SoK: Preventing Transaction Reordering Manipulations in Decentralized Finance

Lioba Heimbach, Roger Wattenhofer

This Systematization of Knowledge (SoK) paper provides a comprehensive framework for understanding and evaluating the various strategies proposed to combat transaction reordering manipulations—often referred to as Blockchain Extractable Value (BEV) or Miner Extractable Value (MEV)—within the Decentralized Finance (DeFi) ecosystem.

TL;DR

The paper systematizes the "invisible tax" of transaction reordering in DeFi and evaluates current mitigation schemes across seven critical dimensions. Its core finding is a "no free lunch" reality: every existing solution—from trusted third parties to algorithmic committees—forces a trade-off between security, decentralization, or network performance.

What Problem Does This Paper Address?

In traditional finance, "front-running" (trading ahead of a known client order) is illegal. In DeFi, it is a structural byproduct of how blockchains work. Because miners/validators have total control over the order of transactions in a block, they (or bots watching the "mempool") can insert their own trades before or after a user's transaction to siphon off profit.

This leads to three main types of attacks:

  1. Fatal Front-running: Causing a victim's transaction to fail entirely while the attacker profits.
  2. Front-running: Worsening the price for a victim's trade.
  3. Back-running: Profiting from the price impact or arbitrage opportunity created by a victim's trade.

The challenge is that DeFi lacks a central authority to enforce "fairness." The paper addresses the gap in understanding how different technical fixes for these attacks impact the broader blockchain ecosystem.

Methods

The authors establish a multi-dimensional evaluation framework consisting of seven measures:

  • Decentralization: Does the scheme introduce central points of failure?
  • Security: How easily can an attacker bypass the protection?
  • Scope: Does it stop all BEV types or just specific ones (e.g., sandwich attacks)?
  • Jostling: Does it prevent "gas wars" where users overbid to get priority?
  • Goodput: Does it maintain a high number of genuine transactions per block?
  • Delay: Does it increase the time from submission to execution?
  • Cost: Does it increase transaction fees for the user?

The paper then categorizes and maps state-of-the-art mitigation strategies—including Optimized Trade Execution, Professional Market Makers, Trusted Third Parties, eUTXO models, Algorithmic Committees, and Commit & Reveal schemes—against these seven metrics using "spider web" (radar) charts.

Datasets

As a Systematization of Knowledge (SoK) paper, this work does not introduce a new training dataset. Instead, it synthesizes and analyzes data from:

  • Existing DeFi Protocols: Uniswap, Curve, Balancer, Sushiswap, and Aave.
  • Mitigation Services: Flashbots, Eden, OpenMEV, and CowSwap.
  • Academic Proposals: A2MM, Tesseract, Fairy, and various Byzantine consensus protocols.
  • Market Data: Historical BEV extraction figures (e.g., exceeding $100M monthly on DEXes).

Key Insights and Ideas

  • The Popularity of Centralization: Currently, "Trusted Third Party" ordering (like Flashbots) is the most widely adopted solution. While it excels in performance (low delay, high goodput), it is in "stark contradiction" to the trustless nature of blockchain.
  • The eUTXO Bottleneck: The paper highlights why the eUTXO model (used by Cardano) is naturally resistant to front-running: it only allows one interaction per pool per block. However, this creates a massive "goodput" problem, making it impractical for high-volume DeFi.
  • Commit & Reveal Trade-offs: While "On-chain Commit & Reveal" offers high security by hiding transaction contents, it introduces unacceptable delays (waiting multiple blocks for the reveal) and doubles transaction costs.
  • The "Byzantine" Committee Risk: Many decentralized solutions rely on a committee to order transactions. The authors point out that if these committees are rational (profit-seeking), they can still collude to extract BEV, meaning decentralization doesn't automatically guarantee security.

Key Results

The paper’s primary result is a comparative matrix (Figure 11) showing that no single approach excels in all areas. * Optimized Trade Execution (e.g., splitting trades) is the most "blockchain-friendly" but has the narrowest Scope. * Trusted Third Parties provide the best user experience (low Delay, low Jostling) but fail on Decentralization. * Off-chain Commit & Reveal is identified as having the "greatest promise" for a broad scope of protection, though it currently struggles with committee-based security concerns.

Limitations and Future Work

The authors acknowledge that their evaluation assumes a "rational" actor model, which may not capture every edge case in adversarial environments.

Future Work: The paper concludes that a "suitable approach" has yet to emerge. Future research should focus on leveraging the high Scope of commit-reveal schemes while finding ways to reduce the Delay and Cost penalties. The authors emphasize that for DeFi to truly compete with centralized finance, it must solve the reordering problem without sacrificing the performance users expect.

Summary updated

Abstract

User transactions on Ethereum's peer-to-peer network are at risk of being attacked. The smart contracts building decentralized finance (DeFi) have introduced a new transaction ordering dependency to the Ethereum blockchain. As a result, attackers can profit from front- and back-running transactions. Multiple approaches to mitigate transaction reordering manipulations have surfaced recently. However, the success of individual approaches in mitigating such attacks and their impact on the entire blockchain remains largely unstudied. In this systematization of knowledge (SoK), we categorize and analyze state-of-the-art transaction reordering manipulation mitigation schemes. Instead of restricting our analysis to a scheme's success at preventing transaction reordering attacks, we evaluate its full impact on the blockchain. Therefore, we are able to provide a complete picture of the strengths and weaknesses of current mitigation schemes. We find that currently no scheme fully meets all the demands of the blockchain ecosystem. In fact, all approaches demonstrate unsatisfactory performance in at least one area relevant to the blockchain ecosystem.
BibTeX
@inproceedings{heimbach2022sok,
  author = {Lioba Heimbach and Roger Wattenhofer},
  title = {{SoK: Preventing Transaction Reordering Manipulations in Decentralized Finance}},
  booktitle = {AFT},
  year = {2022},
  doi = {10.1145/3558535.3559784},
}

Loading related content...

Figure