Sports Analytics: Big Data in the Big Time
summary
Covers concrete methods and results: uses OPS outperforms complex regressions for hitter value, applies regression and FanGraphs data, uses walks‑strikeouts‑HR models and BABIP to show luck in ERA, and runs Monte Carlo on Barry Bonds to predict extreme-lineup runs. Introduces win‑probability‑added (Mike Trout +5.3 wins, Cabrera +4.8, Verlander +4.0), points‑per‑play for NFL (Spiller > Peterson), 70% fourth‑and‑2 conversion rule, and adjusted plus‑minus plus lineup optimization that produced a +46 per‑48‑min playoff impact.
AI-written summary · about
chapters (9)
- 0:00Sports analytics intro
- 10:41Baseball analytics
- 21:53Player and contract eval
- 31:08Football analytics
- 39:11NFL draft strategy
- 40:13Basketball box score limits
- 46:13Basketball metrics and lineups
- 53:57Mavericks case study
- 56:04Q&A
related talks
33mUsing Data Science to Better Evaluate American Football Players
41mThe evolution of basketball with data science
41mFootball's data analytics in python
01h 20mBayesian Inference for Fun and Profit
22mNeural Networks for Longitudinal Data Analysis
19mFootball Analytics with the Soccermatics Package
08mData in Sports
41m"R" You Ready for Some Football? Hacking Fantasy Sports with Open Source Software