How results are decided
The full mechanism is in Resolution and voids. This is the short version for someone deciding whether to trust it.
| Item | Rule |
|---|---|
| Who proposes | Three resolver signers per chain, operated by Pools Markets. Addresses on Contracts and addresses. |
| Quorum | 2 of 3 |
| Challenge window | 30 seconds between proposal and finalisation |
| What a signature binds | Market, outcome, evidence hash, resolver epoch |
| If no valid result | Void after the resolution deadline, full refund, no fee. Anyone can trigger it. |
| Can discovery change an open market | No |
| Can a finalised result be changed | No |
What you are trusting
Resolvers are trusted to report honestly from the evidence their rules name. That is a real assumption, and it is stated rather than dressed up.
What bounds it: the evidence hash is committed on chain with the result and can be checked against the named source; a wrong proposal can be discarded for 30 seconds; the resolver set can only be rotated by a new epoch, and signatures from an old epoch are worthless; and the failure mode of a silent or offline resolver set is a refund, not a payment to the wrong side.
What the evidence is
- Launch events from the launchpad’s contracts, read from finalized blocks.
- Swap logs from PancakeSwap V2 or PumpSwap, replayed from finalized blocks over the whole observation window.
- For the “up 15% or more” questions on Robinhood Chain, venue observations sampled at fixed times and anchored to confirmed blocks.
Not evidence: a screenshot, a page’s DOM, an unanchored API response, or a model’s opinion.
Last updated on