Home

PredPal.com
daily predictions
PredPal/Results
all past predictions and actual results (json)
View Download
blog.predpal.com
our methodologies and reasearch findings

Introduction
- Overview
  An overview of how machine learning and data science techniques can be used to solve it.
- EDA & Data Visualization in Python
  Perform exploratory data analysis on historical soccer dataset
- Home Bias Betting
  Predict game results without any data science knowledge
- Flat Bet
  Betting strategy without any data science knowledge
Classification Task
- Evaluation Metrics
  Assessing the performance of prediction models
- Logistic Regression
  Build a game prediction model using Logistic Regression
- Decision Tree and Random Forest
  Build game prediction models using Decision Tree and Random Forest
- Gradient Boosting Tree
  Build a game prediction model using GBT
More Machine Learning Algorithms
- K Near Neighbors
  Build a game prediction model using KNN
- Poisson Regression
  Build a game prediction model using Poisson Regression
- Deep Learning Models
  Build a game prediction model using deep learning
- Review
  A comparison of different ML algorithms using historical dataset
Bank Roll Management
- Percent Betting
  Introduce Percent Betting
- Kelly Criterion
  Introduce Kelly Criterion as a betting strategy
- Martingale
  Introduce Martingale as a betting strategy
- Reinforcement Learning
  Introduce Reinforcement Learning as a betting strategy
More Topics
- Feature Engineering
  Creating relevant features from raw data to improve the performance of machine learning models
- Ranking Algorithms
  Algorithmically-generated sports rating systems differ from standings by using advanced calculations and metrics