# SBD Total Predictor

A linear regression model that predicts a powerlifter's total (Squat + Bench + Deadlift) 
based on competition attributes such as bodyweight, age, sex, and equipment category. 
Built on a real-world U.S. powerlifting competition dataset (via Kaggle) to explore how 
well simple linear models can capture strength-performance relationships.

## What it does
- Cleans and preprocesses raw competition data (handles missing values, outliers, categorical encoding)
- Trains a scikit-learn linear regression model to predict SBD total from lifter attributes
- Evaluates performance using R² and residual analysis
- (Optional, if applicable) Extends to per-lift prediction (squat, bench, deadlift individually)

## Tech stack
- Python, pandas, NumPy, scikit-learn
- Jupyter notebooks for EDA and experimentation
- Git/GitHub for version control

## Why
Built as a hands-on project to apply core ML fundamentals — data cleaning, feature 
engineering, train/test splits, model evaluation — to a domain I'm personally invested 
in (powerlifting and strength training).