$1.5 Billion Unlocked in a Week. Almost Nobody Claimed It.

The most widely quoted number in crypto tokenomics is one that almost never happens.

Roughly $1.5 billion of new supply came unlocked across HYPE, SUI and ENA in the first week of September, according to BeInCrypto's tally, led by Hyperliquid releasing 9.92 million HYPE — about $797 million — to core contributors on 6 September. Unlock trackers lit up. Threads warned about dilution. The figure travelled everywhere, because $797 million is a number that travels.

Here is the number that didn't travel. In a prior Hyperliquid unlock, roughly 1.75% of the released tokens were actually claimed. On that basis the September event put something closer to $14 million of newly movable supply into the world, not $797 million, and the gap between those two figures is not a rounding error. It is the entire thesis.

Unlocked, claimed and sold are three different things

A vesting schedule releases tokens into a claimable state. Somebody then has to sign a transaction to take custody. Somebody then has to decide to sell.

Each of those is a separate event with a separate probability, and the market routinely collapses all three into one headline as though a cliff date were a market order. Tokenomist's records show the Hyperliquid schedule releasing to core contributors on a linear basis, with the September tranche representing about 2.37% of the 464.91 million tokens already circulating. Two percent is dilution. It is not a crisis, and it is certainly not $797 million of sell pressure.

The honest counterpoint: a low claim rate is not safety. Unclaimed tokens are not cancelled. They sit in a queue, fully vested, waiting on a decision that can be made any morning by someone who has been watching the price for two years. Deferred overhang is still overhang, and the fact that insiders did not take custody in March tells you what they thought in March.

There is a structural reason claim rates stay low that nobody wants to say out loud. Claiming is legible. The moment a contributor pulls tokens out of a vesting contract, every analytics dashboard in the industry timestamps it, attributes it to a cohort, and produces a chart captioned "insiders are moving." A recipient who intends to sell eventually has a straightforward incentive to leave the balance where it is until the attention moves elsewhere. Low claim rates may be measuring discretion rather than conviction, and those look identical until they don't.

Pump.fun already ran this experiment in public

July gave the cleanest test yet. Pump.fun's twelve-month cliff expired, releasing 57.279 billion PUMP worth about $86.5 million across 121 wallets, as crypto.news reported — the largest single Solana unlock of the month and a textbook setup for a collapse.

The token went up more than 12% in the following day.

Part of that was an ongoing buyback absorbing supply. Part of it was that the release came in smaller than the version circulating in threads. And part of it was simply that becoming transferable is not the same as being transferred, which on-chain movement afterwards did not show in any volume worth panicking about.

One unlock is an anecdote. Two are a pattern worth pricing. But the lesson is narrower than "unlocks don't matter" — it is that the market has no reliable way to distinguish a cliff that will be sold from one that will not, so it prices all of them identically and is wrong most of the time in the same direction.

The date is trivia. The schedule is the product.

What actually separates a survivable unlock from a fatal one is visible months earlier, and it has nothing to do with the calendar.

Is the schedule enforced by a contract or by a promise? Are the recipients disclosed, or is "core contributors" doing a lot of work in a footnote? Can anyone check the remaining balance without asking the team? A cliff that anybody can verify in a block explorer — through a lock contract on a service like Team Finance or equivalent tooling — gets priced as a known quantity. A cliff that exists in a PDF gets priced as a rumour, which is why identical dilution produces wildly different reactions across two projects in the same week.

That asymmetry is the thing worth acting on. Every project raising now decides, at launch, whether its future unlocks will be events the market can model or events the market has to guess at. Launch infrastructure that ships enforced vesting and published allocations by default — the TrustSwap Launchpad among them — is not solving a marketing problem. It is deciding, two years in advance, which kind of headline the token gets on its cliff date.

August through early September moved more than $1.28 billion of supply into claimable state across dozens of projects, and the aggregate market barely registered it. Either the dilution is genuinely immaterial, or it is accumulating somewhere nobody is counting, in wallets that have not moved yet and will not move until the reason to move is obvious to everyone at once.

Both readings fit the data. Only one of them ends quietly, and the market will not know which it bought until the first large holder decides the wait is over.