Pokerbots Competition MIT 2025

Welcome to the Pokerbots Competition MIT 2025 repository. This project is designed for the MIT Pokerbots competition, where participants develop autonomous poker-playing bots. The competition involves creating a bot that can play a variant of Texas Hold'em with unique rules and strategies.

Project Structure

The project is organized into several key components:

• Engine: The core game engine that simulates poker rounds and manages game states.
• Player: The bot logic that decides actions based on the current game state.
• Skeleton: Contains the base classes and interfaces for actions, states, and bot behavior.
• Logs: Game logs for analyzing bot performance and decision-making.

Key Files and Directories

Engine

The engine.py file contains the main game engine logic. It handles the game flow, including dealing cards, managing player actions, and determining the outcome of each round.

startLine: 1
endLine: 33

Player

The player.py file contains the logic for the poker bot. It includes methods for preflop and postflop decision-making, hand evaluation, and board analysis.

startLine: 1
endLine: 16

Actions and States

The skeleton directory includes definitions for actions (e.g., Fold, Call, Raise) and game states (e.g., GameState, RoundState). These are used to structure the bot's decision-making process.

startLine: 1
endLine: 10

Game Logs

The gamelog.txt file records detailed logs of each game round, including player actions, card distributions, and outcomes. This is useful for debugging and improving bot strategies.

startLine: 1
endLine: 60

Game Rules

The competition uses a variant of Texas Hold'em with the following rules:

• Each player is assigned a bounty rank, which affects the payout if their bounty card appears during the round.
• The starting stack for each round is 400 chips, with blinds set at 1 (small blind) and 2 (big blind).
• Players can perform actions such as Fold, Call, Raise, and Check based on the current game state.

For more detailed rules, refer to the GAME_RULES section in the player_chatbot/player.py file.

startLine: 35
endLine: 55

Getting Started

To get started with the project, follow these steps:

1. Clone the Repository: Clone the repository to your local machine using git clone.
2. Install Dependencies: Ensure you have Python and necessary libraries installed. Use pip install -r requirements.txt if a requirements file is provided.
3. Run the Bot: Execute the main script to start the bot and simulate games. Use the provided scripts or create your own to test different strategies.

Contributing

Contributions to the project are welcome. If you have ideas for improving the bot's strategy or optimizing the engine, feel free to submit a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for more details.

References

For more information on the algorithms and strategies used in this project, refer to the following papers:

• [1] Martin Zinkevich et al., "Regret minimization in games with incomplete information," 2008.
• [2] Marc Lanctot et al., "Monte Carlo sampling for regret minimization in extensive games," 2009.

startLine: 73
endLine: 77

We hope you enjoy participating in the Pokerbots Competition MIT 2025 and look forward to seeing your innovative strategies!