Celtics Player Action Profiles, 2025-26

Generated 2026-05-19. Source: stats.nba.com/stats/synergyplaytype via nba_api 1.10.2, already on disk at data/raw/action_types/nba_api/. All 22 player-level files parsed. 14 Boston players returned. Companion deliverables:

• data/celtics_player_action_profiles_all.csv (126 rows, long format)
• data/celtics_player_action_strengths_weaknesses.csv (23 rows)
• data/celtics_player_action_matrix.csv (23 rows, wide-pivot)

Sample rule: any total-possession count below 10 is flagged in the long CSV and not used to pick "best/worst" actions in the strengths/weaknesses table. In the report tables, I note small-sample cells inline.

---

1. Executive verdict

• Boston's best PNR ball-handler in the playoffs was Pritchard (1.172 PPP, 88th percentile, rank 9 of 65) by a wide margin. Tatum, Brown, and White all sat between the 34th and 38th percentile.
• Boston's best roll man in both regular season and playoffs was Queta (86th and 96th percentile). Garza was a second elite RS roll-man (82nd) but did not produce a ≥10-poss playoff row. Vucevic's roll-man data was 18th percentile RS and 32nd playoffs.
• Boston had a deep elite spot-up corps: Hauser, Pritchard, Scheierman, and Simons all >78th percentile RS. White was a surprisingly weak spot-up shooter (36th RS, 15th playoffs).
• Boston's isolation hierarchy was inverted relative to public perception: Pritchard 90th pctile, Simons 82nd, Brown 77th, Tatum 13th, White 4th. Tatum's isolation has been the worst high-volume iso line on the roster two years running.
• Brown was the team's most efficient post-up option in the playoffs (93rd pctile, 1.167 PPP), not Vucevic or Queta.

2. Data source and coverage

• 22 player-level Synergy JSON files (11 play types × 2 season types).
• Boston players returned: 14 (Tatum, Brown, White, Pritchard, Queta, Garza, Vučević, Hauser, Scheierman, Simons, Walsh, Minott, Harper, González). Banton is in the BBR roster but had zero qualifying Synergy rows.
• Playoff GP for most players: 7. Tatum 6 (sat one game). Garza produced no qualifying playoff rows (DNP or below the qualifying threshold across all 11 play types). Simons same.

3. Player-by-player regular season profile (≥10 total possessions)

Compact table; full data in the matrix CSV. Volume = total possessions; PPP percentile is NBA Synergy's published number.

Lead handlers

Wings / shooters

Bigs

4. Player-by-player playoff profile (≥10 total possessions)

7-game series. Rows below threshold flagged.

5. Tatum profile

Highest-volume action (RS, 16 GP): PRBallHandler 7.4/g, 118 poss, 0.966 PPP, 77th percentile. Reasonable.

Highest-volume action (playoffs, 6 GP): PRBallHandler 9.7/g, 58 poss, 0.776 PPP, 38th percentile. Volume UP, efficiency DOWN.

Best regular-season actions: Spotup 1.244 (91st pctile, 42 poss), PRBallHandler 0.966 (77th), Postup 1.105 (72nd, 19 poss), OffScreen 1.158 (76th, 19 poss). Tatum is a good off-ball player when not asked to create.

Hurt-Boston action: Isolation 0.671 (13th pctile, 82 poss regular season). Two-year red flag. He should not iso.

Playoff off-ball was outstanding: Spotup 1.667 PPP (98th pctile, 15 poss), Transition 1.533 (96th pctile, 15 poss). Small samples but extreme efficiency. Off-ball Tatum was Boston's best playoff offensive weapon by percentile, by a clear margin.

Role suggestion: Lower iso share. Lower on-ball PNR share if the 6-game playoff PPP signal holds. Run him into more Spotup / OffScreen / Transition catches; he was a 98th- and 96th-percentile finisher in those.

6. Brown profile

Highest-volume action (RS): PRBallHandler 6.4/g, 442 poss, 0.932 PPP, 70th percentile. The most utilized action on the roster.

Highest-volume action (playoffs): Isolation 6.6/g, 46 poss, 0.935 PPP, 60th percentile. Note: ISO overtook PNR-handler in the playoff share. PRBallHandler was 6.0/g, 42 poss, 0.762 PPP, 36th percentile.

