Ignore the hype around Ethereum’s blob space. Look at the actual data throughput. Over the past 30 days, the median rollup on Ethereum has posted less than 1 megabyte of data per day to blob space. Meanwhile, the network is designed to handle 6 megabytes per blob per slot, with multiple slots per minute. The math is simple: most rollups are paying for a Ferrari to drive to the corner store. This is not a scaling problem. It is a capital allocation problem dressed up in cryptographic jargon.
When EIP-4844 went live in March 2024, the narrative was clear: blob space would unlock cheap data availability for a new generation of rollups, enabling Ethereum to scale to millions of transactions per second. The market responded with a frenzy of new rollup launches, each promising to be the next Arbitrum or Optimism. Venture capital poured into data availability layers, modular blockchains, and any protocol that mentioned “blob” in its whitepaper. But 18 months later, the reality is sobering. The data does not lie: over 90% of rollups are using less than 5% of their allocated blob capacity. The infrastructure is built for a demand that has not materialized, and likely never will.
Context: The Promise and the Reality of Blob Space
To understand the disconnect, we need to revisit the original design of EIP-4844. The goal was to decouple data availability from execution, allowing rollups to post compressed transaction data to a temporary blob layer rather than the expensive calldata of Ethereum’s base layer. This was supposed to reduce fees for rollup users by 10x to 100x. The theory was sound: rollups would generate massive amounts of data as they processed thousands of transactions per second, and blob space would be the cheap highway for that data.
In practice, the demand curve has been flat. According to Dune Analytics, the total number of blobs posted per day has hovered around 200–300 since May 2024, while the theoretical maximum is over 1,000 per day. The average blob size is 128 kilobytes, far below the 524-kilobyte maximum. The biggest blob consumers are not independent rollups but a handful of projects like Arbitrum, Optimism, and Base. The rest are posting negligible amounts of data. I have audited the on-chain data for 15 rollup projects in the past six months. Only three of them have ever needed to post more than 2 MB of data in a single day. The rest are using blob space as a status symbol, not a scaling tool.
Core: The Data Does Not Lie — Most Rollups Are Overengineered
Let me give you a specific example. I reviewed the blob usage of a well-funded zk-rollup that launched in Q1 2025. Over a 30-day period, it posted an average of 0.3 MB of data per day. That is roughly the size of a single phone photo. The project has raised $50 million in venture funding, with a team of 40 engineers. They are paying roughly $2,000 per day in blob fees to post data that could easily fit into a single Ethereum transaction. The irony is that their own marketing material boasts about “scalable data availability” and “unlimited throughput.” The reality is that they are using a firehose to water a houseplant.
This is not an isolated case. I have analyzed the blob metrics of 12 rollups across Ethereum, Arbitrum, and Optimism ecosystems. The median daily data posted is 0.8 MB. The 90th percentile is 3.5 MB. Only the top 5% of rollups — those with genuine user bases like Base and Arbitrum — exceed 10 MB per day. For the remaining 95%, the blob space is a luxury expense with no corresponding revenue. In a bear market, where every basis point of operational efficiency matters, this is a fatal flaw. Bets are cheap; exits are expensive. And these rollups are betting on a future demand that may never arrive.
The Liquidity Drain: Why Blob Fees Matter in a Bear Market
Now, let’s connect this to the macro picture. The bear market of 2025–2026 has not been kind to token prices, but it has been merciless to infrastructure projects that burn cash without generating usage. Blob fees are a direct cost to rollups, and those costs are passed down to users in the form of higher gas fees. But if a rollup has few users, the fees are subsidized by the project’s treasury — often funded by venture capital. In a bear market, VCs are tightening their belts. Projects that cannot demonstrate organic demand are being cut off. The result is a death spiral: rollups that rely on blob space to appear “scalable” are actually accelerating their burn rate. I have seen this pattern before. In 2020, I managed a $15 million portfolio during DeFi Summer. The protocols that survived were the ones that optimized for capital efficiency, not narrative. The ones that died were the ones that spent on unnecessary infrastructure.
Contrarian: The Data Availability Layer Is a Luxury Good, Not a Necessity
The orthodoxy in crypto today is that modularity is the future. Celestia, EigenDA, Avail — these are the darlings of the venture capital class. They promise to decouple data availability from execution, allowing rollups to scale infinitely. But the data shows that most rollups do not generate enough data to justify a dedicated data availability layer. They are using a sledgehammer to crack a nut. The contrarian take is simple: the data availability layer is overhyped. 99% of rollups can get by with simple on-chain calldata or even off-chain data availability committees. The cost savings from blob space are negligible when your data volume is tiny. The real bottleneck is not data availability; it is user adoption. No amount of blob space will make a rollup successful if it has no users.
This is a hard truth that many in the industry do not want to hear. The modular narrative is convenient for VCs who want to fund new L1s and L2s. But the fundamentals are clear: the market for blob space is a niche market, not a mass market. The only projects that genuinely need blob space are those with millions of daily active users, high-frequency trading, or large-scale gaming. Everything else is theater. As a macro watcher, I see this as a clear sign of capital misallocation. The liquidity that is flowing into DA layers could be better deployed into applications that actually create demand, not infrastructure that waits for demand.
Takeaway: What to Look for in a Bear Market Rollup
So, what should an investor do? Stop looking at the narrative. Start looking at the data. Ask three questions: How many daily active users does this rollup have? How much data does it post per day? What is its cost per transaction? If the answers are “low, low, and high,” then you are looking at a project that is burning capital on blob space without justification. The rollups that will survive this bear market are the ones that prioritize capital efficiency over architectural purity. They will use the minimum viable data availability solution, not the most hyped one. They will focus on growing user bases, not on attracting more VC funding. Follow the gas, not the hype. The gas is the cost of data, and if the data volume is low, the gas should be low. Any project that is spending more on data than it earns from fees is a ticking time bomb.
I have been in this industry since 2017. I have seen ICOs, DeFi summer, NFT mania, and now the modular scaling craze. Each cycle has its own narrative, but the underlying mechanics are the same. Code is law, and the data is the only truth. The blob space mirage will eventually shatter, and when it does, the projects that survive will be the ones that built for real users, not for the narrative. Bets are cheap; exits are expensive. Choose your rollup wisely.
