By Will Strauch

Inspiration

The inspiration for my project was to try and learn more about how to evaluate soccer players because of how difficult it currently is as well as improve how scouting soccer players can be done. Using data to evaluate soccer players is difficult because of how little can be recorded for every player so finding different statistics to evaluate players besides goals and assists is important.

What it does

Helps determine important statistics as well as predicts effectiveness for players currently playing in Europe. Uses these predictions to determine the best young talents and best countries at producing talents.

How I built it

I built a Random Forest regressor model to try and predict a player's match rating. Using this I was able to determine the important statistics. The predictions allowed me to figure out who the best predicted players were, and which countries produced the best talent.

Challenges I ran into

Some challenges that came up during this project was learning how hard it is to wrangle multiple datasets together. Making sure data was matching up when merging data frames was quite challenging but also rewarding. It was also difficult to learn new libraries, such as Plotly, because of the new syntax that needed be learned.

Accomplishments that I'm proud of

I am proud of this project because the model that was created produced results that were consistent with what I expected which was very cool. I was also proud of being able to make interesting visualizations using a new library because that was quite challenging.

What I learned

I learned a lot about how important the pre-processing of data is, I spent a lot more time cleaning and merging data frames together than I expected so I definitely learned how beneficial this skill is to have.

What's next for Analyzing Soccer Statistics

The next step for this project would be to bring more leagues into the model. The results were very euro-centric because it was built up using the top five European soccer leagues so adding in more leagues around the world would give a better snapshot of the current reality of soccer talent around the world.

Built With

Share this project:

Updates