Bankless co-founder David Hoffman sold all of his ETH in late May, and the rotation has since outperformed Ether by a wide margin over the following three and a half months.
According to TechFlowPost, Hoffman wrote on May 21 that the mood on crypto Twitter had changed and that he had sold the last of his ETH. By early June, he had laid out the full path of the portfolio shift on X: after exiting ETH, he split the capital into two buckets, one deployed immediately across four assets and the other held back for dollar-cost averaging before being committed fully to LIT.
How the capital was redeployed
The article says around 50% of the funds were allocated right after the ETH sale to four names: VVV, NEAR, ZEC, and HYPE. Hoffman explicitly said in a post that he bought NEAR at roughly $1.4.
The remaining 50% was initially set aside for DCA. In his words, it was money he planned to use to slowly buy something that had not yet moved. That capital ultimately went entirely into LIT, the token tied to Lighter, an on-chain perpetuals exchange built on zkRollup.
TechFlowPost said Hoffman’s rationale for buying LIT was unusually direct. Exchanges, in his view, remain the best business model in crypto; Lighter’s buyback pace was about twice that of HYPE; zk circuits let users verify without permission whether the exchange is following its own rules; and the product offered lower latency, a better fee structure, and support for a wider set of assets, including pre-IPO markets.
When asked how to choose between LIT and HYPE, Hoffman said LIT was both HYPE’s beta and its alpha. By late June, he had posted that he regretted not buying more LIT.
Five assets, four themes
The source article presents the trade as a clear investment framework. ZEC represented a bet on the privacy theme. HYPE and LIT were tied to structural growth in on-chain derivatives. NEAR was a bet on cross-chain infrastructure and the AI agent theme. VVV represented decentralized AI inference.
In that framing, the five positions covered four distinct narrative tracks, and none of them relied on Ethereum L1 valuation.
Performance snapshot
TechFlowPost compared the prices around Hoffman’s disclosed entries with prices in early September. The ETH exit happened in late May, the first wave of entries took place from late May to early June, and the LIT DCA continued into mid-June. The article uses approximate early-June prices as the cost basis and compares them with the latest prices in early September.
- ETH: sold at about $2,100 and later traded around $2,450, for a gain of roughly 17%.
- ZEC: entered at about $540 and later rose above $1,200, up more than 120%. The article says it briefly touched $1,200 on Sept. 6. It links the move to the Aug. 25 launch of Grayscale’s spot ZEC ETF, ZCSH, on NYSE Arca. Two weeks after launch, AUM had grown from $300 million to $460 million, alongside $46 million in short liquidations.
- HYPE: entered at about $56 and later traded near $87, up about 55%. It hit an all-time high of $89.54 on Sept. 6. The article says Hyperliquid’s cumulative token burn has exceeded $4 billion and that average daily protocol revenue was holding around $2.26 million.
- LIT: accumulated through DCA at roughly $1.5 to $2 on average and later traded around $4.7, implying gains of about 135% to 210%. It reached a record $4.95 on Sept. 5. The piece describes LIT as Hoffman’s largest single position, accounting for 50% of total capital and delivering the biggest absolute profit in the basket.
- NEAR: bought at about $1.4 and later traded around $2.37, up about 69%.
- VVV: entered around $16 to $18 and later traded near $17, roughly flat. The article notes that VVV hit an all-time high of $21.32 on June 3, making it the weakest performer among the five holdings.
Using a rough portfolio estimate, the article assumes that 50% of the capital was split equally across VVV, NEAR, ZEC, and HYPE, with each taking 12.5%, while the other 50% went to LIT. On a midpoint basis, it estimates the portfolio returned around 90% to 120%. ETH gained about 17% over the same period, meaning the basket outperformed ETH by at least 70 percentage points.
Why the trade worked, according to the article
TechFlowPost argues that the main takeaway is not just the magnitude of the gains in each token. It is the way Hoffman’s selection logic was validated by the market over the next three and a half months.
For ZEC, the article says the surge had a visible path. Grayscale’s ETF filing had already been moving through the SEC process since November last year, and the Aug. 25 launch turned a long-developing institutional variable into realized price action. When Hoffman bought in May, ZEC had already climbed from the $30 range at the start of the year to above $500. In the article’s telling, he was betting that the certainty of the catalyst outweighed the short-term price risk.
For LIT, the source says the logic is even more revealing. While much of the market was talking about HYPE as the on-chain equivalent of a major derivatives venue, Hoffman picked a smaller, earlier-stage rival with what the article describes as a more aggressive product structure. The framework was to find a higher-elasticity beta in the same sector, supported by faster buybacks, a transparency premium from zk verification, and the possibility of user migration driven by lower fees. The article adds that, looking back, LIT has risen about 500% from its bottom to current levels.
NEAR’s return, at about 69%, ranked fourth within the basket. Even so, the article points to a possible hidden feedback loop. It says NEAR Intents has acted as a tollbooth during the ZEC rally, with ZEC-related trading pairs accounting for nearly 40% of NEAR Intents’ total volume. As ZEC rises, trading flow through the Zashi wallet and NEAR Intents increases, which lifts fee revenue for NEAR Intents and strengthens buyback support for NEAR.
VVV was the only position that did not clearly deliver. TechFlowPost says the decentralized AI inference narrative around Venice AI spiked briefly in early June and then lost momentum, and Hoffman did not publicly signal additional buying.
From L1 valuation to application-layer revenue
The article’s broader argument is that Hoffman’s result points to a shift in where crypto markets are assigning value. In its view, attention is moving from L1 valuation toward application-layer revenue.
The ETH he sold was an L1 asset, with valuation tied to network effects, the developer ecosystem, and the gas burn mechanism. Among the five assets he bought, the valuation anchors were different: HYPE and LIT were tied to verifiable protocol revenue and buybacks; ZEC was tied to institutional demand through the ETF and to observable on-chain privacy usage data, including shielded supply share; NEAR was tied to transaction volume and fees in cross-chain settlement; and VVV was tied to actual usage of AI inference services.
In the article’s framing, what links these assets is that their valuation depends on activity that can be independently checked on-chain, rather than on promises about the future.
TechFlowPost adds that Hoffman’s ETH sale was not presented as a verdict against Ethereum itself. The article says he still believes Ethereum will win. What changed, in this reading, was the market environment: in a liquidity-scarce market, belief alone does not produce returns, while verifiable cash flow and measurable demand do.