Open vs. Sealed: Auction Format Choice for Maximal Extractable Value

Battle Count: 59

The paper is highly relevant to crypto/DeFi quantitative trading and MEV extraction strategies. It provides actionable guidance for MEV searchers and block builders on auction format selection, quantifies the revenue impact of mechanism choice ($10-18M foregone), and characterizes competition intensity across MEV types. However, it focuses on mechanism design rather than direct trading strategy development. The findings inform how searchers should bid in different auction formats and how builders should structure their auctions, which directly affects MEV extraction profitability and competitive dynamics in crypto markets.

The theoretical framework requires solving ODEs for equilibrium bid functions under affiliation, computing conditional distributions via Monte Carlo integration over the common factor posterior, and evaluating order statistics of log-normal distributions. The simulation grid (11 values of n × 10 values of rho × 10^6 draws) is computationally intensive but parallelizable. The empirical analysis (fitting log-normal, computing Gini, bribe percentages) is straightforward. The main complexity lies in the numerical ODE solver for affiliated FPSB/Dutch equilibria and the cubic spline interpolation of bid functions.

The paper uses the publicly available libmev dataset, specifies all simulation parameters (grid of n and rho, 10^6 draws per cell), provides MLE estimates for log-normal parameters, and details the Gaussian common factor model. However, no code repository is explicitly linked, and the ODE-based equilibrium computation for affiliated FPSB/Dutch auctions requires numerical implementation that is described but not provided as code. The simulation methodology is well-documented enough for replication.