Paper: arXiv 2404.09090

Authors: Erhan Bayraktar, Asaf Cohen, April Nellis

Abstract

We investigate the behavior of liquidity providers (LPs) by modeling a decentralized cryptocurrency exchange (DEX) based on Uniswap v3. LPs with heterogeneous characteristics choose optimal liquidity positions subject to uncertainty regarding the size of exogenous incoming transactions and the prices of assets in the wider market. They engage in a game among themselves, and the resulting liquidity distribution determines the exchange rate dynamics and potential arbitrage opportunities of the pool. We calibrate the distribution of LP characteristics based on Uniswap data and the equilibrium strategy resulting from this mean-field game produces pool exchange rate dynamics and liquidity evolution consistent with observed pool behavior. We subsequently introduce Maximal Extractable Value (MEV) bots who perform Just-In-Time (JIT) liquidity attacks, and develop a Stackelberg game between LPs and bots. This addition results in more accurate simulated pool exchange rate dynamics and stronger predictive power regarding the evolution of the pool liquidity distribution.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematical frameworks like mean-field games and Stackelberg equilibrium, which are computationally dense. However, it also demonstrates strong empirical rigor by calibrating models with real Uniswap v3 data and validating simulated dynamics against observed pool behavior.

Research Flowchart

  flowchart TD
  A["Research Goal: Model & Calibrate LP Behavior<br>in Uniswap v3 & MEV JIT Attacks"] --> B["Key Methodology: Mean-Field Game &<br>Stackelberg Game Theory"]
  B --> C["Input: Historical Uniswap Data<br>Calibrates LP Characteristics Distribution"]
  C --> D["Computational Process: Simulate<br>Liquidity & Exchange Rate Dynamics"]
  D --> E{"Outcome 1: Base Model Validation<br>Consistent with Observed Pool Behavior"}
  D --> F["Input: MEV Bot Parameters"]
  F --> G["Computational Process: Stackelberg Game<br>LPs vs. JIT Liquidity Attacks"]
  G --> H{"Outcome 2: Enhanced Model<br>Accurate Dynamics & Stronger Predictions"}

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a mean‑field game model of heterogeneous liquidity providers in Uniswap v3 and extends it with a Stackelberg game that includes MEV bots performing just‑in‑time liquidity attacks.

Source sentence
We investigate the behavior of liquidity providers (LPs) by modeling a decentralized cryptocurrency exchange (DEX) based on Uniswap v3

Assumes 6

  • Each LP has a fixed amount of capital (valued in token B) that it can contribute to the pool. model
    Source sentence
    Furthermore, we will assume that each LP has a fixed amount of capital (as valued in terms of token B) that it is able to contribute to the pool.
  • LPs value their capital using the market exchange rate m* which is assumed impervious to manipulation. model
    Source sentence
    We assume that LPs value their total capital using the market exchange rate, which we denote as m*, because it is impervious to manipulation from within the pool.
  • LPs have a belief parameter δ about future market exchange rate movements and model m* dynamics with a geometric Brownian motion. statistical
    Source sentence
    we assume that, based on δ, the LP believes that the market exchange rate will evolve according to m*_t = m*_{t-1} e(α+δ/1000−σ^2)Δt+σΔW_t.
  • The distribution of LP types Θ is separable, complete, atomless, and each type is drawn independently. model
    Source sentence
    Assumption 3.1. The space of characteristics/types, supp(Θ), is a separable and complete metric space. We also impose that Θ is atomless.
Show 2 more
  • The time horizon for the LP optimization problem is fixed at T = 2700 blocks. model
    Source sentence
    We will also hold the time horizon of the problem fixed for all players at T = 2700.
  • Bots consider a fixed gas fee G in their profit calculation, while LPs do not. execution
    Source sentence
    Note that we choose to have Bots to consider gas fees (G, which we set as fixed for now) while LPs do not.

Data

  • Etherscan transaction data · Etherscan crypto-onchain public
    Source sentence
    All transaction data is obtained from the Etherscan API...
  • Coinmetrics market exchange rates · Coinmetrics market-prices public
    Source sentence
    exchange rate data was obtained from the Coinmetrics API...
  • GitHub code and datasets · GitHub synthetic public
    Source sentence
    Our code and the necessary datasets can be found on Github.

Recognised sources: Etherscan CoinGecko

Costs

Unclear - sources disagree The platform classifier and the paper-text extraction give different answers, so no value is asserted.

Platform classifier: Not applicable

Paper text extraction: Partial - Bots incur a fixed gas fee G (average $10‑$30 per swap) while LPs are assumed not to pay gas fees.

