Discretionary Vesting
Discretionary vesting is a distribution structure that authorises a beneficiary to claim up to a scheduled maximum without obliging it to claim any of it. The contract sets a ceiling, the beneficiary chooses the amount and often the timing, so the projected unlock is an upper bound rather than an expectation of supply that will actually reach the market.
TradFi parallel: Like a vested but unexercised stock option: the right is live and the terms are known, but nothing touches the share count until the holder decides to exercise.
Key Takeaways
- 01Discretionary vesting grants the right to claim up to a maximum without any obligation to claim it, so the schedule sets a ceiling rather than a forecast
- 02A pure linear unlock model miscategorises any token with this structure and consistently overstates supply pressure
- 03Hyperliquid's whitepaper implies a uniform monthly projection of roughly 9.92 million HYPE from a 238 million core contributor allocation over 24 months, while actual claims have run between 1.4% and 17.6% of that figure
- 04Unlocked and claimed are separate states: through March 2026 roughly 3.19 million HYPE had been claimed against a combined ceiling near 39.67 million over the four months with verified on-chain amounts, a cumulative claim rate around 8%
- 05The correct output is a range: the projected unlock as the ceiling and worst case, the announced claim as the expected case, with the gap between them read as behavioural evidence
- 06The structure is not confined to team allocations. Aster's airdrop emissions are discretionary, with much of the supply held under team-controlled programmes rather than entering circulation on schedule
How It Works
Most unlock modelling assumes vesting is mechanical. Tokens accrue on a schedule, the schedule executes, supply arrives. Discretionary vesting breaks that assumption at the last step by separating authorisation from execution. Tokenomist states the consequence plainly: any token with a discretionary distribution mechanism, where team members have the right to claim up to a maximum but are not obligated to claim the full amount, will be miscategorised by a pure linear unlock model. The model is not wrong so much as mis-specified. It reports a ceiling and labels it a forecast.
Hyperliquid is the worked example. Its whitepaper implies a uniform monthly projection of roughly 9.92 million HYPE, derived by spreading a 238 million core contributor allocation over 24 months. Actual behaviour has looked nothing like that line. Verified claim amounts run 1,745,746 HYPE in November 2025, 1,125,766 in January 2026, 140,333 in February and 173,217 in March, with roughly 330,000 announced for April 2026, about $12.1 million against a projected 9,916,666 HYPE worth around $364 million. Across tracked months the claimed amount has ranged from 1.4% of the projected figure to 17.6%. The team announces a specific amount on the 6th of each month, and the Hyper Foundation has stated that most of the vesting actually completes between 2027 and 2028, a shape no 24-month linear model reproduces.
The residue accumulates as a distinct supply state. Across the four months to 23 March 2026 with verified on-chain amounts the ceiling totalled roughly 39.67 million HYPE, four tranches of 9,916,666, against about 3.19 million actually claimed, a cumulative claim rate around 8%. The remainder stays inside the 238 million HYPE core contributor allocation. A figure of 405.41 million is sometimes set against that 3.19 million as though it were the denominator, but it is total unlocked supply across every allocation, most of it the 310 million genesis airdrop that has been liquid since December 2024, so it is the wrong base for a core contributor claim rate. Unlocked and circulating are not the same thing under this structure, and a ceiling-only model books the whole authorisation as delivered supply when in practice it is an overhang that remains claimable at the beneficiary's discretion, indefinitely and at a time of its choosing. That is a different risk from a dated cliff, and it should be modelled differently.
The practical reading is a range rather than a point. The projected unlock is an authorisation, so treat it as the ceiling and the worst case. The announced claim is an execution commitment, so treat it as the expected case. Presenting both is what makes the gap legible, and the gap is information rather than noise: a consistent pattern of small claims against a high ceiling is a behavioural data point, and a team that consistently claims close to its ceiling produces a different one. Announcing also creates accountability, because a team that states 330,000 and then claims a million on-chain has made a public commitment and visibly broken it. Nor is the structure limited to team allocations. Aster's airdrop emissions are discretionary, which Tokenomist notes reduces mechanical sell pressure, and much of that supply moves through team-controlled distribution rather than free-floating circulation.
Real World Examples
Hyperliquid: a projection and a claim that differ by orders of magnitude
View →For April 2026 the Hyper Foundation announced roughly 330,000 HYPE, about $12.1 million. The whitepaper-derived projection for the same month was 9,916,666 HYPE, around $364 million. Both figures are correct. One is what the contract permits, the other is what the team said it would take, and only the second speaks to the supply likely to arrive, though a stated intention is not a settled outcome until the claim executes on-chain.
