Over-collateralization scores higher.
Quality of reserve assets. Cash and short-term treasuries score highest. Commercial paper, crypto, or opaque assets score lower.
Frequency and depth of reserve attestations. Monthly point-in-time < monthly with methodology < continuous proof-of-reserves.
Reputation and independence of the auditing firm. Big Four > mid-tier > self-attested.
Quality of custodial infrastructure. Regulated custodians with bankruptcy remoteness score highest.
Distribution of reserves across banking partners. Single-bank dependency is penalized.
Execution Cost
5 sub-scoresAssesses how easily the stablecoin can be traded or redeemed at par. Considers DEX pool depth, CEX pair availability, Curve pool dominance, and direct issuer redemption capacity. Scores are adjusted relative to market cap — a $100M coin needs less absolute depth than a $10B coin.
Sub-Scores
Total value locked in decentralized exchange pools. Measures slippage at standard trade sizes ($100K, $1M, $10M).
Number and quality of centralized exchange trading pairs. Weighted by exchange tier and volume.
Presence and balance in Curve Finance pools — the benchmark venue for stablecoin liquidity.
Direct-to-issuer redemption availability, speed, minimums, and fee structure.
Raw liquidity volume independent of market cap normalization. Captures systemic importance.
Operational Controls
9 sub-scoresEvaluates the organizational and legal structure protecting holders. Covers legal entity structure, key person risk, transparency practices, communication quality, freeze/blacklist capabilities, custodian governance, and sanctions compliance posture.
Sub-Scores
Jurisdiction of incorporation, entity type, regulatory licenses held, and bankruptcy remoteness of reserves.
Concentration of control. Single-signer multisigs, founder dependency, and team depth are evaluated.
Public availability of reserve reports, smart contract audits, team identities, and operational documentation.
Quality and timeliness of issuer communications during normal operations and crisis events.
Whether the issuer can freeze/blacklist addresses. Assessed as a risk factor with nuance — regulatory compliance vs. censorship.
Governance practices of custodians holding reserves. Independence, audit rights, and insurance coverage.
OFAC/EU sanctions screening practices and historical compliance record.
Day-to-day operational controls: multisig thresholds, timelock delays, upgrade mechanisms.
Ability to survive adverse events: team departure, regulatory action, market stress.
On-Chain Health
5 sub-scoresReal-time signal of operational health derived from on-chain data. Tracks peg stability across timeframes, mint-burn activity patterns, freeze events, and supply volatility. This component updates most frequently and serves as an early warning system.
Sub-Scores
Maximum and average deviation from target peg over the trailing 24 hours.
Maximum and average deviation from target peg over the trailing 7 days. Captures sustained stress.
Ratio of minting to burning activity. Extreme imbalances signal potential stress or rapid growth.
Recent address freeze/blacklist events. Frequency and magnitude are assessed.
Rate of supply change. Rapid supply contractions are penalized more heavily than growth.
Operating History
4 sub-scoresTime-weighted evidence of reliability. Longer operating history, clean depeg record, consistent audit trail, and sustainable growth trajectory all contribute. New coins start with a track record penalty that decays over time.
Sub-Scores
Time since launch. Logarithmic scaling — the first year matters most, subsequent years add diminishing credit.
Historical depeg events weighted by severity and recency. A 5% depeg 3 years ago matters less than a 2% depeg last month.
Consistency and quality of smart contract and reserve audits over time.
Supply growth rate and pattern. Organic, steady growth scores higher than volatile spikes.
Yield Sustainability
3 sub-scoresFor yield-bearing stablecoins, evaluates whether the yield source is sustainable and transparent. Non-yield stablecoins receive a neutral score. Unsustainable yield (funded by token emissions or opaque sources) is penalized.
Sub-Scores
Origin of yield. T-bills and lending protocols score higher than token emissions or undisclosed sources.
Whether yield generation exposes holders to additional smart contract, counterparty, or market risk.
