nflverse

A Set of Functions to Efficiently Scrape NFL Play by Play Data

python port of nflreadr package for loading nflverse data

Efficiently download nflverse data

A set of functions to visualize National Football League analysis in 'ggplot2'

Models for Fantasy Football Expected Points

Use publicly available data to track NFL betting trends.

Weekly NFL fantasy point predictions from nflverse data. As-of feature engineering, temporally validated RF models vs a causal baseline, GMM draft tiers, artifact-backed evaluation with rolling-origin folds, autonomous weekly scoring via GitHub Actions. FastAPI on Hugging Face + Next.js on Vercel, $0/month.

Interactive NFL fourth-down decision audit using nfl4th and nflverse data. Compare coach decisions with modeled win-probability recommendations, identify conservative and aggressive misses, and explore coach decision fingerprints.

Tiered fantasy football draft board from open NFL data, tuned to one league's exact settings — an orchestrator plus five Claude Code sub-agents over a DuckDB warehouse.

Track the weekly and yearly record pace and performance of NFL quarterbacks from 1999 to present!

Chat with NFL play-by-play data in plain English — an LLM translates questions to DuckDB SQL over nflverse parquet and answers back. FastAPI + React, runs local or hosted models.

Defensive blitz-rate predictability across 32 NFL defenses, 2022-2025

Verified NFL stats and Sleeper league context, callable from ChatGPT and Claude. Free, no account, no API key.

NFL play-by-play → BigQuery: idempotent nflverse ingestion + LLM-friendly docs + verification

NFL data analytics and fantasy football TUI — polars + textual + nflverse

Predicts week-by-week 2026 stats for Dallas Cowboys skill players (QB/RB/WR/TE), factoring in opponent defensive efficiency and roster continuity for both the opposing defense and Dallas's O-line. Built with nflreadpy.

Machine learning project that predicts 2026 fantasy football production, including player floor, median, and ceiling outcomes, to identify draft value in a 10-team PPR league. 8/18-9/7

AI pre-snap run/pass prediction & defensive tendency engine for football. LightGBM on nflverse data, time-split validated (69.7% acc, calibrated).

NFL 4th-down decision audit. Win-probability model + coach scorecards with bootstrap confidence intervals.

A win-probability model for every one of the NFL's 15,545 fourth downs since 2015 - and what it says coaches are still getting wrong. Python, scikit-learn, decision analysis.