Aggression in Playstyle and Football Performance

Inspiration

We wanted to try the sports analytics track, partially because we didn't really know what kind of data we wanted to analyze otherwise. We had no idea how to do data science or sport analytics, but when we saw the "aggression" column in the player attributes dataset, we thought it was funny because it stood out from other attributes, and built off that.

What it does

Visualizes how the aggression of a football team and the aggression of their opponents is related to the score outcomes of football games. Predict the ideal aggression for playing against a given team.

How we built it

Python, using numpy, matplotlib, seaborn, pandas

Coded in python and juypter, utilizing pandas and numpy for csv processing, matplotlib and seaborn for visualization. Utilizes expected aggression values and k-means clustering to predict ideal playstyle aggression to use when playing against a given opponent.

Challenges we ran into

Data cleaning was a pain. Dealing with missing values and extra columns was a pain. Calculating the average aggression of a team's players (team aggression) based on multiple aggression ratings from different dates was a pain. Data is fun!

Accomplishments that we're proud of

Getting something to work at all! We had no python experience going into this.

What we learned

How to utilize numpy, matplotlib, seaborn, and pandas. Data cleaning and processing. Don't use R-studio, it takes too much memory.

What's next for Aggression in Playstyle and Football Performance

Chilling

Built With

• juypter
• python