Whether current yield levels can persist. Historical yield stability and source durability.
Adoption Signal
2 sub-scoresMeasures ecosystem integration depth and institutional confidence. Protocol integrations indicate developer trust. Institutional adoption signals real-world validation. This component has the lowest weight because adoption follows quality — it should confirm, not drive, the score.
Sub-Scores
Number and quality of DeFi protocol integrations. Aave, Compound, MakerDAO listings weighted more than niche protocols.
Signals of institutional use: custody support, payment network integration, treasury adoption by DAOs and corporates.
Scoring Formulas
The final trust score is not simply a weighted average. Three adjustment mechanisms protect against misleading scores: a drag penalty for weak components, a reserve cap for undercollateralized coins, and lifecycle caps for defunct or winding-down coins.
Trust Score Composition
Step 1: Weighted Sum
pre_adjustment = round(Σ(component_score × weight))
Step 2: Drag Penalty
trust_score = round(pre_adjustment × (1 − drag_penalty))
Step 3: Reserve Cap
trust_score = applyReserveCap(trust_score, reserve_health)
Step 4: Lifecycle Cap
trust_score = applyLifecycleCap(trust_score, lifecycle)
Drag Penalty
Any component scoring below 50 creates “drag” on the overall score. The penalty is quadratic — a component at 45 contributes minimal drag, but a component at 10 creates substantial drag. Higher-weighted components produce proportionally more drag.
For each component where score < 50:
deficit = 50 − score
weight_factor = (component_weight / total_weights) × 7
component_drag = weight_factor × deficit² / 2500
Final penalty:
drag_penalty = min(Σ component_drag, 0.20)
Maximum drag penalty is capped at 20%. A coin with all components above 50 has zero drag.
Reserve Cap
No coin can score well overall if its reserves are critically weak. The reserve cap enforces hard ceilings based on reserve health.
| Condition | Maximum Score |
|---|---|
| Reserve Health < 25 | 50 |
| Reserve Health < 35 | 60 |
| Reserve Health < 45 | 75 |
Lifecycle Caps
Coins in terminal lifecycle states are capped regardless of component quality.
| State | Maximum Score |
|---|---|
| Defunct | 15 |
| Sunset | 35 |
| Wind Down | 50 |
Grade Scale
Numeric scores map to letter grades for quick comparison. Grades follow S&P-style letter conventions familiar to institutional finance professionals.
The investment-grade boundary sits at 70 (BBB), consistent with the S&P BBB− convention. Coins at 70+ have demonstrated adequate backing, governance, and operational maturity for institutional use with appropriate limits. Below 70 is speculative grade — material weaknesses exist that require explicit risk budgets.
Exceptional. Cash-equivalent treatment for core treasury and settlement.
Very strong. Suitable for core holdings with standard limits and monitoring.
Strong. Approved for institutional use with normal risk limits.
Adequate. Lowest investment grade — tighter limits, periodic review required.
Material weaknesses. Tactical or restricted use only.
Vulnerable. Explicit risk budgets required, not suitable for treasury.
Very weak. Distressed-like risk, no buy-and-hold usage.
Extreme fragility. Avoid for institutional programs.
Default or failed. Collapsed, de-pegged, or fundamentally broken.
Not Rated. Insufficient data to assign a defensible grade. Requires manual review.
NR Threshold
If a coin's confidence score falls below 0.40, it receives an NR (Not Rated) designation regardless of its numeric score. NR is not a grade — it means we lack sufficient data to assign a defensible rating. NR coins require manual review before inclusion in any institutional program.
Scale Design
The 9-grade scale was validated across 4 independent AI models (Grok, OpenAI o3, Gemini 2.5 Pro, Perplexity) using structured prompts against our score distribution. All models agreed on: 9–10 grade scale (not 7 or 21), no +/− modifiers at current coverage (82 coins), NR as a separate designation, and S&P-style letter grades for institutional familiarity. The only divergence was IG boundary placement (70 vs 75) — 3 of 4 models recommended 70, which we adopted. Full validation outputs are archived for audit.
