Passes (passes, handoffs, inbounds) and drives are checked against the NBA's own per-game counts: about 98% / 93% and 92% / 91% (found / right). A drive starts 20+ ft from the rim, gets within 18 ft and isn't a fast break.
Ball screens: an on-ball screen, how the defense played it (over, under, switch, trap) and what the screener did after (roll, pop, short roll). The short bar on the screener is the screen itself: square to the defender he's screening, on that defender's side, at the angle they met; it's dashed when the screener was still moving (over 5 ft/s, about walking pace: a slip or a brush screen, about 1 in 4). Each one also shows how deep the screener's defender was at the screen ("big 7 ft below the screen"): drop here means 10+ ft below, as Shax charts it (a higher drop is "level" to him); other charters call a shallower big a drop too, so the number lets you read it either way. Only screens that led to baskets can be checked against the NBA (about 80%).
Shot types: each shot's play type (spot up, P&R ball handler / roll man, handoff, off-ball screen, cut, iso, post up, transition, putback) and contest (open / light / heavy / blocked), in Shax's vocabulary. The play type comes from a model trained on both hand charters, Shax's Knicks (10 games) and David Lee's 2025-26 Hawks (31 games, a season without the NBA's shot details), reading what the dots show before the shot: the catch, the shooter's path, the screens and handoffs around him. On games it never trained on, it agrees with Shax ~79% of the time and with David ~71% (the hand-written rules it replaced: ~69% / ~59%). Most of the rest is a screen the dots can't confirm (ghost screens, some curls) or a judgement call on which action gets the credit; contest ~62% with Shax's (open vs contested ~73%); the contest is the NBA feed's closest defender joint to the ball. Where the feed has no shot details for a shot (OKC–DAL and most of LAL–LAC, Oct 28), the contest marked est. is estimated from the dots around the shot (nearest defender to the shooter and the ball): it agrees with the measured contest on 90% of the shots that have one, and is left blank when less sure (about half). Estimates aren't included when you search by contest.
Help (experimental): a defender who leaves his man toward a driver or a roller (HELP: gets to him or in front of him; TAG: meets the roller; STUNT: a fake at him), the dashed line pointing to the man he left. Checked so far against the analysts' posts. Ball-screen coverage uses Shax's vocabulary: the handler's defender (over / under) | the big (drop / level / hedge-blitz / switch), tuned on his Knicks charting and checked on @dlee4three's Hawks and on video (over / under 79% / 88%, the big 70% on Shax's; the big is 65% on @dlee4three's in his own terms: he has no 'level', so a big who stays back is a drop at any depth). A hedge / blitz call needs the big up near the screen's level, within reach of the ball while the handler is still out high, and moving toward the ball: a big who stands there or backs up while the play comes to him is level or drop. Topside: on a side screen the handler's defender set up 2+ ft on the top side, the ICE setup; only about 1 in 4 is an ICE in his charting (a defender who then goes over looks the same beforehand), so it's a flag, not a call. Trap: both screen defenders within 6 ft of the ball handler for 0.6 s or more while he is still out around the three-point line (22+ ft from the rim), in the 2.4 s after the screen, and the big still on him when he gives the ball up: two on the ball, high, held. A hedge recovers before the pass; a blitz stays. It marks about half of the blitzes both charters logged, and about 1 in 8 of their hedges too, so it's a flag, not a blitz call.
Off-ball screens (experimental): two teammates without the ball meeting, the slower one the screener. Set: the screener stood within 3 ft of the line the cutter's defender ran and that defender got within 4 ft of him (a legal screen puts the body in the defender's way). Unconfirmed: either failed, so the tracking doesn't confirm the screener was in the way; some are still real screens (a defender who goes wide around one never reaches it), others are teammates passing near each other. About 63 a team-game, 41 of them set. Against the NBA's screen assists, keeping only set ones is right a bit more often (81% vs 80%) but misses more (76% vs 78% found): the NBA credits only screens that lead straight to a basket and no public count covers the rest, so ✓ / ✗ reviews here are the real check.
Cuts (experimental): an off-ball player getting 8+ ft closer to the rim at 10+ ft/s and ending within 12 ft (the paint, short corner or dunker spot), named by context: cut in the half court, rim run in transition (within 6 s of gaining the ball), crash (going to the rim with a teammate's shot in the air), roll (right after setting a ball screen). No public count of cuts exists; against two hand charters' cut finishes (Shax's Knicks, @dlee4three's Hawks) we find a cut on 40% / 33%. Most of the rest are dump-offs to a big already under the rim, or catches on the move outside the paint.
Handoffs: a dribble handoff when the giver had nearly stopped (3 ft/s or less) and then screened the receiver's defender (within 4 ft in the next 1.2 s), else a plain handoff.
Named sets (experimental) come from matching templates, mostly unchecked. Reviewing them with ✓ / ✗ is how they get checked. A few are checked against the charters: SLOB / BLOB, an inbound pass by a player standing past the sideline / baseline in the frontcourt, finds 98% of Shax's and 87% of @dlee4three's, and gets side vs baseline right 96%+ of the time. Iverson cut: a player runs across the top of the key, above the elbows, from one wing to the other off a screen and catches; it finds 9 of @dlee4three's 13 tracked Iverson cuts. Elevator doors needs both screeners side by side together with the cutter coming up between them, so it's rare (it used to fire on any two screens, which called Iversons elevators).
Made shots: rings burst from the rim in the scorer's colour as the ball goes in, the score updates at that moment, and playback eases for a beat. Until the other team's inbound pass the ball is greyed out and marked as dead, so the walk out of bounds doesn't look like a live ball after a miss.
Players: the strip under the court shows who's on the floor (headshot, jersey number, name). Hover a headshot or a dot to find that player and their path; click or tap to keep them highlighted (several at once is fine). Names labels every dot.
3D is the Discord /replay3d view: pins as tall as the listed height that rise on jumps; contested shots show the defender's measured hand height in purple.
All games builds a queue across every tracked game (newest first). For detected actions each clip centers on one detection, and Y / N marks it and moves on: the fastest way to review a layer.
Credits: all of the data (tracking, play-by-play, games, players, scores, headshots) is the
NBA's, and credit for it goes to the NBA.
The shot play types, contest levels and ball-screen coverage words follow
Shax (@ShaxNBA)'s charting vocabulary,
and the agreement numbers above are measured against his hand-charted Knicks games and
David Lee (@dlee4three)'s hand-charted 2025-26 Hawks.