Using Data Science to Better Evaluate American Football Players
summary
Shows how play‑by‑play charting plus tenth‑of‑a‑second tracking data produce concrete metrics: builds expected points added and a model that yields PFF wins above replacement used to price contracts, uses XGBoost on NextGen tracking to find that off‑the‑snap speed is the top predictor of pressure rate, demonstrates a year‑to‑year stable “bite distance” for play‑action, reproduces Tyreek Hill’s value vs cheaper replacement, and ensembles charting tracking and text scouting to improve projections.
AI-written summary · about
chapters (9)
- 0:00Intro
- 2:05Learning Objectives
- 3:33Initial Inspiration
- 11:13Where we are -Analytics/Sports
- 13:03Pass Rushing
- 17:50Wide Reciver Deep Speed
- 19:53What the Rest of this Talk is About
- 25:35Data Mining
- 27:41Conclusion
related talks
56mSports Analytics: Big Data in the Big Time
41mFootball's data analytics in python
31mBeyond a Bit Fit
42mHow Deep Learning is Revolutionizing Sports Analytics
41mThe evolution of basketball with data science
32mCutting Edge Football Analytics in Python
39mPredicting sports winners using data analytics with pandas and scikit-learn
31mFootball is complex, but your code doesn’t have to be — meet DataBallPy and a practical deep dive into pressing