This project focuses on analyzing NFL tackling data to gain insights into player and team performances. The analysis includes exploratory data analysis (EDA), player and team-level analysis, tackling type and success factors, and the development of new metrics to evaluate tackling efficiency.
The analysis utilizes several datasets:
tackles.csvgames.csvplays.csvplayers.csvtracking_week_{i}.csvfor weeks 1 to 9
Data is mounted from Google Drive and processed using Python libraries including Pandas, NumPy, Matplotlib, and Seaborn.
- Identification of unique games, plays, and players.
- Analysis of the distribution of tackles, assists, and missed tackles.
- Exploration of player heights, weights, and their correlation with tackling efficiency.
- Calculation of tackling efficiency and ranking of players based on various metrics.
- Investigation into how player attributes like height and weight impact tackling efficiency.
- Examination of tackling proficiency among different teams.
- Analysis of total tackles and missed tackles for each team.
- Study of different types of tackles (e.g., open field tackles, gang tackles).
- Classification of play types based on play descriptions.
- Creation of a class
nflBigDatafor advanced data analysis. - Implementation of methods to calculate downfield speed, downfield acceleration, distance to ball carrier, and yards forward.
- Development of a custom metric to evaluate the importance of tackles (yards saved).
- Numerous plots and visualizations to support the analysis, including histograms, scatter plots, and heatmaps.
- Comparison of traditional tackling efficiency with the new yards saved metric.
- Visualization of player performance metrics for better understanding.