Hyperliquid: the unclaimed overhang
View →Across the four months to 23 March 2026 with verified on-chain amounts, core contributors claimed roughly 3.19 million HYPE against a combined ceiling near 39.67 million, a cumulative claim rate around 8%, with on-chain claims totalling $119.52 million. The unclaimed remainder sits inside the 238 million core contributor allocation. A 405.41 million figure quoted alongside it is total unlocked supply across every allocation, most of it the genesis airdrop, not a core contributor denominator. Supply that is unlocked but unclaimed is claimable at any time, which makes it a standing overhang rather than a past event.
Hyperliquid: reading the claim rate as a trend
View →The tracked series peaks at 17.6% of the projected ceiling in November 2025 and troughs at 1.4% in February 2026, with March at about 1.7% and April at roughly 3.3%. April is nearly double March and the second consecutive month-over-month increase, since February's 140,333 HYPE had already risen to March's 173,217. The absolute level stays far below the ceiling, but the direction is the part worth monitoring.
Aster: discretionary emissions outside a team allocation
View →Aster directs 53.5% of supply to airdrops and community incentives, with monthly emissions around $42.58 million, roughly 2.17% of circulating supply. Tokenomist notes these tokens often do not enter circulation immediately because they remain under team-controlled airdrop programmes, and lists discretionary airdrop emissions as a factor reducing mechanical sell pressure.
Why an announcement is worth more than a schedule here
View →Under discretionary vesting the schedule cannot commit anyone, so the informative artefact is the team's own statement plus its on-chain execution. Announce 330,000 and claim a million, and the breach is public and verifiable. Tokenomist tracks each event through Whitepaper Estimate, then Committed, then Completed, so the ceiling, the stated intent and the settled outcome are all visible at once.
Frequently Asked Questions
How is discretionary vesting different from a normal vesting schedule?
Authorisation and execution coincide only where the contract pushes tokens to the beneficiary. Plenty of ordinary fixed-date schedules are pull-based instead, from Sablier and Hedgey streams to merkle claim distributions, and there a separate claim transaction is still required, which is why claim data diverges from unlock data generally rather than only under discretionary structures. What discretionary vesting adds is that the amount is optional too: the contract authorises a maximum for the period, and the beneficiary decides how much of that maximum to take, or whether to take any. The dates still exist, but they describe when a right becomes exercisable rather than when supply arrives.
If the tokens are unlocked but never claimed, are they in circulating supply?
No, and conflating the two is the core modelling error. Hyperliquid's core contributors claimed only about 3.19 million HYPE through March 2026 against a combined ceiling near 39.67 million over the four months with verified on-chain amounts, so most of the authorised amount stayed inside the 238 million core contributor allocation. Those tokens are outside circulation until a claim transaction executes, but they remain claimable at any moment, so they should be carried as a standing overhang rather than either ignored or booked as delivered.
Should I use the projected unlock or the announced claim in a supply model?
Both, as a range. The projected unlock is the authorisation, so it belongs in the model as the ceiling and worst case. The announced claim is the execution commitment, so it belongs as the expected case. For most tokens the two converge and the distinction is academic. For a discretionary structure they can differ by an order of magnitude, and a model built on the ceiling alone will materially overestimate dilution.
How can I tell whether a token uses discretionary distribution?
Compare the projected unlock against what has actually been claimed over several periods. Persistent, large gaps in the same direction indicate the schedule is functioning as a ceiling rather than a mechanism. Tokenomist surfaces this by tracking claims through Whitepaper Estimate, Committed and Completed, and by showing cumulative unlocked against cumulative claimed, where a low claim rate is the signature of the structure.
Does a low claim rate mean the supply risk has gone away?
It means the risk has changed shape, not disappeared. Unclaimed authorised supply accumulates rather than expiring, so restraint today builds a larger claimable balance for later. What the record gives you is a behavioural prior: consistent claiming far below the ceiling is evidence about how the beneficiary behaves, and evidence is not a commitment. The direction of the claim rate is worth watching as closely as its level.
Related Terms
Track on Tokenomist
Supply-side analysis for educational purposes. Not financial advice. Verify assumption and precision labels on the relevant token page.
Tokenomist.ai provides a complete solution for supply-side tokenomics data. Analyze future token emissions, track vesting schedules, and compare standardized tokenomics and allocation across projects to gain actionable insights