Ethereum L2 Throughput Doubled in Q1 — Base and Arbitrum Lead

Rollup activity now accounts for 71% of Ethereum economic security demand. Base added 6.2M weekly active addresses; Arbitrum still owns DeFi TVL.

Ethereum's execution layer is no longer the place most transactions actually happen. According to a fresh snapshot of L2Beat and on-chain analytics this week, rollup throughput for the first quarter of 2026 was double the same period a year ago, and the L1 itself processed only the residual share that hasn't yet been ferried to a cheaper venue.

Base, Coinbase's in-house OP Stack rollup, was the structural winner. It added 6.2 million weekly active addresses in Q1 alone — more than every other rollup combined — and is now the default consumer chain for Coinbase wallet users. Arbitrum, the older incumbent, lost some address share but retained its grip on DeFi TVL, which sits above $14B.

The Bigger Trend

Settlement on Ethereum is increasingly an institutional product, not a retail one. The economic security demand — the number of dollars of value secured per dollar of fee paid — has shifted decisively to L2s, which buy data availability from the L1 in bulk and resell it as throughput.

Rollups are the user-facing product. L1 is the data centre.

The implication for builders: the question is no longer whether to deploy on a rollup, but which one. The question for ETH holders: whether the value capture from being "the settlement layer" is enough.

What to Watch

• Base — Coinbase's integration play. Watch for the next round of consumer apps to ship there first.
• Arbitrum — Still the institutional DeFi venue. Stylus and Orbit chains keep the ecosystem widening.
• OP Stack chains — Optimism, Base, and the Superchain partners are increasingly indistinguishable from a user's perspective.
• Linea, Scroll, ZkSync Era — The ZK cohort gaining on the OP stack in narrow benchmarks.

The Q2 question is whether Base's growth chokes Coinbase's own retail spreads. If yes, the company has its first internal conflict to manage. If no, every other exchange will copy the playbook.

The Engineering Story

The most interesting part of this update is what it takes to ship at the scale these teams now operate at. Coordinated forks, multi-client compatibility, and a release cadence that doesn't break downstream tooling have become as important as the raw protocol improvements themselves.

Validator operator economics keep getting tighter. Margins compressed, hardware demands up, and the long tail of independent operators is consolidating.

Why It Matters for Builders

• Composability. Every upgrade either preserves or breaks the contract surface apps were written against.
• Bandwidth. Throughput only matters if peripheral systems keep up.
• Security model. Modular stacks introduce new trust assumptions.
• Cost. Real per-transaction cost, fully loaded, is an order-of-magnitude conversation now.

Adoption Curve

The application layer responds slowly to L1 improvements. Most apps don't care about a 2x throughput bump because they aren't throughput-bound. The ones that do — orderbook DEXs, on-chain games, micropayment rails — are the leading indicators.

The Competitive Frame

Competition between L1s is no longer a TPS race. It's a developer-experience race.

What Comes Next

Watch the next two upgrade cycles. Most chains have one major release left in the current architecture.

Infrastructure is plumbing. Plumbing matters most when it stops working.