More than 13 million cryptocurrency tokens had stopped trading by the end of 2025, according to CoinGecko data cited by Real Vision founder Raoul Pal, underscoring how quickly the industry’s long tail is being culled as capital and users concentrate on a smaller set of networks.
Pal’s conclusion is stark: most token projects created in each market cycle are unlikely to survive. He compared the sector’s outcome distribution with venture capital, where roughly 80% of projects go to zero, 19% produce some value, and 1% capture the largest gains. In crypto, that division increasingly appears between networks that retain users and stablecoin liquidity during declines and those supported mainly by token launches and temporary speculation.
Pal disclosed an affiliation with the Sui Foundation during the discussion, while saying his data review preceded that relationship.
Blockchains face a platform-market test
Pal argued that general-purpose blockchains should be assessed less like conventional companies and more like platform markets selling blockspace, the transaction capacity that applications and users consume. Under that view, Ethereum, Solana and Sui operate as separate digital economies whose health depends on whether activity, capital and developers remain on-chain.
The comparison changes how token valuations should be approached. A blockchain may have a fast network, substantial venture backing or a large fully diluted valuation, but those attributes alone do not establish lasting demand for its blockspace. Projects must show that users are conducting transactions, deploying capital and building applications in ways that persist beyond a rising token price.
Pal compared the current shakeout with the dot-com era, when the Nasdaq Composite fell 78% from its peak over roughly two and a half years. Many internet companies disappeared, while a smaller group later used infrastructure built during the boom to develop durable businesses.
Crypto networks face a related pattern, though the economics of blockspace add another constraint. Competition between chains can drive the marginal cost of transactions toward very low levels. Pal argued that automated users, including AI agents, could intensify that competition by routing activity toward networks offering the lowest cost and fastest execution.
That dynamic makes fee revenue a less reliable shortcut for judging which chains will endure. It also places pressure on general-purpose networks that lack a distinct user base, application ecosystem or distribution advantage.
Four paths to survival
Pal identified four routes through which a blockchain project could establish a more durable position: operating as a standalone economy, specializing in a particular vertical, functioning as a layer-2 network, or relying on the balance sheet and distribution of a centralized platform.
He cited Ethereum, Solana and Sui as examples of networks pursuing the standalone-economy model. Each seeks to host a broad mix of applications, assets and users rather than serving one narrow product category.
Hyperliquid was presented as an example of vertical specialization, where a network’s identity is tied closely to a particular trading use case. A focused network can avoid competing on every type of application, although its prospects remain linked to the durability of its chosen market.
Layer-2 networks can draw on the ecosystem and settlement infrastructure of a larger base chain. Centralized-platform models, meanwhile, gain distribution and financial support from an associated company. Pal used BNB as an example of an asset whose role is connected to Binance’s platform balance sheet and user network rather than decentralization alone.
In his view, many newer general-purpose chains sit outside those categories. Without a clear structural advantage, they face the risk of becoming interchangeable as applications and capital move to cheaper or more liquid venues.
Low-float launches compound selling pressure
Token distribution can accelerate that process. Pal said he had tracked low-float, high-FDV token launches since the summer of 2024 and found a recurring pattern: tokens begin trading with a small circulating supply and a valuation based on a much larger eventual supply, then fall sharply as markets anticipate future unlocks.
He said such tokens often declined more than 70% soon after listing. The concern is not simply that early holders can eventually sell; it is that the market prices those future sales before the network has demonstrated sufficient organic demand to absorb them.
A low circulating supply can make an early valuation look stronger than the liquid market supporting it. As unlock schedules expand the number of tradable tokens, projects need growing activity, liquidity and user demand to offset the added supply. Without those forces, the token can weaken even if the underlying technology remains operational.
Capital velocity produced an inverse signal
Pal said he examined 12 layer-1 networks from their January 2025 peak through the following 13 months to find metrics that aligned with later token performance. His review found no forecasting value in fee-based yield or in TVL-to-FDV, a commonly cited ratio comparing the assets locked in a network’s applications with its diluted token valuation.
The only measure associated with returns in his sample was capital velocity, which tracks how rapidly capital turns over through on-chain venues. The relationship was inverse: higher turnover corresponded with weaker performance over the following 13 months.
Pal cited Solana, where turnover reached 200 times TVL in January 2025, compared with seven times TVL on Ethereum. High turnover can reflect active trading and liquid markets, but it can also signal a capital base focused on rapid circulation rather than longer-term deployment. That distinction makes raw transaction volumes and headline activity less useful without evidence that users are retaining capital on the chain.
Stablecoins and developers offer a tougher filter
Pal proposed three non-price indicators for assessing potential survivors: resilient economic density per user during drawdowns, rising stablecoin balances while a native token falls, and sustained developer inflows.
In one tracked case between December 2024 and February 2026, Pal said the token price declined 83%, TVL dropped 78%, and active addresses fell by three-quarters. Capital deployed per active user stayed close to $2,100 at both the peak and trough. That pattern suggests that while speculative users left, the remaining users continued to deploy meaningful capital.
Stablecoin balances provide another test because dollar-pegged assets do not decline alongside a network’s native token. Pal said stablecoin balances rose during the downturn across Ethereum, Solana and Sui. In the case where the token fell 83%, stablecoin balances increased by roughly 50%.
Developer activity completes the filter. Developers can move between chains, but application code, tooling, teams and user communities create higher switching costs than short-term capital. Pal described developer migration during bear markets as a more durable expression of commitment than token-price performance.
According to Pal’s review, Ethereum, Solana and Sui were the only three networks that met all three conditions during the down cycle he examined. The standard set by that framework is demanding: a network must keep economically meaningful users, retain stable-value assets and continue attracting builders even when its token is under pressure.
Want deeper context on survivors like Ethereum and Solana? Explore how Ethereum and Solana may perform in 2025.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
