PSEPHOS reads who is holding what, weighs each position by the verified record of the wallet holding it, and declares a result only where the count can carry one. Six hours later it marks its own declaration and publishes how often it was wrong.
Every other tracker counts each large wallet the same, so the loudest position wins the room. A pollster has never worked that way: a respondent is weighted by how representative they have proven to be, and an unweighted poll is not a poll at all. This is that method, turned on open positions.
This only works because Hyperliquid is transparent by design. Every fill on its tape carries the wallets on both sides, every wallet's open positions are readable, and every closing fill is stamped with the profit or loss it made.
So a voter's record is not modelled, inferred or self-reported. It is read off the exchange that settled it. No API key is used anywhere in this office, and nothing is bought.
Nobody is invited onto the roll and nobody is curated onto it. The roll is simply whoever is actually trading — … wallets sighted so far, … of them weighed.
The venue's own tape names the wallets on both sides of every fill. PSEPHOS notes whoever appears — no leaderboard, no curated list, no wallet anybody chose to feature.
A voice is weighed on the edge it has actually made per dollar traded, and on how often it has been right. Never on how large it is. Weight is bounded at both ends so no single voice can decide a result.
Under six voices, or where fewer than a third of them have a proven record, nothing is declared and the board says which of the two it was. A number computed from three unproven wallets is not a weak signal — it is a fabricated one.
Six hours later the declaration is marked against what the price actually did. Moves inside the noise band are neither a hit nor a miss and are excluded from the rate, rather than quietly counted as wins.
A wallet with fewer than twenty closed trades has not earned an opinion yet. It is not scored as a loser and it is not thrown out — it is counted at base weight and reported separately, so every declaration can always say how much of it rests on wallets that have proven nothing.
A tracker that quietly treats an empty record as a zero is not being cautious. It is inventing evidence.
the full roll0 of 0 weighed wallets carry a record strong enough to lift their voice above base weight.
A count nobody marks is a rumour with a number attached. So every declaration is read again six hours later and marked in public — and the office refuses to state a hit rate at all until it has enough marks for one to mean anything.
reading the record…
A count describes where the weighed room is standing right now. It is not a prediction and it is never a recommendation.
Nothing here follows any one wallet. The entire point is that no single voice, however large, carries a declaration alone.
The office publishes its own hit rate, and refuses to state one before it has enough marks to mean something.
Every figure comes from the venue's public API. No keys, nothing bought, nothing taken from anyone's account.