← All articles

#MEV

7 articles

The Invisible Leg: $132 Billion in Non-Atomic Arbitrage and the Eleven Searchers Behind It

The legendary first comprehensive study of CEX-DEX arbitrage on Ethereum reveals that non-atomic arbitrage accounts for more than a quarter of top DEX volume, is dominated by eleven searchers, and creates a centralizing force in block building that subsidizes itself with off-chain profits.

Why Fixed Block Times Are the Best Deal LPs Already Have

This mathematical framework proves that Ethereum's constant 12-second slots uniquely minimize LP losses to arbitrage - and quantifies exactly how sub-slots, faster blocks, and block time distributions affect the cost of providing liquidity.

The Price of 50 Milliseconds: What Latency Is Actually Worth to Validators

Optimum's new analysis puts concrete ETH numbers on the value of network latency reduction - 50ms of faster propagation translates to +0.66% APR and 150 ETH/week in additional MEV bid value across the network.

Phantom Millions: Why Sandwich Attack Statistics Are Broken

Three sandwich attacks reported as $4.7M in profit were actually worth close to zero. A forensic look at how low-liquidity token pricing distorts MEV dashboards - and why it matters for protocol design, regulation, and the dark forest narrative.

JIT Liquidity: Two Economic Regimes, One Block

A study of 442,000 JIT bundles on Uniswap v3 reveals that public and private execution channels create fundamentally different economics - with measurable consequences for builders, searchers, and passive LPs.

Ethereum's Blockspace Market Under the Microscope

A builder, a relay, a pricing researcher, and an EF researcher co-authored a paper on what's broken in Ethereum's blockspace pipeline. Here's what they found - and what they're not saying.

Relay Block Merging: How Losing Builders Still Get Paid

A new mechanism from the Titan team lets relays append transactions from losing builders onto winning blocks - creating more valuable blocks, solving relay economics, and improving censorship resistance. All without changing the protocol.