Beta Free full access during beta. All models, tools, and daily picks included.

NBA · v0 Research

DVP — Defender vs Position

How each NBA team's defense allows points, rebounds, assists, and threes to opposing guards, forwards, and centers per game. Built from the last 14 days of player logs.

Sport NBA MLB · See BvP
What this is. A research surface showing the defensive matchup advantage / disadvantage each NBA team gives to opposing guards, forwards, and centers. Toggle the stat (points, rebounds, assists, threes) and read the "vs Lg Avg" column — positive numbers mean this defense lets that position score MORE than league average (a target for over bettors), negative means LESS (target for unders). Not a betting recommendation; not picks.
DVP isn't published yet. The DVP feed for the NBA has not been written. This page fills in once the first daily build lands.
How this works

For each game in the 14-day window, every player's stat line is credited to their opponent's defense, bucketed by the player's primary position (Guard / Forward / Center). The per-game value is the team's average across all games it has played in the window.

  • Position bucketing — players are sorted into G / F / C by the first letter of their listed position. Combo positions like "G/F" are credited to the G bucket. v0.5 will split Guard / Wing / Big using position_full where it adds precision.
  • vs League Avg — this team's per-game allowed minus the league pool average for that position. Color cues: green = strong defense (below avg), red = weak defense (above avg).
  • Completeness pill — High (5+ games), Medium (3–4), Low (1–2), Insufficient (0).
  • Excluded — preseason games and All-Star events are filtered out at the source.

Refreshes with each NBA pipeline cycle. Source feed: public/nba/dvp.json, built by scripts/basketball/dvp.py --league nba.