DEX · BAL · $0.6B TVL. Balancer is the only protocol in this screen carrying an active notching cap. The November 2025 exploit of its V2 vault — a rounding-error flaw in the stable-pool math that drained well over $100M across chains — is exactly the fatal-flaw condition Methodology v1.0 was written to handle: a weakness that cannot be averaged away by good scores elsewhere.
Solvency is capped at BB (59), and this is the decisive fact of the rating. Before the cap, the weighted inputs would have produced a materially higher score; the cap overrides them because restitution is incomplete and the recovery is only partial. The underlying inputs are themselves weak — a 55% drawdown, the deepest in the universe, and a 4/10 risk score reflecting that the flaw reached core vault accounting rather than a peripheral integration.
Legitimacy screens at A, and the divergence from Solvency is the interesting part. Balancer's audit history was genuinely deep, the team disclosed and coordinated the response publicly, white-hat recovery efforts were real, and governance functioned through the crisis. The framework treats disclosure quality and solvency damage as separate questions — a protocol can handle a disaster honorably and still be uncreditworthy while the hole is open.
Growth screens at B: TVL halved, share down to ~5%, momentum 3/10. The exploit accelerated an erosion already underway as the boosted-pool and veBAL flywheel lost ground to newer AMM designs.
Composite B, inheriting the cap. The screen's read: the cap lifts only on verified full restitution and a re-audited V3 accounting surface. Until then, no amount of legitimacy or engineering pedigree moves the grade — which is precisely what a notching cap is for.
"A chain is as strong as its weakest link." The conservative composite takes the weakest dimension and inherits any notching cap, so a single structural weakness cannot be diluted by strength elsewhere.
