"Now that transactions have moved to L2, does Ethereum mainnet still matter?" That's the wrong question. What matters on mainnet isn't transaction volume, but its ability to collect rent.
According to L2BEAT, all rollups combined processed about 303 million user operations on 11/13, averaging more than 3,500 per second. Ethereum mainnet handled about 25 per second on the same day. There’s no real debate about the activity gap of more than 100x—scaling is supposed to move transactions off-chain.
The question is how much rent L2s pay L1 after moving those transactions off-chain.
Blobscan data shows that since the Dencun upgrade in March 2024, all L2s combined have burned about 4,748 ETH in blob base fees—worth roughly $14.9 million at the prices at the time. That’s over two and a half years. Base alone, the largest contributor, accounts for about 1,694 ETH, or roughly $5.29 million.
Now look at DefiLlama: Base generated about $2.61 million in revenue over the past 30 days. In other words, about two months of Base’s earnings roughly match all the blob fees it paid to Ethereum over two and a half years. Over the same period, Ethereum itself burned about $6.2 million in fees in the past 30 days, and just over $80,000 in the past 24 hours.
That’s the balance-sheet reality behind the narrative that "L2s are Ethereum’s scaling solution": users, fees, and sequencer profits all stay on L2, while L1 gets the prestige of providing security and a little spare change. Blobspace supply is set to a target, and as long as demand doesn’t hit capacity, prices stay at rock bottom. The more L2s thrive, the less likely they are to generate significant ETH burns.
So ETH/BTC’s steady decline over the past few years isn’t entirely down to sentiment. Before touting "record L2 activity" as bullish for ETH, do the rent math. Ethereum either needs blob demand to actually hit capacity, or it needs to bring execution back to mainnet. Otherwise, the more traffic grows, the more it looks like Ethereum is working for other chains.
Data: L2BEAT activity API (11/13 UTC), Blobscan stats (updated 11/14 07:25 UTC), DefiLlama fees API (retrieved 11/14). USD figures are converted by Blobscan using historical prices.
#ETH #L2
According to L2BEAT, all rollups combined processed about 303 million user operations on 11/13, averaging more than 3,500 per second. Ethereum mainnet handled about 25 per second on the same day. There’s no real debate about the activity gap of more than 100x—scaling is supposed to move transactions off-chain.
The question is how much rent L2s pay L1 after moving those transactions off-chain.
Blobscan data shows that since the Dencun upgrade in March 2024, all L2s combined have burned about 4,748 ETH in blob base fees—worth roughly $14.9 million at the prices at the time. That’s over two and a half years. Base alone, the largest contributor, accounts for about 1,694 ETH, or roughly $5.29 million.
Now look at DefiLlama: Base generated about $2.61 million in revenue over the past 30 days. In other words, about two months of Base’s earnings roughly match all the blob fees it paid to Ethereum over two and a half years. Over the same period, Ethereum itself burned about $6.2 million in fees in the past 30 days, and just over $80,000 in the past 24 hours.
That’s the balance-sheet reality behind the narrative that "L2s are Ethereum’s scaling solution": users, fees, and sequencer profits all stay on L2, while L1 gets the prestige of providing security and a little spare change. Blobspace supply is set to a target, and as long as demand doesn’t hit capacity, prices stay at rock bottom. The more L2s thrive, the less likely they are to generate significant ETH burns.
So ETH/BTC’s steady decline over the past few years isn’t entirely down to sentiment. Before touting "record L2 activity" as bullish for ETH, do the rent math. Ethereum either needs blob demand to actually hit capacity, or it needs to bring execution back to mainnet. Otherwise, the more traffic grows, the more it looks like Ethereum is working for other chains.
Data: L2BEAT activity API (11/13 UTC), Blobscan stats (updated 11/14 07:25 UTC), DefiLlama fees API (retrieved 11/14). USD figures are converted by Blobscan using historical prices.
#ETH #L2