Backing Models
KYSC classifies every stablecoin into one of 6 backing model taxonomies. The backing model determines which scoring track is applied and influences how individual sub-scores are weighted within components.
Fiat Custodied
Track ATraditional currency and cash equivalents held in regulated custody. The most common model for high-cap stablecoins.
Examples: USDC, USDT, TUSD
Crypto Collateralized
Track BOver-collateralized with crypto assets, typically through smart contract vaults. Decentralized but volatile collateral.
Examples: DAI, LUSD, sUSD
RWA Yield
Track ABacked by tokenized real-world assets generating yield, such as treasury bills or money market instruments.
Examples: BUIDL, USDY, USDM
Synthetic Hedged
Track BDelta-neutral positions using derivatives to maintain peg. No traditional reserves — stability from hedging.
Examples: USDe
Algorithmic
Track BAlgorithmic supply adjustment mechanisms without full collateral backing. Highest risk category historically.
Examples: AMPL, FEI (historical), UST (collapsed)
Commodity Backed
Track AReserves in physical commodities (gold, silver) or commodity derivatives. Peg target may differ from $1.
Examples: PAXG, XAUT
Dual-Track Scoring
Stablecoins are fundamentally different in their architecture. Comparing a fiat-custodied coin to a crypto-collateralized one using identical criteria produces misleading results. KYSC uses dual-track scoring to address this.
Track A — Centralized
For fiat-custodied, RWA yield, and commodity-backed coins. Emphasizes custodian quality, attestation rigor, banking relationships, and regulatory compliance. Reserve composition is evaluated against traditional finance standards.
Track B — Decentralized
For crypto-collateralized, synthetic hedged, and algorithmic coins. Emphasizes collateralization ratios, liquidation mechanisms, oracle reliability, governance decentralization, and smart contract audit quality.
Hybrid
Some stablecoins blend centralized and decentralized elements (e.g., DAI with RWA collateral). These receive a blended evaluation that applies relevant criteria from both tracks, weighted by the proportion of each architecture.
Confidence Model
Not all scores are created equal. Data availability varies dramatically across stablecoins — a coin with monthly Big Four attestations and deep on-chain data will have a more reliable score than one relying on sparse self-reported information. The confidence level tells you how much to trust the trust score.
Confidence is calculated per-component and then aggregated, weighted by component weights. Each component's confidence reflects the coverage, recency, and quality of its data inputs. A coin can have verified confidence on liquidity (rich on-chain data) but provisional confidence on reserves (no recent attestation).
Strong data coverage across all components. Score is reliable for decision-making.
Good data with some gaps in non-critical components. Score is directionally reliable.
Material data gaps exist. Score may shift significantly as data improves.
4-Factor Confidence Formula
confidence = completeness × 0.40 + freshness × 0.25 + track_record × 0.15 + corroboration × 0.20
| Factor | Weight | What It Measures |
|---|---|---|
| Data Completeness | 40% | Ratio of known data points to total required. Track A needs 16 points; Track B needs 15. |
| Data Freshness | 25% | How recent the attestation or on-chain data is. Attestation ≤15 days = 1.0; Track B on-chain = 0.9 (always fresh). |
| Track Record Length | 15% | Months since launch. ≥60 months = 1.0; <12 months = 0.5. |
| Cross-Source Corroboration | 20% | Whether data is verified by independent sources. Big Four auditor + attestation = 0.95; self-reported = 0.6. |
Confidence is clamped to [0.1, 1.0]. For IMPAIRED lifecycle: confidence × 0.85. Below 0.40 = NR (Not Rated).
Score Adjustments
Before the final composite is calculated, component scores are adjusted for behavioral flags and lifecycle state. These adjustments reflect real-world risks that cut across standard component scoring.