Best regular-season actions: Cut 1.286 (45th pctile, 28 poss), Transition 1.130, Isolation 1.013 (77th, 393 poss), OffScreen 1.007.

Hurt-Boston action: PRBallHandler PPP fell to 36th pctile in playoffs (from 70th RS). Postup was a positive (93rd pctile playoff, 12 poss).

Role suggestion: More Iso vs targetable defenders (RS 77th pctile, playoff 60th — still above average). More Postup vs smaller defenders. Reduce PNR-handler share, especially in late-clock situations.

7. White profile

Highest-volume action (RS): PRBallHandler 5.7/g, 428 poss, 0.923 PPP, 67th percentile.

Highest-volume action (playoffs): Spotup 5.3/g, 37 poss, 0.730 PPP, 15th percentile — a steep regular-season-to-playoffs collapse (RS 36th to playoff 15th).

Best regular-season actions: Transition 1.186 (62nd), OffScreen 1.083 (70th), PRBallHandler 0.923 (67th), Cut 1.389 (66th, small).

Hurt-Boston action: Isolation 0.537 (4th percentile, 67 poss regular season). White should never iso. Playoff Spotup collapse (0.730, 15th pctile) was the single biggest individual-action drop in the team profile.

Role suggestion: Off-ball cutter, transition, secondary PNR- handler. Avoid spot-up over-emphasis in the playoffs (catch-and-shoot under pressure was the issue). Avoid iso entirely.

8. Pritchard profile

Highest-volume action (RS): PRBallHandler 4.4/g, 339 poss, 0.944 PPP, 73rd percentile.

Highest-volume action (playoffs): PRBallHandler 4.1/g, 29 poss, 1.172 PPP, 88th percentile, rank 9 of 65. The cleanest playoff profile on the team.

Best regular-season actions: Isolation 1.110 (90th pctile, 162 poss) — Boston's most efficient iso option by a wide margin. Spotup 1.212 (86th, 331 poss). Misc 0.902 (94th, small).

Hurt-Boston action: None at meaningful volume. Floor is high.

Role suggestion: Pritchard deserved more PNR-handler reps in the playoffs. He took 4.1 per game while Tatum took 9.7 and Brown 6.0. At 88th-percentile playoff PPP, he should have been at 6+ per game.

9. Queta profile

Highest-volume action (RS): Cut 2.5/g, 188 poss, 1.366 PPP, 63rd percentile.

Best regular-season actions: Transition 1.409 (94th pctile, 45 poss), PRRollman 1.348 (86th, 158 poss), Cut 1.366 (63rd, 188 poss), OffRebound 1.234 (70th, 135 poss).

Highest-volume action (playoffs): PRRollman tied with Cut at 2.0/g each (14 poss each). PRRollman PPP 1.786 (96th percentile), Cut 1.500 (79th).

Hurt-Boston action: Spotup 0.850 (15th pctile RS, 22 poss), Postup 0.714 (10th pctile, 15 poss), Misc 0.438 (turnovers, 64% TOV rate). Never let Queta shoot off the catch from outside; never post him up.

Role suggestion: More PNR rolls. More cuts. Avoid post-ups and catch-and-shoots. The 2.0 PRRollman possessions per game against PHI was severe under-deployment of a 96th-percentile finisher.

10. Garza / Vučević / big-man comparison

*Vučević's full 2025-26 MP includes pre-BOS games. He joined Boston late.

Two real elite rollers, one drag. Queta is the everyday roller. Garza is the highest-PPP small-sample roller AND a 95th-percentile spot-up shooter — a unique stretch-and-roll combination at low usage. Vučević's roll-man PPP is 18th percentile RS and 32nd in playoffs; his Postup is 18th. The numbers do not justify his playoff role over Garza.

Vučević's 2.3 PRRollman possessions per game in the playoffs at 0.857 PPP took away minutes from Queta's 2.0 possessions per game at 1.786. That single substitution is the biggest individual-rotation mistake the data flags.

11. Hauser / Scheierman / wing shooting profile

Hauser is Boston's signature spot-up shooter. Scheierman held up in playoff spot-up (78th percentile, 19 poss), the only secondary shooter who maintained efficiency. White's spot-up collapsed; Walsh's collapsed to a 4th-percentile small-sample line.

