GitHub - steliosspap/footballpredictions: This project was developed as part of the Wharton High School Data Science Competition(https://wsb.wharton.upenn.edu/wharton-data-competition/), organized by the Wharton Sports Analytics and Business Initiative. The competition provided teams with simulated soccer league data to develop predictive models for playoff outcomes.

steliosspap / footballpredictions Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

This project was developed as part of the Wharton High School Data Science Competition(https://wsb.wharton.upenn.edu/wharton-data-competition/), organized by the Wharton Sports Analytics and Business Initiative. The competition provided teams with simulated soccer league data to develop predictive models for playoff outcomes.

MIT license

0 stars 0 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 16 Commits
data		data
tests		tests
.gitignore		.gitignore
LICENSE		LICENSE
README		README
football_predictions.py		football_predictions.py
requirements.txt		requirements.txt

Repository files navigation

Football Predictions

Overview

- Developed a logistic regression model using `scikit-learn`
- Applied standard preprocessing and evaluation techniques
- Organized in a single Python script for clarity and accessibility
- Evaluated using accuracy as the primary performance metric

File Structure

- `football_predictions.py` – main script containing data preprocessing, model training, and prediction logic

Example Output

```
Accuracy: 72.4%
Predictions on test data:
[1, 0, 1, 1, 0, ...]
```

Getting Started

To replicate this project:

1. Clone the repository:
   ```bash
   git clone https://github.com/steliosspap/footballpredictions.git
   cd footballpredictions
   ```

2. Install the required Python packages:
   ```bash
   pip install -r requirements.txt
   ```

3. Execute the script:
   ```bash
   python football_predictions.py
   ```

Learning Outcomes

This was one of my first experiences (16) developing a machine learning model from scratch. Key takeaways include:

- Fundamentals of data preparation and feature selection
- Implementation of logistic regression for binary classification
- Basic performance evaluation using test accuracy
- Limitations and potential areas for future improvement

License

This repository is licensed under the MIT License. Please note that this license applies only to the code in this repository and does not extend to the dataset provided by the Wharton High School Data Science Competition.

Data Access

The dataset used in this project was provided by the Wharton High School Data Science Competition. Due to usage restrictions, the full dataset is not included in this repository. Interested individuals can learn more about the competition and data access by visiting https://wsb.wharton.upenn.edu/wharton-data-competition/

Acknowledgments

This project was completed as a learning exercise and was taken part at the Wharton High School Data Science competition. Feedback and suggestions for improvement are welcome, mostly for discussion and personal interest.