2026-27 projection
| GP | MPG | FG% | FT% | 3PM | PTS | REB | AST | STL | BLK | TO |
|---|---|---|---|---|---|---|---|---|---|---|
| 72 | 30.0 | .470 | .750 | 1.4 | 17.5 | 5.4 | 7.2 | 1.2 | 0.3 | 3.1 |
- Role: starting combo guard and primary half-court organizer in many lineups.
- 9-cat: 55-85.
- Points: top 35-55.
- Dynasty: 20-45; July crowdsourced keeper rank No. 24, expert rank No. 40.
Development signal
Castle jumped to 16.7 points, 5.3 rebounds and 7.4 assists in 30.0 minutes in year two while improving to 47.1% from the field. His 40-point, 12-rebound, 12-assist triple-double illustrates the multi-category ceiling, but the season-long 73.4% FT and 3.2 turnovers explain why conventional category rank trails the box-score excitement. ESPN stats, accessed 2026-07-18 · NBA game account, 2026-02-07
The San Antonio Spurs project him for 30 shooting-guard minutes. He will share creation with De'Aaron Fox and Dylan Harper, so projecting another assist jump is less prudent than projecting modest shooting growth.
Ceiling and floor
- Ceiling: top 35 in 9-cat if he reaches 77% FT, 1.7 threes and 1.5 steals without losing assists.
- Floor: top 100 if 2025-26's FT/turnover combination persists and Harper takes more on-ball work.
- Points leagues: rebounds and assists create a stronger floor; turnovers matter only in harsher systems.
- Dynasty: a cornerstone, though not yet the category asset implied by his real-life impact.
Hashtag lists Castle at expert dynasty No. 40 and keeper No. 24; that disagreement is a useful warning against buying solely from age and highlight outcomes. Hashtag dynasty, accessed 2026-07-18 · Hashtag keeper market, updated 2026-07-18
Career record
Pre-NBA: UConn, one season (2023-24). He played 34 games with 30 starts at 27.0 minutes and averaged 11.1 points, 4.7 rebounds, 2.9 assists, 0.8 steals and 0.5 blocks on .472 from the field, .267 from three and .755 at the line. He was Big East Freshman of the Year and Big East All-Freshman, and started for the national championship team. The 26.7 percent three-point mark on a title team is the relevant carryover: he arrived as a downhill creator and defender whose jumper was the open question, and it still is. Draft: 2024, first round, 4th overall by San Antonio — no draft-night trade. He won Rookie of the Year for 2024-25.
| Season | Age | Team | GP | MPG | PTS | REB | AST | STL | BLK | 3PM | FG% | FT% | TO |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2024-25 | 20 | SAS | 81 | 26.7 | 14.7 | 3.7 | 4.1 | 0.9 | 0.3 | 1.2 | .428 | .724 | 2.2 |
| 2025-26 | 21 | SAS | 68 | 30.0 | 16.7 | 5.3 | 7.4 | 1.1 | 0.3 | 1.2 | .471 | .734 | 3.2 |
| Career | SAS | 149 | 28.2 | 15.6 | 4.4 | 5.6 | 1.0 | 0.3 | 1.2 | .448 | .729 | 2.6 |
Fantasy history: Season-end nine-category ranks are not verifiable from public sources here and are not asserted. The two-season shape is legible without them. As a rookie he was a category liability on efficiency — .428 from the field with 2.2 turnovers and thin stocks — while producing points-league value on volume. In 2025-26 the field goal percentage moved from actively harmful to roughly neutral and the assists became a genuine plus, but 3.2 turnovers is a top-of-league figure that eats much of the gain, and 0.3 blocks with 1.1 steals means he contributes little in the two defensive categories. The points-league shape is materially better than the category shape and has been in both seasons: a 16.7-point, 5.3-rebound, 7.4-assist line is a strong points-format asset, and most points systems penalize turnovers far more gently than nine-cat does.
Durability: 81 games as a rookie, 68 in 2025-26. The rookie year was essentially clean. The second season was not: he suffered a left hip flexor strain in the first half against Sacramento in mid-November 2025 that was projected at two to three weeks and cost roughly nine games, and he later dealt with right hip soreness, a pelvic contusion and a thumb injury. None of those are structural and none required surgery, but three separate hip-region issues in one season is a pattern worth naming rather than filing as bad luck, particularly for a guard whose value comes from downhill drives. The dossier's 72-game projection is therefore an assumption of partial recovery toward his rookie availability, not a reading of his most recent season.
Trajectory: The year-two leap is real but narrower than the counting stats suggest, and the distinction matters. Per-36 scoring was essentially flat between the two seasons at roughly 20 points, so the jump from 14.7 to 16.7 was minutes, not scoring growth. The two things that genuinely changed are the passing, which went from about 5.5 to 8.9 assists per 36 — an enormous move that reflects San Antonio handing him the half-court organizing job — and the field goal percentage, which climbed from .428 to .471 on a better shot diet. What did not change is the shooting: 1.2 made threes in both seasons and a free throw percentage stuck in the low 70s, which is the same profile UConn saw. At 21 he sits at the very front of the aging curve with real room, but the 2026-27 setup argues against extrapolating the assist curve any further, because Fox and Harper both need on-ball possessions and the assist spike came partly from availability of touches that will now be contested. The most likely progression is the one the dossier implies: flat-to-slightly-down assists, modest shooting improvement, and a category value that finally depends on whether he can cut the turnovers rather than on whether he can score.
Career sources: ESPN career stats · StatMuse season-by-season · Wikipedia — UConn stats, draft, bio · NBA.com — hip flexor strain timeline
Open questions
- Is his post-All-Star shooting improvement durable at four or more attempts per game?
- Which of Castle, Fox and Harper leads the team in assists when all three are healthy?