Master's Thesis Tobias Krebs
Abstract
Impermanent loss poses a significant risk to liquidity providers in automated market makers (AMMs) especially within blockchain-based prediction markets. Unlike traditional central limit order books where buyers and sellers are matched directly, AMMs rely on liquidity pools and algorithmic pricing. This unique structure, while offering continuous liquidity, exposes liquidity providers to impermanent loss, a temporary loss of funds due to price divergence between their deposited assets and holding them outside the AMM. This Master's thesis investigates strategies to mitigate this financial risk, focusing on prediction markets such as Polymarket. The research systematically addresses three key areas: quantifying impermanent loss, identifying its main contributors, and evaluating dynamic fees as a counteractive measure. The study employs a comprehensive methodology, beginning with an in-depth analysis of historical transaction data from Polymarket to accurately quantify impermanent loss experienced by liquidity providers. Subsequently, it delves into identifying the primary factors that contribute to substantial impermanent loss within these decentralized prediction markets. Building upon these findings, a proof of concept will be developed. This proof of concept will implement dynamic fees, calculated based on the identified contributing market features, and demonstrate their efficacy in reducing impermanent loss when compared to conventional AMM designs. By analyzing the mechanisms of impermanent loss and proposing a novel, adaptive fee structure, this thesis aims to enhance the resilience and sustainability of liquidity provision in AMMs.
Research Questions
- Quantification of impermanent loss for liquidity providers in automated market makers for prediction markets
- What are the main contributors to impermanent loss in blockchain-based prediction markets?
- To what extend can dynamic fees based on the main contributing factors (determined in RQ2) reduce impermanent loss in prediction markets?
| Attribute | Value |
|---|---|
| Title (de) | Kompensation des Impermanent Loss für Automated Market Maker in blockchainbasierten Prognosemärkten |
| Title (en) | Counteracting Impermanent Loss Occurring in Automated Market Makers in Blockchain-Based Prediction Markets |
| Project | |
| Type | Master's Thesis |
| Status | started |
| Student | Tobias Krebs |
| Advisor | Jonas Gebele |
| Supervisor | Prof. Dr. Florian Matthes |
| Start Date | 28.04.2026 (date of LMU registration 16.06.2026) |
| Sebis Contributor Agreement signed on | 30.04.2026 |
| Checklist filled | Yes |
| Submission date | 27.10.2026 (according to LMU regulation) |