Basketball Data Science All studies Scouting one-pagers Code on GitHub

Opponent scouting one-pagers

One page per team for the full 2025-26 season: the daily deliverable of a coaching-analytics seat, generated end-to-end from public data.

Each report reads the way a staff consumes it: the keys up top, then the evidence beneath them. The shot profile prices where a team shoots from using a model whose season-to-season transfer was validated by backtest; personnel is split into shot diet (the part that repeats year over year) and shot making (the part that is half noise), and the making numbers are shrunk by empirical Bayes before any bullet is allowed to call someone a shot-maker, so a hot month cannot drive a game plan. Lineups come from the per-team files, player impact from the opponent-adjusted RAPM model, and every claim in the keys traces to a number in a table below it.

The reports cannot silently drift: the pipeline regenerates the Phoenix report in memory on every family check run and fails unless the committed copy matches byte-for-byte.

Atlanta Hawks

one-pager โ†’

Brooklyn Nets

one-pager โ†’

Boston Celtics

one-pager โ†’

Charlotte Hornets

one-pager โ†’

Chicago Bulls

one-pager โ†’

Cleveland Cavaliers

one-pager โ†’

Dallas Mavericks

one-pager โ†’

Denver Nuggets

one-pager โ†’

Detroit Pistons

one-pager โ†’

Golden State Warriors

one-pager โ†’

Houston Rockets

one-pager โ†’

Indiana Pacers

one-pager โ†’

Los Angeles Lakers

one-pager โ†’

Memphis Grizzlies

one-pager โ†’

Milwaukee Bucks

one-pager โ†’

Minnesota Timberwolves

one-pager โ†’

New Orleans Pelicans

one-pager โ†’

New York Knicks

one-pager โ†’

Oklahoma City Thunder

one-pager โ†’

Orlando Magic

one-pager โ†’

Philadelphia 76ers

one-pager โ†’

Phoenix Suns

one-pager โ†’

Portland Trail Blazers

one-pager โ†’

Sacramento Kings

one-pager โ†’

San Antonio Spurs

one-pager โ†’

Toronto Raptors

one-pager โ†’

Washington Wizards

one-pager โ†’

How they're built โ†’