L2 Transaction Volume, 2025
A watercolor field journal of Layer-2 blockchain data
Look at this trend -- rollup throughput has been climbing steadily since the Dencun upgrade slashed blob fees. The numbers speak for themselves: aggregate L2 transactions per second jumped from 45 to over 120 in a single quarter. That's not incremental growth, that's a phase change.
TPS across major rollups, monthly avg.
Quick thought: blob fees dropped 94% post-Dencun. That's not a correction -- that's a new era.
Here's what matters -- most L2s still run a single sequencer. The trust assumptions haven't changed as much as the throughput numbers. We're building faster trains on the same single track. Three projects are seriously working on shared sequencing, but production-ready deployment is still quarters away.
Status check: centralized sequencers remain the norm in 2025.
Observation: TVL ratio of L2-to-L1 crossed 0.35 this month. First time ever.
Bridge exploits remain the biggest risk vector in the L2 ecosystem. Over $2.1B has been lost to bridge hacks since 2021. But look at the trajectory: canonical bridges with fraud proofs and validity proofs are replacing third-party bridges. The technology is maturing, even if the scars remain visible on the chain.
Bridge exploit losses declining as canonical bridges mature.
The original tribal divide is blurring. Optimistic rollups are adding ZK components for faster finality. ZK rollups are exploring optimistic verification for cost reduction. The endgame isn't one or the other -- it's a hybrid architecture that takes the best of both worlds.
The rollup wars are less about technology and more about ecosystem lock-in now.
Side note: average L2 gas fee is now $0.001. That's basically free. The UX challenge has shifted from cost to fragmentation.
Full-spectrum data paintings from the Layer-2 seas.
Layer 2 isn't the future of Ethereum scaling anymore. It's the present. The data is clear, the trajectory is set, and the watercolors don't lie.
-- End of field notes, 2025