Inspiration
What inspired us was the interest in understanding how sports betting industry works nowadays. We wanted to explore how professionals and analysts predict outcomes for any type of tournaments and how much of that can be driven by data modeling.
What it does
It takes data from all the previous World Cups, such as teams, goals scored, and goals received, and then builds a function using Poisson distribution to predict how many points on average a team would get facing another team.
How we built it
We built it using Python libraries such as pandas, scipy and beautiful soup.
Challenges we ran into
Some challenges we faced, for example, were extracting all the data and finding a pattern to iterate over all the matches.
Accomplishments that we're proud of
Made a predictive model using Monte Carlo simulation.
What we learned
We learned about fetching data from the web, cleaning it, organizing it on data frames, and applying a distribution to build a final product.
What's next for Data Science Prediction Model World Cup Edition
Wait for the official fixtures to be announced to predict the winner of 2026.
Built With
- beautiful-soup
- colab
- pandas
- pycharm
- python
- scypi
Log in or sign up for Devpost to join the conversation.