Bunny Hops and Blockchain Stops: Cross-Chain MEV Detection With N-Hops
Davide Mancino, Hasret Ozan Sevim, Oriol Saguillo Gonzalez
TL;DR
The paper tackles cross-chain MEV by targeting sequence-dependent multihop arbitrage (SDA), analyzing an immense dataset of over $2.49$ billion swaps and $34.84$ million bridge transactions across $12$ blockchains and $45$ bridges to detect SDA opportunities. It formalizes an $n$-hop SDA path as $2n-1$ sequential transactions (swaps and bridges) with strict constraints and a validation function to ensure legitimate cross-chain sequences, and computes profit as $\Pi(A)=v_{out}(t_{2n-1})-v_{in}(t_1)$. The empirical results reveal only 10 multihop SDA instances (8 three-hop, 2 four-hop), supporting a power-law decline in feasibility with hop count and highlighting that higher-hop strategies face increasing latency and risk; profitability remains variable with 60% of detected paths profitable. The work emphasizes the challenges of multihop cross-chain arbitrage and proposes extending the framework to sequence-independent arbitrage and more detailed latency/fee modeling to better understand the evolving landscape as cross-chain interoperability improves.
Abstract
This student paper introduces a novel methodology for the detection and analysis of multihop cross-chain arbitrage opportunities, wherein multihop denotes arbitrage sequences involving more than two transactional steps across distinct blockchain networks, executed using sequence-dependent strategies. Utilizing a comprehensive dataset comprising over 2.4 billion transactions recorded between September 2023 and August 2024 (encompassing 12 blockchain platforms and 45 cross-chain bridges) we design and implement an algorithm capable of identifying, sequence-dependent arbitrage paths spanning multiple ecosystems. Our empirical analysis demonstrates that such arbitrage opportunities are exceedingly infrequent, underscoring the inherent challenges associated with multihop execution in cross-chain environments.