{"paper_id":"2508.04003","title":"The Marginal Effects of Ethereum Network MEV Transaction Re-Ordering","authors":["Bruce Mizrach","Nathaniel Yoshida"],"paper_date":"2025-08-06","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2508.04003v1","page_url":"https://thequant.space/flowcharts/the-marginal-effects-of-ethereum-network-mev-transaction-re/","scores":{"empirical_rigor":7,"math_complexity":4,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper estimates the marginal effects of MEV transaction re‑ordering on block placement and quantifies associated costs to participants.","quote":"We estimate the marginal effects of MEV transaction re‑ordering and find that in our baseline model, investors would pay nearly $14 million to ensure that they remained in the first quartile of the block.","summary":"Two MEV builders now produce nearly 80% of Ethereum blocks. Block builders have the ability to reorder transactions on the blockchain in a way that can be harmful to participants. We estimate they would pay in the aggregate nearly $14 million per month to ensure that they remained in the first…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The error term follows a normal distribution, allowing use of an ordered probit model.","type":"model","stated":"explicit","quote":"where F is the cumulative normal distribution function."},{"assumption":"Transaction characteristics are captured by dummy variables for mempool position, DEX/MEV involvement, and maximum fee per gas.","type":"model","stated":"explicit","quote":"X contains three dummy variables corresponding to a transaction’s position in the mempool, dummy variables denoting a transaction going to and from a DEX, to and from an MEV agent, and a measure of the maximum fee in gas units a participant will commit to the transaction."},{"assumption":"Gas fees are measured in ETH and converted to USD using CoinGecko spot prices.","type":"data","stated":"explicit","quote":"To put these gas fees into context, we first convert them to USD using Coin Gecko prices."},{"assumption":"Back‑run transactions incur higher gas fees than other legs of the sandwich.","type":"other","stated":"explicit","quote":"One important thing we have identified is that the back‑run transaction, in which the attacker flattens their position and records a profit, pays substantially higher gas fees than other transactions."},{"assumption":"Sandwich cost and number of sandwich transactions drive the skewness of marginal effects.","type":"statistical","stated":"explicit","quote":"The cost of the sandwich is highly statistically significant and positively correlated with marginal effects, indicating that costlier sandwich transactions are subject to more reordering."}]},"datasets":{"status":"available","items":[{"name":"Ethereum on‑chain transaction data (October 2024)","provider":"Google Big Query / Etherscan","kind":"crypto-onchain","access":"public","period":"October 2024","universe":"All Ethereum transactions","quote":"The sandwich positions are from EigenPhi and the gas fees are from Google Big Query. The sample is the month of October 2024.","known_id":"etherscan"},{"name":"MEV builder addresses","provider":"MEVBoost.pics","kind":"crypto-onchain","access":"public","period":null,"universe":null,"quote":"tify MEV builders are from MEVBoost.pics."},{"name":"DEX and CEX address list","provider":"Brian Lect’s Github","kind":"crypto-onchain","access":"public","period":null,"universe":null,"quote":"Addresses for centralized exchanges (CEX) and decentralized exchanges (DEX) are from Brian Lect’s Github,26 and all active addresses were confirmed manually from Etherscan."},{"name":"ETH price data","provider":"CoinGecko","kind":"market-prices","access":"public","period":"October 2024","universe":"ETH/USD","quote":"The average ETH price for the month of October 2024 was $2,520.05","known_id":"coingecko"}],"known":[{"id":"etherscan","name":"Etherscan","provider":"Etherscan","kind":"crypto-onchain","access":"public","quote":"We use average daily gas costs of 22.45 Gwei which we obtain from Etherscan and the closing Ethereum price of $2,597.34 from CoinGecko.","mentions":2,"used_in":1},{"id":"coingecko","name":"CoinGecko","provider":"CoinGecko","kind":"market-prices","access":"public","quote":"We use average daily gas costs of 22.45 Gwei which we obtain from Etherscan and the closing Ethereum price of $2,597.34 from CoinGecko.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"n/a","evidence_card":"yes","evidence_card_quoted":true},"conflict_kind":"applicability","card_detail":"Back‑run gas fees average 0.0248 ETH (~$62.70) per transaction; total gas fees for sandwich attacks estimated at $7.89 million in October 2024.","card_quote":"The average ETH price for the month of October 2024 was $2,520.05, so the back run transaction fees averaged $62.70. For the 125,829 sandwich attacks in the month, we","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"Block builders collectively pay about $14 million per month to stay in the first quartile of block placement.","kind":"statistical","quote":"We estimate they would pay in the aggregate nearly $14 million per month to ensure that they remained in the first quartile of the block."},{"finding":"Sandwich attacks occur on average more than once per block, with 124,946 attacks in October 2024.","kind":"performance","quote":"Sandwich attacks, in which a transaction is front‑run, are frequent, averaging more than one per block."},{"finding":"Front‑run transactions are about 75.86% more likely to appear in the first block quartile; back‑run about 72.46%.","kind":"statistical","quote":"The marginal effects for back run transactions are similarly large, averaging 72.46% for October 2024."},{"finding":"Each additional sandwich transaction in a block raises marginal effects by ~0.04075 (≈4.1% higher chance of first‑quartile placement).","kind":"statistical","quote":"An additional sandwich transaction within a block is associated with an increase in the marginal effects of 0.04075. This means that an additional sandwich transaction within a block corresponds to a transaction being approximately 4.1% more likely to appear in the first quartile of block transactions."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":19,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:30:30Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}