Skip to content

Latest commit

 

History

History
56 lines (40 loc) · 1.52 KB

README.md

File metadata and controls

56 lines (40 loc) · 1.52 KB
Raw

NHL Prediction Model 3.0

A python machine learning model to predict the outcomes of NHL games.

Table of Contents
  1. Overview
  2. Built With
  3. Dependencies
  4. Acknowledgements

Overview

A machine learning model designed to predict the outcome of NHL matches. The model has been trained on data from 2010 to 2022 predicting the expected goals for and against. Using a combination of those predictions and ELO ratings, I then calculate offensive and defensive strengths for each team, with which I can predict the outcome of upcoming games, and determine whether my odds have an edge over Vegas.

(back to top)

Built With

(back to top)

Dependencies

  • numpy
  • pandas
  • seaborn
  • matplotlib.pyplot
  • sklearn.linear_model.Ridge
  • sklearn.metrics.mean_squared_error
  • sklearn.metrics
  • re

(back to top)

Acknowledgements

Data sourced from naturalstattrick.com

(backop)