Analyzing Ethereum block builders: clustering strategies and profitability in the MEV- Boost ecosystemWolley, Benjamin Scott (Avtor) Škulj, Damjan (Mentor) EthereumMEV-boostblock-buildingk-means clusteringprincipal components analysisThis thesis examines the bidding behaviours and success metrics of Ethereum MEV-boost block builders during November 2023 and April 2024. These builders are responsible for the construction of over 90% blocks within the Ethereum blockchain system, generating tens of millions in profit each month. By employing Principal Components Analysis and K-means clustering on data from over 120 unique builders each month, the study identifies patterns in bidding strategies correlating with profitability and block-winning frequency. Key variables include bidding frequency, timing, and profit metrics. The research reveals a high concentration of success among a small subset of builders with Gini coefficients of 0.92 and 0.86 for the respective months. Successful builders demonstrated higher bidding intensity, greater slot participation, and more frequent bids within slots they participated in. Despite observing several significant changes in the total bids and bids per slot from November to April, profit distribution remained highly concentrated. These findings highlight the dominance of established entities in the Ethereum MEV-boost block-building ecosystem and suggest potential strategies improvements for less successful builders.B. S. Wolley20242024-10-17 08:30:04Magistrsko delo/naloga164136UDK: 004(043.2)VisID: 169280COBISS_ID: 216069123sl