Source sentence (the authors' words)
the average fee for a Uniswap currency swap or liquidity adjustment hovers between $10 -$30

Finds 4

  • The calibrated mean‑field equilibrium reproduces observed pool liquidity and exchange‑rate dynamics. predictive-accuracy
    Source sentence
    the equilibrium strategy resulting from this mean-field game produces pool exchange rate dynamics and liquidity evolution consistent with observed pool behavior.
  • When LPs anticipate bot attacks, their collective profits increase by approximately 3%. performance
    Source sentence
    when LPs choose their liquidity positions while anticipating Bot attacks, they are able to increase their collective profits by approximately 3%.
  • Including bots improves the MAPE between simulated pool and market exchange rates from 0.1357% to 0.0732%, close to the observed 0.063%. predictive-accuracy
    Source sentence
    the MAPE between the simulated pool exchange rate and the market exchange rate improved from the previous value of 0.1357% to 0.0732%, which is very close to the observed MAPE
  • Risk‑averse LPs concentrating liquidity in anticipation of bots reduce overall MAPE by 46.1% across simulations. performance
    Source sentence
    This led to a 21.54% increase in the accuracy of the predicted liquidity distribution and a 46.1% decrease in the overall MAPE between the market and pool exchange rates throughout our simulations.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 17 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 2 papers in the archive (2 by independent authors)

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (48)

  1. Infinite Dimensional Analysis: A Hitchhiker’s Guide (2007)
  2. Improved Price Oracles: Constant Function Market Makers (2020) — link
  3. An analysis of Uniswap markets (2021) — link
  4. Liquidity Provision by Automated Market Makers (2020) — link
  5. Integrals of set-valued functions (1965) — link
  6. Approximate Purification of Mixed Strategies (1983)
  7. Nonparametric Adaptive Robust Control Under Model Uncertainty (2022) — link
  8. Uniswap Liquidity Provision: An Online Learning Approach (2023) — link
and 40 more
  1. High-Roller Impact: A Large Generalized Game Model of Parimutuel Wagering (2017) — link
  2. Mean Field Game Approach to Bitcoin Mining (2020) — link
  3. Advances in Blockchain and Crypto Economics (2023) — link
  4. The Adoption of Blockchain-based Decentralized Exchanges: A Market Microstructure Analysis of the Automated Market Maker (2021) — link
  5. Liquidity Provision on Blockchain-based Decentralized Exchanges (2023) — link
  6. The Paradox Of Just-in-Time Liquidity in Decentralized Exchanges: More Providers Can Sometimes Mean Less Liquidity — → 2311.18164
  7. Decentralised Finance and Automated Market Making: Predictable Loss and Optimal Liquidity Provision — → 2309.08431
  8. A Machine Learning Approach to Adaptive Robust Utility Maximization and Hedging (2021) — link
  9. Bitcoin Mining and Electricity Consumption (2022) — link
  10. A Machine Learning Method for S tackelberg Mean Field Games (2023) — link
  11. Strategic Liquidity Provision in U niswap v3 (2023) — link
  12. Credible Decentralized Exchange Design via Verifiable Sequencing Rules (2023) — link
  13. Liquidity effects of trading frequency (2017) — link
  14. Endogenous Formation of Limit Order Books: Dynamics Between Trades (2018) — link
  15. Games with Incomplete Information Played by `` B ayesian" Players, I-III. Part I. T he Basic Model (1967)
  16. Games with Incomplete Information Played by `` B ayesian" Players, I-III. Part II. B ayesian Equilibrium Points (1968)
  17. Games with Incomplete Information Played by `` B ayesian" Players, I-III. Part III. T he Basic Probability Distribution of the Game (1968)
  18. Eliminating Sandwich Attacks with the Help of Game Theory (2022) — link
  19. Risks and Returns of U niswap V3 Liquidity Providers (2022)
  20. The Potential of Self-Regulation for Front-Running Prevention on Decentralized Exchanges (2023)
  21. Behavior of Liquidity Providers in Decentralized Exchanges (2021) — link
  22. Core and Equilibria of a Large Economy (1974)
  23. Large population stochastic dynamic games: closed-loop McKean-Vlasov systems and the Nash certainty equivalence principle (2006)
  24. Monopoly without a Monopolist: An Economic Analysis of the B itcoin Payment System (2021) — link
  25. Optimal Trading in Automatic Market Makers with Deep Learning — → 2304.02180
  26. Disentangling Decentralized Finance (DeFi) Compositions (2023) — link
  27. Stochastic Differential Mean Field Game Theory (2015) — link
  28. Mean field games (2007) — link
  29. On a theorem of S chmeidler (1984) — link
  30. Non-Cooperative Games (1951)
  31. Flashbots—Frontrunning the MEV Crisis (2020) — link
  32. The Conceptual Flaws of Decentralized Automated Market Making (2023) — link
  33. Computing pure Bayesian-Nash equilibria in games with finite actions and continuous types (2013) — link
  34. Private Information and Pure-Strategy Equilibria (1982)
  35. Verallgemeinerung eines in der Statistik ben "o tigten Satzes der Ma \ss theorie (1963) — link
  36. An Iterative Method of Solving a Game (1951)
  37. A Mean Field Games Model for Cryptocurrency Mining (2023) — link
  38. Uniswap v3 Core (2021) — link
  39. Demystifying Just-in-Time (JIT) Liquidity Attacks on Uniswap V3 (2023) — link
  40. High-Frequency Trading on Decentralized On-Chain Exchanges (2021) — link

3 of 48 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.