IPD Simulator

Python simulator for the Iterated Prisoner's Dilemma, inspired by classic IPD tournaments.

This project collects a large set of strategies and runs them against each other in tournament-style matchups.

Contents

• 30+ strategy functions in IPD_tournament/IPDStrategies.py
• match simulation logic in IPD_tournament/IPD_Simulations.py
• a small entry script in IPD_tournament/IPDtournament_main.py

---

• IPD_tournament/IPDtournament_main.py Current entry point. By default it runs:

ClassicTourney(Player_list, 200, 4)

• IPD_tournament/IPD_Simulations.py Core tournament logic:

  • Fight(player1, player2, playto) runs a head-to-head match
  • RoundRobin(playerlist, playto) computes average performance across the full pool
  • ClassicTourney(playerlist, playto, permutations) runs a bracket-style tournament

• IPD_tournament/IPDStrategies.py Strategy definitions and registration lists.

Scoring

Moves are represented as:

• 1 = cooperate
• 0 = defect

Payoffs: Standard

• (1,1) -> (3,3)
• (1,0) -> (0,5)
• (0,1) -> (5,0)
• (0,0) -> (2,2)

Strategy Format

Each strategy is a Python function that receives:

strategy(player, score, modifiers=[None])

Where:

• player is player 1 or 2
• score is the prior match history as a list of (p1_move, p2_move) tuples
• the return value is usually a single move or a short list of upcoming moves

Strategies are registered in:

• Player_dict
• Player_list

So adding a new strategy usually means:

1. Define the function in IPDStrategies.py
2. Add it to Player_dict
3. Run the main script again