Behavioral Flag Adjustments
Flags derived from a coin's classification data apply targeted penalties to specific components.
| Flag | Component | Delta |
|---|---|---|
| CEX counterparty exposure | Backing Quality | −5 |
| Bridged asset | On-Chain Health | −8 |
| Active DeFi strategy | Backing Quality, On-Chain Health | −5 each |
| Rebasing token | Operating History | −5 |
Lifecycle State Adjustments
Coins in distressed lifecycle states receive additional component-level penalties before scoring.
| State | Component | Effect |
|---|---|---|
| Restricted | Execution Cost | −15 |
| Restricted | Operational Controls | −10 |
| Impaired | Operating History | −10 |
| Wind Down / Sunset | Adoption Signal, Yield Sustainability | Set to 0 |
Jurisdictions
A stablecoin's trust profile changes depending on where you operate. MiCA-compliant coins score higher in the EU. GENIUS Act-aligned issuers score higher in the US. KYSC computes jurisdiction-adjusted scores for 9 regions by applying regulatory compliance bonuses and penalties to the global baseline score.
Jurisdiction adjustments affect the Governance component most heavily, but can also impact Reserve Health (e.g., reserve composition requirements) and Track Record (e.g., licensing history).
| Code | Jurisdiction | Framework | Description |
|---|---|---|---|
| GLOBAL | Global (Baseline) | — | No jurisdiction-specific adjustments. The raw composite score. |
| US | United States | GENIUS Act | Federal stablecoin legislation requiring reserve composition, redemption rights, and issuer licensing. |
| EU | European Union | MiCA | Markets in Crypto-Assets Regulation. E-money token classification, reserve requirements, and whitepaper obligations. |
| UK | United Kingdom | FCA | Financial Conduct Authority oversight. Focuses on consumer protection and systemic stability. |
| SG | Singapore | MAS | Monetary Authority of Singapore framework. Emphasis on reserve composition and redemption guarantees. |
| HK | Hong Kong | HKMA | Hong Kong Monetary Authority stablecoin licensing regime. Focus on reserve quality and issuer fitness. |
| AE | UAE | CBUAE/VARA | Central Bank of the UAE and Virtual Assets Regulatory Authority dual oversight. |
| JP | Japan | JFSA | Japan Financial Services Agency. Strict requirements on asset segregation and issuer capital. |
| BR | Brazil | BCB | Central Bank of Brazil. Emerging regulatory framework for digital payment tokens. |
Regulatory Frameworks
KYSC evaluates compliance against 6 major regulatory frameworks. Each framework defines specific requirements for reserve composition, attestation, governance, and consumer protection. Compliance (or non-compliance) with these frameworks drives jurisdiction-specific score adjustments.
GENIUS Act
USEnactedFederal Payment Stablecoin Act. Requires full reserve backing, monthly attestation by registered public accounting firm, and direct redemption rights. Applies to issuers above $10B via federal charter; state-level for smaller issuers.
MiCA
EUEnactedMarkets in Crypto-Assets Regulation. Classifies stablecoins as e-money tokens (EMTs) or asset-referenced tokens (ARTs). Requires white paper, reserve custody in EU credit institutions, and limits on non-EUR stablecoin transaction volumes.
FCA
UKEnacted (effective 2027)Financial Conduct Authority regime. Enacted but not yet enforced — authorization window Sep 2026 – Feb 2027, effective Oct 2027. Stablecoins used for payments must meet reserve, redemption, and governance requirements.
MAS
SGIn forceMonetary Authority of Singapore Single Currency Stablecoin Framework. Requires SGD-pegged or G10-currency-pegged coins to hold reserves in cash/equivalents, with daily valuation and segregated custody.
HKMA
HKIn forceHong Kong Monetary Authority licensing regime for fiat-referenced stablecoin issuers. Requires a Hong Kong entity, adequate reserves, and compliance with anti-money laundering standards.
