Base and Arbitrum hold two-thirds of Layer 2 value: what the concentration tells us
L2BEAT data shows Ethereum's rollup market shrinking in dollar terms while two chains tighten their grip. We look at why distribution beats technology.
LAYER 2ETHEREUMANALYSIS
Sofia Marchetti
10/28/20263 min read
Ethereum's Layer 2 market looks very different from the crowded landscape promised two years ago. According to L2BEAT, the projects it tracks secured about $39.1 billion at the end of October. Base accounted for roughly $15.1 billion of that and Arbitrum One for about $9.9 billion. Together the two chains hold close to 64% of all value secured by Ethereum scaling networks, and no rival comes close to either of them.
The headline number is also falling. In early October, L2BEAT put total value secured across tracked projects near $43.6 billion, with Base at about $16.2 billion and Arbitrum One at $11.4 billion. Part of the decline reflects a weaker ether price, since most assets on rollups are priced in ETH or ETH-based tokens. But price alone does not explain why some networks lost a fraction of their deposits while others emptied out.
A steep drop after the top two
The gap between the leaders and the rest of the table is the most striking feature of the data. OP Mainnet, in third place, secures about $1.7 billion, roughly a ninth of Base. Mantle sits near $1.3 billion and the trading-focused Lighter just above $1 billion. After that, values fall into the hundreds of millions: Starknet, Ink, Linea and ZKsync Era each hold between roughly $250 million and $370 million. Scroll, Unichain and Taiko Alethia are lower still.
This shape matters because value secured is the closest thing the sector has to a measure of trust. Users park assets where liquidity is deep, withdrawals are routine and the applications they want already live. Each dollar that moves to Base or Arbitrum makes the next dollar more likely to follow, while smaller chains struggle to reach the depth that keeps trading costs low and lending markets solvent.
Distribution, not technology, picked the winners
The two leaders are both optimistic rollups, and both were among the networks L2BEAT moved to its Stage 1 tier, which requires a working permissionless proof system. Yet zero-knowledge rollups, often described as the more advanced design, sit far down the ranking. The deciding factor has been access to users. Base is wired into Coinbase's exchange and wallet products, so a retail customer can move funds onto it without ever touching a bridge. Arbitrum built a deep DeFi base early and now licenses its technology to others, including Robinhood, whose own chain went live on mainnet in July.
Starknet shows what happens when a network lacks that kind of pipeline. After its community backed a non-binding proposal to research becoming a separate, quantum-resistant Layer 1, value secured on the chain fell almost 40% in a week, according to L2BEAT figures reported by CryptoTicker. Starknet now holds just under 1% of the tracked total. Whatever the long-term merits of the plan, the immediate reaction showed how quickly capital leaves a network when its future relationship with Ethereum becomes uncertain.
The bottom of the table is thinning out. Blast, which held more than $2 billion at its June 2024 peak, announced on October 2 that it would shut down after its locked value fell to around $32 million, based on DefiLlama data. Abstract, the chain behind the Pudgy Penguins ecosystem, plans to close on December 15. Both had attracted deposits with incentives that faded once the rewards ended.
What to watch into 2027
Three signals will show whether concentration deepens or eases. The first is whether corporate chains keep their users once incentives stop. Robinhood Chain has posted multi-million-dollar fee days, according to Bitquery estimates, but its value secured is still small next to Base. The second is how mid-sized networks such as OP Mainnet, Mantle and Linea respond. Mergers of applications, like the recent Aerodrome and Velodrome combination, hint that liquidity fragmented across many chains is becoming harder to defend.
The third signal is Ethereum itself. Developers have postponed testing of the Glamsterdam upgrade, and proposals to make the base layer faster and cheaper could pull some activity back from rollups altogether. If that happens, the weaker Layer 2 networks will feel it first.
For now the data points to a simple conclusion. The Layer 2 market has become a two-chain market with a long tail. Builders choosing where to launch, and users deciding where to hold assets, are increasingly choosing liquidity and distribution over technical promises. The networks below the top two need a clear reason to exist beyond being another place to deploy the same applications.