Methodology

How EVA Hub measures and reports validator sentiment on Ethereum Improvement Proposals.

Important Framing

EVA Hub measures sentiment (support/opposition/neutrality toward a proposal), not operational readiness (testing complete, deployment timeline, blockers resolved). Results should be interpreted as "what operators think about this EIP" rather than "whether operators are ready to run it."

What We Measure

Stake-Weighted Sentiment - Each signal is weighted by the total ETH staked by the validators the signaler controls. An operator controlling 1,000 validators (32,000 ETH) has proportionally more weight than an operator controlling 1 validator (32 ETH).

Participation Breadth - We report unique participant count alongside stake-weighted results to provide context on how distributed sentiment is across the validator set.

Network Coverage - We report participating stake as a percentage of total estimated network stake, providing context for how representative results are.

Eligibility

Who Can Signal - Verified Ethereum validator operators who can prove control of one or more active validators on the Beacon Chain.

Verification Method - Operators connect the wallet that originally deposited ETH to the Beacon Chain deposit contract and sign a SIWE message proving ownership. The backend verifies the depositor address on-chain and links all associated validators.

Represented ETH Calculation - Calculated from the effective balance of linked validators at the stake snapshot timestamp for each signal window.

Signal Windows

Timing - Each EIP has a defined signal window (typically 7 days). Operators may cast or update their signal any time during the window. Only the final signal counts.

Stake Snapshots - Represented ETH is calculated from a stake snapshot taken at window open. The snapshot block number and timestamp are disclosed on each EIP page.

Results Publication - Results are published within 15 minutes of window close unless an incident is triggered requiring review.

Known Limitations

  • Results represent only participating validators, not all network validators
  • Stake delegation may allow single entities to appear as multiple participants
  • EVA Hub cannot fully detect all forms of duplicate representation
  • Sentiment does not indicate operational readiness or technical feasibility
  • Results are non-binding and should not be interpreted as governance decisions

Incident Handling

When integrity issues are identified, EVA Hub may:

  • Add incident banner - Flag results with context about the issue
  • Pause window - Temporarily halt signaling while investigating
  • Extend window - Add time if significant issues affected participation
  • Void window - Invalidate results entirely (with public rationale)

All incidents are logged publicly with timestamps and resolution status.

Data Handling

What We Collect - Wallet address, validator indices, signal choices, and timestamps. No email or personal identifying information required.

What Is Public - Aggregate results only. Individual signals are private by default.

Retention & Deletion - Signal records are retained indefinitely for result integrity. You can request deletion of your personal data at any time.

View full privacy policy

Questions about our methodology? Contact us on X