Basel III/IV
GlobalIn forceBank capital classification for crypto-asset exposures. Stablecoins meeting strict criteria qualify for Group 1b (lower capital charge). Non-qualifying stablecoins face Group 2 treatment (1,250% risk weight).
Named Profiles
Different users care about different risks. A payment processor needs liquidity above all else. A corporate treasury needs reserve quality and governance. KYSC offers 5 named profiles that re-weight components to match specific use cases. The Default profile uses the published weights. Other profiles apply overrides.
| Profile | Use Case | Weight Adjustments |
|---|---|---|
| Default | Standard weights as published. No adjustments. | None |
| Payment | Optimized for payment corridors and settlement. Emphasizes liquidity and on-chain reliability. | Liquidity +, On-Chain Health +, Reserve Health − |
| Treasury | Optimized for corporate treasury and reserve management. Emphasizes reserve quality and governance. | Reserve Health +, Governance +, Adoption Signal − |
| DeFi | Optimized for DeFi composability and protocol integration. Emphasizes governance transparency and ecosystem adoption. | Governance +, Adoption Signal +, Track Record − |
| Institutional | Optimized for regulated institutional use. Maximum emphasis on reserves, compliance, and auditor quality. | Reserve Health +, Governance +, Yield Sustainability − |
Profile weight adjustments are relative to Default weights. The exact override values are calibrated to maintain meaningful score differentiation without creating extreme distortions. Scores across profiles remain on the same 0–100 scale.
Worked Example: USDC
Walk-through of how USDC's trust score is computed, from raw inputs through component scoring, drag penalty, reserve cap, and final grade assignment.
Step 1: Component Scores
| Component | Weight | Score | Weighted |
|---|---|---|---|
| Backing Quality | 0.30 | 90 | 27.0 |
| Execution Cost | 0.18 | 80 | 14.4 |
| Operational Controls | 0.12 | 90 | 10.8 |
| On-Chain Health | 0.12 | 96 | 11.5 |
| Operating History | 0.12 | 91 | 10.9 |
| Yield Sustainability | 0.09 | 92 | 8.3 |
| Adoption Signal | 0.07 | 100 | 7.0 |
| Pre-Adjustment Score | 90 | ||
Step 2: Drag Penalty
All 7 components score above 50, so no drag penalty applies. drag = 0.00
Step 3: Reserve Cap
Reserve Health = 92, well above all cap thresholds (25, 35, 45). No cap applies.
Step 4: Lifecycle Cap
Lifecycle = ACTIVE. No cap applies.
Result
Confidence: 0.97 (verified). No flags. No caps applied. The score reflects strong performance across all 7 components with Execution Cost (80) as the lowest — still well above the drag threshold.
Methodology Changelog
Material changes to the scoring methodology are documented here. Minor constant adjustments and data pipeline improvements are not listed.
- • Grade scale redesigned: 10 grades → 9 (C grade removed). Multi-model validation (4 AI models).
- • Investment-grade boundary moved from 80 to 70 (BBB).
- • Commodity-backed scoring module added (PAXG, XAUT).
- • Composition quality: sovereign risk adjustments for government bonds and cash deposits.
- • Era-segmented track record scoring for coins that changed backing models.
- • Peg deviation scoring now supports non-USD peg targets.
- • Coverage expanded to 82 coins across 6 backing models.
- • Dual-track scoring (Track A / Track B) introduced.
- • 6 backing model taxonomy replacing 3-category system.
- • Drag penalty and reserve cap mechanisms added.
- • Confidence model: 4-factor weighted formula replacing binary data coverage.
- • 5 named profiles (Default, Payment, Treasury, DeFi, Institutional).
- • 9 jurisdiction-specific compliance evaluation.
- • Initial 5-component scoring (reserve, liquidity, governance, on-chain, track record).
- • 7-grade scale (A through F).
- • Single-track scoring for all backing types.
- • Coverage: 30 coins.