How we calculate depeg risk scores for each token
For a quick reference of weights and tier thresholds, see How the score is computed in Docs.
Each token receives a score from 0-100 based on weighted risk signals. The score maps to a tier:
The final score is a weighted sum of normalized risk signals (each 0-1). Weights sum to 90%, leaving room for future features.
Beyond the base score, we track how quickly risk is changing. Rapidly rising scores get boosted for earlier warnings.
Different token types have different expected behaviors. The model adjusts for this:
To prevent alert fatigue, tier transitions require sustained conditions:
