I separate buybacks already happening from promises that still need something to happen first. That distinction runs through Lark Davis's Q4 watchlist: ENA, HYPE, STONK, AERO and Sui. His case rests on fee-funded purchases, changes to token releases, a merger and a privacy feature awaiting production readiness. The figures and schedules below come from his account; his predictions remain predictions.
Ethena: an unlock ends, but supply keeps coming
Ethena's ENA case combines newly available supply with a proposed source of buying demand. An unlock makes previously restricted tokens transferable, while vesting releases an allocation over time. Neither requires recipients to sell, but both increase what they could sell.
October 5 was the final original investor unlock: about 1.4 billion ENA, or about 14% of circulating supply, released at once. The previous schedule would have released those tokens monthly until March 2028. The change brought about 17 months of potential selling pressure into a single day.
I would not confuse the end of that schedule with the end of token releases. Team vesting did not accelerate. About 94 million ENA remained in the immediate team release, and team and foundation allocations continue at around 134 million tokens a month into early 2028. Davis describes that as $100 million worth of monthly selling pressure if recipients sell.
In late August, the foundation bought remaining locked tokens from large seed wallets that had sold after the October 2025 peak. Holders who had never sold declined the same offer. The buyout's size and price were not disclosed, so the remaining investor overhang cannot be treated as resolved.
Stablecoin X, with NASDAQ ticker USDe, holds about 3 billion ENA, roughly a fifth of supply. Its lock was also due to lift, although any sale still needs the foundation's written approval. That restriction creates an approval step, not a permanent bar on sales.
The proposed buying demand comes from Ethena's fee switch. Governance approved it on September 2 with zero votes against it. Once active, the mechanism would direct 95% of net revenue toward programmatic ENA buybacks, purchases made under a defined rule.
Activation requires USDe supply to average $7.5 billion over 14 days. Recent supply was around $5 billion, with incentives winding down after more than $750 million in rewards. I would keep approval and activation separate: the vote passed, but the supply condition still has to be met.
The creator considers Ethena a bear-market survivor whose business model has held up despite earlier criticism. The narrower question for ENA is whether USDe reaches the threshold and subsequent revenue-funded purchases create meaningful demand alongside continuing releases.
Hyperliquid: buybacks depend on trading volume
Hyperliquid already has a buyback mechanism. Its perpetual futures contracts generally have no fixed expiry date. Traders can use leverage, which magnifies exposure and can make losses develop quickly.
Reported 30-day perpetual trading volume was about $215 billion, about 35% of decentralized perpetual volume and more than three times the next venue's volume. Lifetime volume exceeded $5 trillion. Open interest reached a record of around 8.8 billion in late September; that measures contracts still open, rather than all trades made during a period.
Average daily revenue rose from about 1.5 million in Q2 to about 3 million in Q3, with some August days topping 5 million. I read those figures as evidence of recent activity, not a promise that the pace will continue.
Almost 99% of fees flow to the Assistance Fund, which buys HYPE on the open market. Cumulative buybacks were around $1.28 billion, and weekly purchases increased from roughly 5 million to 20 million in August. Here, customer activity already funds token purchases rather than waiting for an activation condition.
US spot ETFs, a NASDAQ-listed treasury vehicle and a Binance listing broaden access to HYPE. The creator argues that this larger potential buyer base strengthens the case. Easier access does not establish how much lasting demand those routes will attract.
I would separate those developments from the institutional-buyer theory. Hyperliquid Labs is unstaking 3.75 million HYPE for a private sale to one institution. ICE CEO Jeffrey Sprecher described Hyperliquid as bigger than NASDAQ in May. The creator connects the two and speculates that ICE, the Intercontinental Exchange and owner of the New York Stock Exchange, might be the buyer.
The earlier comment does not identify the transaction's counterparty. That theory needs direct confirmation before it carries the same weight as exchange activity or completed buybacks.
Competition can reduce trading volume and therefore the money available for purchases. The creator also warns that the rally to around 100 bucks may already reflect substantial good news. A working buyback mechanism does not settle whether the token's price assumes too much future growth.
STONK: burns depend on speculative activity
STONK is the token behind STONK Fun, a Solana launchpad that launched in July. Launchpads help new tokens enter the market, so their income depends on people wanting to create and trade those assets.
STONK Fun generated about $32 million in revenue after launch, with about 60% directed toward buying STONK and burning it. Buying creates demand; burning permanently removes the purchased tokens from usable supply.
Lifetime buybacks were around 19.3 million across roughly 218,000 swaps. A mid-September weekly revenue spike of about $5 million ranked roughly 10th on DefiLlama's protocol income list. An early September snapshot put lifetime revenue around $18.6 million, so revenue accumulated quickly during that stretch.
I would focus on whether the activity lasts. A rush of launches can generate substantial fees without establishing a stable customer base. Purchases and burns can work as intended even while demand for the underlying service falls sharply.
The creator warns that launchpad revenue is fickle and that STONK Fun competes with the well-funded Pump Fun. In their assessment, declining interest in launching tokens against stocks or crypto coins would also reduce burn activity.
Burning tokens does not require buyers to value the remaining supply more highly. If demand falls faster than supply, the price can weaken despite continued burns. A cumulative total therefore cannot replace evidence about current activity.
The creator calls STONK Fun the only distinctly newer, more speculative entry among the five because its longevity remains uncertain. Early fee generation establishes initial interest; repeat activity after the launch excitement would provide stronger evidence of a lasting business.
Raydium: an indirect route with shared exposure
Raydium is an additional idea attached to STONK, not a separate member of the five-token list. The creator describes RAY as a picks-and-shovels approach: owning exposure to infrastructure that benefits from another business's activity, rather than relying only on that business itself.
STONK Fun's volume also feeds Raydium's fee income. RAY is described as trading at about 6X trailing revenue, with buybacks already live. Trailing revenue measures past income, so the multiple's usefulness depends on whether that period offers a reasonable guide to what comes next.
I would be careful with a revenue multiple during a launch-driven surge. The arithmetic can be correct while the assumption is weak: a token can look inexpensive against recent earnings that later fall.
Q2 gross protocol revenue was about $17 million, compared with $114 million in Q3 2025, a rise the creator attributes to STONK Fun. Cumulative holder revenue was around $65 million. These measures should stay separate: gross protocol revenue concerns income at the protocol level, while holder revenue concerns value attributed to holders.
The creator expects Raydium to remain even if STONK Fun fails, but also expects a substantial revenue hit. I see that as shared exposure through a different business structure, not an escape from launchpad risk. An infrastructure business can have other uses while depending heavily on one application for its latest growth.
Aerodrome: the merger needs users on new chains
Aerodrome and Velodrome are scheduled to merge on October 21 into one protocol with one AERO token, live on seven chains from day one. The creator regards this as the clearest dated event on the list.
The networks are Base, Ethereum, Optimism, Arc, Ink, Robinhood chain and Arbitrum. Staked AERO holders will direct rewards across them. Staking commits tokens under the protocol's rules; here, it also gives holders influence over where rewards go.
The venues already handle about 17% of EVM spot volume, and the team wants to triple that share. EVM, the Ethereum Virtual Machine, is the execution environment used by Ethereum and compatible networks. Spot trading concerns the assets themselves, unlike the perpetual contracts discussed earlier.
I would keep the team's goal separate from a forecast, as the creator does. Serving more networks creates more opportunities to attract users, but those users still have alternatives.
Reported revenue was about $15 million over 30 days and roughly $485 million cumulatively, with holders already receiving a share. Almost all of that activity comes from Base. The creator's interest lies in extending that business beyond the network where it has built most of its activity.
Success elsewhere depends on product-market fit: whether users find the product useful enough to keep using it. The creator warns that success on one chain does not ensure success on another, where competition, liquidity and user preferences can differ.
AERO jumped substantially after the announcement. The creator warns of a possible buy-the-rumor, sell-the-news reaction, with anticipation lifting the price before traders take profits at the event. I would judge the technical merger separately from whether it attracts additional business.
The creator predicts that meaningful volume on the new chains after October 21 could produce a substantial upward repricing. That remains an adoption-dependent prediction; sustained activity, not the announcement alone, would support it.
Sui: confidential transfers still need audits
Sui's proposed catalyst is a product feature rather than a revenue distribution rule. Confidential transfers would encrypt balances and transfer amounts on-chain while leaving sender and receiver information visible. Issuers would control who can see the sensitive data.
The creator frames the feature as compliance-friendly privacy for payments, payroll and treasury activity. It would restrict access to financial details without concealing every detail about the participants.
I find that distinction more useful than a broad privacy claim. Visible participants with hidden amounts are different from full anonymity. Assessing the feature means checking both what it protects and who can reveal that information.
Gasless stablecoin transfers are already live on Sui mainnet. Gas is the charge associated with executing network activity, so gasless transfers concern the user's transaction experience. Confidential transfers address a separate question: who can inspect the financial details.
The confidential-transfer devnet, a development environment, has been live since June. Testnet is expected later this year, allowing developers to examine behavior before production deployment. Mainnet, the live production network, is expected to follow audits.
Co-founder Aman expects the feature before year-end. At the end of September, however, it was unaudited and not production-ready. I would treat the date as a target, not evidence that the remaining technical work is complete.
Audits can identify security weaknesses but cannot guarantee flaw-free software. Implementation matters especially when users rely on a feature to protect sensitive information. The creator's Q4 test is whether confidential transfers clear testnet and receive audits on time.
The creator compares Sui's potential comeback with Solana's trajectory across cycles and predicts that the feature could become a major catalyst. That analogy does not prove adoption. Development progress comes first; commercial importance depends on whether people find the capability useful.
Compare the evidence, not just the catalysts
A governance vote establishes approval. A completed buyback establishes that purchases occurred. A merger date is a planned event, while a development environment establishes that work is underway. None proves that the next stage will succeed.
I also distinguish business income from value reaching token holders. Purchases, burns and distributions connect revenue to a token in different ways. The rules governing that connection matter as much as the income figure.
Even a direct connection leaves room to overestimate future growth. A genuine improvement in a business can coexist with a token price that expects the improvement to last longer than it does.
Thin liquidity can make entering or exiting more expensive than the displayed price suggests. Smart-contract failures can disrupt activity or put assets at risk. Market sentiment can change before a scheduled milestone arrives.
I use those risks to decide what to examine: the rules linking fees to holders, the security work behind a deployment, or actual activity after an event. They are more useful as research questions than as a repeated warning beside every token.
Track what would change the case
A watchlist becomes useful when it records what would change your assessment. Write down each thesis's dependency and the observation that would support or weaken it, rather than relying on the latest price movement.
I would keep confirmed information separate from the creator's interpretation and future targets. The institutional-buyer theory, for example, belongs apart from completed purchases and reported trading activity.
Keep measurement labels beside revenue figures. Daily averages, weekly spikes and cumulative totals cover different periods; protocol income and holder income concern different recipients. Comparing unlike figures can manufacture an apparent increase that the numbers do not establish.
Before connecting a wallet to a crypto service, check the site address independently and examine the permissions requested. Protect your private key, the secret that authorizes control of your assets.
For the next claim, ask what would make it wrong and whether it needs several favorable developments in sequence. I would put Ethena's unresolved activation condition directly on the watchlist: USDe supply averaging $7.5 billion over 14 days.