12. Which actions Boston should increase

1. Queta PRRollman. 96th playoff percentile at 2.0 possessions/game is severe under-deployment of an elite finisher.
2. Pritchard PRBallHandler. 88th playoff percentile at 4.1 / game while Tatum (38th) ran 9.7 and Brown (36th) ran 6.0.
3. Pritchard Isolation. 90th RS percentile at 2.1/game. Highest iso efficiency on the roster.
4. Garza PRRollman and Spotup. 82nd and 95th RS percentile at 1.6 and 1.0/game respectively. Stretch-and-roll combination is underused. (Note: no qualifying playoff rows — confirm whether coaches benched him in the series or whether sample fell below qualifying threshold.)
5. Tatum off-ball (Spotup, OffScreen, Transition). 98th/76th/96th playoff percentile across these three. Use him as the finisher more than the creator if the playoff signal holds.
6. Brown Postup vs mismatches. 93rd playoff percentile at 1.7/game, 12 poss. Underused leverage.
7. Hauser Transition. 94th playoff percentile at 2.0/game. Run- the-floor weapon.

13. Which actions Boston should reduce

1. Tatum Isolation. 13th RS percentile at 5.1/game, 82 poss. He was somewhat better in the small playoff sample (64th, 18 poss) but the RS volume is the issue. Stop iso-ing Tatum.
2. White Isolation. 4th RS percentile at 0.9/game, 67 poss. Should approach zero.
3. Vučević PRRollman. 18th RS / 32nd playoff percentile. Boston replaced Queta minutes with Vučević in the playoff PNR roll-man slot — that swap was negative-EV.
4. Hauser PRRollman. 20th RS percentile at 0.7/game, 53 poss. He is not a roll man. Whatever lineup constructs put him in that role should stop.
5. White Iso/Spotup playoff over-share. The playoff Spotup line (0.730, 15th pctile, 37 poss) is the team's biggest individual- action collapse. Some of that volume should shift to Hauser or Scheierman.
6. Brown PNR-handler late-clock. 36th playoff percentile at 6.0/g. The volume is fine; the efficiency under pressure is the issue. Move some of those reps to Pritchard.

14. RealGM-ready summary

The actual 2025-26 Celtics player action data refuses several pieces of the conventional read of this team. Pritchard - not Tatum or Brown - was Boston's most efficient PNR ball-handler in the playoffs at 1.172 PPP and 88th percentile, but he took 4.1 possessions per game while Tatum took 9.7 (at 38th percentile) and Brown took 6.0 (at 36th). Queta was a 96th-percentile playoff PNR roll-man at 1.786 PPP and got two possessions per game; Vučević took 2.3 in the same action at 0.857 PPP. Tatum's regular-season isolation profile sits at the 13th percentile across 82 possessions and that's a two-year flag, not a one-year blip - he should not iso. Tatum off-ball, by contrast, finished playoff Spotup possessions at 98th percentile and Transition at 96th, so the answer is to run him into catch-and-shoot and broken-floor opportunities rather than empty side PnRs. White's playoff Spotup collapse to the 15th percentile is the largest single-action efficiency drop on the team. The good news is the cure for almost all of it is on the roster already: more Pritchard PNR-handler, more Queta and Garza rolls, more Hauser and Scheierman spot-up volume, more Tatum off-ball attacks. Boston has the pieces. The 2025-26 playoff usage didn't deploy them in efficiency order.

---

Sources and methodology

• Long CSV: data/celtics_player_action_profiles_all.csv (126 rows)
• Strengths/weaknesses: data/celtics_player_action_strengths_weaknesses.csv
• Matrix: data/celtics_player_action_matrix.csv
• Raw JSON: data/raw/action_types/nba_api/*_P_offensive_PerGame.json
• Companion: reports/celtics_pnr_dho_action_analysis_REPAIRED.md, reports/celtics_pnr_ballhandler_outcome_analysis.md
• Sample threshold: ≥10 total possessions to count as a "real" volume row
• Playoff GP: 7 for most players, 6 for Tatum, 0 qualifying for Garza / González / Harper / Minott / Simons