Skip to content

RussellParco/MLBSimulator

Repository files navigation

MLBSimulator

Description

This python project is built to simulate MLB baseball games. We first create two customised teams composed of 9 hitters and a pitcher to compete against one another. Using the MLB Stats API we can choose to use up-to-date current season stats or previous seasons stats. This allows us to build teams made from current stars like MIke Trout to basbeall legends like Babe Ruth. We then find the average statistics for MLB players, using the average stats and the individual stats of a hitter, pitcher pair we are able to find an effective estimate of the probabilities of all the outcomes for the matchup between the hitter and pitcher. From this we can use simple base running rules and how to score runs to simulate the rest of the game.

Background

This project takes inspiration from the vast sabermetrics community, r\Sabermetrics, and their work to better explain the sport that so many love, including myself. Further research was done to find an effective way to simulate a hitter, pitcher matchup. Some of the following papers provided insight on both simulating the matchups and base runners.
https://tht.fangraphs.com/10-lessons-i-learned-from-creating-a-baseball-simulator/
https://cdn.shopify.com/s/files/1/1633/3369/files/VersionFinale.pdf

How To Run

Download the files and install required libraries

pip install json
pip install requests 
pip install random

Finally to run,

python main.py

About

MLB Game Simulator

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages