Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Political Leanings in Web3 Betting: Decoding the Interplay of Political and Profitable Motives

Hongzhou Chen, Xiaolin Duan, Abdulmotaleb El Saddik, Wei Cai

TL;DR

This work introduces the Political Betting Leaning Score ($PBLS$), a quantitative measure that decodes political leanings from on-chain betting activity in Polymarket, leveraging rational-choice theory and revealed preferences. The methodology combines extensive data collection, validation against polls, large-scale feature engineering (825 features with 533 PBLS-significant signals), and a predictive model (XGBoost with RFE) to estimate $PBLS$ for over 15,000 addresses, alongside a 2022 Senate case study to examine the interaction of political and profitable motives. Key findings show high predictive accuracy of political markets ($DLS\approx0.74$; $R^2\approx0.50$ for accuracy drivers), distinct behavior of political bettors versus general users, and a robust link between higher $PBLS$ and positive profits, with nuanced temporal dynamics around election events. The study demonstrates the value of blockchain data for decoding complex online behavior, informs prediction-market design, and opens avenues for cross-market validation and integration with traditional data sources to enhance forecasting and mechanistic understanding of decision-making in decentralized ecosystems.

Abstract

Harnessing the transparent blockchain user behavior data, we construct the Political Betting Leaning Score (PBLS) to measure political leanings based on betting within Web3 prediction markets. Focusing on Polymarket and starting from the 2024 U.S. Presidential Election, we synthesize behaviors over 15,000 addresses across 4,500 events and 8,500 markets, capturing the intensity and direction of their political leanings by the PBLS. We validate the PBLS through internal consistency checks and external comparisons. We uncover relationships between our PBLS and betting behaviors through over 800 features capturing various behavioral aspects. A case study of the 2022 U.S. Senate election further demonstrates the ability of our measurement while decoding the dynamic interaction between political and profitable motives. Our findings contribute to understanding decision-making in decentralized markets, enhancing the analysis of behaviors within Web3 prediction environments. The insights of this study reveal the potential of blockchain in enabling innovative, multidisciplinary studies and could inform the development of more effective online prediction markets, improve the accuracy of forecast, and help the design and optimization of platform mechanisms. The data and code for the paper are accessible at the following link: https://github.com/anonymous.

Political Leanings in Web3 Betting: Decoding the Interplay of Political and Profitable Motives

TL;DR

This work introduces the Political Betting Leaning Score (), a quantitative measure that decodes political leanings from on-chain betting activity in Polymarket, leveraging rational-choice theory and revealed preferences. The methodology combines extensive data collection, validation against polls, large-scale feature engineering (825 features with 533 PBLS-significant signals), and a predictive model (XGBoost with RFE) to estimate for over 15,000 addresses, alongside a 2022 Senate case study to examine the interaction of political and profitable motives. Key findings show high predictive accuracy of political markets (; for accuracy drivers), distinct behavior of political bettors versus general users, and a robust link between higher and positive profits, with nuanced temporal dynamics around election events. The study demonstrates the value of blockchain data for decoding complex online behavior, informs prediction-market design, and opens avenues for cross-market validation and integration with traditional data sources to enhance forecasting and mechanistic understanding of decision-making in decentralized ecosystems.

Abstract

Harnessing the transparent blockchain user behavior data, we construct the Political Betting Leaning Score (PBLS) to measure political leanings based on betting within Web3 prediction markets. Focusing on Polymarket and starting from the 2024 U.S. Presidential Election, we synthesize behaviors over 15,000 addresses across 4,500 events and 8,500 markets, capturing the intensity and direction of their political leanings by the PBLS. We validate the PBLS through internal consistency checks and external comparisons. We uncover relationships between our PBLS and betting behaviors through over 800 features capturing various behavioral aspects. A case study of the 2022 U.S. Senate election further demonstrates the ability of our measurement while decoding the dynamic interaction between political and profitable motives. Our findings contribute to understanding decision-making in decentralized markets, enhancing the analysis of behaviors within Web3 prediction environments. The insights of this study reveal the potential of blockchain in enabling innovative, multidisciplinary studies and could inform the development of more effective online prediction markets, improve the accuracy of forecast, and help the design and optimization of platform mechanisms. The data and code for the paper are accessible at the following link: https://github.com/anonymous.
Paper Structure (28 sections, 38 equations, 5 figures, 11 tables)

This paper contains 28 sections, 38 equations, 5 figures, 11 tables.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Measure Political Leanings Online
  4. Polymarket, Web3 Prediction Market
  5. Decoding Online Betting Behaviors
  6. Methodology
  7. Data Collection
  8. Accuracy of Political Betting Markets
  9. Scoring Political Betting Leaning
  10. Validation
  11. Feature Engineering and Correlation
  12. Predicting PBLS for All User Addresses
  13. Method for Case Study
  14. Visualizing PBLS and Profit/Loss
  15. Panel Data Analysis of Price, PBLS, and Holding Size
  16. ...and 13 more sections

Figures (5)

  • Figure 1: Decoding political leanings through on-chain betting behaviors. The blockchain enables transparent and immutable records of anonymous behaviors across markets. By aggregating, we could construct a quantitative measure of political leaning.
  • Figure 2: (a) Distribution of events. Politics-related events rank second in numbers but first in volume. The skewness of count and volume existed across events. (b) Distribution of politics-related markets. Elections markets attract the highest volume and unique user addresses. (c) The accuracy of political prediction markets improved and stabilized after an initial volatility, positively impacted by participation and negative by volume.
  • Figure 3: (a) Yearly user address creation on Polymarket: general vs. political market participants. (b) Realized Profit/Loss distribution: general vs. political betting participants. (c) Number of markets traded: general vs. political betting participants. (d) Volume traded: general vs. political betting participants.
  • Figure 4: (a) Distribution of Political Betting Leaning Scores (PBLS) for user addresses participated in the 2024 U.S. presidential election event on Polymarket, showing a majority with moderate partisan preferences. (b) Comparison of actual and predicted PBLS distributions reveals the effectiveness of the machine learning approach in capturing political leanings. (c) Correlation coefficients between features and PBLS increase from user to event to market levels, highlighting the importance of granular behavioral analysis in decoding political preferences.
  • Figure 5: (a) The distribution of users’ PBLS alongside their realized profit and loss (P/L) in the event. (b) Price dynamics for the Democratic and Republican outcomes in the 2022 U.S. Senate election on Polymarket. The upper panel shows the price trends during the total event, indicating a long-term advantage for the Republican party before the election night (November 8, 2022). The lower panel presents the situation after the election night, revealing a shift towards the Democratic party. The vertical dashed line in both panels marks the election night.